Exhibit 96.7 S-K 1300 TECHNICAL REPORT SUMMARY FOR SIBANYE-STILLWATER ON KELIBER LITHIUM PROJECT, FINLAND Prepared for: Sibanye-Stillwater Limited Report Prepared by: VBKOM (Pty) Ltd EFFECTIVE DATE: 31 December 2024 ISSUE DATE: 25 April 2025 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland ii DATE AND SIGNATURE PAGE This Report titled S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland was prepared for Sibanye-Stillwater. This Report has been compiled in accordance with the United States Securities and Exchange Commission Part 229 Standard Instructions for Filing Forms Regulation S-K subpart 1300. The effective date of this Report is 31 December 2024. The Mineral Resources were prepared by CSA Global South Africa (Pty) Ltd (an ERM Group company), while Mineral Reserves were prepared by VBKOM (Pty) Ltd. Company Authorised Signatory Signature Place and Date VBKOM (Pty) Ltd Otto Wilhelm Warschkuhl /s/ Otto Wilhelm Warschkuhl Centurion, Gauteng, South Africa on 25 April 2025 CSA Global South Africa (Pty) Ltd CSA Global South Africa (Pty) Ltd /s/ CSA Global South Africa (Pty) Ltd Woodmead, Gauteng, South Africa on 25 April 2025 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland iii IMPORTANT NOTICES Sibanye Stillwater Limited (SSW, also referred to as Sibanye-Stillwater, the Company or the Registrant) holds a 79.82% share in the Keliber Lithium Project (‘the Keliber Project’ or ‘the Project’), which is located in Central Ostrobothnia, Finland, through its 100% interest in Keliber Lithium (Pty) Ltd (Keliber). The Keliber Project is in the developed stage and is currently undergoing construction, with the Keliber lithium refinery and the Päiväneva concentrator under construction, and the Syväjärvi open-pit mine infrastructure to commence at the start of Q3 2025. This technical report summary (TRS) builds on the Mineral Resources estimate update TRS (initial assessment format) produced by The ERM International Group Limited (ERM) in 2024 on behalf of SSW according to the United States Securities and Exchange Commission’s (SEC’s) Subpart 1300 of Regulation S-K (S-K 1300). This new TRS was prepared to reflect the material increase in Mineral Reserves that has come about due to incorporating the updated Resource estimate, as reported on in 2024. In addition, the TRS reflects the material changes and progress that have been made in the permitting and construction of the operation, as well as incorporates an updated and enhanced market study section. In an important departure from the previous comprehensive TRS reported on in 2022, the basis of Mineral Reserve declaration has shifted from a spodumene concentrate only, to include the Keliber lithium refinery, producing a final product of lithium hydroxide monohydrate (LiOH.H2O). VBKOM (Pty) Ltd (VBKOM), as Mineral Reserves Qualified Person, have ascertained themselves that the testwork and piloting conducted is sufficient to ameliorate any material risk associated with extraction technology employed. The impact of the changes to the ramp-up schedules has been evaluated and the associated risk has been highlighted. This report includes technical information, which requires subsequent calculations to derive subtotals, totals, and weighted averages. Such calculations may involve a degree of rounding and consequently introduce an error. Where such errors occur, VBKOM does not consider them to be material. The reader and any potential or existing shareholder or investor of SSW are cautioned that SSW are involved in exploration on the Keliber Project, and there is no guarantee that any unmodified part of the Mineral Resources will ever be converted into Mineral Reserves nor, ultimately, extracted at a profit. This report uses a shorthand notation to demonstrate compliance with S-K 1300, for example: [[§229.601(b)(96)(iii)(B)(2)] represents subsection (iii)(B)(2) of section 96 of CFR 229.601(b) (“Item 601 of Regulation S-K”). Subsequent Events Post 31 December 2024 (the effective date of the TRS), SSW have embarked on a detailed review of the anticipated ramp-up schedule of Keliber lithium refinery, as well as the capital estimate to complete the construction of the Project. At the time of reporting (25 April 2025), this review had not yet been concluded. The amended permit of the Keliber lithium refinery was received in January 2025 (4/2025 number: LSSAVI/13185/2024) and became legally valid on 20 February 2025. The chemical permit for the concentrator plant was approved on 3 April 2025. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland iv CO-ORDINATING AUTHORS Author Qualifications & Affiliations Signature Otto Wilhelm Warschkuhl Principal Mining Engineer BEng (Hons) Mining, Pr.Eng. SAIMM, CIM /s/ Otto Wilhelm Warschkuhl Chantelle Obermeyer Principal Geologist BSc (Hons) Geology, MSc Geology Pr.Sci.Nat /s/ Chantelle Obermeyer Ruben Els Principal Mining Engineer BEng Mining Pr.Eng, SAIMM /s/ Ruben Els Brandon Okhuis Senior Mining Engineer BEng (Hons) Mining Cand. Eng. /s/ Brandon Okhuis Mientjie van der Vyver Principal Metallurgist PhD Metallurgy PMP, Pr.Sci.Nat, SAIMM /s/ Mientjie van der Vyver Maria Antoniades Senior Geologist BSc (Hons) (Gly), MSc Env.Sci. Pr.Sci.Nat, GSSA, SAIMM /s/ Maria Antoniades Tiaan Ackermann Industrial Engineer BEng Industrial, MBA Pr.Eng, PRINCE2 /s/ Tiaan Ackermann

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland v DISCLAIMER This Report has been prepared by VBKOM (Pty) Ltd (VBKOM) for Sibanye-Stillwater Limited (SSW). SSW and its contractors have provided certain information, reports and data to VBKOM for preparation of this Report relating to operational activities and methodologies, which, to the best of SSW’s knowledge and understanding, is complete, accurate and true. VBKOM has, as far as is practically possible, verified this information from independent sources after making due enquiry of all material issues required to comply with the requirements of the Item 601 of the United States Securities and Exchange Commission’s Subpart 1300 of Regulation S-K. Unless otherwise expressly stated, the analyses contained in this Report are developed from such information provided by SSW, thus being reliant on the accuracy and completeness of the supplied data. The comments presented in this Report apply to project features and extent of information available as at the time of VBKOM’s investigations, and those reasonably foreseeable. VBKOM shall not be responsible for any errors or omissions in the supplied information and shall not accept any consequential liability arising from commercial decisions or actions resulting from them. The findings, conclusions, and opinions of VBKOM are based on the scope of services as defined within the contractual agreement with SSW. VBKOM accepts no liability for damages, if any, suffered by any third party as a result of decisions made or actions based on this Report. No warranty or guarantee is made by VBKOM with respect to the completeness or accuracy of the legal aspects of this document, be it express or implied. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland vi EXECUTIVE SUMMARY [§229.601(b)(96)(iii)(B)(1)] I. Introduction This Technical Report Summary (TRS) of the Keliber Lithium Project (‘the Keliber Project’ or ‘the Project’) was compiled by VBKOM (Pty) Ltd (VBKOM) on behalf of Sibanye-Stillwater Limited (SSW, also referred to as the Company) according to Item 601 of the United States Securities and Exchange Commission’s (SEC’s) Subpart 1300 of Regulation S-K (S-K 1300). SSW holds a 79.82% share in the Keliber Project, which is located in Central Ostrobothnia, Finland, through its 100% interest in Keliber Lithium (Pty) Ltd (Keliber). This report is the third TRS for the Keliber Lithium Project and supports the disclosure of Mineral Resources and Mineral Reserves at 31 December 2024. The Mineral Resources and Mineral Reserves have been prepared and reported according to the requirements of S-K 1300. Mineral Resources have been prepared by ERM International Group Limited and are fully incorporated into this Report. II. Effective Date [§229.1302(b)(iii)(3)] The effective date of the TRS is 31 December 2024, which satisfies the S-K 1300 requirement of a current report. III. Property Description and Ownership The Keliber Lithium Project is situated in Central Ostrobothnia in Western Finland in the municipalities of Kaustinen, Kokkola, and Kruunupyy. There are nine elements to the Project: • Seven spodumene exploration or mining properties at Syväjärvi, Rapasaari, Länttä, Outovesi, Emmes, Leviäkangas, and Tuoreetsaaret; • The Keliber lithium concentrator at Päiväneva next to the Rapasaari mining property; and • The Keliber lithium refinery at the Kokkola Industrial Park (KIP). The figure overleaf depicts the location of the various Project elements. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland vii Location of the Various Project Elements Location of the various Project elements. Keliber own land at Syväjärvi (47.39 ha (~28%) of the current mining area of 166.3 ha), at Rapasaari (400.15 ha (~82%) of the current mining area of 488.97 ha) and at Outovesi (41.73 ha (~20%) of current claim areas of 209.67 ha). Compensation to the landowners is due on the valid mining and exploration permits, according to the Mining Act (621/2011) which was amended by Act 505/2023. Compensation on the granted exploration permits and the applications will become due once the permits become legally valid. IV. History None of the properties have previously been mined, although the mining rights to the Länttä, Emmes, and Syväjärvi deposits were first owned by Suomen Mineraali Oy, then by Paraisten Kalkkivuori Oy and, from the early 1960s to the early 1980s, by Partek Oy. These rights expired in 1992, and the areas were unclaimed until 1999 when Olle Siren, together with private partners, claimed the Länttä deposit and later the Emmes deposit. From 2003 to 2012, the Geological Survey of Finland (GTK) held ownership of the Syväjärvi and Rapasaari deposits. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland viii V. Geology and Mineralisation The Keliber Project is located in the Kaustinen Lithium Pegmatite (KLP) Province of western Finland, covering an area of about 500 km2. The underlying geology comprises Palaeoproterozoic (1.95–1.88 Ga) supracrustal rocks of the northern Pohjanmaa Belt, which forms a 350 km long and 70 km wide arcuate belt between the Central Finland Granitoid Complex to the east and the Vaasa Granitoid Complex in the west. The Pohjanmaa Belt is host to several pegmatite groups/provinces, and the northern parts of the belt have been intruded by several Lithium-Cesium-Tantalum (LCT)-type pegmatites, with a majority of those belonging to the KLP Province being of the albite/spodumene type. At least ten individual pegmatite deposits have been discovered to date within the KLP, with most having subsequently been evaluated by drilling methods only. Historical exploration comprised identification and mapping of spodumene-bearing pegmatite boulders as outcropping pegmatites, and their host rocks are rare, most being covered by 3–18 m of overburden (OVB) comprising surficial sediments (mostly glacial till) followed by drill testing of the targets. Historical drilling in the period 1960–1980 and, more recently by GTK and Keliber (ongoing), has resulted in the delineation of seven discrete LCT pegmatite deposits to a relatively high level of confidence, namely Syväjärvi, Rapasaari, Länttä, Emmes, Outovesi, Tuoreetsaaret, and Leviäkangas. All of the pegmatites that have been discovered and evaluated to date within the Kaustinen area have very similar mineralogy and are dominated by albite (37–41%), quartz (26–28%), K-feldspar (10–16%), spodumene (10–15%), and muscovite (6–7%). Internal pegmatite zonation, as seen in many other similar LCT-type pegmatites, is poorly developed to absent from the Kaustinen pegmatites, with spodumene being the only lithium-bearing mineral that is of economic interest and is generally homogeneously distributed throughout most of the pegmatites. Several deposits display frequent inclusion or incorporation of host rock xenoliths within the modelled pegmatites and represent a form of internal dilution to the pegmatites. At most of the deposits, no weathering is observed; however, at the Rapasaari deposit, partial weathering or fracture oxidation occurs to a depth of 20–30 m, but it is understood this is rare and does not appear to significantly alter the spodumene or affect the lithium grades. VI. Status of Exploration, Development, and Operations With the exception of some shallow surface reverse circulation drilling completed by GTK over the Syväjärvi and Leviäkangas deposits, all drilling on the Project has been completed using diamond core drilling methods. Diamond core drilling has been the only exploration method used to generate the geological, structural, and analytical data over the deposits, which have been used as the basis for Mineral Resource estimation over each of the deposits defined to date. Keliber’s systematic hard rock-focussed lithium exploration programme in the Kaustinen area has successfully delineated the seven discrete spodumene-bearing LCT pegmatite deposits for which Mineral Resource estimates have been reported. Older drilling phases from the 1960s to early 1980s were executed by Suomen Mineraali Oy and Partek Oy, targeting the Emmes, Länttä, Leviäkangas, and Syväjärvi deposits. This was followed by GTK which completed drilling over the Syväjärvi and Rapasaari deposits between 2004 and 2012. Since 1999, Keliber has completed extensive drilling programmes focusing on delineating Mineral Resource estimates over each of these deposits, including the Outovesi deposit that Keliber discovered in 2010 and Tuoreetsaaret discovered in 2020.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland ix The work completed to date has captured all the important variables (mineralogical, structural, and lithological) required to properly define host pegmatite/s attitude and, importantly, spodumene or grade distribution within the various pegmatites that host each deposit. The drilling by Keliber has also served to validate the historical drilling completed. Summary of drilling completed over the Keliber Lithium Project (source: Keliber, 2023). Historical & GTK Keliber Total No. of Drill Holes Length (m) No. of Drill Holes Length (m) No. of Drill Holes Length (m) Syväjärvi 91* 4,197 170 19,385 261 23,582 Rapasaari 26 3,655 321 56,651 347 60,306 Länttä 54 3,494 54 5,691 105 9,185 Emmes 31 3,348 23 2,937 54 6,285 Outovesi – – 24 1,816 24 1,816 Tuoreetsaaret 2 270 103 24,143 105 24,413 Leviäkangas 106** 5,850 49 5,127 155 10,977 Total 310 20,814 744 115,750 1,051 136,564 * Includes 57 percussion holes. ** Includes 60 percussion holes. In February 2022, Keliber issued a final Definitive Feasibility Study (WSP Global Inc., 2022) based on the forecast production of 15,000 tpa of battery-grade lithium hydroxide (BG). The Definitive Feasibility Study is considered to be the equivalent of a Feasibility Study (FS) as defined in the S-K 1300 Definition Standards. This study used the FS issued in February 2019 as basis for most of the technical work. In 2022, SRK Consulting South Africa (Pty) Ltd (SRK) reviewed this FS and classified it as a pre-feasibility study (PFS) in the Amended 2022 Keliber TRS. As part of the Amended 2022 Keliber TRS, Mineral Resources as at 31 December 2022 were reported for Syväjärvi, Rapasaari, Länttä, Emmes, Outovesi, Tuoreetsaaret, and Leviäkangas. Following successful exploration drilling, a materially revised and increased Mineral Resource estimate was conducted during 2023, which was the subject of a Technical Report Summary filed in April 2024. Keliber’s involvement in the Project began in 1999 when a group of investors led by Mr Olle Siren began evaluation of the Länttä deposit, where drilling commenced in 2004. Keliber then extended their exploration efforts to the rest of the Kaustinen region, where it has completed acquisition of exploration rights and extensive drilling programmes over all of the deposits, including the discovery of the Outovesi deposit in 2010 and Tuoreetsaaret in 2020. The Keliber Lithium Project consists of the Mineral Resource properties around Kaustinen, the Keliber Lithium Concentrator at Päiväneva near Kaustinen, the Keliber lithium refinery at Kokkola, and ongoing exploration activities. VII. Mineral Resource Estimates The in situ Mineral Resources at 31 December 2024 are summarised in the table below and overleaf on an attributable basis (Sibanye-Stillwater attributable ownership is 79.82%) and are reported exclusive of Mineral Reserves. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland x The Mineral Resources are reported above a cut-off of 0.5% Li2O. The majority of the declared Mineral Resources are classified as Inferred Mineral Resources (55%), with the remaining split between Measured Mineral Resources (7%) and Indicated Mineral Resources (38%). The majority of the Measured Mineral Resources and Indicated Mineral Resources are converted to Proven Mineral Reserves and Probable Mineral Reserves and are, therefore, excluded from the table overleaf. Keliber Mineral Resources, exclusive of Mineral Reserves*, at a 0.5% Li2O cut-off as at 31 December 2024 and reported on a 79.82% ownership basis. Deposit Mineral Resource Classification Tonnes (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) *Rapasaari Measured 0.05 0.59 1.27 1.7 Indicated 0.85 0.52 1.13 23.6 Measured + Indicated 0.90 0.53 1.14 25.3 Inferred 1.50 0.58 1.25 46.3 *Syväjärvi Measured 0.09 0.56 1.21 2.6 Indicated 0.19 0.59 1.28 6.1 Measured + Indicated 0.28 0.58 1.26 8.7 Inferred 0.28 0.58 1.24 8.6 Tuoreetsaaret Measured - - - - Indicated 0.33 0.43 0.94 7.6 Measured + Indicated 0.33 0.43 0.94 7.6 Inferred 1.38 0.40 0.87 29.5 Länttä Measured 0.33 0.59 1.27 10.4 Indicated 0.57 0.55 1.18 16.6 Measured + Indicated 0.90 0.56 1.21 27.0 Inferred 0.35 0.54 1.16 10.0 Emmes Measured - - - - Indicated 0.67 0.62 1.33 21.9 Measured + Indicated 0.67 0.62 1.33 21.9 Inferred 0.29 0.61 1.31 9.5 Outovesi Measured - - - - Indicated 0.13 0.64 1.38 4.4 Measured + Indicated 0.13 0.64 1.38 4.4 Inferred 0.12 0.67 1.44 4.3 Leviäkangas Measured - - - - Indicated 0.19 0.55 1.19 5.7 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xi Deposit Mineral Resource Classification Tonnes (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) Measured + Indicated 0.19 0.55 1.19 5.7 Inferred 0.55 0.47 1.00 13.6 Total Measured 0.47 0.58 1.26 14.7 Indicated 2.93 0.55 1.19 86.0 Measured + Indicated 3.41 0.56 1.20 100.7 Inferred 4.48 0.51 1.10 121.9 Notes: 1. Mt is million tonnes, kt is thousand tonnes, LCE is lithium carbonate equivalent. (conversions used: Li2O = Li x 2.153; LCE = Li x 5.324). 2. Figures have been rounded to the appropriate level of precision for the reporting of Mineral Resources. 3. Mineral Resources are stated as in situ dry tonnes; figures are reported in metric tonnes. 4. The Mineral Resource has been classified under the guidelines of S-K 1300. 5. The Mineral Resource has demonstrated reasonable prospects for economic extraction based on conceptual mining and costs parameters. 6. Mineral Resources are reported on a 79.82% ownership basis. 7. Concentrator recovery of 88% and refinery recovery of 86%. VIII. Mineral Reserve Estimates The S-K 1300 TRS for the Keliber Lithium Project provides an assessment of the Syväjärvi and Rapasaari deposits with focus on Mineral Reserve estimation, pit optimisation, geotechnical analysis, and life of mine (LOM) planning. The study focused on improving the Project confidence with updated geological models, economic parameters, and mining methodologies. The stakeholders of this Project dedicated considerable time and effort to establish updated modifying factors, mine design parameters, and scheduling criteria, all of which were essential in developing a robust business case and, ultimately, enabling the conversion of Mineral Resources into Mineral Reserves. Modifying Factors The Mineral Reserves estimated for the Syväjärvi and Rapasaari deposits were calculated using the modifying factors determined during the study to demonstrate the limits of the deposit which is to be exploited. The Syväjärvi and Rapasaari mining areas are defined by two key property limits: the mining lease boundaries and the pit constraint boundaries. The pit constraint boundaries, which lie within the mining lease boundaries of the respective mining areas, serve as a hard limit for the ultimate pit extents for each pit, ensuring compatibility with planned and approved infrastructure. The 2022 FS pit surface extents formed the basis for defining these constraints. Where feasible, the pit areas were expanded beyond the 2022 FS outlines without impacting infrastructure, maintaining appropriate stand-offs. The pit optimisation study and final pit designs were limited to the pit constraint perimeter strings, which ultimately limited the potential Mineral Reserves within the pits. Mining losses and dilution were addressed through block model regularisation using a dominant mass approach, re- blocking to a smallest mining unit (SMU) size of 5 x 5 x 2.5 m, deemed practical based on historical observations and selective mining practices. This SMU size balanced equipment productivity and dilution control, with dilution applied on a zero-grade basis for non-target rock types. The re-blocked model provided a realistic representation of ore and waste mixing, revealing global dilution and losses: Syväjärvi experienced 11.08% dilution and 14.97% losses, while S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xii Rapasaari values are 11.66% dilution and 15.67% losses. These outcomes were considered acceptable for the planned selective mining approach. Li2O Cut-Off Grade Ore and waste reclassification was determined by a marginal lithium cut-off value, calculated considering concentrator plant throughput following the re-blocking process. Key inputs included the diluted feed grade from the block model and processing recovery curves derived from metallurgical testwork (1.06% Li₂O grade). The marginal cut-off grade (COG) accounted for all processing and selling costs, excluding mining costs, and was aligned with plant capacity to achieve a target of 15,000 tpa of lithium hydroxide. The Mineral Resources COG was set at 0.5% Li₂O, with the lower marginal COG reflecting dilution and reallocated blocks. The Syväjärvi marginal COG was 0.19% Li₂O based on financial parameters, while the Rapasaari marginal COG of 0.30% Li₂O was constrained to the concentrator feed required to produce the targeted 15,000 tpa of lithium hydroxide monohydrate. Ore blocks were reclassified based on the marginal COG. Mining The 2024 study presents significant advancements in geotechnical engineering compared to previous assessments. A major improvement involved new core logging conducted within ±50 metres of pit slopes, which provided better modelling of fault (FLT) and shear zones. The open-pit optimisation was completed using Deswik Pseudoflow which generated economic pit shells according to various revenue factors. Key considerations for pit design and improved mining included revised pit slope angles from the updated geotechnical assessments, optimised mining sequencing to ensure efficient waste and ore mining, and selective mining considerations. The Syväjärvi pit was designed as a single large pit, with strategic ramp placements ensuring safe and efficient haulage. In contrast, Rapasaari’s pit, being much larger, was designed incorporating three pushbacks which enable phased ore extraction to maintain steady production rates with mostly constant stripping ratios. Ramp widths for both pits were fixed at 16 m and 28 m for single and dual access, respectively. Ramp designs were limited to 1:10 in order to ensure optimal haul truck productivities. The Syväjärvi and Rapasaari pit designs are illustrated in the figure overleaf.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xiii Plan view Isometric view Plan view Isometric view Open-Pit Mine Layout – Syväjärvi and Rapasaari Syväjärvi and Rapasaari open-pit designs. A drill, blast, load, and haul sequence are employed, with selective mining techniques considered to reduce dilution and maximise ore recovery. The equipment fleet has been selected to balance efficiency, selective loading practices, and optimal production throughput. The primary loading units will include backhoe excavators fitted with 4.6 m³ 40m200 m200m 200 m 500 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xiv buckets for ore and 5.8 m³ buckets for waste, while haulage will be managed using 64-tonne rear-dump trucks for waste and 41-tonne articulated trucks for ore. The Syväjärvi pit/mine was limited to a 540 ktpa production rate due to the limitations set on the environmental permit. No blending of material from different deposits was allowed. The Rapasaari open pit is scheduled to commence at the tail end of Syväjärvi to ensure sufficient waste stripping for continuous crusher feed once Syväjärvi’s ore feed stops. The Project targeted LiOH.H2O production of 15 ktpa in the Keliber Lithium Project LOM production schedule, which is achieved throughout the LOM with the exception of 2030 when transitioning from Syväjärvi to Rapasaari production. The combined LOM production for Syväjärvi and Rapasaari spans from 2025 to 2044, with plant feed ending in April 2045, amounting to a total of 19.8 years. Over the mine life, Syväjärvi is expected to produce 3.15 Mt of ore at an average Li₂O grade of 1.03%, while Rapasaari will yield 11.28 Mt of ore at 0.93% Li₂O. The stripping ratio for Syväjärvi is 5.14, whereas Rapasaari’s overall stripping ratio is 7.63. The overall production schedule for both pits and the annual Product schedule are shown in the figures below. Annual LOM Ore and Waste Production Schedule by Pit Annual LOM ore and waste production schedule by pit. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xv Annual LOM Final Product (LiOH.H2O) Annual LOM final product (LiOH.H2O). Mineral Reserves The updated Mineral Reserve estimates reflect significant advancements in geological modelling and mine planning. Compared to the 2023 study, the Syväjärvi and Rapasaari deposits have an increase in Mineral Reserves of 3.81 Mt (41%) and an increase of 98 kt LCE (46%). This improvement was driven by the adoption of a regularised block model with updated dilution estimation which enhanced the accuracy of reserve estimation. For Syväjärvi, the Proven Mineral Reserves and Probable Mineral Reserves amount to 2.91 Mt at an average grade of 1.06% Li₂O. The Rapasaari deposit holds 10.06 Mt of Proven Mineral Reserves and Probable Mineral Reserves at 0.94% Li₂O. The open-pit Mineral Reserves for the Keliber Syväjärvi and Rapasaari operations are summarised in the table below. The Mineral Reserves are reported as delivered to the concentrator plant or associated stockpile, considering the applied modifying factors and SSW's 79.82% attributable interest in Keliber. Mineral Reserves for Keliber’s Syväjärvi and Rapasaari open-pit operations as at 31 December 2024. Syväjärvi Mineral Reserves as at 31 December 2024 - Reported on a 100% attributable basis 100% Mineral Reserves as at 31 December 2024 - Reported on a 79.82% attributable basis 79.82% Mineral Reserves Class Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Proven Mineral Reserves 1.92 0.54 1.16 55 1.53 0.539 1.16 44 Probable Mineral Reserves 0.99 0.40 0.86 21 0.79 0.402 0.86 17 Total/avg 2.91 0.49 1.06 76 2.32 0.492 1.06 61 Rapasaari Mineral Reserves as at 31 December 2024 - Reported on a 100% attributable basis 100% Mineral Reserves as at 31 December 2024 - Reported on a 79.82% attributable basis 79.82% Mineral Reserves Class Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Proven Mineral Reserves 2.42 0.48 1.03 62 1.93 0.478 1.03 49 Probable Mineral Reserves 7.63 0.43 0.92 173 6.09 0.427 0.92 138 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xvi Total/avg 10.06 0.44 0.94 235 8.03 0.439 0.94 188 Syväjärvi & Rapasaari Mineral Reserves as at 31 December 2024 - Reported on a 100% attributable basis 100% Mineral Reserves as at 31 December 2024 - Reported on a 79.82% attributable basis 79.82% Mineral Reserves Class Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Proven Mineral Reserves 4.34 0.50 1.09 117 3.47 0.50 1.09 93 Probable Mineral Reserves 8.62 0.42 0.91 195 6.88 0.42 0.91 155 Total/avg 12.96 0.45 0.97 311 10.35 0.45 0.97 248 Notes: 1. The Mineral Reserves estimate is reported in accordance with the requirements of S-K 1300 2. The Mineral Resources were reported exclusive of the Mineral Reserve 3. The Mineral Reserves are reported as run of mine (ROM) material delivered to the concentrator plant, or related ROM stockpile 4. Tonnage estimates are in metric units and reported as million tonnes (Mt) 5. Numbers may not add up due to rounding 6. Mineral Reserves are also reported on a 79.82% attributable to Sibanye-Stillwater basis 7. The Mineral Reserves are subject to EP approvals 8. Measured Mineral Resources converted to Proven Mineral Reserves 9. Indicated Mineral Resources converted to Probable Mineral Reserves 10. No Inferred Mineral Resources included in the Mineral Reserves 11. COG for Syväjärvi at 0.20% Li₂O 12. COG for Rapasaari at 0.30% Li₂O 13. Li (%) content was calculated by multiplying the Li₂O (%) content by a factor of 0.465 14. LCE content was calculated by multiplying the Li (%) content by a factor of 5.323 15. The Mineral Reserves are inclusive of the Keliber lithium refinery of which the process is likely to be the first implementation of this specific flowsheet. While the individual unit processes are not novel, the pilot trials have significantly de-risked the flowsheet. The Project’s financial model includes a reasonable ramp-up for the hydrometallurgical plant, and sensitivities to address commissioning risks were completed to incorporate the process. 16. The 31 December 2024 interpretation of the environmental permit regarding the production of 540 ktpa at Syväjärvi is under review. The inclusion of a 10% increase in the production limitation is also subject to environmental controls and approval. Conversions In line with industry standards, the total metal content of lithium Mineral Resources and Mineral Reserves is expressed as LCE, which is one of the final products produced in the lithium mining value chain. LCE is determined by converting the in situ lithium content using a factor of 5.323. To obtain lithium hydroxide monohydrate from LCE, the value is divided by a factor of 0.88. Lithium itself is derived from lithium oxide using a conversion factor of 0.465. IX. Metallurgy and Mineral Processing There are three main process stages for producing lithium hydroxide from the Keliber Lithium Project ores: • Concentration (producing a spodumene concentrate from the ore); • Conversion (converting the spodumene from the unreactive α phase to the reactive β phase); and • Hydrometallurgical processing (leaching, purification, and crystallisation stages to extract the Li from the β spodumene) and precipitate it as high purity lithium hydroxide monohydrate.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xvii Ore Sorting Initial ore-sorting tests explored optical and laser technologies. Following comprehensive comparative testing, X-Ray Transmission (XRT) ore sorting was selected as the preferred method for the Keliber Project due to its superior performance. Pilot-scale XRT tests on Syväjärvi ore samples demonstrated a promising sorting efficiency of 73%. There is a risk that some variability in efficiency may occur across the Syväjärvi deposit, and it is recommended that ore- sorting variability tests be conducted across the Syväjärvi deposit. Although it has been initially assumed that similar efficiency rates will apply to other deposits, additional pilot-scale and variability tests are recommended to confirm and potentially enhance performance across all ore types. This proactive approach will help ensure robust performance across the project. The test feed was created using an artificial blend of ore and waste rock, which may not perfectly represent run-of- mine material. Therefore, pilot ore-sorting tests using actual mined ore from all deposits are recommended to validate performance under real operating conditions. These efforts, combined with selective mining and stockpiling strategies, are expected to support consistent and efficient ore sorting throughout the project lifecycle. Flotation Flotation has been successfully tested at bench scale across most deposits and at pilot scale on samples from Länttä, Syväjärvi, and Rapasaari, providing a solid foundation of understanding for flotation parameters. Building on this, pilot- scale testing is recommended for the remaining primary ore bodies to further optimise flotation performance. Comprehensive ore variability tests have already been conducted on Rapasaari samples, representing four distinct mineralised material types. These tests highlighted the importance of understanding spatial variability. It is recommended that similar variability testing programmes will be implemented for all other deposits. This will support the development of robust geo-metallurgical models, ensuring efficient and predictable flotation outcomes across the entire project. Conversion Conversion testwork was carried out by FLSmidth at their facility in the USA. The concentrate was calcined using a two-stage cyclone preheater rotary kiln system with 2 hours residence time, and the burning zone solids temperature was generally maintained between 1,050–1,100°C. These conditions resulted in an overall average alpha-to-beta conversion level of approximately 97% as measured by the conventional sulphuric acid solubility method. Some variation in optimum operating conditions were observed for concentrates from the different orebodies. Hydrometallurgy Outotec and Keliber have been working on the hydrometallurgical process since 2002. In June 2018, Keliber completed an FS for a project to produce battery-grade lithium carbonate from spodumene-rich pegmatite. However, following further market studies, it was decided to consider the production of battery-grade lithium hydroxide monohydrate (LiOH·H2O). A series of tests were completed to determine the production parameters of lithium hydroxide from spodumene ore, including the proprietary Metso Outotec lithium hydroxide process at pilot scale. The key process stages for lithium hydroxide production are an initial alkaline leaching stage, using sodium carbonate and undertaken at high temperature and pressure, reacting the Li in the spodumene to solid-phase lithium carbonate, S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xviii followed by reacting this material with lime to produce lithium hydroxide in solution. The hydroxide is then crystallised out of solution following purification to form the final product. The soda leach developed by Outotec has been successfully demonstrated at pilot scale on Syväjärvi and Rapasaari beta-spodumene concentrate. SRK recommended that other concentrates should also be subjected to conversion and hydrometallurgical testing but highlighted that, as the spodumene pegmatites of the Kaustinen area are understood to resemble each other petrographically, mineralogically, and chemically, it is likely that their concentrates will perform similarly to those from Syväjärvi and Rapasaari. SRK recommended that the mineralogical and chemical similarity of other concentrates be assessed and that they be subjected to conversion and hydrometallurgical testing if significantly different to Syväjärvi or Rapasaari. Three different commercially available spodumene concentrates were tested in 2024 at the Metso Research Center after calcination at FLSmidt for hydrometallurgical testwork. The lithium concentrations of the different calcines were as follows: Sayona 2.50%, AMG 2.51%, and Sigma 2.54%. These corresponded with the average Li2O concentration of 5.4% for Sayona and AMG and 5.5% for Sigma. Based on the results of the mineralogical investigation as well as the batch leaching tests, calcination of the Sayona and AMG concentrates was successful, whereas there was potentially some over-calcination of the Sigma concentrate. After batch testwork to fix the parameters, the continuous LiOH pilot test was operated successfully. The main process stages in the pilot were soda leaching, cold conversion, secondary conversion, ion exchange, and LiOH crystallisation. The operation of the Metso LiOH process was demonstrated with selected feed materials with good results. Inclusion of the second LiOH crystallisation process for the Rapasaari trial proved that deleterious elements are removed to comply with the battery-grade specifications. Based on VBKOMS's review and assessment, there is sufficient evidence in the pilot plant results to give confidence that the design recovery figure can be achieved in practice, following a suitable ramp-up period. Although the final products from the pilot-scale tests did not meet the Lithium specification for the final product grade, this was due to insufficient drying of the final crystals. The drying technology being implemented is successfully operated in other plants and, therefore, not deemed a significant risk. The Keliber Project is likely to be the first implementation of the Metso Outotec soda pressure leaching technology. While the individual unit processes are not novel, and while the Syväjärvi (2020), Rapasaari (2022), and commercial concentrates (2024) pilot trials have significantly de-risked the flowsheet, a residual risk remains as it does with the first implementation of any novel technology. These risks were tested as sensitivities in the financial model, demonstrating that the risks are sufficiently low to warrant Mineral Reserve disclosure based on lithium hydroxide monohydrate production. In addition, Metso Outotec also provides a process guarantee, although such a guarantee does not ultimately guarantee that a process will work so much as it defines the extent of financial compensation that will apply should it not. Importantly, it should be noted that the Mineral Reserves for Keliber project have been declared on the basis that a ready market exists for battery-grade Lithium hydroxide monohydrate. Mineral Processing In February 2022, Keliber issued an FS and have undertaken engineering studies to produce 15,000 tpa of BG. Mined ore will be beneficiated at the Päiväneva concentrator located near the Rapasaari mine. Flotation concentrate will be transported to the Keliber lithium refinery where lithium hydroxide monohydrate will be produced as final product. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xix The selected overall flowsheet comprises a conventional spodumene concentrator which includes crushing, ore sorting, grinding, and spodumene recovery by flotation. Flotation concentrate is calcined to convert alpha-spodumene to beta-spodumene. The converted spodumene concentrate will be processed via the patented Metso Outotec soda pressure leach to produce lithium hydroxide monohydrate. The previously reported ramp-up of 27 months, which diverged from the approved FS ramp-up schedule of nine months, was based on an initial due diligence review completed early in 2022 by SSW. In parallel, detailed due diligence done by SRK (UK) during the same period suggested a six-month concentrator ramp-up and a nine-month refinery ramp-up to be reasonable. Since then, SSW have put in significant work to understand the asset better and have given diligent technical support to ensure a ramp-up more in line with the FS. A follow-up review was done by SRK (UK) in 2023 supporting an eight-month concentrator ramp-up and a nine-month refinery volume ramp-up. One major change since then is the fact that the refinery will not start up with third-party spodumene anymore but will operate from day one with ore originating from Keliber’s own operations. It must also be noted that pre-commissioning of equipment (standalone) and cold commissioning of processes in the refinery are planned to start during March 2025 and will continue until construction is completed by the end of 2025. Simultaneously, training of operators will continue. Although SSW acknowledge the risk of commissioning both plants near parallel with each other, additional allowances were made in the estimate to cater for separate commissioning teams. It must also be noted that detailed work around commissioning and ramp up timeframes and costs are in progress by the time of the release of this report and will continue as such for the remainder of 2025. This is a significant change from the previously reported ramp-up schedule and sensitivities in this regard to evaluate the robustness of the financial model relayed positive results. Modelled recovery parameters for each deposit are included in the Technical Economic Model. Sensitivities on the final product quality, ramp-up periods, and throughput have been run to evaluate the impact of said deviations on the financial viability of the Project. X. Infrastructure Construction of the Keliber lithium refinery commenced in 2023 and is nearing completion, with limited cold commissioning activities planned to start in March of 2025 continuing for the rest of the year when main process mechanical completion expected during Q4 2025. Hot commissioning is anticipated to commence during Q1 2026 and ramp up to start during Q2 2026. The infrastructure development of the Päiväneva concentrator plant started in October 2023 and main process mechanical completion is planned to be completed in December 2025. Production of the Syväjärvi mine is planned for Q3 2025. All relevant infrastructure between the two mining sites to haul the run of mine (ROM) to the concentrator plant will also be included in the scope of the mine development. The construction of the bulk infrastructure is planned to be completed by May 2025. Construction of the tailings storage facility (TSF) is planned to be completed in time for the hot commissioning of the concentrator plant. The current schedule is tight, and any delay with the development infrastructure and the TSF will cause a delay in the hot commissioning of the plants. XI. Permitting Requirements Keliber have obtained mining rights for Syväjärvi, including the Syväjärvi Auxiliary Area, Länttä, with an expiry date of 20/03/2027, and also for Rapasaari. In the end of 2024, Keliber is holding prospecting rights for twenty-seven S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xx exploration areas that are legally valid for a range of dates from 2025 to 2028. These include Emmes 1, Outovesi, Rapasaari, and Syväjärvi 3–4. All permits are in place to commence development and operations at Syväjärvi. Good progress is being made with obtaining outstanding necessary permissions for the Päiväneva concentrator and Rapasaari mining operations. The current permitting status is summarised in the table overleaf. Permitting situation of Keliber Project as of 31 December 2024. Production Site Permit Status Date Syväjärvi Mine Environmental and water permit Valid 20/02/2019 Exception permit to moor frogs Valid 01/02/2020 Exception permit for diving beetles Valid 21/07/2020 Mining permit Valid 13/12/2018 The right of use of the mining area Valid 09/08/2021 Mining Safety Permit Valid 13/10/2021 Rapasaari Mine Environmental and water permit Valid, partly returned for additional permitting 28/12/2022 Mining Permit Valid 23/03/2022 The right of use of the mining area Valid Mining Safety Permit Not started - Länttä Mine Environmental and water permit Permitting not started - Mining Permit Valid 23/05/2006 Mining Safety Permit Not started - Outovesi Mine Environmental and water permit Permitting not started - Mining Permit Permitting not started - Mining Safety Permit Permitting not started - Päiväneva Concentrator Environmental and water permit Valid, partly returned for additional permitting 28/12/2022 Mining Permit (included in Rapasaari mining area) Valid 23/03/2022 Building permit Valid Chemical permit In progress - Keliber Lithium Refinery Environmental and water permit Valid 28/06/2022 Building permit Approved 02/08/2024 Chemical Permit In progress - XII. Capital and Operating Costs Keliber presents capital expenditure (Capex) which includes the establishment of the open pits, the capital for the Päiväneva concentrator, and the Keliber lithium refinery. All data provided in this chapter was sourced from Keliber

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxi and the updated 26 January 2025 Technical Economic Model titled “Keliber_Economic_Model_v4.0_LoM2024_quick_update_20250126_Ore_Reserve_Final”. Capital is modelled as from 31 October 2024, and capital before this date is considered sunken cost and therefore not included in the NPV result. The Keliber FS was completed, approval for the implementation of the project was given on 3 October 2022. The Capital Cost Estimate and Operating Cost Estimate accuracy of ±15% is applied and overall Project contingency of ≤10% could be achieved. The mining operations will be contractor mining and detailed contractor costs were received in 2023. The capital costs and operating costs were received from the client and are at FS or higher level of accuracy. It should be noted, however, that estimation of capital and operating costs is inherently a forward-looking exercise. These estimates rely upon a range of assumptions and forecasts that are subject to change depending upon macro- economic conditions, operating strategy, and new data collected through future operations. Therefore, changes in forward-looking assumptions can result in capital and operating costs that deviate more than 15% from the costs forecast herein. Certain risks were identified during the review of the financial model and, as a result, sensitivities were modelled to understand the magnitude of the risks, as described in section 18.9. The table below is a high-level summary of the Capex for the Project. The total Capex for the Project amounts to EUR651m, as shown in the table below. Capex summary. Item Total Capex (MEUR) % Development Phase 0 Direct Costs 465 71 Syväjärvi Mine 18 3 Rapasaari Mine 68 10 Concentrator Plant (Päiväneva Site) 228 35 LiOH Plant, Kokkola Site 152 23 Indirect Costs 159 24 Engineering & Construction Services 41 6 Site Facilities During Construction 10 2 Construction Equipment 13 2 Spare Parts 26 4 Commissioning 1 0 Owners' Cost 18 3 Contingency 45 7 Grants + Adjustments 4 1 Plant-Related Costs 27 4 Concentrator Plant (Päiväneva Site) 5 1 LiOH Plant, Kokkola Site 22 3 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxii Item Total Capex (MEUR) % Total Capex 651 100 Initial Capex is scheduled to be spent from October 2024 to December 2026 with no historical Capex being included in the financial model, and Sustaining Capex is scheduled throughout the Project life period from October 2024 to June 2045. The total operating expenditure (Opex) for the Project amounts to EUR2,293m and includes variable costs, fixed costs, royalties, and other fees, as shown in the table below. Total Opex. Item Total (MEUR) % Mining Opex 515 22 Crushing & Sorting and Concentrator Opex 217 9 Conversion & Keliber lithium Refinery Opex 885 39 Other Variable Opex 101 4 Freight and Transportation 32 1 Fixed Costs 520 23 Royalties and Fees 23 1 Total Opex 2,293 100 XIII. Economic Analysis VBKOM received and reviewed the financial model, Keliber_Economic_Model_v4.0_LoM2024_ quick_update_20250126_Ore_Reserve_Final, from Keliber Lithium Mine. This model was interpreted to determine the economic viability of the Project when updating and declaring Mineral Reserves since previous declarations. The Discounted Cash Flow (DCF) method is used to calculate the Net Present Value (NPV) and Internal Rate of Return (IRR) and, as a result, the intrinsic value of the Keliber Lithium Mine in real terms. The NPV is derived post-royalties and -tax from pre-debt real cash flows after considering Opex, Capex for the mining operations, Concentrator, and Keliber lithium refinery and using forecast macro-economic parameters. Cash Flow Forecast The undiscounted free cash flow (FCF) is shown in the table below and figure overleaf. FCF summary. Item MEUR Revenue 4,796 Opex (2,293) Inventory Movement 0 EBITDA 2,503 Capex (651) S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxiii Item MEUR FCF before Tax 1,852 Corporate and Mineral Tax (345) FCF after Tax 1,507 The payback period of the Project in an Equity-funded scenario is 5.4 years after the start of production in May 2026, therefore, payback will be in October 2031, as shown in the figure overleaf. FCF Forecast FCF forecast. ( 400) ( 300) ( 200) ( 100) - 100 200 300 400 2 0 2 4 2 0 2 5 2 0 2 6 2 0 2 7 2 0 2 8 2 0 2 9 2 0 3 0 2 0 3 1 2 0 3 2 2 0 3 3 2 0 3 4 2 0 3 5 2 0 3 6 2 0 3 7 2 0 3 8 2 0 3 9 2 0 4 0 2 0 4 1 2 0 4 2 2 0 4 3 2 0 4 4 2 0 4 5 2 0 4 6 2 0 4 7 2 0 4 8 M EU R Cashflow Forecast Revenue OPEX Inventory Movement Working Capital CAPEX Corporate & Mineral Tax EBITDA Free Cash Flow after Tax S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxiv Cumulative FCF Forecast Cumulative FCF forecast. The NPV and IRR are derived post-royalties and -tax, from pre-debt real cash flows after considering operating costs, capital expenditures for the mining operations, concentrator, and Keliber lithium refinery, and using forecast macro- economic parameters. The NPV is calculated as EUR411.67 million and the IRR is 16.87% over the evaluation period, as shown in the table below. Key metrics. Key Metrics Unit Value NPV, 8% @31 Oct 2024 EURm 411.67 IRR % 16.87 Capital Payback Period Years 5.4 LOM Period Years 20 Evaluation Period Years 25 Total Mined Ore Tonnes 14,426,296 Total Product Sold Tonnes 291,403 Sensitivity Analysis The following risks were identified during the review of the financial model and, as a result, sensitivities were modelled to understand the magnitude of the risks: (1 000) ( 500) - 500 1 000 1 500 2 000 2 0 2 4 2 0 2 5 2 0 2 6 2 0 2 7 2 0 2 8 2 0 2 9 2 0 3 0 2 0 3 1 2 0 3 2 2 0 3 3 2 0 3 4 2 0 3 5 2 0 3 6 2 0 3 7 2 0 3 8 2 0 3 9 2 0 4 0 2 0 4 1 2 0 4 2 2 0 4 3 2 0 4 4 2 0 4 5 2 0 4 6 2 0 4 7 2 0 4 8 M EU R Cumulative Cashflow Forecast Cumulative Free Cash Flow after Tax Production Start May 2026 Payback Oct 2031

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxv • Plant ramp-up curve; • Prices; • Discount rate; • Exchange rate; and • Logistics cost. The sensitivity analysis results showed a positive NPV in most cases, thus enabling VBKOM to sign off on the Mineral Reserves. XIV. Key Risks The top risks of the Project, as identified by SSW, are listed below. • Analcime sand disposal, permitting needed for disposal area in case it cannot be disposed in harbour or any other permitted location; • Concentrator environmental permit not valid when the operations are planned to commence; • Construction of concentrator tailings facilities is not ready for the commissioning; • Delay in refinery commissioning and ramp-up; • Lithium hydroxide market price lower than expected; • Unfavourable USD/EUR exchange rate; and • Wastewater treatment planned at Päiväneva concentrator cannot reach permit limits for nitrogen. Additional risks, identified by VBKOM, are: • Variability in the feed size due to blasting and double handling could impact the overall efficiency of the ore sorters due to too many fines bypassing the sorters. Similarly, poor liberation of waste rock due to blasting could lead to misplacement of material and lower overall efficiencies on the sorters. Despite spodumene mineralisation being generally homogeneously distributed throughout most of the pegmatites, the contamination caused by the inclusion of host rock xenoliths and wall rock material with ore material will impact the metallurgical recovery of spodumene during flotation and metallurgical processing. The current mine schedule caters for selective mining supported by ore sorting to mitigate these risks and executing the plan is imperative to limit the impact of contamination on the recovery of spodumene. • The Keliber Project is likely to be the first implementation of the Metso Outotec lithium hydroxide flowsheet. While the individual unit processes are not novel, and while the Syväjärvi (2020) and Rapasaari (2022) pilot trials have significantly de-risked the flowsheet, a residual risk remains, as it does with the first example of any novel technology. • The orebody orientation and thickness require strict grade-control measures to ensure the dilution is kept to a minimum. This is planned for. • Restrictions on waste dumping with deleterious elements could impact the sequence of mining when ensuring the deleterious elements’ limits are adhered to. • A significant number of skilled personnel will be required to develop and work at the operations. Labour availability could impact planned production and build-ups. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxvi • The Syväjärvi waste rock dump (WRD) height requirement increased by 6 m, and this amendment to the permit is required. The Rapasaari WRD requires extension beyond the current mining area; approval of this is required to ensure sufficient space is available. • Hazardous waste material is defined as sulphur >1.0% and arsenic >100 ppm. In the previous study, the impact of arsenic was overlooked, but has since been addressed. For Syväjärvi, the storage requirement marginally increased, however, at Rapasaari, the storage requirement increased to above 8 MLCM. Areas were identified to host the hazardous material, with associated environmental controls. This requires amendment to permitting. XV. Conclusions and Recommendations Permits Development and operations at Syväjärvi are fully permitted, while processes for obtaining outstanding permissions for the Päiväneva concentrator and Rapasaari mining operations are well underway. Additional design work has been triggered by the finalisation of the Keliber lithium refinery environmental permit and issue of the Rapasaari and Päiväneva permit, as well as the implementation of compliance measures. The extractive waste management plan and closure plan must be reviewed to support implementation and to reflect design changes. Subject to specific technical issues associated with permit changes and potential delays caused by ongoing permitting processes (3 existing permit updates and one for the analcime sand storage area), no material issues linked to ESG have been identified. Studies and preparation for new or amendment applications as required must be done timeously and in accordance with legislation and regulations to avoid Project delays. Permitting processes will be further facilitated through continuous stakeholder and authority engagement. Geology and Mineral Resources ERM were not involved in any of the exploration conducted but conducted a site visit and reviewed the exploration completed to date and the supporting documentation provided by SSW. According to ERM, the exploration data that have been captured to date (consisting primarily of drilling data) are of sufficient quality to be used in Mineral Resource estimation and for the purposes used in this TRS. Overall, ERM consider the data used to prepare the geological models and MRE to be accurate and representative and have been generated with industry-accepted standards and procedures. ERM consider the MRE to be representative of the informing data and that the data are of sufficient quality to support the MRE for each of the deposits classified into the Measured Mineral Resources, Indicated Mineral Resources, and Inferred Mineral Resources categories. ERM note that there is room for improvement in the exploration process such as implementation of a fit-for-purpose relational database with timely backups, which will ensure a robust and secure database going forward and relevant workflows. Investigation into the use of hyperspectral core scanning to aid geological logging and material characterisation (from a geological, processing, and environmental perspective) should also be considered. ERM also recommend some improvements to the quality assurance and quality control (QAQC) protocols, and these include resolving the apparent under-reporting of Keliber’s reference materials through additional certification, inclusion of additional certified reference materials across a broader lithium grade range, and more frequent check lab assays. Other considerations include the sampling and assay protocols for sampling of pegmatite and host rock to compile a robust dataset of deleterious elements from an environmental and processing perspective. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxvii Mining and Mineral Reserves The Mineral Resource model estimation methodology was altered from the previous study, whereby internal dilution was included in the estimation process, resulting in more defined lithium grades. The resource models were regularised to 5 m x 5 m x 2.5 m, incorporating dilution and losses. The modifying factors for both Syväjärvi and Rapasaari were investigated in depth during this study. The dilution and losses from the regularisation process resulted in 11.08% dilution and 14.97% losses for Syväjärvi and 11.66% dilution and 15.67% losses for Rapasaari. The modifying factors also include perimeter constraints whereby the extent of the pits was limited in their extents. Marginal COGs were determined as 0.188% Li₂O for Syväjärvi and 0.3% Li₂O for Rapasaari. The ore definition was altered, taking the marginal COGs into account. The pit optimisation for both Syväjärvi and Rapasaari was completed based on a constrained scenario, hence, the shell selection was limited. The 2024 geotechnical study presents significant advancements in geotechnical engineering compared to previous assessments. A major improvement involved new core logging conducted within ±50 metres of pit slopes, which provided better modelling of FLT and shear zones. This resulted in the geotechnical parameters complying with a feasibility level of accuracy. The geotechnical study resulted in an increase in slope angles for both Syväjärvi and Rapasaari. The Syväjärvi pit was designed as a single large pit; in contrast, Rapasaari’s, being much larger, was designed to incorporate three pushbacks which enables phased ore extraction to maintain steady production rates with mostly constant stripping ratios. The combined LOM production for Syväjärvi and Rapasaari spans from 2025 to 2044, with plant feed ending in April 2045. Syväjärvi is expected to produce 3.15 Mt of ore at an average Li₂O grade of 1.03%, while Rapasaari will yield 11.28 Mt of ore at 0.93% Li₂O. The stripping ratio for Syväjärvi is 5.14, whereas Rapasaari’s overall stripping ratio is 7.63. The Syväjärvi WRD elevation was lifted by 6 m due to the increase in tonnage from the FS work. The OVB dump elevations were also increased to ensure sufficient capacity. The Rapasaari WRD requires additional space due to the increased waste stripping. No backfill was included in the Project. The primary loading units will include backhoe excavators fitted with 4.6 m³ buckets for ore and 5.8 m³ buckets for waste, while haulage will be managed using 64-tonne rear-dump trucks for waste and 41-tonne articulated trucks for ore. No Inferred Mineral Resources were included, only Measured Mineral Resources were converted to Proven Mineral Reserves, and only Indicated Mineral Resources were converted to Probable Mineral Reserves. The Rapasaari pushback 2 design requires adjustment due to instances in the production schedule where only one access is available. It is recommended to optimise the pushback selection of Rapasaari. The orebody orientation and thickness require strict grade-control measures to ensure the dilution is kept to a minimum. The Syväjärvi WRD height requirement increased by 6 m, and this amendment to the permit is required. The Rapasaari WRD requires extension beyond the current mining area; approval of this is required to ensure sufficient space is available. Similarly, amendments to the OVB dumps are also required. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxviii The increase in hazardous waste to 8 MLCM for Rapasaari requires an environmental impact assessment (EIA) and application for an EP. Processing Although the testwork on the ore sorters proved positive, the feed to the ore-sorting test equipment comprised only of artificial blends of ore and waste rock. There is a risk that performance on ROM ore may be less efficient than on the artificial composite ore feed due to particles not being liberated. The final concentrate quality and Li recovery are highly dependent on the feed grade and ROM particle size distributions (PSDs’). As part of the mining schedule, provision was made for various stockpiles to manage the feed grade. However, the impact of rehandling of the material on the PSD of the plant feed, which has a direct impact on the bypass stream and milling plant feed qualities, has not been evaluated. The principal processing risk lies with the implementation of the Metso-Outotec hydrometallurgical process. The Keliber Project is likely to be the first implementation of this specific flowsheet. While the individual unit processes are not novel, and while the 2020 and to a slightly lesser extent the 2023 pilot trials have significantly de-risked the flowsheet, a residual risk remains, as it does with the first example of any novel technology. The ramp-up schedule (3 months hot commissioning followed by 9 months ramp-up) is in line with the 2022 FS study ramp-up for internal Keliber ore, bar the fact that the refinery will not start up with third party spodumene anymore (previously planned for up to 12 months) but will operate from day one with ore originating from Keliber’s own operations. The 12-month period with third party concentrate would have allowed time to address all the operating issues, usually experienced during the start-up of high-pressure leaching plants. Sibanye-Stillwater acknowledge the risk of commissioning both plants near parallel with each other and additional allowances were made in the estimate to cater for separate commissioning teams. This is a significant change from the previously reported ramp-up schedule and sensitivities in this regard were tested to evaluate the robustness the financial model and relayed positive results. Given the short periods allocated for construction of the flotation tailings pond, there is no flexibility to accommodate any delays in construction of this facility. To mitigate this, a contractor with a proven track record has been appointed by Keliber. Acceptance has been received to use the peat liner solution for the flotation tailings facility. Detailed planning is required to manage the ramp-up of both plants simultaneously. Training of key personnel before hot commissioning is key to ensure the successful execution of the ramp-up plan. Control of feed grade and feed PSDs is critical to ensure final product qualities. Additional testwork to understand the impact of Rapasaari ore on the concentrate recoveries is recommended. Economic Analysis VBKOM reviewed the financial model developed by Keliber and are satisfied with the positive NPV of EUR411.67 million, IRR of 16.87%, and Payback Period of 5.4 Years. Total Capex amounts to EUR651 million with Opex amounting to a cumulative EUR2,293 million over the LOM period. Capital is modelled as from 31 October 2024, and capital before this date is considered sunken cost and therefore not included in the NPV result. Sensitivities were evaluated to test the robustness of the Project’s economic viability, given technical risks and other economic drivers. Most sensitivities resulted in a positive NPV, which shows a robust business case for the Project. It is recommended that the ramp-up schedules of the two plants and the Rapasaari recovery be critically reviewed.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxix TABLE OF CONTENTS DATE AND SIGNATURE PAGE ................................................................................................................................ II IMPORTANT NOTICES .......................................................................................................................................... III CO-ORDINATING AUTHORS ................................................................................................................................. IV DISCLAIMER ......................................................................................................................................................... V EXECUTIVE SUMMARY ........................................................................................................................................ VI I. Introduction ..................................................................................................................................................... vi II. Effective Date .................................................................................................................................................. vi III. Property Description and Ownership ............................................................................................................ vi IV. History ........................................................................................................................................................... vii V. Geology and Mineralisation .......................................................................................................................... viii VI. Status of Exploration, Development, and Operations ................................................................................. viii VII. Mineral Resource Estimates ......................................................................................................................... ix VIII. Mineral Reserve Estimates .......................................................................................................................... xi IX. Metallurgy and Mineral Processing .............................................................................................................. xvi X. Infrastructure ................................................................................................................................................. xix XI. Permitting Requirements .............................................................................................................................. xix XII. Capital and Operating Costs.......................................................................................................................... xx XIII. Economic Analysis ...................................................................................................................................... xxii XIV. Key Risks .................................................................................................................................................... xxv XV. Conclusions and Recommendations .......................................................................................................... xxvi TABLE OF CONTENTS ....................................................................................................................................... XXIX LIST OF FIGURES ............................................................................................................................................. XXXV LIST OF TABLES ................................................................................................................................................ XLII LIST OF EQUATIONS ........................................................................................................................................ XLVI LIST OF APPENDICES ........................................................................................................................................ XLVI 1 INTRODUCTION ....................................................................................................................................... 1 1.1 Registrant ................................................................................................................................................ 1 1.2 Terms of Reference and Purpose of the Report ..................................................................................... 2 1.3 Qualified Persons’ Qualifications and Site Visit ...................................................................................... 2 1.4 Effective Date .......................................................................................................................................... 6 1.5 Sources of Information ........................................................................................................................... 6 1.6 List of Units, Abbreviations & Acronyms, and Chemical Formulae ........................................................ 7 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxx 2 PROPERTY DESCRIPTION ........................................................................................................................ 17 2.1 Location and Area of the Property........................................................................................................ 17 2.2 Ownership ............................................................................................................................................. 19 2.3 Finland Regulatory Environment .......................................................................................................... 20 2.4 Mineral Deposit Tenure ........................................................................................................................ 27 2.5 Surface Rights ....................................................................................................................................... 30 2.6 Environmental Authorisations .............................................................................................................. 30 2.7 Property Encumbrances and Permitting Requirements ....................................................................... 31 2.8 Significant Factors and Risks Affecting Access, Title ............................................................................. 32 3 ACCESSIBILITY, CLIMATE, LOCAL RESOURCES, INFRASTRUCTURE, AND PHYSIOGRAPHY ........................... 34 3.1 Topography, Elevation, and Vegetation ............................................................................................... 34 3.2 Access to the Properties ....................................................................................................................... 34 3.3 Climate and Length of Operating Season ............................................................................................. 35 3.4 Local Resources and Infrastructure ...................................................................................................... 35 4 HISTORY ................................................................................................................................................ 37 4.1 Previous Operations, Operators ........................................................................................................... 37 4.2 Exploration and Development Work .................................................................................................... 37 5 GEOLOGICAL SETTING, MINERALISATION, AND DEPOSIT ......................................................................... 39 5.1 Regional, Local, and Project Geology .................................................................................................... 39 5.2 Internal Pegmatite Zonation and Mineralogy ....................................................................................... 50 5.3 Weathering ........................................................................................................................................... 51 5.4 Mineralisation Style and Deposit Type – LCT Pegmatites ..................................................................... 51 5.5 General Lithium Mineral Processing Considerations for Hard Rock Deposits ...................................... 54 5.6 Mineral Concentrates ........................................................................................................................... 55 6 EXPLORATION ........................................................................................................................................ 57 6.1 Non-Invasive Exploration Activities ...................................................................................................... 57 6.2 Drilling, Logging, and Sampling ............................................................................................................. 59 6.3 Geotechnical and Hydrogeological Drilling ........................................................................................... 67 6.4 QP’s Opinion on the Exploration........................................................................................................... 67 7 SAMPLE PREPARATION, ANALYSES, AND SECURITY ................................................................................. 68 7.1 Sample Preparation Methods and Quality Control Measures .............................................................. 68 7.2 Sample Preparation, Assaying, and Laboratory Procedures ................................................................. 68 7.3 QAQC Measures .................................................................................................................................... 69 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxi 7.4 Adequacy of Sample Preparation, Security, and Analytical Procedures .............................................. 77 7.5 Unconventional Analytical Procedures ................................................................................................. 78 8 DATA VERIFICATION ............................................................................................................................... 79 8.1 Data Verification Procedures ................................................................................................................ 79 8.2 Site Visit ................................................................................................................................................ 79 8.3 Check Logging, Database Verification, and Validation ......................................................................... 86 8.4 QP Opinion and Recommendations ...................................................................................................... 93 9 MINERAL PROCESSING AND METALLURGICAL TESTING ........................................................................... 94 9.1 Nature and Extent of Testing and Analytical Procedures ..................................................................... 94 9.2 Recovery Dependencies in Mineral Processing of Syväjärvi, Rapasaari, and Länttä .......................... 126 9.3 Adequacy of Data ................................................................................................................................ 132 9.4 Comment ............................................................................................................................................ 136 10 MINERAL RESOURCE ESTIMATES .......................................................................................................... 139 10.1 Introduction ........................................................................................................................................ 139 10.2 Database ............................................................................................................................................. 139 10.3 Database Validation ............................................................................................................................ 140 10.4 Topography ......................................................................................................................................... 140 10.5 Geological Interpretation .................................................................................................................... 140 10.6 Geological Modelling .......................................................................................................................... 141 10.7 Compositing ........................................................................................................................................ 148 10.8 Exploratory Data Analysis ................................................................................................................... 148 10.9 Top Caps .............................................................................................................................................. 148 10.10 Variography ......................................................................................................................................... 148 10.11 Block Model ........................................................................................................................................ 152 10.12 Grade Estimation ................................................................................................................................ 152 10.13 Validation ............................................................................................................................................ 155 10.14 Density ................................................................................................................................................ 158 10.15 Mineral Resource Classification .......................................................................................................... 160 10.16 Reasonable Prospects for Economic Extraction ................................................................................. 162 10.17 Mineral Resource Statement .............................................................................................................. 163 10.18 Comparison with the Previous MRE ................................................................................................... 168 10.19 Risks .................................................................................................................................................... 172 11 MINERAL RESERVE ESTIMATES ............................................................................................................. 173 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxii 11.1 Key Assumptions, Parameters, and Method for the Estimation of Mineral Reserves ....................... 173 11.2 Open-Pit Optimisation ........................................................................................................................ 181 11.3 Open-Pit Design .................................................................................................................................. 189 11.4 Mineral Reserve Classification ............................................................................................................ 197 11.5 Mineral Reserve Estimates ................................................................................................................. 197 12 MINING METHODS ............................................................................................................................... 215 12.1 Rock Engineering – Syväjärvi .............................................................................................................. 215 12.2 Rock Engineering – Rapasaari ............................................................................................................. 238 12.3 Hydrogeology and Hydrology ............................................................................................................. 265 12.4 LOM Production Schedule .................................................................................................................. 267 13 PROCESSING AND RECOVERY METHODS ............................................................................................... 297 13.1 Concentrator ....................................................................................................................................... 297 13.2 Lithium Hydroxide Refinery ................................................................................................................ 305 13.3 Plant Commissioning and Ramp-Up ................................................................................................... 311 14 PROJECT INFRASTRUCTURE .................................................................................................................. 313 14.1 Mine Layout and Operations .............................................................................................................. 313 14.2 Surface Infrastructure and Bulk Services ............................................................................................ 317 14.3 Plant Infrastructure ............................................................................................................................. 319 14.4 Tailings Storage Facility ....................................................................................................................... 323 15 MARKET STUDIES ................................................................................................................................. 325 15.1 Uses of Lithium Hydroxide Monohydrate ........................................................................................... 325 15.2 Market Overview ................................................................................................................................ 325 15.3 Market Entry Strategy and Product Specifications ............................................................................. 338 15.4 Material Contracts .............................................................................................................................. 341 15.5 Commodity Market Assessment ......................................................................................................... 342 15.6 Prices ................................................................................................................................................... 344 16 ENVIRONMENTAL STUDIES, PERMITTING AND PLANS, NEGOTIATIONS OR AGREEMENTS WITH LOCAL INDIVIDUALS OR GROUPS .................................................................................................................... 347 16.1 Relevant Environmental Issues and Results of Studies Conducted .................................................... 347 16.2 Water Management ............................................................................................................................ 350 16.3 Potentially Sulphate Soils .................................................................................................................... 353 16.4 Acid-Producing Waste Rock ................................................................................................................ 353 16.5 Waste Disposal .................................................................................................................................... 353

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxiii 16.6 Environmental Site Monitoring ........................................................................................................... 354 16.7 Social and Community Aspects ........................................................................................................... 354 16.8 Closure Aspects ................................................................................................................................... 355 16.9 Environmental, Social, and Governance Summary ............................................................................. 356 17 CAPITAL AND OPERATING COSTS .......................................................................................................... 357 17.1 Capital Costs ........................................................................................................................................ 357 17.2 Operating Cost .................................................................................................................................... 358 17.3 Financial Costs Indicators .................................................................................................................... 361 17.4 Accuracy of Estimates ......................................................................................................................... 365 18 ECONOMIC ANALYSIS ........................................................................................................................... 367 18.1 Principal Assumptions ......................................................................................................................... 367 18.2 Macro-Economic Forecasts ................................................................................................................. 368 18.3 Working Capital ................................................................................................................................... 369 18.4 Recoveries ........................................................................................................................................... 369 18.5 Discount Rate ...................................................................................................................................... 370 18.6 Cash Flow Forecast ............................................................................................................................. 370 18.7 Net Present Value ............................................................................................................................... 372 18.8 Regulatory Items ................................................................................................................................. 372 18.9 Sensitivity Analysis .............................................................................................................................. 373 18.10 Economic Analysis Conclusions ........................................................................................................... 380 19 ADJACENT PROPERTIES ........................................................................................................................ 381 20 OTHER RELEVANT DATA AND INFORMATION ........................................................................................ 382 20.1 Project Implementation ...................................................................................................................... 382 20.2 Exploration Programme and Budget .................................................................................................. 382 20.3 Risk Review ......................................................................................................................................... 383 21 INTERPRETATION AND CONCLUSIONS .................................................................................................. 391 21.10 Permitting ........................................................................................................................................... 391 21.11 Geology and Mineral Resources ......................................................................................................... 391 21.12 Processing ........................................................................................................................................... 393 21.13 Mining and Mineral Reserves ............................................................................................................. 393 21.14 Economic Analysis ............................................................................................................................... 394 22 RECOMMENDATIONS ........................................................................................................................... 395 22.1 Permitting ........................................................................................................................................... 395 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxiv 22.2 Geology and Mineral Resources ......................................................................................................... 395 22.3 Processing ........................................................................................................................................... 396 22.4 Mining and Mineral Reserves ............................................................................................................. 396 22.5 Economic Analysis ............................................................................................................................... 396 23 RELIANCE ON INFORMATION PROVIDED BY REGISTRANT ...................................................................... 397 24 DATE AND SIGNATURE PAGE ................................................................................................................ 398 24.1 Resources ............................................................................................................................................ 398 24.2 Reserves .............................................................................................................................................. 399 REFERENCES .................................................................................................................................................... 400 APPENDICES .................................................................................................................................................... 404 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxv LIST OF FIGURES Figure 1-1: Location of the Keliber offices and the Syväjärvi pilot mining area. .............................................................. 4 Figure 1-2: Keliber mining areas visited. ........................................................................................................................... 5 Figure 1-3: Keliber processing areas visited. ..................................................................................................................... 6 Figure 2-1: Locality plan. ................................................................................................................................................. 18 Figure 2-2: Mine and exploration sites. .......................................................................................................................... 19 Figure 2-3: Group ownership structure. ......................................................................................................................... 20 Figure 2-4: Simplified permitting process for mining projects in Finland. ...................................................................... 26 Figure 2-5: Permit areas. ................................................................................................................................................. 29 Figure 5-1: Regional geological map of the Kaustinen Lithium Province within the Pohjanmaa Belt (source: SRK, 2023 – modified after Ahtola et al., 2015). .............................................................................................................................. 40 Figure 5-2: Map showing the location of the various pegmatite groups within the broader Pohjanmaa Belt. Note: The yellow box around the Kaustenin Province in the north is host to the albite-spodumene pegmatites that form part of the Keliber Project. The complex pegmatite group (red) may also host lithium mineralisation (source: Alviola et al., 2001). .............................................................................................................................................................................. 41 Figure 5-3: Geology of the Keliber Project showing the map of pegmatite deposits (source: SRK, 2023 (modified after Ahtola et al., 2015)). ....................................................................................................................................................... 42 Figure 5-4: Syväjärvi – 3D view of modelled pegmatites looking southwest (source: ERM). ......................................... 43 Figure 5-5: Rapasaari – 3D view of modelled pegmatites looking northwest (source: ERM). ....................................... 44 Figure 5-6: Länttä – 3D view of modelled pegmatites looking northeast (source: ERM). .............................................. 45 Figure 5-7: Emmes – 3D view of modelled pegmatite looking north-northwest. Note the reverse FLT displacing the pegmatite (source: ERM). ............................................................................................................................................... 46 Figure 5-8: Outovesi – 3D view of modelled pegmatite looking north-northwest (source: ERM). ................................ 47 Figure 5-9: Leviäkangas – 3D view of modelled pegmatite looking north-northwest (source: ERM). ........................... 48 Figure 5-10: Tuoreetsaaret – plan view of modelled pegmatites (source: ERM). .......................................................... 49 Figure 5-11: Tuoreetsaaret – 3D view of modelled pegmatites looking south (source: ERM). ...................................... 50 Figure 5-12: Example of weathered pegmatite from Rapasaari (hole RA14; box 1 (~11–15 m depth)). Although the core is broken, the spodumene looks largely unaltered and lithium grades through this zone (samples 40582 to 40584) range from 0.52–0.86% Li (or 1.13–1.86% Li2O) and average 0.64% Li (1.38% Li2O). .................................................... 51 Figure 5-13: Idealised schematic model in profile or plan showing the regional zonation in a pegmatite field around a parental granite intrusion. Note: The rare-element suites of the most enriched pegmatites in each zone are indicated with the most prospective pegmatites located in distal areas compared to the parental granite (source: London, 2016). ........................................................................................................................................................................................ 53 Figure 5-14: Sketches showing the shapes of (A) a vertical en Echelon series of intrusions which are joined at depth (Fossen, 2010) and (B) a more shallowly dipping series of veins exposed at surface, with blind intrusions at depth (source: unknown). ......................................................................................................................................................... 54 Figure 5-15: Spodumene-quartz intergrowth seen in thin section (source: Scogings et al., 2016). ............................... 55 Figure 6-1: Geological map showing distribution of mapped spodumene pegmatite boulders in relation to pegmatites (source: SRK, 2023 (Keliber)). .......................................................................................................................................... 57 Figure 6-2: Regional distribution of Li in till in relation to known lithium deposits (source: Ahtola et al, 2015). .......... 58 Figure 6-3: Map showing historical, GTK, and Keliber drilling at Syväjärvi (source: Keliber, 2023). .............................. 60 Figure 6-4: Map showing GTK and Keliber drilling at Rapasaari (source: Keliber, 2023)................................................ 61 Figure 6-5: Map showing historical, GTK, and Keliber drilling at Länttä (source: Keliber, 2023). .................................. 62 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxvi Figure 6-6: Map showing historical and Keliber drilling at Emmes (source: Keliber, 2023). .......................................... 63 Figure 6-7: Map showing Keliber’s drilling at Outovesi (source: Keliber, 2023). ............................................................ 64 Figure 6-8: Map showing GTK and Keliber drilling at Tuoreetsaaret (source: Keliber, 2023). ....................................... 65 Figure 6-9: Map showing historical, GTK, and Keliber drilling at Leviäkangas (source: Keliber, 2023). ......................... 66 Figure 7-1: Reference material control charts from 2010 to 2020 in analytical order for the Keliber reference materials (source: Keliber, 2023). ................................................................................................................................................... 70 Figure 7-2: Observations of AMIS0355 values since its introduction to laboratory internal QC protocol in 2016. Blue dashed line is uncertified value for fusion method and green dashed line is certified value for 4-acid digestion (source: Keliber, 2023). ................................................................................................................................................................. 71 Figure 7-3: Summary of core replicate results for the period 2010–2023 using fusion method 720P (source: Keliber, 2023). .............................................................................................................................................................................. 73 Figure 7-4: Summary of laboratory pulp duplicate pairs for the period 2010–2023 using fusion method 720P (source: Keliber, 2023). ................................................................................................................................................................. 73 Figure 7-5: Absolute value of relative difference between pulp re-assays and reference sample vs Li% for the period 2010–2023 using fusion method 720P (source: Keliber, 2023). ..................................................................................... 74 Figure 7-6: Inter-laboratory checks conducted in 2014 by Labtium (fusion method 720P) vs ALS (4-acid method) (source: Sandberg, 2014). ............................................................................................................................................... 75 Figure 7-7: (A) Plot of 2022 inter-laboratory check – Kuopio (blue) and Oulu (orange). (B) Normal distribution of 2022 inter-laboratory check showing relative difference of paired samples. ......................................................................... 76 Figure 8-1: (A) Keliber’s core processing facility at Kaustinen; (B) Angled core racks used for core processing and logging. ............................................................................................................................................................................ 80 Figure 8-2: Keliber’s core receipt and storage facility adjoining the processing facility. (A) Stacked core boxes, and (B) sealed crates of coarse and pulp rejects received back from the laboratory. ................................................................ 81 Figure 8-3: (A) Core saw, and (B) cut sample in baskets prior to density measurements and packing. ........................ 81 Figure 8-4: Photo of drill hole S76 (Syväjärvi), checked in July 2023. ............................................................................. 82 Figure 8-5: Host rock outcrops from Syväjärvi. (A) Plagioclase-bearing porphyrite (metavolcanic) (WPT840), and (B) sulphide-bearing mica schist (metasediment) (WPT844). .............................................................................................. 85 Figure 8-6: Photo looking east of host schists and thin northerly-dipping pegmatite veins in hanging wall to main spodumene pegmatites exposed in water-filled pit at entrance to the portal at Syväjärvi. .......................................... 85 Figure 8-7: Tuoreetsaaret: (A) Erratic of spodumene-bearing pegmatite in forest with (B) large spodumene lathes (>20 cm long) showing uniform crystal orientation (interpreted to be perpendicular to host rock contacts) (WPT848). .... 86 Figure 8-8: Hole RA-14 (Box 32) with an interval of unsampled pegmatite logged as SPG and host rock. .................... 87 Figure 8-9: Hole S-22 (Box 19) showing samples 30371 (60.3–61.2 m) logged as muscovite pegmatite and 30372 (61.2– 62.5 m) logged as SPG with high lithium content. .......................................................................................................... 87 Figure 8-10: Comparison of recent (blue) and historical (green) Li2O% assay data within the interpreted spodumene pegmatite zones at Rapasaari (top), Syväjärvi (middle), and Emmes (bottom). ............................................................ 92 Figure 9-1: Spodumene pegmatite at the end of the tunnel (top) and numbered ore piles before transport to GTK Mintek (bottom). ............................................................................................................................................................ 97 Figure 9-2: Syväjärvi pilot sample location – plan (top) and long section (bottom) views. ............................................ 98 Figure 9-3: Lithium recovery as a function of feed grade (source: Keliber 2019 and 2022 FS Reports). ..................... 102 Figure 9-4: Lithium recovery as a function of concentrate grade (source: Keliber 2019 and 2022 FS Reports). ......... 103 Figure 9-5: Variability in Rapasaari flotation recovery (source: Keliber 2022 FS). ....................................................... 106 Figure 9-6: Variability in Outovesi flotation recovery (source: Keliber 2022 FS). ......................................................... 107 Figure 9-7: Syväjärvi pilot tests 2019. ........................................................................................................................... 109

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxvii Figure 9-8: Sixty-five (65)-litre autoclave used in the semi-continuous pilot processing. ............................................ 120 Figure 9-9: Simplified process flowsheet of LiOH*H2O production. ............................................................................. 122 Figure 9-10: Lithium recovery at 4.5% Li2O in the concentrate vs lithium grade in the feed. ...................................... 127 Figure 9-11: Grade recovery curves of the geo-metallurgical dilution study. .............................................................. 128 Figure 9-12: Recovery at 4.5% Li2O against MgO% of the feed sample........................................................................ 130 Figure 9-13: Fitted lines for lithium recovery into the spodumene concentrate vs wall rock dilution. ....................... 130 Figure 10-1: Oblique views of the modelled pegmatites for Emmes, Länttä, Leviäkangas, and Outovesi (INT = internal xenolith; MPEG = muscovite pegmatite; SPEG = spodumene pegmatite) (after ERM 2024). ...................................... 142 Figure 10-2: Plan view of the modelled pegmatites at Syväjärvi (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) (oblique view) (after ERM 2024). .................................................................................. 143 Figure 10-3: View looking west showing the modelled pegmatites at Syväjärvi (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) and topography (green) (oblique view) (after ERM 2024). ...... 144 Figure 10-4: Plan view of the modelled pegmatites at Tuoreetsaaret (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) (oblique view) (after ERM 2024). ............................................................... 145 Figure 10-5: View looking west showing the modelled pegmatites at Tuoreetsaaret (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) and topography (green) (oblique view) (after ERM 2024). ...... 145 Figure 10-6: Plan view showing pegmatites at Rapasaari from numeric modelling (left-hand side) and pegmatite zones (right-hand side) (INT = internal xenolith; MPEG = muscovite pegmatite; SPEG = spodumene pegmatite); cross-section location in green (oblique view) (after ERM 2024). ...................................................................................................... 147 Figure 10-7: Cross-section looking north showing pegmatite zones at Rapasaari relative to drill holes (INT = internal xenolith; MPEG = muscovite pegmatite; SPEG = spodumene pegmatite) (oblique view) (after ERM 2024). .............. 147 Figure 10-8: Variogram models for lithium oxide at Rapasaari (after ERM 2024)........................................................ 150 Figure 10-9: Variogram models for lithium oxide at Syväjärvi (after ERM 2024). ........................................................ 151 Figure 10-10: Swath plots for lithium oxide at Rapasaari; composites as orange line; block estimates as black line (after ERM 2024). .................................................................................................................................................................... 156 Figure 10-11: Swath plots for lithium oxide at Syväjärvi; composites as orange line; block estimates as black line (after ERM 2024). .................................................................................................................................................................... 157 Figure 10-12: Scatterplot of Li2O grade vs SG at Rapasaari (bivariate fit of SG by Li2O zone = 1) (spodumene pegmatite) (after ERM 2024). .......................................................................................................................................................... 158 Figure 10-13: Regression of Li2O grade bins vs average SG at Rapasaari (bivariate fit of mean SG by Li2O) (after ERM 2024). ............................................................................................................................................................................ 159 Figure 10-14: Mineral Resource classification at Rapasaari with drill hole collar locations (after ERM 2024). ........... 161 Figure 10-15: Mineral Resource classification at Syväjärvi with drill hole collar locations (after ERM 2024). ............. 162 Figure 11-1: Syväjärvi pit constraint outline. ................................................................................................................ 175 Figure 11-2: Rapasaari pit constraint outline. ............................................................................................................... 175 Figure 11-3: Histogram showing the global resource dilution at Syväjärvi (left) and Rapasaari (right). ...................... 177 Figure 11-4: Syväjärvi recovery curve. .......................................................................................................................... 178 Figure 11-5: Rapasaari recovery curve. ......................................................................................................................... 179 Figure 11-6: Geotechnical slope sectors for Syväjärvi (source: Geotec Africa CC). ...................................................... 183 Figure 11-7: Geotechnical slope sectors for Rapasaari (source: Geotec Africa CC). .................................................... 184 Figure 11-8: Pit-by-pit graph for the Syväjärvi optimisation......................................................................................... 187 Figure 11-9: Comparison between the 2023 FS pit design and the selected 2024 pit shell. ........................................ 187 Figure 11-10: Pit-by-pit graph for the Rapasaari optimisation. .................................................................................... 188 Figure 11-11: Comparison between the 2022 FS pit design and the selected 2024 pit shell. ...................................... 189 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxviii Figure 11-12: Standard dual- and single-lane ramp configurations. ............................................................................ 191 Figure 11-13: Syväjärvi open-pit design. ....................................................................................................................... 194 Figure 11-14: Rapasaari ultimate open-pit design. ....................................................................................................... 195 Figure 11-15: Rapasaari pushback design. .................................................................................................................... 196 Figure 11-16: Syväjärvi in situ Mineral Resources. ....................................................................................................... 198 Figure 11-17: Syväjärvi in situ Mineral Resources inclusive and exclusive of LOM plan. ............................................. 200 Figure 11-18: In situ Mineral Resources to Mineral Reserves for Syväjärvi. ................................................................ 203 Figure 11-19: Rapasaari in situ Mineral Resources. ...................................................................................................... 204 Figure 11-20: Rapasaari in situ Mineral Resources inclusive and exclusive of LOM plan. ............................................ 206 Figure 11-21: In situ Mineral Resources to Mineral Reserves for Rapasaari. ............................................................... 208 Figure 11-22: Syväjärvi Mineral Reserves reconciliation: 2024 vs 2023. ...................................................................... 211 Figure 11-23: Rapasaari Mineral Reserves reconciliation: 2024 vs 2023. .................................................................... 212 Figure 12-1: The selected sections of slope-proximate boreholes of which the geotechnical data, in particular the discontinuity orientation logs, were used in the 2024 upgrade study (oblique view). The “boreholes count” and “total logged metres” refer to the original larger database. .................................................................................................. 217 Figure 12-2: The spatial distribution of the ten drill holes from which cores were selected for relogging shown in relation to the June 2019 proposed Syväjärvi pit design (oblique view). ..................................................................... 218 Figure 12-3: Lithologies cutting the June 2019 design Syväjärvi open-pit slopes. Light green = metavolcanic rock; orange = spodumene pegmatite; pink = plagioclase porphyrite; dark green = sulphidic schist; blue = mica schist. Plan view. ...................................................................................................................................................................................... 221 Figure 12-4: Four continuous broken rock zones interpreted or inferred as FLTs shown in the June 2019 proposed Syväjärvi pit. View is to the west (direction 295°). ....................................................................................................... 222 Figure 12-5: Syväjärvi drill-hole-based GSI values of the main rock types (sill = plagioclase porphyrite). The mean values are indicated by the red diamonds. The median values are indicated by the lines that cross inside of the coloured boxes. The box encloses the interquartile range around the median. The whiskers extend out to vertical lines that mark the extreme value extents. ........................................................................................................................................... 226 Figure 12-6: Hydraulic conductivity, K, in m/s, measured in monitoring bores in Syväjärvi rocks in Q1 of 2021 (Figure 18-83 copied from the 2022 draft DFS Report, Volume 4, Chapter 18). ...................................................................... 228 Figure 12-7: Pseudo-PoFs from (2nd iteration) stereographic analyses of kinematic failure mechanisms for BFAs from 60° to 90°. ..................................................................................................................................................................... 231 Figure 12-8: The sensitivity of the pseudo-PoF for planar sliding to BFA in SDS 1A, all other factors at mean (expected) values. BFA ≤ 65° is recommended since this is a (secondary) ramp slop and considering the relatively wide variation in FO dip angle. ............................................................................................................................................................. 232 Figure 12-9: The sensitivity of the pseudo-POF for wedge sliding to BFA in SDS 1A, all other factors at mean (expected) values. BFA ≤ 65° is recommended since for this (secondary) ramp slope, considering the relatively wide variation in the FO and ST (set 1 m) dip angles, but the scatter (lack of clustering in sets) of other structures (like JN and JN1) which may combine with set 1 m to form wedges. ................................................................................................................ 233 Figure 12-10: Syväjärvi east slope base case set up in Slide2. OSA is 49°. Blue = mica schist; dark green = sulphidic schist; pink = plagioclase porphyrite. The blue line (W) is the simulated elevated groundwater profile applied in all analyses. FS ≈ 3.2. ......................................................................................................................................................................... 234 Figure 12-11: Syväjärvi west slope base case set up in Slide2. Green = metavolcanite; dark green = sulphidic schist; pink = plagioclase porphyrite. The blue line (W) is the simulated elevated groundwater profile applied in all analyses. OSA is 42°. FS ≈ 8.0. .............................................................................................................................................................. 234 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xxxix Figure 12-12: Syväjärvi final SDSs and SDDIR ranges based on the August 2024 FS pit design. This includes minor refinements to the July 2024 design, which were the basis for second iteration slope stability analyses. ................. 236 Figure 12-13: The August 2024 FS design of the Syväjärvi pit viewed to the south-west, direction 255°, which is the average approximate dip direction of inferred Syväjärvi FLTs SJF1, SJF2, and SJF3 (numbered red drops). The dashed red lines show the projected lines of intersection of the inferred FLTs (or broken rock zones) with the west and north slopes. The numbered red drops point in the approximate south-westerly dip directions. SJF4 is near-vertical. ...... 238 Figure 12-14: Geotechnical core logs (2014–2023) limited to be within maximum 50 m distance (inside and outside) from the April 2024 provisional Rapasaari pit slopes. Holes are coloured with the calculated RQD values. .............. 241 Figure 12-15: The spatial distribution of the ten selected drill cores which were re-logged, in relation to the 2021 design version Rapasaari open pit. Holes are coloured with the logged lithologies. .............................................................. 242 Figure 12-16: Rapasaari provisional open-pit design (April 2024) and lithologies cutting the pit slopes. Orange = spodumene pegmatite; pink = plagioclase porphyrite; dark green = sulphidic schist; light green = metavolcanite; blue = mica schist; grey = OVB. ............................................................................................................................................. 244 Figure 12-17: Zones with lower RQD and lower GSI values interpolated and extrapolated as broken rock zones, inferred as FLTs or brittle shear major geological structures (named and numbered as “Rapasaari FLT-shear” (FS1 to FS5, or FLT- SHR1 to FLT-SHR5)). ...................................................................................................................................................... 245 Figure 12-18: Rapasaari drill-hole-based GSI values of the main rock types (sill = plagioclase porphyrite). The mean values are indicated by the red diamonds. The median values are indicated by the lines that cross inside of the coloured boxes. The box encloses the interquartile range around the median. The whiskers extend out to vertical lines that mark the extreme value extents. ........................................................................................................................................... 248 Figure 12-19: Hydraulic conductivity, K, in m/s, measured in monitoring bores in Rapasaari rocks in Q1 of 2021 (Figure 18-73 copied from the 2022 draft DFS Report, Volume 4, Chapter 18). ...................................................................... 251 Figure 12-20: PoF for range of BFA from DIPS® kinematic stability analyses for the RA east slope SDS-1A to SDS-1E. Red cells exceed the target DAC by all criteria; green cells meet the target DAC by all criteria; yellow cells exceed but are within 5% of SDS-specific target criteria and warrant further assessment. ................................................................. 253 Figure 12-21: PoF for range of BFA from DIPS® kinematic stability analyses for the RA west slope SDS-2A and SDS-2B, and of the small NW satellite pit. Red cells exceed the target DAC by all criteria; green cells meet the target DAC by all criteria; yellow cells exceed but are within 5% of SDS-specific target criteria and warrant further assessment. ....... 254 Figure 12-22: Sensitivity analyses of the (pseudo-) probability of planar sliding versus slope dip (BFAs from 50° to 90°) in the average SDDIR of 180° of SDS-1A, all other factors kept constant, show PoF ≈ 15% for BFA = 70°. ................. 254 Figure 12-23: Sensitivity analyses of the (pseudo-) probability of wedge sliding versus slope dip (50° ≤ BFA ≤ 90°) in the average SDDIR of 180° of SDS-1A, all other factors kept constant, show PoF ≈ 28% for BFA = 70°. ........................... 255 Figure 12-24: Sensitivity analyses of the (pseudo-) probability of planar sliding versus slope dip (BFAs from 50° to 90°) in the average SDDIR of 215° of SDS-1B, all other factors kept constant, show PoF ≈ 17% for BFA = 70°. ................. 255 Figure 12-25: Sensitivity analyses of the (pseudo-) probability of wedge sliding versus slope dip (50° ≤ BFA ≤ 90°) in the average SDDIR of 180° of SDS-1A, all other factors kept constant, show PoF ≈ 42% for BFA = 70°. This was further analysed with SWEDGE®. .............................................................................................................................................. 256 Figure 12-26: The probability of wedge failures in Rapasaari SDS-1B nominally reduces from ±44% to PoF ≈ 36% if the minimum wedge size to consider as relevant is increased from 1 m3 to 2 m3, or from 0.026 MN to 0.052 MN. ........ 258 Figure 12-27: The probability of wedge failures in Rapasaari SDS-1B further reduces (nominally) from ±36% to PoF ≈ 32% if the minimum wedge size to consider as relevant is increased from 2 m3 to 5 m3, or from 0.052 MN to 0.13 MN, which is still quite small. ............................................................................................................................................... 259 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xl Figure 12-28: The Rapasaari east slope model in Slide2®, analysed with central case strength parameters. Blue = mica schist; red = pegmatite. The blue line (W) is the simulated elevated groundwater profile, applied in all analyses. Central case FS is slightly above 2.0. ......................................................................................................................................... 260 Figure 12-29: The Rapasaari west model in Slide2®, analysed with central case strength parameters. Blue = mica schist; green = vulcanite; pink = plagioclase porphyry; red =pegmatite. The blue line (W) is the simulated elevated groundwater profile, applied in all analyses. Central case FS is above 4.0 .................................................................. 260 Figure 12-30: The final Rapasaari SDSs based on SDDIR ranges scaled from the (shown) July 2024 pit design. ......... 262 Figure 12-31: Annual LOM ore and waste production schedule by pit. ....................................................................... 271 Figure 12-32: Annual LOM crusher feed from stockpiles. ............................................................................................ 272 Figure 12-33: Annual LOM mill feed. ............................................................................................................................ 272 Figure 12-34: Annual LOM final product (LiOH.H2O). ................................................................................................... 273 Figure 12-35: Syväjärvi monthly stockpile levels. ......................................................................................................... 274 Figure 12-36: Rapasaari monthly stockpile levels. ........................................................................................................ 275 Figure 12-37: Monthly contaminant grade in crusher feed from stockpiles. ............................................................... 275 Figure 12-38: High sulphide- and arsenic-bearing waste rock excavated annually. ..................................................... 276 Figure 12-39: Waste rock storage facilities at Syväjärvi. .............................................................................................. 284 Figure 12-40: Waste rock storage facilities at Rapasaari. ............................................................................................. 285 Figure 12-41: Syväjärvi Pit: dewatering routes to dewatering pond (source: Nurizon Consulting Engineers). ........... 287 Figure 12-42: Rapasaari Pit: dewatering routes to dewatering pond per mining phase (source: Nurizon Consulting Engineers). .................................................................................................................................................................... 288 Figure 12-43: Surface drainage berm and earth channel drawing (source: Nurizon Consulting Engineers). ............... 289 Figure 12-44: SMU matrix for practical and productive loader selection. .................................................................... 292 Figure 12-45: Syväjärvi haulage routes. ........................................................................................................................ 293 Figure 12-46: Rapasaari haulage routes. ...................................................................................................................... 294 Figure 12-47: Syväjärvi truck fleet. ............................................................................................................................... 295 Figure 12-48: Rapasaari truck fleet. .............................................................................................................................. 296 Figure 13-1: Päiväneva concentrator – simplified block flow diagram. ........................................................................ 298 Figure 13-2: Basic ore-sorting operating principle........................................................................................................ 299 Figure 13-3: Keliber lithium refinery BFD. .................................................................................................................... 306 Figure 13-4: Plant ramp-up schedules. ......................................................................................................................... 312 Figure 14-1: General layout of the Syväjärvi mine site. ................................................................................................ 314 Figure 14-2: General layout of the Rapasaari mine site. .............................................................................................. 315 Figure 14-3: Proximity of Syväjärvi to Rapasaari, also indicating the haul road that will be constructed to transport the ore to the concentrator plant. ...................................................................................................................................... 316 Figure 14-4: Layout of the Päiväneva concentrator site. .............................................................................................. 318 Figure 14-5: Stormwater management plant at the Päiväneva concentrator site. ...................................................... 319 Figure 14-6: GA drawing of the Keliber lithium Refinery situated in the KIP at Kokkola. ............................................. 321 Figure 14-7: Photo of the Keliber lithium Refinery construction site in the KIP at Kokkola. ........................................ 322 Figure 14-8: Overall layout of the Keliber lithium Refinery at the KIP site. .................................................................. 323 Figure 14-9: Schematic of the TSF located within the Päiväneva plant area. ............................................................... 324 Figure 15-1: Global lithium demand outlook. ............................................................................................................... 326 Figure 15-2: Automotive battery-grade lithium demand by powertrain. .................................................................... 327 Figure 15-3: RoW light vehicle production. .................................................................................................................. 328 Figure 15-4: Global BEV battery demand by chemistry. ............................................................................................... 329

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xli Figure 15-5: BEV production by lithium precursor. ...................................................................................................... 330 Figure 15-6: Primary lithium supply by region. ............................................................................................................. 333 Figure 15-7: Primary lithium supply, including possible projects. ................................................................................ 334 Figure 15-8: Primary lithium supply from Europe, including projects. ......................................................................... 335 Figure 15-9: Lithium supply potential from EoL BEVs by region. .................................................................................. 336 Figure 15-10: All-in sustaining costs: 2024E ................................................................................................................. 337 Figure 15-11: All-in sustainable costs: 2030F ................................................................................................................ 338 Figure 15-12: LiOH.H2O technical- and industrial-grade specification (source: SSW). ................................................. 339 Figure 15-13: LiOH.H2O battery-grade specification (source SSW). ............................................................................. 340 Figure 15-14: Lithium supply vs demand. ..................................................................................................................... 343 Figure 15-15: Lithium supply-demand balance. ........................................................................................................... 343 Figure 15-16: Lithium supply vs demand in 2040, including projects........................................................................... 344 Figure 17-1: Monthly Capex. ......................................................................................................................................... 358 Figure 18-1: FCF forecast. ............................................................................................................................................. 371 Figure 18-2: Cumulative FCF forecast. .......................................................................................................................... 371 Figure 18-3: Plant ramp-up curve. ................................................................................................................................ 374 Figure 20-1: Risk severity heat map. ............................................................................................................................. 383 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xlii LIST OF TABLES Table 1-1: QP and responsibility. ...................................................................................................................................... 3 Table 2-1: Key environmental legislation........................................................................................................................ 24 Table 2-2: Valid mining permits as at 31 December 2024. ............................................................................................. 27 Table 2-3: Valid exploration permits as at 31 December 2024....................................................................................... 27 Table 2-4: Exploration permits under application as at 31 December 2024. ................................................................. 28 Table 2-5: Environmental permitting status of Keliber Lithium Project. ......................................................................... 31 Table 2-6: Permitting situation of Keliber Project as of 31 December 2024. ................................................................. 31 Table 4-1: Previous operators. ........................................................................................................................................ 37 Table 4-2: Summary of the sampling and ground geophysics (after Ahtola et al., 2015). ............................................. 38 Table 5-1: Summary of chemical composition and density of the main lithium minerals associated with pegmatites.52 Table 6-1: Summary of drilling completed over the Keliber Lithium Project (source: Keliber). ..................................... 59 Table 7-1: Summary of expected values for Keliber’s internally sourced reference materials and commercially sourced AMIS0355 (source: Keliber, 2023). ................................................................................................................................. 69 Table 7-2: Performance of AMIS0355 and Keliber’s reference materials over the period 2016 to 2023 at Labtium (source: Keliber). ............................................................................................................................................................. 71 Table 8-1: List of drill holes field-checked during site visit. ............................................................................................ 83 Table 8-2: List of drill holes (geological and sample logs and assay certificates) checked against drill holes during site visit. ................................................................................................................................................................................. 86 Table 8-3: Checks conducted on drill holes from various campaigns. ............................................................................ 88 Table 8-4: Summary of historical drill data review within interpreted spodumene pegmatite zone. ........................... 93 Table 9-1: Syväjärvi comminution characteristics. ......................................................................................................... 99 Table 9-2: Summary of flotation results. ...................................................................................................................... 101 Table 9-3: TOMRA ore-sorting balance 2021. ............................................................................................................... 115 Table 9-4: Modal composition of the waste rocks of Syväjärvi, Länttä, and Rapasaari. .............................................. 129 Table 9-5: Recovery parameters. .................................................................................................................................. 131 Table 9-6: ICP-OES results of the crystallisation product from all three pilot plant trials. ........................................... 135 Table 10-1: Microsoft Access® databases by date. ....................................................................................................... 140 Table 10-2: Microsoft Access® databases – drilling and assay summary. .................................................................... 140 Table 10-3: Example from Rapasaari of grouping simplified lithologies for modelling. ............................................... 141 Table 10-4: Naive statistics for Li2O%. .......................................................................................................................... 148 Table 10-5: Composite statistics for Li2O%. .................................................................................................................. 148 Table 10-6: Variogram parameters for Li2O%. .............................................................................................................. 149 Table 10-7: Block model parameters. ........................................................................................................................... 152 Table 10-8: Search parameters. .................................................................................................................................... 154 Table 10-9: Comparison between the input composites and ordinary kriged estimates. ........................................... 155 Table 10-10: Conceptual parameters used to determine RPEE. ................................................................................... 163 Table 10-11: Keliber Mineral Resources, exclusive of Mineral Reserves, at a 0.5% Li2O cut-off as at 31 December 2024 and reported on a 79.82% ownership basis. ................................................................................................................ 164 Table 10-12: Keliber Mineral Resources, inclusive of Mineral Reserves, at a 0.5% Li2O cut-off reported on a 100% basis ...................................................................................................................................................................................... 165 Table 10-13: Keliber Mineral Resources, inclusive of Mineral Reserves, at a 0.5% Li2O cut-off reported on a 79.82% attributable ownership basis. ....................................................................................................................................... 166 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xliii Table 10-14: Lithium product conversion matrix. ......................................................................................................... 167 Table 10-15: Comparison between the 2024 and 2023 MREs (exclusive of Mineral Reserves) at 0.5% cut-off. ......... 170 Table 11-1: Summarised block model properties – sub-celled geological models. ...................................................... 173 Table 11-2: Summarised block model properties – regularised geological models. .................................................... 174 Table 11-3: Rock codes. ................................................................................................................................................ 177 Table 11-4: Cut-off calculation parameters. ................................................................................................................. 181 Table 11-5: COG per pit used in optimisation. .............................................................................................................. 181 Table 11-6: Pit optimisation input parameters: financial parameters. ........................................................................ 184 Table 11-7: Pit optimisation input parameters: mining cost. ....................................................................................... 185 Table 11-8: Pit optimisation input parameters: processing cost. ................................................................................. 186 Table 11-9: Pit optimisation input parameters: selling costs. ...................................................................................... 186 Table 11-10: Optimisation analysis comparison for Syväjärvi. ..................................................................................... 188 Table 11-11: Optimisation analysis comparison for Rapasaari. .................................................................................... 189 Table 11-12: Ramp width design parameters. .............................................................................................................. 191 Table 11-13: Recommended parameters for open-pit designs – Syväjärvi. ................................................................. 191 Table 11-14: Recommended parameters for open-pit designs – Rapasaari. ............................................................... 192 Table 11-15: Syväjärvi LOM pit design diluted tonnes. ................................................................................................ 194 Table 11-16: Rapasaari LOM pit design diluted tonnes. ............................................................................................... 196 Table 11-17: In situ Mineral Resources categorised by Mineral Resource classification at Syväjärvi. ......................... 199 Table 11-18: Syväjärvi in situ Mineral Resources exclusive and inclusive of mine plan. .............................................. 201 Table 11-19: Syväjärvi diluted Mineral Resources in the LOM plan after SMU re-blocking and including all Li2O-bearing spodumene pegmatite blocks. ...................................................................................................................................... 202 Table 11-20: Syväjärvi diluted Mineral Resources in the LOM plan after marginal cut-off.......................................... 202 Table 11-21: In situ Mineral Resources categorised by Mineral Resource classification at Rapasaari. ....................... 205 Table 11-22: Rapasaari in situ Mineral Resources exclusive and inclusive of mine plan. ............................................ 207 Table 11-23: Rapasaari diluted Mineral Resources in the LOM plan after SMU re-blocking and including all Li2O-bearing spodumene pegmatite blocks. ...................................................................................................................................... 207 Table 11-24: Rapasaari diluted Mineral Resources in the LOM plan after marginal cut-off. ....................................... 208 Table 11-25: Mineral Reserves for Keliber’s Syväjärvi and Rapasaari open-pit operations as at 31 December 2024. 209 Table 11-26: Reconciliation between 2023 Mineral Reserves and 2024 Mineral Reserves. ........................................ 210 Table 12-1: UCS test results. ......................................................................................................................................... 223 Table 12-2: Summary of the DST samples by rock type and structure type. Samples for saw-cut DST were intact rock, to be sawn parallel to the FO or ST rock texture. ......................................................................................................... 224 Table 12-3: Residual Friction Angles (Phir) derived from DST on saw-cut intact rock samples, after applying Hencher s correction. EP Laboratory calculated Phir by Mohr-Coulomb curve fit to three test stages. ....................................... 224 Table 12-4: Peak friction angles derived from DST on structure types in four rock types, after applying the Hencher correction. Structure types are named according to the 2024 relogged cores from which the samples were taken. 225 Table 12-5: Mean and standard deviations of peak friction angles by rock type derived from DST, after applying Hencher’s correction (green numbers), compared with frictional angles estimated with the BB criterion at normal stress 400 kPa, to approximately match the mean normal stress applied in the DST (left three columns). Mean Phi-BB are within one standard deviation of Phi-DST, except for PP which is within two standard deviations. ..................... 225 Table 12-6. Summary of the discontinuity sets identified in Syväjärvi open pit in the SDSs. ....................................... 227 Table 12-7: Typical FoS and PoF acceptance criteria values for open-pit slopes at various scales proposed in Table 9.9 in the Guidelines for Open Pit Slope Design. ............................................................................................................... 229 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xliv Table 12-8: Slope design configurations, viz. BH, BFA, and BWs, and ISAs recommended for the Syväjärvi FS design. NB: Possible impacts of modelled FLTs Syväjärvi-FLT1, 2 and 3 on SDS-2B and SDS-2C ramp slopes must be mitigated with the pit design and/or managed during mining. Current estimated confidence levels in these 3D models are Inferred to Indicated. To be upgraded from quarry face mapping. ................................................................................................ 237 Table 12-9: UCS laboratory test results. ....................................................................................................................... 247 Table 12-10: Summary of the direct shear tested sample numbers by rock type and structure type. ........................ 247 Table 12-11: Residual friction angles derived from saw-cut rock samples, after applying Hencher’s correction. E- Precision Laboratory calculated Phir by Mohr-Coulomb curve fit to results of three test stages at increasing normal stresses. ......................................................................................................................................................................... 247 Table 12-12: Peak friction angles derived from DST on structure types in four rock types, after applying the Hencher correction. ..................................................................................................................................................................... 247 Table 12-13: Mean and standard deviations of peak friction angles by rock type derived from DST, after applying Hencher’s correction, compared with frictional angles estimated with the BB criterion at normal stress 400 kPa, to approximately match the mean normal stress applied in the DST (left three columns). ............................................. 248 Table 12-14: Summary of the discontinuity sets identified in the Rapasaari open pit in the defined SDSs................. 249 Table 12-15: Deterministic (FoS) and probabilistic (PoF) DAC from bench to overall slope scales proposed in Table 9.9 of the Guidelines for Open Pit Slope Design (Read & Stacey, 2009). ........................................................................... 252 Table 12-16: Slope design configurations recommended for the Rapasaari DFS pit optimisation and design. SDS locations refer to Figure 12-30. NB: Possible impacts of modelled FLTs RA-FLT1 to FLT5 on ramp slopes must be mitigated with the pit design and/or managed during mining. Current estimated confidence levels in these 3D models are undefined but are likely Inferred. AFRY to review, upgrade 3D models, and qualify confidence levels and residual risks. .............................................................................................................................................................................. 263 Table 12-17: Summary of groundwater inflows per deposit. ....................................................................................... 265 Table 12-18: Keliber Lithium Project production summary. ......................................................................................... 267 Table 12-19: Production scheduling basis of design summary. .................................................................................... 267 Table 12-20: Design criteria for daily ore production. .................................................................................................. 269 Table 12-21: Contaminants limitations. ........................................................................................................................ 269 Table 12-22: Syväjärvi and Rapasaari time usage model for loading and hauling. ...................................................... 277 Table 12-23: Summary of waste production blasts. ..................................................................................................... 279 Table 12-24: Summary of ore production blasts. ......................................................................................................... 280 Table 12-25: Stockpile grade ranges. ............................................................................................................................ 282 Table 12-26: WRD design parameters. ......................................................................................................................... 283 Table 12-27: OVB dump design parameters. ................................................................................................................ 283 Table 12-28: Syväjärvi scheduled waste material volumes versus waste storage capacity. ........................................ 284 Table 12-29: Rapasaari scheduled waste material volumes versus waste storage capacity. ....................................... 286 Table 12-30: Barloworld CAT bucket size calculation. .................................................................................................. 290 Table 12-31: Syväjärvi and Rapasaari load and haul input parameters. ....................................................................... 294 Table 13-1: Key process design criteria – concentrator. ............................................................................................... 302 Table 13-2: Concentrator reagents and consumables. ................................................................................................. 304 Table 13-3: Labour requirements for the Päiväneva concentrator plant. .................................................................... 305 Table 13-4: Key process design criteria – lithium hydroxide chemical plant. ............................................................... 309 Table 13-5: LiOH . H2O product specifications. .............................................................................................................. 310 Table 136: Labour requirements for the Keliber lithium refinery. ............................................................................... 311 Table 15-1: Price forecast for LiOH battery grade (USD/t) as reported in January 2025 (source: UBS Bank). ............. 345

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xlv Table 15-2: LiOH prices applied in the financial model. ............................................................................................... 346 Table 16-1: Field studies carried out at Syväjärvi, Rapasaari, and Outovesi mine sites and Vionneva Natura 2000 area. ...................................................................................................................................................................................... 347 Table 17-1: Capex summary. ......................................................................................................................................... 357 Table 17-2: Total Opex. ................................................................................................................................................. 358 Table 17-3: Mining Opex. .............................................................................................................................................. 359 Table 17-4: Crushing, sorting, and concentrator Opex. ................................................................................................ 359 Table 175: Conversion and Keliber lithium Refinery Opex.- ......................................................................................... 359 Table 17-6: Other variable Opex. .................................................................................................................................. 360 Table 17-7: Freight and transportation Opex. .............................................................................................................. 360 Table 17-8: Fixed costs Opex. ....................................................................................................................................... 361 Table 17-9: Royalties and fees Opex. ............................................................................................................................ 361 Table 17-10: LiOH price forecast. .................................................................................................................................. 362 Table 17-11: Unit costs. ................................................................................................................................................ 363 Table 17-12: Steady state indicators. ............................................................................................................................ 365 Table 17-13: Accuracy of estimates. ............................................................................................................................. 365 Table 18-1: LiOH price forecast. .................................................................................................................................... 368 Table 18-2: Exchange rate forecast. .............................................................................................................................. 368 Table 18-3: Modelled lithium recoveries included in the Technical Economic Model. ................................................ 369 Table 18-4: FCF summary. ............................................................................................................................................. 370 Table 18-5: Key metrics. ................................................................................................................................................ 372 Table 18-6: Plant ramp-up curve. ................................................................................................................................. 373 Table 18-7: Plant ramp-up individual sensitivities. ....................................................................................................... 375 Table 18-8: Plant ramp-up combined sensitivities. ...................................................................................................... 377 Table 18-9: Price sensitivity. ......................................................................................................................................... 378 Table 18-10: Discount rate sensitivity. .......................................................................................................................... 379 Table 18-11: Exchange rate sensitivity.......................................................................................................................... 379 Table 18-12: Logistics cost sensitivity. .......................................................................................................................... 379 Table 20-1: Project milestones. .................................................................................................................................... 382 Table 20-2: Risk assessment – medium to high risks. ................................................................................................... 383 Table 20-3: Risk assessment – high risks mitigation measures. ................................................................................... 384 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland xlvi LIST OF EQUATIONS Equation 1: Price adjustment formula for the royalty for the Syväjärvi and Leviäkangas deposits. .............................. 23 Equation 2: Price adjustment formula for the royalty for the Rapasaari deposit. ......................................................... 23 Equation 3: Dissolution of sodium carbonate. .............................................................................................................. 122 Equation 4: Formation of lithium carbonate and analcime solids from β-spodumene................................................ 122 Equation 5: Conversion of calcium hydroxide plus lithium carbonate to lithium hydroxide and calcium carbonate.. 123 Equation 6: Conversion of lithium hydroxide to lithium carbonate. ............................................................................ 123 Equation 7: Final recovery formula applied to mine planning and financial modelling. .............................................. 131 Equation 8: Mean SG dependence on grade at Rapasaari. .......................................................................................... 159 Equation 9: Mean SG dependence on grade at Syväjärvi. ............................................................................................ 160 Equation 10: Recovery from the trendline for the Syväjärvi open pit. ......................................................................... 178 Equation 11: Recovery from the trendline for the Rapasaari open pit. ....................................................................... 178 Equation 12: Lithium hydroxide production. ................................................................................................................ 179 Equation 13: Final mass recovery of lithium hydroxide. ............................................................................................... 179 Equation 14: Marginal COG. ......................................................................................................................................... 180 Equation 15: Conversion of beta spodumene to lithium carbide. ................................................................................ 307 Equation 16: Cold conversion reaction. ........................................................................................................................ 307 Equation 17: Crystallisation reaction. ........................................................................................................................... 308 Equation 18: Formation of lithium cabornate. ............................................................................................................. 309 Equation 19: Formation of aluminium carbonate. ....................................................................................................... 309 LIST OF APPENDICES Appendix A: Risk Register ............................................................................................................................................. 404 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 1 1 INTRODUCTION [§229.601(b)(96)(iii)(B)(2)] 1.1 Registrant [§229.601(b)(96)(iii)(B)(2)(i)] This Technical Report Summary (TRS) was prepared for Sibanye-Stillwater Limited (Sibanye-Stillwater, SSW, also referred to as the Company), a limited public company (JSE: SSW; NYSE:SBSW) with its registered office in South Africa. SSW are involved in the exploration, development, mining, and processing of lithium spodumene mineral deposits in Central Ostrobothnia, Finland. This TRS relates to the Keliber Lithium Project (‘the Keliber Project’ or ‘the Project’), which consists of exploration and planned mining operations around Kaustinen, a planned mineral processing plant at Kaustinen (the Keliber Lithium Concentrator), and a planned conversion plant at Kokkola, the Keliber lithium refinery. The Mineral Resources and Mineral Reserves is declared on the economics of the production of lithium hydroxide monohydrate. SSW announced on 28 November 2022, subsequent to securing an effective controlling interest of 84.96% in Keliber Lithium (Pty) Ltd (Keliber), as announced on 3 October 2022, the approval of capital expenditure of EUR588 million for the Keliber Lithium Project, beginning with the construction of the Keliber lithium refinery. The construction of the Keliber lithium refinery commenced in 2023 and completion and hot commissioning is planned for February 2026. The construction of the bulk infrastructure will be completed by May 2025. The construction of the concentrator started in November 2023, and it is planned to be completed in December 2025. Commissioning of the plant is planned for January 2026. It should be noted that estimation of capital and operating costs is inherently a forward-looking exercise. These estimates rely upon a range of assumptions and forecasts that are subject to change depending upon macro-economic conditions, operating strategy, and new data collected through future operations. Therefore, changes in forward- looking assumptions can result in capital and operating costs that deviate more than 15% from the costs forecast herein. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 2 1.2 Terms of Reference and Purpose of the Report [§229.601(b)(96)(iii)(B)(2)(ii)] 1.2.1 Terms of Reference SSW commissioned VBKOM (Pty) Ltd (VBKOM) to compile this TRS for the Keliber Project according to Item 601 of the United States Securities and Exchange Commission’s (SEC’s) Subpart 1300 of Regulation S-K (S-K 1300), under the Securities Act of 1933 and the Securities Exchange Act of 1934. Mineral Resources have been prepared by ERM International Group Limited (ERM) and are fully incorporated into this Report. 1.2.2 Purpose [§229.601(b)(96)(iii)(B)(2)(v)] This TRS, dated at 31 December 2024, updates the previously filed technical report summaries identified below. The Mineral Resources and Mineral Reserves have been prepared and reported according to the requirements of S-K 1300. • Technical Report Summary, CSA Global South Africa (Pty) Ltd, an ERM Group company, 21 April 2024 (Effective date 31/12/2023) (Report number R142.2024). • Technical Report Summary, SRK Consulting (South Africa) (Pty) Ltd, 13 December 2023 (Effective date (31/12/2022) (Report number 592138). • 2022 Keliber TRS. The Original 2022 Keliber TRS was the first TRS for the Keliber Lithium Project filed by SSW in support of the reporting of Mineral Resources and Mineral Reserves for Keliber. 1.2.3 Compliance This TRS report has been compiled to ensure regulatory compliance. 1.2.4 Notation This report uses a shorthand notation to demonstrate compliance with Item 601 of Regulation S-K 1300 as follows: • [[§229.601(b)(96)(iii)(B)(2)] represents subsection (iii)(B)(2) of section 96 of CFR 229.601(b) (‘Item 601 of Regulation S-K’). 1.3 Qualified Persons’ Qualifications and Site Visit [§229.601(b)(96)(iii)(B)(2)(iv)] 1.3.1 Qualified Persons [§229.1302(b)(1)(ii)] This report was prepared by VBKOM, a third-party consulting firm comprising mining experts in accordance with §2291302(b)(1). SSW have determined that VBKOM meet the qualifications specified under the definition of Qualified Person in §229.1300. References to the Qualified Person, or QP, in this report are references to VBKOM and not to any individual employed by VBKOM. In the case of instances referring to geology, exploration, Mineral Resources and direct relations, the QP refers to ERM and not to any individual employed by ERM.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 3 The QPs have supervised the preparation of this Report and take responsibility for the contents of the Report as set out in Table 1-1. Table 1-1: QP and responsibility. Qualified Person Report Responsibilities Report Sections ERM Mineral Resources • Geological Setting, Mineralisation, and Deposit • Exploration • Sample Preparation, Analyses, and Security • Data Verification • Mineral Resource Estimates • Executive Summary, Conclusions and Recommendations that incorporate any or part of the above subjects. 5, 6, 7, 8, 10, and those parts of Executive Summary, Conclusions and Recommendations that incorporate any or part of the related subjects VBKOM Mineral Reserves • All sections exclusive of those prepared by ERM All sections exclusive of those prepared by ERM The Date and Signature Page can be found on page ii. Neither VBKOM nor personnel nominated for the completion or review of work, including the QPs, have any interest (present or contingent) in Sibanye-Stillwater and its subsidiaries (including Keliber), its directors, senior management, advisers or the mineral properties reported on in this QPR. The proposed work, and any other work done by VBKOM for Keliber, is strictly in return for professional fees. The fees for this engagement are not contingent on any aspect of this report and were determined before commencement of the engagement. Payment for the work is not in any way dependent on the outcome of the work, nor on the success or otherwise of SSW and its subsidiaries' own business dealings. The QPs and authors have no bias with respect to the assets that are the subject of the Report, or to the parties involved with the assignment. There is no conflict of interest in VBKOM undertaking the QPR as contained in this document. 1.3.2 Site Visit 1.3.2.1 Mineral Resource QP A site visit was conducted by the ERM QP from 11 to 13 July 2023. The Project site was visited on 11–12 July 2023 and included: • A visit to Keliber’s geology office and core processing and storage facility in Kaustinen where a selection of drill holes was reviewed against the logs and assay certificates and a review of the logging and sampling protocols. • A field visit to three of the deposits was made, namely Syväjärvi, Rapasaari, and Tuoreetsaaret, during which the location and field confirmation of a selection of drill holes and a number of outcrops around these deposits were completed. • Discussions with Keliber geologists, Pentti Grönholm (Senior Manager Geology) and Joonas Kurtti (Exploration Geologist), regarding the Project exploration history, geology, core processing, operating procedures, and data collection processes. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 4 Keliber’s main office in Kokkola was visited on 13 July 2023. The QP is satisfied that the necessary steps in the data collection process were taken to verify the data used for the Mineral Resource Estimate (MRE). Refer to Chapter 8.2 for more detail on this site visit. 1.3.2.2 Mineral Reserve QP Mr Wilhelm Warschkuhl and Mr George Olivier of VBKOM visited the subject property on 20 and 21 August 2024. The site visit entailed visiting and verifying the mining areas, project infrastructure, and pilot mining area located at Syväjärvi. During the visit, the Keliber lithium refinery was viewed without accessing it as this was prohibited. In addition to the Syväjärvi and Rapasaari mining areas, Länttä and Outovesi mining areas were also inspected. During the site visit, discussions were held regarding the Project with various stakeholders, Pentti Grönholm (Senior Manager Geology), and Sari Koivisto (Senior Manager Mining) to identify Project-related risks. Figure 1-1 shows the Keliber offices and Syväjärvi pilot mining area. Keliber Offices and Syväjärvi Pilot Mining Area Figure 1-1: Location of the Keliber offices and the Syväjärvi pilot mining area. Figure 1-2 shows the mining areas visited for the Keliber Project, namely Syväjärvi, Rapasaari, Länttä, and Outovesi. Keliber offices Pilot mining area S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 5 Keliber Mining Areas Figure 1-2: Keliber mining areas visited. Figure 1-3 shows the processing areas visited for the Keliber Project, namely the Lithium concentrator plant at Päiväneva and the LiOH plant at the Kokkola Industrial Park (KIP). Rapasaari Syväjärvi Länttä Outovesi S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 6 Keliber Processing Areas Figure 1-3: Keliber processing areas visited. 1.4 Effective Date [§229.1302(b)(iii)(3)] The effective date of the TRS is 31 December 2024, which satisfies the S-K 1300 requirement of a current report. 1.5 Sources of Information §229.601(b)(96)(iii)(B)(2)(iii)] VBKOM have made reliance on information provided by professional employees of SSW, as well as the following key reports: • WSP Global Inc. 2022. Definitive Feasibility Study Report – Keliber Lithium Project, 2022. (Referred to as Feasibility Study (FS) in this report. The Definitive Feasibility Study is considered to be the equivalent of a FS as defined in the S-K 1300 Definition Standards adopted 26 December 2018.) • Broda, L., Blomqvist, N., Nieminen, V. 09 February 2025. Definitive Feasibility Study of the Syväjärvi Open Pit Mine (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. • Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study for the Rapasaari Open Pit (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. • Deminey, J.G.L., & Barnard, E. 02 December 2024. Keliber Lithium Mine – Dewatering and Drainage Strategy. Nurizon Consulting Engineers, Revision Number 1, Final. • FLSmidth. January 2024. Evaluation of spodumene alpha to beta phase conversion utilising a cyclone preheater rotary calciner system for Keliber Technologies OY Finland. Project # 9232517826. • Metso. 28 August 2024. Keliber: Recovery of Lithium and production of LiOH·H2O from calcined spodumene concentrates in a continuous pilot. • Rorke, A.J. 09 October 2024. Blast Designs – Syväjärvi and Rapasaari Lithium Mines. AJR, Revision Number 2. Concentrator plant Hydroxide refinery

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 7 • SRK Consulting (South Africa) (Pty) Ltd. December 2023. An Updated Due Diligence Addendum Report on the Keliber Lithium Project, Finland. Additional sources of information and data used in the preparation of the TRS are included in Section References and are duly referenced throughout the report. SSW have confirmed in writing that to their knowledge, the information they provided to VBKOM was complete and not incorrect, misleading or irrelevant in any material aspect. VBKOM have no reason to believe that any material facts have been withheld. 1.6 List of Units, Abbreviations & Acronyms, and Chemical Formulae 1.6.1 List of Units Unit Definition dB Decibel BV Bed Volume cm Centimetre Mm3 Cubic Megametre m3 Cubic Metre m3/d Cubic Metre per Day m3/h Cubic Metre per Hour m3/t Cubic Metre per Tonne ° Degree °C Degree Celsius EP Environmental Permit EUR Euro EUR/a Euros per Annum EUR/m2 Euros per Square Metre EUR/m3 Euros per Cubic Metre EUR/t Euros per Tonne EUR’000/k EUR Thousands of Euros EUR’000’000/M EUR Millions of Euros Ga Giga Annum GWh Gigawatt Hour g/cm3 Grams per Cubic Centimetre g/L Grams per Litre g/m3 Grams per Cubic Metre ha Hectare S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 8 Unit Definition h Hour IRR Inherent risk rating kBCM Kilo-Bank Cubic Metre kg Kilogram kg/h Kilograms per Hour kg/m Kilograms per Metre kg/m3 Kilograms per Cubic Metre km Kilometre km/h Kilometres per Hour kPa Kilopascal kt Kilotonne ktpa Kilotonne per Annum kV Kilovolt kW Kilowatt kWh Kilowatt Hour kWh/t Kilowatt Hour per Tonne L Litre MVA Megavolt Ampere m Metre m/s Metre per Second mamsl Metres Above Mean Sea Level masl Metres Above Sea Level min Minute min/hr Minutes per Hour µg/L Microgram per Litre µm Micrometre mg/L Milligrams per Litre MLCM Million Loose Cubic Metres mm Millimetre MPa Megapascal EURm Millions of Euros Mt Million Tonne Mtpa Million Tonnes per Annum M Molarity/Molar Concentration ppm Parts per Million S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 9 Unit Definition % Per cent wt% Percentage by Weight s Second km2 Square Kilometre m2 Square Metre TWh Terawatt Hour t Tonne t/hr Tonnes per Hour tpa Tonne per Annum t/m3 Tonne per Cubic Metre USD United States Dollar USD/t United States Dollars per Tonne ZAR South African Rand ZAR/g South African Rands per Gram ZAR/t South African Rands per Tonne ZAR/t/10 m South African Rands per Tonne per 10 Metres 1.6.2 List of Abbreviations & Acronyms Abbreviation/Acronym Definition AAS Atomic Absorption Spectroscopy ABA Acid-Base Accounting AK Alholmens Kraft AMIS African Mineral Standards ARDML Acid Rock Drainage and Metal L eaching AVI Regional State Administrative Agency BB Barton-Bandis BD Banding BEV Battery Electric Vehicle BFA Bench Face Angle BG Battery-Grade Lithium Hydroxide BH Bench Height BW Berm Width CAGR Compound Average Growth Rate Capex Capital Expenditure S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 10 Abbreviation/Acronym Definition COG Cut-Off Grade Conc. Concentrate CRM Certified Reference Material DAC Design Acceptance Criteria DAF Dissolved Air Flotation DCF Discounted Cash Flow DLE Direct Lithium Extraction DMS Dense Media Separation DPA Direct Products of Anatexis DST Direct Shear Test EBITDA Earnings Before Interest, Taxes, Depreciation, and Amortisation EGL Effective Grinding Length EIA Environmental Impact Assessment ELY-Keskus Centres for Economic Development, Transport and the Environment (Elinkeino-, Liikenne-, ja Yympäristökeskus) EoL End-of-Life EPCM Engineering, Procurement and Construction Management EQS Environmental Quality Standard EREV Extended Range Electric Vehicle ERM ERM International Group Limited ESS Energy Storage System ETP Effluent Treatment Plant EU European Union FCF Free Cash Flow FHEV Full Hybrid Electric Vehicle FINAS Finnish Accreditation Service FLT Fault FO Foliation FoS Factor of Safety FS Feasibility Study FTIR Fourier Transform Infrared FX Foreign Exchange G&A General and Administrative GA General Arrangement GISTM Global Industry Standard on Tailings Management GPS Global Positioning System

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 11 Abbreviation/Acronym Definition GSI Geological Strength Index GTK Geological Survey of Finland HDEV Heavy-Duty Electric Vehicle HDPE High-Density Polyethylene HLS Heavy Liquid Separation HVAC Heating, Ventilation, and Air Conditioning ICE Internal Combustion Engine ICP-OES Inductively Coupled Plasma Optical Emission Spectroscopy IDW2 Inverse Distance Weighted to the Power of 2 IEC International Electrotechnical Commission IFA Indicative Friction Angle IRR Internal Rate of Return ISA Inter-Ramp Slope Angle ISO International Organization for Standardization Ja Joint Alteration JCS Joint Compressive Strength JN Joint Jr Joint Roughness JRC Joint Roughness Coefficient JS Joint Set JSE Johannesburg Stock Exchange Keliber Keliber Lithium (Pty) Ltd KIP Kokkola Industrial Park KLP Kaustinen Lithium Pegmatite KNA Kriging Neighbourhood Analysis Labtium Eurofin Labtium Group LAeq A-Weighted Equivalent Continuous Sound Level LCE Lithium Carbonate Equivalent LCT Lithium-Caesium-Tantalum LFP Lithium-Ion Phosphate LiDAR Light Detection and Ranging LIMS Low-Intensity Magnetic Separator LMFP Lithium Manganese Iron Phosphate LOM Life of Mine LPG Liquid Petroleum Gas S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 12 Abbreviation/Acronym Definition LT Long-Term M&I Measured and Indicated (Mineral Resources) MCAF Mining Cost Adjustment Factor MHEV Mild Hybrid Electric Vehicle MM Molar Mass MO Marginal Ore MP/MPG/MPEG Muscovite Pegmatite MRE Mineral Resource Estimate MS Mica Schist MV Metavolcanite MVR Mechanical Vapour Recompression NAG Net Acid Generation NIF Near Infrared NPV Net Present Value NYF Niobium-Yttrium-Fluorine NYSE New York Stock Exchange OEM Original Equipment Manufacturer OK Ordinary Kriging Opex Operating Expenditure OSA Overall Slope Angle OVB Overburden p.a. Per Annum PEA Preliminary Economic Assessment PFL Positive Flow Logging PFS Pre-Feasibility Study PG Pegmatite PHEV Plug-In Hybrid Electric Vehicle PoF Probability of Failure PP Plagioclase Porphyrite PPI Producer Price Index PSD Particle Size Distribution QAQC Quality Assurance and Quality Control QC Quality Control QP Qualified Person QPR Qualified Person’s Report S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 13 Abbreviation/Acronym Definition R&D Research and Development RC Reverse Circulation REACH Registration, Evaluation, Authorisation and Restriction of Chemicals RF Revenue Factor RHG Rapasaari High-Grade Ore RL Reduced Level RLG Rapasaari Low-Grade Ore RMG Residual Melts of Granitic Magmatism ROM Run of Mine RoW Rest of the World RPEE Reasonable Prospects for Economic Extraction RQD Rock Quality Designation SAC Supreme Administrative Court SCO Syväjärvi Contaminated Ore SD Standard Deviation SDDIR Slope Dip Direction SDS Slope Design Sector SEC Securities and Exchange Commission SEM Scanning Electron Microscopy SG Specific Gravity SHG Syväjärvi High-Grade Ore S-K 1300 Subpart 1300 of Regulation S-K of the United States Securities and Exchange Commission SLG Syväjärvi Low-Grade Ore SMO Syväjärvi Marginal Ore SMU Smallest Mining Unit SOP Standard Operating Procedure SP/SPG/SPEG Spodumene Pegmatite SRK SRK Consulting South Africa (Pty) Ltd SS Sulphidic Schist ST Schistosity SSW Sibanye-Stillwater Limited SW Southwest TG Technical-Grade Lithium Hydroxide TRS Technical Report Summary TS Topsoil S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 14 Abbreviation/Acronym Definition TSF Tailings Storage Facility UCS Uniaxial Compressive Strength USA United States of America VAC Vaasa Administrative Court VBKOM VBKOM (Pty) Ltd WACC Weighted Average Cost of Capital WHIMS Wet High-Intensity Magnetic Separation WRD Waste Rock Dump XRT X-Ray Transmission 1.6.3 List of Chemical Formulae Chemical Formula Mineral/Oxide (Li,Na)Al(PO4)(F,OH) – LiAl(PO4)(F,OH) Amblygonite-Montebrasite Al Aluminium Al2O3 Aluminium Oxide As Arsenic B Boron Be Beryllium Bi Bismuth Ca Calcium Ca(OH)2 Calcium Hydroxide Cd Cadmium Cl Chlorine Co Cobalt CO2 Carbon Dioxide Cs Cesium F Fluorine Fe Iron Fe2O3 Ferric Oxide FeO Iron Oxide H2O Dihydrogen Oxide/Water H2SO4 Sulphuric Acid HCl Hydrogen Chloride HNO3 Nitric Acid

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 15 Chemical Formula Mineral/Oxide K Potassium K(Li,Al)3(Al,Si,Rb)4O10(F,OH)2 Lepidolite K2O Potassium Oxide KLiFeAl(AlSi3)O10(OH,F)2 K(Al,Fe,Li)3(Si,Al)4O10(OH)F Zinnwaldite Li Lithium Li(Mn,Fe)PO4: LiFePO4 – LiMnPO4 Lithiophilite Li2CO3 Lithium Carbonate Li2O Lithium Oxide Li3PO4 Lithium Phosphate LiAl(PO4)(OH,F) – LiAl(PO4)F Montebrasite-Amblygonite LiAl(Si2O6) Spodumene LiAl(SiO4) Eucryptite LiAl4(Si3Al)O10(OH)8 Cookeite LiAlSi4O10 Petalite LiMnPO4 – LiFePO4 Lithiophilite-Triphylite LiOH Lithium Hydroxide LiOH·H2O Lithium Hydroxide Monohydrate Mg Magnesium MgO Magnesium Oxide Mn Manganese MnO Manganese Oxide Na Sodium Na0.3(Mg,Li)3Si4O10(OH)2 Hectorite Na2O Sodium Oxide NaAlSi2O6·H2O Analcime NaLi2.5Al6.5(BO3)3Si6O18(OH)4 Elbaite/Tourmaline NaOH Sodium Hydroxide Nb Niobium Ni Nickel P Phosphorus P2O5 Phosphorus Pentoxide S Sulphur Sb Antimony Si Silicon S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 16 Chemical Formula Mineral/Oxide SiO2 Silicon Dioxide SO4 Sulphate Ta Tantalum Ta2O5 Tantalum Pentoxide Zn Zinc Zr Zirconium S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 17 2 PROPERTY DESCRIPTION [§229.601(b)(96)(iii)(B)(3)] 2.1 Location and Area of the Property [§229.601(b)(96)(iii)(B)(3)(i)] The Project is in Central Ostrobothnia, western Finland, in the area of the municipalities of Kaustinen, Kokkola, and Kruunupyy, approximately 385 km north-northwest of Helsinki. The Keliber Lithium Project consists of mining operations around Kaustinen, the Keliber lithium concentrator at Päiväneva near Kaustinen, the Keliber lithium refinery at Kokkola, and ongoing exploration activities. There are nine elements to the Project, including: • Seven spodumene exploration or mining properties at Syväjärvi, Rapasaari, Länttä, Outovesi, Emmes, Leviäkangas, and Tuoreetsaaret; • The Keliber lithium concentrator at Päiväneva; and • The Keliber lithium refinery at the KIP. The Finnish national grid coordinate system used is ETRS-TM35FIN. The planned Keliber lithium refinery at KIP is centred at approximately N 7086600 E 306020, while the Päiväneva concentrator is centred on N 7060429, E 343076. The coordinates of the main spodumene deposits are as follows: • Syväjärvi N 7063218 E 341875 Mine/Exploration property • Rapasaari N 7061966 E 343691 Mine/Exploration property • Länttä N 7057934 E 358386 Mine/Exploration property • Outovesi N 7063902 E 338547 Mine/Exploration property • Emmes N 7065038 E 330803 Mine/Exploration property • Leviäkangas N 7060615 E 338135 Mine/Exploration property • Tuoreetsaaret N 7061929 E342665 Mine/Exploration property Figure 2-1 and Figure 2-2 show the location of the operations with local existing infrastructure. The maps presented contain data from the National Land Survey of Finland Topographic Database 06/2014. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 18 Locality Plan Figure 2-1: Locality plan.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 19 Mine and Exploration Sites Figure 2-2: Mine and exploration sites. 2.2 Ownership The mineral rights to the properties are held by Keliber Technology Oy, which is a wholly owned subsidiary of Keliber Oy, as shown in Figure 2-3. SSW hold an effective 79.82% ownership of Keliber Oy as at 31 December 2024. The balance of shares in Keliber Oy are held by the state-owned company Finnish Minerals Group (20%), who are tasked with managing the mining holdings of the Finnish state, and a group of Finnish shareholders (0.18%). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 20 Group Ownership Structure Figure 2-3: Group ownership structure. 2.3 Finland Regulatory Environment [§229.601(b)(96)(iii)(B)(2)(iv)] A brief overview of the regulatory environment in Finland within which Keliber operates and that affects Keliber is summarised below. 2.3.1 Constitution of the Republic of Finland The ultimate source of national law in Finland is the Constitution (731/1999 amended 2018), which defines the basis, structures, and organisation of government and the relationship between the different constitutional organs; it defines the fundamental rights of Finnish citizens and other individuals. Section 20 of the Constitution covers responsibility for the environment; this states that “everyone is responsible for nature, the environment, and the national heritage and that public authorities shall endeavour to guarantee for everyone the right to a healthy environment and for everyone the possibility to influence the decisions that concern their own living environment.”1 Act 505/2023 amended the Mining Act (621/2011) and prescribes how mining activities are conducted in order to meet the objectives of Section 20. This must be read along with the following pertinent legislation: • Finnish Government Decree on mining safety (1571/2011); • Finnish Government Decree on mining activities (391/2012); and • Finnish Government Decree on mine hoists (1455/2011). 1 https://www.refworld.org/pdfid/4e5cf5f12.pdf Keliber Lithium (Pty) Ltd Keliber Technology Oy Sibanye Stillwater Limited Sibanye Battery Metals (Pty) Ltd F I N L A N D R E P U B L I C O F S O U T H A F R I C A 100% 79.82% 100% Keliber Oy 100% Finnish Minerals Group Shareholders 20% 0.18% S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 21 2.3.2 The Mining Act All minerals are owned by the State of Finland. The objective of the Mining Act 505/2023 (Mining Act) is “to promote mining and organise the use of areas required for it, and exploration, in a socially, economically, and ecologically sustainable manner.”2 The Mining Act defines exploration and mining activities; the applicable permitting that is required for each and the validity and obligations thereof; the definition of a mining area and its establishment; the safety and supervision required on a mine; and the termination of mining and returning the possession of the mine when mining has ceased. The Finnish Safety and Chemicals Agency (Tukes) is responsible for issuing the relevant permits required for exploration and mining activities. Once a permit is granted, there is a 37-day period during which an appeal against the permit may be lodged with the Administrative Court. If no appeals are lodged, the permit then becomes legally valid. If an appeal is lodged, resolution of the appeal may delay operations by up to 18 months or longer should the appeal be escalated to the Supreme Administrative Court (SAC) (in that case, potentially up to 30 months). Any person, company or organisation may lodge an appeal, which is normally environmental-related (i.e. noise, pollution, dust, increased traffic, etc.). The permits are described below. 2.3.2.1 Exploration Permit This allows the holder to explore or prospect but not to exploit a deposit. The permit gives the holder the right to: • Conduct exploration; • Explore the structures and composition of geological formations; • Conduct other exploration in order to locate a deposit, investigate its quality, extent, and degree of exploitation; • Build, or transfer to the exploration area, temporary constructions and equipment necessary for exploration activity; and • Conduct other exploration in order to prepare for mining activity. An exploration permit is valid for a maximum period of four years and may be extended by three years until a maximum of fifteen years’ validity has been reached. Extension is dependent on the exploration having been effective and systematic, that all Mining Act obligations and all permit regulations have been complied with, that extension will not cause an undue burden to public or private interests, and that further research is required to confirm whether exploitation is possible. The permit holder has priority for being granted a mining permit. At all times, the property owner retains the right to use and govern the area. 2.3.2.2 Mining Permit A mining permit is required to establish a mine. Once a permit is granted, the permit holder has the right to: • Perform exploration within the mining area; 2 Ministry of Employment and the Economy, Finland. (2011) S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 22 • To exploit: − The minerals found within the mining area; − Any organic or inorganic surface materials, excess rock, and tailings generated as a by-product of mining activities; and − Other materials belonging to the bedrock and soil of the mining area, where their use is required for the mining operations. Permits are usually granted until further notice, but they may be granted for a fixed period of time; the validity of a fixed-term mining permit may be extended if mining has not yet commenced or if operations have been interrupted for a period of five years. The holder may apply to change the size of the area of a mining permit and may also assign the permit to another party. The holder must ensure that mining activities do not damage people’s health; do not impact public safety; do not cause significant harm to or infringe on public or private interests; do not obviously waste mining minerals and do not cause, prevent or hinder the potential future use and/or excavation work at the mine and the deposit. 2.3.2.3 Mining Safety Permit A mining safety permit is required to construct and operate a mine (Mining Act (505/2023) and Mining Safety content requirements as under Regulation (EU) No. 1571/2011). This covers the structural and technical safety of a mine, the prevention of hazards and accidents, and the mitigation of adverse effects of an accident. A mining permit must first be legally binding before a mining safety permit can be issued. 2.3.2.4 Surface Ownership It is not a prerequisite that the entity conducting the exploration and/or mining own the land on which the activities are taking place. However, if the land is privately owned, agreement must be reached with the owner before activities can commence. Conditions to such agreements must be determined and agreed upon by both parties and usually include some form of compensation. 2.3.3 Royalties and Payments Royalties that must be considered for the state of Finland are mined ore from Syväjärvi and Rapasaari mines due to their exploration licences being purchased from the Finnish state. In the agreement between the Government of Finland and Keliber concerning the Leviäkangas and Syväjärvi deposits (dated 19 October 2012) and the agreement between the Government of Finland and Keliber concerning the Rapasaari deposit (signed by the company on 22 October 2014), the following apply: • Keliber will pay EUR0.5 per tonne of ore (that is, the base rate) after the ore has been mined and treated to produce the products and those products sold; • The royalty for the Syväjärvi and Leviäkangas deposits is subject to the following price adjustment formula (Equation 1):

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 23 Equation 1: Price adjustment formula for the royalty for the Syväjärvi and Leviäkangas deposits. 𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑 𝑝𝑟𝑖𝑐𝑒 = (( Y Z ) × S(P) + (1 − S) × (P)) Where: • Z = Index for Base Period (January 2012) • Y = Index for December month preceding the year Royalty is calculated • S = Percentage of Price Subject to Adjustment (100%) • P = Base unit contract Price (EUR0.5) Royalty may be adjusted upward or downward based on the change in the Index from the base value to the December month value preceding the year for which the royalty is to be calculated. The base period for calculating the change will be from the January 2012 date of the Agreement. Royalty payment is payable annually and is made by the end of April the following year. • The royalty for Rapasaari is subject to the following price adjustment formula (Equation 2): Equation 2: Price adjustment formula for the royalty for the Rapasaari deposit. Adjusted price = (( Y Z ) × S(P) + (1 − S) × (P)) Where: • Z = Index for Base Period (September 2014) • Y = Index for December month preceding the year Royalty is calculated • S = Percentage of Price Subject to Adjustment (100%) • P = Base unit contract Price (EUR0.5) Royalty may be adjusted upward or downward based on the change in the index from the base value to the December month value preceding the year for which the royalty is to be calculated. The base period for calculating the change will be from the effective date of the Agreement. Royalty payment is payable annually and is made by the end of April the following year. Index means, in the Syväjärvi and Leviäkangas Agreement, the Producer Price Index (PPI) of the Industry (2000 = 100), and in the Rapasaari Agreement, the PPI of the Industry (2010 = 100). 2.3.3.1 Income Tax Income taxes are based on a company’s net income and are levied as prepayments during the fiscal year, which is generally the calendar year. If the company’s accounting year differs from the calendar year, taxes are based on the S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 24 accounting period or the accounting period that ends during the calendar year. Advance payments may be collected in two or twelve instalments during the year: • Two instalments: if the total tax is ≤EUR2,000, payments are made in the third and ninth months; and • Twelve instalments: if the total tax is >EUR2,000, payments must be made each month, by the 23rd of the month. Corporate income tax is currently at 20%. 2.3.3.2 Carbon Tax Finland introduced a carbon tax in 1990 based on the carbon content of fossil fuels. The average tax is currently EUR62.00 per tonne of CO2. 2.3.4 Environmental Legislation Finland has adopted a comprehensive regulatory framework on environmental issues. Although mostly regulated through national legislation, a large part of Finnish environmental legislation is from European Union (EU) law, either as directly applicable law or through the implementation of EU law. The key national legislation and main environmental regimes in Finland are shown in Table 2-1. Table 2-1: Key environmental legislation. Applicable Act Aspect Governed by the Act Environmental Protection Act (Ympäristönsuojelulaki). Prevention and control of pollution; prevention of generation of waste by certain activities; soil and groundwater conservation and remediation Waste Act (Jätelaki) General prevention of generation of waste and prevention of hazards and harm to human health and the environment Water Act (Vesilaki) Water resource management and control Nature Protection Act (Luonnonsuojelulaki) Nature and landscape conservation Act on Compensation for Environmental Damage (Laki ympäristövahinkojen korvaamisesta) Liability for environmental damage Act on Remediation of Certain Environmental Damage (Laki eräiden ympäristölle aiheutuneiden vahinkojen korjaamisesta) Remediation of damages to biodiversity and certain water systems Act on Environmental Impact Assessment Procedure (Laki ympäristövaikutusten arviointimenettelystä) Environmental impact assessment (EIA) Act on Environmental Impact Assessment of Plans and Programmes of the Authorities (Laki viranomaisten suunnitelmien ja ohjelmien ympäristövaikutusten arvioinnista) EIA concerning certain plans and programmes Land Use and Building Act (Maankäyttö- ja rakennuslaki) Land use and planning Emission Trading Act (Päästökauppalaki) Emissions trading Act on the Use of the Kyoto Mechanisms (Laki Kioton mekanismien käytöstä) Emissions trading Land Extraction Act (Maa-aineslaki). Use and control of certain natural resources S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 25 Applicable Act Aspect Governed by the Act Mining Act (Kaivoslaki) Use and control of mining resources Forest Act (Metsälaki) Use and control of forest resources Chemical Act (Kemikaalilaki)x Use and control of forest resources Gene Technology Act (Geenitekniikkalaki) Genetic engineering Nuclear Energy Act (Ydinenergialaki Nuclear power Act on Operating Aid for Power Generation from Renewable Energy Sources (Laki uusiutuvilla energialähteillä tuotetun sähkön tuotantotuesta) Renewable energy/feed-in tariffs Radiation Act (Säteilylaki) Radiation control 2.3.4.1 Key Regulatory Authorities The main body that develops environmental policy and drafts environmental legislation is the Ministry of the Environment. Other relevant ministries with adjacent competencies are the: • Ministry of Employment and the Economy, which handles policy issues concerning mining and energy (including renewable energy); and • Ministry of Agriculture and Forestry, which handles policy issues concerning the use of water and forest resources. There are several competent authorities that enforce environmental legislation. Generally, the competent supervisory authorities are the regional Centres for Economic Development, Transport and the Environment (Elinkeino-, liikenne- ja ympäristökeskus) (ELY-keskus), and the municipalities. The competent permitting authorities for environmental permits (EPs) are the Regional State Administrative Agencies (Aluehallintovirasto) and the municipalities. There are six Regional State Administrative Agencies (AVIs) in Finland, of which four issue EPs. The AVIs operate as state permit authorities according to the Water Act and the Environmental Protection Act. Under the Environmental Protection Act, they are responsible for processing EP applications for projects that have a major impact on the environment, and they handle all permit applications under the Water Act. The AVIs for Western and Inland Finland are responsible for matters related to EPs for Keliber. 2.3.4.2 Environmental Permitting The Environmental Protection Act governs an integrated permit regime for emissions into air, water, and/or soil and the generation of waste. However, an EP does not necessarily cover all activities on site or even all emissions from the site/operations. Under certain circumstances, the permit process for a water permit under the Water Act is integrated with the permit process for an EP. 2.3.4.3 Environmental Impact Assessments An Environmental Impact Assessment (EIA) must be performed for projects if: S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 26 • The project type is listed in the Environmental Impact Assessment Decree, which contains a list of projects (industrial and construction) that are deemed to have considerable environmental impacts. • The competent authority decides that an EIA must be performed due to the considerable environmental impact of the project, even if the project is not included in the Decree. In addition to the general EIA legislation that applies to projects, public authority plans and programmes also require an EIA under certain circumstances. The most important ones are listed in a separate government Decree. For planning decisions, municipalities are responsible for assessing the environmental impact of the plan under the land use and planning legislation. In addition to typical environmental impacts, impacts on the local economy must also be assessed. If a project or plan may affect the nature conservation values of a Natura 2000 nature conservation site, the impact must be evaluated before the project or plan can be carried out. 2.3.5 Summarised Permitting Process The simplified permitting process for a mining project in Finland is presented in Figure 2-4. Mining Projects’ Permitting Process in Finland Figure 2-4: Simplified permitting process for mining projects in Finland. Key laws and regulations relevant to the Keliber operations include: • Mining legislation including the Mining Act (505/2023); Operation Tu ke s M u n ic ip al it y C en tr e o f Ec o n o m ic D ev el o p m en t, Tr an sp o rt a n d t h e En vi ro n m en t R eg io n al S ta te A d m in is tr at iv e A ge n cy Fi n la n d G o ve rn m en t BuildingPlanning and Engineering Exploration permit Mining Act Mining Safety permit Mining Act Mining permit Mining Act Chemical permit Chemical Safety Act Land use plan upgrade (Regional Planning, Local master plan, Local Detailed plan) Land Use and Building Act Environmental Impact Assessment (EIA) Act on Environmental Impact Assessment procedure Considered in permitting process No complaint possibility Building Permit Land Use and Building Act Dam safety documentatio n assessment Dam Safety Act Environmental and Water Permit Environmental Protection Act Exploration Permits for protected areas The right of use of the mining area Mining Act Permit for producing Uranium and Thorium (if needed)

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 27 • Environmental Protection Act (527/2014) including the Environmental Protection Decree (713/2014), the Water Act (587/2011), and the Environmental Impact Assessment Procedure Act (252/2017); • Dam Safety Act 494/2009; • Chemical legislation including Chemical Act (599/2013), Act on the Safe Handling and Storage of Dangerous Chemicals and Explosives (390/2005), and the Chemical Safety Act (390/2005); • Government Decree on Extractive Waste (190/2013 as amended); • Waste Act (646/2011) and Waste Decree (179/2012); • Nature Conservation Act (1096/1996)/Natura 2000 (Appropriate Assessment); • Fire safety legislation including Rescue Act (379/2011); • Land use and building legislation including the Land Use and Building Act (132/1999); • Air Pollution Control Decree (79/2017); • Decree on the Safe Production and Handling and Storage of Explosives (1101/2015); and • Forest Act (1093/1996). 2.4 Mineral Deposit Tenure [§229.601(b)(96)(iii)(B)(3)(ii)-(iv)] SSW have confirmed to VBKOM that all legal information in this TRS is correct and valid and that the Company in which it has the shareholding (Keliber) has title to the mineral rights and surface rights for the Keliber Lithium Project through its subsidiary Keliber Technology Oy. Exploration and mining permits are issued to, or under application by, the operating company Keliber Technology Oy in terms of Section 34 of the Mining Act. All permits (or applications) are for the commodity lithium. There are exclusion zones or “protection circles” with a diameter of 150 m around residential structures or working farms within the tenement boundaries upholding a requirement of the Mining Act in Finland. Compensation to the landowners, according to the Mining Act, applies to all legally valid mining and exploration permits; compensation for all permit applications or the granted exploration permits will only become due once the permits are legally valid. Keliber hold four valid mining permits located in Syväjärvi, Länttä, and Rapasaari, totalling 712.71 ha. These are listed in Table 2-2. An additional 27 valid exploration permits are held over a total area of 6,738.31 ha, as listed in Table 2-3. Table 2-2: Valid mining permits as at 31 December 2024. Holder Permit Name Permit ID Area (ha) Decision Date Expiry Date Mining Permit Keliber Technology Oy Syväjärvi KL2018:0001 166.30 2018/12/13 2029/01/14 Keliber Technology Oy Syväjärvi (apualue)* KL2021:0003 19.95 2022/02/08 2032/03/18 Keliber Technology Oy Länttä 7025 / KL2016:0002 / KL2021:0002 37.49 2022/02/11 2027/03/21 Keliber Technology Oy Rapasaari KL2019:0004 488.97 2022/03/23 2033/12/22 Total Area 712.71 *Syväjärvi (apualue) is the auxiliary area of the Syväjärvi mining area. Table 2-3: Valid exploration permits as at 31 December 2024. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 28 Holder Permit Name Permit ID Area (ha) Date of Decision Date of Expiry Exploration Permit Keliber Technology Oy Roskakivi ML2016:0020-01 227.18 2021/07/30 2025/09/06 Keliber Technology Oy Haukkapykälikkö ML2011:0002-02 350.32 2021/07/30 2026/07/20 Keliber Technology Oy Pässisaarenneva ML2018:0040-01 22.53 2021/07/30 2027/07/20 Keliber Technology Oy Emmes 1 ML2015:0031-02 19.86 2021/11/01 2026/12/18 Keliber Technology Oy Rytilampi ML2011:0020-02 163.21 2023/09/07 2026/09/06 Keliber Technology Oy Palojärvi ML2018:0091-02 35.55 2024/02/08 2027/03/18 Keliber Technology Oy Kokkoneva ML2018:0055-01 278.61 2024/02/08 2028/03/18 Keliber Technology Oy Länkkyjärvi ML2018:0036-01 361.56 2024/03/14 2028/04/22 Keliber Technology Oy Peikkometsä ML2018:0023-01 770.04 2024/03/14 2028/04/22 Keliber Technology Oy Valkiavesi ML2018:0031-01 860.10 2024/03/14 2028/04/22 Keliber Technology Oy Vehkalampi ML2018:0022-01 1,123.89 2024/03/14 2028/04/22 Keliber Technology Oy Hassinen ML2018:0034-01 300.39 2024/05/17 2028/06/24 Keliber Technology Oy Hyttikangas ML2018:0035-01 238.08 2024/05/17 2028/06/24 Keliber Technology Oy Keskusjärvi ML2018:0033-01 211.08 2024/05/17 2028/06/24 Keliber Technology Oy Matoneva ML2018:0041-01 507.37 2024/05/17 2028/06/24 Keliber Technology Oy Peuraneva ML2018:0032-01 152.67 2024/05/17 2028/06/24 Keliber Technology Oy Östersidan ML2018:0056-01 197.23 2024/05/17 2028/06/24 Keliber Technology Oy Heikinkangas ML2012:0156-02 42.55 2024/06/14 2027/06/13 Keliber Technology Oy Karhusaari ML2012:0157-03 137.91 2024/06/14 2027/06/13 Keliber Technology Oy Leviäkangas 1 ML2013:0097-03 90.70 2024/06/14 2027/06/13 Keliber Technology Oy Outovesi ML2018:0089-02 157.89 2024/06/14 2027/06/13 Keliber Technology Oy Paskaharju ML2016:0044-02 131.71 2024/06/14 2027/06/13 Keliber Technology Oy Päiväneva ML2012:0176-03 52.02 2024/06/14 2027/06/13 Keliber Technology Oy Rapasaari ML2018:0121-02 64.90 2024/06/14 2027/06/13 Keliber Technology Oy Syväjärvi 2 ML2016:0001-02 71.53 2024/06/14 2027/06/13 Keliber Technology Oy Syväjärvi 3-4 ML2018:0120-02 115.75 2024/06/14 2027/06/13 Keliber Technology Oy Timmerpakka ML2019:0010-02 53.68 2024/06/14 2027/06/13 Total Area 6,738.31 Furthermore, Keliber have submitted exploration permit applications for 11 additional areas covering a total of 3,026.59 ha. These areas are listed in Table 2-4. Table 2-4: Exploration permits under application as at 31 December 2024. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 29 Holder Permit Name Permit ID Area (ha) Date of Application Keliber Technology Oy Kellokallio ML2019:0032-01 177.63 2019/04/27 Keliber Technology Oy Peräneva ML2024:0007-01 622.01 2024/03/25 Keliber Technology Oy Buldans ML2020:0001-01 97.98 2020/01/16 Keliber Technology Oy Ruskineva ML2020:0002-01 737.90 2020/01/17 Keliber Technology Oy Timmerpakka 2 ML2020:0025-01 171.36 2020/04/23 Keliber Technology Oy Orhinselkä ML2018:0042-01 222.05 2018/05/08 Keliber Technology Oy Vanhaneva ML2019:0002-01 343.55 2018/09/27 Keliber Technology Oy Arkkukivenneva ML2021:0045-01 83.78 2021/03/31 Keliber Technology Oy Outovedenneva ML2011:0019-03 68.75 2024/07/12 Keliber Technology Oy Emmes 2 ML2019:0052-02 57.80 2024/07/12 Keliber Technology Oy Outoleviä ML2019:0011-02 443.78 2024/07/12 Total Area 3,026.59 The permit and application areas are illustrated in Figure 2-5. Permit Areas Figure 2-5: Permit areas. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 30 2.5 Surface Rights Keliber have ownership of the land area of 41.73 ha in Outovesi, which covers approximately 20% of the current claim areas of Outovesi (209.67 ha). In Syväjärvi, Keliber own a land area of 47.39 ha which corresponds to approximately 28% of the current Syväjärvi mining area (166.3 ha). Furthermore, as a part of proceedings establishing the mining area, Keliber have purchased the majority (82%) of a block of lands at Rapasaari: 400.15 ha of 488.97 ha. Keliber will conduct mining activities in land areas owned by private landowners. The establishment of a mine and undertaking of mining activity are subject to a permit (mining permit). When a permit is granted, it entitles the permit holder to exploit: • The mining minerals found in the mining area; • The organic and inorganic surface materials, excess rock, and tailings generated as a by-product of mining activities (by-product of mining activity); and • Other materials belonging to the bedrock and soil of the mining area, insofar as the use thereof is necessary for the purposes of mining operations in the mining area. Moreover, the mining permit entitles Keliber as a permit holder to perform exploration within the mining area. As compensation to landowners, Keliber will pay annual compensation (excavation fee related to Mining Act 505/2023) to the owners of land included in the mining area. The other specified compensations are agreed upon with the landowners and confirmed officially in the mining area establishment process. All mine sites have road connections (mainly forestry roads) and accessibility to Highway 63, which crosses the area. The land area of the Keliber lithium refinery (125,149 m2) is leased from Kokkolan Energia Oy with a fixed-duration agreement until 31-12-2049, after which the tenancy will continue as valid until further notice. An option for the lease of an additional 33,589 m2 area is included in the agreement. 2.6 Environmental Authorisations Keliber have completed all relevant EIA procedures to proceed with the Project. Keliber hold a valid EP for the Syväjärvi mining operations and a water permit for dewatering Lake Syväjärvi and Lake Heinäjärvi. A valid permit states that the permit decision issued by AVI was appealed, and appeals were processed in the Vaasa Administrative Court (VAC). The Court ruled against appeals and kept AVI’s permit decision in force on 16 June 2021. There were no appeals made to the SAC against the VAC Decision. The Syväjärvi EP became final in July 2021. The environmental permitting of the sulphidic waste rock dump (WRD) to Syväjärvi commenced in October 2024. The target is to permit a storage facility for sulphidic mica schist waste rock in the close vicinity of an open pit to avoid long-distance hauling of the material to Kokkola Harbour, which has the EP to receive the material. The Rapasaari mine EP application was submitted to AVI on 30 June 2021. The Päiväneva concentrator EP was submitted to AVI on 30 June 2021. Concentrator operations require a water permit for raw water intake from Köyhäjoki River, which was included in the EP application. The permit decisions (Environmental permit 208/2022 number: LSSAVI/10481/2021, LSSAVI/10484/2021) from AVI were received on 28 December 2022. For the Keliber lithium refinery located in Kokkola, EPs are in place to start the operation. The EP was approved on 28 June 2022. The EP for Keliber lithium refinery was not appealed and is, therefore, legally valid. There are no known encumbrances to the Keliber lithium refinery.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 31 The EP status is summarised in Table 2-5. Table 2-5: Environmental permitting status of Keliber Lithium Project. Operation Permit No. Receiving Date Status Mining operations Syväjärvi Mine 36/2019 20/02/2019 Legally valid Päiväneva Concentrator 208/2022 206/2024 28/12/2022 23/02/2024 Legally valid, however, magnetic waste stream management needs to be re-permitted (application submitted for authority approval on 30 May 2024) Rapasaari Mine 208/2022 206/2024 28/12/2022 23/02/2024 Legally valid, however, WRD areas need to be re- permitted prior to operational phase being initiated (submission of the application scheduled on for early 2025) Keliber lithium refinery Lithium refinery in KIP 122/2022 28/06/2022 Legally valid 4/2025 14/01/2025 Appeal period ongoing until 20 February 2025 - Legally valid 2.7 Property Encumbrances and Permitting Requirements [§229.601(b)(96)(iii)(B)(3)(v)] The mining and exploration permits and applications related to the Project Areas listed in Table 2-2 to Table 2-4. The EP status is presented in Table 2-5. Additional permits required and their current status are presented in Table 2-6. It should be noted herein that Länttä, Outovesi, and Emmes need to be re-permitted, and all required infrastructure needs to be constructed prior to the operational phase. Thus, these are not short-term deposits to be utilised. Table 2-6: Permitting situation of Keliber Project as of 31 December 2024. Production Site Permit Status Date Syväjärvi Mine Environmental and water permit Valid 20/02/2019 Exception permit to moor frogs Valid 01/02/2020 Exception permit for diving beetles Valid 21/07/2020 Mining permit Valid 13/12/2018 The right of use of the mining area Valid 09/08/2021 Mining safety permit Valid 13/10/2021 Rapasaari Mine Environmental and water permit Valid, partly returned for additional permitting 28/12/2022 Mining permit Valid 23/03/2022 The right of use of the mining area Valid Not available Mining safety permit Not started - S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 32 Production Site Permit Status Date Länttä Mine Environmental and water permit Old permit has to be renewed; NPermitting not started yet - Mining permit Valid 23/05/2006 Mining safety permit Not started - Outovesi Mine Environmental and water permit Permitting not started - Mining permit Permitting not started - Mining safety permit Permitting not started - Päiväneva Concentrator Environmental and water permit Valid, partly returned for additional permitting 28/12/2022 Mining permit (included in Rapasaari mining area) Valid 23/03/2022 Building permit Valid Not available Chemical permit In progress - Keliber lithium refinery EP Environmental and water permit Valid 28/06/2022 Building permit Approved 02/08/2024 Chemical permit In Progress - Keliber’s operations will be governed by framework legislation that includes several laws, acts, decrees, and permits. The legislation and permits that steer Keliber operations are listed in Keliber’s compliance register. To the knowledge of the QPs, there are no known violations directed at or fines issued to Keliber for the subject properties. 2.8 Significant Factors and Risks Affecting Access, Title [§229.601(b)(96)(iii)(B)(3)(vi)] The combined Rapasaari and Päiväneva Environmental permit was appealed by other parties, and the appeal was finalised with the VAC decision on 23 February 2024. The VAC returned the Päiväneva concentrator magnetic waste stream case for additional permitting, which was initiated on 30 May 2024. Magnetic waste stream permitting is currently in the hearing phase. In addition, the VAC returned Rapasaari WRDs for additional permitting. The permit application is scheduled to be submitted for authority approval after the hearing of magnetic waste stream permitting. The permitting authority (AVI) has indicated a maximum of 12 months of processing time for additional permitting. In addition, Keliber have applied enforceable decisions to start the permitted operations, regardless of appeal. The environmental permitting status of the SSW Keliber Lithium Project is presented in Table 2-6. Mining operations will be ramped up in the Syväjärvi and Rapasaari mining lease areas. The Päiväneva concentrator is located in the Rapasaari mining lease area. The Päiväneva concentrator and Rapasaari mining operations can be initiated when the required permitting processes have been finalised (Table 2-6). Operations can be initiated with an enforcement order. The likelihood that the decision of the AVI will be appealed is high in Finland. Thus, the appeals process can cause delays to the start of the operations if, for example, the enforcement order is revoked by the VAC. Operational delays S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 33 are estimated to be up to 18 months. Appeals may be extended to the SAC, in which case the delays could be extended for a further 12 months. Keliber have not identified any material risks relating to access to the sites or title to the mining lease areas. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 34 3 ACCESSIBILITY, CLIMATE, LOCAL RESOURCES, INFRASTRUCTURE, AND PHYSIOGRAPHY [§229.601(b)(96)(iii)(B)(4) 3.1 Topography, Elevation, and Vegetation [§229.601(b)(96)(iii)(B)(4)(i)] The average altitude for Central Ostrobothnia is 75 mamsl; the topography of the Project area is relatively flat with the total difference in elevation between the various sites being in the order of 40 m. The lowest site is Rapasaari at 82.7 mamsl, while the highest is Länttä at 122.0 mamsl. The Perhonjoki River flows north-northeast through the area, decanting into the Gulf of Bothnia north of Kokkola. Numerous streams and lakes of all sizes occur throughout the area. The land is cultivated, especially along the river courses, with most of the remaining land covered with forest. There is no permafrost at these latitudes. The overburden thickness at the mine sites varies in thickness from zero at Syväjärvi and Länttä to 20 m at Rapasaari: • Syväjärvi: 0–10 m; • Rapasaari: 4–20 m; • Länttä: 0–8 m; • Outovesi: 7–13 m; • Leviäkangas: to be determined; and • Tuoreetsaaret: to be determined. 3.2 Access to the Properties [§229.601(b)(96)(iii)(B)(4)(ii)] The Keliber lithium refinery (Conversion plant and Hydrometallurgical plant) is located in the city of Kokkola in Central Ostrobothnia, Western Finland. It is located approximately 6 km from the city centre of Kokkola in the KIP. The Keliber lithium refinery has excellent road and railway connections, and it is located 2 km from the port of Kokkola on the Gulf of Bothnia; the road and rail links between the two are good. Kokkola and Kaustinen are connected by national road 13 and are approximately 46 km apart. Kokkola-Pietarsaari Airport is approximately 13 km south of the city and is serviced by regular Finnair flights as well as charter flights. The Päiväneva concentrator and the proposed mining areas are located to the north, northeast, and east of the city of Kaustinen in the municipalities of Kruunupyy, Kokkola, and Kaustinen in the Central Ostrobothnian region. KIP and the concentrator are approximately 68 km apart. The various mine sites are located close to the Päiväneva concentrator; distances and directions are given from the concentrator site: • Syväjärvi (Kokkola and Kaustinen municipalities) – 3 km north-northeast; accessible via paved national road 63 and gravel forestry road;

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 35 • Rapasaari (Kokkola and Kaustinen municipalities) – 1.5 km northeast; accessible via paved national road 63 and gravel forestry road; • Länttä (Kokkola municipality) – 25 km east-southeast; accessible via paved national road 63 and local road 18097 (approximately the first two kilometres are gravel); • Outovesi (Kaustinen municipality) – 10 km northwest; accessible via paved national road 63 and gravel forestry road; • Emmes (Kruunupy municipality) – 20 km west-northwest; accessible via gravel forestry road, paved national road 63, Emmeksentje road and gravel local road 17947; • Leviäkangas; (Kokkola and Kaustinen municipalities) – 4.5 km northwest; accessible via paved national road 63 and gravel forestry road; and • Tuoreetsaaret (Kokkola and Kaustinen municipalities) – 1.5 km northeast; accessible via paved national road 63 and gravel forestry road. 3.3 Climate and Length of Operating Season [§229.601(b)(96)(iii)(B)(4)(iii)] The climate in Finland is a so-called intermediate climate, combining characteristics of both a maritime and a continental climate. Finland's mean temperature is several degrees higher than in most other continental areas located at the same latitudes. The annual average temperature in the Central Ostrobothnian area is above 3˚C. The coldest time of the year is typically in January or February. The average temperature in February is typically minus 6˚C to minus 8˚C. The warmest time of the year occurs, on average, in July, with the average temperatures above 16˚C. Therefore, no break due to weather is foreseen for operations. The annual precipitation in Central Ostrobothnia varies between 500 mm and 600 mm. The months of February, March, April, and May see the least precipitation, while the precipitation increases towards summer with August typically the wettest month. In Central Ostrobothnia, the number of days with snow cover varies between 110 to 155 days. Snow cover is deepest in late winter, typically in early March, being 300 mm to 400 mm. The windier part of the year is from September to March, with the windiest month being December and the least windy being July. Daylight hours vary from four hours in December/January to 20 hours in June/July. Typically, in Nordic countries, operations continue in subarctic conditions at temperatures below -20°C. It is thus expected that Keliber will operate continually during the year. 3.4 Local Resources and Infrastructure [§229.601(b)(96)(iii)(B)(4)(iv)] 3.4.1 Kokkola Central Ostrobothnia is a province in Western Finland with a surface area of 6,462.96 km². The Central Ostrobothnia province has a population of about 69,000 inhabitants with a population density of 13.7 inhabitants per km2. Kokkola is the largest city of Central Ostrobothnia with approximately 48,000 inhabitants (33.02 inhabitants per km2). The municipality of Kaustinen has about 4,300 inhabitants (12.15 inhabitants per km2). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 36 Kokkola has two institutes offering higher education: the Kokkola University Consortium Chydenius and the Centria University of Applied Sciences. Chydenius has a department of applied chemistry and high-level research in the chemistry of materials including lithium-ion battery materials. Centria offers several Bachelor’s degree programmes e.g. Bachelor’s degree programme in environmental chemistry and technology. In Kokkola, there is also the Federation of Education in Central Ostrobothnia, consisting of seven vocational schools, and an adult education unit which arranges vocational upper secondary education in the region e.g. in process technology. The Keliber lithium refinery is located in the KIP area, which has a significant concentration of chemical industry installations in northern Europe. Seventeen (17) industrial operators and more than 60 service companies are located in the KIP area. The companies on-site directly employ 2,300 people. The KIP area has 700 ha of land zoned for use by the heavy chemical industry. Through the service enterprises in the area, companies have use of, among others, commodity and sewage networks, pipe bridges, railways, a factory fire brigade, and security. In the KIP area, the Keliber lithium refinery will be immediately adjacent to several important resources such as water, steam, electricity, heat, gas (e.g. CO2), and acids (e.g. sulfuric acid), all of which are produced in the KIP area. The Port of Kokkola, which is the third-largest general port in Finland, is located 2 km from the Keliber lithium refinery and 66 km from the Päiväneva site. The Port of Kokkola is the largest port serving the mining industry in Finland. The Port includes general port facilities, mainly for containers, breakbulk cargoes, and so-called light bulk such as limestone. The Port has an all-weather terminal, mainly for containers and breakbulk cargo. The Port of Kokkola also has a deep port for handling bulk cargoes. Central Ostrobothnia is serviced by Kokkola-Pietarsaari Airport which offers both traffic connections and a setting for active general aviation. 3.4.2 Kaustinen Potable water is available from the Kaustinen municipality water pipeline and the Pirttikoski hydroelectric power plant on the Perhonjoki River. The Perhonjoki River, in Kaustinen, supplies power to the main 110 kV power line. The area is also serviced by mobile phone networks from all the main Finnish service providers, as well as a fibre optic network from a local service provider. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 37 4 HISTORY [§229.601(b)(96)(iii)(B)(5)] 4.1 Previous Operations, Operators [§229.601(b)(96)(iii)(B)(5)(i)] None of the properties have previously been mined, although the mining rights to the Länttä, Emmes, and Syväjärvi deposits were first owned by Suomen Mineraali Oy, then by Paraisten Kalkkivuori Oy and, from the early 1960s to the early 1980s, by Partek Oy. These rights expired in 1992, and the areas were unclaimed until 1999 when Olle Siren, together with private partners, claimed the Länttä deposit and later the Emmes deposit (Table 4-1). From 2003 to 2012, the Geological Survey of Finland (GTK) held ownership of the Syväjärvi and Rapasaari deposits. Table 4-1: Previous operators. Deposit Date Operator Länttä, Emmes, Syväjärvi, Leviäkangas 1960–1968 Suomen Mineraali Oy 1963–1999 Paraisten Kalkkivuori Oy (later Partek Oy) All 1992–1999 Unclaimed Länttä 1999 Olle Siren and private partners Emmes After 1999 Olle Siren and private partners Syväjärvi, Leviäkangas, Rapasaari 2003–2012 GTK Länttä, Emmes, Rapasaari, Syväjärvi, Outovesi, remaining exploration areas Not available Keliber (previously known as Keliber Resources Ltd.) Tuoreetsaaret 2020–2022 Keliber (previously known as Keliber Resources Ltd.) Note: Paraisten Kalkkivuori Oy acquired Suomen Mineraali Oy in 1959; both companies operated in the same lithium-potential area under the same umbrella. 4.2 Exploration and Development Work [§229.601(b)(96)(iii)(B)(5)(ii)] Since the discovery of spodumene and beryl mineralisation in the Kaustinen region in the late 1950s, the area began to see systematic exploration being initiated in the 1960s by Suomen Mineraali Oy and Paraisten Kalkkivuori Oy. Due to the lack of outcrop throughout most of the area, surface exploration methods were restricted to spodumene/pegmatite boulder hunting. These results were then used to delineate the source of origin for the boulder fans using palaeo-glacial directions. Apart from the Länttä deposit (discovered as outcrop), this method proved highly successful in the discovery of the Emmes and Syväjärvi deposits by early operators. Between 2003 and 2012, GTK were also very active in the area, with exploration work including boulder mapping, geophysical surveys, till sampling, re-analysis of historical regional till samples, percussion drilling, and diamond core drilling. This work was successful in the discovery of the Rapasaari deposit as well as further delineation of the Syväjärvi S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 38 deposit. Keliber’s involvement in the Project began in 1999 when a group of investors led by Mr Olle Siren began the evaluation of the Länttä deposit, where drilling commenced in 2004. Keliber then extended their exploration efforts to the rest of the Kaustinen region where it completed acquisition of exploration rights and extensive drilling programmes over all of the deposits including the Outovesi deposit which was discovered in 2010 and Toureetsaaret in 2020. GTK carried out an extensive areal geochemical till sampling programme covering the whole of Finland during the 1970s and 1980s. At that time, no analysis for lithium was conducted. Later, GTK re-analysed the old till samples and large geochemical anomalies were discovered in the Kaustinen area. Some of the known deposits are reflected in lithium anomaly maps, but spotty anomalies extend far outside the known deposits, especially to the northwest (WSP Global Inc., 2022). During 2004–2011, GTK carried out 15.5-line kilometres of gravity survey and 4.4 km2 of gravity and magnetic ground geophysical surveys in seven different exploration areas (Table 4-2). A slingram survey was also conducted at Rapasaari. Ground geophysics was surveyed to support geological mapping and to define the borders of the spodumene pegmatites. High-resolution, low-altitude airborne geophysics data for 2004 were also used (Ahtola et al., 2015). The pre-2004 exploration results are limited, and historical sampling data and exploration were not directly for Lithium. Results, therefore, are not considered relevant for this TRS. Table 4-2: Summary of the sampling and ground geophysics (after Ahtola et al., 2015). Drilling Target Period No. of Diamond Drill Holes Ground Geophysics Till Sampling (No. of Samples) RC Drilling (No. of Samples) Total Length (m) Line (km/km2) Method* Leviäkangas 2004–2008 22 2,032 1 km2 mg + gr 60 Syväjärvi 2006–2010 24 2,547 1 km2 mg + gr 56 Rapasaari 2009–2012 26 3,653 2.2 km2 mg + sl + gr 508 TOTAL – 72 8,232 4.4 km2 508 116 * mg = magnetic; sl = slingram; gr= gravity The first drilling programmes were undertaken by Suomen Mineraali Oy in 1961 and were executed using small drill rigs. From 1966 to 1981, a core diameter of 32 mm was used by Suomen Mineraali Oy and Partek Oy. These small- diameter-drilling programmes were executed at Emmes, Länttä, Leviäkangas, and Syväjärvi in the 1960s, 1970s, and at the beginning of the 1980s (WSP Global Inc., 2022). The historical drilling activities undertaken by these operators are summarised in Table 6-1, along with the work undertaken under the ownership of Keliber Oy.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 39 5 GEOLOGICAL SETTING, MINERALISATION, AND DEPOSIT [§229.601(b)(96)(iii)(B)(6) 5.1 Regional, Local, and Project Geology [§229.601(b)(96)(iii)(B)(6)(i) (ii)] The Keliber Project is located in the Kaustinen Lithium Pegmatite (KLP) Province of western Finland, covering an area of about 500 km2 (SRK, 2023 and references therein). The underlying geology comprises Palaeoproterozoic (1.95–1.88 Ga) supracrustal rocks of the northern Pohjanmaa Belt (also referred to as the Ostrobothnia Schist Belt) which forms a 350 km long and 70 km wide arcuate belt between the Central Finland Granitoid Complex to the east (Vaasjoki et al., 2005) and the Vaasa Granitoid Complex in the west (Aviola et al., 2001). The supracrustal rocks of the Pohjanmaa Belt, comprising micas schists/metasediments, gneisses, and metavolcanics (ranging from felsic and intermediate to mafic in composition), metamorphosed to grades ranging from lower to upper amphibolite facies and locally granulite facies conditions 1.89–1.88 Ga ago. Metamorphic grades are lowest in the central and eastern parts of the belt and increase towards the Vaasa Migmatite Complex and southwards in the southern part of the belt (Figure 5-1) (Alviola et al., 2001). The Pohjanmaa Belt is host to several pegmatite groups/provinces, ranging from dominantly complex pegmatites (which may contain spodumene) in the southeast, beryl/beryl-columbite(+phosphate)-type pegmatites in the central regions, beryl-(topaz/andalusite)-bearing pegmatites along the southern margin of the Vaasa Complex, and several Lithium-Caesium-Tantalum (LCT), albite-spodumene-type (Cerny and Ercit, 2005) pegmatites of the Kaustinen Province in the north (Figure 5-2) (Aviola et al., 2001 and Ahtola et al., 2012). The pegmatites of the KLP, dated at 1.79 Ga, were intruded into the Pohjanmaa metasediments just after peak regional metamorphism, with the source rocks of the pegmatites considered to be the contemporaneous large pegmatitic granites and granites found within the Kaustinen region (Figure 5-1 and Figure 5-3). Approximately ten individual pegmatite deposits have been discovered to date within the KLP, with most having subsequently been explored exclusively by drilling methods due to the paucity of outcropping pegmatites and host rocks, most of which are covered by 3–18 m of overburden comprising surficial sediments (mostly glacial till) (Ahtola et al., 2015). The majority of the pegmatites have been intruded at high angles or subparallel to host supracrustal rock foliations. Most pegmatites have a similar mineralogy, dominated by feldspar, quartz, spodumene, and muscovite. The pegmatites are generally poorly zoned, often with a variably developed outer border and marginal zone of quartz- feldspar-muscovite (with little to no spodumene mineralisation) and a mineralised core of quartz-feldspar-spodumene (± muscovite). Historical exploration comprised identification and mapping of spodumene-bearing pegmatite boulders and, supported by more recent drilling by GTK and Keliber (ongoing), has resulted in the delineation of seven discrete LCT pegmatite deposits, namely Syväjärvi, Rapasaari, Länttä, Emmes, Outovesi, Tuoreetsaaret, and Leviäkangas (Figure 5-3). Each of the deposits is characterised by a series of pegmatite veins and dykes with an intrusion geometry controlled by regional structural controls as well as host rock rheology contrasts. The ubiquitous overburden, comprising till and sediments covers most of the region. Project and regional scale geological maps, stratigraphic columns, and regional geological cross sections are not available. However, detailed drilling by both GTK and Keliber has been able to delineate most of the larger individual pegmatites to a relatively high level of confidence. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 40 It is noted that it is an SEC requirement to include a stratigraphic column and regional geological cross-section of the Project area/s. The intrusion type and style of deposit being considered i.e. vein pegmatite and dyke intrusions, means that the inclusion of a stratigraphic column and regional geological cross-section in this TRS are not considered relevant nor would they provide any real technical guidance within the context of the Project geological setting being described in this TRS. Regional Geological Map of the Kaustinen Lithium Province within the Pohjanmaa Belt Figure 5-1: Regional geological map of the Kaustinen Lithium Province within the Pohjanmaa Belt (source: SRK, 2023 – modified after Ahtola et al., 2015). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 41 Map showing the Location of the Various Pegmatite Groups within the Broader Pohjanmaa Figure 5-2: Map showing the location of the various pegmatite groups within the broader Pohjanmaa Belt. Note: The yellow box around the Kaustenin Province in the north is host to the albite-spodumene pegmatites that form part of the Keliber Project. The complex pegmatite group (red) may also host lithium mineralisation (source: Alviola et al., 2001). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 42 Geology of the Keliber Project showing the Map of Pegmatite Deposits Figure 5-3: Geology of the Keliber Project showing the map of pegmatite deposits (source: SRK, 2023 (modified after Ahtola et al., 2015)). 5.1.1 Syväjärvi Geology The Syväjärvi lithium pegmatite deposit (Figure 5-3) is overlain by an average of 8 m (ranging from <1 m to approximately 20 m) of sandy till cover (SRK, 2023 and reference therein)(refer to Figure 5-4). Outcrop within the Project is limited to an isolated exposure of a host rock comprising plagioclase porphyrite (porphyritic metavolcanic). The geological model used to define the morphology, attitude, and thicknesses of the various pegmatites and contact relationships with host rocks was derived entirely from surface drilling. At Syväjärvi, five modelled spodumene-bearing pegmatite veins are intruded into mica schists, metagreywackes, and metavolcanics following a broad antiformal structure forming “saddleback” type reefs. This has resulted in a series of shallow northerly dipping pegmatite veins, the largest of these attaining thicknesses of up to 20 m in places. The strike length totals 365 m for all veins, extending approximately 810 m down dip and to a maximum depth below surface of 170 m. Due to variability in the pegmatite/s strikes and dips, true pegmatite thicknesses were generally 70–80% of drill length. The main pegmatite is relatively flat-lying with shallow to horizontal dips (10–30˚) to the north-northwest (Figure 5-4). Pegmatite contacts are typically sharp with the frequent development of weakly mineralised or un-mineralised zones of muscovite-rich pegmatite within and along the margins of the pegmatites.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 43 Syväjärvi – 3D View of Modelled Pegmatites looking Southwest Figure 5-4: Syväjärvi – 3D view of modelled pegmatites looking southwest (source: ERM). In 2016, Keliber developed an inclined tunnel into the deposit in order to provide a bulk sample for metallurgical testwork. The total length of the tunnel was 71 m, which included a 17 m intersection of the main pegmatite. Here, the pegmatite comprised coarse-grained spodumene, light grey to green in colour with individual spodumene laths displaying lengths varying from 3 cm to as much as 70 cm. Mineralogical analyses by GTK (Ahtola et al., 2015) have shown that the pegmatites are comprised of albite (37–41%), quartz (27%), potassium feldspar (16%), spodumene (13%), and muscovite (6–7%). Accessory minerals are apatite (fluorapatite), Nb-Ta-oxides (Mn- and Fe-tantalite), tourmaline (schorl), garnet (almandine), arsenopyrite, and sphalerite. 5.1.2 Rapasaari Geology The Rapasaari lithium pegmatite deposit (Figure 5-3) is covered by a variable cover of till and overburden averaging 12.5 m, ranging from 2.5 m to 30 m in thickness, and outcrops are rare. In places, the till is overlain by peat which can reach up to 2 m in thickness. The pegmatites that make up the Rapasaari deposit are intruded as a series of curvilinear, structurally controlled pegmatites with variable thicknesses resulting in a series of bifurcating lenses and veins that follow a southwesterly plunging synformal structure. This has resulted in a series of northwest-southeast striking and steeply dipping (>60°) southwesterly dipping pegmatites (Rapasaari East) that become more west-east striking, south- dipping (~30°) pegmatites in the north (Rapasaari North). A few small flat-lying pegmatites occur to the east of the main concentration of pegmatites (Figure 5-5). Pegmatites are generally intruded parallel to the host rocks that are primarily composed of mica schists, metagreywackes, and metavolcanics. In certain places, the mica schists are graphitic and sulphide-bearing, but these are generally isolated. Pegmatite contacts are typically sharp, with the frequent development of weakly mineralised or un-mineralised zones of muscovite-rich pegmatite within and along pegmatite margins. The style of pegmatite emplacement has also resulted in the frequent inclusions/xenoliths/rafts of country rocks throughout all the modelled pegmatites at Rapasaari, with these representing internal dilution to the modelled pegmatites. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 44 The three largest modelled pegmatites vary in thickness from 10 m to 30 m, with most of the minor (modelled) pegmatites having thicknesses of less than 10 m. The strike extent totals 1,300 m for all veins – approximately 700 m in the primary dip orientation (east-west) – to a maximum depth below surface of 330 m. Due to the variability in the strike and dip of the pegmatite/s, true pegmatite thicknesses were generally 70–90% of the drill intercept length. Mineralogical analyses by GTK (Ahtola et al., 2015) have shown that the pegmatites comprise albite (37–41%), quartz (26%), potassium feldspar (10%), spodumene (15%), and muscovite (6–7%). Accessory minerals are apatite (fluorapatite), zinnwaldite, Nb-Ta-oxides (Mn- and Fe-tantalite), beryl, tourmaline, fluorine, garnet (grossular), andalusite, calcite, chlorite, Mn-Fe-phosphate, arsenopyrite, pyrite, pyrrhotite, and sphalerite. In general, spodumene crystals are light greyish green in colour, with the lengths of minerals varying from 2 cm to 10 cm. Rapasaari – 3D View of Modelled Pegmatites looking Northwest Figure 5-5: Rapasaari – 3D view of modelled pegmatites looking northwest (source: ERM). 5.1.3 Länttä Geology The Länttä lithium pegmatite deposit (Figure 5-3) is covered by a thin veneer of sediments and till averaging 5.5 m and ranging from 2 m to 10 m in thickness (Figure 5-6). The deposit was discovered during road excavation work in the 1950s. Subsequent drilling completed by historical operators (Suomen Mineraali Oy and Partek Oy) and Keliber delineated three parallel-trending pegmatite veins with a 420 m northeasterly strike and steep southeasterly dips (>60°) to a maximum depth of 220 m below surface, extending approximately 100 m southeast of the outcrop location (Figure 5-6). The pegmatites reach an individual maximum thickness of 10 m and often show localised bifurcating and S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 45 “boudinaging” (SRK, 2023 and reference therein) (pinching and swelling). As a result, xenoliths of metavolcanic host rocks within the pegmatites are common. Due to variability of the strikes and dips of the pegmatite/s, true pegmatite thicknesses were generally 80–90% of the intercepted drill length. Overburden stripping, completed in 2010, has exposed the pegmatite veins on surface and confirmed the variable widths. Host rocks to the pegmatites are metavolcanic rocks containing lenses of metagreywacke schists and plagioclase porphyrite rocks, with pegmatites intruding parallel to schistosity/cleavage and bedding of the host rocks. The pegmatite host rock contacts are sharp and are typically characterised by the development of a tourmaline-rich band at the contact. Mineralogical analyses by GTK show that the pegmatites comprise albite (40%), quartz (15%), potassium feldspar (15%), spodumene (15%), and muscovite (2%). Accessory minerals include apatite, garnet, beryl, tourmaline, and columbite-tantalite. Spodumene crystals are coarse-grained, elongated, and lath-shaped, with lengths ranging from 3 cm to 10 cm, but often reaching 30 cm. Länttä – 3D View of Modelled Pegmatites looking Northeast Figure 5-6: Länttä – 3D view of modelled pegmatites looking northeast (source: ERM). 5.1.4 Emmes Geology The Emmes lithium pegmatite deposit (Figure 5-3) is largely located under Lake Storträsket, close to the village of Emmes. Overburden thickness is highly variable ranging from 2.5–17 m, reaching 17 m thickness under the lake and 10 m closer to the village, and average overburden thickness is approximately 8 m. Exploration drilling completed to date has identified and delineated a single 400 m long pegmatite vein, striking southeast-northwest, extending down dip for 260 m and dipping at ~45–50° to the southwest and a depth of 225 m below surface (Figure 5-7). A curvilinear reverse fault (FLT) displaces the lower dip extent of the pegmatite with a throw of about 30 m. The FLT strikes northwest in the south and becomes more north-northwest striking to the north and steeply dipping, ranging from S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 46 60° to the northeast in the south, to 70° to the east-southeast in the north. The Emmes pegmatite reaches a maximum thickness of 20 m in places and is intruded into mica schists containing occasional graphitic and sulphidic phases as well as metagreywackes. Spodumene is distributed evenly throughout the central part of the pegmatite and a spodumene-poor muscovite-bearing margin. Contacts with the host rocks are sharp, with true pegmatite thicknesses being generally 70–90% of the intercepted drill length. The spodumene, which is similar to that observed in the other pegmatites, is light grey to green in colour as is modal mineralogy which is dominated by feldspar, quartz, spodumene, and muscovite. No inclusions or xenoliths of country/host rock have been identified within the Emmes pegmatite. Emmes – 3D View of Modelled Pegmatite looking North- Northwest Figure 5-7: Emmes – 3D view of modelled pegmatite looking north-northwest. Note the reverse FLT displacing the pegmatite (source: ERM). 5.1.5 Outovesi Geology The Outovesi deposit (Figure 5-3) was discovered by Keliber in 2010 and is covered by surficial till sediments that range from 8.5 m to 18.5 m and average 14 m in thickness (refer to Figure 5-8). The deposit was subsequently drilled and comprises a single pegmatite vein striking northeast-southwest for approximately 550 m, reaching a maximum thickness of 10 m (Figure 5-8). The vein has a variable dip to the southeast at between ~40–80° to a depth of 100 m

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 47 below surface. Host rocks are dominated by homogenous mica schists and metagreywackes, with the northern parts of the deposit being hosted by more graphite-rich schists. The Outovesi pegmatite has intruded almost at right angles to the host rock fabric, which is different to that at the Länttä and Rapasaari deposits where the pegmatites have generally intruded parallel to host rock fabrics. Contacts with the host rocks are sharp, with true pegmatite thicknesses being generally 90% of drill length. Despite no detailed mineralogy having been completed over Outovesi, the modal mineralogy is anticipated to be very similar to the other deposits, being dominated by albite, quartz, potassium feldspar, spodumene, and muscovite (SRK, 2023) which is supported by observations of the drill core. Spodumene crystals are generally light grey-green in colour with individual spodumene minerals reaching lengths of between 2 cm to 10 cm. It is noted that a later stage, possibly hydrothermal, overprint has resulted in a variable zone of alteration close to the pegmatite contacts, and this has resulted in the alteration of spodumene to a lower tenor Li-bearing muscovite. Outovesi – 3D View of Modelled Pegmatite looking North- Northwest Figure 5-8: Outovesi – 3D view of modelled pegmatite looking north-northwest (source: ERM). 5.1.6 Leviäkangas Geology The Leviäkangas lithium pegmatite deposit (Figure 5-3) is located in the Kaustinen Municipality of Western Finland approximately 10 km north of Kaustinen town. Exploration drilling has identified a single sigmoidal shaped spodumene pegmatite with a strike length of 500 m along a north-northwest strike and dips at between 45° and 60 ° to the west. The pegmatite is conformably intruded into host rocks comprising mica schists interlayered with metagreywacke and black schist layers and, locally, plagioclase porphyritic rocks units are present (PL Mineral Reserve Services, 2016 and SRK, 2023). The thickness of the pegmatite S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 48 varies from a few metres to approximately 20 m (Figure 5-9). The overburden is formed by till with some peat at the surface at Leviäkangas and varies in thickness from 3.5 m to 14 m, averaging 7.5 m. Close to, and at the contact with the host rocks, the spodumene in the pegmatite is altered to muscovite. This persists for a few tens of centimetres up to 1.5 m. In addition, there are a few narrow (0.5–3 m) internal waste zones in the pegmatite comprising mica-schist xenoliths or where the spodumene is replaced by muscovite resulting in low Li2O grades. Spodumene typically occurs as coarse-grained, light greyish-green lath-shaped crystals between 2 cm and 10 cm long and orientated perpendicular to the contacts of the veins with the wall rock. The pegmatite consists predominantly of albite (37–41%), quartz (28%), potassium feldspar (orthoclase) (15%), spodumene (10%), and muscovite (6–7%), with accessory minerals comprising apatite, cassiterite, cookeite, garnet, graphite, Mn-Fe phosphate, montebrasite, Nb-Ta oxides, sphalerite, tourmaline, and zeolite (Ahtola et al., 2015). Leviäkangas – 3D View of Modelled Pegmatite looking North- Northwest Figure 5-9: Leviäkangas – 3D view of modelled pegmatite looking north-northwest (source: ERM). 5.1.7 Tuoreetsaaret Geology The Tuoreetsaaret lithium pegmatite deposit, also located in the Kaustinen Municipality of western Finland (Figure 5-3) was discovered by Keliber using a combination of geological, geochemical, and geophysical data and led to the first intersection by diamond core drilling in March 2020 (SRK, 2023). The deposit comprises three north-south striking and one northwest striking lithium-bearing pegmatite vein-like bodies intruded into country rocks comprising intermediate meta-tuffite, plagioclase porphyrite, mica schist, and sulphide-bearing mica schist. The hanging wall is generally formed by intermediate meta-tuffite and the footwall by mica schist and sulphide-bearing mica schist. Plagioclase porphyrite generally forms the middling between the pegmatite veins. The pegmatite veins and their wall rocks are covered by 10 m to 35 m, averaging 22 m of glacial till with peat at the top. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 49 The pegmatites range in true thickness from <3 m to 40 m and in strike length from 350 m to 900 m. The dip is steep (80–90°) and to the east but is locally variable (Figure 5-10, Figure 5-11). Xenoliths of country rock are common within the pegmatite bodies. The lithium grains (1 mm to 3 mm in length) are significantly smaller than at Leviäkangas, but the grain size does not differ much between the different veins intersected (SRK, 2023 and references therein). Tuoreetsaaret – Plan View of Modelled Pegmatites Figure 5-10: Tuoreetsaaret – plan view of modelled pegmatites (source: ERM). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 50 Tuoreetsaaret – 3D View of Modelled Pegmatites looking South Figure 5-11: Tuoreetsaaret – 3D view of modelled pegmatites looking south (source: ERM). 5.2 Internal Pegmatite Zonation and Mineralogy The pegmatites that have been discovered and evaluated to date within the Kaustinen area have very similar mineralogy and are dominated by albite (37–41%), quartz (26–28%), K-feldspar (10–16%), spodumene (10–15%), and muscovite (6–7%) (Ahtola et al., 2015). Internal pegmatite zonation, as seen in many other similar LCT-type albite spodumene pegmatites, is poorly developed to absent from the Kaustinen pegmatites, with spodumene being the only lithium-bearing mineral that is of economic interest. The poorly developed internal zonation from the contact to the pegmatite centre is variable and may include: • A thin, fine-grained rim and coarser-grained wall zone of variable thickness comprising quartz, K-feldspar (graphic intergrowth may be present), and muscovite; • A coarse-grained spodumene-bearing inner/core zone that contains varying ratios of quartz, K-feldspar, albite, and minor muscovite along with a variety of accessory minerals; and • Banded aplite layers comprising fine-grained quartz, feldspar, and mica are also present within the spodumene-bearing inner zones. It is also noted that this zonation is not symmetrical about the centre of the pegmatite and, while broad zones are recognised and the spodumene mineralisation is generally homogeneously distributed throughout most of the pegmatites, there is a fair amount of local variation in the mineral composition and textures within individual pegmatites. The inclusion or incorporation of host rock xenoliths and wall rock material through dilution will impact the recovery of spodumene during flotation and metallurgical processing. This will require careful selective mining supported by optical or density sorting methods to mitigate the impacts of dilution on the recovery of spodumene.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 51 Other lithium-bearing minerals such as petalite (LiAlSi4O10), lepidolite (K(Li,Al)3(Al,Si,Rb)4O10(F,OH)2), montebrasite- amblygonite (LiAl(PO4)(OH,F) – LiAl(PO4)F), lithiophilite (Li(Mn,Fe)PO4: LiFePO4 – LiMnPO4), zinnwaldite (KLiFeAl(AlSi3)O10(OH,F)2), and elbaite (tourmaline) (NaLi2.5Al6.5(BO3)3Si6O18(OH)4) have been found only in minor or trace quantities. 5.3 Weathering Surficial weathering and alteration of the pegmatite minerals to clays can result in alteration and leaching of the lithium from spodumene and other lithium-bearing minerals, largely depleting the upper weathered portions of pegmatites of lithium or alteration of spodumene to other lithium-bearing minerals that are largely unrecoverable. Incipient alteration may result in partial alteration of spodumene and physical breakdown of spodumene crystals which tends to make recovery of this material difficult due to the production of fines. At most of the deposits, no weathering is observed, however, at the Rapasaari deposit, partial weathering or fracture oxidation occurs to a depth of 20–30 m (PayneGeo, 2022). An example of the deeper weathering from Rapasaari is shown in Figure 5-12, but it is understood this is rare and it does not appear to significantly alter the spodumene or affect the lithium grades. Example of Weathered Pegmatite from Rapasaari Figure 5-12: Example of weathered pegmatite from Rapasaari (hole RA14; box 1 (~11–15 m depth)). Although the core is broken, the spodumene looks largely unaltered and lithium grades through this zone (samples 40582 to 40584) range from 0.52–0.86% Li (or 1.13–1.86% Li2O) and average 0.64% Li (1.38% Li2O). 5.4 Mineralisation Style and Deposit Type – LCT Pegmatites [§229.601(b)(96)(iii)(B)(6)(ii-iii)] A pegmatite is defined as “an essentially igneous rock, commonly of granitic composition, that is distinguished from other igneous rocks by its extremely coarse but variable grain size or by an abundance of crystals with skeletal, graphic, or other strongly directional growth habits. Pegmatites occur as sharply bounded homogenous to zoned bodies within igneous or metamorphic host rocks.” (London, 2008). The main rock-forming minerals in a granitic pegmatite include feldspar, mica (muscovite and biotite), and quartz. Other minerals may occur in economic concentrations and include, but are not limited to, various lithium minerals S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 52 (Table 5-1), beryl, tourmaline, cassiterite, columbite-tantalite, topaz, garnet, and various rare earth minerals. Commercially, spodumene and petalite are the two most important lithium minerals mined from LCT pegmatites. Spodumene concentrates are largely used to produce lithium carbonate or lithium hydroxide for the battery industry whereas petalite, as well as some of the spodumene production, is mostly utilised in the glass and ceramics industry. The feldspar, muscovite, and quartz from the pegmatites also have several industrial and commercial applications. Table 5-1: Summary of chemical composition and density of the main lithium minerals associated with pegmatites. Mineral Chemical Composition Maximum* Li% (calculated) Maximum* Li2O% (calculated) Density Range (g/cm3) (average) Spodumene LiAl(Si2O6) 3.7 8.0 3.15 Lepidolite K₂(Li,Al)5-6(Si6-7Al2-1O20)(OH,F)4 1.39–3.6 3–7.9 2.8–2.9 (2.84) Petalite LiAl(Si4O10) 1.6–2.27 3.4–4.9 2.39–2.46 (2.42) Amblygonite-montebrasite (Li,Na)Al(PO4)(F,OH) – LiAl(PO4)(F,OH) 3.4–4.7 7.4–10.2 3.0 Hectorite Na0.3(Mg,Li)3Si4O10(OH)2 0.54 1.17 2–3 (2.5) Eucryptite LiAl(SiO4) 2.1–5.5 4.5–11.8 2.67 Lithiophilite-triphylite LiMnPO4 – LiFePO4 4.4 9.53 3.34–3.5 Zinnwaldite K(Al,Fe,Li)3(Si,Al)4O10(OH)F 1.59 3.42 2.9–3.1 (3.0) Cookeite (alteration product of spodumene or petalite) LiAl4(Si3Al)O10(OH)8 1.33 2.86 2.67 *Note that the actual lithium concentrations presented represent maximum theoretical lithium content and may be lower due to natural variations in the mineral chemistry. Conversion factor from Li % to Li2O % = Li % x 2.153. Source: www.webmineral.com; BGS, 2016 Most pegmatites occur in swarms or pegmatite fields and occupy areas ranging from tens to hundreds of square kilometres. Pegmatites are classified according to several geological, textural, mineralogical, and geochemical parameters and the accepted classification scheme is described in Černy and Ercit (2005) and London (2008). Three broad pegmatite families are recognised based largely on geochemical (i.e. composition) data, namely LCT, Niobium- yttrium-fluorine (NYF), and mixed LCT-NYF families. Traditional models considered pegmatites to be the product of extreme fractional crystallisation of granites and usually a close association with a parental granite referred to as RMG (residual melts of granitic magmatism) pegmatites (Müller et al., 2022). Often, there is also an increase in the complexity of the internal pegmatite zonation with increasing fractionation. Granites (S-type) derived from melting of metasedimentary rocks in continental collision zones (Černy and Ercit, 2005) often give rise to LCT pegmatite fields that often show a broad geochemical zonation pattern, with pegmatites most enriched in incompatible elements like Li, Cs, and Ta which are typically the furthest away from their cogenetic granite and represent the last phase of crystallisation (Figure 5-13). These pegmatites are often hosted within supracustal rocks (e.g. greenstone belts), comprising mafic volcanics, and igneous equivalents, intercalated with sedimentary rock where peak metamorphic conditions attained are usually upper greenschist to amphibolite facies (London, 2008; Bradley and McCauley, 2016). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 53 Idealised Schematic Model in Profile or Plan showing the Regional Zonation in a Pegmatite Field around a Parental Granite Intrusion Figure 5-13: Idealised schematic model in profile or plan showing the regional zonation in a pegmatite field around a parental granite intrusion. Note: The rare-element suites of the most enriched pegmatites in each zone are indicated with the most prospective pegmatites located in distal areas compared to the parental granite (source: London, 2016). More recently, pegmatite models also include pegmatite formation by anatexis (melting) of suitable metasedimentary (e.g. metasedimentary rocks with evaporite sequences: Simmons and Webber, 2008; London, 2008, 2018; Knoll et al., 2023) and/or meta-igneous rocks (Duuring, 2020; Koopmans et al., 2023), referred to as direct products of anatexis (DPA) type pegmatites (Müller et al., 2022). These pegmatite fields show no systematic zonation and are often restricted to specific structural zones and/or lithologies. Pegmatites sizes may vary from a few metres to hundreds of metres (and sometimes >1 km) in length with variable widths ranging from <1 m to tens of metres (or even hundreds of metres in some rare examples) and may have simple to complex internal structure. The Kaustinen pegmatites are considered to belong to the rare-element pegmatite class of the LCT family of the albite- spodumene type. The albite-spodumene type pegmatites are characterised by a general absence of a systematic internal zonation, although crude zones can be defined on the basis of mineralogy and texture and fine-grained and/or layered aplite zones are commonly distributed within the pegmatite but often towards the lower half of the intrusion. The pegmatites’ shape is usually controlled by existing FLTs, fractures, foliation, and bedding in country rocks (Duuring, 2020) and often form a series separate to semi-contiguous en échelon and crosscutting bodies with sub-horizontal to vertical dips, intruded along extensional fracture sets (Figure 5-14). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 54 Sketches showing the Shapes of (A) a Vertical En Echelon Series of Intrusions which are Joined at Depth and (B) a more Shallowly Dipping Series of Veins Exposed at Surface, with Blind Intrusions at Depth Figure 5-14: Sketches showing the shapes of (A) a vertical en Echelon series of intrusions which are joined at depth (Fossen, 2010) and (B) a more shallowly dipping series of veins exposed at surface, with blind intrusions at depth (source: unknown). The Kaustinen pegmatites are considered to be the products of extreme fractionation of the numerous granites (many being pegmatitic granites) in the Kaustinen area, although there is no clear or well-defined zonation observed within the pegmatite groups to date to prove this and more accurate age determination of the granites and pegmatites is required (Ahtola et al., 2015). 5.5 General Lithium Mineral Processing Considerations for Hard Rock Deposits Lithium minerals such as spodumene and petalite are generally separated from other pegmatite minerals by flotation and gravity separation methods. Hand sorting may be used for very coarse-grained lithium minerals or ore sorting technologies for finer-grained minerals. Low-intensity magnetic separation can be used to remove tramp iron (from grinding balls), while paramagnetic minerals such as tourmaline or garnet may be removed using high-intensity magnetic separators (Garrett, 2004). Downstream processing of lithium mineral concentrates may follow several routes. Typically, to extract lithium from spodumene, the crystal structure of spodumene must be converted from the naturally occurring monoclinic α-form to the tetragonal β-form by roasting to about 1,000°C. This makes the spodumene amenable to leaching with sulphuric acid, which forms soluble lithium sulphate from which lithium carbonate may be precipitated using soda ash. An evaluation of lithium mineral processing for any specific project should address the following points: • What minerals are present in the mineralised rock – if there are several lithium minerals, can they be recovered and processed economically? • How pure are the lithium minerals? For example, there could be small quartz intergrowths that reduce concentrate purity, as with spodumene quartz intergrowths, which typically form as a replacement for petalite (Figure 5-15).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 55 • What liberation methods may be applied e.g. gravity, flotation, and cleaning to produce concentrates of acceptable size distribution and purity? • How does the liberation grind size affect other minerals such as niobium-tantalum minerals that may also be of potential economic interest? Spodumene-Quartz Intergrowth seen in Thin Section Figure 5-15: Spodumene-quartz intergrowth seen in thin section (source: Scogings et al., 2016). As alluded to above, spodumene and other lithium minerals are sold as mineral concentrates. The following general specifications for spodumene concentrates are provided below only as an example. Technical grade SC5 refers to a technical grade spodumene concentrate with a Li2O content of plus 5% Li2O. Technical-grade lithium concentrates are commonly used in the manufacture of glass, ceramics, where discolouration from iron is a concern, and in metallurgical powders. Compositions of technical grade spodumene concentrates range from 4.0% to 7.5% Li2O and require ultra- low levels of iron (<0.05% Fe2O3). Alkaline content for ceramics is also important with <1% combined K2O and Na2O requested by many end-users. Chemical grade SC6 refers to a chemical grade spodumene concentrate with a 6% Li2O content. Chemical-grade concentrates are sold to lithium chemical producers who convert the mineral concentrates into lithium carbonate, lithium hydroxide or lithium metal. The lithium content of these concentrates ranges from 4% to 6% Li2O and no firm iron (but generally <1% Fe2O3), feldspar or other impurity ranges. 5.6 Mineral Concentrates Spodumene concentrates are traditionally the preferred feedstock into the lithium-ion battery supply chain, however, future demand is looking to petalite and lepidolite concentrates as feedstocks to meet demand. Petalite concentrates, although lower grade, between 4.0% and 4.5% Li2O, compared to spodumene concentrates which are >5.5% Li2O, follow the same processing route to either lithium carbonate or lithium hydroxide (i.e. calcination petalite or spodumene converts to β-spodumene which is then leached and processed to the relevant Li- chemical product). The Chinese have, and are currently, using lepidolite as a feedstock for some of their chemical converters. There are also a number of technologies and processes that have been developed by some of the lithium explorers to process S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 56 lepidolite and potentially other feedstocks (e.g. lithium-phosphates and also recycling of lithium-ion batteries), examples include: Lepidico’s L-Max® and LOH-Max® technologies (https://lepidico.com/technology#our-technologies) and Lithium Australia’s LieNA® and SciLeach® technologies (www.lithium-au.com/lithium-chemicals/). These technologies can also handle deleterious elements like fluorine which is associated with the processing of lepidolite and other lithium-bearing micas. In summary, the current lithium market is being driven by forecast demand for the battery market and, while grade and tonnage are important metrics to consider for lithium deposits, other important aspects to consider are mineralogy, mineral textures and variability of these within individual pegmatites and between pegmatites within a particular project, and how these translate into the production of a mineral concentrate and feedstock for either the chemical or technical market. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 57 6 EXPLORATION [§229.601(b)(96)(iii)(B)(7) 6.1 Non-Invasive Exploration Activities [§229.601(b)(96)(iii)(B)(7)(i)] 6.1.1 Geological and Boulder Mapping The lack of outcrop throughout most of the Kaustinen region has necessitated the use of mapping of erratics (glacier- transported boulders and rock fragments) as the primary pegmatite exploration method instead of traditional mapping and outcrop sampling methodologies. This type of litho-geochemical sampling and mapping has been used since the 1960s and remains an effective method for discovering hidden or buried pegmatites. Since Keliber started exploration in 2010, more than 1,500 spodumene pegmatite boulders have been mapped, and the distribution of the boulders and boulder fans have been used to vector to potential pegmatite source areas. All the pegmatite deposits, except for the Länttä deposit (which was discovered through a road excavation), have been discovered through tracing of boulder fans to the northwest (being the regional direction of palaeo-glacial ice movement). The subsequent drilling programmes were then focused and designed around the areas around the northwest end of these boulder fans (Figure 6-1). Geological Map showing Distribution of Mapped Spodumene Pegmatite Boulders in relation to Pegmatites Figure 6-1: Geological map showing distribution of mapped spodumene pegmatite boulders in relation to pegmatites (source: SRK, 2023 (Keliber)). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 58 6.1.2 Geochemical Sampling GTK carried out extensive regional scale till sampling covering the entire country in the 1970s and 1980s, resulting in the collection of over 22,420 samples in the Kaustinen area (Ahtola et al., 2015; Chudasama and Sarala, 2022). The samples were collected from an average depth of 2.4 m and along 500–2,000 m-spaced lines, with sample intervals varying from 100–400 m. Sample lines were orientated perpendicular to the direction of glacial drift (i.e. southwest- northeast). At the time, lithium was not analysed and only in 2010 when GTK reanalysed 9,658 samples from the Kaustinen area was the presence of lithium anomalies around known deposits confirmed (Figure 6-2). These results demonstrated the effectiveness of till geochemistry coupled with boulder mapping as an exploration tool for pegmatite-hosted lithium mineralisation in this environment. Regional Distribution of Li in Till in relation to Known Lithium Deposits Figure 6-2: Regional distribution of Li in till in relation to known lithium deposits (source: Ahtola et al, 2015). A recent study by Chudasama and Sarala (2022) conducted prospectivity mapping of lithium-bearing spodumene pegmatites in the Kaustinen area using regional till geochemical and Light Detection and Ranging (LiDAR)-based glacial geomorphological data. They used Li-pegmatite pathfinder elements (As, Be, Bi, K, Li, Sb, and Zr) for spatial data analysis from the regional till geochemical data and an interpreted NNW-SSE trend of glacial transportation. They made use of three prospectivity methods (weights-of-evidence, logistic regression, and fuzzy models) for mapping the potential of Li-enrichment in the study area. The results of the weights-of-evidence and logistic regression methods were found to be the most accurate and useful for potentially locating areas for detailed ground exploration activities and identification of new Li-rich pegmatites in the Kaustinen area.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 59 6.2 Drilling, Logging, and Sampling [§229.601(b)(96)(iii)(B)(7)(ii) (v) (vi)] With the exception of shallow surface reverse circulation (RC) drilling completed by GTK over the Syväjärvi and the Leviäkangas deposits, all drilling on the Project has been completed using diamond core drilling. Diamond core drilling has been the only method used to generate geological, structural, and analytical data, and these have been used as the basis for Mineral Resource estimation over each of the deposits defined to date (SRK, 2023). Earlier drilling phases conducted in the 1960s to early 1980s, executed by Suomen Mineraali Oy and Partek Oy, focussed on the Emmes, Länttä, Leviäkangas and Syväjärvi deposits. GTK subsequently completed drilling over the Syväjärvi and Rapasaari deposits between 2004 and 2012. Since 1999, Keliber have completed extensive drilling programmes focusing on delineating MREs over each of these deposits, including the Outovesi deposit that Keliber discovered in 2010 and Tuoreetsaaret discovered in 2020. The recent drilling completed by Keliber at Syväjärvi, Rapasaari, Länttä, Leviäkangas, and Emmes has been as infill and extensional drilling to the historical drilling and served to validate the historical datasets. Furthermore, drilling has been carried out at new target areas and for sterilisation drilling in areas where infrastructure is planned. The historical drilling completed in the 1960s through to the 1980s was completed using 32 mm diameter core drilling, with GTK drilling using a 42 mm diameter and Keliber a 50.7 mm core diameter, respectively. The majority of the drill holes were directed to intersect pegmatites at right angles to their orientations, with holes inclined at an average of 45° with an average mean vertical drilling depth of 85 m below surface. Table 6-1 shows details of historical, GTK, and Keliber drilling over each of the deposits. Core recoveries across all deposits for drilling conducted by Keliber are in excess of 98%, with slightly higher losses expected in the top portions of holes from where weathering is present (e.g. Rapasaari, see Section 5.3), but it is not considered that this will materially impact the accuracy or representativeness of the data. Table 6-1: Summary of drilling completed over the Keliber Lithium Project (source: Keliber). Historical & GTK Keliber Total No. of Drill Holes Length (m) No. of Drill Holes Length (m) No. of Drill Holes Length (m) Syväjärvi 91* 4,197 170 19,385 261 23,582 Rapasaari 26 3,655 321 56,651 347 60,306 Länttä 54 3,494 54 5,691 105 9,185 Emmes 31 3,348 23 2,937 54 6,285 Outovesi – – 24 1,816 24 1,816 Tuoreetsaaret 2 270 103 24,143 105 24,413 Leviäkangas 106** 5,850 49 5,127 155 10,977 Total 310 20,814 744 115,750 1,051 136,564 * includes 57 shallow percussion holes ** includes 60 shallow percussion holes 6.2.1 Syväjärvi Drilling Suomen Mineraali Oy discovered the Syväjärvi deposit following boulder mapping, and the first drilling was subsequently completed in 1961, followed by drilling by Partek Oy until the 1980s. Close-spaced drilling was then S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 60 completed by GTK between 2006 and 2010 (SRK, 2023). Following Keliber’s acquisition of the Project in 2012, several drilling campaigns were completed between 2013 and 2023, with the focus of declaring a high-confidence MRE. A total of 261 holes have been drilled over the Project, totalling 23,582 m (Table 6-1 and Figure 6-3). Due to the Project’s location close to and partly under Lake Syväjärvi, drilling was only possible during the winter months when the lake froze and access to it could be achieved. Keliber’s surface drilling was completed on a drill hole spacing ranging from 20 m to 50 m, with most drill holes having easterly azimuths to intersect the shallowly dipping pegmatites as close to their true width/attitude as possible (Figure 6-3). Following the completion of the exploration tunnel, an additional six underground holes were drilled along the plane of the pegmatite to test and validate its up- dip continuity. Map showing Historical, GTK, and Keliber Drilling at Syväjärvi Figure 6-3: Map showing historical, GTK, and Keliber drilling at Syväjärvi (source: Keliber, 2023). 6.2.2 Rapasaari Drilling Rapasaari was discovered in 2009 following a boulder mapping, till sampling, and geophysical programme by GTK. During 2009 and 2011, GTK completed a 26-hole drilling programme. Since Keliber’s acquisition of the Project in 2014, numerous drilling campaigns over the period of 2014 to 2023 have been completed. The focus has been on accurately delineating the Rapasaari deposit geology and structure. A total of 347 holes have been drilled on the Project, totalling 60,306 m (Figure 6-4). Keliber’s surface drilling was completed on a semi-regular grid with holes spaced at between S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 61 30 m and 60 m, although there are some closer-spaced holes in areas, with Rapasaari East drill holes having easterly azimuths and some of the Rapasaari North drill holes having southerly azimuths in order to intersect pegmatites as close to their true width/attitude as possible. Map showing GTK and Keliber Drilling at Rapasaari Figure 6-4: Map showing GTK and Keliber drilling at Rapasaari (source: Keliber, 2023). 6.2.3 Länttä Drilling The Länttä deposit was discovered when a mineralised pegmatite was exposed during road works in the 1950s and was initially drilled by Suomen Mineraali Oy. This initial exploration work included bulk sampling and metallurgical testing in the late 1970s, but no additional work was completed as the Project was considered uneconomic at the time. Keliber acquired the mineral rights to the Project in 1999 and completed more detailed exploration in collaboration with GTK and, in 2010, overburden stripping and exposure of both pegmatite veins was completed. Bulk samples were taken for metallurgical testwork as well as to generate internal certified reference material (CRM) for the Project (SRK, 2023). A total of 105 diamond core drill holes have been drilled over the Project, totalling 9,185 m (Figure 6-5). Keliber’s surface drilling was completed on 40 m-spaced section lines with all drill holes having northwesterly azimuths to intersect pegmatites as close to their true width/attitude as possible (Figure 6-5). Most of the Keliber drilling was done along strike to the southwest of the GTK and historical drilling. Overall drill hole spacing ranges between 10 m and 50 m. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 62 Map showing Historical, GTK, and Keliber Drilling at Länttä Figure 6-5: Map showing historical, GTK, and Keliber drilling at Länttä (source: Keliber, 2023). 6.2.4 Emmes Drilling The Emmes deposit was discovered following boulder mapping completed by Suomen Mineraali Oy in the 1960s. A number of historical drilling phases by Suomen Mineraali Oy, as well as Partek Oy, were completed through to 1981. Following Keliber’s acquisition of the mineral rights in 2012, three drilling programmes, including several ice drilling programmes to validate historical holes as well as to further delineate the extent of the pegmatite beneath Lake Storträsket, have been completed. A total of 54 holes of diamond drill core have been drilled over the Project, totalling 6,285 m (Figure 6-6). Keliber’s surface drilling was completed using a drill hole spacing ranging from 20 m to 50 m along variable spaced lines, largely as infill to the historical drilling, with drill holes having north and northeasterly azimuths to intersect pegmatites as close to their true width/attitude as possible.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 63 Map showing Historical and Keliber Drilling at Emmes Figure 6-6: Map showing historical and Keliber drilling at Emmes (source: Keliber, 2023). 6.2.5 Outovesi Drilling The Outovesi deposit was discovered by Keliber in 2010 and all 31 diamond core drill holes totalling 2,613 m were completed in that year (Figure 6-7). Keliber’s surface drilling was completed on nominal 40 m-spaced section lines, with overall drill hole spacing ranging from 30–50 m along a northeasterly strike. All drill holes had easterly azimuths to intersect pegmatites as close to their true width/attitude as possible. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 64 Map showing Keliber’s Drilling at Outovesi Figure 6-7: Map showing Keliber’s drilling at Outovesi (source: Keliber, 2023). 6.2.6 Tuoreetsaaret Drilling Located between the Syväjärvi and Rapasaari deposits, and despite two poorly directed GTK holes drilled between 2005 and 2009, the Tuoreetsaaret deposit was only discovered by Keliber in 2020 and subsequently drilled during 2021, 2022, and 2023 (refer to Figure 6-8). The drilling is directed towards the east and west and is approximately perpendicular to the orientation of the steeply dipping pegmatites. A total of 103 holes totalling 24,143 m have been completed. The sub-vertical veins are intersected through both east- and west-oriented drill holes on approximately 40 m-spaced fence lines with a nominal 50–60 m spacing along the fences. The veins are reasonably closely spaced at between 10 m to 50 m apart. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 65 Map showing GTK and Keliber Drilling at Tuoreetsaaret Figure 6-8: Map showing GTK and Keliber drilling at Tuoreetsaaret (source: Keliber, 2023). 6.2.7 Leviäkangas Drilling The Leviäkangas deposit was initially identified through boulder mapping and later confirmed by percussion and diamond drilling by Partek. Infill drilling by Keliber was guided by Partek’s results and focussed on the better- mineralised areas intersected. The percussion drilling by Partek is not used in the Mineral Resource estimation. The drill hole spacing is fairly close for the shallowest pegmatite, with fence lines spaced at approximately 20 m and oriented towards the east, perpendicular to the strike of the veins. For the two deeper mineralised pegmatites, the spacing is significantly wider at 50 m to 100 m. Of the 155 drill holes in the vicinity of the deposit (including the percussion drilling), 31 holes have intersected the modelled mineralised pegmatites (Figure 6-9). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 66 Map showing Historical, GTK, and Keliber Drilling at Leviäkangas Figure 6-9: Map showing historical, GTK, and Keliber drilling at Leviäkangas (source: Keliber, 2023). 6.2.8 Sampling Procedures The logging and sampling of diamond drill core by Keliber was completed at Keliber’s core processing and sampling facility in Kaustinen, guided by their Standard Operating Procedures (SOPs) which are aligned with industry-accepted best practice. Lithological logging criteria focused on mineralogical, lithological, and structural variables, with sample intervals varying from 0.2 m to 2.5 m. Mineralogical logging focused on documenting spodumene crystal size, orientation, colour, and estimated quantity. In earlier drilling phases the core was orientated by drillers (every 10–15 m) using the “wax stick method”. However, in more recent phases, post-2016, core orientation has been done using a digital Reflex Act III tool that measures the orientation of drill core for each 3 m run. Once the core had been logged, the core boxes were photographed dry and wet. After core had been marked up for sampling, it was split in half along the long axis using an automatic diamond saw, with half of the core being subject to drying, weighing, measurement of specific gravity (SG), further drying, and then packing into sample bags for dispatch to the laboratory for preparation and analysis. All lithological, structural, mineralogical, density, rock quality designation (RQD) and sampling data were captured into Microsoft Excel® spreadsheets and then compiled into a Microsoft Access® database.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 67 6.2.9 Density Keliber carried out density determinations using the water displacement (Archimedes) method and included the use of two standards that were measured at every 10th sample. The majority of Keliber’s density measurements were collected from pegmatite material, and non-mineralised material (host rock inclusions/xenoliths) was also included. There is a strong correlation with Li2O grade (i.e. spodumene content) and density and, depending on grade (usually 10–20% spodumene), the density can vary between 2.65 g/m3 and 2.80 g/m3. See Section 10.14 for further information. 6.3 Geotechnical and Hydrogeological Drilling To date, no oriented geotechnical drilling has been done for any of the sites, with geotechnical logging carried out on geological drill core (SRK, 2023). As part of the 2022 Feasibility Study (FS), geotechnical logging was conducted during the exploration drilling programmes. The current geotechnical environment at Rapasaari, Syväjärvi, and Länttä sites is understood to PFS level. The intact rock strength parameters for the Syväjärvi site were inferred from those determined from the Rapasaari due to their close proximity to each other in comparison to other mining areas. Geotechnical conditions vary across the different sites, with open-pit reserves having higher geotechnical data confidence due to existing exposures and laboratory testwork. Separate hydrology and hydrogeology studies were conducted as part of the 2022 FS and are discussed at a high level in the mining methods section (Chapter 12) of this TRS. 6.4 QP’s Opinion on the Exploration The Resource QP considers that the drilling and recovery methods employed by Keliber are suitably aligned with industry practice and is not aware of any factors that could affect the reliability or accuracy of the results of the recent exploration targeting the pegmatite-hosted lithium mineralisation. The historical data have undergone a thorough verification process as described in Chapter 8. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 68 7 SAMPLE PREPARATION, ANALYSES, AND SECURITY [§229.601(b)(96)(iii)(B)(8)] 7.1 Sample Preparation Methods and Quality Control Measures [§229.601(b)(96)(iii)(B)(8)(i)] The sample material used for analyses on the Project was sourced from diamond drill core that was split in half using an electric diamond core saw or, in the case of the historical drill core samples, a guillotine. All core handling and sampling were completed at Keliber’s secure core logging and sampling facility in Kaustinen. To monitor the accuracy and precision of the results, Keliber have, since 2013, implemented a quality assurance and quality control (QAQC) SOP for all their drilling programmes on the Keliber Project. The pegmatite sample lengths vary from a nominal 0.5 m to 1.5 m but, based on geological contacts and at the geologist’s discretion, may be <0.5 m and up to +2 m in places (Sandberg, c.2013). Sampling of the host rocks has also been conducted for rock characterisation for environmental and general exploration purposes, but these data have not been used for Mineral Resource estimation purposes. The Quality Control (QC) policy includes the insertion of Keliber’s in-house, locally sourced CRM, blanks, and duplicates into the sampling sequence at a frequency of one in every 20 samples (5%). Duplicate QC samples included replicate samples (quarter core samples) and pulp duplicate samples. Keliber generated three separate in-house CRMs from samples drawn from the Länttä deposit as well as a certified blank material drawn from the Lumppio granite which outcrops in the area. These CRMs (including the blank material) were prepared by an independent laboratory, the Eurofin Labtium Group (Labtium), in Finland. Labtium is ISO/IEC 17025 accredited by FINAS (Finnish accreditation service; Testing laboratory T025) and the routine inductively coupled plasma optical emission spectroscopy (ICP-OES) analytical methods discussed below have accreditation status. A commercially available CRM (AMIS0355) has also been used by Labtium as part of their internal QC when analysing Keliber’s samples. All sealed samples were delivered to Labtium’s independent laboratory in Kuopio, Finland, which has carried out all primary sample preparation and assay for the Project since 2014. 7.2 Sample Preparation, Assaying, and Laboratory Procedures [§229.601(b)(96)(iii)(B)(8)(ii)] All sample preparation and analyses were completed at Labtium’s laboratory facility in Kuopio, Finland. The sample preparation comprises weighing, drying and crushing to a nominal -6 mm, and taking a 0.7 kg split of the coarse crush using a rotary splitter. The coarse-crush split is then pulverised, and an aliquot of 0.2 g is used for analyses. Pulp and coarse reject samples have been retained for reference purposes, future analysis, and possible metallurgical testing. The analytical process used by Labtium (method code 720P) is sodium peroxide fusion (700°C/5 min) followed by dissolving in HCl plus dilution with HNO3 and analysis by ICP-OES. A 27-element suite is routinely assayed using this method with a lithium detection limit of 0.001%. During 2013, samples from Leviäkangas, Länttä, Syväjärvi, and Outovesi were assayed by 4-acid digest (an accepted and suitable method also used in pegmatite-hosted lithium exploration, but may underreport the lithium slightly, usually within 5–10% of results produced by the peroxide fusion method). Considering this, samples within mineralised S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 69 spodumene-bearing pegmatite zones from Leviäkangas and Syväjärvi were subsequently re-assayed by peroxide fusion (720P). Länttä and Outovesi samples assayed by 4-acid were not re-analysed. Check sample assays carried out by ALS Ltd in 2013, using the 4-acid digestion, showed some variance between results, which has been attributed to the efficiency of the 4-acid digest used by ALS Ltd versus the sodium peroxide fusion method used by Labtium. Keliber have, therefore, used the sodium peroxide fusion digest (Labtium code 720P) methodology for all sample analyses as this method provides a more complete digest and hence more accurate analytical results. Multi-element lithogeochemical data generated from sampling of non-pegmatite rocks have been used for rock characterisation for environmental and general exploration purposes. Assay methods used for this work were not used for the MRE. This included an aqua regia digest (method 511P) in 2016 which was replaced in 2019 by a 4-acid digestion (method 306P/M). 7.3 QAQC Measures [§229.601(b)(96)(iii)(B)(8)(iii)] Keliber’s QAQC programme included the insertion of four internal CRMs (comprising three reference materials and a blank material) and quarter-core replicates into the sample batches submitted to the laboratory. The analytical laboratory (Labtium) conducted internal QC through use of one CRM (AMIS0355) and pulp duplicate samples, which are also monitored by Keliber. Small batches of sample pulps have also been sent to check laboratories at various stages. 7.3.1 Certified Reference Materials Keliber have relied on the use of four internal reference materials prepared by Keliber and one external CRM, AMIS0355, supplied by African Mineral Standards (AMIS). The Keliber reference materials comprise three prepared from mineralised pegmatite (CRMs KEL2010-A, -B, and -C), spanning a range in grade from 0.62–1.05% Li (1.33–2.26 % Li2O), and a blank material locally sourced and prepared from unmineralised Lumppio granite (CRM KEL2010-D). The Keliber CRMs were prepared and certified by Labtium (Myöhänen, 2011). The AMIS0355 CRM was included as part of Labtium’s internal QAQC programme and monitored by Keliber (Table 7-1). Table 7-1: Summary of expected values for Keliber’s internally sourced reference materials and commercially sourced AMIS0355 (source: Keliber, 2023). Reference Material Source Digest Method Expected Li Value (%) 2 x SD (%) Expected Range (±2SD) Keliber A Keliber Fusion 1.0504 0.0683 0.9821–1.1187 Keliber B 0.7502 0.0315 0.7187–0.7817 Keliber C 0.6227 0.0262 0.5965–0.6489 AMIS0355 African Mineral Standards (Volte Grande, Brazil) Fusion (uncertified) 0.8063 0.1627 0.6436–0.9690 4-acid (certified) 0.7268 0.0836 0.6432–0.8074 The CRM samples are selected randomly for insertion in a sample stream at a ratio of 1 in 20. Figure 7-1 is a summary of the performance of the Keliber CRM observations assayed by fusion method from 2010 to 2023 related to Syväjärvi, Rapasaari, Emmes, Länttä, Outovesi, Tuoreetsaaret, and Leviäkangas. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 70 Reference Material Control Charts from 2010 to 2020 in Analytical Order for the Keliber Reference Materials Figure 7-1: Reference material control charts from 2010 to 2020 in analytical order for the Keliber reference materials (source: Keliber, 2023). The performance of the three in-house CRMs (A, B, and C) is consistently within a narrow range but below the expected reference values, with some below the lower limit of -2xSD (particularly for B), and with a few values above the expected value (Figure 7-1). The scatter of data for each of the materials since 2014 has been attributed to potential sample inhomogeneity within all three of these CRMs. AMIS0355 has been included in the internal laboratory QAQC programme since 2016 and monitored by Keliber (Figure 7-1). The performance of AMIS0355 is slightly above the expected mean and within the 2xSD limits of the 4-acid method (Figure 7-2). However, when compared to the uncertified fusion value, it is consistently lower, but also within the 2xSD range for the fusion method – similar to the trend observed with the Keliber reference materials (refer to Table 7-2).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 71 Observations of AMIS0355 Values since its Introduction to Laboratory Internal QC Protocol in 2016 Figure 7-2: Observations of AMIS0355 values since its introduction to laboratory internal QC protocol in 2016. Blue dashed line is uncertified value for fusion method and green dashed line is certified value for 4-acid digestion (source: Keliber, 2023). Table 7-2: Performance of AMIS0355 and Keliber’s reference materials over the period 2016 to 2023 at Labtium (source: Keliber). Reference Material Labtium Kuipio for Period 23/09/2016 to 06/08/2021 Oulu for Period 30/09/2021 to 28/08/2023 n Mean 2 x SD Relative Difference from Expected Value (%) n Mean 2 x SD Relative Difference from Expected Value (%) Keliber A 73 1.010 0.062 -3.88 25 0.965 0.096 -8.10 Keliber B 70 0.719 0.042 -4.13 31 0.696 0.076 -7.17 Keliber C 63 0.606 0.028 -2.74 31 0.580 0.062 -6.88 AMIS0355 414 0.758 0.04 -5.95 66 0.717 0.044 -11.11 It is also noted that the “step down” in reported values from late 2021 (observed in both the Keliber reference material and AMIS0355) is attributed to some organisational changes in the laboratory service from Kuopio Eurofins-Labtium to Oulu Eurofins-Ahma due to closing of Eurofins-Labtium Kuopio in September 2021. This negative bias of the reference materials would suggest that the grades reported by Keliber are potentially on the conservative side. It also suggests that the original certified values reported by Labtium in 2011 need to be relooked S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 72 at, and a more robust recertification of the material needs to be conducted. Despite these variances in the expected CRM’s values, they fall within a narrow range (Table 7-2) and are not considered significant enough to impact the quality (accuracy) of analyses generated to date. Although this underreporting may be regarded as minor and the variability low, it is more pronounced in the data reported by the Oulu Laboratory and mainly affects the Tuoreetsaaret samples submitted since 2021. This has been brought to the attention of the exploration team and various actions are being investigated to resolve the matter going forward. The variance from the expected values and changes in the reported Li contents of the reference materials over time have also been observed by other QPs in the past (Payne, 2022 and SRK, 2023). 7.3.2 Blanks Blank pulp (samples containing negligible amounts of the element of interest) is inserted into the sample stream to assess whether any potential contamination has been introduced during the sample preparation stages. The blank used by Keliber was also part of the same suite of in-house CRMs prepared by Labtium. However, results from this CRM do not show any significant contamination throughout all batches prepared by Labtium (Figure 7-1). 7.3.3 Core Replicates and Lab Pulp Duplicates Core replicates comprising quarter-core samples were taken randomly for insertion in the sample stream at a ratio of 1 in 20 and were cut from the remaining half-core, replicating the half-core sample preceding the replication in the sample sequence. Results are plotted in Figure 7-3. The replicates show reasonable repeatability. The variance observed between the replicate pairs is expected and attributed to a combination of the different sample sizes (i.e. half core vs ¼ core) and the coarse-grained and heterogenous nature of the spodumene mineralisation. Annual paired T-tests replicated by Keliber show no statistical significance for the difference of their means (Kurtti, 2019, 2020, 2021 & 2022; Lamberg, 2018). As part of Labtium’s internal laboratory protocol, pulps were selected at a frequency of approximately 1 in 20 as a pulp duplicate for re-assay. The results show an acceptable level of repeatability of laboratory analysis with no observable bias (Figure 7-4). Relative differences of pulp re-assays vs the primary/reference sample show typical scatter to form a Horwitz’s trumpet towards the method detection limit (Figure 7-5). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 73 Summary of Core Replicate Results for the Period 2010–2023 using Fusion Method 720P Figure 7-3: Summary of core replicate results for the period 2010–2023 using fusion method 720P (source: Keliber, 2023). Summary of Laboratory Pulp Duplicate Pairs for the Period 2010–2023 using Fusion Method 720P Figure 7-4: Summary of laboratory pulp duplicate pairs for the period 2010–2023 using fusion method 720P (source: Keliber, 2023). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 74 Absolute Value of Relative Difference between Pulp Re-Assays and Reference Sample vs Li% for the Period 2010–2023 using Fusion Method 720P Figure 7-5: Absolute value of relative difference between pulp re-assays and reference sample vs Li% for the period 2010–2023 using fusion method 720P (source: Keliber, 2023). 7.3.4 Inter-Laboratory Checks Inter-laboratory checks were conducted in 2014 (Sandberg, 2014), 2022, and 2023 (Kurtti, internal report) based on anomalous reference material results. In 2013, ALS laboratory was used for the analysis of exploration samples using a 4-acid digestion method Li-ICP61. However, Keliber QC reference materials used reported significantly lower-than-expected results, hence an inter- laboratory check was conducted by Sandberg (2014). A total of 510 samples, comprising 417 from Syväjärvi and 93 from Leviäkangas, were re-assayed. The selected samples represented the majority of the samples from the spodumene pegmatite (SPG) zones and with grades >0.5% Li2O (>0.23% Li) from these two pegmatites. The results showed the Labtium fusion results for Li2O to be 13.8% (Syväjärvi) and 10.4% (Leviäkangas), higher than the ALS 4-acid digest (Li-ICP61) results (Figure 7-6) (Sandberg, 2014). As a consequence, the assay laboratory and method were changed to Labtium Rovaniemi using the fusion digestion method 720P. The small number of samples not re-analysed at the time were and are not considered to be material to the overall database integrity (Sandberg, 2014). Inter-laboratory checks conducted in 2022 were done as a result of the observed changes in reported reference material values after laboratory change from Eurofins Labtium Kuopio to Eurofins Ahma Oulu (Figure 7-1 and Figure 7-2). Two batches of pulp rejects were selected, one from samples originally assayed in Kuopio and the other from samples originally sent to Oulu, and sent for re-assay by fusion to ALS (Ireland). The results are plotted in Figure 7-7 and show that the change to Ahma Oulu resulted in underreporting of lithium by about 6.3% relative, with less precision compared to ALS, and results reported by Labtium Kuopio are on average 1.85% higher relative to ALS (Keliber comms, 13 March 2024). Additional batches were subsequently sent to CRS Laboratories Oy in Kempele, Finland, and the results showed a similar trend to those reported by ALS.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 75 Inter-Laboratory Checks Conducted in 2014 by Labtium (Fusion Method 720P) vs ALS (4-Acid Method) Figure 7-6: Inter-laboratory checks conducted in 2014 by Labtium (fusion method 720P) vs ALS (4-acid method) (source: Sandberg, 2014). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 76 (A) Plot of 2022 Inter-Laboratory Check – Kuopio (Blue) and Oulu (Orange). (B) Normal Distribution of 2022 Inter- Laboratory Check showing Relative Difference of Paired Samples Figure 7-7: (A) Plot of 2022 inter-laboratory check – Kuopio (blue) and Oulu (orange). (B) Normal distribution of 2022 inter- laboratory check showing relative difference of paired samples. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 77 7.4 Adequacy of Sample Preparation, Security, and Analytical Procedures [§229.601(b)(96)(iii)(B)(8)(iv)] Keliber have developed and implemented well-defined core processing, logging, sampling, and analytical procedures since 2013. The core processing, sampling, and storage facility in Kaustinen is considered a secure facility, with the sample preparation and analytical methodologies considered appropriate for pegmatite-hosted lithium mineralisation. The internal reference materials (which were certified by Labtium), as well as the commercially available CRM AMIS0355, which are used consistently, under report lithium within a relatively narrow range for the fusion method relative to the expected values which, although not considered material to the reliability and accuracy of the assay results, suggests a number of potential issues that warrant further investigation and resolution going forward. These issues could be one or a combination of: • The original certified values reported by Labtium in 2011 need to be relooked at and a more robust recertification of the material conducted. • Changes and tweaks in the analytical methodology by the assay laboratory. • Some of the variance may be attributable to the inhomogeneity of the Keliber reference sample material. It should also be noted that the 2 x SD limits determined by Labtium represent a very small error tolerance ranging from ±4.2% (Keliber-B and C) to ±6.5% (Keliber-A) when assessing their performance compared to the commercially available AMIS0355, which ranges from ±11.5% for the 4-acid method (certified) to 20.2% for the uncertified fusion method. Other AMIS lithium CRMs also have 2 x SD limits/tolerances that range from 7–12%. However, in contrast, recent inter-laboratory checks using ALS (Ireland) suggest that the results up to September 2021 i.e. samples assayed at Labtium, are very similar, and subsequent samples assayed at Ahma Oulu under report lithium by around 6%. As such, the lithium values reported are potentially slightly conservative but of sufficient quality and accuracy for Mineral Resource estimation purposes. The sample database is also considered to be of sufficient quality and accuracy for use in Mineral Resource estimation. The QP recommends the following: • Keliber engage an umpire/check laboratory to analyse an additional subset of the previously analysed samples spread across various deposits and time periods representative of the grade range of the deposits. This should provide a better understanding of whether the apparent underreporting of lithium in the internal reference material and AMIS0355 is related to laboratory performance or improperly determined expected reference values and ranges for these materials or a combination of the above. • Inclusion of additional commercially available CRMs as part of its QC programme in the future across a broader lithium range. • Inclusion of coarse-crush blank material to monitor potential cross-contamination in the crushing stage of the sample preparation, in addition to the pulp blanks currently used by Keliber. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 78 • Further round-robin testing of the in-house reference materials at other laboratories is recommended, should Keliber wish to continue using these. Alternatively, Keliber should consider engaging a commercial laboratory or company (e.g. AMIS) to prepare and certify material from the Keliber deposits. • Continue to engage with Ahma Oulu Laboratory with regards to the performance of the 720P fusion method or look to engage an alternative laboratory for analysis of exploration samples. 7.5 Unconventional Analytical Procedures [§229.601(b)(96)(iii)(B)(8)(v)] No unconventional analytical procedures have been employed by Keliber.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 79 8 DATA VERIFICATION [§229.601(b)(96)(iii)(B)(9) 8.1 Data Verification Procedures [§229.601(b)(96)(iii)(B)(9)(i)] The verification of the Keliber exploration data comprise the following exercises: • A site visit during which Keliber offices in Kokkola and the exploration office and core handling and sampling facility in Kaustinen were visited. During the site visit, a review of Keliber’s exploration history, operating procedures, drill core intersections, and data was conducted, and various areas of the Project sites were visited (Syväjärvi, Rapasaari, Tuoreetsaaret, and the proposed concentrator site). • Review of Keliber’s operating procedures during the site visit; this included documentation and discussions with Keliber staff that guide the core handling, core logging, sampling and QAQC, assay methods, sample logging, and sampling data. The review encompassed both historical and current exploration. • Review of selected drill hole logs (geological, sample, and assay) from all deposit databases against available drill core photography. • Comparison of historical and current assay and geological data (see Section 8.3.4 and completed as part of the geological modelling process discussed in Chapter 10). • Review of company reports summarising the geology, exploration, and QAQC results. • A review of Keliber’s QAQC protocols and implementation thereof related to the monitoring assay laboratory performance and review of the relevant QAQC data (see Chapter 7). ERM have relied on Keliber to provide the necessary assay QAQC plots and compilation of the drilling history. • A review of public domain literature including GTK reports and academic papers covering the exploration history, geology, and evaluation of LCT pegmatites in the Kaustinen area, many of which include references to the deposits referenced in this study. • Review of Keliber’s reports related to the historical and current exploration specific to the deposit data used to inform the MRE. This included a review of the verification process of the data by SRK Consulting South Africa (Pty) Ltd (SRK) (2023). Relevant aspects of the above are detailed below. 8.2 Site Visit A site visit was conducted by the ERM QP from 11 to 13 July 2023. The Project site was visited on 11–12 July 2023 and included: • A visit to Keliber’s geology office and core processing and storage facility in Kaustinen where a selection of drill holes was reviewed against the logs and assay certificates, and a review of the logging and sampling protocols was performed. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 80 • A field visit to three of the deposits, namely Syväjärvi, Rapasaari, and Tuoreetsaaret, during which the location and field confirmation of a selection of drill holes and a number of outcrops around these deposits was completed. • Discussions with Keliber geologists, Pentti Grönholm (Senior Manager Geology) and Joonas Kurtti (Exploration Geologist), regarding the Project exploration history, geology, core processing, operating procedures, and data collection processes. Keliber’s main office in Kokkola was visited on 13 July 2023. 8.2.1 Core Processing and Storage Facility The core processing (Figure 8-1 and Figure 8-2) and storage facility in Kaustinen was inspected with regards to core and sample receipt as well as the flow of core through the various processing stations to sample despatch. The facility comprises a secure weatherproof storage warehouse and adjoining office and work area split into discrete areas to accommodate Keliber’s workflows, namely a core receiving and storage, core logging, sample splitting, sample packing, and density measurement areas. Signage and summaries of key processes are well displayed. During the visit, the facility was clean and orderly. (A) Keliber’s Core Processing Facility at Kaustinen; (B) Angled Core Racks used for Core Processing and Logging Figure 8-1: (A) Keliber’s core processing facility at Kaustinen; (B) Angled core racks used for core processing and logging. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 81 Keliber’s Core Receipt and Storage Facility Adjoining the Processing Facility Figure 8-2: Keliber’s core receipt and storage facility adjoining the processing facility. (A) Stacked core boxes, and (B) sealed crates of coarse and pulp rejects received back from the laboratory. When core arrives at the facility, it is core is loaded onto ergonomically designed, angled, well-lit core racks (Figure 8-1) and inspected, and depth checks, and core markup are done. The core is then logged (collecting lithological, mineralogical, structural, and geotechnical data) onto paper logs and marked up for sampling. Structural logging is currently done using a Reflex IQ-Logger™; previously, these data were captured using a goniometer. The core boxes are then photographed wet and dry. The core to be sampled, comprising pegmatite intervals (including spodumene- bearing and muscovite pegmatite) and selected host rock intervals, is split with a diamond saw in a separate room (Figure 8-3A), and samples are then put into baskets (Figure 8-3B) to dry. Every 10th sample is then taken and density measurements, using a standard Archimedes-type technique of weighing the core dry and wet, are completed. (A) Core Saw, and (B) Cut Sample in Baskets Prior to Density Measurements and Packing Figure 8-3: (A) Core saw, and (B) cut sample in baskets prior to density measurements and packing. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 82 Samples are then packaged and QAQC samples are inserted into the sample sequence as per the operating procedure and submitted for assay. Pegmatite samples are assayed by a peroxide fusion method (method 720P) for lithium and a selection of elements at the assay laboratory in Oulu; host rock samples are assayed by a multi-acid digest for a multi-element suite. The logged core is then packed and stored in the adjacent core storage facility (Figure 8-2A). All coarse and pulp reject material is also returned from the analytical laboratory and stored in crates at the Kaustinen facility (Figure 8-2B). The facility also contains some of the historical core. However, the majority of the older core is stored offsite at a facility in the town of Kemi (300 km north-northeast of Kaustinen), and the core drilled in the period 1960–1980 is stored and curated by GTK. Cores from various drill holes were examined on the logging tables and verified against the geological logs and assay data and summarised below. 8.2.2 Drill Hole Verification and Field Checks A number of drill hole locations and sites of geological interest at Syväjärvi, Rapasaari, and Tuoreetsaaret were visited during the site visit. The drill hole locations were collected by handheld global positioning system (GPS) (Figure 8-4) and hole orientation was measured off the casing left in the top of the holes to mark the holes. These measurements compared well with the data in Keliber’s databases; all holes checked were within 2.5–7.5 m of the surveyed locations. The list of holes with the surveyed locations and measured locations is tabulated in Table 8-1. Outcrop of the host rocks is sparse (Figure 8-5) and largely covered by thickly vegetated soil comprising peat and glacial till and, naturally, no outcrops of pegmatite were observed during the site visit. Thin pegmatite veins in the hanging wall schists at the main Syväjärvi deposit are exposed in the sidewall of the water-filled pit at the entrance to the portal (Figure 8-6). A number of the spodumene-bearing pegmatite erratics, that were key in the discovery of the pegmatites in the area, were observed around Syväjärvi and Tuoreetsaaret (Figure 8-7). Photo of Drill Hole S76 (Syväjärvi) Figure 8-4: Photo of drill hole S76 (Syväjärvi), checked in July 2023.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 83 Table 8-1: List of drill holes field-checked during site visit. Hole ID D e p o si t Surveyed GPS Difference (m) Orientation Comments X (m) Y (m) Long (DD) / X (m) Lat (DD) / Y (m) S-75 Sy vä jä rv i 2490444.2 7062516.6 23.8034 / 2490446.2 63.6637 / 7062516.4 2.0 Measured 64/252 Surveyed 65/270 S-76 2490475.0 7062516.7 23.8041 / 2490478.2 63.6637 / 7062515.7 3.4 Measured 56/260 Surveyed 64/270 S-93 2490443.1 7062500.0 23.8034 / 2490444.4 63.6635 / 7062499.7 1.3 Measured 43/083 Surveyed 45/089 S-160 2490458.1 7062579.7 23.8037 / 2490460 63.6642 / 7062579.4 1.9 Measured 50/090 Surveyed 51/090 Casing removed S-161 2490441.9 7062680.4 23.8034 / 2490443.5 63.6651 / 7062677.8 3.1 Measured 65/090 Surveyed 65/087 RA-17 R ap as aa ri 2492433.9 7060980.0 23.8436 / 2492431.8 63.65 / 7060985.1 5.5 Measured 48/076 Surveyed 45/090 RA-31 2492370.3 7060980.1 23.8423 / 2492366.2 63.6499 / 7060982.3 4.6 Measured 48/078 Surveyed 44/089 RA-33 2492372.4 7060939.8 23.8424 / 2492372.7 63.6496 / 7060942.7 2.9 Measured 42/082 Surveyed 44/089 RA-36 2492395.6 7060900.2 23.8428 / 2492393.7 63.6492 / 7060904.8 5.0 Measured 40/080 Surveyed 45/088 RAPI-20 2492422.0 7060996.0 23.8434 / 2492422.4 63.65 / 7060993.4 2.6 Measured 48/302 Surveyed 46/295 Unmarked in field PD1 2492420.0 7060999.0 23.8434 / 2492419.3 63.6501 / 7061002 3.1 Measured 40/032 Surveyed 45/298 Unmarked in field S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 84 Hole ID D e p o si t Surveyed GPS Difference (m) Orientation Comments X (m) Y (m) Long (DD) / X (m) Lat (DD) / Y (m) RA-256 Tu o re et sa ar et 2491381.9 7060980.1 23.8223 / 2491378 63.65 / 7060986.1 7.1 Measured 65/080 Surveyed 45/090 RA-316 2491402.5 7060939.8 23.8228 / 2491400.3 63.6496 / 7060946.3 6.9 Measured 45/254 Surveyed 44/270 RA-317 2491409.6 7060859.9 23.823 / 2491411.3 63.6489 / 7060867.3 7.6 Measured 40/252 Surveyed 44/271 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 85 Host Rock Outcrops from Syväjärvi Figure 8-5: Host rock outcrops from Syväjärvi. (A) Plagioclase-bearing porphyrite (metavolcanic) (WPT840), and (B) sulphide- bearing mica schist (metasediment) (WPT844). Photo Looking East of Host Schists and Thin Northerly-Dipping Pegmatite Veins in Hanging Wall to Main Spodumene Pegmatites Exposed in Water-Filled Pit at Entrance to the Portal at Syväjärvi Figure 8-6: Photo looking east of host schists and thin northerly-dipping pegmatite veins in hanging wall to main spodumene pegmatites exposed in water-filled pit at entrance to the portal at Syväjärvi. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 86 Tuoreetsaaret: (A) Erratic of Spodumene-Bearing Pegmatite in Forest with (B) Large Spodumene Lathes (>20 cm long) showing Uniform Crystal Orientation Figure 8-7: Tuoreetsaaret: (A) Erratic of spodumene-bearing pegmatite in forest with (B) large spodumene lathes (>20 cm long) showing uniform crystal orientation (interpreted to be perpendicular to host rock contacts) (WPT848). 8.2.3 Site Visit Conclusion The QP is satisfied that the necessary steps in the data collection process were taken to verify the data used for the MRE. 8.3 Check Logging, Database Verification, and Validation The core handling, processing, sampling, and core logging are guided by an SOP (Sandberg, c.2013). Data are collected that assist the interpretation of both the pegmatite and country rock geology. The logging data are collected on Microsoft Excel® sheets and then captured into a Microsoft Access® database (refer to Figure 8-8 and Figure 8-9). During the site visit, a number of drill holes were laid out by Keliber and check logging was conducted against the geological logs, the sample intervals, and assay certificates provided. A summary of the holes checked and the findings are summarised in Table 8-2. Table 8-2: List of drill holes (geological and sample logs and assay certificates) checked against drill holes during site visit. Hole ID Deposit Year Drilled Finding Risk RA-156 Rapasaari 2018 Box 30 – some pegmatite sampling included 2 cm of host rock. Dilution of Li grade and artificial elevation of iron (Fe) in mineralised pegmatite intervals. Box 33 – unsampled pegmatite that appears unmineralised. Missing potentially mineralised pegmatites. RA-14 Rapasaari 2014 Box 2 – core loss interval not captured and sampled across loss; not common. Poor logging and sampling practices BUT core losses now captured in logs and sampling across core losses avoided where possible.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 87 Hole ID Deposit Year Drilled Finding Risk Box 32 – included unsampled thin pegmatite that is logged as mineralised “spodumene pegmatite” (Figure 9-8). Missing mineralised pegmatite assay data. S-22 Syväjärvi 2013 Pegmatite logged interval (50 cm) as unmineralised “muscovite pegmatite” contains 1.91% Li2O (Figure 9-9). Misclassification of pegmatite type in geological models. OV-27 Outovesi 2012 No issues. Hole RA-14 (Box 32) with an Interval of Unsampled Pegmatite Logged as SPG and Host Rock Figure 8-8: Hole RA-14 (Box 32) with an interval of unsampled pegmatite logged as SPG and host rock. Hole S-22 (Box 19) showing Samples 30371 (60.3–61.2 m) Logged as Muscovite Pegmatite and 30372 (61.2–62.5 m) Logged as SPG with High Lithium Content Figure 8-9: Hole S-22 (Box 19) showing samples 30371 (60.3–61.2 m) logged as muscovite pegmatite and 30372 (61.2–62.5 m) logged as SPG with high lithium content. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 88 8.3.1 Observations and Comments Some general observations regarding the logging based on the review of the four drill holes above include: Most of the sampling is done from and to geological contacts and also split according to pegmatite composition i.e. muscovite pegmatite (MPG/MPEG) was logged and sampled separately from the mineralised spodumene-bearing pegmatite (SPG/SPEG). Samples within the large spodumene-bearing pegmatite intervals are split based on mineralogy or texture. This is considered aligned with industry-accepted best practice for logging and sampling of pegmatite- hosted lithium mineralisation. A number of areas of concern, although they do not appear to be a common occurrence and/or are not considered to materially impact the integrity of the databases, include: • A number of the smaller muscovite pegmatites were not sampled. These are usually restricted to thinner pegmatites in the hanging wall or footwall to the spodumene-bearing pegmatites. • Impact of weathering, where present, on the core checked appears to be minimal with regards to the lithium content of spodumene, which is still crystalline but does appear to have some alteration. However, the pegmatite is more friable due to the alteration of the feldspar (altered to a red colour) to clay materials and incipient alteration of the spodumene (see also Section 5.3). This is something that should be looked at going forward, to determine if it poses a risk to the processing and recovery of spodumene from the pegmatites. • Some of the older Keliber sampling was done across core losses, but this practice has largely stopped and core losses are logged in the lithology log and not sampled across. • Some smaller spodumene-bearing pegmatite intervals within host rock are unsampled. 8.3.2 Database Checks [§229.601(b)(96)(iii)(B)(9)(ii)]] Checks of a number of drill holes from the different deposits were done and included drilling from the 1960s, 1980, and more recent drilling, 2012 to 2023 (Table 8-3). The check reviewed the various deposit databases and compared the geological logging and reported lithium assays with the sample intervals against the core photographs supplied by Keliber. Table 8-3: Checks conducted on drill holes from various campaigns. Deposit Hole ID Year Drilled Finding Risk Emmes 61-R6 1961 No issues. 61-R8 No issues. 63-R4 1963 No issues. 63-R6 Sections in interval not sampled (mostly host rock). Pegmatite material all sampled. Poor sample practice but not commonly observed. 63-R10 Core photos but not in database. No central database and no strict version control on databases. 66-R1 1966 Core photos but not in database. No central database and no strict version control on databases 66-R2 No issues. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 89 Deposit Hole ID Year Drilled Finding Risk 67-R19 1967 No issues. 67-R31 Waste interval does not appear sampled based on photos but has assay values. Possibly sampled after photo taken. 80-R12 1980 Not in database. E-1 2014 No issues. E-8 No issues. E-10 No issues. E-13 2019 No issues. E-16 No issues. E-22 No geological and sample log. No pegmatite form. No central database and no strict version control on databases. E-23 Geological and sample log only for SPG entry. No assay data. No central database and no strict version control on databases. Rapasaari RA-15 2014 No issues. RA-289 2021 Thin pegmatites unsampled between 1 cm discrepancy between pegmatite (72.85–73.24) and sampling (72.85– 73.25). Makes geological modelling difficult. RA-370 2023 Small (15–20 cm) unmineralised pegmatites logged as MPG at ~138 m, 156 m, and 158 m not sampled. May miss potentially significant mineralisation. Lantta R4-3 ? No issues. R4-19 ? Historical Drilling Box 8 – pegmatite in photo not logged or sampled ~49.5 m BUT log has a pegmatite logged from 59.5–59.91 m but NOT visible in photos. Inaccurate logging possibly a typo and meant to be 49.5–49.91 m results in inaccurate geological models. Unsampled pegmatites may miss potentially significant mineralisation. L-25 ? No issues. L-40 ? No issues. L-46 ? Not all MPG (unmineralised muscovite pegmatite) intervals sampled. May miss potentially significant mineralisation. Leviäkangas LE-8 2017 No issues. LE-38 2019 No major issues. Some thin unmineralised pegmatites unsampled. May miss potentially significant mineralisation. LE-51 2022 No issues. Outovesi OV-1 ? No issues. OV-10 ? No issues. OV-27 ? Photos but not in database. No central database and no strict version control on databases. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 90 Deposit Hole ID Year Drilled Finding Risk Potentially significant data omitted from database. OV-29 ? No issues. Syväjärvi S-2 2013 No issues. S-18 2013 No issues. S-780 2016 Pegmatite interval in photos from 56.15 – 56.8 m logged as mica schist. Was sampled and carried no grade. Possible typo. Need some internal checks and QC on logging process to avoid errors. Could potentially have been a mineralised pegmatite. S-98 2018 No issues. S-141 2019 No issues. S-159 2023 No photos or lithology logs. Assays present. No central database and no strict version control on databases. Potential data omitted from database. S-162 2023 No issues. Tuoreetsaaret Core photography provided for 5 holes only mostly outside main resource area. RA-259 2020 Some apparently barren pegmatites not sampled. May miss potentially significant mineralisation. RA-260 2020 No issues RA-262 2020 No issues RA-264 2020 Some pegmatites logged as barren MPG not sampled. May miss potentially significant mineralisation. RA-365 2022 Some pegmatites logged as barren MPG not sampled. Top part of pegmatite interval from 117.2–122.75 m not sampled from 117.2–118 m (logged as MPG). May miss potentially significant mineralisation. Makes geological modelling and estimation difficult. 8.3.3 Observations and Comments [§229.601(b)(96)(iii)(B)(9)(iii)] Some general observations regarding the logging based on the review of the databases provided for each of the deposits: • Structure of databases varies from deposit to deposit although lookups for geological logs are generally consistent. • Some data within the databases is not relevant to the specific deposit e.g. Rapasaari database contains drill holes from Tuoreetsaaret. • Muscovite pegmatites are often not sampled. Generally, where they are sampled, they return lithium values of <0.1% Li2O, suggesting the logging is generally accurate. However, there is the chance some lithium

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 91 mineralisation may be missed by not sampling these pegmatite zones (e.g. Hole S-22 interval 60.3–61.2 m was logged at MPG but 1.38% Li2O – see Figure 8-9 and Table 8-1). • Sampling of the historical drill holes consisted of samples taken using a guillotine and did not always produce a consistent sample. It is noted that the historical drilling and data comprise a small proportion of the data used in the MREs. See also Section 8.3.4 reviewing the historical data against the current Keliber data. 8.3.4 Review of Historical Drilling against Keliber Drilling Recent and historical assay data (drilled by operators prior to Keliber Oy) within individual deposits are compared, and results for Rapasaari, Syväjärvi, and Emmes are presented. The statistics of the Li2O% distributions are reviewed, and datasets are filtered for drilling within the interpreted spodumene pegmatite estimation zones. Figure 8-10 shows the Li2O% distributions on probability plots for recent (blue line) and historical (green line) drill assays. The mean grades and the variance between historical and recent data are tabulated in Table 8-4 along with the proportions of historical data. The larger deposits (Rapasaari and Syväjärvi) have a small proportion of historical data (<20%) within the spodumene pegmatite zone and mean grades are within a 20% variance. The distributions for Li2O% are comparable in all cases and the historical data are considered appropriate for use in resource estimation. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 92 Comparison of Recent (Blue) and Historical (Green) Li2O% Assay Data within the Interpreted Spodumene Pegmatite Zones at Rapasaari (Top), Syväjärvi (Middle), and Emmes (Bottom) Figure 8-10: Comparison of recent (blue) and historical (green) Li2O% assay data within the interpreted spodumene pegmatite zones at Rapasaari (top), Syväjärvi (middle), and Emmes (bottom). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 93 Table 8-4: Summary of historical drill data review within interpreted spodumene pegmatite zone. Deposit Parameter Historical Recent Variance on Mean Rapasaari Mean Li2O% 1.34 1.20 11% Number of samples 94 2,010 Proportion of samples 4% 96% Syväjärvi Mean Li2O% 1.12 1.36 -18% Number of samples 153 781 Proportion of samples 16% 84% Emmes Mean Li2O% 1.29 1.27 2% Number of samples 109 89 Proportion of samples 55% 45% 8.4 QP Opinion and Recommendations The QP considers data management an area of future improvement in terms of data integrity and security. The implementation of a fit-for-purpose relational database with timely backups will ensure a robust and secure database going forward. In addition, it will make data extraction, assay management, data interrogation, and export simpler and will avoid version control issues and make auditing more traceable. It is understood that Keliber are in the process of implementing a database solution for the entire Project. The QP was unable to verify that paper logs were accurately transcribed into the digital database, however, during the check of a number of drill holes, some errors have been identified but not considered material to the overall data integrity. The overall work completed to date has captured all the important variables (mineralogical, structural, and lithological) required to properly define the attitude of the host pegmatites and the spodumene or grade distribution within the various pegmatites that host each deposit. It is, however, recommended that Keliber conduct regular reviews and checks of their exploration drill holes in terms of geological logging and sampling as part of their internal QC protocols. Additional checks on whether logs are correctly captured into the database are recommended or the implementation of a digital logging platform where the logging and sampling data are captured directly into the database. The checks done on the Rapasaari, Syväjärvi, and Emmes historical and recent assay data indicate that the datasets are largely comparable and suitable for use in the geological models and MREs. Based on these comparisons, the differences in the historical sampling (use of a guillotine to split core) compared to the more recent exploration (use of a core saw to split the core in half) are noted but are considered to be a minimal risk to the Project. Additional checks on the historical data used in the Lantta and Leviäkangas deposits, which were generated over similar periods to that of Rapasaari, Syväjärvi and Emmes, are recommended going forward. A small number of unsampled pegmatite intervals were noted during the checks and not considered to materially impact the data. However, it is recommended that all mineralised pegmatite intervals, irrespective of size, particularly around larger pegmatites, are sampled as well as apparently unmineralised muscovite pegmatites. Overall, the QP considers the exploration data used in this TRS in support of the mineral estimate to be accurate and representative and to have been generated with industry-accepted standards and procedures. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 94 9 MINERAL PROCESSING AND METALLURGICAL TESTING [§229.601(b)(96)(iii)(B)(10)] The first metallurgical tests were completed in the early 1970s by Paraisten Kalkkivuori Oy. Keliber began studying the deposits in 1999 and between 2001 and 2006, in partnership with Outotec, developed a new lithium carbonate production process. More intensive investigations started in 2014. In June 2018, Keliber completed an FS for a project to produce battery-grade lithium carbonate from spodumene-rich pegmatite deposits in Central Ostrobothnia, Finland. However, following further market studies it was decided to consider the production of battery-grade lithium hydroxide monohydrate (LiOH·H2O), or more simply lithium hydroxide (LiOH), instead of lithium carbonate. A series of tests was completed to determine the production parameters of lithium hydroxide from spodumene ore. Engineering studies were undertaken to produce 12,500 tpa of battery-grade lithium hydroxide (BG) via the following unit processes: • Concentration comprising crushing, optical sorting, grinding, and flotation to produce a spodumene concentrate; • Conversion of the spodumene concentrate from alpha to beta-spodumene by roasting in a rotary kiln; and • Soda leaching in an autoclave and hydrometallurgical processing including solution purification, crystallisation, and dewatering to produce lithium hydroxide. In January 2022, Keliber issued a draft FS (WSP Global Inc., 2022) based on the production of 15,000 tpa of BG. An FS was issued on 1 February 2022. 9.1 Nature and Extent of Testing and Analytical Procedures 9.1.1 Historical Metallurgical Testwork After the initial metallurgical tests conducted in the early 1970s, further investigation was undertaken between 1976 and 1982. The research included mineral processing tests to produce spodumene concentrate as well as its by- products: quartz, feldspar, and mica concentrates. Keliber restarted metallurgical testing in 2003, which led to the preliminary engineering for a spodumene concentrator and a lithium carbonate production plant. Mineral processing included two-stage grinding, gravity separation, de- sliming, pre-flotation, spodumene flotation, and dewatering. Conversion from alpha- to beta-spodumene was undertaken in a rotary kiln, and the hydrometallurgical process included pressure leaching of beta-spodumene in a soda environment, solution purification with ion exchange, and precipitation of lithium carbonate. Subsequent changes to the process route have principally been in the production of lithium hydroxide. 9.1.2 Recent Mineral Processing Testwork The purpose of the mineral processing circuit is to produce spodumene concentrate for the downstream pyrometallurgical and hydrometallurgical processes. Typically, commercial spodumene concentrate would target a grade of 6% Li2O. However, given that concentrate transportation costs to the relatively close KIP are low, the concentrate grade will be a point of optimisation. During the production phase, the concentrate grade will be optimised depending on the grade-recovery relationship and the price of the end product. Generally speaking, the production of lower-grade concentrate will be more feasible at high product price. The level of impurities in the

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 95 concentrate is also important. Keliber testwork programmes have revealed iron, arsenic, and phosphate to be the main impurities in the spodumene flotation concentrate that impact the downstream process. The maximum levels have been indicated at 2% for Fe2O3, 50 ppm for As, and 0.4% for P2O5. The concentrator plant design includes magnetic separation and pre-flotation processes to limit the Fe2O3 and P2O5 respectively, and only As is managed in the mine plan. 9.1.2.1 Länttä Pilot Test in 2015 In the 2015 PFS, samples of Länttä ore were tested at pilot scale. 9.1.2.1.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Three samples with a total mass of 14.8 t and combined grade of 1.27% Li2O, 0.0092% Nb, and 0.0024% Ta were processed through a pilot plant. The 2022 FS Report referenced the primary sample but did not describe the sampling details. 9.1.2.1.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The combined sample of Länttä ore was treated through a pilot mineral processing plant at the mineral processing and materials research unit (Mintec) of the GTK in Outokumpu, Finland. The spodumene concentrates produced were then processed by conversion and hydrometallurgical testing. This is described in Subsection 9.1.3: Laboratory-Scale Conversion Tests. GTK’s quality system consists of the following elements: • GTK’s quality manual; • SOPs; and • Appendices and reference material. The ISO 9001 2015 quality system standard is applied in all production-related activities, such as mapping and measuring, and the mineral technology laboratory’s research and process operations, among others. The quality system describes the flow of GTK’s processes so that everything related to customer service, the reliability and efficiency of operations, and environmental protection is defined to meet the requirements of the standard. 9.1.2.1.3 Mineral Processing Testing and Results The pilot plant included dense media separation (DMS), rod milling with gravity separation, and flotation. Sample preparation for the pilot plant test included crushing and screening into two fractions, 0–3 mm and 3–6 mm. DMS was executed separately for these size fractions. Fines were fed directly to the spodumene flotation circuit with a feed rate of 300 kg/h. Unfortunately, the pilot plant de-sliming cyclones were ineffective, resulting in sub-optimal flotation. Laboratory-scale flotation tests were accordingly used to complement the pilot plant results. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 96 It was shown that the combination of DMS and flotation resulted in a 2% to 3% increase in lithium recovery compared with simple flotation. Test results obtained from combined DMS and flotation indicated a recovery of 85.9% with a 4.59% Li2O in the spodumene concentrate. 9.1.2.2 Syväjärvi Laboratory Tests in 2015 9.1.2.2.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Laboratory-scale tests were carried out on Syväjärvi samples collected from drill cores with an average grade of 1.47% Li2O. Primary sampling details were not described in the 2022 FS Report. 9.1.2.2.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The draft 2022 FS does not state where these tests were conducted, but presumably, they were conducted at the GTK Mintec facility. The spodumene concentrates produced were then processed by conversion and hydrometallurgical testing. This is described in Subsection 9.1.3: Laboratory-Scale Conversion Tests. GTK certification details are provided in Subsection 9.1.2.1: Länttä Pilot Test in 2015. 9.1.2.2.3 Mineral Processing Testing and Results Laboratory-scale testwork included DMS and flotation with the purpose of comparing the metallurgical performance of Syväjärvi ore with the Länttä ore tested earlier in pilot scale. In addition, a concentrate was produced for subsequent leaching tests. Tests confirmed that Syväjärvi ore can be processed using a similar flowsheet to Länttä. Recoveries for a concentrate of 4.5% Li2O were higher than achieved with the Länttä sample: 90.0% when only flotation was applied and 93.5% when both DMS and flotation were used. The phosphorus content of the produced spodumene concentrate was higher in the DMS and flotation alternative: 0.59% P2O5 compared to 0.26% when only flotation was applied. 9.1.2.3 Syväjärvi Pilot Tests in 2016–2017 (PFS) 9.1.2.3.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] For the PFS, a tunnel was mined during the summer of 2016 to extract a bulk sample for pilot plant and other testing. Four breaks were mined from pure spodumene pegmatite at the end of the tunnel, which were separately stockpiled (Figure 9-1). The 160-t bulk sample of Syväjärvi ore had a grade of 1.445% Li2O. A waste rock sample containing 0.188% Li2O was also collected as dilution in mineral processing tests. A plan and long section showing the location of the tunnel relative to the Syväjärvi deposit are shown in Figure 9-2. 9.1.2.3.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] An ore sorting test programme was completed in the TOMRA sorting test facility in Wedel, Germany. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 97 Mineral processing tests were conducted at pilot-scale at the GTK Mintec facility in Outokumpu. The spodumene concentrates produced were further used in laboratory and pilot conversion tests, with the converted concentrate then used in both laboratory- and pilot-scale leaching tests. This is described in this report, Subsection 9.1.2.1: Länttä Pilot Test in 2015. It was noted that the full Keliber process was thus tested at pilot scale. Spodumene Pegmatite at the End of the Tunnel (Top) and Numbered Ore Piles Before Transport to GTK Mintek (Bottom) Figure 9-1: Spodumene pegmatite at the end of the tunnel (top) and numbered ore piles before transport to GTK Mintek (bottom). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 98 TOMRA is certified to the ISO 9001 and ISO 14001 quality system standards. GTK certification details are provided in Subsection 9.1.2.1: Länttä Pilot Test in 2015. 9.1.2.3.3 Optical Sorting Testing and Results Sorting tests were conducted using 4 t of Syväjärvi run of mine (ROM) spodumene-rich ore (20 mm to 100 mm in size) and 500 kg of black waste rock. The focus of the tests was to remove black plagioclase porphyrite waste rock from the plant feed. Syväjärvi Pilot Sample Location – Plan (Top) and Long Section (Bottom) Views Figure 9-2: Syväjärvi pilot sample location – plan (top) and long section (bottom) views. 100 m

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 99 TOMRA’s test set-up included a PRO secondary COLOR NIR, which consists of a colour line scan CCD camera and a near-infrared (NIR) scanner. The combination of these sensors takes advantage of the absorption fingerprint of minerals in the NIR wavelength range and the colour characteristics. Ore sorting was found to be effective in removing black waste rock from the ore feed at different artificial waste rock sorter feed compositions. The sorting results indicate around 12% of the mass and 3% of the Li2O were lost in sorting. After accounting for the 0–20 mm fine fraction that will bypass sorting, there was a mass rejection of 10.1% with 2.2% lithium losses. In 2018, complementary batch-scale tests were carried out on hand-picked samples from Syväjärvi, Länttä, and Rapasaari. In addition to the main aim of verifying the separation of the pegmatite ore from the dark country rock, tests were conducted to separate spodumene pegmatite from barren pegmatite. The separation of black country rocks from pegmatite (ore-bearing and barren) was achieved with the COLOR, NIR, and X-ray transmission (XRT) sensors. It was noted that the laser sensor could also be used to separate the spodumene-bearing ore from the barren pegmatite but that further testing at pilot-scale would be needed to verify the mass balance and possible lithium losses. 9.1.2.3.4 Mineral Processing Testing and Results Pilot plant tests using rod and ball mills operating in closed circuit, gravity concentration, and flotation were conducted in September 2016 at the GTK Mintec facility in Outokumpu. The pilot-scale processing was divided into two campaigns, with the first processing 71 t of material with a 10% waste (country rock) dilution and the second 73 t with a 3% dilution. The flowsheet was based on the Länttä pilot plant test but without DMS due to high P2O5 concentrations seen previously in Syväjärvi concentrate. The following key unit processes were included: • Crushing; • Grinding and classification; • Gravity concentration; • De-sliming; • Pre-float flotation; • Magnetic separation; and • Flotation. Results showed two subsets with one averaging 75% recovery at 5.3% Li2O and the other averaging 82% recovery at 4.7% Li2O. Based on the GTK report, the biggest lithium losses were in the primary de-sliming and the spodumene rougher tails, totalling 9–10%. Abrasion and crushing work indices were determined by Sandvik at their test centre in Svedala using Syväjärvi ore and waste rock samples from the pilot feed material, as summarised in Table 9-1. Table 9-1: Syväjärvi comminution characteristics. Material type Measurement Comment Abrasion Index Syväjärvi ore 0.40 Abrasive S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 100 Material type Measurement Comment Syväjärvi waste rock 0.27 Abrasive Crusher Work Index Syväjärvi ore 12.4 ± 1.9 Soft Syväjärvi waste rock 13.9 ± 1.8 Medium hard Rod Mill Work Index Syväjärvi ore 15.3 Hard Syväjärvi waste rock 16.7 Hard Ball Mill Work Index Syväjärvi ore 18.9 Hard Syväjärvi waste rock 12.6 Medium 9.1.2.4 Laboratory Flotation Tests for Länttä and Syväjärvi in 2016 9.1.2.4.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] More than 50 bench-scale, batch flotation tests were carried out in this phase of the investigation. The programme included the following sample materials: • Länttä deep ore drill core sample; • Syväjärvi drill core sample; • Outotec (TOMRA) sorting testwork samples; • Cyclone overflow from Syväjärvi pilot plant testwork 2016; • Slimes from Syväjärvi pilot plant testwork 2016; and • Upgraded, Syväjärvi pilot concentrate sample. The Länttä drill core sample was collected from three drill cores in the central part of the deposit. The samples were from depth levels between 20 m and 40 m and visible weathering was not observed from the drill cores. The waste rock was excluded from the batch float sample. The Syväjärvi drill core sample was collected from one drill core. The sample contained only spodumene pegmatites, and waste rock was excluded from the sample. The sample was taken well below the surface to compare the effect of weathering with the Syväjärvi pilot processing sample. 9.1.2.4.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The batch flotation tests were carried out at the GTK Mintec facility in Outokumpu. The concentrate was also upgraded for subsequent testing for spodumene conversion. GTK certification details are provided in Subsection 9.1.2.1: Länttä Pilot Test in 2015. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 101 9.1.2.4.3 Mineral Processing Testing and Results The focus of the programme was to optimise flotation conditions with the Syväjärvi and Länttä ore samples. The concentrate was also upgraded for subsequent testing for spodumene conversion. Average flotation results are shown in Table 9-2. It was noted that optimisation of the flotation conditions was generally successful. Table 9-2: Summary of flotation results. Feed Test Product Grade (% Li2O) Syväjärvi Drill core – 3.35 mm 40 Rougher Conc. 3.36 Cleaner Conc. 7 6.15 Calculated Feed 1.46 Länttä Drill core – 3.35 mm 7 Rougher Conc. 2.01 Cleaner Conc. 7 5.59 Calculated Feed 1.20 Syväjärvi Pilot run cyclone O/F 3% wt. 29 Rougher Conc. 3.22 Cleaner Conc. 7 6.29 Calculated Feed 1.36 Syväjärvi SPG 7&8 TOMRA 1 Rougher Conc. 3.37 Cleaner Conc. 7 6.00 Calculated Feed 1.59 9.1.2.5 Geo-Metallurgical Study in 2016–2017 9.1.2.5.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Sampling was designed by Keliber’s Chief Geologist and a total of 18 ore samples were collected from the Syväjärvi, Länttä, and Rapasaari deposits. 9.1.2.5.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The geo-metallurgical tests were carried out at the GTK Mintec facility in Outokumpu. GTK certification details are provided in Subsection 9.1.2.1: Länttä Pilot Test in 2015. 9.1.2.5.3 Geo-Metallurgical Testing and Results The study included modal analysis by Mineral Liberation Analysis, the determination of the chemical composition of spodumene by Energy Dispersive Spectroscopy, grindability tests, and diagnostic flotation tests. The flowsheet and the conditions in the diagnostic test were similar to the test developed for the Syväjärvi ore. The following properties relevant to processing were considered: lithium grade, spodumene grain size, alteration, wall rock type, and dilution percentage. Grindability was seen to correlate with spodumene grade in that the higher the grade, the greater the resistance to being ground. No significant difference was observed between deposits. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 102 In all the ores, the flotation performance was strongly dependent upon the spodumene head grade and wall rock dilution. The recovery of lithium at a concentrate grade of 4.5% Li2O increased with the lithium head grade, as shown in Figure 9-3. The wall rock dilution impacted negatively on the flotation performance. The diagnostic flotation tests showed a significant difference between deposits, with Syväjärvi showing the best performance, followed by Länttä and Rapasaari, as seen in Figure 9-4. Individual flowsheets, processing conditions, and optimisation will, therefore, be needed for each ore to maximise the metallurgical performance. Lithium Recovery as a Function of Feed Grade Figure 9-3: Lithium recovery as a function of feed grade (source: Keliber 2019 and 2022 FS Reports).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 103 Lithium Recovery as a Function of Concentrate Grade Figure 9-4: Lithium recovery as a function of concentrate grade (source: Keliber 2019 and 2022 FS Reports). 9.1.2.6 Laboratory Flotation Tests for Rapasaari in 2017 Exploration and resource drilling of Rapasaari during 2016 and 2017 led to the deposit becoming the biggest ore body of the Keliber Lithium Project. Mineral processing testing was, however, quite limited and, therefore, further testing of Rapasaari was started in July 2017. 9.1.2.6.1 Mineral Processing Testing and Results With the new sample and after optimisation, Rapasaari lithium recovery was reported to be close to that achieved for Syväjärvi. 9.1.2.7 Rapasaari Locked-Cycle Flotation Testwork 2018 (FS) 9.1.2.7.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] The programme was executed with the following Rapasaari sample materials: • Average ore around 100 kg; S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 104 • High-grade ore around 87 kg; and • Waste rock around 40 kg. The 2022 FS Report did not describe drill core sampling details. 9.1.2.7.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The Rapasaari flotation tests were carried out at the GTK Mintec facility in Outokumpu. GTK certification details are provided in Subsection 9.1.2.1. 9.1.2.7.3 Mineral Processing Testing and Results The programme included 16 batch flotation tests to optimise flotation conditions and a locked-cycle flotation test. Mineral liberation analysis was utilised to characterise the average ore, waste rock, and final flotation concentrate mineralogical properties. The results of batch flotation tests indicated that coarser grinding had a positive effect on flotation. Higher waste rock dilution decreased the final concentrate grades and recoveries. Lower collector dosage in the pre-flotation resulted in higher Li2O recovery in the spodumene flotation but slightly higher magnesium grade in the final concentrate. In the locked-cycle test, the required collector dosage was found to be about 20% of that needed in the open circuit. The locked-cycle grade-recovery points showed about 1% higher lithium recovery than the corresponding grade in open circuit. The final concentrate grade for the last five rounds (average) was 4.34% Li2O at 88.36% lithium recovery. 9.1.2.8 Emmes Laboratory-Scale Flotation Tests and Further Optimisation Tests 2018 9.1.2.8.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] The 2022 FS Report noted that, as Emmes ore had yet to be tested by Keliber, a representative sample was collected in 2018. Primary sampling details were not described. The Emmes ore sample had grades of 1.43% Li2O and the wall rock mica schist 0.265% Li2O. In chemical and modal composition, both the ore and wall rock were reportedly typical for spodumene pegmatite deposits of Central Ostrobothnia. Spodumene was the main lithium mineral but traces of cookeite and siclerite were also detected. 9.1.2.8.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The 2022 FS Report does not state where these tests were conducted, but presumably, they were conducted at the GTK Mintec facility. GTK certification details are provided in Subsection 9.1.2.1. 9.1.2.8.3 Mineral Processing Testing and Results The Emmes ore showed a similar flotation response to Syväjärvi. The lithium recovery at 4.5% concentrate grade was 91.8%, and 91.0% at 5.0% grade. Wall rock dilution resulted in an almost linear decrease in the final concentrate grade S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 105 e.g. the final concentrate was 5.8% for the sample with no dilution and 5.0% with 10% dilution. With a fixed concentrate grade, the dilution caused a decrease in recovery, but this was significantly lower for Emmes than Syväjärvi: e.g. when wall rock dilution was increased from zero to 10%, Syväjärvi recovery at 4.5% Li2O decreased from 92.2% to 85.8%, whereas only 0.6% loss to 91.2% was experienced for Emmes. 9.1.2.9 Flotation Tests for Rapasaari and Outovesi 2019 9.1.2.9.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] This programme was initiated in November 2018 and includes ore variability tests on different Rapasaari ore types, initial flotation tests on Outovesi, and a locked-cycle test on a Rapasaari drill core sample. The programme was conducted with the following Rapasaari and Outovesi sample materials: • Average ore 56 kg; • High-grade ore around 8 kg; • Waste rock around 35 kg; • Rapasaari North ore 38 kg; • Rapasaari West 86 kg; • Rapasaari South-West 86 kg; • Outovesi ore 64 kg; and • Outovesi white and black waste rock total 26 kg. Primary sampling details were not described in the 2022 FS Report. 9.1.2.9.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] This programme was initiated in November 2018 and completed in April 2019 at the GTK Mintec facility. GTK certification details are provided in Subsection 9.1.2.1. 9.1.2.9.3 Mineral Processing Testing and Results Modal mineralogy determined that the spodumene content in Rapasaari samples varied from 13.1% to 20.6%. Small contents of some other lithium-containing minerals were also found, including petalite, trilithionite, and triphylite. The primary gangue minerals were plagioclase (25.7–36%) and quartz (26.9–31%). Other gangue minerals were microcline, K-feldspar, and muscovite. The Bond rod mill work index value was found to be 15.3 kWh/t, and the ball mill work index was 15.2 kWh/t. In terms of the JKTech scale, the Rapasaari West sample would be classified as a hard material. Regarding the spatial variability, the best Li2O grades and recoveries were achieved from the Rapasaari North sample and the results from the Rapasaari West sample were quite similar. The recoveries were a bit lower from the Rapasaari Main sample. The poorest results were achieved from Rapasaari South-West as shown in Figure 9-5. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 106 The flotation behaviour of Outovesi mineralised samples was rather like Rapasaari main and Ra-all-2019 composite, as shown in Figure 9-6. Variability in Rapasaari Flotation Recovery Figure 9-5: Variability in Rapasaari flotation recovery (source: Keliber 2022 FS).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 107 Variability in Outovesi Flotation Recovery Figure 9-6: Variability in Outovesi flotation recovery (source: Keliber 2022 FS). Overall, the higher the waste rock dilution ratio the lower the Li2O grades and recoveries in the cleanings, regardless of the sample. Based on the results, it seems clear that the head grade will have an effect on Li2O recoveries. This was confirmed partially by spatial variability testing where the Rapasaari North was found to have the best flotation response and the highest head grade. The grades and the recoveries in the locked-cycle flotation test with the Ra-all-2019 composite were lower in comparison to the single-batch flotation test with the same material. Flotation without de-sliming produced good results, as did magnetic separation on the final spodumene concentrate. The normal slurry density of 30% in the conditioning of the pre-flotation stage seemed to work quite well. It was noted in the 2022 FS Report that such process changes should be considered for future studies and process design. 9.1.2.10 Optical Ore Sorting in 2018 9.1.2.10.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Sorting tests were conducted in November 2018 using Syväjärvi ROM ore (4 mm to 35 mm in size), spodumene-rich material, and black waste rock. Syväjärvi ore samples included spodumene-pegmatite ore (grey-green), and reddish marginal ore (red, light) including muscovite pegmatite and potassium feldspar. Syväjärvi dark side rock sample includes plagioclase-porphyrite and mica schist. Feed sample for the sorting tests included the Syväjärvi ore and marginal ore at the ratio 1:10 with 15% of side rock dilution. Primary sampling details were not described in the 2022 FS Report. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 108 9.1.2.10.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] The sample was crushed and screened at GTK Mintec before dispatch to Binder+Co sorting test facility in Gleisdorf, Austria. Atomic absorption spectroscopy (AAS) and XRF analyses were conducted at the Labtium laboratories of the subsamples for each ore type and side rock. The Certification Body of TÜV SÜD Management Service GmbH certified that Binder GmbH has established and applies a Quality Management System according to ISO 9001:2015. 9.1.2.10.3 Mineral Processing Testing and Results The focus of the tests was to remove black plagioclase porphyrite waste rock from the plant feed. Size classes 12/20 mm and 20/35 mm were washed at a rinsing feeder before the optical sorting. Additional sorting of smaller 4/12 mm size class was sorted without washing; instead, air knives and dust removal were utilised. Ore sorting was found to be effective in removing black waste rock from the artificial composite ore feed. The lithium grade of the reject in the tests was 0.2–0.3% Li2O. Lithium content of the black waste rock in contact with the ore varied between 0.08 and 0.47% Li2O with the average being in the range of 0.24–0.30% Li2O. It was reported that lithium in country rocks was not included in the Mineral Resources nor in the Mineral Reserves. Thus, the recovery of lithium carried by pegmatite was practically 100% in the testwork. 9.1.2.11 Optical Ore Sorting at Redwave in 2019 9.1.2.11.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Sorting tests were conducted in August 2019 using samples of Syväjärvi spodumene-rich material and black waste rock (12.4 mm to 20 mm in size). The focus of the tests was to remove black plagioclase porphyrite waste rock from the plant feed. 9.1.2.11.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] An ore-sorting test programme was completed in the Redwave sorting test facility in Eggersdorf, Austria. The Certification Body of TÜV SÜD Management Service GmbH certified that Redwave, a division of BT-Wolfgang Binder GmbH, has established and applies a Quality Management System according to SCC**:2011. 9.1.2.11.3 Mineral Processing Testing and Results The sample was crushed and screened at GTK Mintec before dispatch to Binder for optical sorting testing that is described in Subsection 9.1.2.1. After completing the testwork at Binder, the same sample was delivered to Redwave to complete the same testwork procedure to support the best sorting equipment selection. Redwave only uses a double-sided Red Green Blue camera as a sensor. The equipment was reported to be robust and suitable for the mining environment. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 109 9.1.2.12 Syväjärvi Pilot Testing in 2019 (FS) 9.1.2.12.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] This pilot campaign processed 89 t of the Syväjärvi ore at a waste rock dilution of 4%, which was the dilution as modelled in the life of mine (LOM) plan of that time. The feed material was the same as that used in the 2016 pilot campaign discussed in Subsection 9.1.2.1. 9.1.2.12.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] This programme was conducted at the GTK Mintec facility in August 2019. GTK certification details are provided in Subsection 9.1.2.1. 9.1.2.12.3 Mineral Processing Testing and Results Pilot tests conducted are shown in Figure 9-7. The minerals processing flowsheet included grinding, de-sliming, pre-float, spodumene flotation, and low-intensity magnetic separation. Overall spodumene recovery was increased by 4% from the previous Syväjärvi pilot to 88%. The recovery increase was achieved by reducing the slime production, optimisation of the pre-float, high-intensity conditioning, and increasing the residence times in spodumene flotation. Syväjärvi Pilot Tests 2019 Figure 9-7: Syväjärvi pilot tests 2019. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 110 9.1.2.13 Dewatering Studies on Syväjärvi Pilot Processing Samples by Outotec in 2019 (FS) 9.1.2.13.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Samples were extracted from the Syväjärvi pilot circuit. 9.1.2.13.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] Outotec representatives were present for several days at GTK Mintec during the Syväjärvi pilot processing in August 2019. Outotec performed dewatering tests of the spodumene concentrate at the Outotec dewatering technology centre in Lappeenranta, Finland. Thickening tests were performed at the Outotec Research Centre in Pori, Finland. Metso Outotec complies with the requirements of international standards for management systems. The majority of Metso Outotec’s major units are certified to ISO 9001 (quality), and the main operational units also have ISO 14001 (environment), ISO 45001 or OHSAS18001 (safety) standards as a framework. 9.1.2.13.3 Mineral Processing Testing and Results Filtration Tests of Spodumene Concentrate The main objective was to determine the moisture content of the cake and to confirm filter cloth selection and maximum filtration capacity for the vacuum belt and vertical pressure filters. A final moisture content of 9.6% was achieved with the vacuum belt filter and 7.3% was achieved with the vertical pressure filter. Both values are below the moisture limit of 10% for the final concentrate before the hot conversion at the Keliber lithium refinery. Thickening Tests Testwork of the pre-float, spodumene flotation feed, tailings, tailings without slime, slime, and spodumene concentrate showed that the materials can be thickened successfully. Keliber wanted to test the settling of flotation tailings with and without slime to inform decision-making on the tailing’s storage designs. Filtration Tests of Flotation Tailings Filtration of the flotation tailings was a continuation of the thickening tests. Keliber wanted to complete testing to support engineering for potential dry stacking of the flotation tailings. Tailings with slime could be dewatered successfully by an Outotec vacuum belt filter (20.6%), filter press filtration (12.1%), and fast-opening filter press filtration (13.3%). Tailings without slime were found to be difficult to filter with coarser particle size distribution (PSD). The following results were achieved: Outotec vacuum belt filter (18.9%) and fast-opening filter press filtration (13.9%). 9.1.2.14 Dewatering Study of the Spodumene Concentrate by Metso Minerals in 2019 9.1.2.14.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Samples were extracted from the Syväjärvi pilot circuit. One barrel containing a 50 kg sample of concentrate was delivered to the Metso Minerals laboratory in Sala.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 111 9.1.2.14.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] During the Syväjärvi pilot processing campaign in 2019, representatives from Metso Minerals visited to witness the pilot plant operation. Metso Minerals proposed vendor testwork to support spodumene concentrate filter selection and sizing data. Details of the Metso Outotec certification are provided in Subsection 9.1.2.13. 9.1.2.14.3 Mineral Processing Testing and Results After thickening and top feed vacuum filtration, the end moisture of the concentrate was measured to be in the range of 10–13%. 9.1.2.15 XRT Ore-Sorting Tests at Outotec (TOMRA) in 2019 9.1.2.15.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] Sorting testing at TOMRA was a continuation of testing described in earlier sections with the same ore and waste rock samples. 9.1.2.15.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] Sorting tests were conducted at Outotec (TOMRA). TOMRA is certified to the ISO 9001 and ISO 14001 quality system standards. 9.1.2.15.3 Mineral Processing Testing and Results The purpose of this testwork was to determine the suitability of a TOMRA® sorting system for the Syväjärvi operation. The tested samples were represented in two size fractions: +12.4–20 mm and +20–35 mm. Before the test, the samples were mixed in a ratio of 79.1% product, 7.9% marginal ore, and 13% waste. Results showed a high lithium recovery of approximately 95% for both size fractions, with mass rejection of 16–19%. The results showed positive amenability of TOMRA XRT ore-sorting technology with Syväjärvi material. Further testing and engineering were, however, recommended for the final flowsheet development of the crushing and sorting circuit. 9.1.2.16 Rapasaari Laboratory-Scale Programme to Control Sulphides at GTK in 2021 9.1.2.16.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] In total, 80 kg of pegmatite ore sample from the Rapasaari deposit was collected from rejects of analytical samples of half-cut drill cores. The 50 kg feed sample for the bench-scale beneficiation tests comprised 47.5 kg Rapasaari ore and 2.5 kg waste rock. After homogenisation, the feed material was divided into suitable 1 kg and 5 kg subsamples for the testwork. In February 2021, approximately 30 kg of additional Rapasaari ore and 3 kg of Rapasaari waste rock were packed and sent to SGS Canada for parallel testwork purposes (Subsection 9.1.2.17). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 112 9.1.2.16.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] This programme was conducted at the GTK Mintec facility. GTK certification details are provided in Subsection 9.1.2.1. 9.1.2.16.3 Mineral Processing Testing and Results Keliber contracted GTK Mintec to study processing alternatives to manage arsenic levels of the final concentrate and develop the process, at bench-scale, for overall arsenic management. The programme included sample preparation, flotation tests, magnetic separation, and gravity separation tests. The arsenopyrite occurred mainly in the waste rock bulk sample, where its content was 0.09%. In addition, the arsenopyrite particles were totally liberated in the ground Rapasaari composite sample. The average Li2O grade in the Rapasaari composite sample was 1.23% and arsenic was 0.021%, calculated from the bench-scale tests. The grain sizes (P80) of spodumene and arsenopyrite were 90 μm and 24 μm in the ground (125 μm) feed samples, respectively. In total, more than 20 bench-scale flotation tests were performed, with combinations of different unit processes to remove arsenopyrite. High gradient magnetic separation was shown to not be an effective method of removing arsenopyrite. Gravity separation by shaking table worked well; however, several cleanings were required in order to avoid spodumene losses. About 50–70% of arsenic could be removed by pre-flotation. Sulphide flotation without pre- flotation and NaOH conditioning had a negative effect regarding selectivity in spodumene flotation. The programme proved that most arsenopyrite could be removed by sulphide flotation. The process was shown to be very sensitive regarding spodumene flotation and it was reported that fresh water should be used in all stages. It was noted that it is essential to remove more than 95% of arsenic prior to spodumene flotation because the arsenopyrite tends to enrich during the spodumene flotation. 9.1.2.17 Rapasaari Laboratory-Scale Programme to Control Sulphides at SGS in 2021 9.1.2.17.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] SGS Minerals were provided with the same sample material as used in the programmes at GTK Mintec (GTK certification details are provided in Subsection 9.1.2.1). 9.1.2.17.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] Keliber sought a second testwork programme and new ideas for the Rapasaari ore flowsheet development in early 2021, especially for arsenic and sulphur management. This programme was conducted at SGS Minerals. SGS Minerals is accredited to the requirements of ISO/IEC 17025 for specific tests listed on their scope of accreditation, including geochemical, mineralogical, and trade mineral tests. 9.1.2.17.3 Mineral Processing Testing and Results The main objective of the metallurgical testwork was to develop an appropriate flowsheet for producing a high-grade spodumene concentrate, with a reasonable recovery, from a composite sample from the Rapasaari deposit. Rejecting S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 113 arsenic and sulphur content was also a focus. Testwork on the composite sample included head characterisation, mineralogical examination, heavy liquid separation, magnetic separation, and flotation. The lithium grade in the composite sample was 1.18% Li2O. The iron content after 5% waste rock dilution was low, at 0.77% Fe2O3. The sample was radioactive-free, with <0.01% Ta2O5. The grades of arsenic and sulphur in the sample were 0.03% and 0.04%, respectively. Heavy liquid separation (HLS) was completed on the Rapasaari ore sample to evaluate the potential of gravity separation. It was concluded that the Rapasaari ore is not amenable to DMS. Wet high-intensity magnetic separation (WHIMS) tests effectively rejected iron-bearing gangues (possibly pyrite, pyrrhotite, and arsenopyrite, iron silicate minerals) from the 100% passing 48 mesh (300 μm) feed. WHIMS at 5,000 Gauss rejected ~20% Fe2O3, 8% As, and 29% S, with only 2.4% of the lithium deporting to the magnetic products. Further optimisation on the magnetic separation tests was recommended to reject more iron-bearing gangues with relatively less lithium loss. Seven flotation tests were carried out after grinding to 100% pass 48 mesh (300 μm). Spodumene upgrading was achieved in a flowsheet that comprised one rougher stage and three cleaner flotation stages. The lithium concentrates produced achieved grades of ~5.4–5.8% Li2O at ~84–92% lithium recoveries. The Fe2O3 and arsenic assays of the flotation concentrate were <1% and <0.005%, respectively, after passing the spodumene concentrate through WHIMS. A final stage of WHIMS on the spodumene concentrate played an important role in lowering the Fe2O3 grade to less than 1% and also decreased the arsenic content. Optimisation was recommended to further reduce lithium loss to magnetic products. 9.1.2.18 Ore-Sorting Testwork at TOMRA for Laser and XRT Sensor Trade-Off 9.1.2.18.1 Representivity of Test Samples [§229.601(b)(96)(iii)(B)(10)(ii)] The sample for this programme was drawn from the 2016 Syväjärvi test mining sample. The sample was loaded and transported to GTK Mintec for sample preparation. For the purpose of this performance test, ore (2,400 kg for size fraction 30–60 mm and 1,200 kg for the finer-grained fraction 15–30 mm) and waste material (250 kg pro fraction) were dispatched to TOMRA. 9.1.2.18.2 Testing Laboratory and Certification [§229.601(b)(96)(iii)(B)(10)(iii)] Testing was conducted at the TOMRA facilities in Hamburg, Germany. TOMRA certification details are provided in Subsection 9.1.2.3. 9.1.2.18.3 Mineral Processing Testing and Results Material in two particle sizes 30–60 mm and 15–30 mm was tested with the PRO Secondary LASER and COM Tertiary XRT by using two settings per technology. With the first, less sensitive setting, black waste represented by porphyrite, metasediments and tuffs was rejected, whereas with the second, more sensitive one, in addition to the black waste, potassium-feldspar was separated. The material sample for the laser sorting was washed since this technique requires clean and wet surfaces. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 114 Two settings were used: • Setting 1 was less sensitive, where dark waste particles, including plagioclase porphyrite, metasediments and tuffs, were rejected. In addition to these rock types, some feldspar without the spodumene inclusions were separated. The application of the more sensitive • Setting 2 allowed the separation of feldspar and pegmatite in addition to the dark waste particles. Based on the mass balances and chemical assays, both of the ore-sorting techniques are suitable for the Keliber Lithium Project operations. Table 9-3 summarises the overall results achieved.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 115 Table 9-3: TOMRA ore-sorting balance 2021. Sensor Set Size (mm) Feed (kg) Feed (Li2O) Product (Li2O) Waste (white) (Li2O) Waste (black) (Li2O) Total Waste (Li2O) Recovery (%) Upgrade Waste Removal (%) Laser 1 30–60 687 1.617 1.939 1.515 0.103 0.223 97 1.20 19 Laser 2 30–60 667 1.432 1.811 0.660 0.097 0.202 97 1.26 24 Laser 1 15–30 362 1.331 1.636 1.590 0.103 0.348 94 1.23 24 Laser 2 15–30 365 1.192 1.640 1.113 0.108 0.438 86 1.38 37 XRT 1 30–60 645 1.480 1.698 1.841 0.138 0.265 97 1.15 15 XRT 2 30–60 642 1.359 1.636 0.750 0.093 0.237 97 1.20 20 XRT 1 15–30 369 1.348 1.742 0.824 0.097 0.196 96 1.29 25 XRT 2 15–30 366 1.234 1.704 0.672 0.097 0.265 93 1.38 33 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 116 Using the same capacities, comparable results were achieved for size fractions 30–60 mm and 15–30 mm. Besides the upgrades of the target element in the product and reduction in the waste fraction, high recoveries were achieved. So, for example, with the more sensitive setting 2, by rejecting 24% of waste with LASER and 20% with XRT, a recovery of 97% has been achieved with both technologies. Results showed high lithium recovery in the range of 86% to 97% for both size fractions, with mass rejection ranging from 15% to 37%. 9.1.2.18.4 Pilot-Scale Testwork on 50 t Rapasaari Deposit Reference is made to a 50 t Rapasaari sample that was processed through a pilot flotation plant in the SRK Due Diligence (2023). The plant was fed at a feed rate of 420 kg/h at a head grade of 1.21% Li2O. The circuit incorporated a sulphide flotation stage to assist with As removal. Final products of up to 5.19% Li2O was produced at a recovery of 75.4%. As none of the previous testwork programmes with Rapasaari ore delivered an acceptable As level in the concentrate, SRK recommended a 3% absolute discount on the Rapasaari recoveries be applied as sensitivity. 9.1.3 Recent Conversion Testwork 9.1.3.1 Laboratory-Scale Conversion Tests Conversion tests carried out prior to 2017 were of small scale. In the Länttä ore pilot test of 2016 (Subsection 9.1.2.1), thermal conversion tests were conducted at 1,000°C, and a retention time of one hour was found to be sufficient to convert alpha-spodumene to leachable beta-spodumene. The Syväjärvi concentrate produced in laboratory-scale tests in 2016 (Subsection 9.1.2.2) was also treated in a furnace prior to autoclave testing. The Syväjärvi concentrate was found to behave in a similar way to the Länttä concentrate. 9.1.3.2 Conversion Pilot for Syväjärvi Concentrate by Metso Minerals in 2017 The spodumene concentrate derived from the Syväjärvi sample processed in the 2016–2017 mineral processing pilot plant (Subsection 9.1.2.3) was tested in three stages: • Firstly, a small amount of concentrate was converted in an indirectly heated laboratory-scale rotary kiln at the Outotec Research Laboratory. A temperature of 1,010°C for 30 minutes was sufficient to convert alpha- spodumene to leachable beta-spodumene; • The second test was carried out in the Development Center, Danville, PA, USA. The sample was prepared by combining two concentrate samples produced in a mineral processing pilot test at GTK (GTK certification details are provided in Subsection 9.1.2.1) – 150 kg from the Outotec Frankfurt research centre and about 400 kg sent from GTK Mintec. For the conversion tests, the as-received subsamples were mixed in weight ratios of 31.36% and 68.64%. Lithium grades of the samples were 2.35% (5.06% Li2O) and 2.34% (5.04% Li2O), respectively. The main target of the test programme was to practise the conversion process and collect operational and material characteristics for the design of commercial conversion equipment for the production of material to achieve greater than 95% alpha- to beta-spodumene conversion. A secondary objective was to produce product for subsequent lithium pressure leaching tests; and • A directly heated rotational drum furnace fired with propane gas was employed to complete eight conversion tests, with temperatures ranging from 1,000°C to 1,075°C. It was concluded that the targeted 95% conversion rate was achieved. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 117 9.1.3.3 Conversion Tests with Syväjärvi and Rapasaari Concentrate 2018 Regarding hydrometallurgical testing for Syväjärvi and Rapasaari concentrates, laboratory conversion tests were also performed by Outotec. The conversion was executed in a batch chamber furnace at temperatures of 990°C, 1,010°C, 1,030°C, and 1,060°C using three hours’ retention time. Conversion of the samples was confirmed by XRD. Scanning electron microscopy (SEM) on the 1,060°C Rapasaari sample showed some melted structures on the spodumene grains, which resulted in lower lithium recoveries in leaching. 9.1.3.4 Conversion Pilot for Syväjärvi Concentrate by FLSmidth in 2018 (FS) The third test series was carried out with a directly fired rotary kiln at FLSmidth Inc. Pyromet Technology testing facilities, Bethlehem, PA, USA. The sample used was concentrate produced in a mineral processing pilot test at GTK (GTK certification details are provided in Subsection 9.1.2.1). The test programme had two targets: (i) to evaluate the physical, thermal, and phase conversion properties of a spodumene concentrate sample, and (ii) to produce bulk converted sample for subsequent pilot-scale lithium pressure leaching tests. According to FLSmidth’s assays, the lithium grade of the bulk sample was 5.57% Li2O analysed with AAS and spodumene content of about 74% to 75% analysed with XRD. The concentrate was held for 30 minutes at 1,100°C. The conversion recovery result was 97.5%. The SG of the material decreased from a feed level value of 3.04 g/cm3 to 2.36 g/cm3, mainly because of the spodumene phase conversion. 9.1.3.5 Conversion Pilot for Syväjärvi Concentrate by FLSmidth in 2019 A pilot test programme was performed to evaluate the conversion of alpha-spodumene to beta-spodumene using a two-stage cyclone preheater rotary calciner system, followed by product comminution using an open circuit ball mill. The material received for this study included ~3,000 kg of flotation concentrate containing 10.6% moisture and 4.75% Li2O. The solids residence time in the rotary kiln was two hours and the burning zone solids temperature was generally maintained between 1,050–1,100°C. These conditions resulted in an overall average alpha- to beta-spodumene conversion level of 96.9%, as measured by the sulphuric acid solubility method. Stable, sinter-free operation of the preheater kiln system was demonstrated along with a high conversion to beta- spodumene when calcining flotation concentrate. The dusting rate was considered very low. Based on the results of the pilot programme, no adjustments are required to the commercial calcining being offered by FLSmidth. 9.1.4 Recent Hydrometallurgical Testing for Production of Lithium Carbonate and Lithium Hydroxide The June 2018 FS considered the production of battery-grade lithium carbonate. However, following further market studies it was decided to consider the production of BG instead of lithium carbonate. As a consequence, much of the hydrometallurgical testwork undertaken was directed at producing lithium carbonate. While not totally applicable, such test programmes are briefly summarised here. 9.1.4.1 Laboratory and Pilot Test for Länttä Concentrate in 2015 The feed material for the testing was from the previous GTK Mintec Länttä 2015 programme (GTK certification details are provided in Subsection 9.1.2.1). Suitable composite samples were prepared so that the feed sample had an average head grade of 4.5% Li2O. Lithium yields in laboratory batch leaching and bi-carbonation tests were low, with 86% being the best result achieved. A higher lithium yield of 91% was, however, obtained in the pilot-plant leaching and bi-carbonation tests. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 118 Ion exchange was used to remove metal impurities such as Ca and Mg from the leach solution. The purified solution from the ion exchange was heated above 90°C to crystallise Li2CO3. The Li2CO3 product contained 17.3% to 18.6% lithium, with phosphorus and silicon being the main impurities. The Bond ball mill work index for the beta-spodumene was determined as 11.51 kWh/t. 9.1.4.2 Laboratory Tests for Syväjärvi Concentrates 2016 The main objective of the programme was to confirm the leaching parameters for the Syväjärvi spodumene flotation concentrate produced in the previous GTK Mintec Syväjärvi 2015 batch flotation. Based on the solid fraction analyses, the lithium leaching yield was 95.6%. 9.1.4.3 Laboratory and Pilot Tests for Syväjärvi Concentrates 2017 Feed material was concentrated that had been converted for subsequent hydrometallurgical tests at Outotec Frankfurt Research Centre (Subsection 9.1.2.3). The programme included the following items: • Thermal conversion tests of alpha-spodumene to leachable beta-spodumene in a laboratory rotary kiln; • Pressure leaching and bi-carbonation tests; • Solid-liquid separation tests of the leach residue (Analcime); • Ion exchange tests; • Crystallisation tests of Li2CO3; and • Solid-liquid separation tests of the lithium carbonate product (analcime). The converted beta-spodumene material was used in leaching and bi-carbonation tests which yielded from 86% to 95% lithium in the batch tests and 84% to 87% in the pilot plant operation. Ion exchange was used to remove metal impurities such as Ca and Mg from the leach solution. The purified solution from the ion exchange was heated above 95°C to crystallise Li2CO3. The Li2CO3 product contained 17.3% to 19.0% lithium, with phosphorus and silicon being the main impurities. In thickening tests, the leach residue slurry settled to an underflow density of 48% and the overflow clarity was between 70 ppm and 250 ppm. The required flocculant dosage was 20 g/t of Superfloc N100. In filtration tests, cake moistures of 30% and 44% were achieved by pressure filtration and vacuum filtration, respectively. No difference was observed in the lithium recoveries between pressure or vacuum filtration for the un- thickened leach residue slurry. With the thickened leach residue slurry, however, the pressure filtration was more efficient and filtration capacities were higher. 9.1.4.4 Laboratory Tests for Syväjärvi Concentrates 2017 The test programme included soda-leaching tests in a batch-scale autoclave of the concentrate that had been converted during the Metso Minerals pilot-scale conversion tests described above (Subsection 9.1.3.2). This programme was conducted at Outotec’s facilities. Alpha-spodumene was not detected in the XRD analysis indicating that the conversion to beta-spodumene was complete. Based on the solid fraction analyses over five batch tests, the lithium yield varied from 79% to 89%.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 119 9.1.4.5 Laboratory Tests for Syväjärvi and Rapasaari Concentrates 2018 The feed material was produced in the previous Syväjärvi 2017 test programme (Subsection 9.1.3). Recent conversion testwork consisted of conversion tests with Syväjärvi and Rapasaari concentrate 2018. The programme comprised laboratory testing of conversion, soda leaching, bi-carbonation, ion exchange, and crystallisation. Lithium carbonate was produced by crystallisation from Syväjärvi and Rapasaari concentrates. Leaching and bi-carbonation tests were conducted on the converted concentrate in a laboratory autoclave. The autoclave temperature was 220°C with carbon dioxide introduced at a pressure of 3 bar and a temperature of 30°C. The following lithium yields were obtained in the programme: • 90–95% for the Syväjärvi_2018 concentrate; • 91–96% for the Syväjärvi_2017 concentrate; and • 88% for the Rapasaari concentrate at 1,010°C and 84% at 1,060°C. SEM study from the calcined Rapasaari concentrate and leach residues revealed that some spodumene grains were covered by the melted phase, which decreased the lithium leaching yield compared to the Syväjärvi concentrates. Lithium carbonate was produced by crystallisation with and without the ion exchange step. Results confirmed that it is possible to produce over 99.5% lithium carbonate end product without the ion exchange step from the Syväjärvi samples. The ion exchange, however, decreased the calcium level from 0.02–0.05% to less than 0.01%. 9.1.4.6 Lithium Hydroxide Pilot Processing at Outotec 2019 – Syväjärvi Metso Outotec's patented Lithium Hydroxide process for the production of BG includes three key unit processes: • Alkaline pressure leaching; • Lime conversion leaching; and • Lithium hydroxide monohydrate crystallisation. Feed to the two-stage alkaline leach process is beta-spodumene concentrate after calcining. Lithium is first extracted using soda ash pressure leaching, resulting in the formation of soluble lithium carbonate (Li2CO3) and mineral component analcime (NaAlSi2O6·H2O) as the main components. In the second stage, lithium carbonate is solubilised in a conversion reaction, producing lithium hydroxide solution and solid calcium carbonate, which will report together with other mineral residues. The alkaline hydroxide and carbonate processing environment ensures very low solubilities of the main impurity elements and compounds, including Fe, Al, Mg, Ca, B, and P, reducing the need for additional impurity removal or precipitation. The Pregnant Leach Solution containing lithium hydroxide is a suitable feed for polishing with Ion Exchange ahead of crystallisation of the final LiOH monohydrate product. The object of the 2019 testwork programme was to study lithium hydroxide production by soda pressure leaching and to produce small amounts of the product for marketing purposes. A beta-spodumene concentrate sample from conversion in a rotary kiln, called 2018 calcine, was the main concentrate used in this work. In addition, comparative hydrometallurgical testwork was carried out with a concentrate from 2017, which was calcined in a chamber furnace in Oberursel, Germany, by Outotec. Based on the chemical analysis of the calcine samples, they had similar compositions. The lithium concentration of the 2018 calcine was 2.55% Li (5.49% S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 120 Li2O) and that of the 2017 calcine was 2.39% Li (5.15% Li2O). Batch tests were carried out for the soda leaching and LiOH conversion process steps to produce information for the pilot operation. The solids analysis showed that 88% lithium extraction was achieved in the LiOH conversion. 9.1.4.7 Lithium Hydroxide Continuous Pilot Processing at Outotec 2020 – Syväjärvi The Syväjärvi beta-spodumene concentrate used in this hydrometallurgical testwork was calcined in the FLSmidth pilot run in 2019. The average lithium concentration of the calcines corresponded to a 4.53% Li2O concentration. Soda leaching, cold conversion, and secondary conversion batch tests were carried out to verify the lithium extraction as well as to produce information for the planning of the continuous pilot. The continuous LiOH pilot was operated for 14 days. The main process stages in the process were soda leaching, cold conversion, secondary conversion, ion exchange, LiOH crystallisation, and mother liquor carbonation. Leaching was carried out in a 65 L titanium autoclave at a target temperature of 220°C and a target residence time of two hours (refer to Figure 9-8). Slurry was flashed off from the autoclave to a flash vessel with 80°C temperature and atmospheric pressure. Slurry from the autoclave was filtered by pressure filter, with solids being washed with water. Sixty-Five (65)-Litre Autoclave used in the Semi-Continuous Pilot Processing Figure 9-8: Sixty-five (65)-litre autoclave used in the semi-continuous pilot processing. Pulped soda leach residue and lime slurry were pumped to the first 20 L cold conversion reactor, from where the slurry was transferred as overflow to the second 20 L reactor and, subsequently, to the filter feed tank. The target temperature in cold conversion was 30°C and residence time was about two hours. Both reactors as well as the filter feed tank were equipped with nitrogen gas feed. Slurry from the cold conversion was filtered by pressure filter, then filtrate was pumped to the secondary conversion feed tank. Solids were washed with water. The wash filtrate was then used in pulping of the soda leach residue and lime slurry preparation. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 121 The filtrate from the second pressure filter and lime slurry were pumped to a 20 L stainless steel reactor for secondary conversion. The residence time in secondary conversion was over two hours and the temperature was ambient. The overflow from the reactor was collected in the feed tank of a filter press polishing filter. Ion exchange was operated continuously during the pilot with a feeding rate based on the availability of feed solution, with two columns in series. Crystallisation was carried out at approximately 77°C. Lithium hydroxide slurry was thickened to a solids concentration of 40–50% and fed to a pusher centrifuge. BG (Na <50 ppm, K <20 ppm) was produced with a higher wash ratio of 0.17 m3/t. Lithium hydroxide monohydrate was dried in a shaking fluid bed dryer. Nitrogen gas was used for drying. It was reported that the objectives of the pilot were mostly met. The consumption of reagents was investigated and the impact of recycling different process streams on the process was observed. The soda leach circuit was stable, with stable concentrations of Na and K. The cold conversion circuit also proved to be stable. The impurities in the cold conversion solutions stabilised at a low level, with Na having a slight increasing trend. The Li extraction in the soda leaching and cold conversion process stages was initially low, but after adjustments were made to the operating conditions, high levels of extraction were achieved. The operation of the crystalliser had challenges related to the differences in the production capacity of the equipment in comparison to the rest of the process. However, the production of approximately 35 kg of moist lithium hydroxide monohydrate product from the centrifuge was achieved. One batch of the centrifuge product was successfully dried with a fluidised bed dryer as well. The product purity achieved with a single crystallisation stage was extremely high. With a second crystallisation, the impurity levels in the crystals could be decreased even further, with Na and K both being <10 ppm. The Si concentration in the crystals was also decreased by the second crystallisation stage. 9.1.4.8 Lithium Hydroxide Continuous Pilot Processing at Outotec 2022 – Rapasaari 9.1.4.8.1 Feed Sample The Rapasaari spodumene concentrate produced in the GTK Mintec pilot plant was calcined in a continuous rotary kiln by FLSmidth in North America, after which it was shipped to Pori, Finland, for hydrometallurgical testwork. 9.1.4.8.2 Test Programme This programme including batch leaching testwork as well as continuous piloting of the Metso Outotec LiOH Process, was carried out between April and June 2022. A simplified block diagram of the process flowsheet is shown in Figure 9-9. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 122 Simplified Process Flowsheet of LiOH*H2O Production Figure 9-9: Simplified process flowsheet of LiOH*H2O production. β-spodumene calcine was initially pulped with water and recycled process solutions. Sodium carbonate was simultaneously fed and dissolved according to the following reaction (Equation 3): Equation 3: Dissolution of sodium carbonate. 𝑁𝑎2𝐶𝑂3(𝑠) → 2 𝑁𝑎 + + 𝐶𝑂3 2− Operating conditions in the pressure-leaching autoclave are typically 200°C to 220°C and approximately 20 bar. β- spodumene reacts to form lithium carbonate and analcime solids according to the following reaction (Equation 4): Equation 4: Formation of lithium carbonate and analcime solids from β-spodumene. 𝐿𝑖𝐴𝑙𝑆𝑖2𝑂6 (𝑠) + 𝑁𝑎2𝐶𝑂3 + 𝐻2𝑂 → 𝐿𝑖2𝐶𝑂3 (𝑎𝑞, 𝑠) + 2 𝑁𝑎𝐴𝐿𝑆𝑖2𝑂6 ∙ 𝐻2𝑂 (𝑠) Some of the lithium remains solubilised, but it is mostly present as a solid lithium carbonate in the leach residue along with analcime. Autoclave discharge is cooled ahead of solid/ liquid separation, with solids content being forwarded to LiOH conversion.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 123 Pressure leach residue is pulped with water and lime. Calcium hydroxide reacts with the lithium carbonate to form more soluble lithium hydroxide and calcium carbonate, which is precipitated. This reaction is carried out at a slightly elevated temperature to limit the solubility of impurities such as aluminium and silicon. The conversion reaction takes place according to the following Equation 5: Equation 5: Conversion of calcium hydroxide plus lithium carbonate to lithium hydroxide and calcium carbonate. 𝐿𝑖2𝐶𝑂3 (𝑠) + 𝐶𝑎(𝑂𝐻)2 (𝑎𝑞) → 2𝐿𝑖𝑂𝐻 (𝑎𝑞) + 𝐶𝑎𝐶𝑂3 (𝑠) After the LiOH conversion, solids and liquids are separated and the solid residue, mainly calcium carbonate and analcime, is the main residue of the process. The filtrate is fed to a secondary conversion stage, where impurities such as Al and Si are removed from the solution by the addition of a small amount of lime. After the secondary conversion, the solution is fed via polishing filtration to ion exchange for the removal of residual divalent metal cations such as calcium and magnesium, before crystallisation. The purified lithium hydroxide solution is fed to the crystallisation stage, where lithium hydroxide monohydrate is crystallised under vacuum from the purified LiOH solution. Some of the crystallisation mother liquor is fed to carbonation. Carbon dioxide gas is fed to the solution and lithium hydroxide is converted to lithium carbonate, which precipitates due to its lower solubility (Equation 6). Equation 6: Conversion of lithium hydroxide to lithium carbonate. 𝐿𝑖𝑂𝐻 (𝑎𝑞) + 𝐶𝑂2 (𝑔) → 𝐿𝑖2𝐶𝑂3 (𝑎𝑞, 𝑠) + 𝐻2𝑂 The product slurry from mother liquor carbonation can be fed to the slurry preparation stage of the soda-leaching circuit. 9.1.4.8.3 Test Results The average lithium concentration of the calcines corresponded to a 5.5% Li2O concentration. Soda leaching, cold conversion, and secondary conversion batch tests were carried out to verify the lithium extraction as well as to produce information for the planning of the continuous pilot. The continuous LiOH pilot was operated for approximately 17 days. The main process stages in the process were soda leaching, cold conversion, secondary conversion, ion exchange, LiOH crystallisation, and mother liquor carbonation as previously described for Syväjärvi. The first stage crystallisation was continuously operated. The average levels of the typical impurities were ~30 ppm Al, 311 ppm Na, 118 ppm Si, and 39 ppm K. During the pilot, three samples of the 1st stage LiOH·H2O products were redissolved and fed to 2nd crystallisation, which was carried out with a rotary evaporator. It was reported that the results of the chemical analyses of the samples were excellent and, in terms of impurity concentrations, the final products corresponded to the specification of battery grade LiOH·H2O provided by Keliber. Almost all impurities were below detection limits, such as Al <5 ppm, K <10 ppm, Cl <20 ppm, F <50 ppm, SO4 <150 ppm, and most of the heavy metals being either <1 ppm or <2 ppm. Some Na and Si S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 124 were detected in two of the samples, but the concentrations were according to the client’s specification, with Na being mostly <30 ppm and Si mostly <20 ppm. Based on LiOH concentrations of both the 1st and 2nd stage samples, there was some residual moisture in the products. 9.1.4.9 Spodumene Alpha to Beta Phase Conversion Tests at FLSmidt January 2024 – Li2O Concentrates 9.1.4.9.1 Feed Sample The material received for this study included three different materials: • ~4,000 kg of Sigma spodumene concentrate containing 0.55 wt% moisture and 5.74 wt% Li2O; • ~5,000 kg of AMG spodumene concentrate containing 2.54 wt% moisture and 5.80 wt% Li2O; and • ~7,000 kg of Sayona spodumene concentrate containing 2.91 wt% moisture and 5.94 wt% Li2O. 9.1.4.9.2 Test Programme The test programme included the conversion of alpha spodumene to beta spodumene using a 2-stage cyclone preheater rotary calciner system. The solids residence time in the rotary kiln was 1.5 hours and the burning zone solids temperature was generally maintained between 1,050–1,100°C. The product of the rotary kilns was milled using an open circuit ball mill. The AMG and Sayona materials were processed as received, but the as-received Sigma feed was too coarse. As a result, the material was sized using a small ball mill. 9.1.4.9.3 Results These conditions resulted in an overall average alpha-to-beta conversion level of 95.5% for Sigma, 96.2% for AMG, and 96.1% for Sayona as measured by the sulfuric acid solubility method. Peak conversions achieved were 95.9% for Sigma, 97.5% for AMG, and 98.4% for Sayona. The spectrometer analyses generally showed complete conversion of alpha spodumene to beta spodumene. The concentrate acted as free-flowing solids on approach to the kiln burning zone and no measurable coating formation or particle sintering was observed. Concentrate flow through the preheater cyclones was very stable for all three materials throughout the duration of the pilot kiln system operation. The bag filter dust capture rate averaged only 6.0% of the spodumene concentrate fed to the system across all products on a dry basis, and the bulk of this dust was returned to the kiln system with a separate bucket elevator. The Sayona material had the highest dust rate at 9.2% while AMG and Sigma were 2.5% and 1.4%, respectively. A total of 600 kg of calcined spodumene from each of the three concentrates was ground in an open circuit ball mill. The purpose of this grinding step was to reduce the product particle size distribution to a level comparable to that of the starting concentrate feed so the products could be shipped off to Keliber for further testing. In all three cases, achieving the same P80 as the feed was not possible (it was finer) even by slowing the mill down and increasing the feed. Emission data were collected during the operation of the preheater kiln system using a combination of conventional gas analysers and a Fourier transform infrared (FTIR) analyser. Stable, sinter-free operation of the preheater kiln system was demonstrated along with a high conversion to beta spodumene when calcining this material. The results obtained from processing all three concentrates are very similar S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 125 to what was obtained when processing the Syväjärvi concentrate and Rapasaari concentrate during previous FLS test campaigns. 9.1.4.10 Lithium Hydroxide Continuous Pilot Processing at Outotec 2024 – Li2O Concentrates 9.1.4.10.1 Feed Sample Three different commercially available calcined spodumene concentrates were received at Metso Research Center for hydrometallurgical testwork (refer to Section 9.1.4.9). The lithium concentrations of the different calcines were as follows: Sayona 2.50%, AMG 2.51%, and Sigma 2.54 %, which corresponded to an average Li2O concentration of 5.4% for Sayona and AMG and 5.5 % for Sigma. 9.1.4.10.2 Test Programme This programme, which included batch leaching testwork as well as continuous piloting of the Metso Outotec LiOH Process, was carried out in 2024. 9.1.4.10.3 Test Results In the batch leaching tests, the Li extraction from the calcined spodumene concentrates was verified. The total Li recoveries in the tests were good, ranging from ~84% to ~91%. The calculations for the total Li recovery included the calcination, soda leaching, and cold conversion as well as the washing of the analcime sand. According to the batch test results, a longer residence time of 2.5 h in soda leaching had a positive effect on Li extraction compared to the shorter residence time of 1.5 h. In cold conversion batch tests conducted at Lhoist, Nordkalk, and SMA Mineral, lime samples were compared, and the lime used didn’t seem to affect the Li extraction significantly and both lime chemicals reacted efficiently in cold conversion. The continuous LiOH pilot was operated successfully. The main process stages in the pilot were soda leaching, cold conversion, secondary conversion, ion exchange, and LiOH crystallisation. Four continuous soda leaching runs were carried out over the course of three weeks, while the rest of the process was operated continuously for approximately 400 h. All main process recycles were included in the pilot operation, including the recycling of soda leach filtrates and wash filtrates back to soda leaching, cold conversion wash filtrates to lime slaking and cold conversion feed slurry preparation, crystallisation mother liquor to soda leaching via carbonation and as received from crystallisation, and mother liquor to secondary conversion as well as Li3PO4 to soda leaching. No significant accumulation of impurities was seen in the different process streams during the pilot. All main process recycle streams form part of the pilot tests and no significant accumulation of impurities was seen in the different process streams during the pilot. The solid analyses from soda leaching showed that the β-spodumene reaction with soda ash to lithium carbonate was efficient, with Li yields ranging between 97% and 100% in the leaching stage. The soda leach filtration carried out on a pressure filter, produced dry soda leach cakes which were fed to cold conversion. Cold conversion was operated smoothly, and the cold conversion filter produced cakes with good washing results. The filtration and washing processes were further investigated in the dewatering campaign. The average total Li recovery was 90.6% with AMG calcine feed, 88.6 % with AMG calcine feed with Li3PO4 circulation, and 90.2 % with Sayona calcine feed. The solution polishing stages successfully decreased the levels of impurities. In the secondary conversion, the average concentrations of Al and Si were decreased from ~50 mg/L to ~8 mg/L and from ~70 to ~16 mg/L, respectively. The ion exchange polished the solution from calcium ions, with the average concentration decreasing from ~30 mg/L in the IX feed to below detection limit in the IX product solution. The ion exchange resins were successfully regenerated during the pilot and, based on the analyses of the IX eluates, the regeneration cycle interval of roughly 100 BV solution S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 126 fed, was suitable. Polished LiOH solution was fed to the preconcentration stage, where water was evaporated from the solution before feeding to the crystalliser. The operation of the pre-concentrator was very consistent, and during the pilot approximately 74% of the solution volume fed to the pre-concentrator was evaporated. The average Li concentration in the pre-concentrated LiOH solution was ~28.5 g/L after startup, and the solution was fed to the LiOH·H2O crystallisation. The crystallisation was continuously operated and, in total, approximately 15.5 kg of crude LiOH·H2O crystals were produced. The results of the chemical analyses of the crystal samples were excellent for crude crystals for both feed calcines and met the final product quality specification with a few exceptions. Na and K concentrations in crude crystals were slightly higher for the Sayona calcine than with AMG calcine. The average impurity levels were Al 11 ppm (AMG) and 13 ppm (Sayona), Na ~330 ppm (AMG) and 490 ppm (Sayona), Si 58 ppm (AMG) and 62 ppm (Sayona), and K 40 ppm (AMG) and 59 ppm (Sayona). There were also trace amounts of Fe in the crude crystals – on average ~4 ppm for both feed materials. Zn was also detected in some crystal samples obtained from Sayona calcine. The concentrations of chlorides, fluorides, and sulphates in the samples were mostly below the detection limit at Cl <20 ppm, F <20 ppm and SO4 <150 ppm. Crystallisation mother liquor was mostly recycled to the crystallisation feed; the rest was bled and recycled upstream in the process, either to the soda leaching process or to secondary conversion. In addition, lithium carbonate was successfully precipitated from the mother liquor in the carbonation stage during the pilot. The goals of the testwork project were achieved and the operation of the Metso LiOH process was demonstrated with selected feed materials with good results. Based on the results of the pilot project, the current process performs well for AMG and Sayona concentrates. 9.2 Recovery Dependencies in Mineral Processing of Syväjärvi, Rapasaari, and Länttä Based on bench-scale and pilot-scale test results undertaken between 2001 and 2017, Keliber developed recovery functions for the main deposits, Syväjärvi, Rapasaari, and Länttä. Not all test results were used, with successful and representative tests being chosen. Based on the test results, it was noted that lithium recovery is dependent on the following key factors: • Deposit from where the sample originated; • Li2O grade of the sample (feed of the test); • Wall rock dilution – wall rock quality and dilution quantity (%); • Scale of the test (laboratory vs pilot); and • Concentrate grade. Keliber’s basic engineering is based on producing 4.5% Li2O concentrate. Therefore, the concentrate grade is fixed at 4.5% and only the effect of other parameters was studied. 9.2.1 Deposit The deposits differ from each other by their flotation response. Test results showed that Syväjärvi performed best. Rapasaari was very similar to Syväjärvi with slightly lower recoveries, but Länttä showed poorer floatation behaviour.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 127 9.2.2 Head Grade The test results confirmed a clear relationship between lithium feed grade and lithium recovery. Laboratory-scale results for pure ore samples without dilution are shown in Figure 9-10. Lithium Recovery at 4.5% Li2O in the Concentrate vs Lithium Grade in the Feed Figure 9-10: Lithium recovery at 4.5% Li2O in the concentrate vs lithium grade in the feed. 9.2.3 Wall Rock Dilution Wall rock dilution reduces the head grade which would result in lower recoveries, but it was shown that the impact is much stronger than the head grade decrease would cause. The grade-recovery curves of the dilution tests included in the geo-metallurgical study are shown in Figure 9-11. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 128 Syväjärvi Rapasaari Länttä Emmes Grade Recovery Curves of the Geo-Metallurgical Dilution Study Figure 9-11: Grade recovery curves of the geo-metallurgical dilution study. Observed differences between the deposit are largely explained by the modal composition of the host rocks, as summarised in Table 9-4. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 129 Table 9-4: Modal composition of the waste rocks of Syväjärvi, Länttä, and Rapasaari. Mineral Deposit Syväjärvi Länttä Länttä Rapasaari Waste/Country Rock Plagioclase Porphyrite %) Amphibolite %) Tourmaline (%) Mica Schist (%) Quartz 8.90 6.60 13.21 30.85 Plagioclase 46.63 33.46 4.49 13.92 Micricline 1.32 0.44 0.07 1.75 Spodumene 0.36 0.00 0.01 0.02 Muscovite 0.19 0.08 10.26 15.09 Epidote 1.58 6.10 1.86 0.00 Biotite 18.60 8.15 13.97 34.91 Tourmaline 0.00 1.79 45.05 2.38 Amphiboles 19.24 40.05 0.04 0.03 Other Mafics 1.62 2.37 0.80 0.34 Other 1.56 .96 10.24 0.71 TOTAL 100.00 100.00 100.00 100.00 Mafic Minerals 39.46 52.36 59.86 37.66 Sheet Silicates 18.79 8.23 24.23 20.00 The impact of dilution on metallurgical results was also shown to be dependent on the MgO content. Diluted feed samples consistently yielded final concentrates with elevated MgO levels, as demonstrated in Figure 9-12, compared to tests without dilution. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 130 Recovery at 4.5% Li2O against MgO% of the Feed Sample Figure 9-12: Recovery at 4.5% Li2O against MgO% of the feed sample. Fitted lines for lithium recovery into the spodumene concentrate vs wall rock dilution in the feed sample are shown for Syväjärvi, Rapasaari, and Länttä in Figure 9-13. Fitted Lines for Lithium Recovery into the Spodumene Concentrate vs Wall Rock Dilution Figure 9-13: Fitted lines for lithium recovery into the spodumene concentrate vs wall rock dilution.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 131 9.2.4 Ore Sorting Ore sorting operates practically in particle size fractions 20–40 mm and 40–100 mm, whereas the 0–20 mm particle size fraction is not sorted due to small particle size and, thus, will bypass sorting. Based on Syväjärvi pilot ore mass balance, it was shown that ore sorting was capable of removing 10.9% of the mass when the wall rock dilution was 15%. This equates to 73% efficiency in the sorter. Thus, it was assumed that 73% of the waste rock is removed from all ore types by the ore sorter, while fines are bypassed. The plant was designed to handle a maximum of 30% dilution. The average mining dilution planned in the mine plan are close to 20%. It should be noted that all sorting tests were conducted with artificilay combined samples, and that liberation of ore based on the blasting regime can impact the efficiency of the sorter. The generation of fines during blasting and double handling of material can also lead to an increase in by-passed material. There is a risk that ore sorting efficiency will vary across the Syväjärvi deposit and that other deposits will not perform with the same efficiency. 9.2.5 Scale-Up from Laboratory- to Full-Scale Keliber considered a number of factors in comparing laboratory and pilot-scale test results. This included slime removal, flotation residence time, losses in cleaning stages, entrainment, rheological factors, and others. Given challenges such as operating cyclones at pilot scale, it was considered fair to assume that full-scale operations could be optimised and lithium losses could be minimised. Therefore, it was estimated that the scale-up factor from the laboratory to the full scale would be slightly lower than observed, and a conservative value of 1.27 percentage points was used for the concentrator plant. No information on the Keliber lithium refinery scale-up factors have been provided. 9.2.6 Summary Recovery Functions [§229.601(b)(96)(iii)(B)(10)(iv)] The final concentrator recovery formula applied to mine planning and financial modelling is as follows (Equation 7): Equation 7: Final recovery formula applied to mine planning and financial modelling. 𝑅𝑒𝑐𝑜𝑣𝑒𝑟𝑦 = 100 − 𝑃1 × (𝑂𝑟𝑒 𝐺𝑟𝑎𝑑𝑒)𝑃2 − 𝑃3 × (% 𝐷𝑖𝑙𝑢𝑡𝑖𝑜𝑛) − 𝑃4 Where: • P1 = Grade parameter 1; multiplier • P2 = Grade parameter 2; exponent • P3 = Dilution parameter • P4 = Scale-up parameter Individual parameters per deposit are shown in Table 9-5. Table 9-5: Recovery parameters. Parameter Syväjärvi Länttä Rapasaari Outovesi Emmes P1 (%) Grade parameter 1; multiplier 10.6 15.0 11.3 11.3 11.3 P2 (%) Grade parameter 2; exponent -0.88 -0.88 -0.88 -0.88 -0.88 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 132 Parameter Syväjärvi Länttä Rapasaari Outovesi Emmes P3 (%) Dilution parameter -0.33 -0.557 -0.272 -0.26 -0.06 P4 (%) Scale-up parameter -1.27 -1.27 -1.27 -1.27 -1.27 86% recovery for the Keliber lithium refinery was used for a 4.5% Li2O concentrate. This is in line with the recoveries observed during the pilot plants tests. 9.3 Adequacy of Data 9.3.1 Ore Sorting Ore sorter performance was based on pilot-scale tests undertaken at equipment manufacturers’ test facilities (Binder+Co, Redwave, and TOMRA). Early tests focussed on optical sorting for the removal of dark waste from ore. More recent tests have assessed laser and XRT sorting. In all cases, the feed to the ore sorting equipment comprised an artificial blend of Syväjärvi ore and waste rock. Optical ore sorting was found to be effective in removing black waste rock from the composite ore feed. With laser and XRT sorting, dark waste particles including plagioclase porphyrite, metasediments, and tuffs, were rejected. Based on pilot-scale XRT ore-sorting test results conducted on the Syväjärvi bulk ore sample, it was concluded that ore sorting is 73% efficient. There is a risk that ore sorting efficiency will vary across the Syväjärvi deposit. Accordingly, it is recommended that ore sorting variability tests be conducted across the Syväjärvi deposit. It was further assumed that the same efficiency would apply to other sources and ore types. There is a risk that other deposits will not perform with the same efficiency. It is, accordingly, recommended that these deposits be subjected to pilot ore sorting and variability tests using XRT ore-sorting technology. The feed to the ore-sorting test equipment comprised an artificial blend of Syväjärvi ore and waste rock. There is a risk that performance on ROM ore may be less efficient than on the artificial composite ore feed due to particles not being liberated. Accordingly, it is recommended that samples of mined ore from all deposits be subjected to pilot ore-sorting tests using XRT ore-sorting technology. The Syväjärvi pilot test conducted in 2019 reported that de-sliming was more efficient with two-stage de-sliming cyclones. The P80 value of slimes was 7 μm, whereas in the 2016 test, the corresponding P80 value was 16 μm. The smaller sizing reduced the Li2O loss to tailings from 6.3% in the 2016 pilot operation down to 4.7% in the 2019 test. The proposed process route includes two-stage de-sliming with hydrocyclones ahead of flotation, but no specific allowance has been made in recovery estimates for de-sliming losses. 9.3.2 Flotation Since 2015, flotation tests have been conducted on various ores at bench- and pilot scale: • Bench: Länttä, Syväjärvi, Rapasaari, Emmes and Outovesi; and • Pilot: Länttä, Syväjärvi and Rapasaari. Flotation parameters are reasonably well understood, but it is recommended that pilot-scale tests be undertaken on the other main sources of ore. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 133 In 2016–2017, a geo-metallurgical study was undertaken on 18 mineralised samples collected from the Syväjärvi, Länttä, and Rapasaari deposits to assess differences in grindability and flotation performance. Furthermore, ore variability flotation tests were undertaken on Rapasaari samples selected from four different mineralised material types. These tests showed significant variability. It is recommended that similar variability programmes be undertaken on all other deposits to ensure an adequate understanding of spatial variability in flotation performance. Ultimately, this should extend into the development of geo-metallurgical models for all deposits. 9.3.3 Conversion The objective of conversion is to convert alpha-spodumene to leachable beta-spodumene. Since 2016, conversion tests have been conducted on various concentrates at bench- and pilot scale: • Bench: Länttä, Syväjärvi, and Rapasaari; and • Pilot: Länttä, Syväjärvi, and Rapasaari. It was recommended in the 2022 TRS that, although conversion parameters are reasonably well understood, pilot- scale tests should be undertaken on the other main sources of concentrate. Subsequent pilot plant testwork was conducted by FLSmidt (evaluation of spodumene alpha to beta phase conversion utilising a cyclone preheater rotary calciner system for Keliber Technologies OY, Finland. January 2024 (Project no: 9232517826)) on the following concentrates: • ~4,000 kg of Sigma spodumene concentrate containing 0.55 wt% moisture and 5.74 wt% Li2O; • ~5,000 kg of AMG spodumene concentrate containing 2.54 wt% moisture and 5.80 wt% Li2O; and • ~7,000 kg of Sayona spodumene concentrate containing 2.91 wt% moisture and 5.94 wt% Li2O. The preheater kiln system operated smoothly and without sintering, achieving a high conversion rate to beta spodumene when the concentrates were calcined. These results were consistent with prior FLS testing on the Syväjärvi and Rapasaari concentrates. Testwork indicated that the optimum conversion temperature differs for the various ore bodies. No tests were conducted on blends form the various areas, and therefore the mining schedule does not incorporate blending from the two mines. Future inclusions from other areas needs to be tested to determine the impact of blending. 9.3.4 Soda Leaching From 2015 to 2017, bench- and pilot-scale tests were undertaken on Länttä and Syväjärvi concentrates including the major process stages, from the spodumene concentrate conversion to lithium carbonate. In 2018, bench tests were undertaken on Syväjärvi and Rapasaari concentrates, including conversion, soda leaching, bi-carbonation, ion exchange, and lithium carbonate crystallisation. Following the decision to produce lithium hydroxide rather than lithium carbonate, semi-continuous bench-scale tests were undertaken in 2019 to produce lithium hydroxide from a beta-spodumene concentrate that was generated in 2018. This was followed by continuous pilot testing of Syväjärvi concentrate in 2020 and Rapasaari concentrate in 2022. The beta-spodumene concentrates used in this hydrometallurgical testwork were calcined in the FLSmidth pilot runs in 2019 and 2021, respectively. The continuous LiOH.H2O pilot was operated for 14 and 17 days, respectively. The S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 134 main process stages were soda leaching, cold conversion, secondary conversion, ion exchange, LiOH.H2O crystallisation, and mother liquor carbonation. The soda leach developed by Outotec has been successfully demonstrated at pilot scale on Syväjärvi and Rapasaari beta-spodumene concentrate. SRK recommended that other concentrates should also be subjected to conversion and hydrometallurgical testing but highlighted that, as the spodumene pegmatites of the Kaustinen area are understood to resemble each other petrographically, mineralogically, and chemically, it is likely that their concentrates will perform similarly to those from Syväjärvi and Rapasaari. SRK (2023 TRS) recommended that the mineralogical and chemical similarity of other concentrates be assessed and that they be subjected to conversion and hydrometallurgical testing if significantly different to Syväjärvi or Rapasaari. Three different commercially available spodumene concentrates were tested in 2024 at the Metso Research Centre after calcination at FLSmidt for hydrometallurgical testwork. The lithium concentrations of the different calcines were as follows: Sayona 2.50%, AMG 2.51%, and Sigma 2.54 %. These corresponded with the average Li2O concentration of 5.4% for Sayona and AMG and 5.5 % for Sigma. Based on the results of the mineralogical investigation as well as the batch leaching tests, calcination of the Sayona and AMG concentrates had been successful, whereas there was potentially some over-calcination of the Sigma concentrate. After batch testwork to fix the parameters, the continuous LiOH pilot test was operated successfully. The main process stages in the pilot were soda leaching, cold conversion, secondary conversion, ion exchange, and LiOH crystallisation. The operation of the Metso LiOH process was demonstrated with selected feed materials with good results. 9.3.1 Crystallisation and Final Product During the 2020 trial, the operation of the crystalliser was less than ideal. However, despite the challenges, a good quality (almost battery grade) LiOH•H2O product was produced at a concentration > 54.9% LiOH. Overdrying was observed during the trial (refer to Table 9-6). During the 2022 trial, the crystallisation stage operated more seamlessly; furthermore, secondary crystallisation was performed where a LiOH•H2O with less impurities was produced. The product concentrations averaged 54.4% LiOH after the first crystallisation; after the second crystallisation the %LiOH averaged 53.2%.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 135 Table 9-6: ICP-OES results of the crystallisation product from all three pilot plant trials. Purity Impurities LiOH.H2O LiOH* CO2 SO4 2- Cl Al Са Cr Cu Fe K Na Si Zn Keliber Target Specs >99.0 - >0.35 <0.005 <0.005 <0.0005 <0.001 <0.0005 <0.0005 <0.0005 <0.001 <0.005 <0.003 <0.0005 Year Sample Details % % % % % % % % % % % % % % 2020 - 98.930 56.459 - - 0.008 0.001 <0.001 <0.0002 <0.0002 0.001 0.002 0.008 0.011 <0.0005 2022 1st Crystallisatio n 95.307 54.391 0.325 <0.015 <0.002 0.003 <0.001 0.0001 <0.0001 0.001 0.004 0.031 0.011 <0.0005 2nd Crystallisatio n 93.278 53.233 0.220 <0.015 <0.002 <0.0005 <0.001 <0.0001 <0.0001 <0.0002 <0.001 0.003 <0.002 <0.0005 2024 AMG Calcine 96.469 55.054 0.294 <0.015 <0.002 <0.001 <0.001 <0.0001 <0.0001 0.0004 0.004 0.032 0.006 <0.0005 Sayona Calcine 97.504 55.645 0.297 <0.015 <0.002 <0.001 <0.001 <0.0001 <0.0001 0.0004 0.006 0.052 0.006 0.0008 *Pilot Plant Specifications >99 >56.5 <0.008 <0.002 <0.002 <0.0001 <0.0007 <0.003 <0.005 <0.005 <0.0001 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 136 During the 2024 trial, a LiOH•H2O product was produced with low impurity levels, therefore, secondary crystallisation was unnecessary. Similar to previous trials, the product concentration ranged between 54.5-56.9% LiOH (averaging 55.6 %LiOH). The shortfall between the actual %LiOH•H2O produced from testwork and the target LiOH•H2O was attributed to underdrying. This was done deliberately to avoid the formation of LiCO3, which forms preferentially in the presence of LiOH. 9.4 Comment [§229.601(b)(96)(iii)(B)(10)(v)] The spodumene pegmatites of the Kaustinen area resemble each other petrographically, mineralogically, and chemically. They are typically coarse-grained, light-coloured, and mineralogically similar. The main minerals are albite (37–41%), quartz (26–28%), K-feldspar (10–16%), spodumene (10–15%), and muscovite (6–7%), generally in this quantitative order. Studies show that the chemical, mineralogical, and geo-metallurgical differences between the deposits are small. Currently, spodumene (LiAlSi2O6) is the only economic mineral identified in the pegmatite veins. Other lithium minerals, for example, petalite, cookeite, montebrasite, and sicklerite, are found only in trace quantities. Beryl and columbite-tantalite are important trace minerals, with mean grades of the deposits as follows: beryllium 60–180 ppm; tantalum 13–60 ppm and niobium 17–60 ppm. The mean chemical compositions of the spodumene grains from three deposits analysed by GTK (Syväjärvi, Rapasaari, and Leviäkangas) are as follows: • SiO2 64.78% to 65.17%; • Al2O3 26.88% to 27.01%; • FeO 0.29% to 0.55%; and • MnO 0.09% to 0.13%. The Li2O content of spodumene is 7.0%, 7.21%, and 7.22% for Syväjärvi, Rapasaari, and Leviäkangas, respectively. Variation in the grindability between the deposits is small, and geo-metallurgical studies show that the hard component in the ores is spodumene and, therefore, the specific grinding energy shows a positive correlation with the lithium grade. In flotation response, the deposits show small differences, mainly due to variation in the lithium head grade and proportion of gangue dilution. Variation in the ore texture, spodumene grain size, colour or alteration do not have an impact on processability. The wall rock dilution has been found to have a negative impact on flotation, lowering the concentrate grade. In this sense, Syväjärvi, where the wall rock dilution is plagioclase porphyrite, has proven to be slightly easier to process than other deposits hosted by mica schist. Minimising the wall rock contamination in flotation is important and, therefore, selective mining and ore sorting will play a significant role in controlling the flotation feed. Keliber have decided not to utilise sulphide flotation ahead of pre-flotation and spodumene flotation to remove sulphide minerals including arsenopyrite to meet the arsenic in the spodumene concentrate specification of 50 ppm, but to put the following measures in place to manage the quality: S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 137 • Arsenopyrite is occurring in high-grade disseminations in the waste rock. By mine planning and selective mining, high As areas (limit of 200 ppm in ore feed in LOM planning) can be avoided; • ROM stockpiles and feed quality control – maximum As level in ore feed – approximately 200 ppm; • Optimise XRT sorting to remove arsenopyrite from the ore feed effectively; and • Remaining arsenic minerals are associated with apatite minerals. Conditions of the pre-flotation are optimum to remove apatite. Then the arsenic will report in the pre-float concentrate and pre-float pond with HPDE liner. The first two measures have been applied in the LOM plan and mining schedule. Arsenopyrite will, however, enter the flotation plant via the fines bypass stream, and excessive fines generation through blasting and double handling could increase the bypass stream and thus contamination. Testwork indicated that the optimum conversion temperature differs for the various ore bodies. No tests were conducted on blends form the various areas and, therefore, the mining schedule does not incorporate blending from the two mines. Future inclusions from other areas need to be tested to determine the impact of blending. The Keliber Project is likely to be the first implementation of the Metso Outotec soda pressure leaching technology. While the individual unit processes are not novel, and while the Syväjärvi (2020), Rapasaari (2022), and commercially available concentrates (2023/2024) pilot trials have significantly de-risked the flowsheet, a residual risk remains as it does with the first implementation of any novel technology. The Project's ore characterisation and mineralogy tests have provided a thorough understanding of the plant feed characteristics, thereby minimising the impact of this risk as the ore's properties and behaviour have been well- defined. Pilot plant test trials were successfully performed on material from the Syväjärvi and Rapasaari deposits, as well as external samples from North America. Despite variations in feed grades, the pilot plant demonstrated consistent behaviour, showing its adaptability to variability in feed ore characteristics and compositions. The design of the equipment, including spares and ancillary equipment, remains a potential risk as any design flaws in equipment could result in delays and costly rework, ultimately impacting the plant's ability to achieve its design capacity and meet production targets. However, the ramp-up period will be instrumental in assessing whether the installed equipment meets the specified performance requirements, thereby mitigating the risk. A secondary risk relates to the availability of critical spares which are essential for a smooth commissioning and production ramp-up. Provisions have been made to address it in the financial model. The process operating parameters were scaled up from the continuous pilot plant trials, therefore, these remain a potential risk (in line with the equipment design). The operating parameters are to be optimised during the ramp-up phase to ensure maximum lithium extraction. Metso Outotec provided a performance guarantee for production throughput and product quality targets, as well as feed qualities. On the basis of this information, recoveries for the hydrometallurgical plant can be calculated. The guaranteed value for recoveries is 84–86% (varying with the assumed conversion degree in calcination). Such a guarantee does not ultimately guarantee a process that will work so much as it defines the extent of financial compensation that will apply should it not. The availability of skilled internal labour and experienced suppliers has not been assessed in this report. To mitigate this risk, it is suggested that training programmes be implemented and skills transferred from the handover commissioning team to the internal team. This will ensure a seamless transition and equip internal staff with the necessary expertise. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 138 Pressure leach installations often encounter prolonged ramp-up periods and frequent breakdowns during commissioning and ramp-up. Consequently, capital costs escalate (impact the profitability of the operation) and many fail to achieve their design throughput. Therefore, a steady and logical ramp-up approach is recommended for the Keliber Project that prioritises efficient process efficiency optimisation. TEMA, the selected supplier of the dryer, have several installations worldwide, including a LiOH plant in South Korea where deviations to the specifications of <0.1% are consistently met. TEMA has confirmed that similar performance can be expected for the Keliber installation, even outperforming the specified requirement of <0.2% outlined in the Keliber process design criteria.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 139 10 MINERAL RESOURCE ESTIMATES [§229.601(b)(96)(iii)(B)(11) 10.1 Introduction This chapter is adapted from Chapter 11 of the TRS for the year ended 31 December 2023 produced by CSA Global South Africa (Pty) Ltd (ERM) and filed on 21 April 2024 by SSW and with changes to the Mineral Resource statement (Section 10.17) as described below and updates to sections 10.18 and 10.19. The Mineral Resources were estimated for the Keliber Project which includes several target areas and were previously reported with an effective date of 31 December 2023. The following provides a description of that work on which the current Mineral Resource is based. The two primary targets (by lithium content), Rapasaari and Syväjärvi, were updated from extensive infill drilling, while the five smaller targets (Tuoreetsaaret, Länttä, Emmes, Leviäkangas, and Outovesi) were either updated from fewer additional drill holes or were not drilled at all since the previous MRE. Mineral Reserve estimation has recently been completed, and the previously reported Mineral Resources figures have been updated to reflect this. All work was carried out using: • Leapfrog Geo® version 2022.1.0; • Datamine Studio RM® version 1.11.300.0; • Supervisor® version 8.13.1; and • JMP® 17.0. The database was established by the collection, validation, recording, storing, and processing of data, and it forms the foundation for the MREs. SOPs were established to govern the collection of all data, while a rigorous QAQC programme was in place to support the database. The Mineral Resource meets the minimum requirement of reasonable prospects for economic extraction (RPEE). The Mineral Resource is based on geological premises, facts, interpretations, and technical information and uses appropriate estimation methods, parameters, and criteria for the deposit under consideration. The Mineral Resources are estimated from drilling assay grades of lithium oxide – Li2O%. Grade interpolation is achieved using Ordinary Kriging (OK) within mineralised domain boundaries interpreted from logged lithological units (spodumene pegmatite) and grade contouring. A cut-off grade (COG) of 0.5 Li2O% is used to report the Mineral Resources. 10.2 Database Individual Microsoft Access® databases were initially supplied by Keliber Technology Oy for the seven targets with various cut-off dates (Table 10-1). In some cases, a database may contain regional data beyond the target area and these data were filtered from the drill hole summary (Table 10-2). During 2023, updated drilling and assay data were supplied for Rapasaari, Syväjärvi, and Tuoreetsaaret. These data were included in the MREs detailed in this TRS. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 140 Table 10-1: Microsoft Access® databases by date. Target Date File Name Rapasaari 2021 April Rapasaari202104.accdb Syväjärvi 2018 December Syväjärvi_DDH_2018dec.accdb Tuoreetsaaret 2022 May Rapasaari-Paivaneva_ddh_2022may31.accdb Länttä 2018 January Länttä_2017.accdb Emmes 2018 March Emmes_DDH_2018-03.accdb Leviäkangas 2014 Leviäkangas_2014.accdb Outovesi 2017 April Outovesi_2017.accdb Table 10-2: Microsoft Access® databases – drilling and assay summary. Target Drill Hole Count Length Drilled (m) Assay Count Length Assayed (m) Rapasaari 347 60,306.10 5,495 7,230.80 Syväjärvi 261 23,581.56 2,073 2,615.20 Tuoreetsaaret 105 24,413.95 1,920 2,545.85 Länttä 105 8,733.38 792 1,025.40 Emmes 54 6,284,79 454 687.90 Leviäkangas 123 6,823.52 300 393.08 Outovesi 24 1,815.60 237 244.97 Bulk density was determined using the hydrostatic immersion (Archimedes) technique on core samples. 10.3 Database Validation The data were reviewed and validated, with minor changes required for use in Mineral Resource estimation. These included a small number of logging overlaps. Assay and bulk density data were reviewed relative to expected values. The assay data contained no unexpected values. Unexpected low or high values encountered in the bulk density data were ignored for use in further work. 10.4 Topography Topography data were not available to construct a 3D surface for use in Mineral Resource estimation. The area is generally flat in nature, therefore, topography surfaces were constructed from surveyed drill hole collars and deemed to be adequate for Mineral Resource estimation. 10.5 Geological Interpretation [§229.601(b)(96)(iii)(B)(11)(i)] 10.5.1 Lithology The host lithology is pegmatite and country rocks that are, for the most part, some form of metasediment, metavolcanite or plagioclase porphyrite sill. The pegmatite is (lithium) mineralised where spodumene is present and S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 141 referred to as spodumene pegmatite. The unmineralised pegmatite is referred to as muscovite pegmatite. Country rock xenoliths are present within the pegmatite, however, the exact geometry of these is generally unknown due to the drill spacing. A soil (overburden) layer overlays the pegmatite and country rock units and was interpreted and modelled due to its barren nature and distinctly different physical properties. For the most part, weathering is not prominent in the pegmatite and country rock. 10.5.2 Mineralisation The host rock displays typical LCT pegmatite zoning, either as unmineralised border/wall zone or mineralised intermediate/core margin zone. Spodumene is the primary lithium mineral present in the mineralised zone. Onsite geologists logged the pegmatite as either spodumene pegmatite or a variation thereof (mineralised), or muscovite pegmatite (unmineralised) based on the presence of spodumene. For geological modelling, further refinements were made whereby the spodumene pegmatite was defined by both logging and a 0.5% Li2O threshold grade. 10.6 Geological Modelling 10.6.1 Lithology The logged lithologies were grouped into simplified modelling units. An example from the regional Rapasaari database illustrates how this was done (Table 10-3). The ten most common lithologies in the database account for 95.1% of the logged metres. These were grouped into overburden, pegmatite, and country rock (metasediment and metavolcanite). The remaining rock types (4.9% of logged metres) were then assigned to one of these simplified groups. Table 10-3: Example from Rapasaari of grouping simplified lithologies for modelling. Lith. Code Count Length (m) Relative % Frequency Cumulative % Frequency Lithology Description Grouped Lithology MS 8,203 33,966.40 39.2 39.2 Mica schist Meta-sediment OVB 708 7,550.54 8.7 48.0 Overburden Overburden PP 1,727 7,243.75 8.4 56.3 Plagioclase porphyrite Plagioclase porphyrite IT 1,599 6,991.35 8.1 64.4 Intermediate metatuff Meta-volcanite SPG 4,381 6,279.80 7.3 71.7 Spodumene pegmatite Pegmatite GW 1,344 6,029.95 7.0 78.6 Metagreywacke Meta-sediment MS_pfb 778 5,088.85 5.9 84.5 Mica schist with porphyroblasts Meta-sediment IV 833 4,119.23 4.8 89.3 Intermediate metavolcanite Meta-volcanite MPG 4,099 2,989.12 3.5 92.7 Muscovite pegmatite Pegmatite SS 487 2,087.45 2.4 95.1 Sulphidic mica schist Meta-sediment S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 142 10.6.2 Mineralisation A pegmatite model was initially modelled from the group lithology field and further refined into spodumene and muscovite pegmatite. For some targets, multiple pegmatites were present. When this was the case, a categorical code was used to distinguish between the various pegmatites and internal xenoliths. The modelling methodology described above was applied to all targets apart from Rapasaari (Figure 10-1 to Figure 10-5). Oblique Views of the Modelled Pegmatites for Emmes, Länttä, Leviäkangas, and Outovesi (INT = Internal Xenolith; MPEG = Muscovite Pegmatite; SPEG = Spodumene Pegmatite) Figure 10-1: Oblique views of the modelled pegmatites for Emmes, Länttä, Leviäkangas, and Outovesi (INT = internal xenolith; MPEG = muscovite pegmatite; SPEG = spodumene pegmatite) (after ERM 2024).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 143 Plan View of the Modelled Pegmatites at Syväjärvi (Blue = Internal Xenolith; Yellow = Muscovite Pegmatite; Red = Spodumene Pegmatite) (Oblique View) Figure 10-2: Plan view of the modelled pegmatites at Syväjärvi (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) (oblique view) (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 144 View looking West showing the Modelled Pegmatites at Syväjärvi (Blue = Internal Xenolith; Yellow = Muscovite Pegmatite; Red = Spodumene Pegmatite) and Topography (Green) (Oblique View) Figure 10-3: View looking west showing the modelled pegmatites at Syväjärvi (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) and topography (green) (oblique view) (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 145 Plan View of the Modelled Pegmatites at Tuoreetsaaret (Blue = Internal Xenolith; Yellow = Muscovite Pegmatite; Red = Spodumene Pegmatite) (Oblique View) Figure 10-4: Plan view of the modelled pegmatites at Tuoreetsaaret (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) (oblique view) (after ERM 2024). View looking West showing the Modelled Pegmatites at Tuoreetsaaret (Blue = Internal Xenolith; Yellow = Muscovite Pegmatite; Red = Spodumene Pegmatite) and Topography (Green) (Oblique View) Figure 10-5: View looking west showing the modelled pegmatites at Tuoreetsaaret (blue = internal xenolith; yellow = muscovite pegmatite; red = spodumene pegmatite) and topography (green) (oblique view) (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 146 10.6.3 Rapasaari 10.6.3.1 Background The pegmatites at Rapasaari are highly variable in terms of strike, dip, thickness, and zonation. The modelling methodology applied to the pegmatites at other targets was attempted at Rapasaari but did not yield the expected results in terms of the geological interpretation of the deposit. An alternate methodology was applied whereby numeric models were constructed for the grouped pegmatite lithology and then refined for the spodumene, muscovite, and xenolith zones within it. 10.6.3.2 Pegmatite The Rapasaari pegmatite was modelled using the Indicator RBF Interpolant in Leapfrog Geo as follows: • A new field (IND_PEG) was added to the geological log. • Values were assigned based on pegmatite vs non-pegmatite lithologies such that pegmatite lithologies were assigned a value of 1 and non-pegmatites a value of 0. • The Indicator RBF Interpolant used these values to model the pegmatite, with a cut-off value of 1 applied as the threshold. • The pegmatite model was limited to below the base of the overburden. • A non-decaying 3D structural trend (strength = 10), which was constructed based on the previous interpretation of the pegmatite, was applied to guide the dip and strike of the newly interpreted pegmatite. • A surface resolution of 1.7 m and an iso-value of 0.5 were applied for the construction of the mesh. • Output volumes less than 5,000 m3 were automatically discarded. • Isolated volumes greater than 5,000 m3 were manually discarded; this included volumes that did not adhere to the geological interpretation in terms of their geometry (Figure 10-6 and Figure 10-7) and demonstrates the variable nature of spodumene pegmatite relative to the muscovite pegmatite. At other targets, the spodumene pegmatite is generally located within a mineralised intermediate/core margin zone. Spatially, these occur away from country rock contacts, central, relative to the total pegmatite extent. At Rapasaari, this is not always the case, with the spodumene pegmatite occurring randomly relative to the unmineralised muscovite pegmatite and country rock contacts.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 147 Plan View showing Pegmatites at Rapasaari from Numeric Modelling (Left-Hand Side) and Pegmatite Zones (Right-Hand Side) (INT = Internal Xenolith; MPEG = Muscovite Pegmatite; SPEG = Spodumene Pegmatite); Cross-Section Location in Green (Oblique View) Figure 10-6: Plan view showing pegmatites at Rapasaari from numeric modelling (left-hand side) and pegmatite zones (right- hand side) (INT = internal xenolith; MPEG = muscovite pegmatite; SPEG = spodumene pegmatite); cross-section location in green (oblique view) (after ERM 2024). Cross-Section looking North showing Pegmatite Zones at Rapasaari Relative to Drill Holes (INT = Internal Xenolith; MPEG = Muscovite Pegmatite; SPEG = Spodumene Pegmatite) (Oblique View) Figure 10-7: Cross-section looking north showing pegmatite zones at Rapasaari relative to drill holes (INT = internal xenolith; MPEG = muscovite pegmatite; SPEG = spodumene pegmatite) (oblique view) (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 148 10.7 Compositing Core was sampled at various lengths depending on the location of geological contacts; however, 1 m and 2 m sample lengths were the most commonly used to sample the pegmatite. Samples were flagged according to the internal pegmatite lithologies i.e. spodumene or muscovite and were composited to 2 m lengths within these zones. 10.8 Exploratory Data Analysis The assay samples and 2 m composites were selected within the spodumene pegmatite and the statistics were calculated. Drill sample and composite statistics for lithium oxide are listed per deposit in Table 10-4 and Table 10-5. Negative values in the datasets were substituted with the lowest positive value to reflect a barren sampling interval. Missing spodumene intervals were left as absent values. 10.9 Top Caps No top caps were applied to the lithium oxide grades as the distributions were near normal. Table 10-4: Naive statistics for Li2O%. Deposit Samples Minimum Maximum Mean Standard Deviation Coefficient of Variation Rapasaari 2,104 0.02 4.05 1.28 0.54 0.42 Syväjärvi 934 0.03 4.05 1.35 0.58 0.43 Tuoreetsaaret 267 0.03 3.44 0.91 0.49 0.54 Länttä 261 0.02 2.65 1.22 0.42 0.34 Emmes 198 0.07 2.41 1.32 0.49 0.38 Leviäkangas 143 0.03 8.68 1.11 0.50 0.45 Outovesi 84 0.27 2.63 1.42 0.52 0.37 Table 10-5: Composite statistics for Li2O%. Deposit Samples Minimum Maximum Mean Standard Deviation Coefficient of Variation Rapasaari 1,653 0.05 3.24 1.26 0.46 0.37 Syväjärvi 748 0.09 3.29 1.34 0.50 0.37 Tuoreetsaaret 209 0.05 2.28 0.90 0.43 0.48 Länttä 270 0.02 2.22 1.21 0.38 0.32 Emmes 179 0.14 2.41 1.32 0.45 0.34 Leviäkangas 108 0.33 2.42 1.10 0.39 0.35 Outovesi 46 0.49 2.03 1.41 0.40 0.28 10.10 Variography [§229.601(b)(96)(iii)(B)(11)(iv)] Directional semi-variogram models were fitted using Supervisor® software for lithium oxide in the larger deposits (Table 10-6). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 149 Table 10-6: Variogram parameters for Li2O%. Deposit Datamine Rotation Rotation Axis Nugget Structure 1 Structure 2 Sill Range Sill Range Rapasaari 60 Z 0.20 0.31 39 0.49 89 40 X 27 82 180 Y 9 56 Syväjärvi 0 Z 0.25 0.41 27 0.34 85 20 X 23 95 0 Y 7 18 Tuoreetsaaret 10 Z 0.37 0.30 46 0.33 120 140 X 28 60 0 Y 7 40 Länttä 120 Z 0.20 0.25 40 0.55 100 60 X 24 70 0 Y 10 20 Emmes 410 Z 0.2 0.32 77 0.48 109 130 X 29 91 0 Y 10 20 The orientations of the variograms correspond to the orientation (dip/strike) of mineralisation. Nugget values for lithium oxide were generally low. The variogram models for lithium oxide for Rapasaari and Syväjärvi are shown in Figure 10-8 and Figure 10-9. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 150 Variogram Models for Lithium Oxide at Rapasaari (After ERM 2024) Figure 10-8: Variogram models for lithium oxide at Rapasaari (after ERM 2024).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 151 Variogram Models for Lithium Oxide at Syväjärvi (After ERM 2024) Figure 10-9: Variogram models for lithium oxide at Syväjärvi (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 152 10.11 Block Model Block model parameters are listed in Table 10-7. The block models cover mineralisation extents at each deposit and sub-celling is used for greater resolution of the mineralised domains. The blocks were coded according to the appropriate zone codes and weathering domains. Table 10-7: Block model parameters. Deposit Axis Minimum (m) Maximum (m) Distance (m) Block Size (m) Number Blocks Sub-Cell (m) Rapasaari X 2,491,550 2,493,200 1,650 5 330 1.25 Y 7,059,885 7,061,725 1,840 10 184 1.25 Z -340 130 470 5 94 1.25 Syväjärvi X 2,489,380 2,491,220 1,840 5 368 1.25 Y 7,061,470 7,063,330 1,860 10 186 1.25 Z -150 120 270 5 54 1.25 Tuoreetsaaret X 2,490,500 2,492,040 1,540 5 308 1.25 Y 7,059,870 7,061,670 1,800 10 180 1.25 Z -200 170 370 5 74 1.25 Länttä X 2,506,700 2,507,980 1,280 5 256 1.25 Y 7,057,130 7,058,300 1,170 10 117 1.25 Z -160 160 320 5 64 1.25 Emmes X 2,478,845 2,479,865 1,020 10 102 1.25 Y 7,063,055 7,064,000 945 15 63 1.25 Z -220 70 290 10 29 1.25 Leviäkangas X 2,486,390 2,487,250 860 10 86 1.25 Y 7,058,940 7,059,780 840 10 84 1.25 Z -120 100 220 5 44 1.25 Outovesi X 3,338,320 3,339,010 690 5 138 1.25 Y 7,066,480 7,067,250 770 10 77 1.25 Z -40 100 140 5 28 1.25 10.12 Grade Estimation Lithium oxide grades were estimated using OK for the larger domains where a variogram model was fitted to the composite data. Inverse distance weighted to the power of 2 (IDW2) was used to estimate grade for the smaller composite datasets. Kriging neighbourhood analysis (KNA) was carried out with the lithium oxide variogram models using Supervisor® software. Search neighbourhood parameters are listed in Table 10-8. The search orientation was based on the variogram where available and on the modelled orientation of mineralisation for smaller deposits. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 153 Three search passes were completed, with distance of the first pass equal to the variogram ranges, where available, and based on drill spacing in the smaller domains. Search neighbourhoods were informed from the modelled variogram directions and ranges to locate composites for estimation. Minimum and maximum samples in the search were informed by the KNA, where available, and applied to the smaller domains with sensitivity testwork. The Mineral Resources are estimated within the entire constructed block model, but those at a 0.5% Li2O COG are reported in the Mineral Resource statement (Section 10.17). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 154 Table 10-8: Search parameters. Deposit Orientation Search Volume 1 Search Volume 2 Search Volume 3 MAXKEY Datamine Axis Datamine Rotation (°) Ranges Composites Ranges Composites Ranges Composites Min. Max. Min. Max. Min. Max. Rapasaari Z 60 89 12 24 178 12 24 445 4 20 4 X 64 82 164 410 Y 180 56 112 180 Syväjärvi Z 0 85 12 24 170 12 24 425 4 20 4 X 20 95 190 475 Y 0 18 36 90 Tuoreetsaaret Z 10 120 12 24 240 12 24 840 4 20 4 X 140 60 120 420 Y 0 40 80 280 Länttä Z 120 100 12 24 200 12 24 500 4 12 4 X 60 70 140 350 Y 0 20 40 100 Emmes Z 40 109 12 24 218 12 24 545 8 24 4 X 130 91 182 455 Y 0 20 40 100 Leviäkangas Z 80 100 12 24 200 12 24 500 4 12 4 X 130 50 100 250 Y 0 20 40 100 Outovesi Z -60 100 12 24 200 12 24 500 4 12 4 X 70 50 100 250 Y 0 20 40 100

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 155 10.13 Validation [§229.601(b)(96)(iii)(B)(11)(iv)] The block model estimates were validated by: • Global statistics; • Swath analysis; and • Localised visual validation on cross-sections. 10.13.1 Global Statistics Global mean values were calculated for the input composites and output estimates. These were compared to assess the global representivity of the model versus the composites. Variance between the mean estimated and composite lithium oxide grades is less than 5% (Table 10-9). Table 10-9: Comparison between the input composites and ordinary kriged estimates. Deposit Method Mean – 2 m Cut Composites Mean – OK Estimate Relative % Difference Rapasaari OK 1.26 1.22 -3% Syväjärvi OK 1.34 1.31 -2% Tuoreetsaaret OK 0.9 0.86 -4% Länttä OK 1.21 1.19 -2% Emmes OK 1.32 1.29 -2% Leviäkangas IDW2 1.1 1.06 -4% Outovesi IDW2 1.41 1.41 0% 10.13.2 Swath Analysis Swath plots were compiled to validate the estimates on a semi-local scale. Plots for lithium oxide for Rapasaari and Syväjärvi are shown in Figure 10-10 and Figure 10-11, respectively. Trends in the composite grades are reflected in the block estimates. 10.13.3 Localised Visual Validation Cross sections were examined to compare the input composites against the estimated block model. This process validated the model on a local scale when comparing the estimated blocks in the vicinity of the input composites. The process showed an acceptable correlation between composites and estimates. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 156 Swath Plots for Lithium Oxide at Rapasaari; Composites as Orange Line; Block Estimates as Black Line (After ERM 2024) Figure 10-10: Swath plots for lithium oxide at Rapasaari; composites as orange line; block estimates as black line (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 157 Swath Plots for Lithium Oxide at Syväjärvi; Composites as Orange Line; Block Estimates as Black Line (After ERM 2024) Figure 10-11: Swath plots for lithium oxide at Syväjärvi; composites as orange line; block estimates as black line (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 158 10.14 Density Density determinations were conducted using the water displacement (Archimedes) method and included the use of two standards that were measured at every 10th sample. Most of the density determinations were performed on pegmatite material but also included non-mineralised material within the pegmatite (host rock inclusions/xenoliths) and country rock. There is a strong correlation between density and Li2O grade in the Rapasaari pegmatite (Figure 10-12), and a regression analysis was conducted to establish if density could be assigned from a reliable regression formula. Scatterplot of Li2O Grade vs SG at Rapasaari (Bivariate Fit of SG by Li2O Zone = 1) (Spodumene Pegmatite) Figure 10-12: Scatterplot of Li2O grade vs SG at Rapasaari (bivariate fit of SG by Li2O zone = 1) (spodumene pegmatite) (after ERM 2024). Li2O grade bins were assigned based on 0.1% Li2O increments, and the mean density was calculated within each bin. These values were then plotted against one another (Figure 10-13). The analysis demonstrated that a reliable formula could be applied for density estimation within the pegmatite based on the existing density determination data and the Li2O grade. The formula overleaf (Equation 8) was limited by a maximum value of 2.8 t/m3:

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 159 Equation 8: Mean SG dependence on grade at Rapasaari. 𝑀𝑒𝑎𝑛 (𝑆𝐺) = 2.631 + 0.05959 × 𝐿𝑖2𝑂 𝑔𝑟𝑎𝑑𝑒 Linear Fit Mean (SG) = 2.630885 + 0.05959 X Li₂O CLASS Summary of Fit RSquare 0.96185 RSquare adj 0.960383 Root Mean Square error 0.010094 Mean of Response 2.717716 Observations (or Sum Var) 28 Analysis of Variance Source DF Sum of Squares Mean Square F Ratio Model 1 0.06678856 0.066789 665.5265 Error 26 0.00264902 0.000102 Prob > F C. Total 27 0.06943758 <0.001* Parameter Estimates Term Estimate Std Error T Ratio Prob > ItI Intercept 2.630885 0.003891 676.13 <0.0001* Li2O CLASS 0.05959 0.002327 25.60 <0.0001* Regression of Li2O Grade Bins vs Average SG at Rapasaari (Bivariate Fit of Mean SG by Li2O) (After ERM 2024) Figure 10-13: Regression of Li2O grade bins vs average SG at Rapasaari (bivariate fit of mean SG by Li2O) (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 160 The same analysis was conducted for Syväjärvi with the formula below (Equation 9), limited by a maximum of 2.85 t/m3: Equation 9: Mean SG dependence on grade at Syväjärvi. 𝑀𝑒𝑎𝑛 (𝑆𝐺) = 2.636 + 0.0633 × 𝐿𝑖2𝑂 𝑔𝑟𝑎𝑑𝑒 Due to the smaller number of density determinations at the other targets, a reliable regression analysis could not be carried out. The formula for Rapasaari was therefore applied to the spodumene domains at all other targets. Mean density values by rock type were applied for all non-spodumene domains. 10.15 Mineral Resource Classification The quality and quantity of data, geological understanding and continuity, along with grade continuity were considered for Mineral Resource classification. Data are generally of acceptable quality and are sufficient in number to reasonably understand the geological setting and nature of grade continuity. Considering these criteria, the Mineral Resources are classified as follows: • Inferred Mineral Resources are classified up to 30 m beyond drilling data. • Indicated Mineral Resources are classified up to 20 m beyond drilling data and are supported by drilling at a spacing of 40 m x 40 m. • Measured Mineral Resources are classified up to 15 m beyond drilling data and are supported by drilling at a spacing of 30 m x 30 m. At Rapasaari and Syväjärvi, some highly continuous spodumene pegmatite volumes were classified as Measured Mineral Resources at a drill spacing of 40 m x 40 m (Figure 10-14 and Figure 10-15). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 161 Mineral Resource Classification at Rapasaari with Drill Hole Collar Locations (After ERM 2024) Figure 10-14: Mineral Resource classification at Rapasaari with drill hole collar locations (after ERM 2024). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 162 Mineral Resource Classification at Syväjärvi with Drill Hole Collar Locations (After ERM 2024) Figure 10-15: Mineral Resource classification at Syväjärvi with drill hole collar locations (after ERM 2024). 10.16 Reasonable Prospects for Economic Extraction [§229.601(b)(96)(iii)(B)(11)(iii) (vi) (vii)] A long-term LiOH price of USD35,000/t was selected for the determination of RPEE. Conceptual costs and mining parameters were applied (Table 10-10). There is considerable uncertainty associated with the lithium market given the rapid changes in supply and demand, but the assumptions used by the Project are aligned with current forecasts. The long-term price assumption used has been derived from consensus economic forecasts. Inclusion of lithium forecasts is relatively new and recent drops in prices and volatility have occurred between 2021 and 2024, during

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 163 which spot prices ranged between approximately USD9,500/t and USD80,000/t. All extraction is assumed to be by open-pit mining. Table 10-10: Conceptual parameters used to determine RPEE. Parameter Unit Value LiOH product selling price USD 35,000 Exchange rate USD to EUR 1.1 Royalties EUR/t 1.69 Fixed mining cost (ore) EUR/t 2.73 Fixed mining cost (waste) EUR/t 2.91 Fixed mining cost (overburden) EUR/t 0.97 Mining cost (auxiliary) EUR/t 0.07 Processing cost EUR/t 51.50 Lithium yield % 74.3 Pit slope angle ° 45 10.17 Mineral Resource Statement [§229.601(b)(96)(iii)(B)(11)(ii)] The MRE for the Project is reported in accordance with S-K 1300 and was last modelled and updated as at 31 December 2023. However, the latest (2024) Mineral Reserve update required an updated MRE exclusive of Mineral Reserves with an effective date of 31 December 2024. For reporting the Keliber Mineral Resources, the following definition, as set forth in the S-K 1300 Definition Standards, was applied. “A Mineral Resource is a concentration or occurrence of material of economic interest in or on the Earth’s crust in such form, grade or quality, and quantity that there are reasonable prospects for economic extraction. A Mineral Resource is a reasonable estimate of mineralisation, taking into account relevant factors such as COG, likely mining dimensions, location or continuity, that, with the assumed and justifiable technical and economic conditions, is likely to, in whole or in part, become economically extractable. It is not merely an inventory of all mineralisation drilled or sampled.” Mineral Resources are not Mineral Reserves and have not demonstrated economic viability. The reported Inferred Mineral Resources are considered too speculative geologically to have economic considerations applied to them that would enable them to be categorised as Mineral Reserves. There is no certainty that all or any part of this Mineral Resources will be converted into Mineral Reserves as defined by S-K 1300. The in situ Mineral Resources, exclusive of Mineral Reserves, are reported on a 79.82% ownership basis. Mineral Reserves were updated for the Project and used to present a subsequent Mineral Resources exclusive of Mineral Reserves where the updated Mineral Reserve pit shells were used to define the Mineral Reserve-Mineral Resource boundary for reporting purposes. Mineral Resources are reported in accordance with the definitions presented in S-K 1300. The effective date of the Mineral Resource is 31 December 2024 (Table 10-11). The Mineral Resources are reported at a 0.5% Li2O COG. For comparison, Table 10-12 and Table 10-13 present the Mineral Resources inclusive of Mineral Reserves on a 100% basis and 79.82% attributable ownership basis respectively. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 164 Table 10-11: Keliber Mineral Resources, exclusive of Mineral Reserves, at a 0.5% Li2O cut-off as at 31 December 2024 and reported on a 79.82% ownership basis. Deposit Mineral Resource Classification Tonnage (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) Rapasaari Measured 0.05 0.59 1.27 1.7 Indicated 0.85 0.52 1.13 23.6 Measured + Indicated 0.90 0.53 1.14 25.3 Inferred 1.50 0.58 1.27 46.3 Syväjärvi Measured 0.09 0.56 1.21 2.6 Indicated 0.19 0.59 1.28 6.1 Measured + Indicated 0.28 0.58 1.26 8.7 Inferred 0.28 0.58 1.24 8.6 Tuoreetsaaret Measured - - - - Indicated 0.33 0.43 0.94 7.6 Measured + Indicated 0.33 0.43 0.94 7.6 Inferred 1.38 0.40 0.87 29.5 Länttä Measured 0.33 0.59 1.27 10.4 Indicated 0.57 0.55 1.18 16.6 Measured + Indicated 0.90 0.56 1.21 27.0 Inferred 0.35 0.54 1.16 10.0 Emmes Measured - - - - Indicated 0.67 0.62 1.33 21.9 Measured + Indicated 0.67 0.62 1.33 21.9 Inferred 0.29 0.61 1.31 9.5 Outovesi Measured - - - - Indicated 0.13 0.64 1.38 4.4 Measured + Indicated 0.13 0.64 1.38 4.4 Inferred 0.12 0.67 1.44 4.3 Leviäkangas Measured - - - - Indicated 0.19 0.55 1.19 5.7 Measured + Indicated 0.19 0.55 1.19 5.7 Inferred 0.55 0.47 1.19 13.6 Total Measured 0.47 0.58 1.26 14.7 Indicated 2.93 0.55 1.19 86.0 Measured + Indicated 3.41 0.56 1.20 100.7 Inferred 4.48 0.51 1.10 121.9 Notes: 1. Mt is million tonnes, kt is thousand tonnes, LCE is lithium carbonate equivalent. (conversions used: Li2O = Li x 2.153; LCE = Li x 5.324). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 165 2. Figures have been rounded to the appropriate level of precision for the reporting of Mineral Resources. 3. Mineral Resources are stated as in situ dry tonnes; figures are reported in metric tonnes. 4. The Mineral Resource has been classified under the guidelines of S-K 1300. 5. The Mineral Resource has demonstrated reasonable prospects for economic extraction based on conceptual mining and costs parameters. 6. Mineral Resources are reported on a 79.82% ownership basis. 7. Concentrator recovery of 88% and refinery recovery of 86%. Table 10-12: Keliber Mineral Resources, inclusive of Mineral Reserves, at a 0.5% Li2O cut-off reported on a 100% basis Deposit Mineral Resource Classification Tonnes (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) *Rapasaari Measured 2.02 0.59 1.28 63.8 Indicated 6.74 0.56 1.22 202.5 Measured + Indicated 8.76 0.57 1.23 266.3 Inferred 1.88 0.58 1.25 58.0 *Syväjärvi Measured 1.69 0.65 1.40 58.4 Indicated 0.90 0.60 1.29 28.8 Measured + Indicated 2.59 0.63 1.36 87.2 Inferred 0.35 0.58 1.24 10.8 Tuoreetsaaret Measured - - - - Indicated 0.41 0.43 0.94 9.6 Measured + Indicated 0.41 0.43 0.94 9.6 Inferred 1.72 0.40 0.87 36.9 Länttä Measured 0.42 0.59 1.27 13.0 Indicated 0.72 0.55 1.18 20.8 Measured + Indicated 1.13 0.56 1.21 33.9 Inferred 0.44 0.54 1.16 12.6 Emmes Measured - - - - Indicated 0.83 0.62 1.33 27.4 Measured + Indicated 0.83 0.62 1.33 27.4 Inferred 0.37 0.61 1.31 11.8 Outovesi Measured - - - - Indicated 0.16 0.64 1.38 5.5 Measured + Indicated 0.16 0.64 1.38 5.5 Inferred 0.15 0.67 1.44 5.4 Leviäkangas Measured - - - - Indicated 0.24 0.55 1.19 7.2 Measured + Indicated 0.24 0.55 1.19 7.2 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 166 Deposit Mineral Resource Classification Tonnes (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) Inferred 0.69 0.47 1.00 17.1 Total Measured 4.12 0.62 1.33 135.2 Indicated 10.02 0.57 1.22 301.9 Measured + Indicated 14.14 0.58 1.25 437.0 Inferred 5.61 0.51 1.10 152.7 Notes: 1. Mt is million tonnes, kt is thousand tonnes, LCE is lithium carbonate equivalent. (conversions used: Li2O = Li x 2.153; LCE = Li x 5.324). 2. Figures have been rounded to the appropriate level of precision for the reporting of Mineral Resources. 3. Mineral Resources are stated as in situ dry tonnes; figures are reported in metric tonnes. 4. The Mineral Resource has been classified under the guidelines of S-K 1300. 5. The Mineral Resource has demonstrated reasonable prospects for economic extraction based on conceptual mining and costs parameters. 6. Mineral Resources are reported on a 100% basis (refer to Table 10-11 and Table 10-13 for MRE for attributable ownership basis). 7. Concentrator recovery of 88% and refinery recovery of 86%. Table 10-13: Keliber Mineral Resources, inclusive of Mineral Reserves, at a 0.5% Li2O cut-off reported on a 79.82% attributable ownership basis. Deposit Mineral Resource Classification Tonnes (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) *Rapasaari Measured 1.61 0.59 1.28 50.9 Indicated 5.38 0.56 1.22 161.7 Measured + Indicated 6.99 0.57 1.23 212.6 Inferred 1.50 0.58 1.25 46.3 *Syväjärvi Measured 1.35 0.65 1.40 46.6 Indicated 0.72 0.60 1.29 23.0 Measured + Indicated 2.07 0.63 1.36 69.6 Inferred 0.28 0.58 1.24 8.6 Tuoreetsaaret Measured - - - - Indicated 0.33 0.43 0.94 7.6 Measured + Indicated 0.33 0.43 0.94 7.6 Inferred 1.38 0.40 0.87 29.5 Länttä Measured 0.33 0.59 1.27 10.4 Indicated 0.57 0.55 1.18 16.6 Measured + Indicated 0.90 0.56 1.21 27.0 Inferred 0.35 0.54 1.16 10.0 Emmes Measured - - - - Indicated 0.67 0.62 1.33 21.9

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 167 Deposit Mineral Resource Classification Tonnes (Mt) Grade (%Li) Grade (%Li2O) LCE (kt) Measured + Indicated 0.67 0.62 1.33 21.9 Inferred 0.29 0.61 1.31 9.5 Outovesi Measured - - - - Indicated 0.13 0.64 1.38 4.4 Measured + Indicated 0.13 0.64 1.38 4.4 Inferred 0.12 0.67 1.44 4.3 Leviäkangas Measured - - - - Indicated 0.19 0.55 1.19 5.7 Measured + Indicated 0.19 0.55 1.19 5.7 Inferred 0.55 0.47 1.00 13.6 Total Measured 3.29 0.62 1.33 107.9 Indicated 7.99 0.57 1.22 240.9 Measured + Indicated 11.29 0.58 1.25 348.8 Inferred 4.48 0.51 1.10 121.9 Notes: 1. Mt is million tonnes, kt is thousand tonnes, LCE is lithium carbonate equivalent. (conversions used: Li2O = Li x 2.153; LCE = Li x 5.324). 2. Figures have been rounded to the appropriate level of precision for the reporting of Mineral Resources. 3. Mineral Resources are stated as in situ dry tonnes; figures are reported in metric tonnes. 4. The Mineral Resource has been classified under the guidelines of S-K 1300. 5. The Mineral Resource has demonstrated reasonable prospects for economic extraction based on conceptual mining and costs parameters. 6. Mineral Resources are reported on a 79.82% ownership basis. 7. Concentrator recovery of 88% and refinery recovery of 86%. 10.17.1 Conversions Lithium Mineral Resources total metal content is often quoted in Lithium carbonate (Li2CO3) equivalent (LCE), which is one of the final products produced in the lithium mining value chain. LCE is derived by multiplying the in situ Li content by a factor of 5.323. Lithium hydroxide monohydrate (LiOH.H2O) is derived by dividing the LCE by a factor of 0.88. Li2O has been derived from Lithium (Li) by multiplying by a factor of 0.2153. These standardised conversion factors are provided in Table 10-14. Table 10-14: Lithium product conversion matrix. Element/Mineral Li Li2O Li2CO3 Li - 2.153 5.324 Li2O 0.464 - 2.473 Li2CO3 0.188 0.404 - LiOH.H2O 0.165 0.356 0.880 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 168 10.18 Comparison with the Previous MRE The Mineral Resource estimate for 2024 is based on the current 2023 Mineral Resource estimate, but updated to exclude the new 2024 Mineral Reserve estimate. Prior to the 2023 Mineral Resource model update, the MRE was last reported by SRK Consulting in December 2022. In that period, SSW completed infill drilling at Rapasaari and Syväjärvi and extensional drilling at Tuoreetsaaret which is used in the current MRE. On a Mineral Resource inclusive of Mineral Reserve basis, there was an increase in tonnes and grade above cut-off for the Project as a whole in the Measured Mineral Resource and Indicated Mineral Resource (M&I) material, which resulted in a 20% increase in contained metal (on a 100% basis). The increase was primarily driven by the larger deposits, namely Rapasaari and Syväjärvi, which in combination contribute 81% to the Project M&I Mineral Resource. Rapasaari is the largest deposit and contributes 61% to the Project M&I material. There is an 33% increase in contained metal at Rapasaari and 7% increase at Syväjärvi (M&I only, 100% basis). The change in Mineral Resources was informed by the following factors: • The lithium hydroxide monohydrate price used to generate the RPEE shells had increased from USD15,000 to USD35,000 since the previous MRE reporting in 2022, which has resulted in an increase in the reporting pit shells for each of the deposits. The long-term price assumption used was derived from consensus economic analysts’ forecasts. The larger pit shells at Rapasaari and Syväjärvi was the main driver for the overall increase in total in-situ Mineral Resources in the updated MRE. • There was infill drilling at Rapasaari and Syväjärvi, as well as extensional drilling at Tuoreetsaaret. The additional drilling at Tuoreetsaaret converted a proportion of the deposit to Indicated Mineral Resource, where it was all classified as Inferred Mineral Resources in the previous MRE. Mean input composite grades are higher for all three deposits when compared with the previous estimate. • There was a change in methodology for the interpretation of the mineralisation volume since the previous 2022 dated MRE reported by SRK. In the current (2023) Mineral Resource model, ERM refined the purely lithological interpretation of the mineralised boundary with a grade shell based on a 0.5% Li2O threshold. Internal waste in the form of xenoliths was modelled and the volumes excised from the mineralised spodumene pegmatite zone. The mean grade of input composite increased for all deposits in the project as a result of the updated interpretation. • There was a decrease in the contained metal for the Emmes deposit in the M+I category where there was 16% less tonnage due to the reclassification of some of this material as Inferred Mineral Resources. On an attributable basis, exclusive of Mineral Reserves, the year-on-year change (2023 vs 2024) in Mineral Resources was informed by the following factors: • Länttä and Outovesi have been excluded from the Mineral Reserves due to technical considerations. • The Rapasaari and Syväjärvi Mineral Reserves have been updated and increased materially. • At the Rapasaari and Syväjärvi deposits, inferred material that fell inside the Mineral Reserve pit shell have been included in the 2024 Mineral Resource estimate. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 169 Table 10-15 shows a comparison of the 2024 MRE (exclusive of Mineral Reserves) against the 2023 MRE (exclusive of Mineral Reserves) on an attributable ownership of 79.82%. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 170 Table 10-15: Comparison between the 2024 and 2023 MREs (exclusive of Mineral Reserves) at 0.5% cut-off. 31 December 2024 reported on a SSW- attributable ownership of 79.82% 31 December 2023 reported on a SSW- attributable ownership of 79.82% Variance Deposit Mineral Resource Classification Tonnes Grade Grade LCE Tonnes Grade Grade LCE Arithmetic % (Mt) (% Li) (% Li2O) (kt) (Mt) (% Li) (% Li2O) (kt) Tonnes (Mt) LCE Content (kt) (% Li2O) Mass LCE Content Rapasaari Measured 0.05 0.59 1.27 1.7 0.21 0.61 1.31 6.9 -0.16 -5.2 -0.04 -74% -75% Indicated 0.85 0.52 1.13 23.6 1.82 0.55 1.17 52.8 -0.97 -29.3 -0.05 -54% -55% Measured + Indicated 0.90 0.53 1.14 25.3 2.03 0.55 1.19 59.7 -1.13 -34.4 -0.05 -56% -58% Inferred 1.50 0.58 1.25 46.3 1.01 0.58 1.26 31.5 0.49 14.8 -0.01 49% 47% Syväjärvi Measured 0.09 0.56 1.21 2.6 0.11 0.55 1.19 3.3 -0.03 -0.7 0.02 -23% -21% Indicated 0.19 0.59 1.28 6.1 0.37 0.60 1.29 11.7 -0.17 -5.6 -0.01 -47% -48% Measured + Indicated 0.28 0.58 1.26 8.7 0.48 0.59 1.27 15.0 -0.20 -6.3 -0.01 -41% -42% Inferred 0.28 0.58 1.24 8.6 0.21 0.56 1.20 6.1 0.08 2.5 0.04 36% 41% Tuoreetsaaret Measured - - - - - - - - - - - - - Indicated 0.33 0.43 0.94 7.6 0.33 0.43 0.94 7.6 - - - - - Measured + Indicated 0.33 0.43 0.94 7.6 0.33 0.43 0.94 7.6 - - - - - Inferred 1.38 0.40 0.87 29.5 1.38 0.40 0.87 29.5 - - - - - Länttä Measured 0.33 0.59 1.27 10.4 0.16 0.56 1.21 4.7 0.17 5.7 0.06 109% 120% Indicated 0.57 0.55 1.18 16.6 0.55 0.54 1.17 15.8 0.03 0.8 0.01 5% 5%

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 171 Deposit Mineral Resource Classification Tonnes Grade Grade LCE Tonnes Grade Grade LCE Arithmetic % (Mt) (% Li) (% Li2O) (kt) (Mt) (% Li) (% Li2O) (kt) Tonnes (Mt) LCE Content (kt) (% Li2O) Mass LCE content Länttä Measured + Indicated 0.90 0.56 1.21 27.0 0.70 0.55 1.18 20.5 0.20 6.5 0.03 28% 32% Inferred 0.35 0.54 1.16 10.0 0.35 0.54 1.16 10.0 0.001 0.03 - - - Emmes Measured - - - - - - - - - - - - - Indicated 0.67 0.62 1.33 21.9 0.67 0.62 1.33 21.9 - - - - - Measured + Indicated 0.67 0.62 1.33 21.9 0.67 0.62 1.33 21.9 - - - - - Inferred 0.29 0.61 1.31 9.5 0.29 0.61 1.31 9.5 - - - - - Outovesi Measured - - - - - - - - - - - - - Indicated 0.13 0.64 1.38 4.4 0.01 0.66 1.41 0.5 0.12 4.0 -0.03 871% 848% Measured + Indicated 0.13 0.64 1.38 4.4 0.01 0.66 1.41 0.5 0.12 4.0 -0.03 871% 848% Inferred 0.12 0.67 1.44 4.3 0.02 0.67 1.45 0.7 0.10 3.6 -0.01 487% 482% Leviäkangas Measured - - - - - - - - - - - - - Indicated 0.19 0.55 1.19 5.7 0.19 0.55 1.19 5.7 - - - - - Measured + Indicated 0.19 0.55 1.19 5.7 0.19 0.55 1.19 5.7 - - - - - Inferred 0.55 0.47 1.00 13.6 0.55 0.47 1.00 13.6 - - - - - Total Measured 0.47 0.58 1.26 14.7 0.48 0.58 1.25 14.9 -0.01 -0.2 0.01 -2% -1% Indicated 2.93 0.55 1.19 86.0 3.94 0.55 1.19 116.0 -1.00 -30.0 -0.01 -25% -26% Measured + Indicated 3.41 0.56 1.20 100.7 4.42 0.56 1.20 130.9 -1.01 -30.2 -0.002 -23% -23% Inferred 4.48 0.51 1.10 121.9 3.81 0.50 1.07 100.9 0.67 21.0 0.03 18% 21% S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 172 10.19 Risks The following risks related to the MRE have been identified: • Sibanye-Stillwater have adopted forward-looking price assumptions. Any material deviations from these assumptions could impact the Mineral Resources. The assumed lithium hydroxide monohydrate prices are higher than current spot prices, implying a degree of short-term risk should these prices persist and the longer- term forecast not be realised. • The topographic layer that acts as the upper boundary for the MRE model is interpreted from drill collars and is considered adequate at this stage of the Project for resource estimation. Topographic surveys are recommended for future work and mine planning stages. • The continuity of the spodumene pegmatite at Rapasaari: it appears more structurally complex. There is greater uncertainty in the geological model of the mineralised domains and the estimation within these domains. Further infill drilling will increase confidence in the geological model. • Xenoliths within the spodumene pegmatite are internal waste; the drill density is not sufficient to accurately model the volume of these units at this stage. Further infill drilling as the development of the projects progresses, will increase the resolution of the location and extents of xenolith bodies. • Deleterious elements have not been reported as the drilling assay data for these variables are incomplete. SSW are currently undertaking a programme of re-sampling and re-assaying to improve the dataset. Future updates should include estimates for the significant deleterious elements. • Classification categories have been applied to blocks within each deposit to qualify the estimation risk based on the input data. The drilling density is used as the criteria for classification, as listed in Section 10.15, as data collected from the drilling core/chips (logged lithology and assays) are the main inputs for Mineral Resource estimation. A portion of the MRE reported for the Project is classified as Inferred Mineral Resources, which has the lowest drill density at a spacing of >30 m between drill holes. An Inferred Mineral Resource is that part of a Mineral Resource for which quantity and grade or quality are estimated on the basis of limited geological evidence and sampling. The level of geological uncertainty associated with an Inferred Mineral Resource is too high because it is informed by less drilling data to apply relevant technical and economic factors likely to influence the prospects of economic extraction in a manner useful for the evaluation of economic viability. Because an Inferred Mineral Resource has the lowest level of geological confidence of all Mineral Resources, which prevents the application of the modifying factors in a manner useful for the evaluation of economic viability, an Inferred Mineral Resources may not be considered when assessing the economic viability of a mining project and may not be converted to Mineral Reserves. In order to increase confidence and, therefore, classification of Mineral Resources, further work would be required to refine the grade continuity, geological continuity, and RPEE required for Mineral Resources. The style of mineralisation is similar between the deposits, and they are all in relatively close proximity. The continuity of the larger veins in all five of the deposits is demonstrated to be good during the geological modelling, with relatively uncomplicated morphology. Considering the continuity of the pegmatite veins, the risk is considered low. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 173 11 MINERAL RESERVE ESTIMATES [§229.601(b)(96)(iii)(B)(12) Lithium mineralisations in the region are hosted within spodumene-bearing pegmatite dyke intrusions. Keliber are considering open-pit mining in two orebodies, at Syväjärvi and Rapasaari, respectively. The spodumene-rich pegmatite vein-type orebodies are similar in nature, shallow to steeply dipping and fairly narrow and appear to have similar geotechnical characteristics. Engineering study work has been done for the proposed open pits that VBKOM considers to be at FS level of detail and accuracy. 11.1 Key Assumptions, Parameters, and Method for the Estimation of Mineral Reserves [§229.601(b)(96)(iii)(B)(12)(i)] 11.1.1 Geological Resource Models Geological resource models were received from the SSW Project team. The models received were Datamine models with the following naming convention: • Syväjärvi model name: SY_20240328_modfin • Rapasaari Model Name: MD_FIN231123 The geological models are sub-celled models with dimensions and properties as presented in Table 11-1. Table 11-1: Summarised block model properties – sub-celled geological models. Parameter X Y Z Syväjärvi Sub-Celled Minimum coordinate 2,489,380.00 7,061,470.00 -150.00 Maximum coordinate 2,491,220.00 7,063,330.00 91.25 User block size 5.00 10.00 5.00 Minimum block size 1.25 1.25 1.25 Rotation - - - Total blocks 11,399,909 Rapasaari Sub-Celled Minimum coordinate 2,491,550.00 7,059,885.00 -340.00 Maximum coordinate 2,493,200.00 7,061,725.00 102.50 Parent cell size 5.00 10.00 5.00 Minimum block size 1.25 1.25 0.25 Rotation - - - Total blocks 20,318,376 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 174 The geological models received were regularised following quantitative and qualitative assessments of the most optimal block size. The dimensions and properties for the regularised models are as presented in Table 11-2. Table 11-2: Summarised block model properties – regularised geological models. Parameter X Y Z Syväjärvi Regularised Minimum coordinate 2,489,380.00 7,061,470.00 -150.00 Maximum coordinate 2,491,220.00 7,063,330.00 92.50 User block size 5.00 5.00 2.50 Minimum block size 5.00 5.00 2.50 Rotation - - - Total blocks 12,914,151 Rapasaari Regularised Minimum coordinate 2,491,550.00 7,059,885.00 -340.00 Maximum coordinate 2,493,200.00 7,061,725.00 130.00 Parent cell size 5.00 5.00 2.50 Minimum block size 5.00 5.00 2.50 Rotation - - - Total blocks 22,830,712 11.1.2 Modifying Factors [§229.601(b)(96)(iii)(B)(12)(vi)] The ore reserves estimated for the Syväjärvi and Rapasaari deposits were calculated using the modifying factors determined during the study. Significant effort was expended by both the client and the consultants to establish and quantify the modifying factors, and it can be noted that the level of information is adequate in detail to demonstrate the limits of the deposit which is to be exploited. 11.1.2.1 Property Limits and Constraints Two key property limits are of interest for the Syväjärvi and Rapasaari study areas which include the mining lease boundaries and the pit constraint boundaries. The pit constraint boundary of each pit lies within the mining lease boundary and serves as a hard limit for the ultimate pit extents considering all planned and approved roads and infrastructure. The 2022 FS pit surface extents for both Syväjärvi and Rapasaari were used as the basis of the pit constraint perimeter string. Areas were identified where the pit areas could be expanded in comparison to the 2022 FS pit and, thus, the pit constraint perimeter string was expanded along these areas without impacting roads and infrastructure, while maintaining reasonable stand-offs from the planned and approved roads and infrastructure. Figure 11-1 and Figure 11-2 detail the approved pit constraint perimeter strings which were incorporated as a hard limit for the extents of both pits, including the distance by which the previous study pit outlines were expanded.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 175 Syväjärvi Pit Constraint Outline Figure 11-1: Syväjärvi pit constraint outline. Figure 11-2 details the approved pit constraint perimeter string for Rapasaari. Rapasaari Pit Constraint Outline Figure 11-2: Rapasaari pit constraint outline. The pit optimisation was limited to the pit constraint perimeter string, which ultimately limited the potential Mineral Reserves within the pits. 150 m 50 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 176 11.1.2.2 Mining Dilution and Losses Mining dilution occurs as ore and waste material are mixed, which typically occurs during blasting and loading practices. A different dilution estimation approach was followed compared to the 2022 FS. The FS determined dilution whereby internal dilution was added to the geological resource block model estimated Li2O value to account for the diluting between white and black waste rock (Black waste represented rock types such as porphyrite, metasediments and tuffs, whereas the waste declared as white are represented by rock types such as K-feldspar. The Black rock is separated based on colour) inside the mining block. Additionally, 10% external dilution as well as ore losses of 5% were added to each block. For this study, the geological resource block model Li2O value is representative of the in situ grade and does not include any diluting effects outside of the regularisation process. Dilution and losses were quantified by comparing the re-blocked, regularised models to the sub-celled geological models since regularised blocks capture dilution along material contacts. Re-blocking to a regularised model gives a more realistic interpretation of ore and waste mixing along the contact and the expected dilution. Different combinations of block model cell sizes were investigated to determine the impact on dilution, which included “x” and “y” dimensions of 3.75 and 5.0 m, respectively, and “z” dimensions of 1.25 and 2.5 m. A block model cell size of 5 x 5 x 2.5 m was considered practical based on historical practical observations at similar operations which practice selective mining with the envisaged loading equipment. The block model analysis revealed acceptable dilution and losses for a cell size of 5 x 5 x 2.5 m. Pre-determined ore and waste production requirements were used to determine the expected loading equipment size. The pre-determined loading equipment was then used to test various block sizes to determine the smallest mining unit (SMU). The SMU selection matrix considered equipment size practicalities and productivity considerations. The SMU selection process confirmed the regularised block model cell size of 5 x 5 x 2.5 m as practical and acceptable for the pre-determined loading equipment. The dilution consensus considers that the SMU will be mined as a whole and the average grade, ore tonnes, and content for a 5 x 5 x 2.5 m block model cell will be representative of the losses and dilution that will be measured even with the efforts of selective mining along contacts. Dilution was applied on a zero-grade basis for rock types other than spodumene and muscovite pegmatites, which are rock codes 100 and 200, respectively. The global dilution and losses after re-blocking are shown below: • Syväjärvi − Losses: -14.97% − Dilution: 11.08% • Rapasaari − Losses: -15.67% − Dilution: 11.66% The histograms in Figure 11-3 depict the movement in tonnes and grade when the in situ and diluted tonnes are compared. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 177 Global Resource Dilution at Syväjärvi (Left) and Rapasaari (Right) Figure 11-3: Histogram showing the global resource dilution at Syväjärvi (left) and Rapasaari (right). The rock codes shown are defined as per Table 11-3. Table 11-3: Rock codes. Rock Code Description 100 Spodumene Pegmatite 200 Muscovite Pegmatite 300 Country Rock (within pegmatite) 400 Overburden 500 Metasediment 600 Metavolcanite 700 Plagioclase Porphyrite Sill 800 Sulphidic Schist 11.1.2.3 Processing Recoveries 11.1.2.3.1 Concentrate Recovery The following block models were used for to determine the concentrate recoveries for the two mining areas: • Syväjärvi: SYMD_FIN231123 • Rapasaari: MD_FIN231123 Recovery data were used to compile simplified recovery curves for the Li2O concentrate, using the “Feed grade” vs “Corrected Li2O Recovery % in full scale - Final value”. A comparison between the developed algorithms (from testwork) and the simplified recovery curves was done by the Keliber team and approval on the methodology was given. Figure 11-4 shows the recovery data for the Syväjärvi open pit, and a linear trendline was generated with an R2 of 0.96. The recovery equation from the trendline is presented in Equation 10. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 178 Equation 10: Recovery from the trendline for the Syväjärvi open pit. 𝑅𝑒𝑐𝑜𝑣𝑒𝑟𝑦 𝑆𝑦𝑣ä𝑗ä𝑟𝑣𝑖 = 7.7926𝑥 + 80.2 Where: • x refers to the feed Li2O grade. Syväjärvi Recovery Curve Figure 11-4: Syväjärvi recovery curve. Figure 11-5 shows the recovery data for the Rapasaari open pit, and a polynomial trendline was generated with an R2 of 0.998. The recovery equation from the trendline is presented in Equation 11. Equation 11: Recovery from the trendline for the Rapasaari open pit. 𝑅𝑒𝑐𝑜𝑣𝑒𝑟𝑦 𝑅𝑎𝑝𝑎𝑠𝑎𝑎𝑟𝑖 = −9.2526𝑥2 + 26.592𝑥 + 70.402 Where: • x refers to the feed Li2O grade.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 179 Rapasaari Recovery Curve Figure 11-5: Rapasaari recovery curve. 11.1.2.3.2 Phase Conversion Recoveries The concentrate with targeted grade is further processed in a chemical plant to produce LiOH.H2O. The process starts with the conversion of the spodumene concentrate from alpha to beta spodumene by roasting the concentrate in a rotary kiln. A constant conversion degree of 97% was provided by the client in the testwork datasheets. Note that after the completion of the production schedule, the recovery was adjusted to 100% in the financial evaluation. This was confirmed with the metallurgist as the 97% was already accounted for in the 86% recovery of the Keliber lithium refinery mentioned below. 11.1.2.3.3 Lithium Hydroxide Production The converted concentrate (beta spodumene concentrate) is fed to the hydro process to produce lithium hydroxide (refer to Equation 12). Equation 12: Lithium hydroxide production. Li2O + 3𝐻2O = 2LiOH. 𝐻2O A constant lithium yield of 86% is applied as proposed in the datasheets. The final mass recovery of lithium hydroxide is presented in Equation 13. Equation 13: Final mass recovery of lithium hydroxide. 𝐿𝑖𝑂𝐻. 𝐻2𝑂 (𝑡𝑜𝑛𝑛𝑒𝑠) = % 𝐿𝑖2𝑂 × 𝑅𝑂𝑀 𝑡𝑜𝑛𝑛𝑒𝑠 × 𝐶𝑜𝑛𝑐 𝑟𝑒𝑐𝑜𝑣𝑒𝑟𝑦 × 0.97 × 0.86 × 2.8087 Where: • % Li2O refers to the ROM diluted feed grade; • Conc recovery refers to value from either Equation 12 or Equation 13; S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 180 • 0.97 refers to the phase conversion recovery (Note that after the completion of the production schedule, the recovery was adjusted to 100% in the financial evaluation after the Client confirmed that the 97% conversion factor was already accounted for in the 86% recovery factor); • 0.86 refers to the lithium hydroxide recovery; and • 2.8087 is the conversion factor from Li2O to LiOH.H2O using the following molar mass: − Molar mass (MM) (Li2O) = 29.8814 − MM (LiOH.H2O) = 41.96362 x 2 = 83.92724 11.1.3 Li2O Cut-Off Grade [§229.601(b)(96)(iii)(B)(12)(iii)] Re-blocking was based on a dominant mass approach (>50% by mass) whereby blocks were classified as either ore or waste depending on the dominant mass of the block. Reclassification of ore and waste was based on a marginal lithium cut-off value which was calculated based on plant throughput: • The diluted feed grade from the block model was used as an input. • The recovery curve (1.06% test work grade) was used. • The recovery curve was used to adjust variable mining costs used in the COG calculations. • The marginal cut-off was calculated on all processing and selling costs (excluding mining variable and fixed costs). • The COG was correlated to plant capacity, hence, the COG was adjusted to achieve 15,000 lithium hydroxide tonnes. The marginal COG formula is presented in Equation 14. Equation 14: Marginal COG. 𝑀𝑂 𝐶𝑢𝑡 − 𝑜𝑓𝑓 𝐺𝑟𝑎𝑑𝑒 = (𝐶𝑝 + 𝐶𝑟 + 𝐶𝑚𝑐) ÷ (𝑚 x 𝑟 x (𝑃 − 𝐶𝑠)) Where: • Cp is the total metallurgical processing cost (Fixed and Variable) in cost/unit treated (processed) [EUR/t t]; • Cr is the cost of rehandle ore from surface stockpiles to the metallurgical plant in cost/unit treated [EUR/t]; • Cmc is mine closure cost incurred during the LOM in cost/unit treated [EUR/t]; • P is the metal price in revenue/unit [EUR/t]; • Cs is the cost of selling the metal (smelting, refining, royalties, concentrate and metal transport, management fees, off-mine cost, head office cost, marketing, etc.) in cost/unit [EUR/g]; • m is the mine call factor (normally use 100%) or other modifying factors that will impact the COG; and • r is the total metallurgical recovery (%). The costs utilised for calculating the marginal cut-off are presented in Table 11-4. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 181 Table 11-4: Cut-off calculation parameters. Cut-Off Calculation Parameters Units Syväjärvi Rapasaari Metallurgical variable cost EUR/t ROM 14.43 14.23 Metallurgical fixed cost EUR/t ROM 3.27 3.27 Rehandle cost EUR/t ROM 1.24 1.11 Metal price EUR/t product 18,182 18,182 Transport EUR/t Conc 8.86 8.86 Refinery ops EUR/t Conc 368.29 368.29 General and Administrative (G&A) EUR/a 9,181,900 9,181,900 Freight EUR/t product 93.23 93.23 Royalty EUR/t ROM 0.67 0.67 The summary of the COGs and key calculated parameters are presented in Table 11-5. Table 11-5: COG per pit used in optimisation. Parameter Units Syväjärvi Rapasaari Marginal COG % 0.188 0.185 Breakeven COG % 0.206 0.212 Tailings grade % 0.190 0.190 COG required for 15 ktpa LiOH.H2O % 0.100 0.300 Concentrator feed COG % 0.206 0.300 Ore above concentrator feed COG Mt 4.01 15.06 Marginal ore tonnes (Conc feed cut-off – MO cut-off) Mt 0.49 4.79 Material containing Li2O discarded (<MO FCOG) Mt 1.10 13.19 COG used as Pseudoflow input parameter % 0.19 0.30 The COG for the Mineral Resources is 0.5% Li2O; the lower marginal COG calculated in Table 11-5 is a result of dilution and reallocated blocks. The Syväjärvi marginal COG is 0.19% Li2O based on financial parameters, whereas the Rapasaari marginal COG of 0.30% Li2O was limited to the concentrator feed to produce 15 kt of product. 11.2 Open-Pit Optimisation Open-pit optimisation was used to evaluate the economic open-pit sizes for the Mineral Reserves. The pit optimisation analysis was carried out for the two orebodies at Syväjärvi and Rapasaari using Deswik Pseudoflow software to evaluate the economic open-pit sizes on all Mineral Resource classifications. The optimisation method was applied to the regularised mining block models. The Pseudoflow algorithm calculates a series of incremental pit shells given input parameters such as cost, recovery, revenue, pit slope parameters, and other physical constraints. The software calculates the discounted cumulative cash flow indicating the net present value (NPV) of the open pit by using the Direct Block Scheduling algorithm. Each block is analysed individually in the algorithm to define the best target period. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 182 Within each shell, there is an optimum value for an incremental higher commodity price factor. The optimisation pit shell results and selected optimal ultimate pit shells were used to identify the pit limits that guided the practical pit design process for each of the mining areas. 11.2.1 Open-Pit Optimisation Parameters The optimisation parameters include the Mineral Resource estimation block models, all necessary operational costs, time costs, and selling and processing costs of the final concentrate. The input factors used in the optimisation process include: • Overall slope angles (OSAs); • Mining costs including variation by mining bench height or mining cost adjustment factor (MCAF); • Mineral processing costs; • Mineral processing recoveries; • Logistical cost; • Product revenues; and • Discount rate. An additional geotechnical study was performed by Geotec Africa CC to re-evaluate the most suitable open-pit OSAs, bench face angles (BFAs), and geotechnical design parameters. The updated geotechnical parameters utilised for the pit optimisation study for Syväjärvi are shown in Figure 11-6.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 183 Geotechnical OSAs and BFAs per Sector for Syväjärvi Figure 11-6: Geotechnical slope sectors for Syväjärvi (source: Geotec Africa CC). The updated geotechnical parameters utilised for the pit optimisation study for Rapasaari are shown in Figure 11-7. 40 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 184 Geotechnical OSAs and BFAs per Sector for Rapasaari Figure 11-7: Geotechnical slope sectors for Rapasaari (source: Geotec Africa CC). The pit optimisation parameters were adjusted/inflated according to the data provided by Keliber. The summary of the optimisation parameters for Syväjärvi and Rapasaari is presented in Table 11-6, Table 11-7, Table 11-8, and Table 11-9. The financial parameters are highlighted in Table 11-6. The concentrating plant recoveries were applied according to the recovery curves mentioned in Section 11.1.2.3.1 and the Keliber lithium refinery recoveries as discussed in Section 11.1.2.3.2 and Section 11.1.2.3.3. Table 11-6: Pit optimisation input parameters: financial parameters. Financial Parameter Unit Syväjärvi & Rapasaari Value Comment Project base date Date 01 January 2025 Mining costs base date Date 05 October 2023 Processing & selling costs base date Date 01 January 2024 Euro escalation rate % per year 4.0%, 0.9%, 1.4% 2023, 2024, 2025 Exchange rate EUR:USD 1.10 2024 Exchange rate EUR:ZAR 20.00 2024 Discount Rate % 8.00 150 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 185 Financial Parameter Unit Syväjärvi & Rapasaari Value Comment LiOH.H2O Price USD/t ddp 20,000 2025–2028 & Long term Royalties EUR/t Ore 0.66 The mining costs (shown in Table 11-7) are sourced from mining contractor quotes, received in a 2023 mining contractor trade-off study, and inflated to 2024 according to the escalation factors provided by the Keliber team. Table 11-7: Pit optimisation input parameters: mining cost. Mining Costs Unit Syväjärvi Escalated Value Rapasaari Escalated Value Comment Topsoil, clear and grub EUR/m3 1.14 0.94 Load & Haul overburden EUR/m3 1.14 0.94 Load & Haul waste EUR/t 0.95 1.02 Load & Haul sulphidic waste EUR/t 0.94 0.99 Load & Haul ROM ore from pit to Syväjärvi temporary stockpile EUR/t 0.73 - Load & Haul ROM ore EUR/t - 1.45 Drilling & Blasting waste EUR/t 0.99 0.77 Drilling & Blasting ROM ore EUR/t 1.22 1.16 Presplit drilling and blasting EUR/m2 15.19 13.97 Presplit drilling and blasting – per tonne EUR/t 0.04 0.02 VBKOM conversion Surface reference elevation masl 88.40 86.17 Mining vertical cost adjustment factor ZAR/t/10 m 0.73 0.73 Ore rehandling cost from ROM pad stockpiles EUR/t - 1.11 Ore rehandling cost from Syväjärvi stockpile to Crusher EUR/t 1.24 - Dewatering cost EUR/a 160,101 197,475 EUR/t Tot 0.06 0.03 Time costs (Fixed costs) EUR/a 971,234 916,899 EUR/t Tot 0.38 0.16 Auxiliary equipment EUR/t Tot 0.77 0.64 Site establishment EUR 1,528,413 1,528,413 Fleet management, technology, safety, and training EUR/a 295,493 EUR/t Tot 0.04 0.04 Grade control cost EUR/t ROM 0.13 0.20 The processing costs (shown in Table 11-8) were sourced from site personnel and inflated to 2024. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 186 Table 11-8: Pit optimisation input parameters: processing cost. Processing Costs Unit Syväjärvi & Rapasaari Escalated Value Comment Ore sorting variable costs EUR/t ROM incl. below Concentrator variable costs EUR/t ROM 14.23 Concentrator fixed costs EUR/a 2,778,584 EUR/t Tot 0.40 Ore sorter efficiency % 73 Concentrate grade % 4.5 Concentrator recovery % - Recovery curves for Syväjärvi & Rapasaari Crusher and sorting capacities tpa (nominal) 850,000 Grinding / milling capacity tpa (nominal) 750,000 Concentrate tonnes per annum tpa (nominal) 192,340 Final product (LiOH·H2O) tonnes produced tpa (nominal) 15,000 The selling cost (shown in Table 11-9) includes the Keliber lithium refinery and is inflated to 2024. Table 11-9: Pit optimisation input parameters: selling costs. Selling Costs Unit Syväjärvi & Rapasaari Escalated Value Comment Road transport cost and duties EUR/t Conc. 8.86 Refinery processing costs EUR/t Conc. 368.08 G&A EUR/a 9,180,000 EUR/t Tot 1.32 Freight cost EUR/t Prod. 93.23 Conversion degree % 97 Updated to 100% in financial evaluation Hydro Li₂O yield % 86 11.2.1.1 Optimisation Results – Syväjärvi Evaluation The 2024 optimisation considered constrained and unconstrained runs, whereby the constrained runs were limited to the footprint of the 2022 FS pit by means of an approved footprint string (“constrained perimeter string 2024”) which outlined the 2022 FS pit and was expanded, where possible, without impacting planned and approved roads and infrastructure, refer to Figure 11-1. The pit selection criteria for Syväjärvi were based on maximising discounted cash flow, thus the highest value pit shell was selected up to revenue factor (RF) 1.0. The maximum discounted pit value for the constrained runs was achieved for an RF of 1.19, however, RF 1.0 was selected as the optimal pit shell. The RF 0.53 to 1.92 (equal to RF 2.0 shell) pit shells had a value within 1% of the maximum pit value RF 1.19 shell due to the footprint constraint which restricted the shell sizes, hence shells beyond a revenue factor of 1.0 did not notably increase in size, as depicted in Figure 11-8. As a result of this, the RF 1.0 shell was selected for the study.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 187 Pit-by-Pit Graph for the Syväjärvi Optimisation Figure 11-8: Pit-by-pit graph for the Syväjärvi optimisation. The comparison of the previous 2022 FS pit design and the 2024 selected constrained RF 1.0 optimisation pit shell are included in Figure 11-9. In comparison, the 2024 shell increased in size and, predominately, depth of the pit, mainly due to the expanded footprint string, updated geotechnical parameters, economics, and ore classification or COG methodology as well as the dilution approach. Comparison between the 2022 FS Pit Design and the Selected 2024 Pit Shell Figure 11-9: Comparison between the 2023 FS pit design and the selected 2024 pit shell. 50 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 188 The comparison of results between the 2022 FS pit design and the 2024 pit optimisation is included in Table 11-10, based on the mining block model from the 2024 study for both. The 2024 optimisation shell shows an increase in ROM ore tonnes, increasing from 2.73 Mt in 2022 to 3.28 Mt in 2024, indicating an improved ore extraction potential and LOM. However, this increase in ore production is accompanied by an increase in waste tonnes, from 12.40 Mt to 17.31 Mt, resulting in a higher stripping ratio of 5.3 compared to 4.5 in the 2022 FS pit design. Additionally, the COG has been reduced significantly from 0.40% Li₂O to 0.188% Li₂O. Table 11-10: Optimisation analysis comparison for Syväjärvi. Optimisation Year ROM Tonnes Waste Tonnes ROM Ore Grade Syväjärvi Mt Mt Li2O % 2022 Pit Design 2.73 12.40 1.056 2024 Optimisation Shell 3.28 17.31 1.015 11.2.1.2 Optimisation Results – Rapasaari Evaluation The 2024 optimisation for Rapasaari considered constrained and unconstrained runs similar to Syväjärvi, whereby the constrained runs were limited to the footprint of the previous 2022 FS pit by means of an approved footprint string (“constrained perimeter string 2024”) which outlined the 2022 FS pit and was expanded, where possible, without impacting on planned and approved roads and infrastructure, refer to Figure 11-2. The pit selection criteria for Rapasaari were based on maximising cash flow, thus the highest discounted value pit shell was selected for the constrained shells. The maximum discounted pit value for the constrained runs was achieved for an RF of 0.68. The RF 0.49 to 2.0 pit shells had a value within 1% of the maximum discounted pit value RF 0.68 shell due to the footprint constraint which restricted the shell sizes, hence shells beyond an RF of 0.68 did not notably increase in size, as depicted by the pit-by-pit graph in Figure 11-10. As a result of this, the RF 0.68 shell was selected for the study. Pit-by-Pit Graph for the Rapasaari Optimisation Figure 11-10: Pit-by-pit graph for the Rapasaari optimisation. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 189 The comparison between the 2022 FS pit design and the 2024 selected constrained RF 0.68 optimisation pit shell is shown in Figure 11-11. The 2024 pit shell is deeper, primarily due to updated geotechnical parameters and changes in the ore classification and COG methodology. However, the footprint of the 2024 pit shell is not larger than that of the 2022 design. Comparison between the 2022 FS Pit Design and the Selected 2024 Pit Shell Figure 11-11: Comparison between the 2022 FS pit design and the selected 2024 pit shell. The results comparison between the 2022 FS and the 2024 pit optimisation is included in Table 11-11, based on the mining block model from the 2024 study. The 2024 optimisation shell for Rapasaari shows an increase in ROM ore tonnes, increasing from 9.05 Mt in 2022 to 12.09 Mt in 2024, indicating enhanced ore extraction potential and an extended LOM. While waste tonnes have increased from 64.05 Mt to 82.42 Mt, the stripping ratio has slightly decreased from 7.1 in the 2022 FS pit design to 6.8 in the 2024 pit optimisation shell. The ROM ore grade remains relatively stable, with a slight decrease from 0.938% Li₂O in 2022 to 0.936% Li₂O in 2024. The COG has been lowered from 0.40% Li₂O to 0.30% Li₂O. Table 11-11: Optimisation analysis comparison for Rapasaari. Optimisation Year ROM Tonnes Waste Tonnes ROM Ore Grade Rapasaari Mt Mt Li2O % 2022 Pit Design 9.05 64.05 0.938 2024 Optimisation Shell 12.09 82.42 0.936 11.3 Open-Pit Design This section summarises the assumptions and methodology used in the Syväjärvi and Rapasaari open-pit design process. The open pits were designed based on the open-pit optimisation results and geotechnical guidance as shown in Figure 11-13, Figure 11-14, and Figure 11-15. 200 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 190 11.3.1 Mine Design Criteria Pit design criteria and industry best practice parameters were used in the open-pit design process to create viable and practical pit designs for the study. The parameters and guidelines were provided by Lofty Mining (Pty) and Geotec Africa CC after significant test work and analysis were completed in July 2024 for Syväjärvi and August 2024 for Rapasaari. The information for this section has been sourced from: • Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study for the Rapasaari Open Pit (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. Edited by Abraham Saayman, GEOTEC Africa CC. • Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study of the Syväjärvi Open Pit Mine (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. Edited by Abraham Saayman, GEOTEC Africa CC. The pit designs for the study were based on and guided by the selected pit optimisation shells for Syväjärvi and Rapasaari, respectively. The rule-of-thumb guideline is that the designed pits should not vary by more than 10% from the Pseudoflow shell in their vital parameters such as ore- and waste tonnes, metal content, and strip ratio. 11.3.1.1 Factors Considered During Pit Design Phase In generating the pit designs, the following parameters and restrictions were considered along with the specifics of the selected Pseudoflow shells and pushbacks. 11.3.1.1.1 Pit Perimeters The current EPs restrict the pit perimeters (footprint) to those of the pits designed and adhered to during the ultimate pits design phase, ensuring conformance with the EP restrictions (Figure 11-1 and Figure 11-2). 11.3.1.1.2 Ramp and Geotechnical Design Parameters Safety-related aspects of in-pit ramp widths were incorporated into the pit design process. Standard ramp designs were implemented not only in the pits but also in the waste and overburden (OVB) dump areas. A central berm was excluded for the purpose of clearing snowfall. Ramp width requirements were determined based on a 64-tonne class rear dump haul truck (CAT775 model) to guide the ramp width design. The roadway width was calculated as 4.0 times the overall truck width of 5.67 metres for the standard dual ramp and 2.0 times the overall truck width for a single- lane ramp. For the standard dual ramp, a berm along the edge will be constructed using waste rock material. The berms were designed to have a height at least equal to the radius of the truck wheel. A secondary single-lane ramp with a width of 16 metres has been designed at both the Syväjärvi and Rapasaari pits to ensure safety and provide emergency access, extending at least two-thirds of the pit's depth. Single-lane ramps were utilised for the final two benches at the bottom of the pit designs which allows for the retrieval of extra ore at the base of the pits. With the exception of the secondary ramp at Syväjärvi, all ramp daylight positions were expected to remain close to their locations in the previous 2022 FS pit designs, staying within 50 metres, in the case of Rapasaari. The gradient for all ramps, including those on the sides of waste dumps, was designed to be 1:10. The ramp width design parameters for Syväjärvi and Rapasaari are indicated in Table 11-12.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 191 Table 11-12: Ramp width design parameters. Ramp Truck Payload [tonne] Truck Width [m] No. of Truck Widths Roadwa y Width [m] Tyre Diamete r [m] Berm Height [½ Tyre Ø] Berm Base [38° slope] Highwal l Off-Set [m] Edge Off-Set [m] Ramp Design Width [m] Parameter - - - B - C D A E F Standard dual direction without central berm 64 (Cat775 ) 5.67 4.0 22.7 2.1 1.1 2.70 1.0 1.5 28.0 Single lane 64 (Cat775 ) 5.67 2.0 11.3 2.1 1.1 2.70 1.0 1.5 16.0 Figure 11-12 represents a schematic illustration of the standard dual ramp and single-lane ramp width parameters. Standard Dual- and Single Ramp Parameters Road Width Design Requirements for Standard Dual Lane and Single-Lane Ramp Systems Figure 11-12: Standard dual- and single-lane ramp configurations. The open-pit slope sectors and parameters for Syväjärvi are presented in Table 11-13. The slope sectors referred to in the table are depicted in the pit optimisation input parameters and utilised in the optimisation process (refer to Figure 11-6). Table 11-13: Recommended parameters for open-pit designs – Syväjärvi. Syväjärvi Geotechnical Parameters Slope Sector Max. Bench Height Min. Berm Width Max. Inter Stack Angle Max. BFA Ramp 1 Width Ramp 2 Width Rock OSA Over- burden OSA [No.] [m] [m] [°] [°] [m] [m] [°] [°] 1A 20 8.5 48 65 16 - 43 18.5 2A 20 8.5 59 80 28 - 49 18.5 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 192 Syväjärvi Geotechnical Parameters Slope Sector Max. Bench Height Min. Berm Width Max. Inter Stack Angle Max. BFA Ramp 1 Width Ramp 2 Width Rock OSA Over- burden OSA [No.] [m] [m] [°] [°] [m] [m] [°] [°] 2B 20 8.5 59 80 28 28 43 18.5 2C 20 8.5 59 80 28 28 43 18.5 3A 20 8.5 55 75 16 - 48 18.5 3B 20 8.5 55 75 28 28 30 18.5 Notes: 1. Based on upgraded kinematic and wedge stability analyses with selected, upgraded, geotechnical data suites. 2. Inter-stack geotechnical berms (width = 16 m) must be introduced if the inter-ramp stack height(s) increase beyond 80 m + 20 m, thus >100 m. Not applicable if ramp divides a stack. The open-pit slope sectors (Figure 11-7) and slope parameters for Rapasaari utilised in the pit optimisation process are presented in Table 11-14. Table 11-14: Recommended parameters for open-pit designs – Rapasaari. Rapasaari Geotechnical Parameters Slope Sector Max. Bench Height Min. Berm Width Max. Inter Stack Angle Max. BFA Ramp 1 Width Ramp 2 Width Rock OSA Over- burden OSA [No.] [m] [m] [°] [°] [m] [m] [°] [°] 1A 20 10 49 70 32 20 37 18.5 1B 20 10 49 70 32 20 41 18.5 1C 20 10 52 75 28 20 43 18.5 1D 20 10 52 75 28 - 44 18.5 1E 20 10 56 80 28 - 46 18.5 2A 20 10 56 80 28 - 50 18.5 2B 20 10 56 80 - - 50 18.5 3A 20 8.5 59 80 20 - 48 18.5 3B 20 8.5 59 80 20 - 48 18.5 Notes: 1. Based on upgraded kinematic and wedge stability analyses with selected, upgraded, geotechnical data suites. 2. Inter-stack geotechnical berms (width = 16 m) must be introduced if the inter-ramp stack height(s) increase beyond 80 m + 20 m, thus >100 m. Not applicable if ramp divides a stack. 11.3.1.1.3 Minimum Mining Width The minimum mining width to be applied for the bench design was 40 metres, to ensure the operations can be conducted effectively, productively, and safely. This width is determined based on the turning circle clearance diameter of the CAT 775G rear dump truck, combined with the length of the CAT 6015 excavator equipped with its S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 193 boom, stick, and bucket. This minimum width requirement was applied to all benches, except where a bench pinches out at a pit corner. 11.3.1.1.4 Pit-Room Required Pit room is the area in the pit that must be available to carry out the mining process i.e. bench preparation, blast hole drilling, charging up and blast, loading, dewatering, and road construction. During the pit design, ramp access to all benches was ensured, considering the mining processes that could necessitate multiple access points to each bench over its lifespan. 11.3.1.1.5 Surface Restrictions Cognisance was taken of the existing and planned surface infrastructure around the pits in addition to the surface footprints described in 11.3.1.1.1 above. 11.3.1.1.6 Surface Topography The surface topography (LiDAR) utilised for this Project is dated 2016, and the designed pit surface was adjusted to match the topography. The borrow pit currently being mined for construction aggregate at Syväjärvi was depleted from the surface topography, dated 12 August 2024. 11.3.2 Syväjärvi The selected constrained RF 1.0 shell from the optimisation study was utilised as a guideline in the pit design process for Syväjärvi. A single large pit was designed, with no pushback designs necessary due to the constrained shell's limited footprint. With the orebody being shallowest on the southern side of the pit and consisting of relatively thick lenses dipping approximately 18° northward, the pit access design strategy prioritised targeting the southern side of the pit to enable early access to the ore and to ensure up-dip extraction of the ore. The positioning and switchback of the primary ramp along the south and west highwalls were designed to maximise the steepness of the east and north highwalls, optimising ore extraction at the lower benches of the pit. Secondary access to the Syväjärvi pit was provided on the eastern side of the pit through a 16-m wide single-lane ramp, extending to at least two-thirds of the pit's depth (refer to Figure 11-13). Geological discontinuities (FLTs) were identified in the southwestern area of the pit by the geotechnical engineer. The ramp position and layout were designed to ensure the FLT can be safely navigated. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 194 Plan view Isometric view Open-Pit Mine Layout – Syväjärvi Figure 11-13: Syväjärvi open-pit design. The diluted tonnes within the Syväjärvi LOM pit design are detailed in Table 11-15, with a total of 3.15 Mt of ore and a strip ratio of 5.14. Table 11-15: Syväjärvi LOM pit design diluted tonnes. Syväjärvi Pit Design Diluted Tonnes Classification Tonnes (Mt) Li₂O (%) Ore 3.15 1.03 Waste 16.17 0.00 Total 19.32 0.17 Strip Ratio (t:t) 5.14 - 11.3.3 Rapasaari The selected constrained RF 0.68 shell from the optimisation study was used as a guideline in the pit design process for Rapasaari. The final pit design was developed alongside three pushback designs. For each of the three pushbacks, two dual-access ramps were designed which align with the final ramps of the ultimate pit design. The orebody is composed of steep dipping ore lenses, with a denser concentration of lenses in the northern part of the pit and a less dense distribution towards the southern and western extremities. The selection of pushback areas was guided by the ore-to-waste ratios across the entire pit to ensure a fairly consistent stripping ratio for each pushback. The ramp positions align with the 2022 pit design, as instructed by the client. The dual-access ramps were designed for a width of 28 m to enable two-way traffic, and single ramps were designed for a width of 16 m for one-way traffic (refer to 200 m

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 195 Figure 11-14). The pit was designed according to the OSAs specified for each geotechnical domain for a stack height of 80 m. The OVB removal angle was set at 18°. The satellite pit, located at the northwestern end of the pit, is serviced by a single access ramp and is planned to be mined at a slower rate and as part of pushback 2. The primary access ramps, situated at the central western and northern ends of the pit, will be used during the mining of Pushback 1, which has a limited depth to balance the stripping ratios. As the mining progresses, the primary access ramps at the central western, northern, and eastern ends, along with the satellite pit’s northwestern access, will all be utilised to provide access to the active mining areas while mining Pushback 2, depending on the extraction sequences and varying mining rates across the pushbacks. The central area of pushback two has only one dual-access ramp (eastern primary access ramp) during mining for a limited time period, after which two dual ramps are re-established for mining the remainder of pushback two. This lapse in dual access will be optimised in future studies. Finally, the primary access ramp at the southern end of the pit, together with the eastern primary ramp, is planned to provide access during the final pushback. Sufficient ramp accesses have been planned since the pushbacks will be mined simultaneously to balance the ore requirements considering the varying concentrations of ore lenses across the pit (refer to Figure 11-15). Plan view Isometric view Open-Pit Mine Layout – Rapasaari Figure 11-14: Rapasaari ultimate open-pit design. The three pushbacks are shown in Figure 11-15. 500 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 196 Open-Pit Pushback Layouts – Rapasaari Figure 11-15: Rapasaari pushback design. Table 11-16 below presents the diluted tonnages, grade, and stripping ratio for each pushback. Pushback three has the highest stripping ratio but the lowest ore grade, while pushback one has the highest ore grade and pushback two has the lowest stripping ratio. These pushbacks have been strategically planned to ensure a consistent stripping ratio and ROM ore profile throughout the LOM. To meet the required feed grade, pushbacks two and three will be mined simultaneously for blending purposes. Table 11-16: Rapasaari LOM pit design diluted tonnes. Rapasaari Pit Design Diluted Tonnes Classification Tonnes (Mt) Li₂O (%) Pushback 1 Ore 3,665,705 0.83 Waste 28,592,925 0.02 Subtotal 32,258,630 0.11 Strip Ratio (t:t) 7.80 - Pushback 2 Ore 5,479,261 0.82 Waste 35,422,364 0.03 Subtotal 40,901,625 0.14 Strip Ratio (t:t) 6.46 - Pushback 3 Ore 2,134,858 0.71 Waste 22,092,552 0.02 Subtotal 24,227,410 0.08 500 m 500 m 500 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 197 Rapasaari Pit Design Diluted Tonnes Classification Tonnes (Mt) Li₂O (%) Strip Ratio (t:t) 10.35 - Total Ore 11,279,823 0.93 Total Waste 86,107,841 0.02 Grand Total 97,387,664 0.11 Strip Ratio (t:t) 7.63 - 11.4 Mineral Reserve Classification This section contains forward-looking information related to the Mineral Reserve classification for the Keliber Lithium Project. For estimating the Mineral Reserves for Keliber, the following definition, as set forth in the S-K 1300 Definition Standards, was applied. “A Mineral Reserve is an estimate of tonnage and grade or quality of indicated and measured mineral resources that, in the opinion of the qualified person, can be the basis of an economically viable project. More specifically, it is the economically mineable part of a measured or indicated mineral resource, which includes diluting materials and allowances for losses that may occur when the material is mined or extracted.” Mineral Reserves are subdivided into classes of Proven Mineral Reserves and Probable Mineral Reserves, which correspond to Measured Mineral Resources and Indicated Mineral Resources, respectively, with the level of confidence reducing with each class. Mineral Reserves are always reported as the economically mineable portion of a Measured and/or Indicated Mineral Resources and take into consideration the mining, processing, metallurgical, economic, marketing, legal, environmental, infrastructure, social, and governmental factors (the “Modifying Factors”) that may be applicable to the deposit. For the purpose of this document only, Measured Mineral Resources were converted to Proven Mineral Reserves and all Indicated Mineral Resources were converted to Probable Mineral Reserves. 11.5 Mineral Reserve Estimates [§229.601(b)(96)(iii)(B)(12)(ii)] This section summarises the process followed in converting Mineral Resources to Mineral Reserves for Syväjärvi and Rapasaari for the Keliber 2024 study. 11.5.1 Syväjärvi Mineral Resource to Mineral Reserve Process 11.5.1.1 Syväjärvi In situ Mineral Resources Figure 11-16 illustrates plan view and cross-sectional representations of the Syväjärvi Mineral Resources, categorised by Mineral Resource classification within the Syväjärvi mine boundary. The cross-sectional side view of the Syväjärvi Mineral Resources, viewed from the west, highlights the vertical and lateral distribution of the Mineral Resources beneath the surface, emphasising the spatial relationship between the classified Mineral Resources and the surface topography. The Measured Mineral Resources are distributed throughout the extent of the orebody. The orebody at Syväjärvi extends approximately 750 metres in length and 350 metres in width, with a maximum depth below the S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 198 surface of nearly 180 metres. The In situ Mineral Resources were reported as per the resource block model and the associated ore definitions. Plan view Sideview from West Plan and Side Views of the In situ Mineral Resources, Categorised by Mineral Resource Classification, shown in Relation to the Mine Boundary and Surface Topography – Syväjärvi Figure 11-16: Syväjärvi in situ Mineral Resources. The Syväjärvi in situ Mineral Resources comprise a total of 2.95 Mt with an average grade of 1.35% Li2O. The table indicates that most of the material is classified as Measured Mineral Resources (57%), followed by Indicated Mineral Resources (31%) and Inferred Mineral Resources (12%). Detail of the in situ Mineral Resources categorised by Mineral Resource classification at Syväjärvi is provided in Table 11-17. The in situ Mineral Resources were reported as per the resource block model and the associated ore definitions. 500 m

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 199 Table 11-17: In situ Mineral Resources categorised by Mineral Resource classification at Syväjärvi. Syväjärvi In situ Mineral Resources Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Measured Mineral Resources 1.69 0.65 1.40 58 Indicated Mineral Resources 0.90 0.60 1.29 29 Inferred Mineral Resources 0.36 0.58 1.24 11 Total 2.95 0.63 1.35 98 Note: In situ Mineral Resources include Geological loss. 11.5.1.2 Syväjärvi In situ Mineral Resources Exclusive and Inclusive of Mine Plan The in situ Mineral Resources, both exclusive and inclusive of the Syväjärvi LOM pit design, are shown in Figure 11-17. The figure highlights that the majority of the Mineral Resources are located within the LOM pit design. However, some Mineral Resources are situated at deeper elevations or beyond the pit boundaries. A portion of these Mineral Resources could potentially be extracted through open-pit mining if the current surface-constrained pit limits were removed. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 200 Isometric view Plan view In situ Mineral Resources Inclusive and Exclusive of LOM Pit Design – Syväjärvi Figure 11-17: Syväjärvi in situ Mineral Resources inclusive and exclusive of LOM plan. 250 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 201 Table 11-18 summarises the in situ Mineral Resources both inside and outside the Syväjärvi LOM plan, categorised by Mineral Resource classification. The total in situ Mineral Resources equal 2.95 Mt with an average grade of 1.35% Li2O. Of this total, 2.37 Mt (80%) are within the LOM plan with an average grade of 1.37% Li2O, predominantly consisting of Measured Mineral Resources and Indicated Mineral Resources. Only 0.13 Mt (6%) of the material within the LOM plan is classified as Inferred Mineral Resources. Outside the LOM plan, 0.58 Mt of Mineral Resources remain, averaging 1.25% Li2O, with the majority classified as Indicated Mineral Resources and Inferred Mineral Resources. Table 11-18: Syväjärvi in situ Mineral Resources exclusive and inclusive of mine plan. Syväjärvi Mineral Resources Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) In situ Mineral Resources Outside Mine Plan Measured Mineral Resources 0.11 0.56 1.20 3 Indicated Mineral Resources 0.24 0.59 1.28 8 Inferred Mineral Resources 0.23 0.57 1.23 7 Subtotal 0.58 0.58 1.25 18 In situ Mineral Resources Inside Mine Plan Measured Mineral Resources 1.58 0.66 1.41 55 Indicated Mineral Resources 0.66 0.60 1.30 21 Inferred Mineral Resources 0.13 0.58 1.24 4 Subtotal 2.37 0.64 1.37 80 Grand Total 2.95 0.63 1.35 98 Note: In situ Mineral Resources include Geological loss. 11.5.1.3 Syväjärvi Diluted Mineral Resources in Mine Plan The process of converting in situ Mineral Resources to diluted Mineral Resources classification involves the following steps: • Re-blocking to SMU block size, accounting for losses and dilution; • Applying a processing (marginal) cut-off, which considers all processing costs and refinery product revenue; and • Including all Li2O-bearing spodumene pegmatite SMU blocks. Table 11-19 presents the results of this process, outlining the diluted Mineral Resources included in the Syväjärvi mine plan. Due to the re-blocking process, which defines classification based on dominant mass, some portions of the Mineral Resources are classified as “non-class.” The non-class material is blocks containing mostly waste material. This is despite these blocks containing Li2O-bearing spodumene pegmatite, which was included in the ore classification process prior to the application of marginal lithium value COG as indicated in Table 11-20. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 202 Table 11-19: Syväjärvi diluted Mineral Resources in the LOM plan after SMU re-blocking and including all Li2O-bearing spodumene pegmatite blocks. Syväjärvi Diluted Mineral Resources in Mine Plan – After Mining Losses and Dilution & Spodumene > 0 Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Measured Mineral Resources 1.58 0.66 1.41 55 Indicated Mineral Resources 0.66 0.60 1.30 21 Inferred Mineral Resources 0.13 0.58 1.24 4 Non-class 1.06 0.01 0.03 1 Grand Total 3.43 0.44 0.96 81 The diluted Mineral Resources, categorised by Mineral Resource classification in the Syväjärvi LOM plan, are summarised in Table 11-20. This is after applying the marginal lithium value COG of 0.188% Li2O and reclassifying all diluted ore blocks by dominant mass and reassigning the Mineral Resources classification. Measured Mineral Resources and Indicated Mineral Resources account for 2.91 Mt (92%) at a weighted average grade of 1.06% Li2O, while Inferred Mineral Resources in the LOM plan comprise 0.24 Mt (8%) at a weighted average grade of 0.67% Li2O. Table 11-20: Syväjärvi diluted Mineral Resources in the LOM plan after marginal cut-off. Syväjärvi Diluted Mineral Resources in Mine Plan – After Marginal COG Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Measured Mineral Resources 1.92 0.54 1.16 55 Indicated Mineral Resources 0.99 0.40 0.86 21 Measured Mineral Resources + Indicated Mineral Resources 2.91 0.49 1.06 76 Inferred Mineral Resources 0.24 0.31 0.67 4 Grand Total 3.15 0.48 1.03 80 11.5.1.4 Syväjärvi Mineral Resource to Mineral Reserve Figure 11-18 illustrates the waterfall graph summarising the process of converting in situ Mineral Resources to Mineral Reserves for Syväjärvi, including the relevant Li2O and lithium grades.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 203 Waterfall Graph Summarising the Process of Converting In situ Mineral Resources to Mineral Reserves for Syväjärvi Figure 11-18: In situ Mineral Resources to Mineral Reserves for Syväjärvi. 11.5.2 Rapasaari Mineral Resource to Mineral Reserve Process 11.5.2.1 Rapasaari In situ Mineral Resources The in situ Mineral Resources at Rapasaari are illustrated in Figure 11-19. The figure illustrates both plan and side view of the in situ Mineral Resources at the Rapasaari site, classified according to Mineral Resource confidence levels. The plan view presents the spatial distribution of Measured Mineral Resources, Indicated Mineral Resources, and Inferred Mineral Resources within the mine boundary, extending approximately 1.5 km along strike from north to south. Most of the Measured Mineral Resources are concentrated in the northern and central parts of the orebody. The side view from the west provides a cross-sectional perspective, highlighting the orebody depth and relationship to the surface topography, extending down to a maximum depth of 290 m. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 204 Plan view Sideview from West Plan and Side Views of the In situ Mineral Resources, Categorised by Mineral Resource Classification, shown in Relation to the Mine Boundary and Surface Topography – Rapasaari Figure 11-19: Rapasaari in situ Mineral Resources. The Rapasaari in situ Mineral Resources total 10.83 Mt with a weighted average grade of 1.23% Li₂O. As shown in Table 11-21, the majority of the material is classified as Indicated Mineral Resources (62%), followed by Measured Mineral Resources (19%) and Inferred Mineral Resources (19%). Further details in terms of lithium grade and lithium carbonate equivalent (LCE) content on the in situ Mineral Resources at Rapasaari, categorised by Mineral Resource classification, are provided in Table 11-21. 500 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 205 Table 11-21: In situ Mineral Resources categorised by Mineral Resource classification at Rapasaari. Rapasaari In situ Mineral Resources Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Measured Mineral Resources 2.02 0.59 1.28 64 Indicated Mineral Resources 6.75 0.56 1.21 203 Inferred Mineral Resources 2.06 0.57 1.23 63 Total 10.83 0.57 1.23 329 Note: In situ Mineral Resources incl. Geological loss 11.5.2.2 Rapasaari In situ Mineral Resources Exclusive & Inclusive of Mine Plan Figure 11-20 presents the in situ Mineral Resources, both within and beyond the Rapasaari LOM pit design. The figure illustrates that most of the Mineral Resources, particularly the Measured Mineral Resources, are contained within the LOM pit. In contrast, the Mineral Resources located outside the mine plan are primarily classified as Indicated Mineral Resources and Inferred Mineral Resources. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 206 Isometric view Plan view In situ Mineral Resources Inclusive and Exclusive of LOM Pit Design – Rapasaari Figure 11-20: Rapasaari in situ Mineral Resources inclusive and exclusive of LOM plan. 500 m

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 207 Table 11-22 provides a summary of the in situ Mineral Resources at Rapasaari, both within and beyond the LOM plan, categorised by Mineral Resource classification. The total in situ Mineral Resources amount to 10.83 Mt, with an average grade of 1.23% Li₂O. Of this, 8.50 Mt (79%) are contained within the LOM plan, averaging 1.24% Li₂O, predominantly consisting of Measured Mineral Resources and Indicated Mineral Resources (90% of the total). Inferred Mineral Resources within the LOM plan account for 0.86 Mt (10%). Outside the LOM plan, 2.33 Mt of Mineral Resources remain, with an average grade of 1.23% Li₂O. These Mineral Resources are primarily classified as Indicated Mineral Resources and Inferred Mineral Resources. Table 11-22: Rapasaari in situ Mineral Resources exclusive and inclusive of mine plan. Rapasaari Mineral Resources Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) In situ Mineral Resources Outside Mine Plan Measured Mineral Resources 0.07 0.59 1.27 2 Indicated Mineral Resources 1.07 0.52 1.13 30 Inferred Mineral Resources 1.20 0.58 1.24 37 Subtotal 2.33 0.55 1.19 69 In situ Mineral Resources Inside Mine Plan Measured Mineral Resources 1.95 0.59 1.28 62 Indicated Mineral Resources 5.68 0.57 1.23 173 Inferred Mineral Resources 0.86 0.57 1.22 26 Subtotal 8.50 0.58 1.24 261 Grand Total 10.83 0.57 1.23 329 Note: In situ Mineral Resources include Geological loss. 11.5.2.3 Rapasaari Diluted Mineral Resources in Mine Plan Similar to the process utilised at Syväjärvi, the conversion of in situ Mineral Resources to diluted Mineral Resources involved re-blocking to SMU size to account for dilution and losses, applying a processing (marginal) cut-off that factors in costs and revenue, and incorporating all Li2O-bearing spodumene pegmatite SMU blocks. Table 11-23 summarises the resulting diluted Mineral Resources within the Rapasaari mine plan, reflecting these adjustments. Some blocks, despite containing Li2O-bearing material, are classified as “non-class” due to the re-blocking method, which determines classification based on dominant mass. Table 11-23: Rapasaari diluted Mineral Resources in the LOM plan after SMU re-blocking and including all Li2O-bearing spodumene pegmatite blocks. Rapasaari Diluted Mineral Resources in Mine Plan – After Losses and Dilution & Spodumene > 0 Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Measured Mineral Resources 1.95 0.59 1.28 62 Indicated Mineral Resources 5.66 0.57 1.23 173 Inferred Mineral Resources 0.86 0.57 1.22 26 Non-class 5.14 0.04 0.09 11 Grand Total 13.61 0.37 0.80 271 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 208 The diluted Mineral Resources for Rapasaari, classified within the LOM plan, are summarised in Table 11-24. Following the application of a marginal lithium value cut-off and reclassification based on dominant mass, Measured Mineral Resources and Indicated Mineral Resources total 10.06 Mt (89%) at a weighted average grade of 0.94% Li₂O. Inferred Mineral Resources account for 1.23 Mt (11%) with an average grade of 0.85% Li₂O. Table 11-24: Rapasaari diluted Mineral Resources in the LOM plan after marginal cut-off. Rapasaari Diluted Mineral Resources in Mine Plan – After Marginal COG Classification Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Measured Mineral Resources 2.42 0.48 1.03 62 Indicated Mineral Resources 7.63 0.43 0.92 173 Measured Mineral Resources + Indicated Mineral Resources 10.06 0.44 0.94 235 Inferred Mineral Resources 1.23 0.39 0.85 26 Grand Total 11.28 0.43 0.93 261 11.5.2.4 Rapasaari Mineral Resource to Mineral Reserve Figure 11-21 illustrates the waterfall graph summarising the process of converting in situ Mineral Resources to Mineral Reserves for Rapasaari, including the relevant Li2O and lithium grades. Waterfall Graph Summarising the Process of Converting In situ Mineral Resources to Mineral Reserves for Rapasaari Figure 11-21: In situ Mineral Resources to Mineral Reserves for Rapasaari. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 209 11.5.3 Mineral Reserves Estimates: Syväjärvi and Rapasaari The open-pit Mineral Reserves for the Keliber Syväjärvi and Rapasaari operations are summarised in Table 11-25. The Mineral Reserves are reported as delivered to the concentrator plant or related stockpile, based on the modifying factors discussed in the previous sections and the attributable interest of Sibanye-Stillwater in Keliber. Table 11-25: Mineral Reserves for Keliber’s Syväjärvi and Rapasaari open-pit operations as at 31 December 2024. Syväjärvi Mineral Reserves as at 31 December 2024 - Reported on a 100% attributable basis 100% Mineral Reserves as at 31 December 2024 - Reported on a 79.82% attributable basis 79.82% Mineral Reserves Class Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Proven Mineral Reserves 1.92 0.54 1.16 55 1.53 0.539 1.16 44 Probable Mineral Reserves 0.99 0.40 0.86 21 0.79 0.402 0.86 17 Total/avg 2.91 0.49 1.06 76 2.32 0.492 1.06 61 Rapasaari Mineral Reserves as at 31 December 2024 - Reported on a 100% attributable basis 100% Mineral Reserves as at 31 December 2024 - Reported on a 79.82% attributable basis 79.82% Mineral Reserves Class Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Proven Mineral Reserves 2.42 0.48 1.03 62 1.93 0.478 1.03 49 Probable Mineral Reserves 7.63 0.43 0.92 173 6.09 0.427 0.92 138 Total/avg 10.06 0.44 0.94 235 8.03 0.439 0.94 188 Syväjärvi & Rapasaari Mineral Reserves as at 31 December 2024 - Reported on a 100% attributable basis 100% Mineral Reserves as at 31 December 2024 - Reported on a 79.82% attributable basis 79.82% Mineral Reserves Class Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Tonnes (Mt) Li (%) Li₂O (%) LCE (kt) Proven Mineral Reserves 4.34 0.50 1.09 117 3.47 0.50 1.09 93 Probable Mineral Reserves 8.62 0.42 0.91 195 6.88 0.42 0.91 155 Total/avg 12.96 0.45 0.97 311 10.35 0.45 0.97 248 Notes: 1. The Mineral Reserves estimate is reported in accordance with the requirements of the S-K 1300 Code 2. The Mineral Resources were reported exclusive of the Mineral Reserve 3. The Mineral Reserves are Reported as ROM material delivered to the concentrator plant, or related ROM stockpile 4. Tonnage estimates are in metric units and reported as million tonnes (Mt) 5. Numbers may not add up due to rounding 6. Mineral Reserves are also reported on a 79.82% attributable to Sibanye Stillwater basis 7. The Mineral Reserves are subject to EP approvals 8. Measured Mineral Resources converted to Proven Mineral Reserves 9. Indicated Mineral Resources converted to Probable Mineral Reserves 10. No Inferred Mineral Resources included in the Mineral Reserves 11. COG for Syväjärvi at 0.20% Li₂O 12. COG for Rapasaari at 0.30% Li₂O 13. Li (%) content was calculated by multiplying the Li₂O (%) content by a factor of 0.465 14. LCE content was calculated by multiplying the Li (%) content by a factor of 5.323 15. The Mineral Reserves are inclusive of the Keliber lithium refinery of which the process is likely to be the first implementation of this specific flowsheet. While the individual unit processes are not novel, the pilot trials have significantly de-risked the flowsheet. The S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 210 Project’s financial model includes a reasonable ramp-up for the hydrometallurgical plant, and sensitivities to address commissioning risks were completed to incorporate the process. 16. The 31 December 2024 interpretation of the environmental permit regarding the production of 540 ktpa at Syväjärvi is under review. The inclusion of a 10% increase in the production limitation is also subject to environmental controls and approval. 11.5.4 Mineral Reserves Estimate: 2023 vs 2024 Reconciliation The comparison between the previous Mineral Reserves (2023) and the current Mineral Reserves (2024) is shown in Table 11-26. The major differences between the two are: • Mineral Resources estimation methodology; • Geotechnical parameters; • Pit constraint perimeter increased marginally; and • Economical parameters. Table 11-26: Reconciliation between 2023 Mineral Reserves and 2024 Mineral Reserves. Syväjärvi Mineral Reserves as at 31 December 2023 Reported on a 100% attributable basis Mineral Reserves as at 31 December 2024 Reported on a 100% attributable basis Mineral Reserve Class Tonnes (Mt) Li (%) Li2O (%) LCE (kt) Tonnes (Mt) Li (%) Li2O (%) LCE (kt) Proven Mineral Reserves 1.63 0.52 1.12 44 1.92 0.54 1.16 55 Probable Mineral Reserves 0.50 0.42 0.91 13 0.99 0.40 0.86 21 Total/avg 2.13 0.50 1.07 56 2.91 0.49 1.06 76 Rapasaari Mineral Reserves as at 31 December 2023 Reported on a 100% attributable basis Mineral Reserves as at 31 December 2024 Reported on a 100% attributable basis Mineral Reserve Class Tonnes (Mt) Li (%) Li2O (%) LCE (kt) Tonnes (Mt) Li (%) Li2O (%) LCE (kt) Proven Mineral Reserves 2.13 0.46 0.98 51 2.42 0.48 1.03 62 Probable Mineral Reserves 4.89 0.40 0.87 105 7.63 0.43 0.92 173 Total/avg 7.02 0.42 0.90 157 10.06 0.44 0.94 235 Syväjärvi & Rapasaari Mineral Reserves as at 31 December 2023 Reported on a 100% attributable basis Mineral Reserves as at 31 December 2024 Reported on a 100% attributable basis Mineral Reserve Class Tonnes (Mt) Li (%) Li2O (%) LCE (kt) Tonnes (Mt) Li (%) Li2O (%) LCE (kt) Proven Mineral Reserves 3.76 0.48 1.04 97 4.34 0.50 1.09 117 Probable Mineral Reserves 5.39 0.41 0.87 117 8.62 0.42 0.91 195 Grand Total/avg 9.15 0.44 0.94 213 12.96 0.45 0.97 311 It is important to note that the 2023 Mineral Reserves were reported with a different resource model. The impact of the 2024 regularised block model was analysed by reporting tonnages within the 2022 FS pit design. Hence, the increase of 0.42 Mt (shown in Figure 11-22) and the increase of 1.15 Mt (shown in Figure 11-23) are the result of changes to the resource model estimation methodology. The comparison, therefore, shows the difference in Mineral Reserves for 2023 and 2024 on a comparable basis.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 211 Syväjärvi Mineral Reserves Reconciliation: 2024 versus 2023 Figure 11-22: Syväjärvi Mineral Reserves reconciliation: 2024 vs 2023. The reconciliation for Rapasaari is shown in Figure 11-23. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 212 Rapasaari Mineral Reserves Reconciliation: 2024 versus 2023 Figure 11-23: Rapasaari Mineral Reserves reconciliation: 2024 vs 2023. 11.5.5 Conversions In accordance with industry standards, the total metal content of lithium (Li) Mineral Resources and Mineral Reserves is reported as LCE, a key final product in the lithium mining value chain. LCE is calculated from in situ lithium content by applying a conversion factor of 5.323. To derive lithium hydroxide monohydrate (LiOH·H₂O) from LCE, the value is divided by a factor of 0.88. Lithium (Li) itself is derived from lithium oxide (Li₂O) using a conversion factor of 0.465. 11.5.6 QP Comments The stakeholders of this Project invested significant time and effort to define the modifying factors, mine design parameters, and scheduling parameters to develop a business case and, ultimately, convert Mineral Resources to Mineral Reserves. The key operational risks that could impact the Mineral Reserves are listed below: • Commodity prices and exchange rate assumptions: SSW have adopted forward-looking price assumptions. Any material deviations from these assumptions could impact the Mineral Reserves. • Geology and grade control: The orebody orientation and thickness require strict grade-control measures to ensure the dilution is kept to a minimum. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 213 • Mining operational cost: Mining costs utilised during the pit optimisation are based on the owner versus contractor trade-off study completed in 2023. • Infrastructure timeline: Delays in the completion dates of the Project infrastructure could impact negatively on the production ramp-up. • Waste rock specification: Restrictions on waste dumping with deleterious elements could impact the sequence of mining when ensuring the deleterious elements’ limits are adhered to. • Permitting: Although several of the required operating permits have been obtained, potential timing delays due to public objections and appeals could impact construction timelines. EP conditions could also be strenuous, impacting or delaying planned mining operations. • Concentrator plant throughput: The efficiency of the plant is highly dependent on the close management of the Blackrock dilution in the plant feed as too much dilution will impact the efficiency of the ore sorters negatively. • Concentrator recovery: Good control of the blasting parameter is also necessary to limit the amount of Blackrock bypassing to the flotation plant as this will have a negative impact on the flotation recovery. In addition, mica will impact the operation of the plant negatively, as build-up of the mica in the plant will reduce the plant's availability. In addition, all mica that enters the flotation plant will report in the product, and this will have a negative impact on the recovery. Therefore, close monitoring of mica in the grade control is recommended. • Keliber lithium refinery feed characteristics: The ore characterisation and mineralogy tests have provided a thorough understanding of the plant feed characteristics, thereby minimising the impact of this risk as the ore's properties and behaviour have been well-defined. Pilot plant test trials were successfully performed on material from the Syväjärvi and Rapasaari deposits as well as on external samples from North America. Despite variations in feed grades, the pilot plant demonstrated consistent behaviour, showing its adaptability to variability in feed ore characteristics and compositions. • Keliber lithium refinery engineering design: The design of the equipment, including spares and ancillary equipment, remains a potential risk as any design flaws in equipment could result in delays and costly rework, ultimately impacting the plant's ability to achieve its design capacity and meet production targets. However, the ramp-up period will be instrumental in assessing whether the installed equipment meets the specified performance requirements, thereby mitigating the risk. A secondary risk relates to the availability of critical spares which are essential for a smooth commissioning and production ramp-up. This risk is considered low as provisions have been made to address it in the financial model. • Keliber lithium refinery scaling up from pilot plant test trials: Scaling up the plant from pilot plant test trials poses a potential risk, as inaccurate scaling can impact equipment design. This has not been evaluated by the VBKOM team. Metso has extensive experience, and the use of proven models enables them to accurately scale up plants. Furthermore, their successful track record of operational scaled-up plants worldwide demonstrates their proficiency in this area. • Keliber lithium refinery process flow diagram: The lithium extraction methods utilised at Keliber are not new. The novelty of the process is with regard to its exclusion of hazardous acids and sulphates (which are commonly used at other lithium extraction plants), rendering it a more environmentally friendly option as produced effluent is free from harmful minerals and safer for disposal. Therefore, the impact of this risk is minimised by Keliber’s use of proven technologies albeit using different reagents. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 214 • Human capital: A significant number of skilled personnel will be required to develop and work at the operations. Labour availability could impact planned production and build-ups. • Waste storage facilities and capacity: The Syväjärvi WRD height requirement increased by 6 m, and this amendment to the permit is required. The Rapasaari WRD requires extension beyond the current mining area; approval of this is required to ensure sufficient space is available. • OVB waste storage capacity: The Syväjärvi final OVB dump requires environmental approval to ensure sufficient dumping space is available. • Hazardous waste storage facilities and capacity: Hazardous waste material is defined as sulphur >1.0% and arsenic >100 ppm. In the previous study, the impact of arsenic was overlooked. For Syväjärvi, the storage requirement marginally increased, however, at Rapasaari, the storage requirement increased to above 8 MLCM. Areas were identified to host the hazardous material, with associated environmental controls. This requires amendment to permitting. • ROM ore stockpile facilities and capacity: Any delays in the permitting of the Syväjärvi ROM Ore stockpile facilities could impact production build-up, potentially impacting the production ramp-up. The Rapasaari Ore stockpile size needs to be optimised and sufficient space needs to be identified.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 215 12 MINING METHODS [§229.601(b)(96)(iii)(B)(13) The mining methods previously selected include conventional truck and shovel operation as the most suitable open- pit mining method for Syväjärvi and Rapasaari (SRK, 2023). A truck and shovel operation refers to the use of large, generally rigid body, off-highway haul trucks that are loaded with blasted rock by large shovels or excavators. This combination of mining equipment is a proven technology and is used in many open-pit mines throughout the world. The key points of a truck and shovel operation are: • The truck and shovel combination is a known and proven mining method, capable of handling most rock types in Finland. Potential mining contractors have suitable equipment readily available; • The haulage and loading equipment can handle both free-dig and blasted material; and • The ability to produce the total annual mining rates is anticipated. In-pit ramps and waste rock haul roads are designed for off-highway trucks with a payload of 65 t. The pits are designed with a 20 m bench height (BH) and will be excavated in 10 m benches. Waste rock maximum particle size is not limited. The Syväjärvi operation is limited to a 540 ktpa ROM (post ore sorter) ore production rate due to the limitations set on the EP. No blending of material from different deposits was applied. The Rapasaari open pit is scheduled to be mined after the Syväjärvi deposit is fully mined out at full capacity. The Project targeted LiOH.H2O production of approximately 15 ktpa in the LOM production schedule. 12.1 Rock Engineering – Syväjärvi [§229.601(b)(96)(iii)(B)(13)(i)] The information for this section has been sourced from: • Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study of the Syväjärvi Open Pit Mine (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. Edited by Abraham Saayman, GEOTEC Africa CC. Updated geotechnical information has been obtained to enhance the geotechnical information presented as part of the 2022 FS, as described below. Independent review found that some aspects of the 2022–2023 geotechnical study for the Syväjärvi pit design were not up to industry FS reliability norms. Most notably, the frictional shear strength(s) of geological structures (natural defects), which largely dictate the bench-berm designs in such strong rock slopes, was not based on (empirical) estimation from available geotechnical core logs data, nor calculated from shear tests data. Field index measurements and/or laboratory direct shear tests (DSTs) of structure surfaces were lacking. The results of kinematic and wedge stability analyses, hence, bench-berm designs and the resulting inter-ramp slope angles (ISAs), were thus not to FS confidence levels. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 216 Key objectives with the 2024–2025 geotechnical study were to address these shortcomings and to upgrade other aspects to industry FS norms. The widely applied handbook Guidelines for Open Pit Slope Design (Read & Stacey, Guidelines for Open Pit Slope Design, 2009), a culmination of the multinational Large Open Pit Project coordinated by the Australian CSIRO, provided guidance and metrics for this. 12.1.1 Models and Data Compilation and Augmenting Comprehensive geological, geotechnical, and oriented core logs data from several drill campaigns (from 2014 to 2022) provide very good 3D spatial coverage of the proposed small Syväjärvi open pit. The following summarise the main activities and processes undertaken in this FS upgrade stage to compile models and data: 1. Geological domains were defined based on the 3D lithological and structural models, which models were developed by AFRY (AFRY Oy, 2021) from geophysical surveys and from geological and geotechnical core logs. The main rock types in which the slopes are to be developed are shown in Figure 12-3 in the next section. 2. Review of the Syväjärvi structural geological modelling report (AFRY Oy, 2021) and qualitative and quantitative review of 3D models of major structures. a. AFRY interpreted broken rock zones from RQD geotechnical core logs and from Geological Strength Index (GSI) values derived from RQD and Q-system geotechnical logs; and b. Apparent continuous, broken rock zones are inferred as potential geological FLTs or brittle shear zones, and four such FLTs were modelled in 3D. These are shown in Figure 12-4 in the next section. 3. Screened, with sophisticated tools like LEAPFROG®, all lithological, oriented structures, and geotechnical logs compiled in a comprehensive Access Database, to: a. Incorporate only rock fabric (minor geological structures) data within about 50 m (+50 m and -50 m) from the proposed (June 2019) pit slopes. The selected holes and intervals are shown in Figure 12-1; b. This is to focus on data pertinent to the pit slopes and to minimise (irrelevant) data distant from the slopes, adding variability to and possibly diffusing or masking joint sets and/or rock fabric characteristics; c. Cross-coding of oriented core structures logs with rock types from the lithological logs; and d. Review core logs against core photos and screening out intact or annealed (closed) textures like foliation (FO), schistosity (ST), and/or banding (BD) orientation measurements from the data, leaving only data on naturally open geological structures (i.e. rock fabric discontinuities) of the same, and of FLTs and joints (JNs). 4. Detailed geotechnical relogging of the selected ±100 m-long core sections from 10 selected holes to remeasure structure orientations and characterise geotechnical (surface) properties of specific structure types according to standard Q-system parameters. These relogged holes are shown in Figure 12-2. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 217 Selected Sections of Slope-Proximate Boreholes of which the Geotechnical Data, in Particular the Discontinuity Orientation Logs, were Used in the 2024 Upgrade Study (Oblique View) Figure 12-1: The selected sections of slope-proximate boreholes of which the geotechnical data, in particular the discontinuity orientation logs, were used in the 2024 upgrade study (oblique view). The “boreholes count” and “total logged metres” refer to the original larger database. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 218 The Spatial Distribution of the Ten Drill Holes from which Cores were Selected for Relogging shown in Relation to the June 2019 Proposed Syväjärvi Pit Design (Oblique View) Figure 12-2: The spatial distribution of the ten drill holes from which cores were selected for relogging shown in relation to the June 2019 proposed Syväjärvi pit design (oblique view). 5. The geotechnical core log datasets were processed and analysed in detail to statistically determine rock quality parameters (Q-system parameters) of structure types in the main rock (lithology) types. a. Rock mass characteristics parameters RQD and GSI were not reprocessed and derived from the selected sub-database. RQD and GSI values from the 2022–2023 rock mechanical assessment are regarded as representative of the total Syväjärvi pit area, hence, also of the smaller sub-database proximate to the pit slopes; and b. Logged Joint Alteration (Ja) and Joint Roughness (Jr) parameters were applied to calculate Indicative Friction Angles (IFAs) of structures with an empirical function, generally suited up to PFS level, but which was not applied in previous geotechnical assessments. 6. Selection for direct shear testing of samples of specific structure types and intact rock of the main rock types in which the west and east pit slopes will be developed. For the 2024 upgrade of the Syväjärvi FS: a. Saw-cut surfaces of 3 x Mica Schist (MS), 3 x Metavolcanic (MV), 3 x Plagioclase Porphyry (PP), and 2 x Sulphidic Schist (SS) intact rock samples were shear tested (3 load stages each test); b. DSTs of 60 x open structures [FO, ST, and JNs were direct shear tested (3 load stages each test); c. Samples were couriered to and tested by accredited E-Precision Laboratory in Perth, Western Australia. These were the first DSTs of geological structures done for the Syväjärvi (and Rapasaari)

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 219 Projects. The sample numbers by rock and structure types are summarised in Table 12-2 in the next section; d. The DST results were processed and analysed to derive shear strength input parameters specific to each of the defined Geological Domains. 7. Rock quality Q-parameters summarised from processed geotechnical logs and “residual friction angle” data from DST on intact rock samples were applied in the Barton-Bandis (BB) empirical function to estimate friction angles (effective shear strengths) by structure type in the main rock types. Suggested friction-reduction structure scale (structure size) adjustments were not applied, firstly because: a. Even slight larger scale undulations, waviness, or curvature, will more than offset scale strength reduction factors; and b. No mapping data are available on structure lengths and shape (1st order roughness). Guesstimating “typical” lengths of structure types is not to FS reliability. Bench faces must be mapped during early pit development to provide these data. 8. The rock fabric datasets selected from oriented core were processed and analysed with stereograms to identify and define structure (JN) sets in each of the defined Geological Domains. 9. The hydrological, hydrogeological, and groundwater modelling sections of the 2022 draft FS Report (Volume 4, Chapters 18–19) were reviewed. The Syväjärvi geohydrological data and models are based on investigations of: a. Trial pits in the till OVB; b. Eleven (11) groundwater observation wells installed from 2014 to 2020 in the Syväjärvi area; c. Hydraulic conductivity field measurements of the till OVB with slug tests done in nine groundwater observation wells. (Hydraulic conductivity of a rock mass is measured as groundwater flow rate (K) in m/s. Generally equivalent to permeability. Transmissivity of a formation is expressed as flow through a unit square area, as m2/s.); d. Hydraulic conductivity (permeability) field measurements of (fractured) bedrock to depth of the then considered underground mine, by means of Positive Flow Logging (PFL) measurements in four drillholes done in 2021 Q1. These drillholes were inter alia selected so that the hydraulic conductivity of interpreted major structures (inferred FLTs) and the different rock types could be measured; e. Pumping tests and drawdown measurements performed around each of the four PFL-measured boreholes. These investigations and interpretation of results are described in detail in Appendix 1B- A1 of Vol 4 of the draft (January 2022) FS report; f. Detailed spatial and statistical analyses of geomechanical RQD core logs by elevation (depth) and lithology, as indication of hydraulic conductivity and spatial variability therein. (RQD is a measure of the degree of natural fracturing of a rock mass.); and g. AFRY’s 2023 finite element numerical model with groundwater (pore pressures) was revisited to consider whether revision may be warranted. 12.1.2 Key Findings and Geomechanical Input Parameters 12.1.2.1 Lithological Models 1. Shallow (<2 m thick) peat deposits overlie tillite OVB (assorted layers of silt, clay, sand, and sandy till). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 220 2. The tillite OVB thickness varies between 12 and 16.5 m over the total Syväjärvi Project area but is generally 9±2 m deep around the projected Syväjärvi pit perimeter. 3. The eastern pit slope will be developed in metasedimentary rock mica schist (MS) with minor plagioclase porphyrite intrusive rock (the PP sill) in the slope toe. 4. The western pit slope will be developed in meta-volcanic rock, or meta-tuff (named “metavolcanite” or “metatuffite”, code MV), with minor plagioclase porphyrite intrusive rock (the PP sill) in the slope toe. 5. The PP and pegmatite (PG) intrusive bodies play a minor role in slope design as these make up very small proportions of the lithologies in which the slopes will be developed (shown in Figure 12-3). 6. No upgrades of the 3D lithological models were deemed necessary for geotechnical purposes. 12.1.2.2 Major Structures Models 1. AFRY interpolated and extrapolated intersections of broken rock and zones of lower RQD and GSI values and developed 3D models of these as FLTs and/or brittle shear zones (AFRY Oy, 2021). 2. High-level review (GEOTEC Africa, June 2024) of these Syväjärvi major structures models found that: a. At least 15 more cored holes were drilled after the 3D models were developed in October 2021; b. More core holes through the modelled structures did not intersect broken rock or (logged) FLTs at the model depths, than holes which did and matched the models; c. Targeted geotechnical drilling in autumn 2023 confirmed some broken rock or FLT zones but refuted others; d. Four more FLTs were logged (shown in Figure 12-4) in holes S-158B and S-160 near the northeast pit corner, which do not match any 3D model. (All but one of these FLTs are harmlessly deep below the proposed pit slope.); S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 221 Lithologies Cutting the June 2019 Design Syväjärvi Open-Pit Slopes Figure 12-3: Lithologies cutting the June 2019 design Syväjärvi open-pit slopes. Light green = metavolcanic rock; orange = spodumene pegmatite; pink = plagioclase porphyrite; dark green = sulphidic schist; blue = mica schist. Plan view. 100 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 222 Four Continuous Broken Rock Zones Interpreted or Inferred as FLTs shown in the June 2019 Proposed Syväjärvi Pit Figure 12-4: Four continuous broken rock zones interpreted or inferred as FLTs shown in the June 2019 proposed Syväjärvi pit. View is to the west (direction 295°). e. This review concluded that the current Syväjärvi major structures (FLT) models are only at Inferred confidence level; f. Drone photography and geological mapping of the Syväjärvi quarry faces in August 2024 to verify modelled FLTs Syväjärvi F1 and Syväjärvi F2, which should by now be exposed, had limited success due to safety and access limitations (AFRY Oy, August 2024). The reliability of these models could not be upgraded. And an unmodelled near-vertical FLT was found near the northeastern corner of the rock quarry. (This is narrow and has favourable orientation relative to the slope.); and g. Time constraints precluded detailed review of the Syväjärvi 3D FLT or broken rock zone models by an expert structural geologist, inter alia by inspecting core photos and/or the stored cores. 3. The August 2024 FS Syväjärvi pit designs largely mitigate potential risks associated with these inferred FLTs (shown in Figure 12-4), with prudent positioned and wide main ramp on the southwest and west slopes, plus adding a secondary (narrow) access ramp on the east slope. 12.1.2.3 Intact Rock-, Structures-, and Rock Mass Strengths 1. Intact rock strength data were obtained from the 2016 (first quantity numbers below) and the 2020 (second quantities below) laboratory strength testing programmes. Most samples were from the Rapasaari Project, but this is quite near the Syväjärvi Project, and the rock types are the same: a. 20 + 36 x density (SG) values, 56 in total; b. 20 + 36 x Uniaxial Compressive Strength (UCS) values, 56 in total; c. 20 + 36 x Young’s Moduli (E) values, 56 in total; d. 20 + 36 x Poisson’s ratio values, 56 in total;

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 223 e. 16 + 22 x Crack Initiation (CImin and CImax and Crack Damage datasets, 38 in total, although several results were noted as unreliable due to the failure mechanisms; and f. 20 + 36 x Tensile strength values (T) from Brazilian disk tests, 56 in total. A summary of the values for the UCS results by rock type is shown in Table 12-1. 2. Laboratory shear tests of structures were lacking in the previous geotechnical studies. Samples of representative structures and rock types were selected from cores of the twelve relogged geotechnical holes for geomechanical testing by accredited and highly sophisticated E-Precision Laboratory in Perth, Western Australia. Sixty (60) samples of the dominant structure types in the three main rock types in which the slopes are to be developed were sent for DSTs. Normal stress ranges appropriate for shallow (shear) failures on structures in mining bench scale were specified to E-Precision Laboratory. These sample suites are summarised in Table 12-2. Eleven intact rock samples of the same rock types were also sent for DST on saw-cut surfaces to derive residual (or basic) friction angles. These are summarised in Table 12-3. Peak and residual friction angles of structures were derived by analysing these DST results by structure types and by rock types. The results are summarised in Table 12-4. These show negligible differences in friction angles among FO, JN, and ST planes. For this reason, friction angles can and were applied by main rock types or geological domains. Even these Phi differ only by two degrees between the MS + SS (west) and MV (east) domains (refer to Table 12-5). 3. Friction angles of structures were also estimated with the industry-accepted BB empirical function (Barton & Chaubey, 1977) (Barton, 1987). Key input parameters to the BB function are Joint Roughness Coefficients (JRCs) and Joint Compressive Strengths (JCSs) (Barton & Bandis, 1990): a. JRC were derived from geotechnical core logs of 12 holes core intervals which were relogged in 2024; and b. JCS were estimated by downgrading UCS values from the (2016 and 2020) laboratory tests to account for weathering of structure surfaces, using factors in a nomogram suggested by Stacey and Page (1986). Defect friction angles thus estimated from geotechnical core log data with the BB empirical method compare quite accurately with the laboratory DST results (see Table 12-3 and Table 12-4). 4. The laboratory-measured friction angles (Phi-DST) were applied in kinematic and limit equilibrium slope stability analyses: a. Phi = 36° ± 4° is applied in Mica Schist (MS, Phi-DST = 35°) + Sulphidic Schist (SS, Phi-DST = 37°); and b. Phi = 37° ± 4° is applied in Metavolcanics (MV). Table 12-1: UCS test results. Rock Codes Sample No. Average UCS (MPa) Std Dev UCS (MPa) MS 8 92 48 MV 9 140 42 PG 8 158 47 PP 7 211 26 SS 3 155 20 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 224 Table 12-2: Summary of the DST samples by rock type and structure type. Samples for saw-cut DST were intact rock, to be sawn parallel to the FO or ST rock texture. Lithology Structure Type Number of Samples Sulphidic schist Total 18 JN 6 ST 10 Saw-cut 2 Mica schist Total 19 JN 7 ST 9 Saw-cut 3 Metavolcanite Total 21 JN 11 FO 7 Saw-cut 3 Plagioclase porphyrite Total 13 JN 4 FO 6 Saw-cut 3 Table 12-3: Residual Friction Angles (Phir) derived from DST on saw-cut intact rock samples, after applying Hencher s correction. EP Laboratory calculated Phir by Mohr-Coulomb curve fit to three test stages. Rock Codes Average Residual Friction Angles φres Std Dev. Residual Friction Angles φres MS 28° ±1° MV 30° ±2° PP 30° ±2° SS 30° ±0° Note: All the values are corrected in accordance with Hencher’ s proposed method. This compensates for effects of dilation when top and bottom halves of a sample rides over asperities during shearing. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 225 Table 12-4: Peak friction angles derived from DST on structure types in four rock types, after applying the Hencher correction. Structure types are named according to the 2024 relogged cores from which the samples were taken. FQ JN ST Rock Codes Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Total Count of S-DST MS 21 35 3.3 27 35 3.7 48 MV 21 35 3.3 33 38 3.9 54 PP 18 38 3.6 12 37 3.0 30 SS 18 36 2.7 30 38 2.2 48 Grand Total 39 36 3.7 84 37 3.7 57 37 3.3 180 Note: All the values are corrected in accordance with Hencher’ s proposed method. This compensates for effects of dilation when top and bottom halves of a sample rides over asperities during shearing. Table 12-5: Mean and standard deviations of peak friction angles by rock type derived from DST, after applying Hencher’s correction (green numbers), compared with frictional angles estimated with the BB criterion at normal stress 400 kPa, to approximately match the mean normal stress applied in the DST (left three columns). Mean Phi-BB are within one standard deviation of Phi-DST, except for PP which is within two standard deviations. Rock Codes DST Applied Normal Stress Range σn (kPa) Peak Friction Angle from DST (°) BB Calculation Min. Mean Max. Mean Standard Deviation Estimated Friction Angle (°) @ 400 kPa MS 101 394 831 35 4 34 MV 101 382 846 37 4 37 PP 101 424 849 37 3 43 SS 98 386 872 37 3 38 5. Rock mass quality by rock type, quantified with GSI values, derived from geotechnical drill core logs, show that rocks are generally massive (intact) or blocky, with wide JN spacings. Statistics show that all the rock types have lower quartile values above GSI-value 70, and spodumene pegmatite and plagioclase porphyrite have GSI-values above 80 (refer to Figure 12-5 for a graphic summary of results). 6. In combination with high intact rock strengths (50 MPa < UCS < 250 MPa), these high GSI values denote strong rock masses. Slope stability will consequently be dictated by the rock fabric (“minor” structures like FO, ST, and JN planes) on bench scale, and rock mass failures will be highly unlikely. The 2024–2025 upgraded geotechnical assessment accordingly focussed on defining structures sets and applying representative frictional shear strengths of structures to analyse the probabilities of structurally controlled planar, wedge, and toppling failure mechanisms on bench scale. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 226 Syväjärvi Drill-Hole-Based GSI Values of the Main Rock Types (Sill = Plagioclase Porphyrite) Figure 12-5: Syväjärvi drill-hole-based GSI values of the main rock types (sill = plagioclase porphyrite). The mean values are indicated by the red diamonds. The median values are indicated by the lines that cross inside of the coloured boxes. The box encloses the interquartile range around the median. The whiskers extend out to vertical lines that mark the extreme value extents. 12.1.2.4 Rock Fabric Sets in Slope Design Sectors The more prominent rock fabric sets appeared to be spatially ubiquitous, with similar orientations throughout the east (MV) and west (MS) geological domains. This obviates the need to subdivide the geological domains into smaller geotechnical domains. The mean and standard deviation dip directions and dip angles of the defined structures sets are summarised in Table 12-6. 1. Study of the structures datasets show that west-dipping ST and (sparse) FO planes will largely determine the east slope bench-berm configurations. 2. FO and ST dip favourably south-westwards into the west slope but may combine with JNs to cause minor toppling failures. This favours the west and south-west slopes for placing the main ramp. 3. Less prominent sets (low-density clusters containing few structures) occur in some slope design sectors (SDSs) and were accounted for in stability analyses. 4. All structures, whether in the defined sets or “random JNs” i.e. structures not grouped into the defined sets, were accounted for in kinematic analyses of each SDS.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 227 Table 12-6. Summary of the discontinuity sets identified in Syväjärvi open pit in the SDSs. Domain Slope Sector Joint Set (JS) Dip Direction (°) Dip (°) East 1A JS1 244 57 JS2 133 84 JS3 175 34 1B JS1 249 71 JS2 152 11 West 2A JS1 245 60 JS2 270 68 JS3 071 18 2B JS1 255 62 JS2 272 75 JS3 351 84 2C JS1 164 67 JS2 164 53 JS3 019 49 JS4 331 33 North and South Transition Sectors 3A JS1 255 65 JS2 093 12 3B JS1 141 81 JS2 297 2 12.1.2.5 Hydrogeological Parameters and Groundwater Models 1. There is a narrow natural wetland northwest of the planned open pit, and a shallow, larger, constructed (built with bunds) wetland north of the open pit. 2. Observations in observation-monitoring bores indicate a very shallow groundwater level, approximately 0 to 4 m below the ground surface, thus, in OVB. 3. Slug tests in OVB yielded low hydraulic conductivity of K = 7.7 x 10-7 m/s (hydraulic conductivity ≈ permeability). 4. Hydrogeological flow rate measurements in deep bores showed that the hydraulic conductivity of the bedrock decreases with depth – see Figure 12-6. a. Hydraulic conductivity below vertical depth at 50 mamsl is generally very low, from K ≈ 1 x 10-9 to K≈1 x 10-7 m/s – see Figure 12-6; b. K ≈ 1 x 10-8 equates flow rate of groundwater through the rock mass of about 3.2 metres / year, thus, extremely slow; c. These very low permeability values imply that passive drawdown of groundwater around the pit will be extremely slow. Hence, that the drawdown “cone” around the pit will generally be steep and be close to the slopes; and S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 228 d. This motivates for analysing overall slope stability with elevated groundwater levels close to the pit slopes, which is detailed in the Syväjärvi FS geomechanical report. 5. Very low transmissivity values (refer to Figure 18-80 in the 2022 draft FS Report, Volume 4, Chapter 18) indicate that very low groundwater inflow volumes into the pit can be expected, except for potential temporary and localised slightly higher inflows from some major geological structures (the broken rock zones) and through closely jointed pegmatite intrusions, where such intersect the slopes. 6. Good correlation hydrogeological field measurements and the derived hydraulic conductivities of bedrock were found with RQD values. a. Lower RQD values of more fractured rock zones match higher hydraulic conductivities; and b. This implies localised drawdown of groundwater level and localised depressurisation of slopes and structures should occur where the lower RQD broken rock zones intersect the slopes. Hydraulic Conductivity, K, in m/s, Measured in Monitoring Bores in Syväjärvi Rocks in Q1 Of 2021 Figure 12-6: Hydraulic conductivity, K, in m/s, measured in monitoring bores in Syväjärvi rocks in Q1 of 2021 (Figure 18-83 copied from the 2022 draft DFS Report, Volume 4, Chapter 18). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 229 12.1.3 Design Acceptance Criteria Various industry Design Acceptance Criteria (DAC) cited in the Guidelines for Open Pit Slope Design (Read & Stacey, Guidelines for Open Pit Slope Design, 2009) were considered, from which Table 12-7 is copied. The wide range for probabilities of failure applied in industry on bench scale (25% ≤ probability of failure (PoF) ≤ 50% in Table 12-7), indicates the need for considering mining-area-specific and slope-specific factors in deciding on appropriate DAC. Table 12-7: Typical FoS and PoF acceptance criteria values for open-pit slopes at various scales proposed in Table 9.9 in the Guidelines for Open Pit Slope Design. Slope Scale Consequence of Failure Minimum Static FoS Maximum PoF Bench Low-high 1.1 25–50% Inter-ramp Low 1.15–1.2 25% Medium 1.2 20% High 1.2–1.3 10% Overall Low 1.2–1.3 15–20% Medium 1.3 5–10% High 1.3–1.5 ≤ 5% The consultants studied the July 2024 Syväjärvi PEA pit redesign, took account of the revised main ramp layout and proposed secondary access ramp, and applied engineering judgement to select DAC regarded as appropriate for each SDS. The following DAC were targeted for the FS design of the Syväjärvi pit: Bench scale: 1. PoF ≤ 25% DAC for slope sectors of mining areas which will be accessed with a single (main) ramp only. 2. 25% < PoF ≤ 35% DAC for slope sectors of mining areas which will be accessed and mined with dual (two, or more) ramps on opposite slopes. Adequate access redundancy justifies accepting higher PoF. (All slopes in the Syväjärvi FS pit carry ramps hence this criterion was applied throughout.) 3. PoF ≤ 45% DAC for slope sectors which will not carry long-term ramps. (All slopes in the Syväjärvi FS pit carry ramps hence this criterion is redundant.) Inter-ramp slopes: Minimum Factors of Safety FS ≥ 1.2 – 1.3 and/or maximum PoFs ≤ 10%, based on potential high consequences should an access ramp be disrupted, which is conservative considering the dual ramp access. Overall slopes: Minimum FS ≥ 1.5 and/or PoF ≤ 5%, which DAC are generally targeted when the consequences of failure may be high (refer to Table 12-7). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 230 12.1.4 Slope Stability Analyses The preliminary (2023) pit design was divided into six SDSs according to slope orientations (slope dip directions, SDDIRs). These SDSs were reviewed and refined as the pit design was upgraded (from the 2023 to the July 2024 PEA design). The July 2024 (and September 2024 FS) SDSs are shown in Figure 12-12. Reputable software DIPS® of the Rocscience® Group was used to assess and quantify the likelihoods of kinematic wedge, planar, and toppling failure mechanisms in each of these SDSs as pseudo PoFs. The probabilities of wedge failures and likely impacts thereof on bench backbreak were specifically analysed for selected SDS with Rocscience’s SWEDGE® (surface wedge) software. This was also used to determine practical and prudent berm widths (BWs). Groundwater was omitted from bench-scale kinematic analyses, because benches will generally be drained to the shallow depths of these failure mechanisms, due to gravitational stress relief and of blast damage opening structures in the near-slope rock. No bench-scale planar or wedge failure mechanisms which involve modelled FLTs were identified. Localised ravelling may occur in and near wider FLTs or brittle shear zones. Kinematic (DIPS®) and wedge (SWEDGE®) analyses of selected SDS were repeated and advanced when the SDS were modified based on the July 2024 PEA design. The results are summarised in Figure 12-7. Since the July 2024 PEA design incorporates dual access ramps on opposite slopes in the pit, the DAC explained above are applied in the colour scheme. All slope sectors are ramp-slopes in this design. Thus, PoF ≤ 35% are highlighted green and PoF > 35% are marked red. PoF ≤ 25% is coloured light green to emphasise very low PoF. 35% < PoF ≤ 40% are coloured orange to indicate further assessment may be warranted if BFA may be steepened. (Some rounding influences the orange cells.) PoF for flexural toppling is not rated (coloured grey) as this failure mechanism is regarded as unlikely, given the high intact rock strengths. Lastly, overall slope stability was confirmed with 2D Limit Equilibrium analyses using Rocscience’s SLIDE2® software. For these analyses, a conservative, very-close-to-the-slope groundwater table was assumed (refer to Figure 12-8 to Figure 12-11). The findings proved stability and confirmed results of Finite Element numerical modelling analyses which AFRY did in 2023.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 231 Pseudo-PoFs from (2nd Iteration) Stereographic Analyses of Kinematic Failure Mechanisms for BFA from 60° to 90° Figure 12-7: Pseudo-PoFs from (2nd iteration) stereographic analyses of kinematic failure mechanisms for BFAs from 60° to 90°. 90° 85° 80° 75° 70° 65° 60° 55° 50° 45° Wedge Sliding 65% 62% 57% 51% 42% 32% 20% 10% 4% 1% Planar Sliding (All) 51% 50% 47% 42% 33% 25% 18% 10% 2% 1% Direct Toppling (Intersection) 2% 2% 1% 1% 1% 0% 0% 0% 0% 0% Flexural Toppling (All) 1% 1% 1% 1% 1% 1% 1% 1% 1% 1% Wedge Sliding 3% 2% 2% 1% 1% 1% 0% 0% 0% 0% Planar Sliding (All) 7% 5% 4% 3% 3% 2% 2% 1% 1% 0% Direct Toppling (Intersection) 12% 11% 11% 11% 11% 11% 10% 10% 10% 9% Flexural Toppling (All) 42% 42% 42% 40% 36% 28% 16% 6% 3% 1% Wedge Sliding 10% 8% 7% 5% 4% 3% 3% 2% 2% 1% Planar Sliding (All) 20% 6% 5% 4% 4% 4% 3% 2% 2% 2% Direct Toppling (Intersection) 14% 14% 14% 14% 13% 13% 13% 13% 13% 13% Flexural Toppling (All) 23% 23% 22% 20% 17% 11% 5% 3% 1% 0% Wedge Sliding 27% 25% 22% 20% 17% 13% 11% 9% 7% 2% Planar Sliding (All) 38% 35% 24% 22% 22% 19% 19% 16% 14% 3% Direct Toppling (Intersection) 6% 5% 4% 4% 3% 3% 3% 3% 2% 2% Flexural Toppling (All) 8% 8% 8% 8% 5% 5% 3% 3% 3% 3% Wedge Sliding 37% 34% 30% 26% 22% 18% 14% 10% 6% 2% Planar Sliding (All) 18% 9% 6% 4% 3% 1% 1% 1% 0% 0% Direct Toppling (Intersection) 17% 12% 10% 9% 8% 7% 6% 5% 3% 2% Flexural Toppling (All) 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Wedge Sliding 8% 6% 6% 3% 3% 3% 0% 0% 0% 0% Planar Sliding (All) 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Direct Toppling (Intersection) 28% 22% 22% 22% 22% 22% 22% 22% 22% 22% Flexural Toppling (All) 11% 11% 11% 11% 11% 11% 11% 0% 0% 0% NORTH TRANSITION (Secondary Ramp) / 3A (Slope DDIR=155°±20°; lateral limit 20° SOUTH TRANSITION (Primary Ramps) / 3B (Slope DDIR=275°±20°; lateral limit 20° Domain/Sector/Failure Type Bench Face Angle EAST (Secondary Ramp) / 1A (Slope DDIR=255°±10°; lateral limit 20°) WEST (Primary Ramp) / 2A (Slope DDIR=085°±10°; lateral limit 20°) SOUTHWEST (Primary Ramps) / 2B (Slope DDIR=060°±5°; lateral limit 20°) SOUTH (Primary Ramps) / 2C (Slope DDIR=005°±5°; lateral limit 20°) S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 232 The Sensitivity of the Pseudo-PoF for Planar Sliding to BFA in SDS 1A, All Other Factors at Mean (Expected) Values Figure 12-8: The sensitivity of the pseudo-PoF for planar sliding to BFA in SDS 1A, all other factors at mean (expected) values. BFA ≤ 65° is recommended since this is a (secondary) ramp slop and considering the relatively wide variation in FO dip angle. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 233 The Sensitivity of the Pseudo-PoF for Wedge Sliding to BFA in SDS 1A, All Other Factors at Mean (Expected) Values Figure 12-9: The sensitivity of the pseudo-POF for wedge sliding to BFA in SDS 1A, all other factors at mean (expected) values. BFA ≤ 65° is recommended since for this (secondary) ramp slope, considering the relatively wide variation in the FO and ST (set 1 m) dip angles, but the scatter (lack of clustering in sets) of other structures (like JN and JN1) which may combine with set 1 m to form wedges. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 234 Syväjärvi East Slope Base Case Figure 12-10: Syväjärvi east slope base case set up in Slide2. OSA is 49°. Blue = mica schist; dark green = sulphidic schist; pink = plagioclase porphyrite. The blue line (W) is the simulated elevated groundwater profile applied in all analyses. FS ≈ 3.2. Syväjärvi West Slope Base Case Figure 12-11: Syväjärvi west slope base case set up in Slide2. Green = metavolcanite; dark green = sulphidic schist; pink = plagioclase porphyrite. The blue line (W) is the simulated elevated groundwater profile applied in all analyses. OSA is 42°. FS ≈ 8.0.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 235 12.1.5 Summary Slope Recommendations and the 2024 FS Pit Design BFAs were determined for 20 m-high benches to meet the target DAC. Pragmatic BWs were determined to catch 80% or more rubble from bench-scale planar, wedge, and/or toppling failures, plus allow safety margins against rock roll- out and scatter. Rubble from larger failures, although reckoned unlikely, can be caught on next-lower berms. The recommended slope configurations by SDSs (see Figure 12-12) are summarised in Table 12-8. VBKOM, with input from GTA, designed the Syväjärvi pit according to these parameters. Specific attention was paid with ramps planning to mitigate residual risks related to the modelled (inferred) FLTs. LOFTY and GTA reviewed the design. The August 2024 FS design is geotechnical compliant. The main ramp from which the Syväjärvi pit will be mined will be 28 m-wide, running north to south on the west and south-western slope sectors (SDS-2A, 2B, and 2C), with a switchback in the southern pit corner (SDS3A). The second south to north leg is on the lower west slope, with narrower (20 m-wide) sections around the lower north slope and along the bottom of the east slope to the final pit-bottom cut at -50 m reduced level (RL). The upper leg of the main ramp will inevitably intersect the shallow (±30°) south-westerly dipping inferred Syväjärvi FLTs 1, 2, and 3, but the slope orientations of SDS-2B and 2C result in favourable intersection angles. The lower section of the main ramp on the west slope is planned to be securely (± one BH) below the projected line of intersection of Syväjärvi FLT 2 (see Figure 12-13). A 16 m-wide secondary ramp is planned on the upper east slope and around the north end of the pit to terminate (or be mined out) on 20 m RL on the northern end of the west slope. This access redundancy will provide extra surety against production disruptions from rock falls onto or out of the main ramp, and an extra safety margin for vehicles, equipment, and personnel exit in case of emergency. The dual access ramps planned on opposite slopes of the pit allow slightly less strict DAC on bench scale. There are no OSA limitations since OSAs result from ISAs plus ramps and the ISA result from the bench-berm configurations. The OSA may reduce if geotechnical berms must be added. That would apply if the pit became deeper or should the east ramp be removed, so that a stack height exceeds 100 m. Analyses of the overall slopes with elevated groundwater tables simulated very close to the slopes confirmed insensitivity of these strong rock masses to pore-water pressures, at the designed OSA. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 236 Syväjärvi Final SDSs and SDDIR Ranges based on the August 2024 FS Pit Design Figure 12-12: Syväjärvi final SDSs and SDDIR ranges based on the August 2024 FS pit design. This includes minor refinements to the July 2024 design, which were the basis for second iteration slope stability analyses. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 237 Table 12-8: Slope design configurations, viz. BH, BFA, and BWs, and ISAs recommended for the Syväjärvi FS design. NB: Possible impacts of modelled FLTs Syväjärvi-FLT1, 2 and 3 on SDS-2B and SDS-2C ramp slopes must be mitigated with the pit design and/or managed during mining. Current estimated confidence levels in these 3D models are Inferred to Indicated. To be upgraded from quarry face mapping. Stack Height (m) Inter-Stack BW (m) 80 16 Slope Sector Location Refer to Figure CC Slope Dip Direction (°) Max. BH (m) Max. BFA (°) Min. BW (m) Max. ISA (°) Approx. Rock Crest RL (m) Approx. Toe RL (m) Rock OSH (m) Number Stacks Total ISB-Width (m) OVB Angle (°) 1A E Footwall 255 ± 10 20 65 8.5 48 75 -20 95 1.2 0.0 18.5 2A W Highwall 085 ± 10 20 80 8.5 59 80 -20 100 1.3 0.0 18.5 2B SW Highwall 060 ± 5 20 80 8.5 59 75 -40 115 1.4 0.0 18.5 2C S Highwall 005 ± 5 20 80 8.5 59 75 -40 115 1.4 0.0 18.5 3A N Transition 155 ± 20 20 75 8.5 55 75 0 75 0.9 0.0 18.5 3B S Transition 275 ± 20 20 75 8.5 55 75 20 55 0.7 0.0 18.5 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 238 August 2024 FS Design of the Syväjärvi Pit viewed to the South-West, Direction 255°, which is the Average Approximate Dip Direction of Inferred Syväjärvi FLTs SJF1, SJF2, and SJF3 Figure 12-13: The August 2024 FS design of the Syväjärvi pit viewed to the south-west, direction 255°, which is the average approximate dip direction of inferred Syväjärvi FLTs SJF1, SJF2, and SJF3 (numbered red drops). The dashed red lines show the projected lines of intersection of the inferred FLTs (or broken rock zones) with the west and north slopes. The numbered red drops point in the approximate south-westerly dip directions. SJF4 is near-vertical. 12.2 Rock Engineering – Rapasaari [§229.601(b)(96)(iii)(B)(13)(i)] The information for this section has been sourced from: • Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study for the Rapasaari Open Pit (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. Edited by Abraham Saayman, GEOTEC Africa CC. Updated geotechnical information has been obtained to enhance the geotechnical information presented as part of the 2022 FS, as described below. Independent review found that some aspects of the 2022–2023 geotechnical study for the Rapasaari pit design were not up to industry FS norms. Most notably, the frictional shear strength(s) of geological structures (natural defects), which largely dictate the bench-berm designs in such strong rock slopes, was not based on (empirical) estimation from available geotechnical core logs data, nor calculated from shear tests data. Field index measurements and/or laboratory DSTs of structure surfaces were lacking. The results of kinematic and wedge stability analyses, hence, bench-berm designs and the resulting ISAs were, thus, not to FS confidence levels.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 239 Key objectives with the 2024–2025 geotechnical study were to address these shortcomings and to upgrade other aspects to industry FS norms. The widely-applied handbook Guidelines for Open Pit Slope Design (Read & Stacey, Guidelines for Open Pit Slope Design, 2009), culmination of the multinational Large Open Pit Project coordinated by the Australian CSIRO, provided guidance and metrics for this. 12.2.1 Models and Data Compilation and Augmenting Comprehensive geological and geotechnical and oriented core logs data from several drill campaigns (from 2014 to 2022) provide very good 3D spatial coverage of the (2021 design) proposed larger and deeper Rapasaari open pit. The following summarise the main activities and processes undertaken in this FS upgrade stage to compile models and data: 1. Geological domains were defined based on 3D lithological and structural models developed by AFRY from geophysical surveys and from the geological and geotechnical core logs (AFRY , October 2020). The main rock types in which the slopes are to be developed are shown in Figure 12-16. 2. Review of 3D models of the Rapasaari structural geological modelling report (AFRY , October 2020) and high- level review of the 3D models of major structures were done. a. AFRY interpreted broken rock zones from RQD geotechnical core logs and from GSI values derived from RQD and Q-system geotechnical logs; and b. Apparent continuous, broken rock zones are inferred as possible geological FLTs or brittle shear zones, and five such FLTs/shears were modelled in 3D. These are shown in Figure 12-17. 3. Screened, with sophisticated tools like LEAPFROG®, all lithological, oriented structures, and geotechnical logs compiled in a comprehensive Access Database. a. Incorporate only rock fabric (minor geological structures) data within about 50 m (+50 m and -50 m) from the 2021 proposed pit slopes. The selected holes and core intervals are shown in Figure 12-14; b. This is to focus on data pertinent to the pit slopes and to minimise (irrelevant) data distant from the slopes, adding variability to and possibly diffusing or masking JN sets and/or rock fabric characteristics; c. Cross-coding of oriented core structures logs with rock types from the lithological logs; and d. Review core logs against core photos and screening out intact or annealed (closed) textures like FO, ST, and/or BD orientation measurements from the data, leaving only data of naturally open geological structures (i.e. rock fabric discontinuities) of the same, and of FLTs and JNs. 4. Detailed geotechnical relogging of the selected ±100 m-long core sections from ten (10) selected holes to remeasure structure orientations and characterise geotechnical (surface) properties of specific structure types according to standard Q-system parameters. These relogged holes are shown in Figure 12-15. 5. The geotechnical core log data sets were processed and analysed in detail to statistically determine rock quality parameters (Q-system parameters) of structure types in the main rock (lithology) types. a. Rock mass characteristics parameters RQD (displayed in Figure 12-14) and GSI values calculated from geotechnical core logs as part of the 2022–2023 rock mechanical assessment are representative of the total Rapasaari pit area, hence, also of the smaller sub-database proximate to the pit slopes. b. Q-system parameters Ja and Jr from the historic geotechnical core logs were applied to calculate IFAs of structures with the empirical function IFA ≈ ATan (Jr/Ja). This is useful (up to PFS level) as preliminary estimate of the frictional shear strength of defects, but IFA were not estimated and applied in previous geotechnical assessments. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 240 6. Selection for direct shear testing samples of specific structure types and intact rock of the main rock types in which the west and east pit slopes will be developed. For the 2024 upgrade of the Rapasaari FS: a. Saw-cut surfaces of 4 x Mica Schist (MS) and 4 x Metavolcanic (MV) intact rock samples (8 samples in total) were shear tested (3 x load cycles each test); b. DSTs of 26 x open structures [FO, ST, and JNs were direct shear tested, of which 12 were of MS and 14 were of MV (3 x load cycles each test)]; c. Samples were couriered to and tested by accredited E-Precision Laboratory in Perth, Western Australia. These were the first DSTs of geological structures done for the Rapasaari (and Syväjärvi) Projects. The sample numbers by rock and structure types are summarised in Table 12-10; d. The DST results were processed and analysed to derive shear strength input parameters specific to each of the defined Geological Domains. 7. Rock quality Q-parameters summarised from processed geotechnical logs and “residual friction angle” data from DST on intact rock samples, were applied in the BB empirical function to estimate friction angles (effective shear strengths) by structure type in the main rock types. Suggested friction-reduction structure scale (structure size) adjustments were not applied, firstly because: a. Even slight larger scale undulations, waviness, or curvature, will more than offset scale strength reduction factors; and b. No mapping data are available on structure lengths and shape (1st order roughness). Guesstimating “typical” lengths of structure types is not to FS reliability. Bench faces must be mapped during early pit development to provide this data. 8. The rock fabric datasets selected from oriented core were processed and analysed with stereograms to identify and define structure (JN) sets in each of the defined Geological Domains. 9. The hydrological, hydrogeological, and groundwater modelling sections of the 2022 draft FS Report (Volume 4, Chapters 18-19) were reviewed. The Rapasaari geohydrological data and models are based on investigations of: a. Trial pits in the till OVB; b. 21 groundwater observation wells installed from 2014 to 2020 in the Rapasaari area; c. Hydraulic conductivity field measurements of the till OVB with slug tests done in nine groundwater observation wells (Hydraulic conductivity of a rock mass is measured as groundwater flow rate K in m/s. Generally equivalent to permeability. Transmissivity of a formation is expressed as flow through a unit square area, as m2/s); d. Hydraulic conductivity (permeability) field measurements of (fractured) bedrock to depth of the then considered underground mine, by means of PFL measurements in five drillholes (RA-93, RA-145, RA- 155, RA-189, and RA-291), done in 2021 Q1. The drillholes were inter alia selected so that the hydraulic conductivity of the interpreted major structures (inferred FLTs) and the different rock types could be measured; e. Pumping tests and drawdown measurements performed around each of the five PFL-measured boreholes. These investigations and interpretation of results are described in detail in Appendix 1B- A1 of Vol 4 of the draft (January 2022) FS Report; f. Detailed spatial and statistical analyses of geomechanical RQD core logs by elevation (depth) and lithology, as indication of hydraulic conductivity and its spatial variance thereof; and S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 241 g. AFRY’s 2023 finite element numerical model with groundwater (pore pressures) was revisited to consider whether revision may be warranted. Geotechnical Core Logs (2014–2023) Limited to be within Maximum 50 m Distance (Inside and Outside) from the April 2024 Provisional Rapasaari Pit Slopes Figure 12-14: Geotechnical core logs (2014–2023) limited to be within maximum 50 m distance (inside and outside) from the April 2024 provisional Rapasaari pit slopes. Holes are coloured with the calculated RQD values. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 242 The Spatial Distribution of the Ten Selected Drill Cores which were Re-Logged, in Relation to the 2021 Design Version Rapasaari Open Pit Figure 12-15: The spatial distribution of the ten selected drill cores which were re-logged, in relation to the 2021 design version Rapasaari open pit. Holes are coloured with the logged lithologies. 12.2.2 Key Findings and Geomechanical Input Parameters 12.2.2.1 Lithological Models 1. The tillite OVB thickness varies from 1 to 33 m over the total Rapasaari Project area but is generally 9±2 m deep around the projected Rapasaari pit perimeter. Only shallow (<2 m thick) peat deposits occur sporadically in the open pit and the planned northern waste rock dump footprints. 2. The eastern pit slopes will be developed in metasedimentary rock, the predominant rock type in the Rapasaari terrain (mica schist, code MS). 3. The western pit slopes will be developed in meta-volcanics (named “Metavolcanite”, code MV). 4. Plagioclase porphyrite intrusive rock (the PP sill) and the pegmatite dykes and sills (various PG codes, the ore rock) occur mainly in the (folded) contact between the MS and MV, which strikes through the middle of the main (or northern) Rapasaari pit. The PP and PG intrusive bodies play a negligible role in slope design as these make up very small proportions of the lithologies in which the slopes will be developed. 5. The lithologies in which the pit slopes are to be developed and according to which the geological domains are defined are shown in Figure 12-16. No upgrades of the 3D lithological models were deemed necessary for geotechnical purposes.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 243 12.2.2.2 Major Structures Models 1. AFRY interpolated and extrapolated intersections of broken rock and zones of lower RQD and GSI values and developed 3D models of these as FLTs and/or brittle shear zones (October 2020), of which FLT-shear number two represents the fold limb contact between the MS and MV in the Rapasaari main pit. These (inferred) major structures models are shown in Figure 12-17. 2. Only a high-level independent overview of these Rapasaari major structures models was done to date. However, if findings of the more detailed review of Syväjärvi FLT models also apply to Rapasaari, then: a. The current Rapasaari major structures (FLT) models are likely (also) only at Inferred Confidence levels. SSW regard detailed review and upgrade of the Rapasaari major structures models as very important but currently not urgent; consequently, the upgrade will be done later. 3. The large-scale fold limb (as FLT-shear 2) interpreted and modelled through the Rapasaari main pit (the northern lobe) should have only minor (and manageable) influence on slope stability by virtue of its steep westerly dip and favourable, near-orthogonal, intersection with the northern and south-eastern pit slopes. 4. However, typical closer-spaced jointing (fractured rock) in a fold hinge zone can affect fragmentation and displacement (throw) from production blasting, hence, can have grade control implications. These considerations warrant further structural geological assessment of this major (closer fractured) geological structure. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 244 Rapasaari Provisional Open-Pit Design (April 2024) and Lithologies Cutting the Pit Slopes Figure 12-16: Rapasaari provisional open-pit design (April 2024) and lithologies cutting the pit slopes. Orange = spodumene pegmatite; pink = plagioclase porphyrite; dark green = sulphidic schist; light green = metavolcanite; blue = mica schist; grey = OVB. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 245 Zones with Lower RQD and Lower GSI Values Interpolated and Extrapolated as Broken Rock Zones, Inferred as FLTs or Brittle Shear Major Geological Structures Figure 12-17: Zones with lower RQD and lower GSI values interpolated and extrapolated as broken rock zones, inferred as FLTs or brittle shear major geological structures (named and numbered as “Rapasaari FLT-shear” (FS1 to FS5, or FLT-SHR1 to FLT-SHR5)). 12.2.2.3 Intact Rock Strengths, Discontinuity Shear Strengths, and Rock Mass Strengths 1. Intact rock strength data were obtained from the 2016 (first quantity numbers below) and the 2020 (second quantities below) laboratory strength testing programmes. Most of these samples were from the Rapasaari Project: a. 20 + 36 x density (SG) values, 56 in total; b. 20 + 36 x UCS values, 56 in total; c. 20 + 36 x Young’s Moduli (E) values, 56 in total; d. 20 + 36 x Poisson’s ratio values, 56 in total; e. 16 + 22 x Crack Initiation (CImin and CImax and Crack Damage data sets, 38 in total, although several results were noted as unreliable due to the failure mechanisms; and f. 20 + 36 x Tensile strength values (T) from Brazilian disk tests, 56 in total. A summary of the values for the UCS results by rock type is shown in Table 12-9. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 246 2. Laboratory shear tests of geological structures were lacking in the previous geotechnical studies. Samples of representative structures and rock types were selected from cores of the ten relogged geotechnical holes for geomechanical testing by accredited and highly sophisticated E-Precision Laboratory in Perth, Western Australia. 3. Twenty-six (26) samples of the dominant structure types in the two main rock types in which the Rapasaari pit slopes are to be developed were sent for DSTs. Normal stress ranges appropriate for shallow (shear) failures on structures in mining bench scale were specified to E-Precision Laboratory. These sample suites are summarised in Table 12-10. 4. Eight (8) intact rock samples of the same two rock types were also sent for DST on saw-cut surfaces to derive residual (or basic) friction angles. These are summarised in 5. Table 12-11. 6. Peak friction angles of structures were derived by analysing these DST results by structure types and by rock types. The results are summarised in Table 12-12. These show negligible differences in friction angles among FO, JN, and ST planes. 7. For this reason, friction angles were applied by main rock types or geological domains. Even these Phi differ only by one degree between the MS (west) and MV (east) domains – refer to Table 12-13. 8. Friction angles of structures were also estimated with the industry-accepted BB empirical function (Barton & Chaubey, 1977) (Barton, 1987). Key input parameters to the BB-function are JRCs and JCSs (Barton & Bandis, 1990). a. JRC were derived from geotechnical core logs of 10 holes core intervals which were relogged in 2024; and b. JCS were estimated by downgrading UCS values from the (2016 and 2020) laboratory tests (Table 12-9) to account for weathering of structure surfaces, using factors in a nomogram suggested by Stacey and Page (1986). 9. Friction angles of structures from DST results corelate accurately with friction angles estimated with the BB empirical method from geotechnical core log data. Refer to Table 12-13. 10. The laboratory-measured friction angles (Phi) were applied in kinematic and limit equilibrium slope stability analyses: a. Phi = 37° ± 4° is applied in Mica Schist (MS); and b. Phi = 36° ± 3° is applied in Metavolcanics (MV). 11. Rock mass quality by rock type, quantified with GSI values, derived from geotechnical drill core logs, show that rocks are generally massive (intact) or blocky, with wide JN spacings. Statistics show that all the rock types have lower quartile values GSI > 70, and plagioclase porphyrite has GSI > 80 (refer to Figure 12-18 for a graphic summary of results). 12. In combination with high intact rock strengths (50 MPa < UCS < 250 MPa), these high GSI values denote strong rock masses. Slope stability will consequently be dictated by the rock fabric (“minor” structures like FO, ST, and JN planes) on bench scale, and rock mass failures will be highly unlikely. The 2024–2025 upgraded geotechnical assessment accordingly focussed on defining structures sets and applying representative frictional shear strengths of structures to analyse the probabilities of structurally controlled planar, wedge, and toppling failure mechanisms on bench scale.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 247 Table 12-9: UCS laboratory test results. Rock Codes Sample No. Average UCS (MPa) Std Dev. UCS (MPa) MS 6 108 44 MV 9 140 43 PG 8 158 47 PP 7 211 26 SS 3 155 20 Table 12-10: Summary of the direct shear tested sample numbers by rock type and structure type. Lithology JN Type Number of Samples Mica schist Total 16 JN 7 ST 5 Saw-cut 4 Metavolcanite Total 18 JN 7 FO 7 Saw-cut 4 Table 12-11: Residual friction angles derived from saw-cut rock samples, after applying Hencher’s correction. E-Precision Laboratory calculated Phir by Mohr-Coulomb curve fit to results of three test stages at increasing normal stresses. Rock Codes Average Residual Friction Angles φres Std Dev. Residual Friction Angles φres MS 29° ±4° MV 29° ±1° Note: All the values are corrected in accordance with Hencher’ s proposed method. This compensates for effects of dilation when top and bottom halves of a sample rides over asperities during shearing. Table 12-12: Peak friction angles derived from DST on structure types in four rock types, after applying the Hencher correction. FQ JN ST Rock Type Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Total Count of S-DST MS 21 39 3 15 34 2 36 MV 21 35 2 21 36 3 42 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 248 FQ JN ST Rock Type Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Count of S-DST Average of Hencher Phi Std Dev. of Hencher Phi Total Count of S-DST Grand Total 21 35 2 42 37 3 15 34 2 78 Note: Sample numbers = Count of S-DST/3. Table 12-13: Mean and standard deviations of peak friction angles by rock type derived from DST, after applying Hencher’s correction, compared with frictional angles estimated with the BB criterion at normal stress 400 kPa, to approximately match the mean normal stress applied in the DST (left three columns). Rock Codes From DST Applied Normal Stress Range σn (kPa) Peak Friction Angle from DST (°) BB Calculation Min. Avg. Max. Avg. Std Dev. Estimated Friction Angle (°) MS 102 390 813 37 4 41 MV 101 385 814 36 3 41 Rapasaari Drill-Hole-Based GSI Values of the Main Rock Types (Sill = Plagioclase Porphyrite) Figure 12-18: Rapasaari drill-hole-based GSI values of the main rock types (sill = plagioclase porphyrite). The mean values are indicated by the red diamonds. The median values are indicated by the lines that cross inside of the coloured boxes. The box encloses the interquartile range around the median. The whiskers extend out to vertical lines that mark the extreme value extents. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 249 12.2.2.4 Rock Fabric Sets in SDSs The prominent rock fabric sets appeared to be spatially ubiquitous, with similar orientations throughout the eastern (MS) and western (MV) geological domains. This obviates the need to subdivide the geological domains into smaller geotechnical domains. The mean and standard deviation dip directions and dip angles of the defined structures sets are summarised in Table 12-14. 1. The MS and MV are foliated schistose rocks, but oriented core structures data show that ST and FO planes are sparser than JNs3 hence should play lesser role in east slope bench stability than foreseen from previous assessments, which were based on the total (unsegregated) Rapasaari database. 2. Moderately steep south-westerly dipping, predominantly JNs, in two prominent sets plus abundant randomly- oriented JNs will largely determine (dictate) the east slope in MV configurations, through planar and wedge failure mechanisms. 3. Jointing in the western MV geological domain is sparser (widely to very widely spaced) and the few FO planes which may be developed dip favourably south-westwards into the slope. Minor toppling may be the primary failure mechanism. This favours the western slopes for placing long-term ramps. 4. Less prominent sets (low-density clusters containing few structures) occur in some SDSs and were accounted for in stability analyses. All structures in defined sets plus all “random JNs” i.e. structures not grouped into the defined sets, were considered and accounted for in kinematic analyses of each SDS. Table 12-14: Summary of the discontinuity sets identified in the Rapasaari open pit in the defined SDSs. Domain Slope Sector JN Set Dip Direction (°) Dip (°) East Main Pit 1A JS1 315 34 JS2 181 55 JS3 214 42 1B JS1 189 49 JS2 227 54 JS3 258 67 JS4 006 63 1C JS1 237 65 JS2 203 50 JS3 113 6 1D JS1 286 82 JS2 291 4 JS3 197 41 JS4 136 90 3 It is possible that core loggers misnamed FO and ST as JN, but this is unimportant as the same friction angle was found for FO, ST, and JN from DST, and applied in kinematic analyses. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 250 Domain Slope Sector JN Set Dip Direction (°) Dip (°) 1E JS1 310 90 JS2 210 82 JS3 334 14 West Main Pit 2A JS1 234 65 JS2 103 43 JS3 339 12 JS4 269 74 2B JS1 197 50 JS2 250 82 JS3 306 85 JS4 200 16 Satellite Pit 3A JS1 299 81 JS2 142 2 JS3 236 41 3B JS1 206 50 JS2 337 50 JS3 263 90 12.2.2.5 Hydrogeological Parameters and Groundwater Models 1. Observations in observation bores indicate a generally very shallow groundwater level, approximately 0.3 to 3 m below the ground surface in peat and tillite OVB. 2. Slug tests in tillite OVB yielded low hydraulic conductivity (K) values, viz. K = 6.3x10-7 m/s to K = 1.5x10-6. 3. Hydrogeological flow rate measurements in deep bores showed that the hydraulic conductivity of the bedrock is greater nearer surface and decreases at depth – see Figure 12-19. a. Hydraulic conductivity below vertical depth of about 50 masl are generally very low, with from K ≈ 1.5 x 10-9 to K≈1.5 x 10-07; b. K ≈ 1 x 10-8 equates flow rate of groundwater through the rock mass of about 3.2 metres/year, thus extremely slow; c. These very low permeability values imply that passive drawdown of groundwater around the pit will be extremely slow. Hence, that the drawdown “cone” around the pit will generally be steep and be close to the slopes; and d. This motivates for analysing overall slope stability with elevated groundwater levels close to the pit slopes, which is detailed in the Syväjärvi DFS geomechanical report. 4. No unique or anomalously high or low hydraulic conductivity (K) values were found for any specific main lithology. K-values of the pegmatites are generally the lowest at K ≈ 1.5 x 10-9. 5. Very low transmissivity values (refer to Figure 18-70 in the 2022 draft FS Report, Volume 4, Chapter 18) indicate that very low groundwater inflow volumes into the pit can be expected, except for potential

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 251 temporary and localised slightly higher inflows from some major geological structures (the broken rock zones) and through closely jointed pegmatite intrusions, where such intersect the slopes. 6. Good correlation hydrogeological field measurements and the derived hydraulic conductivities of bedrock were found with RQD values. a. Lower RQD values of more fractured rock zones match higher hydraulic conductivities; and b. This implies localised drawdown of groundwater level and localised depressurisation of slopes and structures should occur where the lower RQD broken rock zones intersect the slopes. Hydraulic Conductivity, K, in m/s, Measured in Monitoring Bores in Rapasaari Rocks in Q1 Of 2021 Figure 12-19: Hydraulic conductivity, K, in m/s, measured in monitoring bores in Rapasaari rocks in Q1 of 2021 (Figure 18-73 copied from the 2022 draft DFS Report, Volume 4, Chapter 18). 12.2.3 DAC and Slope Stability Analyses Various industry DAC cited in the Guidelines for Open Pit Slope Design (Read & Stacey, Guidelines for Open Pit Slope Design, 2009), from which Table 12-15 is copied], were considered. Wide range probabilities of failure applied in industry on bench scale (25% ≤ PoF ≤ 50% in Table 12-15), indicate consideration of mining area specific and slope- specific factors when deciding appropriate DAC. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 252 Table 12-15: Deterministic (FoS) and probabilistic (PoF) DAC from bench to overall slope scales proposed in Table 9.9 of the Guidelines for Open Pit Slope Design (Read & Stacey, 2009). Slope Scale Consequence of Failure Minimum Static FoS Maximum PoF Bench Low-high 1.1 25–50% Inter-ramp Low 1.15–1.2 25% Medium 1.2 20% High 1.2–1.3 10% Overall Low 1.2–1.3 15–20% Medium 1.3 5–10% High 1.3–1.5 ≤ 5% The consultants studied the July 2024 RA PEA pit redesign, specifically the proposed ramps layouts, and applied engineering judgement to select DAC regarded as appropriate for each SDS. Based on the PoF ranges suggested in Table 12-15 and in Table 9-10 in the Guidelines, the following DAC were targeted for FS design of the Rapasaari pit: Bench scale: 1. PoF ≤ 25% DAC for slope sectors of mining areas which will be accessed with a single (main) ramp only. 2. PoF ≤ 35% DAC for slope sectors of mining areas which will be accessed and mined with dual (two, or more) ramps on opposite slopes. Adequate access redundancy justifies accepting higher PoF. 3. PoF ≤ 45% DAC for slope sectors which will not carry long-term ramps. A second criterion applied in each case when assessing the results of stability analyses is the likely backbreak widths of probable bench-scale failures compared to the ramp and/or BWs. Inter-ramp slopes: Minimum Factors of Safety FS ≥ 1.2–1.3 and/or maximum PoFs ≤ 10%, based on potential high consequences should an access ramp be disrupted, which is conservative when considering the dual ramp access. Overall slopes: Minimum FS ≥ 1.5 and/or PoF ≤ 5%, which DAC are generally targeted when the consequences of failure can be high (refer to the bottom row in Table 12-15). 12.2.4 Slope Stability Analyses and Key Results The preliminary (2023) pit design was divided into nine SDSs according to slope orientations (SDDIRs). These SDSs were refined as the pit design was upgraded (from the 2023 to the July 2024 PEA design). The July 2024 (and September 2024 FS) SDS are shown in Figure 12-30. Highly regarded and widely used software DIPS® of the Rocscience® Group was used to quantify the likelihoods of kinematic wedge, planar, and toppling failure mechanisms in each of these SDSs as pseudo-PoFs. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 253 The probabilities of wedge failures and likely impacts thereof on bench backbreak were specifically analysed for selected SDS with Rocscience’s SWEDGE® (Surface Wedge) software. This was also used to determine practical and prudent BWs. Groundwater was deliberately omitted from bench-scale kinematic analyses, because benches will generally be drained to the shallow depths of these failure mechanisms, due to gravitational stress relief and of blast damage opening structures in the near-slope rock. No bench-scale planar or wedge failure mechanisms which involve modelled FLTs were identified. Localised ravelling may occur in and near wider FLTs or brittle shear zones. The results of first-iteration stereographic kinematic bench stability analyses are summarised in Figure 12-20 and Figure 12-21. Stereographic kinematic (DIPS®) analyses were repeated and wedge (SWEDGE®) analyses advanced for SDS-1A and SDS-1B because the sector boundaries and slope directions were modified based on the July 2024 PEA design. Results of DIPS® analyses of SDSs SDS-1A and SDS-1B are shown in Figure 12-22 to Figure 12-25. PoF for Range of BFA from DIPS® Kinematic Stability Analyses for the RA East Slope SDS-1A to SDS-1E Figure 12-20: PoF for range of BFA from DIPS® kinematic stability analyses for the RA east slope SDS-1A to SDS-1E. Red cells exceed the target DAC by all criteria; green cells meet the target DAC by all criteria; yellow cells exceed but are within 5% of SDS- specific target criteria and warrant further assessment. 90° 85° 80° 75° 70° 65° 60° Wedge Sliding 44% 42% 40% 38% 35% 32% 28% Planar Sliding (All) 58% 53% 47% 44% 41% 37% 33% Direct Toppling (Intersection) 3% 3% 3% 2% 2% 1% 1% Wedge Sliding 56% 53% 50% 46% 42% 38% 33% Planar Sliding (All) 21% 21% 20% 18% 17% 16% 13% Direct Toppling (Intersection) 2% 1% 1% 1% 1% 1% 1% Wedge Sliding 20% 17% 14% 12% 9% 7% 5% Planar Sliding (All) 50% 38% 21% 11% 7% 4% 2% Direct Toppling (Intersection) 10% 9% 7% 5% 3% 2% 1% Wedge Sliding 34% 30% 25% 20% 15% 11% 8% Planar Sliding (All) 48% 37% 23% 16% 12% 8% 6% Direct Toppling (Intersection) 3% 2% 2% 1% 1% 1% 1% Wedge Sliding 22% 16% 11% 8% 6% 5% 4% Planar Sliding (All) 28% 17% 12% 8% 6% 5% 4% Direct Toppling (Intersection) 17% 16% 14% 13% 12% 11% 9% SOUTH PIT SOUTHEAST - 1E (Slope DDIR=315°; lateral limit 20°) - 1 Ramp Domain - Slope Sector / Failure Type Bench Face Angle MAIN PIT NORTH - 1A (Slope DDIR=195°; lateral limit 20°) - 2 Ramps MAIN PIT EAST -1B (Slope DDIR=215°; lateral limit 20°) - 2 Ramps MAIN PIT SOUTHEAST - 1C (Slope DDIR=305°; lateral limit 20°) - 2 Ramps SOUTH PIT EAST - 1D (Slope DDIR=255°; lateral limit 20°) - 1 Ramp S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 254 PoF for Range of BFA from DIPS® Kinematic Stability Analyses for the RA West Slope SDS-2A and SDS-2B, and of the Small NW Satellite Pit Figure 12-21: PoF for range of BFA from DIPS® kinematic stability analyses for the RA west slope SDS-2A and SDS-2B, and of the small NW satellite pit. Red cells exceed the target DAC by all criteria; green cells meet the target DAC by all criteria; yellow cells exceed but are within 5% of SDS-specific target criteria and warrant further assessment. Sensitivity Analyses of the (Pseudo-) Probability of Planar Sliding versus Slope Dip (BFAs from 50° to 90°) in the Average SDDIR of 180° of SDS-1A, All Other Factors Kept Constant, show PoF ≈ 15% for BFA = 70° Figure 12-22: Sensitivity analyses of the (pseudo-) probability of planar sliding versus slope dip (BFAs from 50° to 90°) in the average SDDIR of 180° of SDS-1A, all other factors kept constant, show PoF ≈ 15% for BFA = 70°. 90° 85° 80° 75° 70° 65° 60° Wedge Sliding 17% 14% 11% 9% 7% 6% 4% Planar Sliding (All) 27% 22% 18% 16% 14% 13% 10% Direct Toppling (Intersection) 8% 7% 6% 5% 5% 4% 4% Wedge Sliding 21% 17% 14% 11% 9% 7% 5% Planar Sliding (All) 32% 20% 12% 8% 5% 3% 2% Direct Toppling (Intersection) 11% 8% 7% 5% 4% 3% 3% Wedge Sliding 11% 7% 3% 2% 1% 1% 0% Planar Sliding (All) 17% 13% 5% 3% 1% 1% 1% Direct Toppling (Intersection) 9% 6% 4% 3% 3% 3% 2% Wedge Sliding 17% 14% 12% 11% 9% 7% 7% Planar Sliding (All) 43% 29% 19% 10% 10% 10% 10% Direct Toppling (Intersection) 11% 9% 7% 5% 4% 3% 2% SATELLITE PIT SOUTHEAST - 3B (Slope DDIR=305°; lateral limit 20°) - 1 Ramp Domain - Slope Sector / Failure Type Bench Face Angle MAIN PIT WEST - 2A (Slope DDIR=070°; lateral limit 20°) - 1 Ramp SOUTH PIT WEST - 2B (Slope DDIR=110°; lateral limit 20°) - No Ramp SATELLITE PIT NORTHWEST - 3A (Slope DDIR=125°; lateral limit 20°) - 1 Ramp

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 255 Sensitivity Analyses of the (Pseudo-) Probability of Wedge Sliding versus Slope Dip (50° ≤ BFA ≤ 90°) in the Average SDDIR of 180° of SDS-1A, All Other Factors Kept Constant, show PoF ≈ 28% for BFA = 70° Figure 12-23: Sensitivity analyses of the (pseudo-) probability of wedge sliding versus slope dip (50° ≤ BFA ≤ 90°) in the average SDDIR of 180° of SDS-1A, all other factors kept constant, show PoF ≈ 28% for BFA = 70°. Sensitivity Analyses of the (Pseudo-) Probability of Planar Sliding versus Slope Dip (BFAs from 50° to 90°) in the Average SDDIR of 215° of SDS-1B, All Other Factors Kept Constant, show PoF ≈ 17% for BFA = 70° Figure 12-24: Sensitivity analyses of the (pseudo-) probability of planar sliding versus slope dip (BFAs from 50° to 90°) in the average SDDIR of 215° of SDS-1B, all other factors kept constant, show PoF ≈ 17% for BFA = 70°. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 256 Sensitivity Analyses of the (Pseudo-) Probability of Wedge Sliding versus Slope Dip (50° ≤ BFA ≤ 90°) in the Average SDDIR of 180° of SDS-1A, All Other Factors Kept Constant, show PoF ≈ 42% for BFA = 70°. This was Further Analysed with SWEDGE® Figure 12-25: Sensitivity analyses of the (pseudo-) probability of wedge sliding versus slope dip (50° ≤ BFA ≤ 90°) in the average SDDIR of 180° of SDS-1A, all other factors kept constant, show PoF ≈ 42% for BFA = 70°. This was further analysed with SWEDGE®. The high pseudo-PoF for wedge failures in SDS-1B prompted more detailed sensitivity analyses with SWEDGE®. First results and various statistics (displayed and studied in histograms and scatter plots) showed that the PoF at targeted BH = 20 m, BFA = 70°, and BW = 10 m, is largely due to high proportion of insignificant small wedges. If multiple scaling parameters are entered in SWEDGE®, only the scaling parameter which gives the smallest wedge will be applied for a given wedge. All other scaling values will be ignored. For this reason, it was decided to apply to minimum wedge weight (or volume) as only definable in practice and defensible criterion in the absence of JN length data. This is also a practical consideration for berm design and for operations. The BFA ≤ 60° indicated by the 1st iteration kinematic and wedge analyses is a difficult bench angle to mine and can generally not be presplit-blasted. The minimum BFA at which 20 m-long blastholes can be drilled for presplit blasting is 65°, in the consultants’ experience. Even this BFA is difficult to drill accurately, and hole blockages can be common. Such flat BFA requires short stab (infill) holes in trim or buffer blasts between the first row of full BH holes and the bench crest, to achieve reasonable fragmentation. Stab holes can often result in extensive damage to bench faces. The PoF reduces substantially, by 10% from PoF ≈ 36% to PoF ≈ 26% for BFA = 70°, if the minimum significant wedge size is increased from 2 m3 to 5 m3. Key inputs and results of these cases are compared in Figure 12-26 and Figure 12-27. A rockfall of 5 m3 would entail only about 14 tonnes of rubble, which can be easily pushed off the ramp or loaded away with a medium front-end loader (e.g. in the CAT 980 XE class). Accepting or discounting such minor rockfalls to attain the targeted BFA = 70° bench design can thus be justified. This is a practical bench angle for presplit and wall-control trim or buffer blasting. Even narrow berms would catch such small wedge failures, but conservative S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 257 10 m-wide berms are advised to allow flexibility to lower BFA in SDS-1A and SDS-1B if proved desirable or necessary during mining. Lastly, inter-ramp and overall slope stability was confirmed with 2D Limit Equilibrium analyses using Rocscience’s SLIDE2® software. For these analyses, conservative high and close-to-the-slope groundwater table were assumed. Figure 12-28 and Figure 12-29 show the analysed cross-sections with the simulated conservative groundwater table and the resulting factors of safety. All FSs were well below the target DAC for the overall slopes and confirmed the insensitivity to groundwater of the slopes in these strong rocks. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 258 The Probability of Wedge Failures in Rapasaari SDS-1B Nominally Reduces from ±44% to PoF ≈ 36% if the Minimum Wedge Size to Consider as Relevant is Increased from 1 m3 to 2 m3, or from 0.026 MN To 0.052 MN Figure 12-26: The probability of wedge failures in Rapasaari SDS-1B nominally reduces from ±44% to PoF ≈ 36% if the minimum wedge size to consider as relevant is increased from 1 m3 to 2 m3, or from 0.026 MN to 0.052 MN.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 259 The Probability of Wedge Failures in Rapasaari SDS-1B Further Reduces (Nominally) from ±36% to PoF ≈ 32% if the Minimum Wedge Size to Consider as Relevant is Increased from 2 m3 to 5 m3, or from 0.052 MN to 0.13 MN, which is Still Quite Small Figure 12-27: The probability of wedge failures in Rapasaari SDS-1B further reduces (nominally) from ±36% to PoF ≈ 32% if the minimum wedge size to consider as relevant is increased from 2 m3 to 5 m3, or from 0.052 MN to 0.13 MN, which is still quite small. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 260 The Rapasaari East Slope Model in Slide2®, Analysed with Central Case Strength Parameters Figure 12-28: The Rapasaari east slope model in Slide2®, analysed with central case strength parameters. Blue = mica schist; red = pegmatite. The blue line (W) is the simulated elevated groundwater profile, applied in all analyses. Central case FS is slightly above 2.0. The Rapasaari West Model in Slide2®, Analysed with Central Case Strength Parameters Figure 12-29: The Rapasaari west model in Slide2®, analysed with central case strength parameters. Blue = mica schist; green = vulcanite; pink = plagioclase porphyry; red =pegmatite. The blue line (W) is the simulated elevated groundwater profile, applied in all analyses. Central case FS is above 4.0 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 261 12.2.5 Summary Slope Recommendations and Review of the September 2024 FS Design The final recommended slope configurations by SDSs (see Figure 12-30) are summarised in Table 12-16. VBKOM, with input from GTA, designed the Rapasaari pit according to these parameters. Specific attention was paid in ramps planning to mitigate residual risks related to the modelled (inferred) FLTs. The Rapasaari main pit i.e. the main northern sector, will be accessed and mined from a 28 m-wide ramp on the western slope (in SDS-2A), as well as a 28 m-wide ramp entering on the north slope (SDS-1A) and running along the northeastern (SDA-1B) and eastern (SDS-1C) SDSs (see Figure 12-30). This access redundancy, and complementary safety benefits, allowed some leeway in strictness of DAC. Kinematic PoF, also tested in detail with discrete wedge analyses, of SDS-1A, 1B, and 1C met the PoF ≤ 35% criterion. The PoF found for the west slope SDS-2A is much lower and comfortably met DAC of PoF ≤ 25%. The PoF of SDS-1D, 1E, of SDS-2B, and SDS-3A and 3B of the small satellite pit in the northwest, all meet the DAC of PoF ≤ 25% on bench-scale for the configurations recommended in Table 12-16. The LE analyses results yielded Central Case Factors of Safety well above the DAC of FS ≥ 1.5 and lower-case FS ≥ 1.3 for overall slopes, even with elevated groundwater tables. There are no OSA limitations since OSA in strong rock slopes result from ISAs plus ramps and the ISA result from the bench-berm configurations. The OSA may reduce if geotechnical berms must be added. That would apply if a stack height should exceed 100 m. Flexibility of +20 m is allowed in maximum stack height, as such changes are likely to occur only near the end of mining operations if flats may be mined out to gain more ore. It is important to consult the notes following Table 12-16. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 262 The Final Rapasaari SDSs Based on SDDIR Ranges Scaled from the (Shown) July 2024 Pit Design Figure 12-30: The final Rapasaari SDSs based on SDDIR ranges scaled from the (shown) July 2024 pit design. 4 0 0 m

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 263 Table 12-16: Slope design configurations recommended for the Rapasaari DFS pit optimisation and design. SDS locations refer to Figure 12-30. NB: Possible impacts of modelled FLTs RA-FLT1 to FLT5 on ramp slopes must be mitigated with the pit design and/or managed during mining. Current estimated confidence levels in these 3D models are undefined but are likely Inferred. AFRY to review, upgrade 3D models, and qualify confidence levels and residual risks. Max. Stack Height (m) Inter- Stack BW (m) 80 16 Slope Sector Number SDDIR (°) BH (m) BFA (°) 2023 Min. BW (m) Max. ISA (°) Approx. Rock Crest RL (m) Approx. Toe RL (m) Rock OSH (m) Number Stacks Total ISB- Width (m) Inter- stack Berm RL RAMP 1 Width (m) Ramp 2 Width (m) Rounded Rock OSA (°) OVB Angle (°) 1A 180 ± 20 20 70 10.0 49 85 -30 115 1.4 0.0 N/A 32.0 20.0 37 18.5 1B 215 ± 20 20 70 10.0 49 85 -90 175 2.2 0.0 N/A 32.0 20.0 41 18.5 1C 295 ± 30 20 75 10.0 52 85 -80 165 2.1 0.0 N/A 28.0 20.0 43 18.5 1D 255 ± 20 20 75 10.0 52 80 -20 100 1.3 0.0 N/A 28.0 44 18.5 1E 315 ± 30 20 80 10.0 56 80 -20 100 1.3 0.0 N/A 28.0 46 18.5 2A 070 ± 30 20 80 10.0 56 80 -90 170 2.1 0.0 N/A 28.0 50 18.5 2B 110 ± 30 20 80 10.0 56 75 -20 95 1.2 16.0 0.0 50 18.5 3A 125 ± 30 20 80 8.5 59 75 10 65 0.8 0.0 N/A 20.0 48 18.5 3B 305 ± 30 20 80 8.5 59 75 10 65 0.8 0.0 N/A 20.0 48 18.5 See next page for important notes. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 264 12.2.5.1 Important Notes Regarding Application of the Slope Design Parameters: 1. The bench and berm configurations must take precedence over other parameters. 2. Numbers of ramps in SDSs and layouts are based on the 2024-07-24 pit design. 3. The 32 m-wide ramp legs advised in SDS-1A and SDS-1B are to minimise impact on traffic from backbreak of possible wedge failures. a. These two ramp legs may, at the discretion of the mine planning engineers, be designed at the standard 28m-width; If b. Approximate synchronised development to about 20 m RL of the access ramps in east SDS-1A and 1B, and in west SDS-2A, will provide access redundancy. 4. The 16.0 m-wide inter-stack safety berm advised on 0 m RL in SDS-2B may be reconsidered and omitted if good rock conditions and slope monitoring justify this when mining reaches 20 m RL. 5. Short and/or narrow ramps for ore routing in the pit bottom are excluded from OSA calculations. 12.2.6 Key Opportunities and Primary Risks The 10 m-wide design BWs in the western (highwall) SDSs, SDS-2A and SDS-2B, may prove to be conservative. This is partly to compensate for the pronounced directional bias in rock fabric data due to preponderance of exploration drilling in north-easterly direction (refer to Figure 12-14). Additional drilling data and/or on face mapping data may prove it feasible to reduce the BWs which can present opportunity for steepening the inter-ramp and overall west slopes. The current low confidence level in 3D major structures models, and the possible occurrence of yet undefined FLTs, may pose risks to slope stability at inter-ramp slope scale. One hundred and ninety (190) more diamond cored holes (numbers RA-289 to RA-466 plus twelve RAPI- diamond core holes) were drilled from 2021 to 2023 to further explore the Rapasaari deposit. These were drilled after completion of the (October 2020) structural geology report and 3D models. Data from geological and geotechnical logs of these additional holes can provide a valuable additional basis for first-iteration review of the 3D models. The 3D broken rock zones or FLTs must be reviewed in detail and confidence ratings assigned. All available data must be reviewed and logged FLTs must be assessed in cores. It may prove necessary to do targeted core drilling to validate specific models and/or upgrade FLT sections to Indicated or Measured confidence levels near the projected pit slopes. The pit design can then be reviewed to address possible adverse FLT intersections. Typical closer-spaced jointing and sheared FO planes (fractured rock) along the fold limb which strikes north-north- westerly through the Rapasaari main pit (FLT-shear model RA-FS2) can affect fragmentation and throw from production blasting. Most pegmatite orebodies are along this major structure. Excessive muck movement can cause ore and waste mixing, hence, grade control complications. These considerations warrant further structural geological assessment and rock quality characterisation of this major geological structure. A multi-disciplinary opportunities and risks workshop is advised. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 265 12.3 Hydrogeology and Hydrology [§229.601(b)(96)(iii)(B)(7)(iii)] All the deposits are located within the bedrock of volcanic and metamorphosed lithological units with low hydraulic conductivity. Higher hydraulic conductivities are associated with bedrock fracturing and faulting. RQD data analysis suggests the Rapasaari and Syväjärvi rock mass is more intensely fractured in the upper part (above 50 mamsl) and less so at depth. The OVB at all the ore deposit sites contains till and peat of varying thickness. Field hydraulic testing and water level observations completed thus far are concentrated on the Rapasaari and Syväjärvi mine sites. Slug testing to determine the hydraulic conductivity of the OVB till layer was carried out at Rapasaari and Syväjärvi. The results from the two sites are of the same order of magnitude with an average of 6.3 to 7.7 x 10-7 m/s, which is a relatively low hydraulic conductivity. The RQD data were used as a proxy for hydraulic conductivity by establishing a correlation with the hydraulic conductivity measurements. The approach followed seems reasonable, although a clearer description of the methodology and derivation of parameters from flow logging and RQD is required. The water table is shallow and close to surface. Recharge from precipitation is assumed to be relatively high at 50% of precipitation. Most of the recharge is assumed to flow laterally in the topmost surficial OVB layer. The interaction between surface water bodies and the groundwater is unknown; however, it is clear that the OVB plays an important role in conveying recharge to local streams and lakes that are fed by groundwater. 12.3.1 Groundwater Inflows The groundwater inflows into the various mines were estimated using a numerical groundwater model for Rapasaari and Syväjärvi (Table 12-17). The inflows are less than 710 m3/d – the reported, relatively low rate of inflow should not pose a material challenge to mining. The estimates are, however, preliminary but seem reasonable if the hydraulic conductivity is as low as reported. These estimates, hydraulic conductivity and inflows, will need to be updated with site-specific hydrogeological data as these are obtained to meet licencing requirements. The inflows at Rapasaari, however, are expected to peak at approximately 2,035 m3/d and may pose a risk to mining. Active dewatering through, for example, dewatering wells located along the pit perimeter will be required. Not adequately providing and planning for such may cause delays and have a severe impact on mining progress and safety. For the Syväjärvi open pit, a cut-off drain will limit the flow from the upstream catchment to the dewatered Syväjärvi Lake. An embankment is constructed to prevent flow between lakes, and the water table will be maintained at a low level through active dewatering (refer to Table 12-17). Table 12-17: Summary of groundwater inflows per deposit. Deposit LOM Open-Pit Depth Inflows (m3/d) Drawdown Rapasaari 14 (yr 0 to 14) 130 m (-40 masl) ~ 2,035 at year 2.7 Limited drawdown, extends to the edge of Vionneva Natura Syväjärvi 4 (yr 0 to 4) 100 m (-5 masl) ~710 Few hundred metres from pit S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 266 12.3.2 Water Quality The below is summarised from the Keliber Water Management Plan (AFRY, 2021). Generally, the different waters at Rapasaari-Päiväneva are slightly saline and mostly also slightly alkaline. Alkali and alkaline earth metals and aluminium are the cations present in the highest concentrations. Sulphate, chloride, and silicate are the main anions. Rock dump and tailings storage facility (TSF) source terms for the Rapasaari-Päiväneva area were determined through laboratory leach testing and modelling. In developing a mine-wide water management plan, loading to discharge streams was considered for the mine operation phase. Post closure, waste rock, pyrite-containing waste rock, flotation tailings (pre- and post-flush-off situations), pre-float tailings, and pit lake overflow loading were assessed. Nutrient content (nitrogen and phosphorus) is a significant part of the Rapasaari-Päiväneva area water quality and load. The source of the nitrogen is ascribed to explosives, and the phosphorus is believed to be from the mined rock. However, the source of phosphorus is not totally understood. Modelling of the loading associated with the mine-wide water balance reveals that iron and phosphorous may exceed the environmental quality standard (EQS). The total salinity and nitrogen content are also of concern to the watercourse ecology. Based on the environmental impact assessment, the Fe- and P-loading does not pose a risk to the watercourses. However, the modelling exercise suggests that treatment of the water is required to address the nitrogen levels. The pyrite-containing waste rock will be deposited separately from the non-pyrite waste. The pyrite-containing waste is acid-generating, and the seepage from the pyrite-containing waste rock is expected to contain high levels of Fe and increased concentrations of metals and metalloids, such as Cd, Co, Ni, and Zn. Other key water quality parameters in the Rapasaari-Päiväneva area are arsenic, copper, and selenium. In the source term assessments, arsenic and copper appear only in small concentrations because of natural sorption in the waste facilities. They are, however, released to a significant extent during sulphide oxidation. It should be noted that the Water Management Study (AFRY, 2021) indicated that there is uncertainty in the Syväjärvi mine site water quality estimates due to the methodology used for determining water quality. 12.3.3 Water Balance A detailed water balance was prepared for the Rapasaari-Päiväneva Complex as part of the water management study (AFRY, 2021). The model considered groundwater and surface water and was run using several scenarios, including a climate change scenario. The model indicates that while freshwater make-up may be required for the first few years of operation, there will be a water surplus for the remaining years of the operation (i.e. there will be discharge from the site). The risk assessment in the water management plan also states that there might not be enough water to supply the process water requirements during all seasons due to modelled data being used to quantify the Köyhäjoki River flow rate. Once mining plans are developed, the water balance should consider including active dewatering as an alternative, or in addition, to pumping from the pit. Only a high-level water balance is available for the Syväjärvi site. Notably, a due diligence review by SRK in December 2023 flagged that updated water balance calculations have not been completed to verify that staged filling of both the magnetic and pre-flotation tailings ponds can synchronously be achieved. They further noted that significant modification to water structures and infrastructure designs and costings may be required, further emphasising the need for revised and detailed water balance and appropriate 3D volumetric modelling.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 267 12.4 LOM Production Schedule [§229.601(b)(96)(iii)(B)(13)(ii) and (iii)] The production schedule was developed on a monthly and annual basis using Deswik.Sched software. The sequence of mining was to commence at the Syväjärvi pit and then proceed to the Rapasaari pit. The Syväjärvi operation was capped at a total mining rate of 5 Mtpa (ore and waste), constrained by EP limitations, which limit mill feed to 540 ktpa and to 594 ktpa if environmental controls are in place. Syväjärvi ore needs to be depleted before Rapasaari ore is mined; this is due to the restriction of no blending between the two orebodies (as insufficient testwork has been completed). The Rapasaari operation was similarly constrained by EPs to a total mining rate of 8 Mtpa (ore and waste). Rapasaari waste stripping will commence during the final years of Syväjärvi to ensure a smooth transition in plant feed between the two orebodies and also to maintain a fairly consistent mining profile. The project targeted LiOH.H2O (the final product sold) production of 15 ktpa in the LOM production schedule for both Syväjärvi and Rapasaari. The production contributions by operation are summarised in Table 12-18, with the Rapasaari open pit being the largest ore production contributor at 11.3 Mt of ore with a strip ratio of 7.6 and an average grade of 0.9% Li₂O. The combined Syväjärvi and Rapasaari LOM production extends from 2025 to 2044, with the plant feed ending in April 2045. The Syväjärvi pit has the lowest stripping ratio of 5.1 with 3.2 Mt of ore at an average grade of 1.03% Li₂O. Table 12-18: Keliber Lithium Project production summary. Site Total (Mt) Ore Production (Mt) Stripping Ratio Li2O (%) LOM Syväjärvi Open Pit 19.32 3.15 5.1 1.03 Jun 2025 to Jul 2030 Rapasaari Open Pit 97.39 11.28 7.6 0.93 Feb 2028 to May 2044 Total 116.71 14.43 7.1 0.95 Jun 2025 to May 2044 12.4.1 LOM Scheduling This section outlines the requirements, constraints, and parameters applied throughout the LOM production scheduling phases for Syväjärvi and Rapasaari. The production schedules incorporated the LOM pit designs for both operations, including pushback phases at Rapasaari. 12.4.1.1 Scheduling Parameters The basis of design for the Syväjärvi and Rapasaari production schedule is summarised in Table 12-19, Table 12-20, and Table 12-21. The ramp-up to for the chemical plant utilised in the schedule was updated in the economic evaluation. Table 12-19: Production scheduling basis of design summary. Parameters Units Syväjärvi Rapasaari Operations limits and commencement: Maximum ore mining tpa 540,000 (Pre-mill feed) 850,000 (Pre-mill feed) Maximum total mining tpa Ore & waste: 5,000,000 Hard rock waste: 8,000,000 OVB stripping commences Date 01 June 2025 01 February 2028 Hard rock mining commences Date 01 June 2025 01 February 2028 Hazardous hard rock commences Date 01 July 2026 - S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 268 Parameters Units Syväjärvi Rapasaari Waste stripping Period OVB: One month of exposed hard rock Hard rock: One month of exposed ore Work Schedule as per EPs: Blasting Week, days, hours Weekdays 7:00-22:00 Weekdays 7:00-22:00 Drilling and secondary breaking Weekdays 7:00-22:00 7 Days/week 7:00-22:00 Loading, hauling ore and waste Weekdays only 24 hrs 24/7 Crusher feed 24/7 Allowed - Expected only 10 to 12 hours/day Days/year lost (uncontrollable & bad weather) Days/year 15 15 COGs: Marginal Ore COG % Li2O 0.19% 0.19% Concentrator feed grade % Li2O 0.21% 0.30% ROM stockpile capacity and grade ranges: Marginal ore % Li2O SMO: 0.15 to 0.20 RMO: 0.15 to 0.30 Low Grade % Li2O SLG: 0.20 to 1.00 RLG: 0.30 to 0.90 High grade % Li2O SHG: >1.00% RHG: >0.90% Ore containing muscovite & other contaminants % Li2O All grades All grades Minimum ROM stockpile levels Time Sufficient to maintain annual concentrator feed 2 Weeks Maximum ROM stockpile levels kt 125 and 250 No limit Päiväneva concentrator limitations Päiväneva concentrator "hot" commissioning Date 01 January 2026 Production ramp-up period (% of full capacity) Concentrator Syväjärvi Mining Month 1 20% 20% Month 2 40% 50% Month 3 60% 85% Month 4 75% 100% Month 5 90% 100% Month 6 100% 100% Päiväneva concentrator operating hours: Primary crushing Ore sorter and wet section Operating time Hours/day 10 to 12 24 Days/Year 340 340 LiOH2O production (Average) tpa 15,000 SMU block size XYZ 5 m x 5 m x 2.5 m Blast block size LWH 50 m x 30 m x 10 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 269 The ROM feed is shown in Table 12-20, highlighting the limitation placed on Syväjärvi. The Rapasaari feed is limited by the capacity of the mill. Table 12-20: Design criteria for daily ore production. Design Criteria for Ore Production Syväjärvi Rapasaari Ore Feed Unit 540 kt limit +10% 750 kt limit Mill feed limit (white rock) tpa 594,000 750,000 Limit with discard tpa 682,759 862,069 Fines to mill tpa 191,172 241,379 Whiterock (WR) excl. fines tpa 402,828 508,621 Whiterock fines tpa 137,644 173,793 Black rock fines tpa 53,528 67,586 Discard tpa 88,759 112,069 WR + fines + discard tpa 682,759 862,069 Max. WR to mill tpa 540,472 682,414 Max. mill feed tpa 594,000 750,000 The ROM feed limitations for contaminates are shown in Table 12-21 below. These contaminates were, subject to the limitations, included in the blending process in the production scheduling. Table 12-21: Contaminants limitations. Parameter Units Limitation Comments Black Rock feed to sorter % 30 Black rock Sulphur ore % No limit Sulphur removed in sorter and flotation to acceptable limits Sulphur waste % 0.8 & 1.0 > 0.8% (Syväjärvi), > 1.0% (Rapasaari) deposited on sulphur dump Arsenic ore ppm 200 200 ppm feed sorter and 50 ppm to mill Arsenic waste ppm 100 Deposited on sulphur dump Iron in ore % No Limit Feed grade >2% - removed in sorter and magnetic separator Iron in waste % No Limit No limit on waste MgO in ore % 0.42 Limit on Ore MgO in waste % No Limit No limit on waste 12.4.1.2 Scheduling Strategy The Syväjärvi production schedule drivers are listed below: • Ore mining direction perpendicular to strike and advancing up-dip (north to south). • Maximum of two active 10 m benches from the ramp to maintain bench access (reverse and forward mining). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 270 • Increase waste mining (additional excavator during initial nine months of schedule) to target 2 Mt low sulphur and arsenic construction waste material (blended to S < 0.3% and As < 19 ppm). • High sulphur and arsenic waste (S > 0.8%, As > 100 ppm) mining commence 01 July 2026. • ROM stockpile maximum limit: Ramp-up to reach 125 kt in quarter one of 2026 and maintain limit until December 2026; thereafter, maximum limit increase to 250 kt. • Sorter feed limit: < 30% black rock in feed. • Sorter feed high contaminants limits: MgO > 0.42%, As > 200 ppm; Fe2O3 limit not specified. • Diluted rock mill feed annual throughput limit: 540–594 ktpa (10% upper limit). • Discontinue plant feed end August 2030 and stockpile remaining ore until the end of LOM (transition to Rapasaari plant feed). • Deplete the ROM stockpile: Resume plant feed from February 2045 to deplete the remaining ore material on the Syväjärvi stockpile when Rapasaari production is complete (no blending of Syväjärvi and Rapasaari plant feed ore permitted). • Provide plant feed for average LiOH.H2O production of 15,000 tpa. The Rapasaari production schedule drivers are listed below: • Ore mining direction perpendicular to strike and advancing from south to north within the northern half of the pit, whereas the southern half of the pit advances from north to south. • Staggered bench approach throughout LOM to assist with temporary ramps when necessary while maintaining access to at least dual permanent ramps with the exception of the satellite pit. • High sulphur and arsenic waste (S > 1%, As > 100 ppm) mined throughout LOM (start to finish). • Monthly mined production tonnage ramps up to align with Syväjärvi’s ramp-down phase and thus fluctuates in order not to exceed a combined total production output. • Maximum permitted total mining limit (ore and waste): 8,000,000 tpa. • January 2029 to June 2030: Production at both Syväjärvi and Rapasaari fluctuates but does not exceed a combined total of 600,000 tonnes per month. • No ROM stockpile limit is applied, allowing the stockpile to grow to approximately 965,000 t. • Sorter feed limit: < 30% black rock in feed. • Crusher feed high contaminants limits: MgO > 0.42%, As > 200 ppm; Fe2O3 limit not specified. • Mill feed annual throughput limit: 750 ktpa. • Provide plant feed for average LiOH.H2O production of 15,000 tpa. 12.4.1.3 Total Material Movement Ore and waste movement were modelled on both a monthly and an annual basis for the respective pits at Syväjärvi and Rapasaari. Waste movement included stripping of topsoil (TS) and OVB, with material directed to designated TS and OVB dumps. Hard rock waste mining was classified by sulphur and arsenic content (low or high) and allocated to the corresponding hard rock waste dumps. No backfilling has been scheduled for the open pits.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 271 Ore mining schedules involved transferring material from the pits to categorised stockpiles on surface close to the pits. These stockpiles were categorised based on high- and low-grade ore as well as high- and low-contaminant ore. The ROM stockpiles were integrated into the overall mine scheduling model in terms of blending plant feed quantities and qualities, stockpile analysis, and meeting product requirements. Marginal ore was considered waste and directed to the waste dump facilities due to its low quantities in the ROM schedule. TS was scheduled for removal in 0.5 m thick layers, OVB and hard rock in 10 m bench intervals, and ore selectively in 2.5 m intervals (flitch heights). The total ore and waste annual material movement from the two deposits is shown in Figure 12-31. Annual LOM Ore and Waste Production Schedule by Pit Figure 12-31: Annual LOM ore and waste production schedule by pit. The annual crusher and mill feed from the Syväjärvi and Rapasaari ROM stockpiles are presented in Figure 12-32 and Figure 12-33, respectively. Throughout the LOM, the annual crusher feed tonnage remains below the crusher's nominal and design capacities, with a Li₂O feed grade ranging from 0.88% to 1.14% (averaging 0.96%). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 272 Annual LOM Crusher Feed from Stockpiles Figure 12-32: Annual LOM crusher feed from stockpiles. The mill feed requirement slightly exceeds Syväjärvi's permit limit of 540 ktpa during the first three years of production, rising to 589 ktpa in the fourth year, which remains within the 10% upper limit of 594 ktpa. For the remainder of the production schedule, the Rapasaari mill feed stays within the mill's grinding design capacity of 725 ktpa. Annual LOM Mill Feed Figure 12-33: Annual LOM mill feed. Figure 12-34 illustrates the annual LiOH.H2O production from the refinery throughout the LOM. After ramp-up, the minimum targeted LiOH.H2O annual production of 15 kt is reached throughout the LOM with the exception of year S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 273 five (2030) when transitioning from the Syväjärvi to the Rapasaari ore production. This is primarily due to the predominance of low-grade ore available on the stockpile at Syväjärvi as it nears the end of its life, combined with restrictions on blending material from the different deposits at Syväjärvi and Rapasaari. Additionally, the limited stockpile capacity at Syväjärvi necessitates ongoing plant feed operations. As a result, the combination of low-grade ore feed and plant capacity constraints leads to product output. The production schedule ramp-up and concentrator feed was altered in the economic evaluation. Annual LOM Final Product (LiOH.H2O) Figure 12-34: Annual LOM final product (LiOH.H2O). The ROM high- and low-grade ore material (as defined in Section 12.4.1.1) from the two deposits was further categorised into ore containing high- and low concentrations of contaminants (As and MgO) for placement on the four different ROM stockpiles for plant feed blending purposes. These include the following: • High-grade, low-contaminant ore (SHG_LowCon and RHG_LowCon) • High-grade, high-contaminant ore (SHG_HighCon and RHG_HighCon) • Low-grade, low-contaminant ore (SLG_LowCon and RLG_LowCon) • Low-grade, high-contaminant ore (SLG_HighCon and RLG_HighCon) • Where: − Syväjärvi (S), and Rapasaari (R) − Low contaminants (LowCon): MgO < 0.42% and As < 200 ppm − High contaminants (HighCon): MgO > 0.42% or As > 200 ppm Figure 12-35 illustrates the monthly stockpile levels and weighted average Li₂O grade (%) for the Syväjärvi deposit at the end of each month. The stockpile levels are categorised into the four grading bins: SHG_LowCon, SHG_HighCon, SLG_LowCon, and SLG_HighCon, with their respective contributions shown in different colours. Initially, stockpile S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 274 levels ramp-up and remain below the 125 kt limit until December 2026, after which the limit increases to 250 kt, with maximum levels peaking slightly above 200 kt. The weighted average Li₂O grade (purple line) initially increases to above 3.0% as high-grade ore is stockpiled from the ROM operation. It then declines as high-grade ore is depleted by late 2029 and early 2030, nearing the end of life of the Syväjärvi pit. By this point, only 123 kt of low-grade, high- contaminant material remains on the stockpile. Plant feed from Syväjärvi is discontinued between 2030 and 2045 to allow production from Rapasaari to commence. In February 2045, plant feed resumes to deplete the remaining Syväjärvi stockpile after Rapasaari production is complete. Syväjärvi Monthly Stockpile Levels per Finger Stockpile Category and Combined Stockpile Li2O Grade Figure 12-35: Syväjärvi monthly stockpile levels. Figure 12-36 illustrates the monthly stockpile levels at Rapasaari over the LOM, similar to the Syväjärvi stockpiles, and categorises them into four grading bins represented by different colours. No limit was set for the Rapasaari stockpile. Initial ROM ore from Rapasaari is stockpiled from February 2029, reaching a peak of 965 kt in 2034. Stockpile levels then decline sharply to 155 kt during the mid-LOM period, which coincides with new pushbacks, increased strip ratios, and lower-grade ore that predominantly does not meet product specifications. These higher stockpile levels were required to ensure a consistent plant feed during this mining phase. The majority of the stockpiled ore consists of low-grade, high-contaminant material, followed by high-grade, high- contaminant material. High-grade, low-contaminant ore remains limited throughout the LOM, with most of it mined towards the end of the LOM from deeper sections of the orebody. The weighted average Li₂O grade of the stockpile fluctuates between 0.64% and 1.15%, with an average of 0.93% across the LOM.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 275 Rapasaari Monthly Stockpile Levels per Finger Stockpile Category and Combined Stockpile Li2O Grade Figure 12-36: Rapasaari monthly stockpile levels. Figure 12-37 presents the MgO and As contaminant grades in the plant feed after blending for the entire LOM. The As grades consistently remain within the upper limit, whereas the MgO grades approach and occasionally exceed the limit during certain periods. Due to the blending strategy and the availability of low-contaminant ore from the Syväjärvi stockpiles, MgO levels are maintained at or below the upper limit throughout Syväjärvi’s production until August 2030. However, a temporary exceedance occurs in 2045 due to the final depletion of high-contaminant ore stockpiles. For Rapasaari, MgO levels surpass the upper limit during a few periods when low-contaminant ore is insufficient in the ROM production, leaving only high-contaminant ore available for blending. Monthly Average MgO and As Contaminant Grades in Syväjärvi and Rapasaari Crusher Feed After Blending vs Limits Figure 12-37: Monthly contaminant grade in crusher feed from stockpiles. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 276 12.4.1.4 High Sulphide- and Arsenic-Bearing Waste Rock The Syväjärvi and Rapasaari deposits comprise waste rock with high sulphide and arsenic content, defined as hazardous waste. These waste rocks will be deposited on separate WRDs for the respective operations. The waste rock is considered to have high sulphide and arsenic content if it meets the following criteria based on blast block size resolution: • Syväjärvi: S > 0.8% or As > 100 ppm • Rapasaari: S > 1.0% or As > 100 ppm Figure 12-38 illustrates the yearly tonnes of excavated waste rock that contains high levels of sulphur and arsenic content. Due to permitting requirements at Syväjärvi, mining of hazardous waste will only commence in July 2026. Rapasaari’s hazardous waste will be mined without any limitations of hazardous waste as the assumption is that the required permits will be in place by the time mining commences at Rapasaari. High Sulphide- and Arsenic-Bearing Waste Rock Mined Annually by Deposit Figure 12-38: High sulphide- and arsenic-bearing waste rock excavated annually. 12.4.2 Production Parameters 12.4.2.1 Operational Parameters for Ore Production Consistent ore production is crucial for maintaining both operational efficiency and economic stability. To achieve the annual target of 15,000 tonnes of LiOH.H2O, a reliable and on-spec ore feed must be maintained each month. The design criteria for ore production from Syväjärvi and Rapasaari are outlined in Table 12-20. Mining benches will be blasted in two stages, each 10 metres high, to improve fragmentation and minimise dilution at the ore-waste boundary. To further reduce dilution, ore will be selectively mined in 2.5-metre flitches. For ore mining, either ADTs with a 41 tonne payload or 64-tonne rigid off-highway trucks will be aligned to the 92-tonne CAT 395 excavator, which has proven effective for selective loading. Waste rock mining is planned to be mined by a 140-tonne CAT 6015 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 277 excavator paired with 64-tonne rigid off-highway trucks. Additionally, Syväjärvi’s ore feed limit has been increased by 10% to compensate for lower feed grade in the latter years, ensuring the final product meets specifications. 12.4.2.2 Operational Parameters for Waste Mining The annual ore and waste mining permit limits are set at 5 Mtpa for Syväjärvi and 8 Mtpa for Rapasaari. The waste mining strategy ensures that enough OVB is removed to expose one month’s worth of hard rock which, in turn, allows for sufficient hard rock removal to keep one month’s worth of ore accessible. Syväjärvi’s and Rapasaari’s waste mining approach, including the handling of hazardous and non-hazardous waste, is detailed in Section 12.4.1.4. Haul roads and in-pit ramps are designed to accommodate rigid-body, off-highway trucks with a 64-tonne payload, such as the CAT 775. 12.4.2.3 Operational Concept Keliber's operating model relies on mining contractors for primary and secondary services, including drilling, blasting, loading, hauling, and associated support tasks. To ensure efficient operations, Keliber will appoint key personnel to oversee contractor activities and manage critical departments. These include Mine Technical Services, responsible for mine planning and resource management; Environmental Services, focusing on permitting and regulatory compliance; and Mine Management, overseeing overall site operations. 12.4.2.4 Operating Hours The operating hours for both Syväjärvi and Rapasaari will comply with the maximum permissible working hours under Finnish regulations and the relevant EP restrictions for each pit. Both operations will run 24 hours a day, with Syväjärvi operating 5 days a week and Rapasaari operating 7 days a week on a 3-shift basis for load and haul and a 2-shift basis for drilling (07h00 to 22h00). Blasting will be permitted only between 07h00 and 22h00 on weekdays. A detailed time usage model for both operations is provided in Table 12-22 below. Table 12-22: Syväjärvi and Rapasaari time usage model for loading and hauling. Description Units Syväjärvi Rapasaari Comments 3 Shifts/day 5 days/week 3 Shifts/day 7 days/week Load & Haul Load & Haul Annual days 365 365 Working days days 260 365 Shifts no. 3 3 Shift duration hrs 8 8 Hrs 07h00 until 22h00 Non-working time (weather) days 15 15 Non-working time (weather) shifts 45 45 Shifts available no. 735 1,050 Hours available hrs 5,880 8,400 On-shift mechanical availability % 85 85 Maintenance off-shift Equipment available time hrs 4,998 7,140 Equipment available time hrs per shift 6.8 6.8 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 278 Description Units Syväjärvi Rapasaari Comments 3 Shifts/day 5 days/week 3 Shifts/day 7 days/week Load & Haul Load & Haul Shift change hrs/shift 0.5 0.5 Outside shift hours Blasting hrs/shift 0.14 0.14 Refuelling hrs/shift 0.15 0.15 Meetings hrs/shift 0.02 0.02 Waiting time (drills) & queuing hrs/shift 0.5 0.5 Waiting for excavator/shovel hrs/shift 0.5 0.5 Total delays per shift hrs/shift 1.81 1.81 Uptime per shift hrs 4.98 4.99 Uptime per day hrs 14.95 14.97 Effective utilisation % 73.3% 73.4% Of availability Operating hours per year hrs 3,662 5,240 Overall utilisation (on-shift) % 62.3% 62.4% Of available hours Overall utilisation (24/7) % 41.8% 59.8% 12.4.2.5 Pre-Production Activities Most of the necessary procurement activities have been completed or are nearing completion. Construction pre- production activities are progressing well at the Syväjärvi pit, with the excavation of a borrow pit and the construction of haul and service roads currently underway. Once the OVB is removed and the rock surface is cleaned, waste rock mining in line with the production schedule will be able to begin. 12.4.3 Drilling and Blasting The information for this section has been sourced from: • Rorke, A.J. 09 October 2024. Blast Designs – Syväjärvi and Rapasaari Lithium Mines. AJR, Revision Number 2. Quantifying dilution was identified as a key aspect of the study, which led to an alternative approach for estimating dilution compared to the 2022 FS. Significant time was also dedicated to refining selective mining practices to ensure a shared understanding of how to minimise dilution during mining operations. Blasting techniques were a major focus, with careful consideration given to methods that would minimise the mixing of ore and waste along the contacts when a blast is detonated. This was done while taking into account the differing dip angles and geometries of the orebodies, with Syväjärvi's orebody having a relatively shallow dip (mostly tabular in nature) and Rapasaari's orebody consisting of steeper dipping lenses. The two pits share similar rock properties, so the blast designs developed for Syväjärvi can be applied to Rapasaari with reasonable confidence that the results and fragmentation will be comparable. A drill hole diameter of 165 mm is recommended for all waste blasts, while 127 mm is suggested for all ore, trim, and pre-split blasts. Except for pre-

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 279 splits, the 20 m high benches will be blasted in separate 10 m increments. Air decks have been incorporated into the designs where ore is vertically interlayered across a blast block. A summary of the production blast designs is provided in Table 12-23 and Table 12-24. There are six basic rock types where blasting will mostly take place. These have different properties that will impact blast fragmentation. They are: • MV: metavolcanite – Waste • MS: mica schist – Waste • SS: sulphide schist – Waste • MP: muscovite pegmatite – Contaminant • PP: plagioclase porphyrite – Waste • SP: spodumene pegmatite – Ore The MP is not considered separately in the designs as it exists as a thin layer at the ore boundaries and has similar properties to the ore. Table 12-23: Summary of waste production blasts. Blast Designs Waste Blasting Unit MV MS SS PP Hole diameter mm 165 165 165 165 BH m 10 10 10 10 Drill angle from vertical ° 10 10 10 10 Stemming m 3.3 3.3 3.3 3.3 Sub-drill m 1.1 1.1 1.1 1.3 Hole depth m 11.4 11.4 11.4 11.7 Burden m 5.3 5.3 5.3 4.5 Spacing m 6.1 6.1 6.1 5.1 Explosive type Bulk Blend Bulk Blend Bulk Blend Bulk Blend Blend ratio % 30 30 30 30 Explosive density g/cm³ 1.2 1.2 1.2 1.2 Charge length m 8.1 8.1 8.1 8.4 Charge deck length m Not air decked Linear charge mass kg/m 25.7 25.7 25.7 25.7 Charge mass per hole kg 208 208 208 208 Charge mass per hole - decked Not air decked Powder factor kg/m³ 0.65 0.65 0.65 0.94 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 280 Table 12-24: Summary of ore production blasts. Blast Designs Ore Blasting Unit SP SP Decked Hole diameter mm 127 127 BH m 10 10 Drill angle from vertical ° 10 10 Stemming m 2.6 2.6 Sub-drill m 1 0.9 Hole depth m 11.3 11.2 Burden m 3.3 2.9 Spacing m 3.8 3.4 Explosive type Bulk Blend Bulk Blend Blend ratio % 30 30 Explosive density g/cm³ 1.2 1.2 Charge length m 8.8 7.2 Charge deck length m 3.6 Linear charge mass kg/m 15.2 15.2 Charge mass per hole kg 208 208 Charge mass per hole - decked 54.6 Powder factor (kg/m³) 1.07 1.11 Fragmentation predictions are derived from the available geotechnical data and visual analysis of the fragmentation results observed in the waste blasts at the Syväjärvi conducted during the construction phase. Drilling quantities are provided for each design in relation to the proposed pattern design. Given the significant differences in the average dip and strike of the orebodies at Syväjärvi and Rapasaari, distinct approaches to grade control will be required for each site. 12.4.4 Material Movement 12.4.4.1 Loading and Hauling Material movement will be via conventional loading and hauling methods, with fleet standardisation considered across both mining areas. Ore will be loaded by a 94-tonne hydraulic excavator with a 4.6 m³ bucket capacity onto articulated dump trucks with a 41-tonne capacity or onto 64-tonne rigid off-highway trucks. Waste will be loaded by a 140-tonne hydraulic excavator with a 5.8 m³ bucket onto 64-tonne rigid off-highway trucks. The selection of the ore loader has been carefully considered to facilitate selective mining. Further details on equipment selection are provided in Section 12.4.11. The distinct colour difference between ore and waste rock is advantageous for selective mining and supports the use of conventional loading and hauling methods. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 281 12.4.4.2 Haul Road Design and Maintenance The widths of both single and dual haul roads have been calculated using standard industry guidelines. A well- constructed haul road is essential for efficient hauling operations. A typical haul road is composed of four distinct layers: sub-grade, sub-base, base course, and surface course. The sub-grade consists of the in situ rock. If the sub-grade is made up of hard, dense rock, minimal fill may be required. When the sub-grade is competent, the base course is typically placed directly on top. This layer consists of high-quality material with the appropriate particle size (usually coarse), as it provides the primary structural strength of the road. The surface course, which is generally made from fine gravel, is placed on top and features closely controlled grading. This layer ensures proper weight distribution, provides a smooth travelling surface, and facilitates drainage. The haul road is designed with a 2–3% cross-sectional slope, which is crucial for effective water drainage, particularly during the snowmelt season. Safety berms have been incorporated along the outer edges of the haul roads, while drainage systems have been integrated within the inner skirts to assist with water runoff. Additionally, the ramps have been designed with a 10% grade. 12.4.5 Dilution Control Syväjärvi's orebody has an approximate dip of 10 to 30 degrees, which is shallower compared to Rapasaari's orebody. Due to this relatively shallow dip, blast analysis has shown that controlling ore dilution at the top and bottom contacts will be challenging. To limit dilution, the proposed methodology involves using decked charges and sequential firing to minimise vertical displacements and prevent the mixing of ore and waste at the shallow dipping contacts. The recommended block orientation is to position the long axis of the blast blocks approximately perpendicular to the average plunge of the orebody, with a minimum block width of 30 metres. Rapasaari has a steeper dipping ore lens with an approximate dip of 30 to 60 degrees (shallow towards the north), which is crucial for generating movement along the strike of the ore and waste to minimise dilution at the contacts. The geometry of the orebody allows for the separation of ore and waste into distinct piles during blasting where both are present. Blast simulations suggest that the free face should be oriented perpendicular to the strike, meaning that mining should proceed north and south, with free faces running east to west across the width of the blast block. In practice, each blast block will have two free faces, positioned opposite to the direction of advance. It is recommended that the blast block size at Rapasaari should maintain a width-to-length ratio of 2:1. The block sizes for both Syväjärvi and Rapasaari have been standardised to 50 m x 30 m x 10 m. 12.4.6 Grade Control It is recommended that grade control is performed daily to assist the process of ensuring a consistent and reliable feed grade to the plant after blending. Grade control will be based on samples taken from the production blast holes and infill drilling ahead of the mining faces. A robust grade control schedule with supporting standards and procedures will be implemented during the production phase. 12.4.7 Primary Crusher Feed and ROM Pad Storage The primary crusher feed refers to the limit between the mining and blending of ore and the concentrator plant. The crusher has been designed with a higher capacity to ensure it can crush ore to a maximum particle size of 700 mm within the required timeframe. Initially, Syväjärvi’s ore will be stockpiled near the waste dump, where a stockpiling pad will be constructed. Once the environmental approval for Rapasaari is obtained, construction of a ROM stockpile facility at the Päiväneva S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 282 concentrator site can proceed. However, it is assumed that the Päiväneva ROM pad will not be available until Syväjärvi is mined out. Therefore, all Syväjärvi ore will be stockpiled at the Syväjärvi ROM pad adjacent to the waste dump and rehandled to the crusher at the Päiväneva concentrator, located approximately 3 km away. The Syväjärvi ROM pad capacity is initially limited to 125 kt until Q3 2026, after which it will increase to 250 kt. The Rapasaari ROM pad has no significant capacity constraints during the scheduling phase and is expected to grow to around 980 kt, ensuring consistent lithium hydroxide production throughout the life of the mine. Sufficient space for the Rapasaari stockpile needs to be confirmed. Ore will be classified into distinct grade ranges through a rigorous grade control process and stockpiled accordingly, as detailed in Table 12-25. Marginal ore will be stored alongside the waste dumps, ensuring future accessibility for concentrator feed. Ore with elevated levels of Muscovite Pegmatite, sulphur, arsenic, iron, or other contaminants exceeding the marginal ore threshold will be segregated into high- and low-contaminant stockpiles. Consequently, the Syväjärvi ROM pad will maintain at least four distinct stockpiles: SLG, SHG, SLowCon, and SHighCon. Similarly, Rapasaari will have RLG, RHG, RLowCon, and RHighCon stockpiles. Table 12-25: Stockpile grade ranges. Stockpile Category Unit Syväjärvi Rapasaari Name Grade Range Name Grade Range Marginal ore % Li₂O SMO 0.15 to 0.20 RMO 0.15 to 0.30 Low-grade ore % Li₂O SLG 0.20 to 1.00 RLG 0.30 to 0.90 High-grade ore % Li₂O SHG > 1.00% RHG > 0.90% Low-contaminated ore % Li₂O SLowCon > 0.20% RLowCon > 0.30% High-contaminated ore % Li₂O SHighCon > 0.20% RHighCon > 0.30% 12.4.8 Ore Blending Ore blending will be essential to maintain a consistent grade for the concentrator, ensuring that the required 15,000 tpa of final product is delivered. The goal of blending is to maximise concentrator recovery. Since ores from the two pits cannot be mixed, blending will be conducted separately for the four stockpiles of each pit (marginal ore not considered for blending). Ore from stockpiles containing contaminants must be blended according to the limits specified in Table 12-25. Blending was managed through the Deswik software, which prioritised meeting grade requirements to ensure that the monthly target for the final product (lithium hydroxide) was achieved. The software minimised the inclusion of contaminants in the feed. For muscovite pegmatites, the blending requirement was set to deplete the stockpile as much as possible, keeping contaminant quantities to a minimum. Although direct tipping from the pit will be implemented once the grade control function is fully established, during this scheduling phase of the study, all ore mined from the pits was stockpiled first. Once the mine is in operation, the grade control officer will provide a daily blending plan, specifying the ratio of truck and front-end-loader loads from the various stockpiles to be fed to the crusher.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 283 12.4.9 Waste Rock Storage Facilities Construction rock is required for constructing facilities and will mostly be mined from Syväjärvi. Approximately 1.5 to 2.0 Mt of construction rock (waste rock) will be mined from Syväjärvi during the Project development phase in the year prior to the LOM production schedule commencement date. During this phase, mined waste rock will be transported to tailings dam raises or dump pad preparations, or crushed for road construction. The balance of the waste rock will be mined from the pits and hauled to separate WRDs or dumps (including TS and OVB material) designated for Syväjärvi and Rapasaari, respectively. Mined material will be transported by off-highway dump trucks from the pits to a level surface on the respective dumps and, subsequently, pushed and levelled using dozers and graders. This method enhances both the safety of operations and the stability of the WRDs. The final slope angle of the waste rock and OVB storages will be maintained at an 18° OSA, ensuring feasible rehabilitation during the closure phase. The waste rock- and OVB dump design criteria for Syväjärvi and Rapasaari are shown in Table 12-26 and Table 12-27, respectively. The Syväjärvi WRD will exceed the 119 mamsl restriction by 6 m; permitting of the additional height will be done in due course. Table 12-26: WRD design parameters. Parameter Unit Syväjärvi Rapasaari BH m 10 10 BW m 18 18 BFA ° 36 36 Ramp width m 28 28 OSA ° 18 18 Height limit mamsl 119 165 The Syväjärvi OVB dumps do not have a height restriction. The Rapasaari OVB dumps exceeded the restrictions shown in Table 12-27 below. The W1 dump is 115 mamsl and the E1 dump is 115 mamsl. Table 12-27: OVB dump design parameters. Parameter Unit Syväjärvi TS Syväjärvi OVB Rapasaari N1 Rapasaari W1 Rapasaari E1 Ramp width m 16 28 16 28 28 OSA ° 18 18 18 18 18 Height limit mamsl N/A N/A 100 100 102 12.4.9.1 Syväjärvi Waste Rock Storage The surface plan layout for waste material storage at Syväjärvi is depicted in Figure 12-39 and detailed in Table 12-28. Most of the waste material storage facilities are planned to be located outside the footprint of the unconstrained RF 2.0 pit optimisation shell. The only exception is a small portion of Topsoil Dump and OVB Dump 1. This area is restricted by environmental permitting constraints. OVB will be dumped in two phases. OVB Dump 2 must be delayed until the end of October 2025. This is when the environmental permitting process is anticipated to be completed, allowing for S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 284 expansion from OVB Dump 1. Provision is made for an isolated hazardous waste rock dump towards the south-east of the main hard rock waste dump. Plan View of the Waste Storage Facilities and their Respective Design Capacities at Syväjärvi Figure 12-39: Waste rock storage facilities at Syväjärvi. Table 12-28: Syväjärvi scheduled waste material volumes versus waste storage capacity. Material Schedule Volume In situ [m3] Schedule Volume Swelled 25% [m3] Dump Capacity [m3] Remaining Dump Contingency [m3] TS 41,282 51,602 177,786* 126,184 OVB 562,275 702,843 817,780 114,937 Hazardous waste rock 309,937 387,422 463,125 75,703 Waste rock 5,233,236 6,541,545 7,544,116 786,446 Discard black rock 172,845 216,056 MO 55 69 Grant Total 6,319,629 7,899,537 9,002,806 1,103,270 *Note: The current TS already dumped is 124 kBCM, hence the total space remaining is 1.6 kBCM. 250 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 285 12.4.9.2 Rapasaari Waste Rock Storage Rapasaari has not been considered as a source of construction rock in this phase of the study. The pit contains approximately 9 million cubic metres of hazardous waste (S>1% and As>100ppm), which will be mined throughout the LOM. The planned layout for waste material storage at Rapasaari is illustrated in Figure 12-40 and detailed in Table 12-29. All waste storage facilities will be positioned outside the unconstrained RF 2.0 pit optimisation shell, with an additional 35-metre buffer. While all WRDs are within the mining lease area, the Phase 2 waste dump lies outside it and will only be utilised once the Phase 1 dump reaches full capacity. To meet storage requirements, two OVB dumps have been planned. As the TS dump has insufficient capacity, any excess TS will be allocated to the OVB dumps, which have surplus space. These dumps, situated on either side of the pit, will be utilised based on the shortest hauling distance relative to the active mining blocks. The Phase 2 Waste Dump permitting will require an EIA and application for an environmental permit. Plan View of the Waste Storage Facilities and their Respective Design Capacities at Rapasaari Figure 12-40: Waste rock storage facilities at Rapasaari. 500 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 286 Table 12-29: Rapasaari scheduled waste material volumes versus waste storage capacity. Material Schedule Volume In situ [m3] Schedule Volume Swelled 25% [m3] Dump Capacity [m3] Remaining Dump Contingency [m3] TS 374,457 468,071 129,712 -338,359 OVB 5,695,816 7,119,770 7,527,830 408,060 Hazardous waste rock 6,987,485 8,734,356 8,981,955 247,599 Waste rock 20,795,277 25,994,096 28,927,857 2,241,985 Discard black rock 553,352 691,690 MO 69 86 Grant Total 34,406,455 43,008,069 45,567,355 2,559,285 12.4.10 Mine Dewatering and Water Management The information for this section has been sourced from: • Keliber_DFS_Volume_1 to _Executive Summary_February_01_2022_(final) in PDF format Volume_1 to _7 • Deminey, J.G.L., & Barnard, E. 02 December 2024. Keliber Lithium Mine – Dewatering and Drainage Strategy. Nurizon Consulting Engineers, Revision Number 1, Final. The water management chapter of this TRS provides a detailed description of the mine site water management plan. To minimise environmental impacts, particularly the release of suspended solids into natural water bodies, water management and treatment systems will be constructed prior to other construction activities. During the operational phase, perimeter drains will be implemented to collect runoff from operational areas. Syväjärvi and Rapasaari both have sulphuric waste dumps that pose significant environmental risks. Runoff from these hazardous dumps will be collected through a network of drains and channels designed to capture and direct the runoff into a collection system. The collected runoff will then be pumped to the concentrator plant for treatment. Nurizon Consulting Engineers were appointed to develop a dewatering and drainage concept design, along with a conceptual budget estimate and strategy for the two open pits. The proposed dewatering strategy relies on a mobile, diesel-powered pump that can be repositioned to the lowest point of the pit as mining progresses. A dewatering pipeline will extend from the pit bottom to surface dewatering ponds, following the haul road route and running adjacent to the highwall. The pump systems will be fully automated and controlled from a nearby control room and a front-end loader or dump truck can be used to transport and move the mobile dewatering pump systems around in the pit. The concept design at Syväjärvi is based on the use of a diesel-powered, self-priming pump capable of handling the full head when the pit reaches its maximum depth. Figure 12-41 illustrates the pipeline route from the bottom of the pit to the dewatering pond, covering a total distance of 1,895 m, representing the longest and deepest scenario. Figure 12-41 is for illustration and calculation purposes only. The pipeline will be optimised in operation. The operational philosophy involves throttling the pump during the initial stages of mining, as it will be oversized for the low pump head at that stage. As the pit deepens, the pump will gradually operate within its optimal efficiency curve. To ensure optimal operation, a sump with a capacity to hold three days’ worth of dewatering volume must be constructed within the pit.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 287 The selected dewatering pump is designed to handle sludge and grit for mining pit dewatering. It is sized to pump 20 litres per second at a 140-metre head, providing a 12-hour redundancy based on the maximum groundwater and precipitation values (including snowfall) specified in the AFRY Site Water Management Report when the pit reaches its deepest point. Sufficient contingency has been incorporated into this conceptual design, including an additional backup pump system. This backup system can also be utilised for the initial phase of the Rapsaari pit. Pit Dewatering Pipe Route from the Bottom of the Pit to the Dewatering Pond on Surface at Syväjärvi Figure 12-41: Syväjärvi Pit: dewatering routes to dewatering pond (source: Nurizon Consulting Engineers). The concept design at Rapasaari is based on using two different diesel-powered, self-priming pumps for dewatering at different stages of pit mining. In the first stage, a smaller pump system will be used to dewater the pit to a depth of 140 metres. The operational philosophy for this stage involves throttling the pump during the initial mining phase, as it will initially be oversized for the low pump head. As the pit deepens to 140 metres, the pump will begin operating within its efficient pump curve. For the second stage, a larger pump system will be required to dewater the pit from 140 metres to 180 metres. Unlike the first pump, this system cannot be throttled from the initial stage, necessitating the use of two separate pumps. The second pump will also be throttled in the early stages of this phase and will gradually operate within its efficient pump curve as the pit deepens. Figure 12-42 illustrates the pit depth across the three mining phases, with Phase 2 being the deepest at 180 metres. In this phase, the longest pipeline extends 4,054 metres from the bottom of the pit to the dewatering ponds, as shown in the figure. Phase 2 was considered the worst-case scenario for the pump systems. 20 m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 288 To ensure optimal operation, a sump with a capacity to store three days’ worth of dewatering volume must be constructed within the pit. The dewatering pump selected for Stage 1 of Phase 2 of mining is designed to handle both sludge and grit for pit dewatering. It is sized to pump 72 litres per second at a 100-metre head, providing 12 hours of daily redundancy based on the maximum groundwater and precipitation values (including snowfall) specified in the AFRY Site Water Management Report. For Stage 2 of Phase 2, the selected dewatering pump is sized to pump 72 litres per second at a 180-metre head, also ensuring 12 hours of daily redundancy. Additionally, a backup pump system has been included as a redundancy measure. Pit Dewatering Pipe Routes from the Bottom of the Pit for the Different Mining Phases to the Dewatering Pond on Surface at Rapasaari Figure 12-42: Rapasaari Pit: dewatering routes to dewatering pond per mining phase (source: Nurizon Consulting Engineers). Surface drainage should be managed using berms and earth channels surrounding the open pits, directing stormwater to low points or stormwater catchment dams away from the mining areas. The typical conceptual design of a berm and earth channel is illustrated in Figure 12-43. It should be noted that a detailed design is required to determine the precise dimensions of the channel and berm combination. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 289 Typical Conceptual Design of a Berm and Earth Channel Drawing Figure 12-43: Surface drainage berm and earth channel drawing (source: Nurizon Consulting Engineers). The mining contractor will be responsible for acquiring and maintaining the pump stations and in-pit pipelines, as well as relocating the pump sump as mining progresses deeper. Keliber will provide the electricity cabling for the pump station and the fixed pipelines on the surface. 12.4.11 Open-Pit Fleet The 2022 FS for Syväjärvi and Rapasaari provided an estimation of the LOM fleet size based on the production schedule available at the time. The study focused on selecting the optimal combination of loaders and trucks, as well as establishing realistic performance rates for the chosen truck and loader pair. The equipment selection was verified during this Project. A key aspect of the study was dilution control during ore extraction, which is critical to ensuring the Project's economic success. Minimising dilution requires the correct loading equipment, considering the geometries and layout of the Syväjärvi and Rapasaari orebodies within the surrounding host rock. The approach used to quantify dilution involved several key steps: • Dilution: Regularised blocks were used to capture dilution at the material boundaries; • Re-blocking/regularisation: Blocks were classified as ore or waste based on the dominant material mass; • Reclassification: Blocks were reclassified according to a marginal lithium value cut-off; and • A 5 x 5 x 2.5 m block size was found to provide an acceptable balance between dilution and losses, while also being practical for implementation. Minimising dilution is essential for realising the calculated value of the Project. As a result, considerable attention was given to selecting a suitably sized excavator with proven application in hard rock mining, ensuring it meets the stringent requirements for selective mining. The following two key factors were considered when selecting an appropriate ore loader: • The practicality of equipment size relative to the chosen SMU; and • Productivity considerations to ensure the equipment can operate efficiently at the required ore extraction rates. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 290 Selecting the right loader for optimal production requires matching the machine and, more specifically, the bucket, to the production requirements. Failing to do so can lead to choosing a loader that is too small to meet the required output, resulting in the need for additional loaders at extra cost. On the other hand, selecting a loader that is too large can cause excessive loading times, reducing overall efficiency. To determine the appropriate loader size, the Barloworld CAT methodology was applied (refer to Table 12-30). This approach calculates the nominal bucket sizes needed to achieve predetermined production throughputs based on several factors, including: • Annual production requirements; • Expected operating time – daytime ore loading; • General operational efficiencies; and • Minimum bucket specification density of 2.1 t/m³ Table 12-30: Barloworld CAT bucket size calculation. Basis – Daytime Loading of Ore Syväjärvi Rapasaari Unit Ore Waste Ore Waste In situ density t/m³ 2.72 2.71 2.70 2.71 Swell factor % 25 25 25 25 Loose density t/m³ 2.18 2.16 2.16 2.17 Material class Hard dig Hard dig Hard dig Hard dig Expected bucket fill for material class % 90 90 90 90 Instantaneous cycle time per pass s 23.00 23.00 23.00 23.00 Bench utilisation % 70 70 70 70 Realistic cycle time per pass sec 32.86 32.86 32.86 32.86 Cycle time per pass Min. 0.55 0.55 0.55 0.55 Operator skill efficiency min/hr 50.00 50.00 50.00 50.00 Operator skill efficiency % 83 83 83 83 Machine availability % 85 85 85 85 General operating efficiency % 90 90 90 90 Effective cycles per hour no. 70.00 70.00 70.00 70.00 Annual production required T 700,000 5,000,000 850,000 8,000,000 Annual hours - ore loading Hrs 1,831 3,662 2,620 5,240 Hourly production required t/hr 382 1,365 324 1,527 Hourly production required - x2 waste excavators t/hr 683 763 Required payload m³ 5.46 9.75 4.64 10.91 Rule of thumb x1 Ore Exc x2 Wst Exc x1 Ore Exc x2 Wst Exc Bucket payload volume m³ 2.51 4.51 2.14 5.03

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 291 Basis – Daytime Loading of Ore Syväjärvi Rapasaari Unit Ore Waste Ore Waste Nominal bucket size – required m³ 2.79 5.01 2.38 5.59 Selected bucket size m³ 4.60 5.80 4.60 5.80 Bucket type Severe Duty Extreme Duty Severe Duty Extreme Duty Bucket spec density t/m³ 2.10 2.70 2.10 2.70 Machine type CAT395 CAT6015 CAT395 CAT6015 The selected bucket sizes are 4.6 m³ for ore and 5.8 m³ for waste. Key considerations in the selection of the bucket sizes and applicable equipment include: • Ore-loading bucket – 4.6 m³ – Larger bucket then required to account for selective loading; • Waste loading bucket – 5.8 m³ – On the basis of x2 waste excavators (it is good practice to plan for x2 waste excavators to eliminate the risk of complete production stoppages for major breakdowns during the process of loader selection); • Bucket size specification – Buckets selected according to the material density of 2,100 kg/m³, rock classification, etc; • CAT 395 has proven application in hard rock selective mining within South African open-pit mines; and • VBKOM additions to CAT methodology include: bench utilisation for the purpose of practical rates and considering x2 waste excavators as a rule of thumb. The selected excavators were tested for their suitable application for the 5 m x 5 m x 2.5 m SMU. The criteria included: • Equipment size practicality: − SMU block height should be lower than the maximum digging depth − SMU block height should be greater than or equal to 50% of the productive BH − SMU block width (“X” and “Y”) should be greater than the track width − Loading bucket width should be less than 75% of the “X” extend or “Y” extend • Productivity considerations: − SMU volume should be larger than 10 standard bucket loads − SMU should be the same or larger than the smallest block model block − SMU blocks should preferably be an interval of the original block model sizes The CAT 395 and CAT 6015B were identified as suitable for mining the SMU size of 5 m x 5 m x 2.5 m, hence 2.5 m flitches for ore were identified as both practical and productive, which is supported by historically proven application observed within South African open-pit hard rock mines (refer to Figure 12-44). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 292 SMU Matrix for Practical and Productive Loader Selection Figure 12-44: SMU matrix for practical and productive loader selection. A practical equipment rate calculator which has been benchmarked and tested within hard rock open mines, was utilised to calculate the rates for the selected loaders. The CAT 395 was matched with a 41-tonne articulated truck, whereas the CAT 6015B was matched with a 64-tonne rigid off-highway truck. The scheduled rates per material time were as follows: • Spodumene : 330 t/hr – CAT 395 • Country rock : 676 t/hr – CAT 6015 • Overburden : 667 t/hr – CAT 6015 • Overburden : 512 t/hr – CAT 395 • Metasediment : 678 t/hr – CAT 6015 • Metavolcanite : 678 t/hr – CAT 6015 • Plagioclase sill : 678 t/hr – CAT 6015 • Sulphidic schist : 678 t/hr – CAT 6015 Destination scheduling determined the number of trucks required for ore and waste mining throughout the LOM. Deswik LHS (Landform and Haulage System) modelled the haulage profiles, with the haulage networks for Syväjärvi and Rapasaari illustrated in Figure 12-45 and Figure 12-46, respectively. At Syväjärvi, ore will be transported to the S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 293 Syväjärvi stockpile, located next to the Syväjärvi OVB dump. From there, it will be rehandled and hauled to the Päiväneva ROM stockpile facility at Rapasaari towards the south. Meanwhile, ore mined at Rapasaari will be directly hauled to the finger stockpiles at the Päiväneva ROM stockpile facility, where it will be blended before being fed into the crusher. Syväjärvi Haulage Routes Figure 12-45: Syväjärvi haulage routes. 500 m Boundary Waste Rock Dump Hazardous Dump Overburden Dump Overburden Dump Overburden Dump Pit Stockpile Haul road Haul road Haul road Haul road Haul road Haul road Haul road S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 294 Rapasaari Haulage Routes Figure 12-46: Rapasaari haulage routes. The load and haul input parameters are presented in Table 12-31. The truck loading times were derived from VBKOM experience and Lofty internal benchmarks. Table 12-31: Syväjärvi and Rapasaari load and haul input parameters. Description Unit Value Truck maximum speed limit – flat km/h 40 Truck maximum speed limit – downhill km/h 30 Truck loading time (manoeuvre, spot, load, move away) CAT 745 & CAT 395: Ore min 4.3 CAT 745 & CAT 395: Country Rock min 3.6 CAT 745 & CAT 395: OVB min 3.3 CAT 745 & CAT 395: Other Hard Rock min 3.1 CAT 775 & CAT 6015: Country Rock min 4.2 CAT 775 & CAT 6015: OVB min 3.9 500 m WRD Phase 2 WRD Phase 1 Hazardous WRD OB Dump East OB Dump West OB Dump West Pit Stockpile Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road Haul Road

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 295 Description Unit Value CAT 775 & CAT 6015: Other Hard Rock min 4.2 Truck dump time (turn, stop, reverse, lift, empty, lower bucket) min 1.24 Rolling resistance no 2 Figure 12-47 and Figure 12-48 illustrate the annual truck requirements for Syväjärvi and Rapasaari, respectively. In 2026, a total of 11 trucks will be required for Syväjärvi, while in 2031, 10 trucks will be needed for Rapasaari, despite its higher mining volumes. Notably, Rapasaari operates with two additional waste trucks and two fewer ore trucks, as the hauling distance to the finger stockpiles is significantly shorter and the operating hours are higher compared to Syväjärvi. Syväjärvi Truck Fleet Figure 12-47: Syväjärvi truck fleet. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 296 Rapasaari Truck Fleet Figure 12-48: Rapasaari truck fleet. 12.4.12 Labour Mining operations will be carried out by contractors, with labour staffing in accordance with the terms outlined in the 2025 tender submission documents. Keliber will appoint key personnel to management positions to oversee and closely monitor the mining activities. It is anticipated that Keliber will fill the following positions: • 1x Mine Manager • 1x Mine Planning Engineer • 2x Mine Geologists • 1x Mine Surveyor • 1x Mine Supervisor • 1x Geotechnician • 1x Technician 12.4.13 Mining Costs [SR4.3(vii), SR5.6(iii)] Mining operating costs for pit optimisation have been provided by Keliber, sourced from mining contractor quotes. For detailed unit mining costs, please refer to the Financial Evaluation chapter. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 297 13 PROCESSING AND RECOVERY METHODS Mined ore will be beneficiated at the Päiväneva concentrator located near the Rapasaari mine. Flotation concentrate will be transported to the Keliber lithium refinery where lithium hydroxide monohydrate will be produced as final product. The selected overall flowsheet comprises a conventional spodumene concentrator which includes crushing, ore sorting, grinding, and spodumene recovery by flotation. Flotation concentrate is calcined to convert alpha-spodumene to beta-spodumene. The converted spodumene concentrate will be processed via the patented Metso Outotec soda pressure leach to produce lithium hydroxide monohydrate. 13.1 Concentrator 13.1.1 Concentrator Throughput and Design Specifications [§229.601(b)(96)(iii)(B)(14)(ii)] Concentrator process design is based on the results of the testwork described in the 2022 FS. Metso Outotec have used the testwork data as a basis to provide basic engineering for the spodumene concentrator. The concentrator is designed for a nominal ore throughput of 680,000 tpa and a design throughput of 815,000 tpa, with a head grade before ore sorting of 1.13% Li2O, and 1.2% Li2O after ore sorting. The design basis for the spodumene concentrator is to produce a flotation concentrate containing 4.5% Li2O for the downstream lithium hydroxide production process. In the production phase, the lithium oxide grade of the concentrate will be a process optimisation point depending on ruling economic factors. In this regard, testwork and design have covered the concentrate grade range from 4.5% to 6.0% Li2O. Keliber testwork programmes have revealed iron, arsenic, and phosphate to be the main impurities of the spodumene flotation concentrate for the downstream process. The maximum levels have been indicated to be 2% for Fe2O3, 50 ppm for As, and 0.4% for P2O5. Concentrate will be dewatered and filtered to have an average moisture content of 10%. The indicated moisture level is the highest allowed moisture for the concentrate preheating phase. Gravity concentration to produce a Nb-Ta concentrate is not included in the flowsheet of the concentrator because it was found not to be economically feasible for Syväjärvi ore. However, the required space for a gravity circuit has been reserved within the concentrator building 13.1.2 Process Description The flowsheet of the spodumene concentrator includes the following unit process operations: • ROM pad for short-term ore storage before feeding the primary crusher; • Material handling equipment to feed the blasted ore to the primary crusher by ore trucks or front-end-loader; • Primary crushed ore silo with 20 hours capacity; • Crushing to produce a crushed product size of 80% passing (P80) 12 mm; • Ore sorting to remove black waste rock and increase the lithium oxide grade of the concentrator ore feed; • Rod mill feed silo with 12 hours capacity at the design throughput rate of 100 t/hr; S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 298 • Rod milling in open circuit. The 3.0 x 4.45 m effective grinding length (EGL) rod mill will be equipped with a 470 kW motor; • Ball milling in a closed circuit with hydrocyclones to produce P80 grind size of 150 μm to flotation feed. The 3.6 x 5.6 m EGL ball mill will be equipped with a 1,100 kW motor; • Low intensity magnetic separator to remove process iron and magnetic gangue minerals before de-sliming; • Two-stage de-sliming before spodumene flotation. The first stage de-sliming cluster will include seven ten- inch hydrocyclones (five operating and two standby) and nine six-inch hydrocyclones in the second de-sliming stage (six operating and three standby); • Pre-flotation to reject apatite, micas, and hornblende. Pre-flotation will be operated in reverse flotation mode, where flotation overflow is rejected and pumped to tailings handling; − Pre-flotation includes roughing and one-stage cleaning flotation. Rougher flotation includes four 20 m3 tank cells in series and cleaning includes two 1.5 m3 tank cells; • Rougher scavenger flotation (5 x 50 m3 tank cells) to produce spodumene rougher concentrate; • Four-stage cleaning flotation (13 x 10 m3 tank cells) to produce final spodumene flotation concentrate; • Dewatering of final spodumene concentrate by thickening (13 m diameter) and a pressure filtration (PF 55/60 M15) to obtain final concentrate with a moisture content of 10%; • Reagent dosing system for the concentrator; • Tailings from the concentrator will be deposited in conventional tailing ponds; and • Tailings from the Keliber lithium refinery (analcime). A simplified block diagram of the concentrator is presented in Figure 13-1. Simplified Block Flow Diagram of Päiväneva Concentrator Figure 13-1: Päiväneva concentrator – simplified block flow diagram.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 299 13.1.2.1 Primary Crushing and Raw Material Storage Ore at a top size of 700 mm is fed from the ROM pad to the feed bin by a front-end loader or ore trucks. Ore is fed to the primary jaw crusher via a scalping screen. Undersize rocks will bypass the crusher while oversize rocks are crushed. A rock breaker will be installed next to the jaw crusher to handle blockages in the jaw crusher. Primary crushing capacity will exceed downstream secondary crushing capacity as it is to be utilised only on day shift. Crusher feed will be measured and automatically controlled. Crusher product particle size is approximately 70 mm. The bypass stream and crushed ore report to a sacrificial conveyor equipped with a tramp iron magnetic separator and metal detector. Tramp metal is collected and recycled off-site. Crushed ore will report to a storage silo with 20 hours live capacity. The primary crushing building will be equipped with a floor pump for housekeeping purposes and a bridge crane and hoist for maintenance purposes. A centralised dedusting system will be installed for personnel safety and housekeeping purposes. Suction points will be mounted at rock transfer points. Dust-laden air will be filtered, with the filter discharge being recycled to the process. 13.1.2.2 Ore Sorting and Secondary Crushing The basic principle of ore sorting is shown in Figure 13-2. Basic Ore-Sorting Operating Principle Figure 13-2: Basic ore-sorting operating principle. Different sensor technologies can be incorporated into ore sorting including Colour (reflection, absorption, transmission), Laser (monochromatic reflection/absorption), NIR spectrometry (reflection, absorption), Electromagnetic (conductivity, permeability), Radiometric (radiation), X-Ray Fluorescence, and XRT. XRT is based on relative atomic density differences and has been selected based on test results. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 300 Primary crushed ore is fed to a double-deck vibrating screen, which separates ore into three fractions. Oversize material of P80 approximately 80 mm is forwarded to the secondary crusher, with secondary crushed material being recycled to the double-deck screen. Material from the second screen plate is directed to the ore-sorting separation screen. Oversize reports to a washing stage ahead of the coarse ore sorter, while undersize reports to a washing stage ahead of the fine ore sorter. Each ore sorter separates waste from ore in its respective size fraction. Rejected material is fed to a stockpile for transport out of the concentrator area, while accepted material is combined and forwarded to tertiary crushing. The undersize fraction from the double-deck screen is directed to the fine bypass conveyor and is combined with tertiary crushed material. 13.1.2.3 Tertiary Crushing Secondary crushed and sorted ore reports to a vibrating screen. The oversized material, P80 approximately 25 mm, is directed to the tertiary crusher. Tertiary crusher discharge is circulated back to the sorting accept conveyor. Vibrating screen undersize, P80 12 mm, is conveyed to the mill feed silo. 13.1.2.4 Grinding and Classification The grinding circuit includes a 3.0 m x 4.45 m rod mill fitted with a 470 kW motor and a 3.6 m x 5.6 m ball mill fitted with a 1,100 kW motor. The rod mill is designed to process 83 tonnes/hour at 75% solids in open circuit ahead of secondary ball milling. The ball mill operates at 65% solids in closed circuit with cyclone and screens. Target solids content for classification is 50 wt%. The cyclone battery comprises two operational and one standby cyclone. Cyclone overflow at target P80 particle size reports to the grinding product pump sump. Cyclone underflow is pumped to ultrafine screens (three operational and one standby). Screen undersize at target P80 of 150 μm flows by gravity to the grinding product pump sump with cyclone overflow. Oversize material from the screens reports to the ball mill. 13.1.2.5 Magnetic Separation Final milled product is pumped to magnetic separation using a low-intensity magnetic separator (LIMS) to avoid spodumene losses. The magnetic fraction that includes process iron and magnetic minerals will be pumped to a lined tailings pond together with pre-float concentrate. The non-magnetic fraction is forwarded to de-sliming. 13.1.2.6 De-Sliming and Pre-Flotation De-sliming consists of two pumps and de-sliming hydrocyclones installed in series. The non-magnetic stream from the magnetic separator is pumped to the first de-sliming hydrocyclone cluster, which consists of seven 10-inch hydrocyclones (five operational and two on standby). Primary cyclone underflow is directed to a pre-flotation conditioner. Primary cyclone overflow is pumped to the second de-sliming hydrocyclone bank, which consists of nine 6-inch hydrocyclones (six operational and three on standby). Secondary cyclone underflow is combined with primary cyclone underflow in the conditioner tank. Secondary cyclone overflow is pumped to the spodumene tailings pump sump. The combined de-sliming cyclone underflow is mixed in the first conditioner tank with sodium hydroxide for pH regulation to approximately pH 10, and then it is fed to the second conditioner. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 301 The purpose of pre-flotation is to decrease the amount of phosphorous in the final concentrate. Fatty acid is applied to the second conditioner tank. Slurry gravitates to the pre-flotation stage, with emulsifier being fed to the feed box. Pre-flotation includes rougher with four TC-20 tank cells in series and cleaner flotation with two OK-1.5 in series. Combined underflows from the pre-float rougher and cleaner stages are pumped to the spodumene flotation feed thickener. The overflow from the cleaner flotation is pumped to a separate lined TSF. Solid’s content of the tailings is approximately 17% and the mass recovery is 1% with apatite recovery of 32%. 13.1.2.7 Flotation Feed Thickening Pre-float tailings are thickened to 60% solids in an 18 m thickener ahead of spodumene rougher flotation. Thickener underflow is pumped to rougher flotation via the attrition conditioner and the overflow is pumped to the water treatment plant and, from there, to the process water tank. 13.1.2.8 Spodumene Flotation Thickened spodumene flotation feed at 60% solids is pumped to an attrition conditioner. In the first conditioner, pH is regulated with sulphuric acid, targeting a pH of 7. From the first conditioner, slurry is directed to the second conditioner where fatty acid is introduced to the slurry. Emulsifier is added to the slurry from the attrition conditioner as it flows to rougher flotation, which comprises one bank of five 50 m³ rougher flotation cells. The combined concentrate from rougher flotation is pumped to the first cleaner flotation. Tailings are pumped to the tailings thickener. Spodumene cleaner flotation includes four stages. The first stage includes five 10 m³ cells, the second stage has three 10 m³ cells, the third stage has three 10 m³ cells, and the fourth stage has two 10 m³ cells. The underflow of the first cleaner is pumped back to the flotation feed thickener. The overflow is pumped to the second cleaner flotation. The concentrate from the second cleaner is pumped to the third cleaner flotation, and tailings flow back to the first cleaner via gravity. Concentrate from the third cleaner is pumped to the fourth cleaner flotation, and tailings flow back to the second cleaner via gravity. Concentrate from the fourth cleaner is pumped to the concentrate thickener, and tailings flow back to the third cleaner via gravity. 13.1.2.9 Tailings Thickening Tailings from rougher flotation and de-sliming cyclones are pumped to the 12 m diameter tailing’s thickener. Thickener underflow at 60% solids is pumped to the tailings dam. The overflow is pumped to the water treatment plant and, from there, to the process water tank. 13.1.2.10 Concentrate Thickening The final flotation concentrate is pumped to the 13 m diameter concentrate thickener. The overflow from the thickener is pumped to the water treatment plant and, from there, to the process water tank. The underflow at 60% solids is pumped to the concentrate filter feed tank by an underflow pump. 13.1.2.11 Concentrate Filtration and Concentrate Storage Filter cake at 10% design moisture is dropped to cake discharge chutes after filtration and then conveyed to the concentrate stockpile. The filtrate is pumped to the concentrate thickener feed box. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 302 The concentrate storage facility will provide sufficient material for two days (48 h) of operation, providing a buffer between the concentrator and the Keliber lithium refinery. Spodumene concentrate will be loaded by a front-end loader and transported by truck to the receiving concentrate storage at the Keliber lithium refinery. 13.1.2.12 Particle Size and On-Stream Slurry Analysers Metso Outotec PSI 500 particle size analyser is an online size measurement system for mineral slurries. It is used in controlling wet mineral processes, primarily grinding, classification, re-grinding, and thickening. Samples for particle size analyses are taken from the LIMS feed, the first de-sliming cyclone overflow, and the second de-sliming cyclone overflow. Courier 8 is an on-stream slurry analyser that can measure element concentrations in slurries for up to 12 samples. It was designed for on-stream measurement of light elements and is suitable for Li measurements. Simultaneously, up to 20 element concentrations and solids content can be measured from one sample. Samples for the element analyses are taken from the sample divider, pre-float tailings, spodumene flotation tailings, spodumene rougher flotation concentrate, spodumene first cleaner flotation tailings, and final concentrate. Samples from slimes, whole pre-float concentrates stream, and sampler whole magnetic fraction stream report to the multiplexer. Streams from rougher concentrate, first cleaner tailings, and all the analysed streams are returned to the process by pump. Samples from the final concentrate will be pumped back to the concentrate pump sump. 13.1.3 Process Design Criteria Key concentrator process design criteria are shown in Table 13-1. Table 13-1: Key process design criteria – concentrator. Description Unit Value Plant design capacity tpa 815,000 Ore moisture % 5 Head grade (ROM ore) % Li2O 1.13 Head grade (after ore sorting) % Li2O 1.20 Lithium recovery % 88 Sorting and crushing availability % 85 Crushing circuit P80 mm 12 Concentrator availability % 93 Bond Abrasion index 0.4 Bond Crushing Work index (Syväjärvi ore) kWh/t 12.4 ± 1.9 Bond Rod Mill Work index (Syväjärvi ore) kWh/t 15.3 Bond Ball Mill Work index (Syväjärvi ore) kWh/t 18.9 De-sliming 1 cut size μm 30 De-sliming 2 cut size μm 7 Pre-float feed P80 μm 130 Pre-float slurry density % solids 30

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 303 Description Unit Value Spodumene flotation feed P80 μm 150 Spodumene flotation slurry density % solids 30 Final spodumene flotation mass pull % 23.5 Final spodumene concentrate grade % Li2O 4.5 Final spodumene concentrate production t/hr 23.5 Bond Crushing Work index (Syväjärvi ore) kWh/t 12.4 ± 1.9 Bond rod mill index of Länttä ore was determined as 12.6 kWh/t and the ball mill work index at 17.1 kWh/t. Bond rod and ball mill indexes are 15.3 kWh/t and 15.2 kWh/t for Rapasaari ore, respectively. The geo-metallurgical study showed that the grindability only varies a little between deposits and correlates with spodumene grade (higher spodumene grade means higher resistance for grinding). As the mineralogical differences between the deposits are small, it was considered that grindability would be within these ranges in untested Emmes and Outovesi ore. 13.1.4 Requirements for Energy, Water, and Consumables [§229.601(b)(96)(iii)(B)(14)(iii)] The following services will be provided: • Flotation air; • Plant and instrument air; • Raw water; • Process water; • Sealing water; • Warm water; • Potable water; and • Fire water. 13.1.4.1 Power In terms of the 2022 FS Report, electric power to the Päiväneva concentrator will be supplied from a local distribution grid at Kaustinen, owned and operated by a local utility company. At the supply end, the transmission cable will be connected to the 110 kV distribution grid through a 16 MVA main transformer. Power will be transmitted to the Päiväneva concentrator via a 33 kV underground transmission cable. At the Päiväneva site, the external power supply will be connected to a 33 kV main distribution switchgear, from which the power will be further distributed to local process substations. The Päiväneva concentrator power requirements are estimated at 11,410 kW. 13.1.4.2 Raw Water Pumping and Treatment A raw water pumping station will be constructed at the Köyhäjoki River for raw water pumping to the chemical raw water treatment plant. Raw water pump dimensioning is based on an estimated flow rate of 150 m3/h. Raw water from Köyhäjoki River will be micro-filtered and preheated to 10°C before being chemically treated with a precipitation S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 304 chemical and pH-adjusted with NaOH. Preheated and chemically treated water will be pumped to three DynaSand contact filters for humus removal. Sludge handling includes a Lamella clarifier, sludge thickener, and drying by sludge centrifuge. 13.1.4.3 Process Water Treatment The process water treatment plant consists of two similar dissolved air flotation (DAF) units, which remove suspended solids and colloid material from liquid by means of air bubbles that attach to agglomerates and raise them to surface. The flotation basin is equipped with a surface sludge removal system. Sludge is removed by gravity. Clarified water will be pumped to the process water tank. 13.1.4.4 Pre-Float Water Treatment The techniques applied to remove arsenic species include oxidation, coagulation-flocculation, flotation, and pressure sand filtration. The first oxidation stage is done by a prefabricated bottom diffuser/aeration system ahead of an aerated coagulant tank. Common coagulants used for arsenic are iron salts. The precipitation of ferric arsenate is commonly done at pH 4–5. To ensure the stability of ferric arsenate, an excess amount of iron must be dosed compared to the amount of arsenic. After coagulation and flocculation, the suspended solids are removed by micro-flotation followed by pressure sand filtration as a final polishing stage. 13.1.4.5 Excess Water Treatment In the recycled water treatment plant, the equipment will be similar to the pre-float water treatment, with larger equipment sizes. 13.1.4.6 Potable Water Potable water will be drawn from the municipal water system. 13.1.4.7 Fire Water Fire water pumps will be located at the water treatment plant. Water from fresh water and process circulation ponds can be used as fire water. 13.1.4.8 Online Water Management Tool Development for the Concentrator Keliber have started development work for concentrator water management. The purpose of the management tool is to provide real-time, concentrator-wide water balance management, control, and reporting including what-if scenarios. The tool will summarise real-time weather data and online process data from the concentrator automation system to provide visualisation and simulation as well as reporting. 13.1.5 Reagents and Consumables Table 13-2 summarises the reagents and consumables for the concentrator. Table 13-2: Concentrator reagents and consumables. Description Unit Value Rod Mill Grinding Media Consumption g/t 593 Ball Mill Grinding Media Consumption g/t 690 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 305 Description Unit Value Caustic Soda Consumption (NaOH) g/t 500 Sulphuric Acid Consumption (H2SO4) g/t 50 Rape Fatty Acid Consumption g/t 1,390 Emulsifier Consumption g/t 155 Flocculant Consumption g/t 80 13.1.6 Labour Requirements The labour requirement for the Päiväneva concentrator plant is shown in Table 13-3. Table 13-3: Labour requirements for the Päiväneva concentrator plant. Manning Plan Number of Employees Site Manager, Concentrator 1 Development Engineer, Concentrator 1 Operation Engineer, Concentrator 1 Process Engineer, Concentrator (incl. Tailings and Dams) 1 Process Engineer, Tailings and Dams 1 Geotechnician, Tailings and Dams 1 Metallurgist 1 Shift Supervisors, Concentrator 5 Operators, Concentrator (experienced) 10 Site Manager, Concentrator 15 13.1.7 Concentrator Plant Costs [SR4.3(vii), SR5.6(iii)] For detailed operating costs for the Päiväneva concentrator plant, please refer to the Financial Evaluation chapter. 13.2 Lithium Hydroxide Refinery 13.2.1 Throughput and Design Specifications [§229.601(b)(96)(iii)(B)(14)(ii)] [SR5.3(iii)] Keliber’s lithium refinery at the KIP, Kokkola, is designed with a feed capacity of 156,000 tpa of spodumene concentrate, which translates to an annual lithium hydroxide monohydrate production of 15,000 tonnes at 99.0% LiOH.H2O purity for the final product. A simplified block flow diagram of the Keliber lithium refinery is shown in Figure 13-3. Key unit processes in the production of lithium hydroxide are summarised as follows: S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 306 13.2.1.1 Concentrate Receipt The spodumene concentrate will be transported approximately 66 km by truck to Keliber lithium refinery at Kokkola Harbour. The capacity of the receiving concentrate storage facility is sufficient for two weeks of operation. The storage capacity will provide flexibility to blend different concentrate qualities and ensure stable operation for the downstream lithium hydroxide production process. The blending and homogenisation might be required to control the lithium oxide grade and levels of impurities in the feed to the Keliber lithium refinery. Keliber lithium refinery simplified block flow diagram Figure 13-3: Keliber lithium refinery BFD.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 307 13.2.1.2 Spodumene Calcination (Conversion) Alpha-spodumene is converted to beta-spodumene in a direct heated rotary kiln. The rotary kiln will be fired with liquid petroleum gas (LPG) and operate at 970–1,075°C. A rotary cooler unit is used to cool the converted beta-spodumene down to between 80°C and 90°C before pulping with the circulating liquids filtrate and wash water from the 1st stage (autoclave) residue filtration, in two agitated tanks in series. 13.2.1.3 Pressure Leaching of Beta-Spodumene Primary leaching is a soda-leaching autoclave operating at a temperature of 215°C and 20–22 bar gauge pressure. High-pressure steam is used to maintain the temperature. Beta-spodumene will be converted to Li2CO3 and analcime as a by-product according to Equation 15: Equation 15: Conversion of beta spodumene to lithium carbide. 2𝐿𝑖𝐴𝑙𝑆𝐼2𝑂6 (𝑠) + 𝑁𝑎2𝐶𝑂3 + 2𝐻2𝑂 ↔ 𝐿𝑖2𝐶𝑂3 (𝑠) + 2𝑁𝑎𝐴𝑙𝑆𝑖2𝑂6 ∙ 𝐻2𝑂 (𝑠) Slurry from the autoclave is released by pressure difference into two-stage flashing. 13.2.1.4 Soda Leach Residue Filtration Autoclave slurry solid/liquid separation is done with two parallel pressure filters. Soda leach residue consists mainly of solid analcime (NaAlSi2O6*H2O), Li2CO3, quartz, and other gangue minerals. This is pulped with leach residue wash water and forwarded to LiOH conversion. Part of the filtrate is recycled for filter manifold flushing after the filling/filtration step. The manifold flushing stage pushes residual solids to the chambers and filtrate to the filtrate tank. Residual manifold flushing liquid is released from the pipeline to an agitation tank, from which the slurry is returned to the filter feed tank. Spent cloth wash waters are recycled to the cake wash tank. Part of the filtrate and wash water is fed back to calcine grinding and pulping. The rest of the water is fed to effluents, to control the leach circuit water balance. The amount of process bleed to effluent is greatly dependent on the lithium grade of the calcined beta-spodumene feed. The lower the Li2O grade, the higher the consumption of cake wash waters. 13.2.1.5 LiOH Conversion Pulped soda leach slurry, lime slurry, and wash waters from leach residue filtration are fed to conversion reactors. Conversion is done preferably below 30°C to minimise solubilisation of aluminium and silica. Li2CO3 in soda leach solids reacts with Ca(OH)2 according to Equation 16: Equation 16: Cold conversion reaction. 𝐿𝑖2𝐶𝑂3 (𝑠) + 𝐶𝑎(𝑂𝐻)2 (𝑠) ↔ 2𝐿𝑖 + 𝑂𝐻− + 𝐶𝑎𝐶𝑂3 Only LiOH is water soluble, and the others are insoluble. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 308 13.2.1.6 Leach Residue Filtration and Handling After conversion leach, the slurry is fed to leach residue filters for solids/liquid separation using two parallel pressure filters. Filter cake consisting mainly of analcime, calcium carbonate, quartz, and other gangue is discharged. Filtrate is fed through polishing filtration to ion exchange. Wash water is fed to lime slurrying and 1st stage residue pulping. 13.2.1.7 Polishing Filtration Feed from the secondary conversion is fed through a polishing filtration stage where suspended solids are removed from the lithium hydroxide solution. The polishing filtration is carried out in LSF-type polishing filters. One will be in operation, and one will be on standby. 13.2.1.8 Ion Exchange Ion exchange is done in three fixed-bed columns connected in series for removal of elevated multi-elements e.g. Ca and Mg from the solution before LiOH crystallisation. The regeneration cycle starts with the pre-wash stage, where 2 M sodium hydroxide solution is fed to the column, mainly to displace most of the lithium bound to the resin with sodium ions. After pre-wash follows the first displacement wash with demi water. A short backwash with demi water is done after the first displacement wash to backwash the resin bed and remove any air bubbles and possible channelling. Elution of the metals is done with excess 2 M hydrochloric acid solution. The resin functional groups are simultaneously converted to acid form. The acidic eluate stream, containing mainly calcium, sodium, and potassium as chlorides, is fed to effluent treatment. After a second displacement wash with demi water, neutralisation of the resin to lithium form is done with process lithium hydroxide solution. After regeneration, the column is connected as the last column in the series. 13.2.1.9 Crystallisation of Lithium Hydroxide Lithium hydroxide is crystallised from the lithium hydroxide solution by means of pre-evaporation in a mechanical vapour recompression (MVR) falling film evaporator, followed by an MVR crystalliser. Lithium hydroxide LiOH*H2O is crystallised according to Equation 17: Equation 17: Crystallisation reaction. 𝐿𝑖+ + 𝑂𝐻− 𝐻2𝑂 ↔ 𝐿𝑖𝑂𝐻 ∙ 𝐻2𝑂 (𝑠) Lithium hydroxide slurry from the crystallisation stage is fed to a centrifuge where solids are separated from the mother liquor and washed. Moist cake is dried in a fluidised bed dryer and packed into big bags for shipment to market. Most of the mother liquor is circulated back to conversion in order to control the soluble concentration of Al, CO3 2- , and Si in solution. A small portion of the mother liquor is fed to bleed treatment: carbonation and conversion. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 309 13.2.1.10 Crystallisation Bleed Treatment Lithium in crystallisation bleed liquor is recovered as lithium carbonate. In the carbonation step, carbon dioxide is fed pH-controlled to mother liquor bleed from the crystallisation batch reactor. Lithium carbonate precipitates from the concentrated lithium hydroxide solution according to Equation 18: Equation 18: Formation of lithium cabornate. 2𝐿𝑖+ + 2𝑂𝐻 + 𝐶𝑂2 (𝑔) ↔ 𝐿𝑖2𝐶𝑂3(𝑎𝑞, 𝑠) + 𝐻2𝑂 Aluminium also precipitates with the lithium as carbonate according to Equation 19: Equation 19: Formation of aluminium carbonate. 3𝐴𝑙3+ + 6𝑂𝐻− + 3𝐶𝑂2 ↔ 𝐴𝑙2(𝐶𝑂3)2 (𝑠) + 3𝐻2𝑂 Carbonation is done at elevated temperatures to minimise lithium solubility. 13.2.1.11 Effluent Treatment Bleed from soda process filtrate and IX eluate acids are constantly pumped to the effluent storage tank. In the effluent pre-treatment area, lithium is recovered from the effluent liquor by precipitation as lithium phosphate with sodium phosphate solution feed. After precipitation, filtration and washing steps follow. Filtrate is further treated in an electrochemical water treatment process in order to precipitate soluble arsenic from the effluent stream. Solids in the effluent stream are removed by dissolved air flotation and pressure sand filtration. The treated water is discharged to a municipal wastewater treatment plant. 13.2.2 Process Design Criteria Key process design criteria for the lithium hydroxide chemical plant are shown in Table 13-4. Table 13-4: Key process design criteria – lithium hydroxide chemical plant. Parameter Unit Value Processing Rate (dry) tpa 156,000 Concentrate Grade % Li2O 4.5 Concentrate Moisture % H2O 11 Concentrate Fineness P80 microns 150 Plant Operating Time h 7,500 Overall Plant Availability % 85.6 LiOH . H2O Production tpa 15,000 Recovery from concentrate to LiOH . H2O product (incl. calcining) % 86 The final product specifications are detailed in Table 13-5. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 310 Table 13-5: LiOH . H2O product specifications. Parameter Unit Technical Grade** Target Specification (Battery Grade) in Design Criteria LiOH . H2O wt% >99 >99 LiOH wt% > 56.5 <56.5 CO2 wt% <0.35 <0.3*** Na wt% <0.03 <0.0058*** Al <0.005 K wt% <0.01 <0.0025*** Fe wt% <0.001* <0.0005 Cu wt% <0.0005 Cr <0.0005 Zn wt% <0.0005 Ca wt% <0.02* <0.001 SO4 2- wt% <0.03 <0.0011*** Cl- wt% <0.01 <0.004*** Si wt% <0.003 Insoluble in HCl <0.005 * Calculated ** Halmek Lithium *** Crystallisation plant vendor 13.2.3 Requirements for Energy, Water, and Consumables [§229.601(b)(96)(iii)(B)(14)(iii)] 13.2.3.1 Power In terms of the 2022 FS Report, electric power to the lithium chemical plant will be provided by a subsidiary of a local power utility company. To enable continuous production during planned grid maintenance works, the plant will have two independent 20 kV connections to the external grid. Both power connections will be able to supply the plant at full capacity. From the main distribution switchgear, the power will be further distributed to the plant process substations by means of 20 kV underground cables. Finally, the power will be transformed to 400 V and 690 V levels in the proximity of the consumers. Kokkola plant power requirements are estimated at 12,450 kW. 13.2.3.2 Keliber lithium refinery – Site Services The KIP site will provide existing infrastructure to supply the required site services. All process water qualities can be purchased from the water treatment plant operated by KIP Service Oy and process steam from the power plant of Kokkolan Energia Oy. The following site services will be provided: • Process water;

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 311 • Demineralised water; • Cooling water; • Sealing water; • Potable water; • Fire water; • Compressed and instrument air; and • Process steam. 13.2.4 Labour Requirements The labour requirements for the Keliber lithium refinery are presented in Table 136. Table 136: Labour requirements for the Keliber lithium refinery. Manning Plan Number of Employees Site Manager, Lithium Refiner 1 Process Engineer, Plant 1 Hydrometallurgist, HSC Chemistry™ Software 1 Metallurgist 1 Process Engineer, Calcining 1 Production Engineer, Keliber lithium Refinery 1 Process Engineer, LiOH 1 Process Engineer, Soda Leaching 1 Water Treatment Engineer, Effluent Treatment Plant (ETP) 1 Production Supervisor 2 Shift Supervisors, Keliber lithium Refinery 5 Operators, Keliber lithium Refinery (experienced) 15 Operators, Keliber lithium Refinery (new) 40 Keliber lithium Refinery Total 71 13.2.5 LiOH Processing Costs [SR4.3(vii), SR5.6(iii)] For detailed operating costs, please refer to the Financial Evaluation chapter. 13.3 Plant Commissioning and Ramp-Up [§229.601(b)(96)(iii)(B)(14)(iv)] First ore is planned to be processed through the Päiväneva concentrator in January 2026. Six months have been allowed to reach design capacity as shown in Figure 13-4. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 312 Hot commissioning of the Keliber lithium refinery is planned for February 2026 for three months, whereafter ramp-up at 15% throughput with concentrate from the Päiväneva concentrator is planned from May 2026 to 100% after nine (9) months have been allowed to reach design capacity, as shown in Figure 13-4. This ramp-up schedule is in line with the 2022 FS study ramp-up for internal Keliber ore, bar the fact that the refinery will not start up with third-party spodumene anymore (previously planned for up to 12 months) but will operate from day one with ore originating from Keliber’s own operations. The 12-month period with third-party concentrate would have allowed time to address all the operating issues, usually experienced during the start-up of high-pressure leaching plants, with limited impact on the production schedule of the mine or concentrator plant. Plant Ramp-Up Schedules Figure 13-4: Plant ramp-up schedules. The following sensitivities on the ramp-up assumptions have been tested to evaluate the impact deviations on the financial viability of the Project: • Increase technical grade pricing for 12-month period; and • Increase 9-month ramp-up to 21 months. The results are discussed in Chapter 18. 0% 20% 40% 60% 80% 100% 120% Jan-26 Feb-26 Mar-26 Apr-26 May-26 Jun-26 Jul-26 Aug-26 Sept-26 Oct-26 Nov-26 Dec-26 Jan-27 CP Plant LHP Plant S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 313 14 PROJECT INFRASTRUCTURE [§229.601(b)(96)(iii)(B)(15)] 14.1 Mine Layout and Operations The open-pit mines and concentrator are situated in Central Ostrobothnia in Western Finland (Figure 2-1 and Figure 2-2). Kokkola is the largest city in the area and the port has all the facilities for overseas shipments; it is ice-free all year. The nearest airport is Kokkola-Pietarsaari, which is serviced by Finnair as well as charter flights. The general layouts of the Syväjärvi mine site are shown in Figure 14-1. The major infrastructure at the Syväjärvi open- pit mine comprises of: • Roads - access roads to the mine, a haul road to the Päiväneva concentrator and internal roads, • Security, • Storm water management, silt traps and safety berms, • Waste rock facility, and • ROM stockpiles areas. The stormwater management infrastructure has been constructed except for the new ROM stockpile area. The construction of the safety berms as indicated in Figure 14-1 has been completed and the silt traps are in construction. The current roads are only suitable for light vehicles. Haul road construction to the ore stockpiles and Rapasaari will commence as soon as the contractor has been appointed. Mining is planned to commence mid 2025 with excavating waste for construction material. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 314 General Layout of the Syväjärvi Mine Site Figure 14-1: General layout of the Syväjärvi mine site. The general layouts of the Rapasaari mine site are shown in Figure 14-2. The following mining infrastructure will be completed during 2025: • ROM pad – indicated in green on Figure 14-2; and • Roads – Haul road from Syväjärvi and road to Kokkola (Figure 14-3). The rest of the mining infrastructure around Rapasaari will be completed in time to execute the LOM plan.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 315 General Layout of the Rapasaari Mine Site Figure 14-2: General layout of the Rapasaari mine site. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 316 Proximity of Syväjärvi to Rapasaari, also Indicating the Haul Road that will be Constructed to Transport the Ore to the Concentrator Plant Figure 14-3: Proximity of Syväjärvi to Rapasaari, also indicating the haul road that will be constructed to transport the ore to the concentrator plant. 14.1.1 Explosives, Fuel Supply, and Storages Fuel storage tanks and refuelling stations are planned to be located at the contractor’s maintenance facility. All trucks will refuel at the refuelling station, while crawler-mounted machinery will be refuelled by a fuel truck. The refuelling station will be designed with a concrete foundation and liner in accordance with the standard SFS 3352:2014/A1:2020. Any spillages will be directed to oil separation chambers for treatment. Explosive storage facilities are planned for both the Syväjärvi and Rapasaari mine sites. The supply and storage of explosives will be managed by the contractor or explosive manufacturers, and the design of the storage facilities will comply with regulatory requirements: • Act on the safe handling and storage of dangerous chemicals and explosives (390/2005); S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 317 • The government decree on safety requirements for the manufacture, handling, and storage of explosives (1101/2015); and • The government decree on the control of the manufacture and storage of explosives (819/2015). 14.2 Surface Infrastructure and Bulk Services The Keliber Lithium Concentrator at Päiväneva is located 18 km from the municipality centre of Kaustinen, in close proximity to the Rapasaari mine site (Figure 14-2). Figure 14-4 shows the overall layout of the Päiväneva Concentrator site. The bulk infrastructure for the Päiväneva concentrator has commenced and includes: • Road constructions and alterations are already completed and include the following: − Road construction to Syväjärvi and Rapasaari mines; − A new road and intersection arrangement to access the Päiväneva plant; and − A new road arrangement at the location of the Kokkola plant. • Raw water pumping station at Köyhäjoki; the civil works are 80% completed and installation is planned to be completed by 5/2025; • Water treatment plant; civil works 60% completed. The installation is planned to be completed by 7/2025; • Stormwater management; construction of the stormwater management infrastructure has been completed. The stormwater management plant is shown in Figure 14-5; • One 19 km 20 kV power transmission line from the Keliber Lithium Project substation in Kaustinen to the Päiväneva site. The progress is 95% and is planned to be completed by 3/2025; • Main electrical substations, electrical distribution, offices, laboratory: − Main electrical substation to be completed 3/2025; and − Electrical distribution to be completed in phases. • Office and laboratory to be completed by 7/2025; and • Contractor laydown areas including the following: − Workshops; − Change houses, and − Mining offices and parking. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 318 • Layout of the Päiväneva Concentrator Site Figure 14-4: Layout of the Päiväneva concentrator site.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 319 • Stormwater Management Plant at the Päiväneva Concentrator Site Figure 14-5: Stormwater management plant at the Päiväneva concentrator site. 14.3 Plant Infrastructure The required infrastructure for the Concentrator and equipment includes: • Crushing, ore storage, and ore sorting; • Grinding and classification; • Magnetic separation; • De-sliming; • Pre-flotation and spodumene flotation; • Concentrate dewatering and filtration; • Concentrate storage: − Tailing ponds: two tailing ponds for processing residues, flotation tailings, and pre-flotation tailings and two water ponds for pit water and process water circuit; and − Small thermal plant to produce heat. Construction of the concentrator commenced in Q4 of 2023 and hot commissioning is planned for January 2026. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 320 The Keliber lithium Refinery is situated in the KIP at Kokkola. A general arrangement (GA) drawing of the Keliber lithium refinery is presented in Figure 14-6, the construction site is pictured in Figure 14, and the overall layout is shown in Figure 14. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 321 GA Drawing of the Keliber lithium Refinery Situated in the KIP at Kokkola Figure 14-6: GA drawing of the Keliber lithium Refinery situated in the KIP at Kokkola. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 322 Photo of the Keliber lithium Refinery Construction Site in the KIP at Kokkola Figure 14-7: Photo of the Keliber lithium Refinery construction site in the KIP at Kokkola.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 323 Overall Layout of the Keliber lithium Refinery at the KIP Site Figure 14-8: Overall layout of the Keliber lithium Refinery at the KIP site. Most of the required external site services for the operation, such as security and fire brigade, are available at the KIP. The plant has all the required infrastructure for concentrate conversion and hydrometallurgical processing including an effluent treatment plant, LPG storage and handling facilities, main electrical substations, electrical distribution, offices, and a laboratory. The infrastructure and engineering designs encompass the required infrastructure for the establishment of processing operations and the surface mine sites at a feasibility level of detail, and all necessary logistics have been considered. 14.4 Tailings Storage Facility The TSF for the Keliber Lithium Project is located within the Päiväneva plant area, to the east of the main mill area and south of the site’s WRD (refer to Figure 14-9). The TSF site is located directly north of ancient forest areas which are home to the Siberian flying squirrel. This has necessitated a redesign of the latter stages of the previous TSF construction configuration to minimise impacts on the squirrels’ habitat. The TSF design work has been undertaken in three stages: a Preliminary Design undertaken in 2020 which sets out the key engineering aspects, a FS design undertaken in 2022 in which additional engineering was completed to advance all aspects of the design, and a construction design (Detailed Design) covering the Flotation Tailings Pond area during 2023. Construction of the recycle water pond was completed in December 2024. AFRY have continued to produce Approved-for-Construction level of documentation to support the construction of the TSF. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 324 Schematic of the TSF Figure 14-9: Schematic of the TSF located within the Päiväneva plant area. The TSF has been designed as three standalone compartments, namely a Flotation Tailings Pond designed to store 8.3 Mt of flotation tailings produced over the LOM, a Pre-Flotation Tailings Pond design to store 58,000 m3 of tailings. Respectively in adjacent cell with capacity of 59,000 m3 will be constructed for magnetic fraction. Water Pond designed to store 0.13 Mm3 of excess water from the Flotation Tailings Pond. The latest designs show these to provide sufficient storage for tailings and pond water over the LOM, however, updates are required to the Pre-Floatation Tailings Pond design to ensure staged construction is appropriate, following a recent decision to segregate Pre-Flotation and Magnetic waste streams and store in separate ponds. The Pre-Flotation Pond design has yet to receive the necessary approvals from the Environmental Regulator to proceed with construction. Keliber have engaged a third-party engineer to review the design work completed to date and provide additional data to the Regulator to prove that the design meets the requirements set out in the EP. Third- party quality control is a requirement in the EP clauses. All required authority approvals are in place for the pre- flotation pond. Final approval for the magnetic fraction pond is pending. Bottom structure is similar to the pre-flotation pond which makes permitting easier. The final approval is expected to be granted this spring, 2025. Acceptance has been received to use peat liner solution for the flotation tailings facility. 350m S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 325 15 MARKET STUDIES [§229.601(b)(96)(iii)(B)(16)] The summary below is based on SFA Oxford’s January 2025 Lithium Market Outlook to 2040. This market analysis conducted for Sibanye-Stillwater covered the period up to 2040. This period will also coincide with the bulk of the financial pay-back period for the Keliber Project. In addition, consensus price forecasts, dated December 2024, as compiled and released by UBS Bank in January 2025, are referenced in the Market Balance & Price outlook section. The Keliber Project is a vertically integrated project that includes the mining of spodumene ore, concentrating the ore, and then conversion of spodumene concentrate into BG. 15.1 Uses of Lithium Hydroxide Monohydrate Lithium hydroxide is used to produce cathodes for rechargeable lithium-ion batteries, especially for use in electric vehicles as it produces the best power density. This results in better battery capacity, better safety features, and a longer-lasting battery than either lithium carbonate batteries or standard lead-acid batteries. It is also used as a thickener in lubricating grease as it is resistant to water and high temperatures and can sustain extreme pressures. It is mainly used in the automobile and automotive industries. Other uses for lithium are in mobile phones, electronic devices, laptops, and digital cameras. Within the Li-ion battery value chain, lithium is used in the manufacturing of cathode materials, electrolytes, and anode materials. 15.2 Market Overview The global market continues to evolve rapidly, with oil supermajors, major mining companies, and auto giants all investing in lithium supply, including Saudi Aramco, ExxonMobil, Rio Tinto, General Motors, and Tesla. This financial clout could have a massive impact on future lithium supply. The first pure-play direct lithium extraction (DLE) operation started production in 2024: Eramet’s Centenario Ratones. Although still in ramp-up phase, this brings DLE technology one step closer to full commercialisation and potentially becoming mainstream or even a supply disruptor. Already dominant downstream, China’s upstream influence is increasing significantly through asset acquisition in Africa and South America, as well as through sizeable expansion of domestic production. China has grown its share of raw lithium supply to 40% (20% domestic; 20% Chinese-owned overseas) from 30% in 2020. Energy storage demand projections have doubled owing to a step change in 2024 when lower battery cell and solar panel costs led to a large increase in demand. Lithium use in energy storage systems is now estimated to reach ~1.4 Mt LCE by 2040, comprising >40% of gross demand growth in the 2030s. Lithium-ion phosphate (LFP) batteries have experienced a resurgence in China, despite previously appearing to be at risk of being phased out, owing to improvements in the energy density of LFP cells, such as by the addition of manganese (LMFP). LFP cathodes (including lithium manganese iron phosphate (LMFP)) are now estimated to comprise ~40% of global battery electric vehicle (BEV) battery demand in 2040. 15.2.1 Demand Outlook The lithium market is estimated to have grown 20% year-on-year in 2024, with demand exceeding the 1 Mt LCE mark for the first time. Growth was driven primarily by automotive batteries in China. Deployment in energy storage systems (ESSs) also boomed as grid-scale developments leveraged improved project economics of solar plus storage, driven by S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 326 a fall in solar and lithium-ion battery prices. Demand is projected to increase to over 4,800 kt LCE by 2040, representing a compound average growth rate (CAGR) of 15% over the 2020–2040 period. Automotive batteries are estimated to account for two-thirds of the demand growth going forward (refer to Figure 15-1). The risk to the outlook from this segment comes from BEVs in the light-duty vehicle segment comprising passenger cars and small trucks/vans. Demand from the ESS segment is the fastest growing (~30% p.a.), but its start from a low base means the segment has historically been a small contributor to global demand. The segment is now expected to represent ~15% of global demand in 2025. The outlook is highly reliant on the outlook for lithium-ion battery demand materialising and so remains highly sensitive to the pace of electrification in the global vehicle fleet and the use of lithium-ion batteries in ESS to alleviate intermittency issues with renewable power generation. Automotive battery production is estimated to rise to 2.7 TWh by 2030 and 4.4 TWh by 2040, from just over 1.0 TWh in 2024. BEVs are expected to represent 90% of the demand growth over this period, owing to their higher production volumes (relative to heavy-duty electric vehicles (HDEVs)) and greater battery packs (relative to all other powertrains). Global Lithium Demand Outlook Figure 15-1: Global lithium demand outlook. HDEVs are the second largest growth segment (~6%), despite their lower production volumes, owing to their larger average battery pack sizes (~250 kWh). Electric bikes, including 3-wheelers, are estimated to contribute 2% to growth in demand with their higher production volumes (predominantly in southeast Asian markets), offsetting their smaller battery pack sizes. Total lithium consumption is projected to grow in line with demand for automotive batteries, to almost 2.0 Mt LCE by 2030 and 3.2 Mt LCE by 2040, with the powertrain split also being proportional to the battery demand split (refer to Figure 15-2).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 327 Lithium-ion batteries are expected to remain the technology of choice over the forecast period, as incremental gains in the technology are expected to outweigh the capital requirements to competitively scale up competing technologies (e.g. sodium-ion). Automotive Battery-Grade Lithium Demand by Powertrain Figure 15-2: Automotive battery-grade lithium demand by powertrain. Almost equal numbers of BEVs and internal combustion engine (ICE)-based vehicles are forecast to be produced in 2040. Full hybrid electric vehicles (FHEVs) and mild hybrid electric vehicles (MHEVs) are rapidly gaining share from IC- only, reaching parity in 2040. China is forecast to produce nearly a third of its light vehicles with an ICE by 2040, but half are hybrids with small batteries, and hybrids are a bridge technology to full electrification, growing to 20% by 2031 and declining thereafter. Europe is the only region currently forecast to be ~100% BEV by 2040. In Japan and the rest of the world (RoW), ICE- based vehicles are still expected to dominate in 2040 (refer to Figure 15-3). China is forecast to remain the largest market, but its dominance is expected to drop. Hybrids (including plug-in hybrid electric vehicles (PHEVs)) are expected to dominate growth (+3.2 million), followed by BEVs (+1.0 million). Hybrids are ahead of BEVs by volume and percentage growth in all regions excluding China. The global light-vehicle production forecast/outlook to the late 2030s is being revised as lower, as downside risks outweigh upside risks. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 328 RoW Light Vehicle Production Figure 15-3: RoW light vehicle production. Several largely interconnected risks combine, leading to a more cautious view of the medium-term market and hence, vehicle production. Interest rates and inflation have risen in many major markets, limiting vehicle affordability. As a result, automakers are more inclined to pursue margin over volume; this strategy is particularly evident whenever supply chain issues arise. The domestic market potential for China is considered more fragile, impacted by the pull- forward of maturing government neighbourhood electric vehicle subsidies, weaker economic expectations, and demographic challenges combined. As a result, automakers are increasingly pursuing growth by export, with tariff risks. Trade tariffs throughout the supply chain are expected to play a growing medium-term role in vehicle sourcing. The EU’s recent decision to impose substantial tariffs on BEV imports from China is the highest profile. Tariffs may adversely impact margins and affordability, but most global original equipment manufacturers (OEMs) are reasonably placed to adjust sourcing patterns appropriately. Automakers aim to meet their near-term CO2 emissions targets with a higher proportion of hybrids in their portfolio and slightly lower BEV shares. By the turn of the decade, BEV shares will need to increase for compliance as emission allowances tighten beyond the reduction capacity of MHEVs and FHEVs. PHEV and extended range electric vehicle (EREV) production forecasts are revised up, while BEV forecasts are revised down owing to increasing EREV model availability and consumer barriers to BEV adoption remaining for longer, namely pricing and charging infrastructure. Both 48V MHEVs and IC-only vehicles are expected to lose ground as, despite their relatively lower purchase price, other stronger hybrid powertrains (PHEV and EREV) deliver the lower CO2 emissions to meet targets. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 329 Global BEV Battery Demand by Chemistry Figure 15-4: Global BEV battery demand by chemistry. Mid- and high-nickel cathodes are forecast to represent two-thirds of EV battery demand. Globally, two-thirds of global BEV battery demand is currently expected to require the lithium hydroxide precursor during cathode fabrication. There is high regional variation in the chemistry outlook, with China much more skewed towards no nickel chemistries (e.g. LFP) and the West skewed towards Ni-rich chemistries (refer to Figure 15-4). In China (and much of Asia, including India), smaller cars are more prevalent, with lower average driving speeds and shorter average driving distances. Furthermore, relative to their American and European counterparts, these consumers are more sensitive to price. This is well suited for the LFP technology. There is high uncertainty in the deployment rate of LFP outside of China owing to the lack of established manufacturing capacity, volatile trade tensions, high rate of research and development (R&D) and industrialisation of competing technologies, and higher performance expected from consumers. The likelihood of disruptive deployments of non-lithium-based batteries in automotive applications is expected to be very low – especially outside China. The largest threat comes from sodium ion. However, successfully scaling sodium- ion cell production to be economically competitive remains a challenge (high setup Capex). Furthermore, in a cycle of low commodity prices, the urgency to substitute away from lithium is reduced. There is also a risk to the long-term chemistry split due to incremental and iterative improvements to existing, lithium-based, cells. Dry electrode manufacturing, for example, could drastically lower the cell manufacturing cost by significantly reducing the input energy required during cathode formulation. Further advancements in the energy density of LFP cells, for example through the addition of manganese, also presents a path to EV/ICE price parity – a point at which EV uptake will happen more organically. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 330 BEVs with cathodes relying on LiOH are predicted to represent 58% of the total market by 2040 (refer to Figure 15-5). The cathodes which require LiOH are typically those with nickel content exceeding 60%, with nickel cobalt aluminium oxide (NCA) and nickel-rich formulations of nickel manganese cobalt oxide (NMC 721 and NMC 811) being current examples. The switch to higher nickel cathodes is motivated by greater electric range requirements as higher nickel content enables higher energy density. Demand for battery-grade LiOH is projected to reach 1,290 kt LCE by 2030 and 1,950 kt LCE by 2040 (Figure 15-5). Growth is forecast to come from the automotive sector, primarily BEVs, but hybrids (PHEVs, FHEVs and MHEVs), e- buses, and e-bikes are also expected to contribute. Despite lower energy density requirements for e-buses and e- bikes, making nickel-rich cathodes unlikely to be used for them, lower specification (cheaper) battery-grade LiOH is expected to generate some demand. Automotive sector LiOH demand is sensitive to EV penetration in Europe and North America, owing to the established industry expertise in these regions being focussed on nickel-based cathode formulations. As such, slower EV sales growth in these two regions presents a downside risk to the long-term lithium hydroxide demand outlook. BEV Production by Lithium Precursor Figure 15-5: BEV production by lithium precursor.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 331 Energy storage demand is expected to more than double from previous forecasts. Global battery-storage installations are now estimated to reach 735 GWh by 2030, up 55% from the previous forecast of 475 GWh for the same year. This uplift has been driven by more deployments in 2024 due to a fall in lithium-ion cell prices and increasing average storage durations. The technology outlook has not been changed, with lithium-ion batteries expected to remain the main player in the sector over the outlook period with a market share of 95% through to 2030 and 90% thereafter through to 2040. Lithium consumption from ESSs is now expected to reach around 470 kt LCE by 2030 and 1,360 kt LCE by 2040, more than double the previous forecast. The ESS demand outlook is more at risk for technological disruption from alternative battery technologies. Grid-scale ESS deployments are more concerned with project economics and space is typically less of an issue – meaning minimum energy density requirements are lax (relative to EVs, at least). In practice, technological selection will be based on what is commercially available, and the order of magnitude advantage presented in the economies of scale from lithium-ion manufacturing (from the EV sector) will limit the cost advantage presented from sodium-ion cells from a lower bill of materials. The greater risk in the forecast is from slower deployments of renewable power generation. The ESS outlook is intrinsically linked to the renewable power generation capacity as requirements for battery storage capacity are based on the additional grid flexibility requirements arising from the intermittency from (variable) renewable power. EV penetration and ESS outlook present the greatest uncertainty. EV Penetration and Powertrain Mix The outlook assumes global BEV penetration of 32% and 51% by 2030 and 2040, respectively, up from an estimated 13% in 2024. This is based on the aggregate of regional production factoring in near- and medium-term OEM production plans and longer-term EV penetration policy mandates (where applicable). The risk in this forecast arises from misses to near-term EV sales as this lowers the base for growth in the future (as highlighted by the revisions). This has been the case for much of 2024, especially in Europe where year-on-year EV growth is more than half of what it was in 2023 (+195,000 vs +470,000 units). Cathode Technology Uncertainty in the battery chemistry outlook comes primarily from three things: iterative improvements to existing processes (e.g. dry electrode manufacturing), breakthroughs in previously phased-out technology (e.g. LFP resurgence in China), and industrialisation of next-generation technologies (e.g. solid-state batteries). In Europe and North America, existing manufacturing expertise is very focused on nickel-based cathodes which will influence the long-term technology direction. For example, Umicore and BASF are both looking at mid-Ni cathodes (Mn-rich) as their mass- market offering. Another evolving dynamic is the trade tensions between China and the West. With the Chinese government’s latest proposal to control LFP technology exports, matching China’s LFP manufacturing cost will be extremely difficult for any emerging player in the West. Energy Storage There are two main sources of uncertainty, despite upgrades to the outlook. First, the upgrade is tied to higher expectations for renewable power – so lithium demand is sensitive to total ESS battery demand. Also, the LiB share of the technology is uncertain beyond 2030 due to competition from emerging technologies (e.g. Na-ion). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 332 Lithium demand in 2030 is not overly sensitive to any single region or variable, despite growth being reliant on vehicle electrification and energy storage systems. For example, a 10% lower penetration in China by 2030, equivalent to approximately 1.2 million fewer BEV units, would lower global lithium demand by 2.0% (equivalent to ~54 kt LCE). The greatest demand sensitivity is to the sources of battery demand, with the geographical ranking being the absolute size of the LiB market (i.e. China > Europe > North America). This is quite intuitively expected, owing to the dominance of lithium-ion batteries in total battery demand. At present, the combined versatility and cost of lithium-ion batteries leave little substitution potential for the technology. While firms are attempting to scale up alternatives such as sodium-ion, the downside risk to lithium remains relatively minimal, as highlighted by the low ESS sensitivity to LiB share. For example, even in the extreme case that sodium-ion batteries take a 50% share of the ESS market by 2030 (~367 GWh), gross lithium demand would only reduce by 8% (owing to the demand growth from other segments over the period). Furthermore, the LiB industry will continue to benefit from scaling demand from the EV sector (owing to low technical suitability of non-lithium competition), making it challenging for market disruption from Na-ion. Unlike nickel and cobalt, lithium demand is not too sensitive to an increased market share of LFP batteries, as there is little change in lithium loadings between the different lithium-ion battery chemistries. There are, however, implications for the preferred lithium precursor based on LFP share. LFP requires lithium carbonate, whereas Ni-rich cathodes require lithium hydroxide. Sensitivity for the precursor is correlated to absolute EV battery demand. 15.2.2 Supply Outlook 15.2.2.1 Primary Global lithium supply more than doubled to 985 kt LCE in 2023, up from 402 kt in 2020, largely owing to greater production in Australia, China, and Chile where numerous new mines and expansions have been commissioned over the past few years in response to strong BEV demand growth and the price spike in 2022. Most of this additional supply came from new or restarted spodumene mines, lepidolite mines in China, and evaporation-based continental brine operations. Primary lithium production is forecast to peak at 1.55 Mt LCE in 2027 from existing operations, a growth of 12% p.a. following the ramp-up of new mines in Africa and Australia, and brine operations in Chile and Argentina (refer to Figure 15-6). Again, most of these additional volumes are expected to come from new spodumene (including spodumene- petalite) mines and new or expanding continental brine operations. These new mines include Mt. Holland and Kathleen Valley in Australia, Goulamina in Mali, and various mines in Zimbabwe. Recent price-induced mine closures, mainly in Australia, China, and Zimbabwe, are likely to partially moderate hard rock-based supply growth. New brine operations include Centenario-Ratones, Tres Quebradas, and Sal de Oro (all in Argentina), while expansions are ramping up at the Salar de Atacama in Chile, and Olaroz and Cauchari- Olaroz, both of which are in Argentina. End-of-life (EoL) mine closures are set to reduce output over the long term, with supply from existing operations decreasing by 2% p.a. between 2027 and 2040 to 1.15 Mt LCE (-407 kt) (Figure 15-6). Africa, Australia, and Brazil are expected to account for most of the lost production during this period, with EoL mine closures anticipated at Mt. Marion (Australia); Arcadia, Kamativi, and Sabi Star (all Zimbabwe); and Xuxa and Mibra (both Brazil). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 333 Primary Lithium Supply by Region Figure 15-6: Primary lithium supply by region. Various advanced projects have the combined potential to boost supply well beyond 3 Mtpa by 2030 (Figure 15-7). Additionally, there are more than 50 ‘possible’ projects worldwide that have progressed beyond FS stage and could come on-stream in the medium term. Most of these are either spodumene or continental brine deposits, located in a variety of places including Australia, Argentina, North America, Africa, Brazil, Chile, and Europe (Figure 15-6). However, some projects are based on other lithium-bearing minerals, such as lepidolite, zinnwaldite, petalite or jadarite. There is also a growing number of advanced oilfield and geothermal brine projects located in North America and Europe. These projects might supply an extra 1.3 Mt LCE by 2030, which would expand total production to ~3.4 Mt (Figure 15-7). North America comprises the largest share of global output potential from these projects at close to 400 kt LCE at peak levels, followed by Australia (290 kt), Argentina (260 kt), Europe (230 kt), and Africa (190 kt). However, low-risk projects account for only 35% of ‘possible’ supply, producing an average of 377 ktpa but peaking at ~550 kt (Figure 15-7). These options are generally spodumene or continental brine deposits with low execution and processing (technical) risk, particularly restarts or expansions, and are located mainly in Australia, Argentina, and Canada. The main risks to these projects are typically funding and timing. High-risk projects comprise a third of ‘possible’ supply, with potential average volumes of 374 ktpa and a peak of ~490 ktpa (Figure 15-7). These options are generally either less advanced or have considerable execution or technical risk, such as unconventional or unproven processing routes, local issues (e.g. residential or political opposition or conflict), or are estimated to have very high costs (either capital requirements or operating costs). Some of these projects are unlikely to commence production before 2030. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 334 Primary Lithium Supply, including Possible Projects Figure 15-7: Primary lithium supply, including possible projects. European lithium production is limited to one small mine in Portugal, which produces 1 ktpa LCE for use in ceramics manufacturing (refer to Figure 15-8). The Keliber Lithium Project is Europe’s only project under construction and is set to increase output to 15 kt LCE by 2028. In addition, two German projects may supply around 30 kt LCE by 2030, namely Lionheart (Upper Rhine Valley) and Zinnwald. Neighbouring Austria might provide an extra 8 kt LCE before 2030 from the Wolfsberg Project, which has lower technical and execution risks than both of its German peers. Then there are three Iberian projects which could produce 67 kt LCE by 2030, with two in Spain (47 kt): San José and Las Navas, and one in Portugal (20 kt): Barroso. Barroso has the lowest technical risk of the three, although all have faced opposition from local communities. Other notable projects in Europe might provide nearly 60 kt in 2030, namely Cinovec in Czechia, Polokhivske in Ukraine, and the largest project on the continent, Jadar in Serbia. Overall, projects could lift European production to over 230 ktpa in the 2030s (Figure 15-8), but there are many threats to commercial- scale lithium mining throughout Europe.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 335 Primary Lithium Supply from Europe, including Projects Figure 15-8: Primary lithium supply from Europe, including projects. Earlier-stage European projects could supply an additional 134 kt LCE at steady-state production levels, which would increase Europe’s output to ~380 kt. 15.2.2.2 Secondary (Lithium-Ion Battery Recycling) Probable secondary lithium supply could grow to 750 kt LCE by 2040. The total number of BEVs reaching EoL is expected to grow to nearly 5 million by 2030 and over 30 million by 2040, from 0.4 million in 2024 with approximately 65% comprised of economically recyclable cathodes (refer to Figure 15-9). Total lithium supply potential from EoL BEVs is estimated to exceed 1,000 kt LCE by 2040, but probable supply is likely to be close to 70% of that. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 336 Lithium Supply Potential from EoL BEVs by Region Figure 15-9: Lithium supply potential from EoL BEVs by region. 15.2.3 Producer Economics Most operations are currently still profitable despite the significant decline in spot prices over recent years. The first two and a half quartiles of the cost curve are dominated by Lithium Triangle brine operations as well as Greenbushes, which is the largest and highest-grade hard rock mine in the world. Most lithium products are sold on contract rather than on the spot market, so the price received by producers can be greater than the spot price at the time of purchase. As indicated in Figure 15-10, the highest-cost mines are now uneconomical, while many others have had margins squeezed significantly over the past year. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 337 All-In Sustaining Costs: 2024E Figure 15-10: All-in sustaining costs: 2024E Concentrate producers suffered the most in 2024 as spodumene concentrate spot prices declined to less than USD1,000/t. Life for these mines, which typically occupy the fourth quartile of the cost curve, is unlikely to improve in 2025 with spot prices predicted to average just USD925/t. Costs are not expected to escalate dramatically over the long term, despite inflation remaining stubbornly high in many countries currently. In some cases, such as Argentina, local cost increases are anticipated to be offset by local currency depreciation versus the USD. Therefore, operating costs are estimated to remain below USD15,000/t in 2030 for almost all operations. The addition of many new projects is set to stretch out the cost curve, mostly entering in quartiles two to four, with costs comparable to existing operations. All-in-sustaining costs are estimated to stay under USD20,000/t in 2030 for even the highest-cost producers. Royalties alter the cost curve considerably, with the Chilean brine operations moving up from the first quartile to the fourth owing to the very high royalties in the country. Conversely, operations with very low royalties (e.g. Upper Rhine Valley in Germany) shift down the cost curve into the first quartile. Most of these projects are needed to meet net demand in 2030, meaning prices will need to rise back up to above USD20,000/t (refer to Figure 15-11). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 338 All-In Sustainable Costs: 2030F Figure 15-11: All-in sustainable costs: 2030F Current project incentive prices are generally between USD10,000/t and USD20,000/t LCE and are forecast to rise by roughly 30% by 2030. Prices are forecast to rise sufficiently each year to incentivise enough projects to meet future demand growth, assuming two years of construction and commissioning. However, near-term prices may need to be higher still to enable junior companies to secure funding for initial capital expenditure. 15.3 Market Entry Strategy and Product Specifications [§229.601(b)(96)(iii)(B)(16 (i))] Keliber’s lithium refinery in Kokkola will commence production during the first half of 2026 and is expected to reach the full run-rate corresponding to the annual capacity of 15,000 tonnes of Lithium hydroxide monohydrate during the first half of 2027. During the first 12 months of production, the refinery is expected to produce technical-grade lithium hydroxide (TG) and, thereafter, BG. Refer to Figure 15-12 and Figure 15-13 for the respective specifications. These are the products Keliber are currently planning on selling.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 339 LiOH.H2O Technical- and Industrial-Grade Specification Figure 15-12: LiOH.H2O technical- and industrial-grade specification (source: SSW). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 340 LiOH.H2O Battery-Grade Specification Figure 15-13: LiOH.H2O battery-grade specification (source SSW). Keliber’s market entry strategy will have two phases as the target customers for TG and unqualified BG differ from the target customers for qualified BG. Sale of TG and unqualified BG will continue until the anticipated qualification period required for long-term off-take contracts (LT BG contracts) is satisfied. 15.3.1 Phase 1 Keliber expect to sell the TG and unqualified BG volumes during 2026 and 2027 until the qualification processes required under the LT BG contracts are completed. These volumes are expected to be sold through one or two traders and potentially to a direct customer. The volume split between these channels has not been firmly set and Keliber will aim to retain a level of flexibility in allocating the volumes between the traders, based on the available end-customer prices. Keliber have signed an indicative term sheet with a trader and a Memorandum of Understanding with a S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 341 potential direct customer. TG has a wide geographical target market, and Keliber have received indications of interest from several Asian and European countries. 15.3.2 Phase 2 Keliber expect to enter into two to three long-term BG contracts during 2026 and that these contracts will cover approximately 80–100% of the Lithium hydroxide production. The rest of the product is expected to be sold on the spot market through a trader or possibly using an already identified third-party trading platform. Sale of commercial volumes under the long-term BG contracts will commence after the product qualification with the long-term BG customers is finalised. Keliber expect this to take place towards the end of 2027, depending on the time that it takes the lithium refinery to reach BG product quality, and the qualification processes can be completed. 15.4 Material Contracts [§229.601(b)(96)(iii)(B)(16)(ii)] Material contracts are required for Keliber´s mining operations, the concentrator, and the lithium refinery. The preparation and execution of the required contracts is progressing according to schedule. All contracts are negotiated on an arm´s length basis with external counterparties. 15.4.1 Mining Operations Since the mining operations will be outsourced, the most important contract is the Mining Operations Service Contract. The service will be executed by a main contractor and the agreement has been negotiated and approved. 15.4.2 Concentrator The material contracts for the concentrator are: • Heating plant: An external supplier will invest in a heating plant at the concentrator and Keliber will buy the required heat as a service. The contract is in place. • Chemicals and reagents: Several different kinds of chemicals and reagents are required for the production process at the concentrator. The most important are fatty rape acid, caustic soda, sulfuric acid, ferric chloride/sulphate, and different types of flocculants and polymers. The required contracts are scheduled for signature in May 2025. • Grinding media: Grinding rods and balls are used for crushing and grinding ore in the grinding mills. The contract for grinding media is scheduled for signature in June. 15.4.3 Lithium Refinery The material contracts for the lithium refinery are: • Natural gas: Natural gas will be utilised in the high-temperature conversion process. The supply contract is in place. • Chemicals and reagents: Several different kinds of chemicals and reagents are required for the production process at the lithium refinery. The most important are a) soda ash (contract has been signed), b) caustic soda, hydrochloric acid, and sulfuric acid (the contracts are scheduled for signature in May 2025, and c) quick lime and trisodium phosphate (the contracts are scheduled for signature in June 2025). S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 342 • Utilities: The contracts for most of the required process utilities have been signed i.e. water, district heating, nitrogen, and carbon dioxide. The contract for steam is under negotiation and is scheduled for signature in May 2025. • Wastewater disposal: The contract for the disposal and treatment of process wastewater is under negotiation with the City of Kokkola. • Analcime sand: Analcime sand will be formed as a secondary output from the production process at the refinery. A contract with the City of Kokkola, under which the analcime sand would be utilised for the expansion of the city´s port area, is in the signing process. 15.4.4 Concentrator and Lithium Refinery Jointly • Electricity: Electricity sourcing for Keliber´s operations will be managed by an external partner. The partner has been selected, and the production time risk policy and electricity strategy are under development in collaboration with the selected partner. • Spare parts and maintenance: The original equipment manufacturers have been selected, and different types of spare parts have been bought or are under negotiation. All critical contracts related to spare parts and maintenance are scheduled for signature by September 2025. • Logistics: Important logistics contracts are a) transport of own concentrate from the concentrator to the lithium refinery (the contract is scheduled for signature in June 2025), and b) transport of final product to end- customers (the contract will be initiated when the customer base has been decided). • Services: Services like area, road and facility maintenance, cleaning, waste and process sludge management, and security are under or scheduled for negotiation. 15.5 Commodity Market Assessment The market is likely to remain well supplied in the 2020s, despite the price-related risks to supply, with large inventories covering any short-term squeezes and any prolonged tightness triggering higher prices and a rapid supply response. Therefore, any potential shortfalls are likely to be relatively short-lived (refer to Figure 15-14). But higher prices are needed to incentivise projects to meet longer-term demand, following recent price-induced supply cuts and project delays. Only half of ‘probable’ and low risk ‘possible’ supply is incentivised at current lithium prices (~USD10,000/t Li2CO3). BEVs are set to drive medium-term demand growth, comprising two-thirds of growth from 2024 to 2030, even accounting for the recent slowdown in rampant sales. However, the downside risk exists that BEV sales (and demand growth) fail to match projections in the near term, partly offsetting supply cuts (refer to Figure 15-15). Energy storage systems are emerging as a key demand sector this decade, accounting for 20% of growth between 2024 and 2030, following a recent step change in demand levels. However, there are downside risks, such as the substitution threat from rival technologies (e.g. Na-ion batteries). The supply gap risk rises in the 2030s as requirements continue to grow strongly, again driven by BEVs (~50%) and energy storage (~40%), while mine depletion kicks in, reducing supply output. The market will become more reliant on higher risk, currently earlier-stage projects, to meet demand. Potential supply from Li-ion battery recycling is not enough to fill the long-term gap but, nevertheless, could become an important contributor to supply in the 2030s, potentially equivalent to supply from existing operations by 2040 (refer to Figure 15-16).

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 343 Rising prices are expected to accelerate project advancement before 2030, helping to fill the sizeable shortfalls thereafter. Near-term prices may push back the higher-risk projects, meaning the supply profile is better aligned with demand. Lithium Supply vs Demand Figure 15-14: Lithium supply vs demand. Lithium Supply-Demand Balance Figure 15-15: Lithium supply-demand balance. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 344 Lithium Supply vs Demand in 2040, including Projects Figure 15-16: Lithium supply vs demand in 2040, including projects. 15.6 Prices Rising prices are expected to incentivise investment in earlier-stage projects in the late 2020s, meaning they can potentially come on-stream by the mid-2030s when they are likely required to fill the supply gap. However, owing to uncertainty and just how undeveloped these options are currently, there is a high execution risk. Conversely, an increasing number of deposits are being explored and quantified worldwide even during the low-price cycle. There are more than enough lithium resources, they just need higher prices to motivate development. Many countries and companies are discovering substantial lithium deposits, like China, which recently announced a significant expansion of its own resources. Despite lower prices, exploration and DLE technology development continues unabated, particularly in places like the USA, China, India, and Saudi Arabia. DLE technology could enable a more rapid supply response and unlock new sources of lithium, such as unconventional brine deposits, reducing the risk of future shortfalls. However, DLE technologies are still relatively unproven at commercial scale, with some still S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 345 under development, so considerable technical risk remains, particularly as China is threatening to ban outside use of its advanced DLE technologies. Higher prices may also incentivise lithium recovery from recycled Li-ion batteries, although future recycling potential is highly dependent on BEV penetration, battery life, and second-life usage. Lower BEV penetration, longer battery lives, and greater second-life usage would limit potential secondary supply of lithium. The most recently available (January 2025) bank consensus metal price forecast for Lithium Hydroxide Battery Grade (USD/t), as compiled by UBS Bank, is provided in Table 15-1 below. Table 15-1: Price forecast for LiOH battery grade (USD/t) as reported in January 2025 (source: UBS Bank). Entity Date of Report 2025 (USD/t) 2026 (USD/t) 2027 (USD/t) 2028 (USD/t) 2029 (USD/t) LT/LT (2025 Real) (USD/t) BoA - Merrill Lynch 16-Dec-24 9,500 9,500 20,167 20,790 21,414 Bell Potter 7-Oct-24 12,500 15,000 19,500 21,000 21,000 BMO Capital 13-Sep-24 15,750 18,878 19,333 19,799 18,000 Canaccord 14-Oct-24 13,571 14,286 20,357 24,286 CIBC 17-Dec-24 13,788 14,956 16,761 Citi 8-Dec-24 10,300 12,000 Deutsche Bank 6-Oct-24 13,000 Jefferies 7-Oct-24 13,539 15,422 17,304 17,304 17,304 JP Morgan 11-Dec-24 10,625 18,000 Macquarie 8-Nov-24 10,975 12,000 15,625 19,625 National Bank of Canada 16-Dec-24 10,673 12,610 14,935 17,510 21,115 RBC 5-Dec-24 10,656 12,338 12,900 17,500 17,500 HSBC 21-Oct-24 13,300 17,500 Scotia 24-Sep-24 23,000 23,000 23,000 23,000 UBS 17-Jan-25 11,000 12,000 12,000 14,000 16,000 19,000 Average 12,812 14,576 17,444 19,090 16,000 19,593 Based on this analysis, the LT price assumption of USD20,000/t for lithium hydroxide in the open-pit optimisation exercise in support of the Mineral Reserve estimation is deemed appropriate and conservative. For the purpose of financial modelling over the LOM, the following price deck (Table 15-2) has been adopted, which is well supported by the market analysis. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 346 Table 15-2: LiOH prices applied in the financial model. Sales Product Start Date Currency 2026 2027 2027– 2032 2032– 2045 2045 2045– 2047 Ramp-up start flag Jan’26 USD/t EUR/t First LiOH sales, Technical grade May’26 USD/t 15,478 17,708 EUR/t 13,685 15,496 Switch to LiOH Sales Battery grade Contract Aug’27 USD/t 19,041 EUR/t 16,659 Switch to LiOH Sales Battery grade Contract & Spot Mix Aug’32 USD/t 19,384 EUR/t 16,525 Switch to Purchased Concentrate, Technical Grade LiOH Sale Jun’45 USD/t 18,027 EUR/t 15,368 Switch to LiOH sales, Battery Grade Contract Dec’45 USD/t 19,384 EUR/t 16,525

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 347 16 ENVIRONMENTAL STUDIES, PERMITTING AND PLANS, NEGOTIATIONS OR AGREEMENTS WITH LOCAL INDIVIDUALS OR GROUPS [§229.601(b)(96)(iii)(B)(17)] Keliber have completed all relevant EIA procedures to proceed with the Keliber Lithium Project. Keliber hold a valid EP for the Syväjärvi mining operations and a water permit for dewatering Syväjärvi and Heinäjärvi lakes. A valid permit states that the permit decision issued by the AVI was appealed and appeals were processed in the VAC, which ruled against appeals and kept AVI’s permit decision in force on 16 June 2021. No appeals were made to the SAC against the VAC’s decision. The Syväjärvi EP became legally valid in July 2021. The Rapasaari mine EP application was submitted to AVI on 30 June 2021. The Päiväneva concentrator EP was submitted to AVI on 30 June 2021. Concentrator operations require a water permit for raw water intake from the Köyhäjoki River, and that permit application was also submitted to AVI on 30 June 2021. Keliber received EPs for the Rapasaari mine and Päiväneva concentrator in December 2022 (Environmental permit 208/2022 number: LSSAVI/10481/2021, LSSAVI/10484/2021). The permit decision issued by the AVI was appealed, and appeals were processed in the VAC – the decision was received 23.2.2024 (206/2024). Some permit regulations e.g. Rapasaari WRD areas, were returned to AVI reprocessing. The Keliber lithium refinery has a legally valid EP (122/2022 number: LSSAVI/17444/2020). Keliber submitted an amendment application to AVI on 26 June 2024 in order to change the cooling and stormwater discharge point in Kokkola Harbour. 16.1 Relevant Environmental Issues and Results of Studies Conducted 16.1.1 Groundwater Studies In support of baseline assessment for the EIA, Syväjärvi, Rapasaari, Outovesi, and Päiväneva groundwater samples were collected from observation wells during the years 2018–2020. The groundwater quality sample results were compared to the Decree of the Ministry of Social Affairs and Health (1352/2015, amendment 683/2017) chemical quality standards and objectives for drinking (potable) water. Results indicated that groundwater quality in most samples meets drinking water quality standards with the exception of the elements iron and manganese. Elevated iron and manganese are the result of higher chemical oxygen demand and low oxygen levels. This is a result of the impact of humus-contained waters from the surrounding peat lands. Natural concentrations of ammonium also exceed recommendations for household water quality. 16.1.2 Biodiversity Starting from 2014, several studies concerning vegetation, habitats, flora, and fauna have been carried out. The list of studies which have been conducted over the years at the mine sites and surrounding areas is provided in Table 16-1. Table 16-1: Field studies carried out at Syväjärvi, Rapasaari, and Outovesi mine sites and Vionneva Natura 2000 area. Study Target Executor Study Period SYVÄJÄRVI Vegetation 2014–2015, 2020 Ahma Ympäristö 2015, Envineer Oy 2020 Habitats 2014–2015, 2020 Ahma Ympäristö 2015, Envineer Oy 2020 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 348 Study Target Executor Study Period Nesting birds 2014, 2020 Ramboll Finland 2014e, Envineer Oy 2020 Moor frog 2014–2024 Ramboll Finland 2014cd, Tutkimusosuuskunta Tapaus 2015– 2019, Saarikivi, J. 2020–2024 Bats 2014, 2020 Ramboll Finland 2014a, Envineer Oy 2020 Siberian flying squirrel 2014, 2020, 2021 Ramboll Finland 2014, Envineer Oy 2020, Saarikivi, J. 2021 Eurasian otter 2023, 2024 Keliber 2023–2024 Predaceous diving beetles 2018–2020 Tutkimusosuuskunta Tapaus, 2018, 2019, Saarikivi, J. 2020 Dragonflies 2018–2020 Tutkimusosuuskunta Tapaus, 2018, 2019, Saarikivi, J. 2020 Fish 2014, 2023–2024 Nab Labs 2014, Eurofins Ahma 2023-24 Benthic invertebrate fauna 2014, 2020, 2024 Ahma 2015, Vahanen Environment Oy 2020, Eurofins Ahma 2024 Diatoms 2014, 2020, 2024 Eloranta 2014, Vahanen Environment Oy 2020, Eurofins Ahma 2024 RAPASAARI JA OUTOVESI Vegetation 2014–2015, 2020 Ahma Ympäristö 2015, Envineer Oy 2020 Habitats 2014–2015, 2020 Ahma Ympäristö 2015, Envineer Oy 2020 Nesting birds 2014, 2020 Ramboll Finland 2014e, Envineer Oy 2020 Moor frog 2014–2024 Ramboll Finland 2014cd, Tutkimusosuuskunta Tapaus 2015– 2019, Saarikivi, J. 2020–2024 Bats 2014, 2020 Ramboll Finland 2014a, Envineer Oy 2020 Siberian flying squirrel 2014, 2020, 2021 Ramboll Finland 2014, Envineer Oy, Saarikivi, J. 2021 Predaceous diving beetles 2018–2020 Tutkimusosuuskunta Tapaus 2018, 2019, Saarikivi, J. 2020 Dragonflies 2018–2020 Tutkimusosuuskunta Tapaus, 2018, 2019, Saarikivi, J. 2020 Fish 2014, 2020, 2024 (Rapasaari) Nab Labs 2014, AFRY Finland Oy 2020b, Eurofins Ahma 2024 Benthic invertebrate fauna 2014, 2020, 2024 (Rapasaari) Ahma 2015, Vahanen Environment Oy 2020, Eurofins Ahma 2024 Diatoms 2014, 2020, 2024 (Rapasaari) Eloranta 2014, Vahanen Environment Oy 2020, Eurofins Ahma 2024 VIONNEVA NATURA AREA Nesting birds 2014–2018, 2020, 2024 Tikkanen and Tuohimaa 2014, Ramboll 2016, 2018, Envineer Oy 2020, Ramboll 2024 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 349 In the following text, the directive habitat species, as identified during the above studies, and Keliber’s actions to protect the habitats are presented. • Moor frog − Keliber have built four moor frog (Rana arvalis) ponds outside the Syväjärvi mine site. The purpose of the ponds is to secure favourable conservation status and to provide moor frogs a place to breed and rest, thus improving the habitat of the moor frogs in the area. • Siberian flying squirrel − The Siberian flying squirrel (Pteromys volans) is a species classified as vulnerable and is strictly protected by the Habitats Directive. Elsewhere in the EU, the Siberian flying squirrel only occurs in Estonia. − Keliber consider in its operations designs, as far as possible, the facilitation of the preservation of the old-growth forest area where the flying squirrel was detected. In response to interactions with ecologists, Keliber have, in their 2021 design engineering work, relocated the south dam wall of the TSF further away from the ancient forest area. − In 2025, Keliber bought out 17.5 hectares of forest, mostly consisting of less than 70 years old forest stand, which includes and is connected to the known flying squirrel habitat. • Bats − All of the bat species present in Finland are listed in the Habitats Directive IV(a) of the EU’s Convention on Biological Diversity. − Keliber endeavour to facilitate, as far as possible, preservation of and creation of new bat resting places. • Eurasian otter − In the field survey carried out during the EIA in 2020, traces of otters (Lutra lutra) were observed on the snow on the shores of streams Näätinkioja (also named as Kärmeoja) and Köyhäjoki, which are located south of the Päiväneva concentrator area. − The 2020 field survey of otters was the first of its kind conducted in the area. − Keliber have carried out more field surveys to obtain more precise information on where otters live and breed. With more precise information, Keliber may help to protect and preserve the otter population in the area. − In addition to field surveys conducted by Keliber, trail cameras have been used to monitor otters since 2023. − As of (February) 2025, several otters, including at least one female with pups, continue to inhabit both Rapasaari and Syväjärvi areas. • Golden eagle − The golden eagle (Aquila chrysaetos) is not listed in the Habitat Directive Annex IV(a) but is classified as vulnerable in Finland. − To protect and improve the golden eagle territory at Vionneva, Keliber have taken the following actions: o Artificial nests have been built further away from mine sites; o Artificial feeding during wintertime has been started to improve the success of nesting; and o Satellite tracking of known eagles. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 350 16.1.3 Air Quality AFRY Finland Oy have modelled potential dust impacts of Syväjärvi and Rapasaari mine operations and Päiväneva concentrator operations with the results reported in: Keliber Technology Oy, Rapasaaren ja Syväjärven kaivosten pölypäästöjen leviämismallinnus, AFRY Finland Oy, 4.11.2021 (Finnish). AFRY calculated the dust dispersion using the Breeze Aermod model tool developed by the USA Environmental Protection Agency. According to a report dated 4.11.2021, modelled results show that respirable particulate matter (PM10) limits are not exceeded at the nearest holiday homes in any modelled situations due to the mining activities at Syväjärvi and Rapasaari and the concentrator operations in Päiväneva. 16.1.4 Noise AFRY Finland Oy have carried out a noise model for Keliber with the results reported in Finnish in the report: Keliber Technology Oy, AFRY Finland Oy 2.11.2021. The modelling was done by using SoundPlan v8.2 noise calculation software. The report is part of the EP application for Rapasaari mine and the Päiväneva concentrator. Noise model results for the Rapasaari mine and Päiväneva concentrator plant have been compared to noise limit values stated in the Syväjärvi EP decision. Based on the noise modelling results reported by AFRY, the results for the average noise level i.e. the A-weighted equivalent continuous sound level (LAeq), are below the average noise level of the limit values for Syväjärvi. According to the modelling, the Vionneva Natura 2000 area could be affected by noise levels greater than 50 dB, especially in the early years of the Rapasaari mine operations when the waste rock area is still shallow. As the Rapasaari mine progresses, noise impacts on the Vionneva Natura area are reduced. 16.2 Water Management Keliber have developed a detailed document, The Site Water Management Plan, which combines the Project site water management data into one document and includes subsequent modelling and assessment tasks: • Rapasaari mine site hydrogeological hydrogeological modelling; • Rapasaari-Päiväneva area source term models (water qualities and quantities for extractive waste facilities, pit, and underground mine), operational, and post-closure phases; • Rapasaari-Päiväneva complex site-wide water balance modelling; • Syväjärvi open-pit hydrogeological modelling; • Site hydrogeological assessment of the Länttä, Outovesi, and Emmes mine sites; • Water quality summaries for Syväjärvi, Länttä, Outovesi, and Emmes mine sites (based on existing data); and • Rapasaari-Päiväneva complex, conceptualisation of site water management related components. Owing to the change in mine plans and infrastructure designs, updated hydrogeological modelling at Rapasaari and Syväjärvi is required. The amendments may also trigger adjustments to water treatment requirements and water management infrastructure, including ponds and treatment facilities. Currently, water treatment facilities may be inadequate and water management feature designs may be under (or over) estimated, as noted by SRK in the 2023 due diligence. This may directly impact budgets and Project timelines.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 351 16.2.1 Surface Waters and Groundwater All planned mine sites are in the River Perhonjoki catchment area. Syväjärvi mine is in the catchment of the River Ullavanjoki while Rapasaari mine, concentrator, and Hoikkaneva disposal site are in the catchment of the River Köyhäjoki. River Ullavanjoki starts from Lake Ullavanjärvi, which is upstream of Syväjärvi mine and, therefore, Syväjärvi mine has no impact on Lake Ullavanjärvi. Syväjärvi has a valid environmental and water permit (LSSAVI/3331/2018, 20 February 2019 and administrative court decision 16 June 2021, 21/0097/3). The permit consists of permit conditions including water management principles, permit conditions for dewatering and sediment removal from the lakes Syväjärvi and Heinäjärvi, and acceptable emission levels. The Syväjärvi mine site water management system has been designed to meet the requirements of permit conditions. All water management structures and water quality monitoring are determined in the EP. When executed accordingly, the risks to the environment, water bodies, or flora or fauna are mitigated. After the Syväjärvi mining operation has ended, the drainage pumping of open-pit water is stopped, allowing surface and groundwater to enter the pit. By slowing water flow through the Syväjärvi open pit, some modifications to the surface watershed area can be achieved e.g. limit flow within ditches. In this way, water quality can be controlled. In the early phase of post-closure, when the open pit is filling with water, any excess water is discharged in a controlled manner through a wetland to remove any solids. The open pit is estimated to take around 5–10 years to fill. Estimation of groundwater inflow to the open pit, when it is at its largest, is 710 m3/d. The dewatering amount including direct precipitation into the pit is approximately 840 m3/d. In this dewatering amount, evaporation is assumed to be 50% of the total precipitation. The radius of the drawdown cone is a few hundred metres from the pit. As explained in the AFRY report of Syväjärvi Hydromodel, pit dewater flow is directed to sedimentation ponds and on to a wetland before flowing to Ruohojärvenoja Ditch. Separate EP applications for the Rapasaari mine and Päiväneva concentrator were submitted to the AVI on 30 June 2021 and approved on 28 December 2022. The permit decision issued by AVI was appealed and appeals were processed in the VAC; the decision was received 23.2.2024 (206/2024). Some permit regulations were returned to AVI for reprocessing. Water management and water quality before, during, and after the operational phase of the Rapasaari-Päiväneva complex is described in detail in the Water Management Plan by AFRY Finland Oy. Raw water needed for concentrator processes is pumped from the River Köyhäjoki at Jokineva. The discharge point for wastewater is downstream of the water intake. The waters will be treated centrally in the Päiväneva water treatment plant where there are unit processes for chemical treatment of raw water as well as biological nitrogen removal from mine water. Treated water is recycled back to the process or discharged to the environment via pipeline to River Köyhäjoki. The discharge rate to Köyhäjoki is limited by the flow of the river and cannot exceed 20% of daily (24-hour average) average river flow. According to Rapasaari Numerical Groundwater Flow Modelling by AFRY, groundwater inflow into the Rapasaari open pit when the pit is at its largest (southern open-pit extension included) will be 2,680 m3/d. The dewatering amount including precipitation into the pit at this stage will be approximately 3,100 m3/d. Mine water is pumped to the mine water pond then to the nitrogen removal process and, from there, to the recycle water pond from which it can be released as effluent to Köyhäjoki Stream. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 352 According to the AFRY waste management plan for Rapasaari mine 05-11-2021, some seepage water from waste rock areas and the TSF will occur. From waste rock areas, seepage flows to the open pit and from the TSF, seepage enters groundwater where it dilutes. The sealed bottom structure of the pre-float and magnetic plants tailings facility and the pyrite- and arsenic-containing WRD area will minimise seepage water effectively. 16.2.2 Effects on Surface Waters During the construction phase, digging and relocation of soil could impact water quality of the Näätinkioja Stream by temporarily increasing turbidity and concentration of suspended solids in the stream. The impact is minimised by preparing first the sedimentation ponds to collect runoff water from the area. All waters at the construction area are discharged through a wetland treatment area to Näätinkioja Stream. During the operation phase, the effluent from the Rapasaari-Päiväneva complex is treated, collected to the recycle water pond, and then discharged into the Köyhäjoki at Jokineva via pipeline. The decision on the location of discharge was made because the Köyhäjoki is a much larger river than Näätinkioja Stream and, during the EIA process, a trout population was found to live and spawn in the Näätinkioja. The volume of effluent discharged into the Köyhäjoki will peak in the years 8 to 10 of operation at around 170–200 m3/h. The concentrations of contaminants in the water bodies during that period were modelled and compared to Finland’s national reference values. In the absence of national reference values, international values such as those from the European Chemicals Agency, the US Environmental Protection Agency, and the Canadian Council of Ministers of the Environment were used. The studied contaminants consisted of more than 40 elements and minerals. The modelling was conducted for three spots: 1) at the discharge in the Jokineva, 2) 10 km downstream from Jokineva, and 3) just before the Köyhäjoki flows to the lake chain 20 km from Jokineva. Cobalt, zinc, and vanadium exceeded the national reference values but, for cobalt and zinc, even the baseline concentrations are above the reference value. It is notable that the national reference values are for soluble concentration while the modelling was conducted for total concentrations and is therefore conservative. Nutrient loading (P and N) from the Rapasaari-Päiväneva complex to the Köyhäjoki was compared to total annual nutrient loading based on VEMALA modelling, which is an operational, national-scale nutrient loading model for Finnish watersheds operated and developed by the Finnish Environment Institute. Based on calculations by AFRY Finland Oy, the Rapasaari-Päiväneva complex is expected to release less than 10% nitrogen and less than 5% phosphorus to the Köyhäjoki during the operational years 8–10. According to VEMALA, the total current annual nitrogen loading to Köyhäjoki Agriculture is the main source of both nitrogen (40% of the annual N load) and phosphorus (54% of the annual P Load) in the Köyhäjoki catchment area. After mine closure, discharge to the Köyhäjoki in Jokineva will cease and the Rapasaari pit will be allowed to fill naturally with water. During and after the filling, leaching of nutrients and contaminants into the Näätinkioja Stream could occur. The modelling for water quality was conducted for three post-closure phases. Phosphorus concentration increased from 20–25 μg/L and nitrogen from 8–68 μg/L, depending on the phase. Such a low increase in nutrient loading to Näätinkioja Stream does not adversely affect water quality or the flora and fauna in the stream. In each post-closure phase, cobalt slightly exceeds the reference value, which is 0.5 μg/L according to the publication of the Ministry of Environment, but the baseline value is 0.45 μg/L. The increase in concentration for other elements is negligible. An ecological status assessment and assessment of impacts from mining operations on the ecological status of surface waters from the Rapasaari-Päiväneva complex was conducted, the full report of which is available in Finnish and is included in the EP application for Rapasaari mine and the Päiväneva concentrator. According to the assessment, water discharge from the Rapasaari-Päiväneva complex does not have a negative impact on the ecological status of surface S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 353 water bodies in the discharge area or further downstream. The implementation of the Päiväneva production area will not hinder the achievement of water management, marine conservation objectives or the implementation of water protection action plans. Furthermore, the recreational use of the waters downstream of the Päiväneva production area, for recreational fishing and crayfishing, is not expected to be adversely affected. 16.3 Potentially Sulphate Soils The GTK conducted a sulphate soil survey in 2014 at the Rapasaari, Syväjärvi, Outovesi, and Länttä mine sites. The GTK study assessed the potential risk of soil acidification due to land use or drainage. Acid sulphate soils are known to pose a risk of acidification to soil and water bodies if non-oxidised sulphide-rich soil layers below the water table are exposed to oxidation. Typically, these layers or soil masses are oxidised during drainage or excavation of the soil. AFRY Finland Oy conducted a sulphate soil survey at the Päiväneva concentrator area in 2020 [21]. In total, 21 soil samples from four locations were taken and analysed for total sulphur content and acid-producing potential with a net acid generation (NAG) test. According to the AFRY report, the test results show the soil is not naturally acid- producing. More studies in 2023 of sulphate soils have been conducted, showing some results of acid-producing potential. 16.4 Acid-Producing Waste Rock At Syväjärvi, pyrite-containing mica-schist makes up 2% of the waste rock and is potentially acid-producing. At Rapasaari, pyrite-containing waste rock makes up 1% of the waste rock and is potentially acid-producing. This is kept as low as possible through the mine schedule and stockpile management. As shown in the EIA of 2020, the acid-producing and neutralising potential for waste rock has been determined by acid-base accounting (ABA) tests. Some of the Syväjärvi mica schist and intermediate metatuffic/meta vulcanite were classified as potentially acid-producing and pyrite-containing mica schist as acid-producing. In their 2023 due diligence, SRK noted the testing indicated elevated arsenic and sulphur in Syväjärvi waste rock lithologies, exceeding limits included in the EP. At Rapasaari, only mica schist was classified as acid-producing. Outovesi samples were all classified as acid-producing. Keliber will install the following structures at waste rock areas of potential acid production. To prevent acid leachate from entering soil or groundwater from acid-producing waste rock areas, a mineral sealing layer will be built on top of the subsoil moraine. A bentonite mat will be laid on top of the moraine layer as well as an HDPE membrane, protected by a geotextile (sizing according to the material supplier's instructions) or a layer of sand. Pre-filling will be done with waste rock with the pre-fill layer acting as both protection of the sealing structure and as an access and working platform for machinery. This applies to Syväjärvi and Rapasaari mine sites where acid-producing waste rock is likely to be encountered. The handling of acid-producing waste rock and waters generated in these areas is described in detail in the waste management plan by AFRY Finland Oy. 16.5 Waste Disposal Government Decree 190/2013 for extractive waste applies to the preparation and implementation of an extractive waste management plan; the establishment, management, decommissioning, and after-care of an extractive waste disposal site; the recovery of extractive waste in an opencast mine; and the monitoring, supervision, and control of the management of extractive waste. An extractive waste management plan is mandatory in order to start mining operations, and the plan is also a mandatory part of the EP application. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 354 Keliber have extractive waste management plans done for Syväjärvi mine, Rapasaari mine, and the Päiväneva concentrator areas where mining waste storage facilities are located. Reports, in the Finnish language, are: • AFRY Finland Oy 2025, Kaivannaisjätteen jätehuoltosuunnitelma Päiväneva; • AFRY Finland Oy 2025, Syväjärven kaivoksen kaivannaisjätteen jätehuoltosuunnitelma; and • AFRY Finland Oy 2025, Rapasaaren kaivoksen kaivannaisjätteen jätehuoltosuunnitelma. 16.6 Environmental Site Monitoring In Finland, the site monitoring will be regulated by the EP decision and by the monitoring authority (ELY). An applicant suggests a monitoring plan, and the monitoring authority accepts it. The plan addresses site monitoring during construction works, operations, the closure phase, and after closure. Discharge and environmental impact monitoring has been carried out since 2022 in accordance with the monitoring programme approved by the ELY. This includes water discharge monitoring during construction work to Ruohojärvenoja ditch and Näätinkioja ditch and impacts monitoring. At Syväjärvi, monitoring is done according to the monitoring plan and according to regulations issued in the EP and the Administrative Court decision. For Rapasaari and Päiväneva, a monitoring plan during construction work has been approved by ELY 20.12.2023, and a monitoring plan for the operational phase will be submitted to ELY in early 2025. Similarly, a monitoring plan was submitted for Hoikkaneva in December 2024. Keliber will join with other operators for the monitoring programme of the Perhonjoki River area, which includes water quality monitoring, diatom, sediment, and fish monitoring. Keliber have joined the air quality bioindicator monitoring programme that is in place at the Kokkola and Pietarsaari areas. In Kokkola, Keliber have joined the groundwater and air quality monitoring programmes. Keliber will also join the sea water quality monitoring programme as well as the KIP’s joint noise monitoring campaign. Biodiversity monitoring is presented in the Biodiversity Management Plan. 16.7 Social and Community Aspects Residential surveys have been conducted during the years 2014–2018, and the latest survey took place during the EIA process for Syväjärvi, Rapasaari, and Outovesi in 2020. Respondents to the 2020 survey were mostly recreational users (33%), permanent residents (23%), and others (23%). A majority of the 98 respondents live within a two-kilometre radius of a Project site. The majority of respondents felt that the impacts of the Project were positive (43%). Employment for the Project was perceived to be the most important effect (49%) and, secondly, environmental management and sustainable development (42%). Regional development was also seen as a positive impact. On the negative side, respondents saw potential negative impacts to surface waters and possible contamination, damage to natural values, impacts on the ecosystem, dust and noise impacts, and possible impacts after closure. Maintaining communication with stakeholders as per the Keliber Stakeholder Engagement Plan, meeting its regulatory commitments, and ensuring that it is transparent in reporting will allow mitigation of social and community-related risks.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 355 16.7.1 Recreational Use According to the results of the 2020 residents' survey carried out in connection with the EIA process, the Syväjärvi, Rapasaari, and Outovesi mining areas are considered important for recreational purposes, in particular for hunting, berry picking, and mushroom picking. Although, according to public sources, there are no official recreational areas or routes in the mining areas. In stakeholder meetings with local people, the recreational use of the areas and the limitations that come along with mining activities have not been raised as a major issue. Although mining areas limit recreational activities and may cause nuisance in terms of noise and artificial lighting, the areas required by mining are moderate in size. Near the Rapasaari-Päiväneva complex, peat production in an area of 350 ha has been carried out for years, resulting in manmade landscape, dust, and noise that already affect recreational use. 16.7.2 Land Use, Economic Activity, and Population The industrial structure of Central Ostrobothnia is characterised by the metal, wood, process, and chemical industries. The construction, services, and manufacturing sectors also have a large employment impact. Agricultural production is concentrated in the dairy, beef, and potato sectors. Peat production plays an important role in the energy supply of Central Ostrobothnia. In the hierarchy of the service network in Central Ostrobothnia, Kokkola is the commercial centre of the region and Kannus and Kaustinen are subcentres. It is estimated that mining, concentrator, and Keliber lithium refinery operations will directly employ approximately 170 personnel and 50 contractors. Keliber will use subcontractors for excavation and transportation. Employment impact was seen as one of the most important positive impacts of the Project. Mining activities and the concentrator plant operations are in accordance with the current regional plan, and therefore, the Project is consistent with and supports the planned land use. Alholmens Kraft (AK) is a significant user of peat and has its own peat production areas at Päiväneva. The Project concentrator plant location is partly on AK’s land. Keliber have purchased areas needed for its operations from AK in a mutual understanding. Forestry at the mining areas will cease, and losses have been or will be compensated to owners. Compensation has been agreed upon in the process of establishing mining areas. The procedure is explained in the Land Acquisition and Livelihood Restoration framework. Päiväneva is currently not a pristine habitat but an industrial peat production area. Production is coming to an end, whereupon the area can be used as a concentrator plant site. Other areas surrounding the Project, mostly peat production and fur farming, can continue in the vicinity despite the mining activities, with no adverse impacts (e.g. dust and noise impacts) from mining expected. No other economic activities are known to exist in the Project area, which could be significantly affected. The Project is seen to have a positive impact on the region. 16.8 Closure Aspects In Finland, a closure plan for a mine is part of the EP application, and the plan must be updated as the operation progresses. The final closure plan will be presented to the authorities at the end of the operation. The overall objective of the closure works is to bring the site into as stable a state as possible, both physically and chemically, in line with the provisions of the legislation and addressing specific requirements of the local environment. Keliber have a conceptual closure plan for the mining areas. Closure reports, in the Finnish language, are: S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 356 • AFRY Finland Oy 2025, Päivänevan rikastamoalueen sulkemissuunnitelma; • AFRY Finland Oy 2025, Syväjärven kaivoksen sulkemissuunnitelma; and • AFRY Finland Oy 2025, Rapasaaren kaivoksen sulkemissuunnitelma. On a general level, closure activities comprise the covering of waste rock areas and TSF, making open pits safer, and dismantling building and other structures unless those can be reused for some other land use activity. The conceptual closure plan for Rapasaari-Päiväneva was updated by AFRY Finland Oy in 2024/2025. The closure plan will be updated during operations, and a final closure plan will be submitted before operations cease and closure commences. The closure plan addresses the impact of closure on surface waters, groundwater, soil, flora and fauna, conservation areas, air quality, landscape, traffic, and people and society. Keliber plan to present security deposits of EUR4.6 million for Rapasaari mine and EUR3.4 million for the Päiväneva concentrator. The security deposits have not been lodged yet, but Keliber have made provision for them in the financial model (Sheet “Assumptions”, lines 184-191). 16.9 Environmental, Social, and Governance Summary All EIA processes including the required statutory stakeholder consultations have been conducted and finalised in terms of the relevant Environmental Law: Environmental Protection Act (527/2014) for the Rapasaari-Päiväneva complex, Syväjärvi, Rapasaari, Länttä, and Outovesi mine sites, the Keliber lithium refinery, and the Hoikkaneva final disposal site. Keliber have met all regulatory permit requirements, except for magnetic waste stream management in Päiväneva and WRDs in Rapasaari. The QPs are satisfied that Keliber and SSW actively manage the processes and have to date timeously initiated additional processes as required to satisfy all compliance and permitting requirements. The QPs are satisfied that the work executed to date, and current plans to address any issues related to environmental compliance, permitting, and local individuals or groups, are adequate. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 357 17 CAPITAL AND OPERATING COSTS 17.1 Capital Costs The total capital expenditure (Capex) for the Project amounts to EUR651m, as shown in Table 17-1. Table 17-1: Capex summary. Item Total Capex (MEUR) % Development Phase - 0 Direct Costs 465 71 Syväjärvi Mine 18 3 Rapasaari Mine 68 10 Concentrator Plant (Päiväneva site) 228 35 Keliber lithium Refinery site 152 23 Indirect Costs 159 24 Engineering & Construction Services 41 6 Site Facilities During Construction 10 2 Construction Equipment 13 2 Spare Parts 26 4 Commissioning 1 0 Owners' Cost 18 3 Contingency 45 7 Grants + Adjustments 4 1 Plant-Related Costs 27 4 Concentrator Plant (Päiväneva site) 5 1 Keliber Refinery site 22 3 Total Capex 651 100 Initial Capex is scheduled to be spent from October 2024 to December 2026 with no historical Capex being included in the financial model, and Sustaining Capex is scheduled throughout the Project life period from October 2024 to June 2045. Figure 17-1 shows the monthly capital expenditures over the Project life period. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 358 Monthly Capex Figure 17-1: Monthly Capex. 17.2 Operating Cost The total operating expenditure (Opex) for the Project amounts to EUR2,293mand includes variable costs, fixed costs, royalties, and other fees, as shown in Table 17-2. Table 17-2: Total Opex. Item Total (MEUR) % Mining Opex 515 22 Crushing & Sorting and Concentrator Opex 217 9 Conversion & Keliber lithium Refinery Opex 885 39 Other Variable Opex 101 4 Freight and Transportation 32 1 Fixed Costs 520 23 Royalties and Fees 23 1 Total Opex 2,293 100 17.2.1 Mining Mining Opex makes up 22% of total operating costs and amounts to EUR515m over the LOM period, as shown in Table 17-3. A contractor is used to remove topsoil, drill, blast, load, and haul material. No underground mining development has been included as part of the mining method for Syväjärvi and Rapasaari and/or this TRS. 0 10 000 20 000 30 000 40 000 50 000 60 000 70 000 O ct -2 4 Ju l- 2 5 A p r- 2 6 Ja n -2 7 O ct -2 7 Ju l- 2 8 A p r- 2 9 Ja n -3 0 O ct -3 0 Ju l- 3 1 A p r- 3 2 Ja n -3 3 O ct -3 3 Ju l- 3 4 A p r- 3 5 Ja n -3 6 O ct -3 6 Ju l- 3 7 A p r- 3 8 Ja n -3 9 O ct -3 9 Ju l- 4 0 A p r- 4 1 Ja n -4 2 O ct -4 2 Ju l- 4 3 A p r- 4 4 Ja n -4 5 O ct -4 5 Ju l- 4 6 A p r- 4 7 EU R '0 0 0 Capital Expenditure over Project Life

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 359 Table 17-3: Mining Opex. Item Total (MEUR) % Ore Mining 68 13 Waste Mining 437 85 Underground Development - 0 Other Mining Costs 10 2 Total Mining Opex 515 100 17.2.2 Crushing, Sorting, and Concentrator Crushing, sorting, and concentrator Opex makes up 9% of total operating costs and amounts to EUR217m over the LOM period, as shown in Table 17-4. The Concentrator Water Treatment is included in this Opex estimate. Table 17-4: Crushing, sorting, and concentrator Opex. Item Total (MEUR) % Crushing, Sorting, and Stockpiling 6 3 Concentrator - Energy 39 18 Concentrator - Reagents 86 40 Concentrator - Consumables 44 20 Concentrator - Maintenance 24 11 Concentrator Water Treatment - Energy 6 3 Concentrator Water Treatment - Reagents 7 3 Concentrator Water Treatment - Consumables 2 1 Concentrator Water Treatment - Maintenance 3 1 Total Crushing, Sorting, and Concentrator Opex 217 100 17.2.3 Conversion and Keliber lithium Refinery Conversion and Keliber lithium refinery Opex makes up 39% of the total operating costs and amounts to EUR885m over the LOM period, as shown in Table 175. The Keliber lithium refinery water treatment is included in this Opex estimate. Table 175: Conversion and Keliber lithium Refinery Opex.- Item Total (MEUR) % Concentrate - Loading & Transport 30 3 Concentrate - Purchase - 0 Conversion - Energy/Fuel 97 11 Conversion - Other Consumables / Utilities 12 1 Keliber lithium Refinery - Energy 91 10 Keliber lithium Refinery - Steam 165 19 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 360 Item Total (MEUR) % Keliber lithium Refinery - Reagents 284 32 Keliber lithium Refinery - Process Water 4 0 Keliber lithium Refinery - Consumables 6 1 Keliber lithium Refinery - Utilities 14 2 Keliber lithium Refinery - Maintenance 20 2 Keliber lithium Refinery Water Treatment - Reagents 81 9 Keliber lithium Refinery Water Treatment - Consumables 60 7 Keliber lithium Refinery Water Treatment - Energy 12 1 Keliber lithium Refinery Water Treatment - Other Costs 10 1 Total Conversion & Keliber lithium Refinery Opex 885 100 17.2.4 Other Variable Opex Other variable Opex makes up 4% of total operating costs and amounts to EUR101m over the LOM period, as shown in Table 17-6. Table 17-6: Other variable Opex. Item Total (MEUR) % Service & Handling 2 2 Post- and Transport Fee 0.3 0 Environmental Emission Map Upkeeping Fee 0.3 0 Analcime Sand Discharge Fee 99 98 Total Other Variable Opex 101 100 17.2.5 Freight and Transportation Opex Freight and transportation Opex make up 1% of total operating costs and amounts to EUR316m over the LOM period, as shown in Table 17-7. It is also noted that additional freight cost is deducted from LiOH sales revenue amounting to EUR21,8m. This freight cost assumes 100% of product volumes are transported to Antwerp, Belgium, at a cost of EUR75/t, and 0% of product volumes are transported to Shanghai, China, at a cost of EUR150/t. The total effective Freight and Transportation cost is therefore calculated as EUR53m Table 17-7: Freight and transportation Opex. Item Total (MEUR) % Side Rock Transport - 0 Final Product Transport 32 59 Subtotal 32 59 Freight Cost deducted from Revenue 21,8 41 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 361 Item Total (MEUR) % Total Freight and Transportation Opex 53 100 17.2.6 Fixed Costs Fixed cost Opex makes up 23% of total operating costs and amounts to EUR520m over the LOM period, as shown in Table 17-8. Table 17-8: Fixed costs Opex. Item Total (MEUR) % Labour Cost 261 50 District Heat 37 7 G&A 171 33 Property-related Costs 10 2 Others 41 8 Total Fixed Cost Opex 520 100 17.2.7 Royalties and Fees Royalties and fees Opex make up 1% of total operating costs and amounts to EUR23m over the LOM period, as shown in Table 17-9. The royalty rate of EUR0.5 per tonne ore is applied with an adjustment for PPI which results in an effective royalty rate of EUR0.57/t for Syväjärvi volumes and EUR0.59/t for Rapasaari volumes. Table 17-9: Royalties and fees Opex. Item Total (MEUR) % Landowner Payment 6 26 Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) Payment 0.1 0 Property Tax 5 22 Closure Bank Guarantee 3 14 Royalties 8 37 Total Royalties and Fees Opex 23 100 17.3 Financial Costs Indicators 17.3.1 LiOH.H2O Price Based on the analysis described in Chapter 15 of this report and the Consensus Commodity Price Forecasts updated on 17 January 2025, the Lithium hydroxide monohydrate price assumptions in Table 17-10 have been adopted in the financial model. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 362 The average lithium hydroxide price applied in the financial model and calculated as a price mix of Technical Grade, Battery Grade – Contract Basis, and Battery Grade – and Contract Basis, is USD19,194/t LiOH as shown in Table 17-10. Table 17-10: LiOH price forecast. Year Technical Grade LiOH Price, Contract Basis (USD/t) Battery Grade LiOH Price, Contract Basis (USD/t) Battery Grade LiOH Price, Spot Basis (USD/t) Average LiOH Price, Applied Mix Calculation (USD/t) 2026 15,478 16,643 16,643 15,478 2027 17,708 19,041 19,041 18,264 2028 19,459 20,923 20,923 20,923 2029 18,027 19,384 19,384 19,384 2030 18,027 19,384 19,384 19,384 2031 18,027 19,384 19,384 19,384 2032 18,027 19,384 19,384 19,384 2033 18,027 19,384 19,384 19,384 2034 18,027 19,384 19,384 19,384 2035 18,027 19,384 19,384 19,384 2036 18,027 19,384 19,384 19,384 2037 18,027 19,384 19,384 19,384 2038 18,027 19,384 19,384 19,384 2039 18,027 19,384 19,384 19,384 2040 18,027 19,384 19,384 19,384 2041 18,027 19,384 19,384 19,384 2042 18,027 19,384 19,384 19,384 2043 18,027 19,384 19,384 19,384 2044 18,027 19,384 19,384 19,384 2045 18,027 19,384 19,384 19,384 2046 18,027 19,384 19,384 19,384 2047 18,027 19,384 19,384 19,384 Average 19,194 17.3.2 Cost Indicators Unit costs are calculated in the financial model using the following volume drivers: • Total Ore Tonnes: 14,426,296 • Spodumene Concentrate Tonnes: 2,686,409 • Lithium Hydroxide (LiOH) Product Tonnes: 291,403

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 363 Table 17-11 shows the Total Cost per LiOH product tonne as EUR7,869/t LiOH, which consists of: • Variable Costs amounting to EUR6,005/t LiOH; • Fixed Costs amounting to EUR1,786/t LiOH; and • Fees and Royalties amounting to EUR79/t LiOH. Table 17-11: Unit costs. Opex Unit Costs Unit Rate Total Variable Costs EUR/t LiOH 6,005 Mining Cost per Ore Tonne EUR/t ore 36 Mining Cost per Product Tonne EUR/t LiOH 1,767 Crushing, Sorting & Stockpiling per Ore Tonne EUR/t ore 0.40 Crushing, Sorting & Stockpiling per Product Tonne EUR/t LiOH 20 Concentrator, incl. Water Treatment per Tonne Concentrate EUR/t conc 79 Concentrator, incl. Water Treatment per Product Tonne EUR/t LiOH 724 Conversion & Keliber lithium Refinery per Product Tonne EUR/t LiOH 3,038 Other Variable Costs per Product Tonne EUR/t LiOH 348 Freight and Transportation per Product Tonne EUR/t LiOH 109 Total Fixed Costs EUR/t LiOH 1,786 Process Labour Cost per Product Tonne EUR/t LiOH 897 Other Operating Costs per Product Tonne EUR/t LiOH 126 Property-Related Cost per Product Tonne EUR/t LiOH 36 Other Cost per Product Tonne EUR/t LiOH 141 Selling, G&A Costs per Product Tonne EUR/t LiOH 586 Fees & Royalties EUR/t LiOH 79 Total Cost per Product Tonne EUR/t LiOH 7,869 17.3.3 Steady State Indicators The average Opex per Product Tonne analysis is shown in S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 364 Table 17-12, with the Opex per Product Tonne after reaching Steady State Production in 2030 amounting to EUR8,047/t LiOH. The Cash Cost C1 after reaching Steady State Production in 2030 is calculated as EUR7,963/t LiOH. The average Earnings Before Interest, Taxes, Depreciation, and Amortisation (EBITDA) during steady state (2029–2044) is EUR136m. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 365 Table 17-12: Steady state indicators. Opex/t – Average Values Unit Rate Total Average Opex/tonne EUR/t LiOH 7,869 Average Opex/tonne excl. Dev. & Construction EUR/t LiOH 7,809 Average Opex/tonne excl. Prod. Starting Year EUR/t LiOH 7,526 Average Variable Opex/tonne EUR/t LiOH 6,083 Opex/tonne Steady State 2030 EUR/t LiOH 8,047 Lithium Hydroxide Cash Cost C1 (adjusted for NPR) EUR/t LiOH 7,963 Average EBITDA during Steady State (2029–2044) EURm 136 17.4 Accuracy of Estimates The Keliber Lithium Project is in the construction phase of the Syväjärvi mine for the concentrator, Keliber lithium refinery, and associated infrastructure. Rates and estimates were received from the Keliber team (the Company) as part of the financial model that was provided (refer to Table 17-13). The accuracy range of ±15% has been received from Keliber and assumed based on the contingency allowance per cost item, and the level of project development. Table 17-13: Accuracy of estimates. Cost Item Source Confidence Level Capex Initial Capex From Company, Actual expenditure to date, Detail Engineering & Firm Quotes 0% Contingency is applied. -15% to +15% Sustaining Capex From Company, Budget Quotes 20% Contingency is applied. -15% to +15% Opex Drill and Blast From Company, Mining Tender Rate Comparison 10% Contingency is applied. -15% to +15% Load and Haul From Company, Mining Tender Rate Comparison 10% Contingency is applied. -15% to +15% Mining Technical Services From Company, Mining Tender Rate Comparison 10% Contingency is applied. -15% to +15% Crushing and Sorting From Company, Crushing Tender Rate Comparison 10% Contingency is applied. -15% to +15% Concentrator Plant - Energy, Reagents, Consumables, Maintenance From Company, Estimates and Budget Quotes 20% Contingency is applied. -15% to +15% Concentrator Plant Water Treatment - Energy, Reagents, Consumables, Maintenance From Company, Estimates and Budget Quotes 20% Contingency is applied. -10% to +15% S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 366 Cost Item Source Confidence Level Keliber lithium Refinery – Conversion, Energy, Reagents, Steam, Process Water, Consumables, Utilities, Maintenance From Company, Detail Engineering & Firm Quotes 10% Contingency is applied. -10% to +15% Keliber lithium Refinery Water Treatment – Reagents, Consumables, Energy From Company, Detail Engineering & Firm Quotes 10% Contingency is applied. -10% to +15% Freight & Transportation From Company, Estimates 10% Contingency is applied. -10% to +15% Labour From Company, Estimates 10% Contingency is applied. -10% to +15% G&A From Company, Estimates 10% Contingency is applied. -10% to +15%

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 367 18 ECONOMIC ANALYSIS 18.1 Principal Assumptions VBKOM received and reviewed the financial model, Keliber_Economic_Model_v4.0_LoM2024_ quick_update_20250126_Ore_Reserve_Final, from Keliber Lithium Mine. This model was interpreted to determine the economic viability of the Project when updating and declaring Mineral Reserves since previous declarations. The Discounted Cash Flow (DCF) method is used to calculate the NPV and Internal Rate of Return (IRR) and, as a result, the intrinsic value of the Keliber Lithium Mine in real terms. The NPV is derived post-royalties and -tax from pre-debt real cash flows after considering operating costs, capital expenditures for the mining operations, concentrator, and Keliber lithium refinery, and using forecast macro-economic parameters. 18.1.1 Basis of Evaluation of the Mining Assets The financial model developed by Keliber includes the following considerations when determining the economic analysis of the Project: • Conversion of Mineral Reserves: Only Indicated Mineral Resources and Measured Mineral Resources in the LOM plan were considered for conversion to Mineral Reserves. • Financial Model Structure: The cash flows are converted to Euros (EUR) currency and set up in calendar months, from 31 October 2024 to 31 December 2047. • Present Value Date: The financial model is based on a present value date of 31 October 2024. • Discount Rate: A real discount rate after tax of 8% is applied when discounting future cash flows to the Present Value Date of 31 October 2024 and calculating the NPV. • Historical Expenditure: Historical cash flows and expenditures are excluded from the NPV calculation. Therefore, the financial model only considers cash flows from the Present Value Date onwards when determining the economic results. • Inflation: No inflation has been applied in the financial model, therefore, all figures are in Real terms. • Start Dates and Ramp-Up: − The Syväjärvi mine starts stripping waste in June 2025 and ore mining starts in November 2025. − The Rapasaari mine starts stripping waste in February 2028 and ore mining starts in February 2029. − The concentrator ramps up in 6 months, from January 2026 to June 2026. − The Keliber lithium refinery ramps up in 9 months, from May 2026 to January 2027. Therefore, product sales are modelled from May 2026. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 368 18.2 Macro-Economic Forecasts The macro-economic forecasts in the financial model include LiOH Price and USD/EUR Exchange Rate. 18.2.1 LiOH Price Based on the analysis described in Section 15 of this report and the Consensus Commodity Price Forecasts updated on 17 January 2025, the LiOH price assumptions in Table 18-1 have been adopted in the financial model. Based on the analysis described in Chapter 15 of this report and the Consensus Commodity Price Forecasts updated on 17 January 2025, the LiOH price assumptions in Table 18-1 have been adopted in the financial model. Sensitivities were evaluated for this parameter, as shown in Section 18.9 of this report. Table 18-1: LiOH price forecast. Sales Product Start Date Currency 2026 2027 2027– 2032 2032– 2045 2045 2045– 2047 Ramp-up start flag Jan’26 USD/t EUR/t First LiOH sales, Technical grade May’26 USD/t 15,478 17,708 EUR/t 13,685 15,496 Switch to LiOH Sales Battery grade Contract Aug’27 USD/t 19,041 EUR/t 16,659 Switch to LiOH Sales Battery grade Contract & Spot Mix Aug’32 USD/t 19,384 EUR/t 16,525 Switch to Purchased Concentrate, Technical Grade LiOH Sale Jun’45 USD/t 18,027 EUR/t 15,368 Switch to LiOH sales, Battery Grade Contract Dec’45 USD/t 19,384 EUR/t 16,525 18.2.2 Exchange Rate The foreign exchange (FX) assumption for USD-EUR is shown in Table 18-2. Sensitivities were evaluated for this parameter, as shown in Section 18.9 of this report. Table 18-2: Exchange rate forecast. FX Rate: USD-EUR 2026 2027 2028 2029–2047 Exchange Rate (USD-EUR) 1.131 1.143 1.158 1.173 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 369 18.3 Working Capital Working capital was considered consistent with the repayment of current overdrafts at the end of LOM. The following working capital assumptions are modelled: • Sales Inventory turnover is 14 days; therefore, between 50% and 55% of inventory is sold in the current month, depending on the number of calendar days, and the remaining inventory is sold in the next month. • Receivables Average Time to Receive Payment is 45 days. • Payables Average Time to Make Payment is 30 days. 18.4 Recoveries Modelled parameters included in the Technical Economic Model are shown for each deposit in Table 18-3. Recoveries are shown for selected months with block grades around 1% Li2O for comparative purposes. Table 18-3: Modelled lithium recoveries included in the Technical Economic Model. Parameter Unit Syväjärvi Open Pit Jun-28 Rapasaari Open Pit Feb-30 Ore Grade in Block % Li2O 0.99 1.00 Block wall rock dilution % 14.30 21.94 Block mass t 64,142 61,540 Ore Grade % (without dilution) % Li2O 1.16 1.28 Sorter efficiency % % 73.00 73.00 p1 – Grade parameter/multiplier % 10.60 11.30 p2 – Grade parameter/exponent % -0.88 -0.88 p3 – Dilution parameter -0.33 -0.27 p4 – Scale-up parameter for full scale -1.27 -1.27 p5 – Scale-up parameter for pilot scale -5.42 -5.42 Targeted Concentrate Grade % Li2O 4.50 4.50 Corrected Li2O Recovery % in full scale - Final value % 88.00 87.71 Conversion degree % 100 100 Li2O.H2O Yield % 86.00 86.00 Conversion + Li2O.H2O Yield % 86.00 86.00 Global Lithium Yield % 75.68 75.43 LiOH.H2O t 1,349.78 1,303.79 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 370 18.5 Discount Rate A real discount rate post-tax of 8% is applied when discounting future cash flows to the Present Value Date of 31 October 2024 and calculating the NPV. This is an assumption received from Keliber, which has been applied to the Project economic analysis since 2020 and was deemed appropriate in previous TRS submissions. This rate is not based on Keliber’s capital structure, market rate, risk premiums, etc. as would be included in a Weighted Average Cost of Capital (WACC) calculation. Sensitivities were evaluated for this parameter, as shown in Section 18.9 of this report. 18.6 Cash Flow Forecast The undiscounted free cash flow (FCF) is calculated in Table 18-4 and shown in Figure 18-1. Table 18-4: FCF summary. Item MEUR Revenue 4,796 Opex (2,293) Inventory Movement 0 EBITDA 2,503 Working Capital (0) Capex (651) FCF before Tax 1,852 Corporate and Mineral Tax (345) FCF after Tax 1,507 The payback period of the Project in an Equity-funded scenario is 5.4 years after the start of production in May 2026, therefore, payback will be in October 2031, as shown in Figure 18-2.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 371 FCF Forecast Figure 18-1: FCF forecast. Cumulative FCF Forecast Figure 18-2: Cumulative FCF forecast. ( 400) ( 300) ( 200) ( 100) - 100 200 300 400 2 0 2 4 2 0 2 5 2 0 2 6 2 0 2 7 2 0 2 8 2 0 2 9 2 0 3 0 2 0 3 1 2 0 3 2 2 0 3 3 2 0 3 4 2 0 3 5 2 0 3 6 2 0 3 7 2 0 3 8 2 0 3 9 2 0 4 0 2 0 4 1 2 0 4 2 2 0 4 3 2 0 4 4 2 0 4 5 2 0 4 6 2 0 4 7 2 0 4 8 M E U R Cashflow Forecast Revenue OPEX Inventory Movement Working Capital CAPEX Corporate & Mineral Tax EBITDA Free Cash Flow after Tax (1 000) ( 500) - 500 1 000 1 500 2 000 2 0 2 4 2 0 2 5 2 0 2 6 2 0 2 7 2 0 2 8 2 0 2 9 2 0 3 0 2 0 3 1 2 0 3 2 2 0 3 3 2 0 3 4 2 0 3 5 2 0 3 6 2 0 3 7 2 0 3 8 2 0 3 9 2 0 4 0 2 0 4 1 2 0 4 2 2 0 4 3 2 0 4 4 2 0 4 5 2 0 4 6 2 0 4 7 2 0 4 8 M E U R Cumulative Cashflow Forecast Cumulative Free Cash Flow after Tax Production Start May 2026 Payback Oct 2031 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 372 18.7 Net Present Value The NPV and IRR are derived post-royalties and -tax, from pre-debt real cashflows after considering operating costs, capital expenditures for the mining operations, concentrator, and Keliber lithium refinery, and using forecast macro- economic parameters. The NPV is calculated as EUR411.67 million and the IRR is 16.87% over the evaluation period, as shown in Table 18-5. Table 18-5: Key metrics. Key Metrics Unit Value NPV, 8% @31 Oct 2024 EURm 411.67 IRR % 16.87 Capital Payback Period Years 5.4 LOM Period Years 20 Evaluation Period Years 25 Total Mined Ore Tonnes 14,426,296 Total Product Sold Tonnes 291,403 18.8 Regulatory Items 18.8.1 Government Royalties As described in the Mining Tax Act (314/2023) which was incorporated by the Finnish Government on 3 March 2023, a 0.6% value-based royalty is levied on metallic minerals, including lithium. The royalty rate of EUR0.5 per tonne ore is applied with an adjustment for PPI which results in an effective royalty rate of EUR0.57/t for Syväjärvi volumes and EUR0.59/t for Rapasaari volumes. Royalties amount to EUR8,5m in the financial model. 18.8.2 Corporate Taxes The following taxes and fees are applied in the financial model: • Corporate Tax: 20% is applied to taxable income and amounts to EUR311m. • Mineral Tax: 0.6% tax on concentrate is applied on the value of pure Li content in Li2O. Mineral Tax amounts to EUR13,9m in the financial model. • Property Tax: Annual tax rates are calculated for property, as follows: − Land Areas Tax Rate: EUR54,393/a − Concentrator Tax Rate: EUR102,041/a − Keliber lithium refinery Tax Rate: EUR167,957/a − TSF Tax Rate: EUR57,165/a Total property tax amounts to EUR5m. • Other Landowner Fees: Hectare-based landowner fees, REACH payments, and bank closure guarantees amount to EUR9,4m. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 373 18.9 Sensitivity Analysis The following risks were identified during the review of the financial model and, as a result, sensitivities were modelled to understand the magnitude of the risks. 18.9.1 Plant Ramp-Up Curve Sensitivities The ramp-up of the concentrator plant is modelled as 6 months, from January 2026 to June 2026, and the ramp-up of the Keliber lithium refinery is modelled as 9 months, from May 2026 to January 2027, as shown in Table 18-6 and Figure 18-3. Table 18-6: Plant ramp-up curve. Ramp-Up Sequence Concentrator Plant (%) Keliber lithium Refinery (%) Jan-26 20 Feb-26 40 Mar-26 60 Apr-26 75 May-26 90 15 Jun-26 100 20 Jul-26 100 35 Aug-26 100 50 Sept-26 100 70 Oct-26 100 80 Nov-26 100 90 Dec-26 100 95 Jan-27 100 100 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 374 Plant Ramp-Up Curve Figure 18-3: Plant ramp-up curve. These timelines are considered significantly reduced from the previous technical report (refer to section 1.5) assumptions and recommendations that stated 8 months for the concentrator and 24 months for the Keliber lithium refinery. Also, considering the Keliber lithium refinery is a novel process when functioning as an assembled technology, there is a risk that the modelled 100% LiOH product may not be achieved within 9 months. A sensitivity was evaluated on the duration of the Keliber lithium refinery ramp-up, from 9 months to 21 months. Should the Ramp-up curve be delayed or not achieve nameplate production (i.e. 10% less than capacity as referred to in Section 9.4), then only Technical Grade prices will be received for product sold until the Keliber lithium refinery produces a chemical-grade product. A sensitivity was evaluated to extend the technical grade prices from 15 to 21 months in conjunction with the delayed plant ramp-up. Additionally, a 3% yield drop for Rapasaari material could realise if the Keliber lithium refinery recoveries are not achieved as designed. A sensitivity was evaluated to reduce the Rapasaari yield by 3% in conjunction with extending the Technical Grade prices and the delayed plant ramp-up. The following sensitivities were evaluated individually (Table 18-7) and then as combined scenarios: 1. Base case: The financial model as received from Keliber (9 months Keliber lithium refinery ramp-up); 2. Technical grade increased from 15 to 21 months; 3. -10% price discount during the 9 months Keliber lithium refinery ramp-up period; 4. -3% Reduction in Rapasaari yield; 5. 21 months Keliber lithium refinery ramp-up; and 6. The nameplate production is reduced by 10%, with a maximum capacity of 90%. 0% 20% 40% 60% 80% 100% 120% Jan-26 Feb-26 Mar-26 Apr-26 May-26 Jun-26 Jul-26 Aug-26 Sept-26 Oct-26 Nov-26 Dec-26 Jan-27 Ramp-up Curve CP Plant LHP Plant

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 375 Table 18-7: Plant ramp-up individual sensitivities. Parameter Base Case Technical Grade for 21 Months -10% Price Discount Rapasaari 3% Less Yield Adjustment 21 Months Ramp Up 90% Maximum Capacity NPV (EURm) 411.67 400.11 398.03 373.74 323.81 240.17 IRR (%) 16.87 16.71 16.46 16.25 14.03 13.44 Technical-Grade LiOH Price, Spot-Based (EUR/t) - Maximum Price 15,493 16,804 15,368 15,493 15,493 15,493 Battery-Grade LiOH Price, Contract-Based (EUR/t) - Maximum Price 18,069 18,069 18,069 18,069 18,069 18,069 Battery-Grade LiOH Price, Spot-Based (EUR/t) - Maximum Price 16,525 16,525 16,525 16,525 16,525 16,525 The combined scenarios of these individual sensitivities ( S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 376 Table 18-8) are as follows: 1. Base case: The financial model as received from Keliber (9 months Keliber lithium refinery ramp-up); 2. Technical grade increased from 15 to 21 months (9 months Keliber lithium refinery ramp-up); 3. Technical grade increased from 15 to 21 months and -10% price discount during that period (9 months Keliber lithium refinery ramp-up); 4. Technical grade increased from 15 to 21 months and -10% discount during that period (9 months Keliber lithium refinery ramp-up) and -3% reduction in Rapasaari yield; 5. Technical grade increased from 15 to 21 months and -10% discount during that period (21 months Keliber lithium refinery ramp-up) and -3% reduction in Rapasaari yield; and 6. Technical grade increased from 15 to 21 months and -10% discount during that period (21 months Keliber lithium refinery ramp-up) and -3% reduction in Rapasaari yield, the nameplate production is reduced by 10% with a maximum capacity of 90%. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 377 Table 18-8: Plant ramp-up combined sensitivities. Parameter Base Case Technical Grade for 21 Months Technical Grade for 21 Months - 10% Discount Technical Grade for 21 Months - 10% Discount & 9 Months Ramp Up & Rapasaari 3% Less Yield Technical Grade for 21 Months - 10% Discount & 21 Months Ramp Up & Rapasaari 3% Less Yield Technical Grade for 21 Months - 10% Discount & 21 Months Ramp Up & Rapasaari 3% Less Yield & 90% Maximum Capacity NPV (EURm) 411.67 406.15 385.35 347.42 271.52 119.45 IRR (%) 16.87 16.71 16.11 15.48 13.11 10.41 Technical-Grade LiOH Price, Spot- Based (EUR/t) - Maximum Price 15,493 16,804 15,368 15,368 15,368 15,368 Battery-Grade LiOH Price, Contract- Based (EUR/t) - Maximum Price 18,069 18,069 18,069 18,069 18,069 18,069 Battery-Grade LiOH Price, Spot- Based (EUR/t) - Maximum Price 16,525 16,525 16,525 16,525 16,525 16,525 The scenarios showed a positive business case in each case, thus enabling VBKOM to sign off the Mineral Reserves. 18.9.2 Other Sensitivities Other sensitivities were evaluated to test the robustness of the economic viability, and not necessarily due to any perceived technical risks: • Price: The sensitivity of prices was tested in increments of 5% with a range of -25% to +25%. Table 18-9 shows the results of the sensitivity and, notably, the Base Case NPV of EUR411.67 million is still positive at -20% reduced prices, and turns negative at -25%. The upside range shows a theoretical sensitivity should prices include up to +25% of current market prices. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 378 Table 18-9: Price sensitivity. Parameter Base Case -5% -10% -15% -20% -25% 25% 20% 15% 10% 5% NPV (EURm) 411.67 325.40 238.13 150.84 63.31 -24.24 847.62 759.59 672.52 585.98 499.48 IRR (%) 16.87 15.19 13.40 11.52 9.53 7.39 24.60 23.11 21.62 20.10 18.53 Average Technical-Grade LiOH Price, Spot- Based (EUR/t) 17,639 16,757 15,875 14,993 14,111 13,229 1,292 1,241 1,189 1,137 1,085 Average Battery-Grade LiOH Price, Contract- Based (EUR/t) 19,189 18,230 17,270 16,311 15,351 14,392 16,536 15,875 15,213 14,552 13,890 Average Battery-Grade LiOH Price, Spot-Based (EUR/t) 19,233 18,271 17,310 16,348 15,386 14,425 10,604 10,179 9,755 9,331 8,907

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 379 • Discount Rate: A sensitivity on the Discount Rate of 8% (Table 18-10) was evaluated by increasing with increments of 2% up to 16% (double the current rate). The NPV remained positive in these sensitivities. It is not expected that the Discount Rate would be lower than 8%, therefore, an upside sensitivity was not evaluated. Table 18-10: Discount rate sensitivity. Parameter Base Case Discount Rate +2% Discount Rate +4% Discount Rate +6% Discount Rate +8% NPV (EURm) 411.67 278.30 173.78 90.96 24.65 IRR (%) 16.87 16.87 16.87 16.87 16.87 Discount Rate (%) 8 10 12 14 16 • Exchange Rate: A sensitivity on the USD-EUR Exchange Rate of 1.13 (as at 31 October 2024) was evaluated by increasing with increments of 5% up to 20%. The NPV remained positive in these sensitivities (Table 18-11). Table 18-11: Exchange rate sensitivity. Parameter Base Case Exchange Rate +5% Exchange Rate +10% Exchange Rate +15% Exchange Rate +20% NPV (EURm) 411.67 413.04 254.99 186.18 123.31 IRR (%) 16.87 16.91 13.75 12.30 10.91 Exchange Rate at 31 Oct 2024 1.13 1.19 1.24 1.30 1.36 • Logistics Cost: A sensitivity on the Logistics Cost was evaluated to test the impact of Shanghai, China, as a destination for the LiOH project instead of the current destination at Antwerp, Belgium (Table 18-12). The Freight Cost increased from kEUR21,855.21 in a 100% Antwerp scenario to kEUR37,372.79 in a 100% Shanghai scenario – this is a 71% increase in cost; however, the NPV decreased by EUR4.77 million (1%), showing that Logistics Cost is not a major driver. Table 18-12: Logistics cost sensitivity. Parameter 100% to Antwerp (Base Case) 50% to Antwerp; 50% to Shanghai 100% to Shanghai NPV (EURm) 411.67 408.74 406.90 IRR (%) 16.87 16.81 16.78 Freight to Antwerp (EUR/t) 75 75 75 Freight to Shanghai (USD/t) 150 150 150 LiOH Shipment to Antwerp 100 50 0 LiOH Shipment to Shanghai 0 50 100 Freight Cost (kEUR) 21,855 29,614 37,373 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 380 18.10 Economic Analysis Conclusions VBKOM reviewed the financial model developed by Keliber and are satisfied with the positive NPV of EUR411.67 million, IRR of 16.87%, and Payback Period of 5.4 Years. Total Capex amounts to EUR651 million. Opex amounts to EUR2,293 million over the LOM period. Sensitivities were evaluated to test the robustness of the Project’s economic viability, given technical risks and other economic drivers. Most sensitivities resulted in a positive NPV, which shows a robust business case for the Project. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 381 19 ADJACENT PROPERTIES [§229.601(b)(96)(iii)(B)(20)] The Keliber Lithium Project is the most advanced lithium project in the region. It is likely that there is potential for the identification and exploration of additional similar orebodies in the region, including under the current Keliber licence areas, however, there are no other lithium exploration licences held by other companies surrounding the Keliber licence areas. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 382 20 OTHER RELEVANT DATA AND INFORMATION [§229.601(b)(96)(iii)(B)(21)] 20.1 Project Implementation A project implementation plan was prepared by Sweco Oy (Sweco) for the establishment of the Syväjärvi mining site, Päiväneva concentrator site, and Keliber lithium refinery. These sites comprise the initial capital footprint. In 2021 Keliber has selected Sweco as an Engineering, Procurement and Construction management (EPCM) contractor to supply services for the project implementation and reduced the scope of Sweco only as Engineering service provider in late 2022. Keliber took over the procurement and selected Nipromec and Construction Management (CM) service provider at lithium refinery construction and Fimpec as CM service provider at concentrator construction site. Construction of the Keliber lithium refinery commenced in 2023 and is expected to be completed in Q4 2025. The construction of the concentrator plant started in Q4, 2023. Key milestone dates are shown in Table 20-11. Table 20-1: Project milestones. Milestone Milestone Date Keliber lithium Refinery – final acceptance December 2025 Päiväneva Concentrator – start of earthworks October 2023 Päiväneva Concentrator – cold commissioning completed Q4 2025 Keliber lithium Refinery – hot Commissioning February 2026 Päiväneva Concentrator – hot commissioning January 2026 Syväjärvi Mine – start of roads, wetland treatment Q3 2025 Syväjärvi Mine – first ore November 2025 Start of sustaining Capex October 2024(1) End of initial Capex December 2026(1) Rapasaari Mine – start of site work – open pit 2028 Rapasaari Mine – first ore February 2029 (Sources: Keliber, 2024, LOM 2024) 20.2 Exploration Programme and Budget [12.10(e)(i)-(iii), 12.10(h)(vi)], SR3.1(i)-(vi), SR 3.2(i)] Currently, Keliber has an exploration budget for the years 2024–2025. The exploration budget for 2024 was EUR4.1 million and a total of 15,019 m was drilled in 2024. For 2025, Keliber’s annual exploration budget is EUR2.0 million. The coming years exploration budget will be considered according to exploration results and tenement holdings. A total of 7,500 m is planned to be drilled in 2025. Drilling will be focused especially on the Leviäkangas deposit to upgrade mineral resource category, and as a brown field targeting to check the extensions of the known spodumene pegmatite veins and to explore the possible subparallel vein systems at Leviäkangas. Greenfield type of exploration drilling is planned with ca. 2,000 m to the new targets, e.g. the Heikinkangas area. Geochemical exploration will also be conducted using percussion drilling methods to obtain samples from the bedrock surface as well as from the basal

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 383 till. Additional greenfield exploration work will include boulder mapping and surface till sampling. The next updates of Mineral Resource estimations will be finalised in 2026. 20.3 Risk Review 20.3.1 Introduction The following section presents the key interpretations for the risk review for Keliber. The risk review considered documents provided by SSW as well as information available in the public domain. 20.3.2 Risk Assessment A detailed risk assessment was undertaken by SSW and reviewed by the authors of this Report. Figure 20-1 shows the matrix utilised to determine the impact severity of each risk identified, while Table 20-2 shows the risks associated with the highest severity. Risk Severity Heat Map Figure 20-1: Risk severity heat map. Table 20-2: Risk assessment – medium to high risks. Ref ID Risk Description Likelihood 1 – 5 Impact 1 – 5 Risk Severity Score Likelihood x Impact R1 Analcime sand disposal, permitting needed for disposal area in case it cannot be disposed in harbour 4 3 12 R10 Concentrator EP not valid when the operations are planned to commence 3 4 12 R11 Construction of concentrator tailings sand facilities is not ready for the commissioning 3 4 12 R17 Delay in Kokkola commissioning 4 3 12 R47 Lithium hydroxide market price lower than expected 3 4 12 R75 Unfavourable exchange rate USD/EUR 3 4 12 R79 Wastewater treatment planned in Päiväneva cannot reach permit limits for nitrogen 3 4 12 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 384 Ref ID Risk Description Likelihood 1 – 5 Impact 1 – 5 Risk Severity Score Likelihood x Impact R9 Concentrator cost overrun - Insufficient amount of contingency 5 2 10 R68 Syväjärvi waste rock material may not be adequate for construction for the concentrator area 5 2 10 R82 No deposit for analcime sand when the production begins. The cost for using external service is more expensive. 5 2 10 R3 Keliber lithium refinery, engineering battery limits 3 3 9 R16 Debt costs higher than expected 3 3 9 R35 ETP: delay in equipment installation 3 3 9 R38 ETP: unclear battery limits / tie-ins inside the Project 3 3 9 R41 Fire at Päiväneva 3 3 9 R53 Operators leaving company due to delayed commissioning of Kokkola 3 3 9 R57 Project implementation in MO not effective 3 3 9 R59 Päiväneva schedule: unclear authority schedules and requirements 3 3 9 R81 Delayed execution of agreement with Kokkolan Teollisuusvesi and hot commissioning is delayed 3 3 9 R84 Loss of the local acceptance for the Project due to either bad performance or inadequate communication 3 3 9 R12 Cost increases due to need to treat wastewater from Rapasaari waste rock area 2 4 8 R31 Energy price risk 4 2 8 R37 ETP: technology risk 4 2 8 R39 Facilities for control room, laboratory and social rooms are not locked, risk that they are not ready in 9/2025 2 4 8 R51 Management of extractive waste 4 2 8 R55 Permit limit for water intake / outlet in Köyhäjoki is too strict for operations 4 2 8 R61 Rapasaari ore cannot be treated in the concentrator due to high Arsenic content 2 4 8 R62 Refinery capex overrun 4 2 8 R65 Soda leaching fails to perform as expected causing lower profitability 2 4 8 R71 The concentrator plant has serious ramp-up issues, which the Project cannot cope with 2 4 8 R78 Wastewater treatment in Päiväneva cannot reach the permit limits for sulphate 2 4 8 Mitigation measures for the highest ranked risks are provided in Table 20-3. The full risk Register is given in Appendix A. Table 20-3: Risk assessment – high risks mitigation measures. Ref ID Risk Description Risk Severity Score Primary Mitigation R1 Analcime sand disposal, permitting needed for disposal area in case it cannot be disposed in harbour 12 1. Obtain Port's commitment to take the analcime sand produced 2025 - 2026. 2. Progress the permitting as fast as possible. R10 Concentrator EP not valid when the operations are planned to commence 12 1. Application covering concentrator has been submitted in May 2024 and the application covering Rapasaari will only be sent once the hearing of concentrator part is completed. This is done to reduce risk that AVI (permitting authority) would combine the procedures and through that delay the concentrator related decision. 2. Permit application for concentrator was drafted in a safest possible manner to avoid potential delays at AVI or in a possible appeal process. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 385 Ref ID Risk Description Risk Severity Score Primary Mitigation 3. Keliber have been discussing the matter in detail with ELY (monitoring authority) to try to ensure that their approach and opinion (to be given to AVI) would not be overly negative. 4. Once AVI is asking for further information, Keliber are ensuring that our responses are of excellent quality and provided in a timely manner. 5. Additional information provided to AVI. Frequent communication with AVI to ensure that hearing takes place. R11 Construction of concentrator tailings sand facilities is not ready for the commissioning 12 Prepare a detailed plan and schedule of TSF area ELY approvals. The plan must be divided by each pond and specific structure to be approved by ELY. Requires smooth communication with independent quality control (Ramboll) and authorities. R17 Delay in Kokkola commissioning 12 1. Create a ramp-up plan (Cold Commissioning, Hot Commissioning, Ramp-Up. 2. Evaluate resources, training program and flag need of additional people if needed 4. Develop a production plan. 5. Update a sales plan and Financial Plan. 6. Based on 1 - 5 prepare a detailed risk assessment Heikki Pihlaja & Sirpa Olaussen => monthly review by Management Team to be called for 7. Construction: Area managers nominated. Main tasks to control overall progress of their areas. Progress monitored by smaller milestones, critical paths identified. Contractors to provide plans that are followed up. 8. To define end points for cold commissioning, hot commissioning, ramp-up. 9. Revisit once schedule available. R47 Lithium hydroxide market price lower than expected 12 The aim is to conduct the offtakes with market prices which means that Keliber are open for market risk. Fixed priced offtakes and sales would also be a risk. R75 Unfavourable exchange rate USD/EUR 12 Sensitivity scenarios prepared to plan for different FX rates. Current plan and the financial model based on a relatively conservative assumption. R79 Wastewater treatment planned in Päiväneva cannot reach permit limits for nitrogen 12 1. To plan a new outlet / inlet in Perhonjoki (impacts smaller on a large stream), 31.5. a reservation for new outlet is in the sustaining capex. No ongoing actions with permitting. 2. Arrange a visit to Kittilä and Kemi mines to discuss nitrogen removal in their wastewaters. 31.5. completed, it seems that 7,5 mg/l is a very challenging goal, also cost could be higher than expected. 3. Updated modelling (phased closure of the waste rock area) shows that permit condition can be met (requires reaching the 7,5 mg/l in outlet), sustaining capex corrected for capacity (doubled, no engineering yet) 4. Ongoing discussions with explosives providers to find an explosive without nitrogen. 5. Permit appealed by Keliber 6. Basic engineering of the Rapasaari water management should be started 2025 R9 Concentrator cost overrun - Insufficient amount of contingency 10 1. Cost of external aggregates is coming from EP limits of arsenic and sulphur for construction rock. This will be handled as scope change and update of the budget/forecast. 2. All procurements must be done within the budget and in case of overruns handled individually according to company processes described in IMS. 3. Cost control and regular forecast updates needed as normal routines. R68 Syväjärvi waste rock material may not be adequate for construction for the concentrator area 10 1. Obtain ELY approval for the sampling method and frequency so Syväjärvi construction stone can be used. 2. Review the possibility to purchase external material to be able to keep the construction schedule 11.4.2024 in use. 3. Review possibility to apply for a change in permit conditions relating to sulphur & arsenic - Will be done as part of the planned permit amendment applications. 4. Project / geology to set up a working group to manage the construction stone matter. 5. Project to ensure that the instructions for rock quality approval are communicated. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 386 Ref ID Risk Description Risk Severity Score Primary Mitigation R82 No deposit for analcime sand when the production begins. The cost for using external service is more expensive. 10 1. The whole area does not need to be ready at the beginning 2. Ongoing negotiations with the harbour 3. To be assessed how much of the sand is produced during the first year 4. To be assessed whether we need a temporary storage for the ramp-up period (if the sand cannot be placed into the harbour) 5. Sampling plan for the sand for the ramp-up period 6. Inquiring ELY whether the sand from ramp-up time can be placed in the harbour after analysis 7. Goal is to apply EP at the end of February 2024. EIA Report will be given parallel and is identical to EP 8. To find another waste disposal area where we can dispose the sand with a cost. 20.3.3 Overview of Additional Specific Risk Elements The available information for the Project identifies and/or points to the following additional broader risk-related issues: 20.3.3.1 Tenure Currently, there are mining permits in place for Länttä, Syväjärvi, and Rapasaari, and a number of applications have been submitted (as well as prepared, pending submission) for exploration and mining permits. However, there is some uncertainty regarding the time required for the authorities to process the applications. It is understood that Keliber are completing a legal due diligence exercise to understand the permitting risks. Additional permits are required to start Päiväneva concentrator and Rapasaari mining operations, including permitting for disposal area and concentrator EP. Additional risk aspects are described in Chapter 2.8. While these may delay development and/or operations, the resolution of these risks are not required for the declaration of Mineral Resources. There are no known material environment- and social-related risks concerning active Syväjärvi and Rapasaari mine sites. Public perception of potential environmental impact related to mining appears to be changing. Uncertainty regarding potential objections by the public and/or authorities to the award of tenure for each of the applications exists. The relevance of the uncertainty is that the current Project does not appear to have considered scenario models if some of the applications, or specific applications, are either significantly delayed or are wholly unsuccessful. 20.3.3.2 Closure Aspects Risks related to closure and controlling measures are listed in the AFRY plan. Appropriate mitigation measures must be implemented and maintained throughout the operations to reduce or eliminate these risks: • The seepage water quantities from the waste rock facilities, TSF, and pre-float tailings facility may be larger than estimated and the load of harmful substances may be greater than anticipated and, therefore, the impact on soil, groundwater, and nearby surface water may be greater than estimated; • Covered waste rock facilities are exposed to erosion. If erosion occurs, an increase in water flow through the facilities could mobilise harmful substances. Pyrite-containing rock oxidation may also increase; − Risks can be mitigated by following the precautionary principle in planning and assessment, monitoring during construction and closure, and drainage and monitoring through the quarry. • Ramp distortion may damage the cover structure and thus increases risk of contaminant transport which could also pose a hazard to people and animals in the area;

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 387 − Risks can be controlled by supervision during the construction and closure phases. • TSF collapse would cause water and tailings discharge into the environment. This may result in the release of contaminants to soil, groundwater, and surface water (The amount of water in the reservoir will decrease with closure, so the environmental spillage would be less severe than during the production phase.); − Risks can be controlled with dam safety inspections, design and quality control and documentation of design and construction, supervision, and elevations downstream on the excavated embankment. • Possible soil contamination not cleaned after operations. Contaminated soil can impact on groundwater and surface waters; − Risks can be controlled during active operations by preventing spills and leakages. 20.3.3.3 Mineral Resource Estimation The following risks related to the MRE have been identified: • The topographic layer that acts as the upper boundary for the MRE model is interpreted from drill collars and is considered adequate at this stage of the Project for resource estimation. Topographic surveys are available and should be utilised for future work and mine planning stages. • The continuity of the spodumene pegmatite at Rapasaari: it appears more structurally complex. There is greater uncertainty in the geological model of the mineralised domains and the estimation within these domains. Further infill drilling will increase confidence in the geological model. • Xenoliths within the spodumene pegmatite are internal waste; the drill density is not sufficient to accurately model the volume of these units at this stage. Further infill drilling as the development of the projects progresses will increase the resolution of the location and extents of xenolith bodies. • Deleterious elements have not been reported as the drilling assay data for these variables is incomplete. SSW are currently undertaking a programme of re-sampling and re-assaying to improve the dataset. Future updates should include estimates for the significant deleterious elements. • Classification categories have been applied to blocks within each deposit to qualify the estimation risk based on the input data. The drilling density is used as the criteria for classification, as listed in Section 11.15, as data collected from the drilling core/chips (logged lithology and assays) are the main inputs for Mineral Resource estimation. A portion of the MRE reported for the Project is classified as Inferred Mineral Resources, which has the lowest drill density at a spacing of >30 m between drill holes. An Inferred Mineral Resource is that part of Mineral Resources for which quantity and grade or quality are estimated on the basis of limited geological evidence and sampling. The level of geological uncertainty associated with an Inferred Mineral Resource is too high because it is informed by less drilling data to apply relevant technical and economic factors likely to influence the prospects of economic extraction in a manner useful for the evaluation of economic viability. Because an Inferred Mineral Resource has the lowest level of geological confidence of all Mineral Resources, which prevents the application of the modifying factors in a manner useful for the evaluation of economic viability, an Inferred Mineral Resources may not be considered when assessing the economic viability of a mining project and may not be converted to Mineral Reserves. In order to increase confidence and, therefore, classification of Mineral Resources, further work would be required to refine the grade continuity, geological continuity, and RPEE required for Mineral Resources. The style of mineralisation is similar between the deposits, and they are all in relatively close proximity. The continuity of the larger veins in all five of the deposits is demonstrated to be good during the geological modelling, with relatively uncomplicated morphology. Considering the continuity of the pegmatite veins, the risk is considered low. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 388 20.3.3.4 Mineral Reserve Estimation The key operational risks that could impact the Mineral Reserves are listed below: • Commodity prices and exchange rate assumptions: SSW have adopted forward-looking price assumptions. Any material deviations from these assumptions could impact the Mineral Reserves. • Geology and grade control: The orebody orientation and thickness require strict grade-control measures to ensure the dilution is kept to a minimum. • Mining operational cost: Mining costs utilised during the pit optimisation are based on the 2023 contractor quotes estimates. • Infrastructure timeline: Delays in the completion dates of the Project infrastructure could impact negatively on the production ramp-up. • Waste rock specification: Restrictions on waste dumping with deleterious elements could impact the sequence of mining when ensuring the deleterious elements’ limits are adhered to. • Permitting: Although several of the required operating permits have been obtained, potential timing delays due to public objections and appeals could impact construction timelines. EP conditions could also be strenuous, impacting or delaying planned mining operations. • Plant recoveries: The efficiency of the plant is highly dependent on the close management of the Blackrock dilution in the plant feed as too much dilution will impact the efficiency of the ore sorters negatively. • Human capital: A significant number of skilled personnel will be required to develop and work at the operations. Labour availability could impact planned production and build-ups. • Waste storage facilities and capacity: The Syväjärvi WRD height requirement increased by 6 m, and this amendment to the permit is required. The Rapasaari WRD requires extension beyond the current mining area; approval of this is required to ensure sufficient space is available. • OVB waste storage capacity: The Syväjärvi final OVB dump requires environmental approval to ensure sufficient dumping space is available. • Hazardous waste storage facilities and capacity: Hazardous waste material is defined as sulphur >1.0% and arsenic >100 ppm. In the previous study, the impact of arsenic was overlooked. For Syväjärvi, the storage requirement marginally increased, however, at Rapasaari, the storage requirement increased to above 8 MLCM. Areas were identified to host the hazardous material, with associated environmental controls. This requires amendment to permitting. • ROM ore stockpile facilities and capacity: Any delays in the permitting of the Syväjärvi ROM Ore stockpile facilities could impact production build-up, potentially impacting the production ramp-up. The Rapasaari Ore stockpile size needs to be optimised and sufficient space needs to be identified. 20.3.3.5 Tailings Storage Facility Given the short periods allocated for construction of the flotation tailings pond, there is no flexibility to accommodate any delays in construction of this facility. To mitigate this, a contractor with a proven track record has been appointed by Keliber; this contractor has designed and constructed similar structures in Finland. Regarding short-term performance of the proposed compacted peat/till liner system beneath the flotation tailings pond, limited additional testing has been undertaken during the detailed design on peat (or samples of underlying till materials) to confirm field conditions or anticipated hydraulic conductivities under compression. Given the variability S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 389 of the peat in the limited samples tested to date, there is a residual risk of excess seepage from the base of the facility (particularly across areas where peat depths are <0.5 m and target permeability values cannot be assured). Only one of the three till samples tested meets the permeability and fines criteria set out in the EP. Keliber undertook a ground survey programme in late 2024 to study conditions in more detail. The results of the programme are partly pending, and additional testing of the peat properties will be done. Authority approval is in place for the construction of the TSF (EP clause 61). Limited geochemical testwork was previously identified as a risk. Keliber have continued humidity cell testing of the Syväjärvi and Rapasaari pilot processing samples. The samples do not show high potential to produce acid mine drainage. In addition to that, Keliber have completed benchmarking with other Finnish mine projects to mitigate the risk of low performance of the selected peat liner solution at the beginning of production. The peat will be more compacted when more tailings are in the area, and the permeability of the peat will decrease over time. Based on the reference sites and design, the hydraulic conductivity is towards drainage collection ditches and conductivity through peat liner will be minimal or almost zero. Based on this addition testing, the risk is deemed to have been mitigated. The spatial distribution of waste types in the Syväjärvi open pit should be confirmed to ensure there is sufficient inert waste rock for use in embankment construction activities. Alternative borrow source may be required if this cannot be achieved, which could impact construction schedule and result in re-design of embankments and have capital cost implications. Keliber have used external aggregates to mitigate this described risk. Although no tailings settling tests have been completed to confirm drained and undrained densities for all tailings types, it is deemed that the value of 1.4 t/m3 is reasonable for the flotation tailings. While this should be verified, along with the other tailings streams, to ensure that each pond has been sized appropriately, the volumes of the TSF are deemed to be sufficient with this conservative estimate. Additional testing will be completed in 2025. 20.3.3.6 Water Management Significant water bodies are present at the Syväjärvi and Emmes deposits and would require careful management. Flow rate modelling parameters need to be carefully considered to accurately determine the amount of fresh water available, and potential impacts on downstream water quality, need to be carefully investigated. In addition, updated hydrogeological modelling at Rapasaari and Syväjärvi is required, the results of which must be incorporated into operational planning. The impact of amended water balance and infrastructure designs may impact directly on capital and operating costs, as well as Project execution timelines. 20.3.3.7 Processing The feed to the ore-sorting test equipment comprised an artificial blend of Syväjärvi ore and waste rock. There is a risk that performance on mined ore may be less efficient than on the artificial composite ore feed due to poor liberation or variations in feed PSDs’. Despite spodumene mineralisation being generally homogeneously distributed throughout most of the pegmatites, the contamination caused by the inclusion of host rock xenoliths and wall rock material with ore material will impact the metallurgical recovery of spodumene during flotation and metallurgical processing. This will require careful selective mining supported by ore sorting to mitigate the impacts of contamination on the recovery of spodumene. The Keliber Project is likely to be the first implementation of the Metso Outotec lithium hydroxide flowsheet. While the individual unit processes are not novel, and while the Syväjärvi (2020) and Rapasaari (2022) pilot trials have significantly de-risked the flowsheet, a residual risk remains, as it does with the first example of any novel technology. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 390 Potential concerns were noted that the processing plant may not cope with the arsenic levels from Rapasaari material, which may lead to LiOH product falling to technical grade. 20.3.4 Opportunities The inclusion of Keliber into SSW’s Battery Metals assets portfolio and battery metals strategy is an important step in acquiring further downstream exposure to the battery metals value chain. Lithium hydroxide (a chemical needed in the production of the cathode active material in modern high-nickel cathode materials, which provide higher energy density) is expected to become the dominant lithium chemical consumed in battery applications. In the future, Keliber will offer lithium hydroxide, especially for the needs of the strongly growing lithium battery market. The BG produced can be used for the manufacturing of batteries for increasingly electrifying transport (electric and hybrid vehicles) as well as in the production of batteries for energy storage.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 391 21 INTERPRETATION AND CONCLUSIONS 21.10 Permitting All permits are in place to commence development and operations at Syväjärvi. Good progress is being made with obtaining outstanding necessary permissions for the Päiväneva concentrator and Rapasaari mining operations. Finalising the Kokkola EP and issuance (subject to appeal) of the Rapasaari and Päiväneva permit has triggered the need for additional design work in a few areas, as well as the implementation of the measures necessary to ensure compliance with the permit conditions. To support this process, Keliber have: undertaken further studies; implemented internal software systems to track requirements; and started developing ISO aligned management systems with the aim of obtaining certification in 2024. There is a need for review or development of management plans to support implementation and to reflect design changes. This includes the extractive waste management plan and the closure plan (and associated cost estimate). Subject to the specific technical issues associated with permit changes and the potential delays caused by ongoing appeal processes and analcime sand approval processes, no material issues linked to ESG have been identified. 21.11 Geology and Mineral Resources The Keliber Project is located in the KLP Province of western Finland, covering an area of about 500 km2. At least ten (10) lithium-bearing pegmatite deposits were discovered through a combination of till geochemistry coupled with boulder mapping and sampling and, most recently, evaluated by diamond core drilling. Outcropping pegmatites and their host rocks are rare, most being covered by 3–18 m of OVB comprising surficial sediments (mostly glacial till). The spodumene-bearing pegmatite deposits that have been evaluated to date within the Kaustinen area all have very similar mineralogy and are dominated by albite, quartz, K-feldspar, spodumene, and muscovite. These rare element pegmatites belong to the LCT group of pegmatites and to the poorly zoned albite-spodumene subgroup of pegmatites. The presence of numerous contemporaneous granites (many being pegmatitic granites) in the Kaustinen area are thought to be the potential sources of the pegmatites. However, there has been no clear or well-defined zonation observed or sufficiently accurate geochronology of the granite and pegmatites to date to support this, and more recent models related to the origin of pegmatites from direct products of anatexis (i.e. in situ melting of potentially fertile host rocks) should be considered. The pegmatites are mostly moderately- to steeply-dripping and hosted in a sequence of metavolcanic and metasedimentary rocks (mica schists). The pegmatites are generally poorly zoned, often with a variably developed outer border and marginal zone of quartz-feldspar-muscovite (with little to no spodumene mineralisation) and a mineralised core of quartz-feldspar-spodumene (±muscovite). The composition of the spodumene pegmatites in the area are typically coarse-grained, light-coloured, and mineralogically similar comprising, on average, albite (37–41%), quartz (26–28%), K-feldspar (10–16%), spodumene (10–15%), and muscovite (6–7%). Keliber’s exploration has focused on seven (7) of these pegmatite deposits, namely Rapasaari, Syväjärvi, Tuoreetsaaret, Länttä, Emmes, Leviäkangas, and Outovesi. Apart from data generated from OVB stripping at Länttä and the exploration tunnel in Syväjärvi, diamond core drilling, from the 1960s to present day, has been the only method used to generate geological, structural, and analytical data, and these have been used as the basis for Mineral Resource estimation over each of the deposits defined to date. Keliber have been following a well-defined logging, sampling, and analytical procedure since 2014. The sampling and core storage facility in Kaustinen is considered a secure facility with the sample preparation and analytical S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 392 methodologies considered appropriate for the commodity being evaluated (lithium). ERM considers the sample database is of sufficient quality and accuracy for use in Mineral Resource estimation. Since commencement of exploration in the Kaustinen region, Keliber have completed a systematic exploration and Mineral Resource evaluation programme that has been successful in delineating seven discrete spodumene- mineralised pegmatite deposits. The work completed to date has captured the important variables (namely, mineralogical, structural, and lithological) required to properly define the attitude of the host pegmatite/s and, importantly, the spodumene or grade distribution within the various pegmatites that host each deposit. The historical data generated prior to Keliber’s involvement in the Project are also considered to be suitable for inclusion in the database used for Mineral Resource estimation. In ERM’s opinion, the exploration data that have been captured to date (consisting primarily of drilling data) are of sufficient quality to be used in Mineral Resource estimation and for the purposes used in this TRS. The Mineral Resources have been estimated using conventional industry standard techniques, and the continuity of the modelled veins has been adequately demonstrated through the wireframe modelling, which supports the lateral and down-dip continuity of the mineralised veins. The following non-material gaps and risks were identified during the data review and estimation process: • Data management has not been centralised and is susceptible to version control issues and inconsistencies of data structure across the various deposit databases. • Some adjustments to the assay QAQC should be considered and include resolving the apparent underperformance of the internal reference materials against the expected values. Additional recommendations are made in Section 23. • The bounding topography for the MRE model is derived from drill collar data and is considered adequate at this stage of the Project for Mineral Resource estimation. • The geological interpretation at Rapasaari is complex, and there is greater uncertainty on grade continuity compared with the other deposits. • The location and extents of internal waste within the mineralised domains of the various deposits in the form of xenoliths is not accurately resolved at the current drill spacing. • Estimates for deleterious elements have not been included in the resource as the database for these variables is incomplete. In order to increase confidence and, therefore, classification of Mineral Resources, further work would be required to refine the grade continuity, geological continuity, and RPEE required for Mineral Resources. The style of mineralisation is similar between the deposits, and they are all in relatively close proximity. The continuity of the larger veins in all five of the deposits is demonstrated to be good during the geological modelling, with relatively uncomplicated morphology. Considering the continuity of the pegmatite veins, the risk is considered low. The risks related to estimation of Mineral Resources are, therefore, expected for early Project stages. The classification categories applied to the Mineral Resources appropriately qualify the risk with respect to the confidence in the data, interpretation, and the vein and grade continuity. Keliber has an exploration budget of EUR2.0 million for 2025. Geochemical exploration will also be conducted using percussion drilling methods to obtain samples S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 393 from the bedrock surface as well as from the basal till. Additional work will include boulder mapping, surface till sampling, and Mineral Resource estimation. ERM considers the budget to be appropriate. ERM were not involved in any of the exploration conducted but have reviewed the exploration completed to date and the supporting documentation provided by Keliber. The Mineral Resource estimates have been prepared and reported by ERM. Overall, the QPs consider the data used to prepare the geological models and MRE are accurate and representative and have been generated with industry-accepted standards and procedures. 21.12 Processing Although the testwork on the ore sorters proved positive, the feed to the ore-sorting test equipment comprised only of artificial blends of ore and waste rock. There is a risk that performance on ROM ore may be less efficient than on the artificial composite ore feed due to particles not being liberated. The final concentrate quality and Li recovery are highly dependent on the feed grade and ROM PSDs’. As part of the mining schedule, provision was made for various stockpiles to manage the feed grade. However, the impact of rehandling of the material on the PSD of the plant feed, which has a direct impact on the bypass stream and milling plant feed qualities, has not been evaluated. The principal processing risk lies with the implementation of the Metso-Outotec hydrometallurgical process. The Keliber Project is likely to be the first implementation of this specific flowsheet. While the individual unit processes are not novel, and while the 2020 and to a slightly lesser extent the 2023 pilot trials have significantly de-risked the flowsheet, a residual risk remains, as it does with the first example of any novel technology. The ramp-up schedule (3 months hot commissioning followed by 9 months ramp-up) is in line with the 2022 FS study ramp-up for internal Keliber ore bar the fact that the refinery will not start up with third party spodumene anymore (previously planned for up to 12 months) but will operate from day one with ore originating from Keliber’s own operations. The 12-month period with third party concentrate would have allowed time to address all the operating issues, usually experienced during the start-up of high-pressure leaching plants. Sibanye-Stillwater acknowledges the risk of commissioning both plants near parallel with each other and additional allowances were made in the estimate to cater for separate commissioning teams. This is a significant change from the previously reported ramp-up schedule and sensitivities in this regard were tested to evaluate the robustness the financial model and relayed positive results. Given the short periods allocated for construction of the flotation tailings pond, there is no flexibility to accommodate any delays in construction of this facility. To mitigate this, a contractor with a proven track record has been appointed by Keliber. Acceptance has been received to use the peat liner solution for the flotation tailings facility. 21.13 Mining and Mineral Reserves The resource model estimation methodology was altered from the previous study, whereby internal dilution was included in the estimation process, resulting in more defined lithium grades. The resource models were regularised to 5 m x 5 m x 2.5 m, incorporating dilution and losses. The modifying factors for both Syväjärvi and Rapasaari were investigated in depth during this study. The dilution and losses from the regularisation process resulted in 11.08% dilution and 14.97% losses for Syväjärvi and 11.66% dilution and 15.67% losses for Rapasaari. The modifying factors also include perimeter constraints whereby the extent of the pits was limited in their extents. Marginal COGs were determined as 0.188% Li₂O for Syväjärvi and 0.3% for Rapasaari. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 394 The ore definition was altered, taking the marginal cut of grades into account. The orebody orientation and thickness require strict grade-control measures to ensure the dilution is kept to a minimum. The pit optimisation for both Syväjärvi and Rapasaari was completed based on a constrained scenario, hence, the shell selection was limited. The 2024 geotechnical study presents significant advancements in geotechnical engineering compared to previous assessments. A major improvement involved new core logging conducted within ±50 metres of pit slopes, which provided better modelling of FLT and shear zones. This resulted in the geotechnical parameters complying with a feasibility level of accuracy. The geotechnical study resulted in an increase in slope angles for both Syväjärvi and Rapasaari. The Syväjärvi pit was designed as a single large pit; in contrast, Rapasaari’s, being much larger, was designed to incorporate three pushbacks which enables phased ore extraction to maintain steady production rates with mostly constant stripping ratios. The combined LOM production for Syväjärvi and Rapasaari spans from 2025 to 2044, with plant feed ending in April 2045. Syväjärvi is expected to produce 3.15 Mt of ore at an average Li₂O grade of 1.03%, while Rapasaari will yield 11.28 Mt of ore at 0.93% Li₂O. The stripping ratio for Syväjärvi is 5.14, whereas Rapasaari’s overall stripping ratio is 7.63. The Syväjärvi WRD elevation was lifted by 6 m due to the increase in tonnage from the FS work. The OVB dump elevations were also increased to ensure sufficient capacity. The Rapasaari WRD requires additional space due to the increased waste stripping. No backfill was included in the Project. The primary loading units will include backhoe excavators fitted with 4.6 m³ buckets for ore and 5.8 m³ buckets for waste, while haulage will be managed using 64-tonne rear-dump trucks for waste and 41-tonne articulated trucks for ore. No Inferred Mineral Resources were included, only Measured Mineral Resources were converted to Proven Mineral Reserves, and only Indicated Mineral Resources were converted to Probable Mineral Reserves. 21.14 Economic Analysis VBKOM reviewed the financial model developed by Keliber and are satisfied with the positive NPV of EUR411.67 million, IRR of 16.87%, and Payback Period of 5.4 Years. Total Capex amounts to EUR651million. Opex amounts to EUR2,293 million over the LOM period. Sensitivities were evaluated to test the robustness of the Project’s economic viability, given technical risks and other economic drivers. Most sensitivities resulted in a positive NPV, which shows a robust business case for the Project.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 395 22 RECOMMENDATIONS 22.1 Permitting All technical work and preparation for new or amendment applications must be done timeously and in accordance with legislation and regulations as applicable to avoid Project delays. Additionally, continued engagement with regulators and stakeholders must be upheld to facilitate all permitting processes. 22.2 Geology and Mineral Resources ERM recommend that Keliber utilise an additional umpire/check laboratory to analyse a larger subset of the previously analysed samples representative of the grade of the deposits and exploration timespan (2013–2023). They also recommend Keliber include additional commercially available CRMs across a broader lithium grade range as part of its QC programme going forward in order to address the possible negative bias observed in the exploration assay results and more recent increased variance in results with the recent move to the Oulu laboratory. The cost of the umpire laboratory checks is expected to be approximately EUR10k–20k. The cost of commercially available Li CRMs for a three- year period would be approximately kEUR3–5. This is aligned with the previous recommendations by SRK (2023). The implementation of a fit-for-purpose relational database with timely backups will ensure a robust and secure database going forward. In addition, it will make data extraction, assay management, data interrogation, and export simpler and avoid version control issues and make auditing more traceable (cost of approximately EUR15,000 per year for initial implementation and monthly hosting). It is understood that Keliber are in the process of implementing a database solution for the entire Project. Streamline the data generation and capture workflows to integrate directly with the database solution implemented through the use of paperless data capture. Adjust sample protocols to ensure all mineralised pegmatites, irrespective of size, are sampled as well as apparently unmineralised muscovite pegmatites. In doing so, the geological data collected are more robust and easier to model, and the opportunity to miss potentially mineralised pegmatites is minimised. Revisiting the validation and verification of the historical data for Lantta and Leviäkangas against the more recent exploration data generated by Keliber to assess whether there are any gaps or deficiencies in the data, is recommended. Investigation into the use of hyperspectral core scanning to aid geological logging and material characterisation (from a geological, processing, geotechnical, and environmental perspective) should be considered (cost of approximately EUR40,000). Consider developing geo-metallurgical models to improve the understanding of the deposits (deposit intelligence) with respect to variations in potentially deleterious elements (from both a processing and environmental perspective), variations in mineralogy, and amenability to processing, etc. This will require the integration of additional data such as, but not limited to, multi-element data and data collected as part of the core logging process. This may require a review of the of the core logging and sampling protocols (cost of approximately EUR25,000). Look at the potential to produce quartz, feldspar, mica, cassiterite, and columbo-tantalite concentrates as by-products to the production of the spodumene concentrates. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 396 ERM consider there to be potential for definition of additional Mineral Resources through the planned exploration programme and through targeted infill and extension drilling of the already-defined deposits. The estimated exploration programme costing is summarised in Section 20.2. Infill drilling is also part of planned future work and could improve confidence in the size and grade of these deposits and give greater resolution location and extent of internal waste xenolith bodies. Some of this infill drilling was completed in 2024 and additional drilling focussed mostly on the Leviäkangas deposit as well as some in new target areas for 2025. Other recommended data to be collected and considered in future work are: • Topographic surveys are recommended to inform future mine planning activities; and • Compilation of a robust drilling dataset for significant deleterious elements from an environmental and processing perspective. It is understood that Keliber are currently undertaking a programme of re-sampling and re-assaying for this purpose. Future Mineral Resource updates should include estimates for the significant deleterious elements (cost of approximately EUR20,000). 22.3 Processing Detailed planning is required to manage the ramp-up of both plants simultaneously. Training of key personnel before hot commissioning is key to ensure the successful execution of the ramp-up plan. Control of feed grade and feed PSDs is critical to ensure final product qualities. Additional testwork to understand the impact of Rapasaari ore on the recoveries is recommended. 22.4 Mining and Mineral Reserves Geotechnical monitoring is required during operation. The Rapasaari pushback 2 design requires adjustment due to instances in the production schedule where only one access is available. It is recommended to optimise the pushback selection of Rapasaari. The orebody orientation and thickness require strict grade-control measures to ensure the dilution is kept to a minimum. The Syväjärvi WRD height requirement increased by 6 m, and this amendment to the permit is required. The Rapasaari WRD requires extension beyond the current mining area; approval of this is required to ensure sufficient space is available. Similarly, amendments to the OVB dumps are also required. The increase in hazardous waste to 8 MLCM for Rapasaari requires an EIA and application for an EP. 22.5 Economic Analysis VBKOM reviewed the financial model developed by Keliber and are satisfied with the positive NPV of EUR411.67 million, IRR of 16.87%, and Payback Period of 5.4 Years. It is recommended that the ramp-up schedules of the two plants and the Rapasaari recovery be critically reviewed. The discount rate used in the model should be adjusted to the WACC calculated for the Project. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 397 23 RELIANCE ON INFORMATION PROVIDED BY REGISTRANT [§229.601(b)(96)(iii)(B)(25)] VBKOM have relied on information provided by SSW (the registrant) and its advisors in preparing this TRS with regard to the following aspects, which are beyond the scope of VBKOM’s expertise: • Status of permitting and authorisations, as well as legal matters and tenure; • Environmental and social studies; • Economic trends, data, assumptions and commodity price forecasts; and • Marketing information. VBKOM believe it is reasonable to rely upon the registrant for the above information for the following reasons: • Legal matters – VBKOM do not have in-house expertise to confirm that all mineral rights and environmental authorisations/permits have been legally granted and correctly registered. VBKOM would defer to a written legal opinion on the validity of such rights and authorisations, which would be provided by the Company; • Commodity prices and exchange rates – VBKOM do not have in-house expertise in forecasting commodity prices and exchange rates and would defer to industry experts for such information which came via the Company; • VBKOM have reviewed the publicly available data to confirm the data provided by the registrant and is satisfied there is acceptable agreement; and • SSW have confirmed in writing that to their knowledge the information provided by them to VBKOM was complete and not incorrect, misleading or irrelevant in any material aspect. VBKOM have no reason to believe that any material facts have been withheld. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 398 24 DATE AND SIGNATURE PAGE 24.1 Resources This TRS documents and justifies the Mineral Resource statement for SSW’s Keliber lithium project located in Central Ostrobothnia, Finland as prepared by CSA Global South Africa (Pty) Ltd in accordance with the requirements of SK- 1300 and the SAMREC Code. The opinions expressed in this TRS are correct as at the effective date of 31 December 2024. We, CSA Global South Africa (Pty) Ltd, are the Qualified Persons (as defined in SK-1300) who are responsible for authoring this Technical Report Summary in relation to the Keliber Lithium Project Mineral Resource Estimation. We hereby consent to the following: • the public filing and use by Sibanye Stillwater Limited (Sibanye-Stillwater) of the Keliber Lithium Project Technical Report Summary; • the use and reference of our name, including our status as experts or Qualified Persons (as defined in SK-1300) in connection with this Technical Report Summary for which we are responsible; • the use of any extracts from, information derived from or summary of this Technical Report Summary for which we are responsible in the annual report of Sibanye-Stillwater on Form 20-F for the year ended 31 December 2024 (Form 20-F); and • the incorporation by reference of the above items as included in the Form 20-F into Sibanye-Stillwater’s registration statement on Form F-3 (File No. 333-234096) (and any amendments or supplements thereto). This consent pertains to the Keliber Lithium Project Technical Report Summary, and we certify that we have read the Form 20-F and that it fairly and accurately represents the information in the Keliber Lithium Project Technical Report Summary. CSA Global South Africa (Pty) Ltd Authorised Signatory Date: 25 April 2025 (Report Date: 25 April 2025) (Effective Date: 31 December 2024) /s/ Graham M. Jeffress Graham M. Jeffress

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 399 24.2 Reserves This TRS documents and justifies the Mineral Resource and Mineral Reserve statements for SSW’s Keliber project located in Central Ostrobothnia, Finland as prepared by VBKOM in accordance with the requirements of S-K1300 and the SAMREC Code. The opinions expressed in this TRS are correct at the Effective Date of 31 December 2024. We, VBKOM (Pty) Ltd, are the Qualified Persons (as defined in SK-1300) who are responsible for authoring this Technical Report Summary in relation to the Keliber Lithium Project Mineral Reserve Estimation. We hereby consent to the following: • the public filing and use by Sibanye Stillwater Limited (Sibanye-Stillwater) of the Keliber Lithium Project Technical Report Summary; • the use and reference of our name, including our status as experts or Qualified Persons (as defined in S-K1300) in connection with this Technical Report Summary for which we are responsible; • the use of any extracts from, information derived from or summary of this Technical Report Summary for which we are responsible in the annual report of Sibanye-Stillwater on Form 20-F for the year ended 31 December 2024 (Form 20-F); and • the incorporation by reference of the above items as included in the Form 20-F into Sibanye-Stillwater’s registration statement on Form F-3 (File No. 333-234096) (and any amendments or supplements thereto). This consent pertains to the Keliber Lithium Project Technical Report Summary, and we certify that we have read the 20-F and that it fairly and accurately represents the information in the Keliber Lithium Project Technical Report Summary. VBKOM (Pty) Ltd Otto Wilhelm Warschkuhl Authorised Signatory Date: 25 April 2025 (Report Date: 25 April 2025) (Effective Date: 31 December 2024) /s/ Otto Wilhelm Warschkuhl Otto Wilhelm Warschkuhl S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 400 REFERENCES AFRY. October 2020. Structural model of the Rapasaari Li-deposit. Vantaa, Finland: Consultants' report for AFRY Finland Oy. AFRY. 2021. Keliber Lithium Project – Definitive Feasibility Study Site Water Management Plan. Project ID: 101016050- 003. AFRY. 2021. Structural model of the Syväjärvi Li-deposit. Vantaa, Finland: Consultants' report for AFRY Finland Oy. AFRY. August 2024. Structural geological mapping of the Syväjärvi open pit. Vantaa, Finland: Internal Consultants' report for AFRY Finland Oy. AFRY. 2025. Kaivannaisjätteen jätehuoltosuunnitelma Päiväneva. AFRY. 2025. Päivänevan rikastamoalueen sulkemissuunnitelma. AFRY. 2025. Rapasaaren kaivoksen kaivannaisjätteen jätehuoltosuunnitelma. AFRY. 2025. Rapasaaren kaivoksen sulkemissuunnitelma. AFRY. 2025. Syväjärven kaivoksen kaivannaisjätteen jätehuoltosuunnitelma. AFRY. 2025. Syväjärven kaivoksen sulkemissuunnitelma. Ahtola, T. 2012. Emmes Li. In: Eilu, P. et al. Metallogenic areas in Finland. In: Eilu, P. (ed.) Mineral deposits and metallogeny of Fennoscandia. Geological Survey of Finland. Special Paper 53. Ahtola, T. (ed.), Kuusela, J., Käpyaho, A. & Kontoniemi, O. 2015. Overview of lithium pegmatite exploration in the Kaustinen area in 2003–2012. Geological Survey of Finland. Report of Investigation 20. Alviola, R., Mänttari, I., Mäkitie, H. and Vaasjoki, M. 2001. Svecofennian rare-element granitic pegmatites of the Ostrobothnian region, Western Finland; their metamorphic environment and time of intrusion. Geological Survey of Finland. Special Paper 30 (2001). pp. 9-29. Barton, N. 1987. Discontinuities. In F.B. (ed.), Ground Engineers' Reference Book (pp. Chapter 5, 5.1-5.15). London: Butterworths. Barton, N., & Bandis, S. 1990. Review of the predictive capabilities of the JRC - JCS model in engineering practice. Rock Joints. Proceedings of International Symposium (eds N Barton & O Stephansson) (pp. 603-610). Lone, Norway: Balkema, Rotterdam. Barton, N., & Chaubey, V. 1977. The Shear Strength of Rock Joints in Theory and Practice. Rock Mechanics, Vol. 10, 1- 54. Bradley, D., & McCauley, A. 2016. A Preliminary Deposit Model for Lithium-Cesium-Tantalum. Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study for the Rapasaari Open Pit (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. Edited by Abraham Saayman, GEOTEC Africa CC. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 401 Broda, L., Blomqvist, N., Nieminen, V. 10 February 2025. Definitive Feasibility Study of the Syväjärvi Open Pit Mine (Geotechnical Slope Design Parameters and DFS Pit Design Review). AFRY Finland Oy, Version 8. Edited by Abraham Saayman, GEOTEC Africa CC. Černý, P., & Ercit, T. S. 2005. The Classification of Granitic Pegmatites Revisited. The Canadian Mineralogist. 43. Chudasama, B., & Sarala, P. 2022. Mineral prospectivity mapping of lithium- spodumene pegmatites in the Kaustinen region of Finland: Implications for lithium exploration in Finland. Geological Survey of Finland. Open File Report. 6.6.2022. CSA Global South Africa (Pty) Ltd, & ERM Group Company. 21 April 2024. Technical Report Summary. Effective date 31/12/2023, Report number R142.2024. Deminey, J.G.L., & Barnard, E. 02 December 2024. Keliber Lithium Mine – Dewatering and Drainage Strategy. Nurizon Consulting Engineers, Revision Number 1, Final. Duuring, P. 2020. Rare-Element Pegmatites: A Mineral Systems Analysis. Geological Survey of Western Australia. Record 20/7. 13 pp. FLSmidth. January 2024. Evaluation of spodumene alpha to beta phase conversion utilising a cyclone preheater rotary calciner system for Keliber Technologies OY Finland. Project # 9232517826. Fossen, H. 2010. Structural Geology. 2nd edition. Cambridge University Press. 2037 p. Garrett, D.E. 2004. Handbook of lithium and natural calcium chloride (1st ed.): Boston, Mass. Elsevier. 476 p. GEOTEC Africa. June 2024. Review of Syväjärvi 3D fault models. Johannesburg, RSA: Internal Consultants' report for SSW-Keliber. https://www.refworld.org/pdfid/4e5cf5f12.pdf Keliber Oy Lithium Project Definitive Feasibility Study Report. Hatch, 28 February 2019. Keliber Oy Lithium Project Definitive Feasibility Study. Hatch, 14 June 2018. Keliber Oy Lithium Project Economic Model, Version 4.0. Keliber, 26 January 2025. Keliber Technology Oy: Rapasaaren ja Syväjärven kaivosten pölypäästöjen leviämismallinnus. AFRY Finland Oy 4.11.2021 (Finnish). Keliber Technology Oy. AFRY Finland Oy 2.11.2021. Knoll, T., Huet, B., Schuster, R., Mali, H., Ntaflos, T., & Hauzenberger, C. 2023. Lithium pegmatite of anatectic origin – A case study from the Austroalpine Unit Pegmatite Province (Eastern European Alps): Geological data and geochemical modelling. Ore Geology Reviews 154. Koopmans, L., Martins, T., Linnen, R., Gardiner, N.J., Breasley, C.M., Palin, R.M., Groat, L.A., Silva, D., & Robb, L.J. 2023. The formation of lithium-rich pegmatites through multi-stage melting. The Geological Society of America. Geology 52. Kurtti, J. 2019. QAQC 2018 report. Keliber Oy. 17pp. Kurtti, J. 2020. Drilling QAQC Report Pegmatite Assays in 2019. Keliber Technology Oy. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 402 Kurtti, J. 2021. Drilling QAQC Report Pegmatite Assays in 2020. Keliber Technology Oy. Kurtti, J. 2022. Drilling QAQC Report Pegmatite Assays in 2021. Keliber Technology Oy. Lamberg, P. 2018. Summary of the Mineralogical Studies on Keliber Lithium Deposits. London, D. 2008. Pegmatites. The Canadian Mineralogist. Special Publication. 10 pp 347. London, D. 2016. Rare-Element Granitic Pegmatites. In: Reviews in Economic Geology. v.18. pp 165-193. Society of Economic Geologists 2016. London, D. 2018. Ore-forming processes within granitic pegmatites: In. Ore Geology Reviews. v.101. pp 349-383. Metso. 28 August 2024. Keliber: Recovery of Lithium and production of LiOH·H2O from calcined spodumene concentrates in a continuous pilot. Ministry of Employment and the Economy, Finland, 2011. Müller, A., Reimer, W., Wall, F., Williamson, B., Menuge, J., Brönner, M., Haase, C., Brauch, K., Pohl, C., Lima, A., Teodoro, A., Cardosa-Fernandes, J., Roda-Robles, E., Harrop, J., Smith, K., Wanke, D., Unterweissacher, T., Hopfner, M., Schröder, M., Clifford, B., Moutela, P., Lloret, C., Ranza, L., & Rausa, A. 2022. GREENPEG – exploration for pegmatite minerals to feed the energy transition: first steps towards the Green Stone Age. Geological Society. London. Special Publications. 526 (June 2023). pp.193-218. PayneGeo (Payne Geological Services Pty Ltd). 2022. Tuoreetsaaret Lithium Deposit Mineral Resource Estimate. Keliber Lithium Project, Finland. p. 89. PL Mineral Reserve Services. 2016. (Lovén, P., & Meriläinen, M.) Mineral Resource and Ore Reserve Estimates of Leviäkangas Lithium Deposits for Keliber Oy. p. 27. Read, J., & Stacey, P. 2009. Guidelines for Open Pit Slope Design. CRC Press. Read, J., & Stacey, P. 2009. Guidelines for Open Pit Slope Design. CRC Press. Rorke, A.J. 09 October 2024. Blast Designs – Syväjärvi and Rapasaari Lithium Mines. AJR, Revision Number 2. Sandberg, E. c.2013. Methods and grade control in drilling and sampling of Keliber Oy. Sandberg, E. 2014. Analytical differences between ALS and Labtium. Scogings, A. Porter, R., & Jeffress, G. (CSA Global Pty Ltd). 2016. Reporting Exploration Results and Mineral Resources for Lithium Mineralised Pegmatites. AIG News. Issue 125. September 2016. SFA Oxford. January 2025 Lithium Market Outlook to 2040. Simmons, W.B., & Webber, K.L. 2008. Pegmatite Genesis: State of the Art. (Paper presented at the symposium “Granitic Pagmatites: the State of the Art”, Porto, May 2007.) European Journal of Mineralogy. v.20. p 421-438. SRK Consulting (South Africa) (Pty) Ltd. 2022. Technical Report Summary. Effective date 31 December 2022, filed as Exhibit 96.7 to Sibanye-Stillwater Limited’s 2022 annual report filed on Form 20-F on 24 April 2023. SRK Consulting (South Africa) (Pty) Ltd. December 2023. An Updated Due Diligence Addendum Report on the Keliber Lithium Project, Finland.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 403 SRK Consulting (South Africa) (Pty) Ltd. 13 December 2023. Technical Report Summary. Effective date 31 December 2022, Report number 592138. Stacey, T. R., & Page, C. H. 1986. Practical Handbook for Underground Rock Mechanics (Vol. 12). Germany: TransTech. UBS Bank. January 2025. Bank consensus metal price forecast for Lithium Hydroxide Battery Grade (USD/t). WSP Global Inc. 2022. Definitive Feasibility Study Report – Keliber Lithium Project, 2022. (Referred to as Feasibility Study (FS) in this report. The Definitive Feasibility Study is considered to be the equivalent of a FS as defined in the S- K 1300 Definition Standards adopted 26 December 2018.) www.webmineral.com; BGS, 2016 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 404 APPENDICES Appendix A: Risk Register Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 1 Analcime sand disposal, permitting needed for disposal area in case it cannot be disposed in harbour We have to stop production in case the harbour will not receive the analcime sand formed at Kokkola Chemical plant and we are not able to complete the permitting process for analcime sand storage are in time. We have a Letter of Intent with Kokkola Harbour and there is space in the existing area for three operating years and the permitting process for the storage area is progressing. //19.10. permitting of Hoikkaneva area ongoing as planned. Negotiations with the Port relating to a binding agreement to receive analcime sand have been commenced. //20.2.2024 21.5. Port has provided an indicative price for receiving analcime sand during 2025-2026. Port has confirmed that they will provide a quote relating to this - expected to be received in May. 26.6. Port's offer received, negotiations ongoing. 11.9.2024 Permitting will not be ready by the time it is required and, hence, solution with Port covering 2025 and 2026 is needed. 1.11.2024 Advanced draft contract with Port and final negotiations during November. Consequence (General Description) Markus Kivimäki 3 4 12 1. Obtain Port's commitment to take the analcime sand produced 2025 - 2026. 2. Progress the permitting as fast as possible. 2 Chemical plant stand- alone startup, product quality We have not piloted the sourced concentrates and ensured the quality of the product. Finding operating parameters could delay ramp-up. Consequence (General Description) Heikki Pihlaja 2 2 4 We will plan pilot programs for the sourced concentrates. The process is quite robust. Purchase specification requires a 3000 kg sample for pilot. However there is a 1 year queue for the pilot (MO Pori, conversion pilot could be faster). This means that if the Chemical refinery is due to be ready in Q2/2025, the samples of the sourced concentrate should be available latest Q2/2023. We have received information from one potential customer that no need of new validation in case concentrate change. Samples from external concentrates are now on the FLS Pyro pilot plant on Whitehall, Pennsylvania. US. FLS Pilot start 8.1.2024 -->Pilot was successful LiOH Pilot will be 15.4 Metso Pori. Samples are there S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 405 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 3 Chemical plant, engineering battery limits Difficulties to incorporate MO engineering to Sweco's engineering leading to capex overrun and schedule delay Consequence (General Description) Pertti Pekkala 3 3 9 1. Sweco engineering to be audited (PP) //4.7.2023 Audit done. Findings about documentation, KPIs, project management and reviewing requirements. 2. Metso engineering has almost completed and key equipment deliveries to the site has started. There are not any more major issues related to the interfaces between the parties. 3. PPE 15.5.204: Some minor issues are still open between Metso/Sweco tie-ins and battery limits like some pump sealing water system changes. However, typically additional challenges are noticed during commissioning phase 4.11.2024 status: engineering has been almost completed and equipment installations have been progressed. Area managers have been nominated and battery limits have been separately reviewed as well as remaining open tasks. Some open topics are related to the utilities tie-ins, because related Hazops have not been completed. 4 Chemical plant, long lead-time items Long equipment delivery times. (indicative quotation states 11-13 months) leading to delayed ramp-up. Procurement packages have been re-scheduled, identified long-lead items have been sourced and ordered. There is still a risk that there are dependencies that have not been detected. Consequence (General Description) Heikki Pihlaja 3 2 6 1. Weekly procurement meetings, schedule meetings for follow-up. 5 Chemical plant, schedule delayed due to delayed process engineering (water treatment) The engineering and construction planning of the new water treatment plant is delayed. Resourcing for project management is scarce. The lead time for filters and evaporation equipment is 50 weeks. If the order is delayed it is possible that cold commissioning is delayed. //12.10. the delay expected to be 0-3 months, most critical thing being the building permit. Hot commissioning can be started without the full water treatment /1.12.2023 hot commissioning not possible without ETP. Current estimate is that the ETP will be finished 3 months after the main process (30.6.2023) Consequence (General Description) Heikki Pihlaja 1 5 5 1. To appoint a project manager for ETP 2. To recruit a process engineer for ETP /done -- 1. To choose the process and prepare a decision proposal, decide and order within May //2.6. JTC instructed to review the process alternatives with a 3rd party and prepare a decision proposal //4.7. 2nd opinion to be achieved 7.7. Opinion to be presented in JTC 14.7. //20.9.2023 decision made (phosphate precipitation + evaporation). 2. To contact other technology providers (ZLD) // 4.7.2023 Done (Alfa Laval, Veolia, GEA). Getting a new technology provider at this stage would cause delay. 6 Chemical plant, wastewater treatment Currently chosen technology is not viable due to accumulation of impurities in the main process (sulfate and chloride) if the reject is circulated back to process. There are three alternative technologies available: ZLD (issues with silica, risk for being Consequence (General Description) Heikki Pihlaja 3 2 6 1. Metso preparing a proposal for Keliber //20230704 Report expected before 7.7.2023 2. SRK will be selected to analyze and prepare a second opinion on the proposal //20230704 Report ready in S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 406 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control technology risk bottleneck, not piloted), sodium triphosphate precipitation (piloted, not reaching the current permit limits, some lithium loss) and sodium aluminate precipitation (no possibility to recycle lithium, benchtop pilot done, partly on synthetic wastewater). Risk that the chosen technology causes increased cost or is difficult to operate. There is no time for conclusive piloting before decision. 1.12.2023 decision made, phosphate precipitation followed by evaporation. Pilot for H1/2024 Pilot test ongoing in Metso Pori and by SSW Denis Beltrami in Belgium). Risk for too hifh Phosforous level after ETP to Kokkolan Vesi 26.4.2024 Friday 3.Other technology providers approached (Alfa Laval, GEA and Veolia) //20230704 Alfa Laval offered a solution (ZLD) with three evaporators (concept). NDA with GEA, no further discussions. Veolia not so interested in sole wastewater treatment, but will continue discussions. Having Jord Proxa along might cause issue with Veolia. Testing and piloting would be necessary, these will require time. Next JTC meeting in July. 4. Opinion for technology solution to be presented in July JTC, budget and schedule for decision in August JTC /20.9.2023 Decided to proceed with phosphate precipitation combined with evaporation. Basic engineering ongoing. 5. Operational cost is in financial model 6. Process guarantees from Metso 7 Commissionin g issues in either Kokkola or Päiväneva disturbances at the other site Consequence (General Description) #VALU E! 8 Commissionin g of both plants is due to take place at same time. Issues at either site cause can cause disturbance in the other site We have limited concentrate storage and buffer. If Kokkola has issues in hot commissioning or ramp-up, we need to have place for the concentrate. If Päiväneva has issues after Kokkola has started commissioning, we might have not enough feed for Kokkola. Consequence (General Description) Erkki Niska #VALU E! 9 Concentrator cost overrun - Insufficient amount of contingency AFE approved on 6th Oct 2023 included Capital cost estimate of 220,9 MEUR and contingency of 16.9 MEUR (7.1%). Contingency usage from beginning of the project (-20.8 MEUR 9/2024). There is approx. value of 20 MEUR as open purchases and signed Consequence (General Description) Ville Vähäkanga s 2 5 10 Cost of external aggregates is coming from environmental permit limits of arsenic and sulphur for construction rock. This will be handled as scope change and update of the budget/forecast during September 2024-December 2024. Other scope change items are listed during October and

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 407 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control a lot of unit priced contracts are ongoing like tailing storage facilities (situation September 2024). Also, considering the relatively low contingency reservation from the beginning 16.9 MEUR/7.1 %. will be informed SSW board in November meeting. All procurements must be done within the budget and in case of overruns handled individually according to company processes described in IMS. Cost control and regular forecast updates needed as normal routines. 10 Concentrator environment al permit not valid when the operations are planned to commence The environmental permit for Rapasaari and Päiväneva was partially reverted to AVI by Vaasa Administrative Court. Keliber has made a new application covering the Päiväneva concentrator part and that permit decision (in enforceable form) is required to commence the production at the concentrator. There is a risk that Keliber does not have an enforceable permit covering the concentrator at the time when operations are planned to commence. Consequence (General Description) Markus Kivimäki 4 3 12 1. Application covering concentrator has been submitted in May 2024 and the application covering Rapasaari will only be sent once the hearing of concentrator part is completed. This is done to reduce risk that AVI (permitting authority) would combine the procedures and through that delay the concentrator related decision. 2. Permit application for concentrator was drafted in a safest possible manner to avoid potential delays at AVI or in a possible appeal process. 3. Keliber has been discussing the matter in detail with ELY (monitoring authority) to try to ensure that their approach and opinion (to be given to AVI) would not be overly negative. 4. Once AVI is asking for further information, Keliber is ensuring that our responses are of excellent quality and provided in a timely manner. 5. Additional information provided to AVI as requested in October. Frequent communication with AVI to ensure that hearing takes place during 2024. 11 Construction of concentrator tailings sand facilities is not ready for the commissionin g Completion of the facilities is moved to summer 2025.The delay coming from missing ELY approvals for construction works in time. If we do not get the authority approvals in time, they will not be ready for the commissioning of the plant when missing approvals. New schedule of TSF area: 1) Tailing storage facility (flotation tailings) 14.10.2024 --> 30.9.2025 2) Linered tailing pond - prefloat 9.9.2024 --> 5.8.2024 3) Linered tailing pond - magnetic fraction --> 5.9.2024 (*) 4) circulating water pond 18.10.2024 --> 5.12.2024 (*) New pond required to meet the environmental permit Consequence (General Description) Ville Vähäkanga s 4 3 12 Prepare a detailed plan and schedule of TSF area ELY approvals. The plan must be divided by each pond and specific structure to be approved by ELY. Requires smooth communication with independent quality control (Ramboll) and authorities. 12 Cost increase due to need to treat waste Rapasaari waste rock cannot be placed as planned due to As content, seepwater needs to be treated Consequence (General Description) Pentti Grönholm 4 2 8 /1.12.2021/Closure plan finished. Only the sulphide- bearing waste rock needs lined area and seepwater treatment. This is considered in the planning and costs. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 408 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control water from Rapasaari waste rock area The humidity cell tests continue. Will be assessed in the closure plan. Humidity cell testing ongoing for 5 wasterock types 6.10.2022 Humidity cell tests still ongoing. Results from spring 2022 support current waste area planning. 12.4.2023 According to Rapasaari Environmental Permit (Dec2022), As-bearing waste rock has to be stored to the waste rock facility areas where are used bentonite liners. A reservation made to CapEx and sustaining CapEx. // 27.12.2023 The risk is unchanged. 13 Cost increase due to need to use more expensive technical solutions for tailings sand and prefloat ponds. We need to prove to the authority that our technical solutions are adequate. If we do not succeed in this we might have to revert to more expensive solutions. Consequence (General Description) Ville Vähäkanga s 1 1 1 Afry is reviewing the permit decisions and plans. Supplier certificate to be acquired (long-time performance) //updated information: cost effect <200 000 € 31.5. Solutions approved by authorities. AFE to be applied for so no budget yet 4.9.2023 KAIELY approved the environmental structures, EPOELY decision pending 14 Cost increase due to new customs or higher toll New customs / higher toll causes increased cost / CapEx Consequence (General Description) Riku Sauso 1 2 2 Follow-up of the situation regularly. Most of our customers and capex purchases within Europe which lowers the risk considerably. 15 Data system loss We loose our data or our data system is not available causing delays in schedule. Our reputation is damaged. Consequence (General Description) Toni Tunkkari 3 2 6 Completed //1.12.2021/ It service provider changed, data transferred to new service provider. Needed: Clear practices and instructions needed. Issues with getting equipment, lack of information in changes 16 Debt costs higher than expected Our profitability decreases because market interests rise raising debt costs. Consequence (General Description) Riku Sauso 3 3 9 Project profitability is assessed with high interest rates. In addition, arranging for the debt funding on the back of SSW support will result in cost efficiencies. Market rates have risen recently, margins and other financing costs stable. 17 Delay in Kokkola commissionin g According to current schedule the date of hot commissioning start is 2.7.2025. Current progress curve shows delay of 2-3 months for construction, further actions needed for acceleration. There are defects in installations that need to be fixed either before ramp-up or later in maintenance shutdowns (welding seams in ETP, separation possibilities against Basis of Design, maintenance issues, some equipment badly placed against the Consequence (General Description) Heikki Pihlaja 3 4 12 1. Create a ramp-up plan (Cold Commissioning, Hot Commissioning, Ramp-Up. Resp. Heikki Pihlaja, support Erkki Niska /11.9.2024 First version ready, schedule ready. Cold commissioning situation OK, hot commissioning still unclear. Toni Lahti nominated as area manager regarding EIA in Kokkola. 2. Evaluate resources, training program and flag need of additional people if needed. Heikki Pihlaja, support Asko S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 409 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control Basis of Design). Ramp-up planning ongoing but not yet detailed. Saastamoinen /11.9.2024 Training mainly scheduled, FLSmidth yet to provide a training plan - Heikki Pekkarinen to contact FLSmidth 4. Develop a production plan 7/2025 - 12/2026. Heikki Pihlaja Erkki Niska /11.9.2024 Ramp-up plan to the end of 2025 5. Update a sales plan and Financial Plan. Thomas Beck & Riku Sauso /11.9.2024 completed. 6. Based on 1 - 5 prepare a detailled risk assessment Heikki Pihlaja & Sirpa Olaussen => monthly review by Management Team /11.9.2024 to be called for 7. Construction: Area managers nominated. Main tasks to control overall progress of their areas. Progress monitored by smaller milestones, critical paths identified. Contractors to provide plans that are followed up. 8. To define end points for cold commissioning, hot commissioning, ramp-up. (Erkki Niska) 9. Revisit once we have the schedule 18 EIA Hoikkaneva: PH value of leachate Consequence (General Description) Lea Nikupeteri 4 1 4 Leachate is collected from the membraned area. Neutralization. 19 EIA Lithium Chemical Plant: Air emissions and traffic Consequence (General Description) Lea Nikupeteri 1 2 2 Dust control/binding (e.g. watering, salting) Conveyor enclosure Canopies of warehouses 20 EIA Lithium Chemical Plant: Chemical and fuel leaks - the consequence depends on the amount of the leak Consequence (General Description) Lea Nikupeteri 3 2 6 21 EIA Lithium Chemical Plant: Fire Consequence (General Description) Lea Nikupeteri 3 2 6 22 EIA Lithium Chemical Plant: Groundwater Consequence (General Description) Lea Nikupeteri 3 2 6 Asphalting/surfacing of areas Drainage of stormwater Safe storage of chemicals Surface height measurement S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 410 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control quality and water level Quality monitoring is part of the joint monitoring of Patamäki's groundwater 2x/year 23 EIA Lithium chemical plant: Quality and temperature of surface waters - quality is of no greater importance - leading water to the sea can cause heat load Consequence (General Description) Lea Nikupeteri 2 3 6 An additional cooling pool will be built if necessary 24 EIA Lithium chemical plant: Road transport - depends on the transported material, incl. raw materials Consequence (General Description) Lea Nikupeteri 2 1 2 25 EIA Lithium chemical plant: Utilization of natural resources - vegetation and soil materials removed from the area (sand, moraine) Consequence (General Description) Lea Nikupeteri 1 3 3 Biodiversity plan 26 EIA Lithium Chemical Plant: Water emissions from process equipment Consequence (General Description) Lea Nikupeteri 2 2 4 Equipment design and spare parts in stock 27 EIA Päiväneva (2020): Animals - moor frogs in Syväjärvi - otters in Päiväneva - eagles - flying squirrels Consequence (General Description) Lea Nikupeteri 1 2 2 A compensation pond made for moor frogs, which are monitored annually. Otter shelves for pipe bridges made for otters. The eagles are artificially fed in winter and artificial nests are placed in the area. The aim is to reduce the noise effects by positioning the piles of side stones. The alignment of the enrichment sand storage was changed due to the flying squirrels' territory. 28 EIA Päiväneva (2020): Dam breach Consequence (General Description) Lea Nikupeteri 4 1 4 Controlling the amount of water (no free water against the dam) Technique Monitoring the condition of the dams with instrumentation

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 411 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 29 EIA Päiväneva (2020): Surface water effects on Ullavanjoki and Köyhäjoki Consequence (General Description) Lea Nikupeteri 2 3 6 Water treatment in the surface drainage field and settling basins Chemical treatment if necessary Water treatment plant 30 Electrical distribution, Kokkola Cold commissioning might be delayed if the electrical distribution is not ready in time. Consequence (General Description) Juha Kerttula 3 2 6 The issue is related to the insufficient installation space for electric power distribution equipment. The issue is considered to be solved by constructing an additional electric building or enclosure to the kiln area. Should this enclosure not be available or delayed, this would have significant impact on the project schedule and indirectly also costs. //30.10.2023 Technical solution exists, no decisions made yet Update 2024-09-23: Additional electric rooms have been ordered and are about to be constructed at the site. The amount of additional consumers is expected to be limited and can thus be handled by minor modifications of the switchgear, if necessary. 31 Energy price risk Energy price significantly higher than expected leading to lower profitability. 9.3.2022 // Availability of LNG and electricity due to war lower, prices rising. Consequence (General Description) Manu Myllymäki 2 4 8 Sensitivity scenarios prepared for different price levels Energy prices to be updated in the financial model needed: Secure the needed energy supply at fixed or semi- fixed pricing terms (MM) //20220707 We have a policy that will be approved and taken into use once the construction timetable is finished To be added on MG follow-up: LNG availability and alternatives //4.10.2022 Reviewed. LNG terminal contract done with Kokkola Energy. Part of LNG already bought from Barents Naturgas (cost connected to Oil & Gas Index). Second part of LNG cost will be connected to TTF-index. Electricity purchasing strategy is created during 2024. 15.3.2024/Manu Myllymäki 32 Environmenta l aspect: Maximizing the water Normal situation, Mining area, Reduction of raw water intake from outside the mining area. Increasing the internal water recycling rate reduces the need for raw water intake from e.g. Köyhäjoki, which reduces the water footprint of mining operations Consequence (General Description) Jaakko Saukkoriipi 1 4 4 Utilization of water fractions within the mining area at the Päiväneva concentrator S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 412 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control recycling rate (Opportunity) 33 ETP civil construction will be delayed Construction schedule is tight, any delay in the construction process will shift the ramp-up. Schedule still on high level. A delay of couple of months is possible. //15.12.2023 we are prepared to start activities earlier (in February) in case we get the construction permit earlier than expected. 13.5.2023 New delay risks (contractual), expected delay now within 1-2 months. Steel structure schedule flagged by Nordec. Mitigating actions still discussed. Not sure what is the impact on ETP schedule. Changed likelihood to3 - delay of >3 months now less likely. Consequence to be reassessed when the mitigating actions are finished. Consequence (General Description) Heikki Pihlaja 4 1 4 1. Constructability reviews to be started 8.11. 2. Construction schedule risk analysis to be made (Kimmo Heikkinen + CM) after constructability review 3. To ensure adequate resources for procurement (Pertti Kalliosalo) //15.12.2023 we are prepared to start in beginning of February. 4. To complete the LOPA analysis as soon as possible. OK (Antti Mäki) According to Pertti Kalliosalo: 5. Earth work contract done. Work Ongoing. 6. Concrete structures contractor selected. Agreement will be sign wk7. 7. Negotiation for steel structures will be 19.2. --> ETP civil construction in on schedule at the moment 34 ETP: Capex overrun risk: budget approved, detailed engineering not ready Cost estimate based on Metso offer and estimate of other costs. No detailed engineering done, there is a risk that there will be more costs due to process safety requirements etc. Consequence (General Description) Heikki Pihlaja 3 1 3 1. To define procurement packages and their costs. DONE (29.11.23 SH) 2. Rejlers to provide budget by 15.11.2023. First revision DONE (29.11.23 SH) 3. Risk to be reassessed after budget is completed. 4. Revised budget to be sent to discipline managers (Jani P) Done 5. To reassess the situation after risk assessments (LOPA) DONE Vaikuttavuus liiketoiminta lowered from 4-->3 35 ETP: delay in equipment installation Process equipment schedule is tight, fast delivery already accounted for from technology supplier (Metso). Key equipment comes from Metso, some purchased by Keliber. All equipment not yet known, small risk for longer than expected delivery times. EIA is also time critical. Metso issued a statement on Friday, April 5th, stating that equipment deliveries cannot be expedited even with additional funds. Therefore, Metso's equipment deliveries will slightly delay the current installation schedule, and EIA installations will need to be accelerated if we aim to meet the MC Day target on March 14, 2025. Consequence (General Description) Heikki Pihlaja 3 3 9 1. Contract liability: Metso to pay fine if the main process ramp-up is delayed because of delay in ETP 2. Equipment installation: possible to choose from more than one suppliers 15.12.2023 one supplier considered due to small size of the project. //15.12.2023 note: if scope increases, schedule does not, this might have impact on cost. 3. EIA to be included in the schedule (Juha Kerttula) //15.12.2023 scheduled 4. Procurement to be included in the schedule (Juha Luikku) //15.12.2023 Schedule based on procurement packages. 5. Expediting: tight follow-up of Metso equipment schedule (Sami Heikkinen) S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 413 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 36 ETP: Initial data from procurement to Rejlers delayed Keliber procurement needs to provide vendor data for further engineering. Any delays will shift the critical path. Consequence (General Description) Heikki Pihlaja 3 2 6 1. Procurement weekly meetings 2. Vendor document schedule to be added to the ETP main schedule (Jussi Parikka) Procurement has provided vendor data for further engineering (29.11.2023 SH) 37 ETP: technology risk Environmental permit requires defining variation in the water treatment process. New pilot planned for H1 2024. It is possible that there are issues in the capacity or wastewater quality. Re- treatment in unit processes possible. There is yet no agreement with the municipal water treatment plant, also no agreement for disposal for the As-containing residue, waste code needed, amount not yet sure. Consequence (General Description) Heikki Pihlaja 2 4 8 1. pilot to be completed in H1 2024 2. possible to double the equipment capacity for EWT, also surplus capacity in Li precipitation 3. A buffer for ~1 day in the main process in case re- treatment needed in ETP. 4. To complete negotiations and have agreement with the wastewater treatment plant (Antti Mäki) 5. To have waste classification for the material (Jussi Ruokanen --> Metso) 6. Agreement for disposing the As containing waste. (Jussi Ruokanen) 38 ETP: unclear battery limits / tie-ins inside the project There are multiple parties in engineering, construction and technology development. If the interfaces are not properly managed, there might be overlapping work, gaps or incompatibility. There has been several cases of delayed information. Sweco and Rejlers are using separate document management systems. Consequence (General Description) Heikki Pihlaja 3 3 9 1. Biweekly tie-in meetings with Sweco and Afry (power needs, overall BoP engineering) (already in practice) 2. RACI matrix to be implemented in the project team (Sami Heikkinen) 3. To ensure that resources are clearly allocated and adequate (Antti Mäki, Jussi Ruokanen --> Sami Heikkinen) 4. Sweco RACI matrix tie-ins to be cross-checked (EIA, DCS, building automation, locks, access control etc) (Pertti Pekkala, Jussi Parikka) //15.12.2023 completed 5. Pipelines listed and clarified 6. RACI-matrixes to be communicated within teams. (Antti Mäki, Jussi Parikka) 39 Facilities for control room, laboratory and social rooms are not locked, risk that they are not ready in 9/2024 Facilities for control room, laboratory and social rooms are not locked due to budget overrun. Schedule may be delayed if the solution is not found promptly, cold commissioning cannot be started in March 2025 if we do not have a control room in end of Sept 2024. Laboratory needed preferably 6 months before hot commissioning start, as well as control room (automation). Also the safety shelter needs to be in place as per permit, it is possible to negotiate this with municipality but not guaranteed. Also ETP construction permit in process at the same time. Consequence (General Description) Heikki Pihlaja 4 2 8 1. Engineering for new solution to be started, engineer to be found (22.9.2023) 2. Laboratory manager to negotiate temporary space for smaller equipment. 3. Schedule to be made (Kimmo H. and Sami H.) 1.12.2023 Estimated completion end of Oct 2024. This means one month delay in EIA needs, also laboratory ramp-up. 4. Ramp-up plan to be made 5. Construction permit to be applied for in December 2023. 6. Construction offers 31.1.2024 7. Building permit for KR2.1 gained 12.2.2024 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 414 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control No schedule made. Resources inadequate, even scheduling not resourced. 8. Laboratory to start method development in local Applied University laboratory 40 Failing to comply with regulations The project has to comply with several regulations, permits and other requirements. If these are not clearly communicated to the project team and followed-up there is a risk we fail to comply. This could cause schedule delay, re-work, cost and authority notices / stoppage of works Consequence (General Description) Sirpa Olaussen 3 2 6 1. A process for communicating and following up the needs has been defined. Responsibilities need to be clarified. Resourcing needs consideration. 12/2023 decision - HSEQ specialists are responsible for authority communication and communicating tasks to the line organization. 26.1.2023 Responsibilities clarified in RACI. Job descriptions to be updated. 2. Model job descriptions to be created (department manager, supervisor, process engineer, senior management): responsibilities EHSQ, HR, finance /12/2023 done 3. Personal job descriptions to be updated and communicated. 2/2024 in progress 4. Arrange an external compliance audit for Päiväneva site during 2025 Process defined and Kokkola starting meeting held. 11.9.2024 Sanction policies still risk. KIWA audit held on constructions sites, some compliance issues identified. 41 Fire at Päiväneva Large scale fire at Päiväneva Peat production area Consequence (General Description) Sirpa Olaussen 3 3 9 Cooperation with AK-Kraft, fire safety authorities, rescue forces (=Pelastuslaitos) and Ullava VPK, 42 High turnover of personnel in beginning of the production We do not have clear employer promise and / or fail to communicate it and do not get employees / get big turnover in the beginning Consequence (General Description) Asko Saastamoin en 3 2 6 We need a clear personnel strategy and communication plan Supervisor info’s started in 2022. Culture is high now, threat for loosing employees low, but to be followed. 43 Inadequate competence to run operations Our competence management process fails and we do not have adequate knowledge for operations. We are highly dependent on technology suppliers´ knowledge, and do not have control over their resources. Consequence (General Description) Asko Saastamoin en 3 2 6 To ensure competence management process and planning. Internal trainings to be planned, strategic needs to be considered. To update personnel plan and ensure we have adequate strategic competence (production planner). Competence management organisation is updated. Personnel plan updated regularly, development and training plan in place, Training methods under development.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 415 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 44 Increased truck traffic to Päiväneva Truck Trafic from and to Päiväneva insreases and start to disturb neighbours and other road users Consequence (General Description) Hannu Hautala 2 3 6 1. use the full length of day to avoid traffic peeks 2. maximize the use own aggregate => cooperation with ELY 3. share fact based information among the local community: Municipality, ELY, neighbours, local/regional voluntary organizations 45 Key people leaving company We fail to reach our goals / schedule due to key person leaving the company Consequence (General Description) Asko Saastamoin en 2 2 4 /2.12. Job descriptions for MG updated. / Succession planning in place - critical positions recognized. Good leadership and engagement. Remuneration based on Labour market and Keliber special situation. Consultants can be used partly. Common practices followed. Notice periods extended. RACI updated. 46 Kokkola refinery: Updated environment al permit needed for drainage pipeline for cooling and storm waters to sea, schedule risk Original and permitted drainage option is not available due to KIP changes. Kokkola Industrial Park has initiated planning process for new drainage pipeline. This change causes need for environmental permit changes. Consequence (General Description) Markus Kivimäki 3 1 3 1. Permit application has been submitted on 26 June 2024. Expected processing time minimum 6 months. In case of an appeal, an enforcement order needed to keep the schedule. 2. Discussions are carried out with KIP relating to using the currently permitted drainage option until the permit amendment is obtained. This has been confirmed. 47 Lithium hydroxide market price lower than expected We fail to reach our revenue targets due to low lithium hydroxide market price. If the price is lower than in the model, there is a risk of project delay. Consequence (General Description) Riku Sauso 4 3 12 The aim is to conduct the offtakes with market prices which means that we are open for market risk. As for now, it seems that the market prices will be sufficiently high (vs our opex level) given the supply/demand balance. Fixed priced offtakes and sales would also be a risk. 48 Lithium hydroxide market vanishes. Business case no longer valid Lithium hydroxide market vanishes. Business case no longer valid Consequence (General Description) Hannu Hautala 5 1 5 1. Battery market will demand mostly LiOH and Keliber project was changed in 2019 to produce LiOH 2. Manage cost efficiency before/during the operative period starting in 2026 49 Lithium loss in new wastewater treatment at Lithium phosphate cannot be effectively recovered in the main process, yield according to pilot about 30 %. This is surplus to the financial model Consequence (General Description) Sami Heikkinen 1 4 4 Tests to re-circulate waste fraction are on going to recover lithium. Based on published information lithium can be recovered Total achievable recovery is still unknown. Continuous proving piloting tests. Can precipitate be S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 416 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control the Chemical Plant circulated to the pressure reactor feed to recover lithium? Option to precipitate major portion of Li with as phosphate, from where Li can be recovered and remaining smaller portion is precipitated as aluminate. First Test results shows Li recovery from lithium aluminate is good in main process. It is far better than Li phosphate. Selected process will be Li phosphate. Li Aluminate Filtration was not working and it carry too much impurities back to main process. To assess possibilities to sell the LiPO4 as by-product (TB) 50 LNG solution not locked, risk for schedule delay due to availability and permitting 19.10.2023 LNG sourcing is locked! Agreement with Kokkolan Energia done! LNG sourcing is not locked. Own terminal is no longer possible at the planned area (due to changes in concentrator storage). Own terminal would require higher tier permitting which takes longer time. LNG terminal as service is prone to appeals (public sourcing), causes schedule risk that is not in our hands If the gas is not available after the mechanical completion, the production cannot be started. Consequence (General Description) Heikki Pihlaja 4 1 4 1. To negotiate (weekly) with Kokkolan Energia. 2. Negotiating together with other users (Yara and Hycamite) 19.10.2023 LNG sourcing is locked! Agreement with Kokkolan Energia done! Also Yara and Hycamite has signed and committed to same agreement. That lowers the risk. 51 Management of extractive waste Pools and waste piles containing mining waste must have sufficient capacity and take into account future needs in time Consequence (General Description) Lea Nikupeteri 2 4 8 We monitor the occupancy rate and anticipate future space needs in time so that the expansion can be approved 52 Mining compensatio ns higher than expected Mining compensation are higher than expected leading to lower profitability Consequence (General Description) Riku Sauso 2 2 4 The royalties and grants to be given are mostly fixed before the actual production starts. Hence there is very little variation expected with the outcome. Needed; The aim is to purchase the Rapasaari areas so that there would be no mining compensations to be paid. 53 Operators leaving company due to delayed commissionin g of Kokkola We have recruited several operators in Kokkola which is now delayed and we cannot offer shift work. This means their earnings are smaller than expected. There is a risk that we lose some experienced operators, which endangers the commissioning of the plants. Our reputation might be damaged which impacts future recruitment. Long-time impact is also expected on work moral. Consequence (General Description), Corporate Image Asko Saastamoin en 3 3 9 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 417 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 54 Other large projects / companies at the same area attract employees Our competitive position is inadequate to attract employees Consequence (General Description) Asko Saastamoin en 2 1 2 needed: A personnel strategy to be created and our attractiveness ensured. Co-operation with education providers at the area, follow-up of the situation //20221214 More projects coming at the area. We still get positive feedback. Critical phase when recruiting employees. 20231017//Keliber extremely wanted; other projects proceeding slowly. 20240108// labour market stabilising; Keliber in good position. Status the same 27082024. 55 Permit limit for water intake / outlet in Köyhäjoki is too strict for operations If the concentrator uptime is 8000h/a the limits are too strict. The planned water intake - outlet in Köyhäjoki is too small for intake and there is also risk we cannot discharge as much wastewater the production would need. We are planning a new intake / outlet in River Perhonjoki (preliminary study made by Afry) but in order to not disturb the ongoing permit process we will start discussions with the authorities after the Rapasaari / Päiväneva permit is granted. It is yet unclear whether a new permit is needed. EIA is not necessarily needed, but the decision will be on the ELY-centre. The new intake/outlet is needed in 10/26. Building takes 10 months and should be started in 1/2026. Risk to be updated when new LOM is published. Permitting should be started ASAP 14.11. Permitting cannot be ready before end of 2026, with appeals 2027. This means that the pipeline is commissioned in 2028, Consequence (General Description) Ville Vähäkanga s 4 2 8 To prepare a schedule and detailed plan (VV) 31.5.2023 Done in January To investigate the landowner situation (MM) Done Discharge modelling (KW) 31.5.2023 no information Updating permits (KW) 31.5.2023 no proceeding To decide a forum to follow-up the issue and prepare a report to MG (VV) 9.5. Not possible to start EIA until the permit is under appeal process. 31.5. KW stated that no EIA needed. Problems at operative years 6 onwards depending on LOM 4.9.2023 Discussed in permitting meeting in June. Budgeted at title level. 23.10.2023 Permitting to be started 4.12.2023 decision made to apply for permit (22.11.2023) 14..11.2024 EIA being prepared. To be completed 12/2025. Permit ready 12/2026. Pipeline commissioned 2028 56 Personnel risk - no background check As for now we do not require safety reports from recruits. There is a risk we recruit someone who can damage the project Consequence (General Description) Markus Kivimäki 3 1 3 To define which positions require safety reports. To arrange planning meeting with the authorities To apply to be SUPOs client Discussion with HR/ASa to ensure that HR takes this into consideration in the recruitment processes. OK. 57 Project implementati on in MO not effective Project implementation not effective enough. Diversified MO Operation model. No ownership or overall management. Layout design has been earlier low quality. This may lead to further delays in engineering 4.7.2023: Risk realized in wastewater treatment; main process not compatible with the proposed process. Consequence (General Description) Pertti Pekkala 3 3 9 1. MO operation model audit related to the industrial water treatment. Special support from other business lines. 2. To ensure that deviations are written and reported to M:O (IX area, wastewater treatment) //4.7.2023 Variation notices done? (PP). 4.12.23 - associated Variation Notices has been done. Observed deviations have been also reported. 4.12.2023 Metso has nominated own project manager for S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 418 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control Effluent Treatment Plant sub-project. Also Engineering Management function is enforced. 22.01.2024 Keliber has also nominated project manager for ETP portion enabling more efficient management of MO sub-project. 20.2.2024 Keliber Project Management has contracted MO management and improvement areas have been outlined. 28.03.2024 Equipment delivery schedule has been reviewed and speeding up actions have been asked from Metso. 20.5.2024 Almost all AFC status GA drawings have been issued by Metso enabling progress of Rejlers detail engineering 4.11.2024 All ETP process equipment have been delivered and installation works are progressing. Piloting has been completed and study engineering has been planned to reduce effluent phosphor levels meaning that still some process technical challenges are present 58 Project schedule delay or cost escalation due to unclear responsibilitie s and decision- making process Unclear responsibilities or decision-making process within the company and with partners causes delay or cost escalation. Instructions are not always known or understood. Also there is lot of new resources coming in, and the resourcing is delayed. Consequence (General Description) Heikki Pekkarinen 2 3 6 1. Change the project execution model from EPCM to owners team driven E + P + CM model. //done Q4/2022 2. Update responsibility matrix //done Q1/2023 3. Strengthen owners team and contract CM-services //done Q1/2023 4. Update Project specification //done Q1/2023. 6. Write, approve and communicate FRAMEWORK/PEP // done 3/2023 5. Ensure participation at change management process 6. Update & communicate RACI and job descriptions 7. Call a meeting for defining responsibilities (SO, MK ,KW, HP, PP, LL, PK) 8. Implement detailed JTC-meeting and reporting practice 4/2023 9. To be decided: a rolling practice for handling deviations and processes / instructions - how to manage the amount of newcomers? Part of weekly meeting agenda? (HP&LL) 10. To update job descriptions, instructions and induction materials. 11. Appointed new mining manager and VP Lithium Refinery (13.5.2024) 12. Using Primavera as schedule control tool.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 419 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 13. To review onboarding and change management process 59 Päiväneva schedule: unclear authority schedules and requirements There are several unclarities in both the scheduling of EPOPELY statements and requirements. Pond area design has been in the EPOELY for a year for review Surprises coming, for example EPOELY requiring further studies for a structure already approved by KAIELY. Planning in terms of procuring design and other material to be approved by ELY is not on a sufficient level and requires improvement. Consequence (General Description) Jaakko Saukkoriipi 3 3 9 1. More high level discussion with the authorities (requirements, processes, timetables) 2. Intensifying cooperation between project and environment to ensure that project delivers the plans early enough and in a form acceptable to authorities. 3. Project to improve the planning relating to timing of designs and authority approvals. 60 Päiväneva: electricity needs rise and the planned power inlet is not adequate The power needs can increase in planning and without adequate control it is possible that the inlet is not adequate. According to planning the power need is 10-12 MW, with heat plant 14-16 MW. The current overhead line is for 16 MW. The permitting of a new overhead line could take 2-3 years. The planned line is ours and the landowners are not required to give land to us (compared to network companies) Note: Contract for cabling in the beginning of 2024. Note: For Rapasaari underground mode power will be needed, then additional power is needed. This will be actual in operative years 5-6 so there is possibility plan mitigating actions (Fingrid project, new overhead line) Consequence (General Description) Juha Kerttula 4 1 4 1. Engineering of the plants is finished. Open issues mainly in heat tracing (pipeline lengths, chemicals that need to be kept in +20 C) 2. Heat plant will be using for both electricity and pellets 3. Electrical engineering of the plant is practically completed and only a small amount of additional consumers is expected, if any. 4. Some additional power is possible to achieve with thicker cable. -> A 20 kV 3x1000 mm2 main supply cable is being constructed instead of 3x800 mm2 initially designed allowing to supply 20 MVA. 5. Local grid operator is planning to enhance local power distribution grid increasing the output of the present overhead line up to 5 MVA vs. present 1 MVA. This connection could be used in parallel with the main supply cable to feed the electric boiler plant. 61 Rapasaari ore cannot be treated in the concentrator due to high Arsenic content Rapasaari ore has higher as content than Syväjärvi. If the arsenic cannot be removed at the concentrator the product will not be battery grade, also the waste fractions will have higher As- content. A pilot was finished in 2022 to test sulphide flotation but the results were not conclusive for investment. Other possibilities are gravity separation and more efficient sorting with XRD. Rapasaari is planned to be started in 2026 with 3 mth campaigns Consequence (General Description) Ville Vähäkanga s 4 2 8 1. To consider whether changing the LOM is possible (waste rock in Syväjärvi might restrict) (PG) 2. To update the environmental permit for Syväjärvi to allow more mining (KW) //4.1.2023 in progress 3. To investigate the possibility for another pilot in beg of 2024 (would require starting infrastructure works in Rapasaari) (VV, PG) 4. To start laboratory scale pilot program in 2023 4.1.2023 in progress. 28.6.2023 pilot completed 12,5 WMT material used, waiting for results./4.9.2023 report being reviewed 5. To evaluate the risk for not getting environmental permit for Päiväneva because of lacking information (VV, S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 420 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control annually. If the start is postponed, we have more time for engineering (change of LOM). KW) //4.1.2023 permit acquired 6. Sorting tests at Tomra Q1 VV / KL : 9.5. tests successful. No process changes expected. Waste rock sorting to be studied. It is expected that sorting combined with prefloat process we will be able to process the Rapasaari ore 28.6. XRD sorting planned /4.9.2023 with XRD it is possible to separate As-bearing fractions. 62 Refinery capex overrun Capex budget is exceeded, 7.9 Me of contingency left. Additional investments added costs. Unit based contracts cause uncertainty. Quality of the forecast process not adequate. Change of scope process ongoing but no guarantee of results. Consequence (General Description) Heikki Pihlaja 2 4 8 1. Implement strict cost control 2. Variation notice form up dated, a comment for whether the notice is considered in the forecast required. 3. Change of scope evaluation 63 Schedule and cost risk due to Ukrainian war We need to evaluate effects on schedule and costs. Materials costs rising Schedule risk is medium Consequence (General Description) Heikki Pekkarinen 2 2 4 Risk can be partly mitigate in purchasing plan. We need to evaluate the impacts and control mechanisms available 64 Schedule delay due to land acquisition issues in Päiväneva We fail to reach project schedule target due to delays in land acquisition in Päiväneva Consequence (General Description) Manu Myllymäki 3 1 3 Landowners have been contacted. needed: The roads are going to be wider, new contracts needed. 5.9.2022: Road contracts in control. Rapasaari Mining area land agreement with private land owners in control, with Alholmens Kraft some cost challenges because of changed energy situation in Europe. 21.11.2022 More peat measurements are done and reported from Rapasaari area. One target is to purchase larger land area from AK than earlier estimated. Negotiations of larger area is about to start and Markus is leading AK negotiations. Private land owners are in control. 9.3.2023 Negotiations with Alholmens Kraft is progressing. 31.5.2023 The agreement can be signed in Friday 2.6. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 421 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 65 Soda leaching fails to perform as expected causing lower profitability Chosen soda leaching technology fails to perform at expected level - quality lower or costs higher. The process is not yet in industrial use. Consequence (General Description) Sami Heikkinen 4 2 8 Good piloting program in progress and SRK involved (MM 31.8.2019). Three step pilot program done successfully -> risk lowered remarkably (MM 26.5.2020) /2.12.2021/ Rapasaari pilot to be completed by 6/2021 Process guarantees from technology provider, high interest from TPs side to ensure success. Pilots to be planned for Rapasaari and possible even for commercial concentrates /7.10.2022 Process guarantees achieved. Some risk in recovery. 2.6.2023 Performance criteria included in agreement. 66 Supplier risk: we choose a supplier that falls under sanction policy The process to evaluate suppliers concerning EU and USA sanctions requires further improvement. We might choose a partner that either already is or is at risk to fall under sanctions or get material from a country that is under sanctions policy. 14.11.2024 all partners ensured that they are not used sanctioned steel. Processes in place. Consequence (General Description) Markus Kivimäki 4 1 4 Currently all the vendors selected for the investment project are checked against the sanctions using our E(PCM) contractor and all the potential raw material suppliers are reviewed by SSW Protection Services. Actions to create a internally managed & controlled process for sanction checks / KYC and then implementing it is ongoing. Service provider for vendor DD's (sanction checks) selected, implementation to start in March. //20.2.2024 Implementation of internal solution has been completed and procurement is carrying out sanctions checks before contracts are entered into. 26.6.2024 67 Syväjärvi environment al permit update is delayed Syväjärvi environmental permit update is not obtained when needed - The storage area for sulphide-containing mica schist is not permitted and will be needed in H2/2026. Consequence (General Description) Markus Kivimäki 2 2 4 1. The environmental permit application is submitted in a manner (scope) which should allow AVI (permitting authority) to issue the decision in a timely manner (e.g. Keliber is not applying for increased mining volumes) 2. To start discussions with Kokkola Port to deliver the sulphide-bearing material for Port extension 3. Co-operate with AVI and ELY during the permitting process and consider possibility to (partially) construct the area before the permit decision is received 68 Syväjärvi waste rock material may not be adequate for construction for the The authority has set a limit value of 19 ppm of As and 0,1 % of S for the material to be used as construction material in the Rapasaari area. Ely has expanded this requirement also for the Syväjärvi waste rock and required a report of the content of As and S in the rock being used. In case external material is needed there will be a cost increase and schedule delay. According to our surveys we do have enough material that fills the Consequence (General Description) Heikki Pekkarinen 2 5 10 1. Obtain ELY approval for the sampling method and frequency so that we can use the Syväjärvi construction stone - OK 2. Review the possibility to purchase external material to be able to keep the construction schedule 11.4.2024 in use - OK 3. Review possibility to apply for a change in permit conditions relating to sulpher & arsenic - Will be done as S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 422 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control concentrator area requirements. ELY has been provided with Keliber's response showing that the material from Syväjärvi fulfills the requirements. Unless ELY disputes Keliber's view, the matter can be considered closed. //20.2.2024 14.3.2024: The material quarried currently has too high sulphur content, quarrying stopped. 21.5.2024 ELY approved the proposal on testing and use of construction stone. 30.8.2024 Current view is that the need of external stone for Päiväneva construction is ~6 M€. There are further needs for Malmitie and other construction for mines. part of the planned permit amendment applications. 4. Project / geology to set up a working group to manage the construction stone matter. 5. Project to ensure that the instructions for rock quality approval are communicated (VV) 69 The analcime produced when using external spodumene cannot be placed in the harbour We are starting the production with purchased spodumene. The environmental permit is applied for analcime produced from own ore. We do not have information for the spodumene quality. The permit conditions need to be checked: is there a difference where the analcime comes from. Consequence (General Description) Markus Kivimäki 4 1 4 1. Check the permit conditions 2. Get the information of the expected quality of the purchased concentrate 3. We are getting a quotation from a service provider in case the analcime sand does not fit the permit limitations. 70 The concentrator / ponds are built on significant orebody A risk that sterilization drilling has been inadequate and we find significant orebody under concentrator / ponds Consequence (General Description) Pentti Grönholm 4 2 8 Sterilization drillings have been made, not deep ones, but considered adequate at this stage. //6.10.2022 Some additional drilling was carried out at Päiväsaari area during June 2022 with no indications for significant orebody. Drilling to be continued according to plan. // 25.9.2023 The surface levels of concentrator pond areas have been checked by sterilisation drilling in 2020-2022. No indications of significant ore bodies in the areas at shallow levels. // 27.12.2023 No changes in the risk. 71 The concentrator plant has serious ramp- up issues, which the project cannot cope with According to plan all sites are to be ramped-up at same time, during Q3/2024. Our processes and resources are not adequate to cope if all sites have serious ramp-up issues. //4.1.2023 There is yet no schedule for Päiväneva //31.5.2023 No new information. //2.11.2023 Päiväneva AFE approved. //10.1.24 Päiväneva ramp-up schedule 1/2026 --> Consequence (General Description) Ville Vähäkanga s 3 2 6 1. Kokkola will be ramped up earlier with external concentrate 2. Testing and commissioning plan ready by 11/2024 (VV) 3. Personnel plan to ensure that there is enough operators. Training program under development (with KPEDU) 4. When the solution and schedule for Kokkola ETP and Päiväneva start are clarified, the ramp-up and concentrate need will be reassessed.

S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 423 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 72 The refinery plant has serious ramp- up issues, which the project cannot cope wit According to plan all sites are to be ramped-up at same time, during Q3/2024. Our processes and resources are not adequate to cope if all sites have serious ramp-up issues. //4.1.2023 There is yet no schedule for Päiväneva //31.5.2023 No new information. //2.11.2023 Päiväneva AFE approved. Consequence (General Description) Heikki Pihlaja 3 2 6 1. Kokkola will be ramped up earlier with external concentrate 2. Testing and commissioning plan (resources and sequence by Q1/2024, structure (index) ready by 12/2023 (SH) 3. Personnel plan to ensure that there is enough operators. Training program under development (with KPEDU) 4. When the solution and schedule for Kokkola ETP and Päiväneva start are clarified, the ramp-up and concentrate need will be reassessed. 73 The tailings storage facilities are too small if LOM extended due to increased ore reserves The tailings ponds will be too small if LOM is extended due to new ore findings Consequence (General Description) Ville Vähäkanga s 1 4 4 It is possible to place the tailings in Syväjärvi pit after operations. It is also possible to fill the Rapasaari UG with the tailings. VV to present the situation and plans to MG // 30.8.2022 This will be acute in 7 yrs time, will be reviewed regularly 31.5.2023 Risk reassessed 74 Uncontrolled culture No common understanding of company culture (values, ways of working, following what has been agreed, respecting other employees and stakeholders). This could lead actions that are contradict to what we say and weaken trust to us. Several new employees arriving during the year 2024, so this needs increased attention. Consequence (General Description) Asko Saastamoin en 3 2 6 Develop the Keliber way. Training has been started, and importance of culture taken from the induction. With present action and development the culture is supported. Culture training/ discussion in employee meetings underlining every ones responsibility in building a good work place. Supervisor culture training y2025. Supportive action i.e. 1:1 meetings to be in place. 75 Unfavourable exchange rate USD/EUR Unfavourable EUR/USD course causes lower profitability Consequence (General Description) Riku Sauso 4 3 12 Sensitivity scenarios prepared to plan for different FX rates. Current plan and the financial model based on a relatively conservative assumption. Sales agreement hedging as the main mitigating action 76 Unsuitable person recruited We fail to reach our goals / schedule due to recruiting unsuitable person in key position Consequence (General Description) Asko Saastamoin en 2 2 4 Good resource plans, all recruitments via MG, according to plans and job descriptions. Recruiting controlled and according to process, resources updated. 77 Wastewater limit compliance in the Chemical Plant We are not able to comply with the environmental permit regulations because the limits are stricter than those we applied for (250 mg/l vs. 200 mg/l). Reached levels were 160-250 mg/l. This will lead to exceeding the permit limits. We are preparing engineering for aluminate precipitation that will Consequence (General Description) Sami Heikkinen 3 1 3 Test work and conceptual study services for LiAlO2 recovery (SH, completed) Continued discussion with the authorities (KW) 1.12.2023 Technology changed. S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 424 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control reach the Li level required. 4.1.2023 The limit value can be reached according to the tests. 1.12.2023 Phosphate precipitation with evaporation chosen as technology. The limit values will be met. 78 Wastewater treatment in Päiväneva cannot reach the permit limits for sulphate Sulphuric acid is used for pH regulation, main source of sulphate is from Rapasaari waste rock area. There is no removal mechanism for sulphate and the permit limit is 150 mg/l. Planning a process step for sulphate requires update for environmental permit and may affect CapEx and schedule. To achieve <100 mg/l a RO process should be used. Consequence (General Description) Ville Vähäkanga s 4 2 8 1. To decide whether apply for change for permit //31.5. Permit appealed for 2. To plan new inlet / outlet in Perhonjoki, when the impact is lower even on higher concentrations (ongoing) 4.12. decision of permit application done 22.11., permit will be applied for. 3. Budgetary offer from Aquaflow / veolia for sulphate removal process. Sustaining CapEx 3M€ addition. A faster route could be to rent the RO equipment (5000 € / week) /4.9.2023 SRK reviewed the scoping study, no major flaws reported 4. Closure of the waste rock area phased (Afry modelling from 5/2023) 5. New production plan in progress (will be finished before the end of the year 2023) 79 Wastewater treatment planned in Päiväneva cannot reach permit limits for nitrogen The technology planned for nitrogen reduction is new in mining industry in Finland and the limit value of 7,5 mg/l might be challenging, Consequence (General Description) Ville Vähäkanga s 4 3 12 1. To plan a new outlet / inlet in Perhonjoki (impacts smaller on a large stream), 31.5. a reservation for new outlet is in the sustaining capex. No ongoing actions with permitting. 4.12.2023 permit will be applied for. 2. Arrange a visit to Kittilä and Kemi mines to discuss nitrogen removal in their wastewaters. 31.5. completed, it seems that 7,5 mg/l is a very challenging goal, also cost could be higher than expected. 3. Updated modelling (phased closure of the waste rock area) shows that permit condition can be met (requires reaching the 7,5 mg/l in outlet), sustaining capex corrected for capacity (doubled, no engineering yet) //31.5. //4.9.2023 SRK reviewed 4. Ongoing discussions with explosives providers to find an explosive without nitrogen. 5. Permit appealed by Keliber //4.9.2023 //13.5.2024 Update probability after Veolia study (LL) S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 425 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 6. Basic engineering of the Rapasaari water management should be started 2025 80 Water management, water storage capacity The volume of the water basins must be constantly monitored in order to keep the water balance under control Consequence (General Description) Lea Nikupeteri 3 2 6 Water balance online monitoring 81 We cannot get the agreement with Kokkolan Teollisuusvesi in time and hot commissionin g is delayed The capacity of the Kokkolan Teollisuusvesi is limited and the investment for added capacity is delayed. We need to have a contract to ensure the availability of water for the process 14.11.2024 We have two possibilities: contract with Teollisuusvesi or own unit. Delivery time for own unit is 12 months Consequence (General Description) Heikki Pihlaja 3 3 9 1. To developed a mitigation plan with schedule and finance. It is possible to have a container solution for water purification at ~1 M€/a rental cost. 2. To get quotes from system providers, discussion with one. (Nina Hjulfors, Heikki Pihlaja, Erkki Niska) 82 We do not have a deposit for analcime sand when the production begins. The cost for using external service is more expensive. Own storage area will be not ready when production starts. Harbour can take the material but on higher cost. Negotiations also going with Vestia. Consequence (General Description) Heikki Pihlaja 2 5 10 1. The whole area does not need to be ready at the beginning 2. Ongoing negotiations with the harbour 3. To be assessed how much of the sand is produced during the first year 4. To be assessed whether we need a temporary storage for the ramp-up period (if the sand cannot be placed into the harbour 5. Sampling plan for the sand for the ramp-up period 6. Inquiring ELY whether the sand from ramp-up time can be placed in the harbour after analysis (and to send the sampling and analysis plan for approval) Kari Wiikinkoski: 7. Goal is to apply environmental permit EP at the end of February 2024. EIA Report will be given paraller and is identical to EP 8. To find another waste disposal area (Kruunupyy Pirilö) where we can dispose the sand with a cost. //11.4.2024 Permit application only possible to be supplied at autumn. SO EIA Text is now in Sharepoint and under review. https://keliberoy.sharepoint.com/sites/Environment/Per mits/Forms/AllItems.aspx?id=%2Fsites%2FEnvironment%2 FPermits%2FHoikkaneva%2FHoikkaneva%20AFRY&p=true &ga=1&LOF=1 S-K 1300 Technical Report Summary for Sibanye-Stillwater on Keliber Lithium Project, Finland 426 Risk Rank ing Risk Name Risk Context Risk Category Risk Owner Inherent Risk Assessment Internal Controls Severity/ Impact Probability/Li kelihood (IRR) Primary Control 83 We fail to comply with the regulations regarding non EU/ETA citizens working on construction sites It is almost certain that there will be non EU/ETA citizens working on the sites. Complying with the regulations concerning contractor's obligations requires competence. There are several cases where worker´s have been underpaid, victimized or been subject to forced labour in Finnish construction sites. In addition to legal consequences there is a reputation risk, since these cases are usually connected to the client. Consequence (General Description) Sirpa Olaussen 2 2 4 1. We require the Vastuu Group report for each subcontractor Good practices: to limit subcontracting, to have a 2nd party audit on site, to have an anonymous channel for whistleblowing, to include information about employees' rights in orientation 6.1.2024 The process for approving sub-contractors has been updated. In induction material there is now information for employee's rights, also information material is to be spread on site. 28.3.2024 An external audit to be arranged during April (safety, compliance to the requirements for foreign workers). Done (Kiwa Inspecta), 6 NCs, corrective actions set. 11.9.2024 External audit arranged, some non- conformances found. Päiväneva to be audited same manned in 2025 84 We loose the local acceptance for the project due to either bad performance or inadequate communicati on Ownership is concentrated to SSW and FMG and SSW communication limitations impact our local communication and information sharing. Earlier local share holders as well known individuals were strong supporters of the project. Increased transportation and traffic will cause incremental risk in spring 2024. Consequence (General Description) Hannu Hautala 3 3 9 1. We intensify our program for public meetings in Kaustinen 2. We intensify our program for public meetings in Kokkola 3. We intensify construction related info-sharing. 4. We develop relationship with local media 5. Allocate a budget for local sponsoring activities e.g. Sibanye Foundation donations 6. Emmes has been subject to complaint, project started at Syväjärvi - Päiväneva. 7. Communications about exploration activities and summer workers 8. Security planning updated for Päiväneva area 9. Conduct a meeting with hunting club 85 We loose the time window for being first Lithium hydroxide producer in Europe We fail to reach the project schedule targets and competitors corner the market securing better contracts Consequence (General Description) Hannu Hautala 1 2 2 1. focus on project execution to start the ramp up in 2025. 2. use external concentrate, sufficient number of suppliers ~3 has been identified and are ready to enter into negotiations 3. monitor the market and understand the market development