High REE Recoveries at Reduced Lixiviant Levels

THIS ANNOUNCEMENT CONTAINS INSIDE INFORMATION FOR THE PURPOSES OF ARTICLE 7 OF REGULATION 2014/596/EU WHICH IS PART OF DOMESTIC UK LAW PURSUANT TO THE MARKET ABUSE (AMENDMENT) (EU EXIT) REGULATIONS (SI 2019/310) ("UK MAR"). UPON THE PUBLICATION OF THIS ANNOUNCEMENT, THIS INSIDE INFORMATION (AS DEFINED IN UK MAR) IS NOW CONSIDERED TO BE IN THE PUBLIC DOMAIN.

NOT FOR RELEASE, PUBLICATION OR DISTRIBUTION, IN WHOLE OR IN PART, DIRECTLY OR INDIRECTLY IN OR INTO THE UNITED STATES, AUSTRALIA, CANADA, JAPAN, THE REPUBLIC OF SOUTH AFRICA OR ANY OTHER JURISDICTION WHERE TO DO SO WOULD CONSTITUTE A VIOLATION OF THE RELEVANT LAWS OF SUCH JURISDICTION.

16 September 2026

Cobra Resources plc

("Cobra" or the "Company")

High REE Recoveries at Reduced Ammonium Sulphate Concentrations

ANSTO test work demonstrates operating flexibility across lixiviant concentration and pH, supporting engineering design and scoping study optionality

Cobra (LSE: COBR), a South Australian mineral exploration and development company, is pleased to report results from 24-hour desorption tests completed by the Australian Nuclear Science and Technology Organisation ("ANSTO") on bulk composite samples from the Boland and Head prospects ("Wudinna Rare Earths Project"). The programme evaluated rare earth element ("REE") recovery across a range of ammonium sulphate ("AMSUL") concentrations and pH conditions using site-sourced saline groundwater.

Results demonstrate that Wudinna recoveries can be managed across a range of lixiviant strengths. Lower reagent concentrations may reduce cost, supply exposure and the salinity load introduced to the aquifer. The results define a range of potential operating conditions for further engineering trade-off studies and will inform the Wudinna Scoping Study assessment of recovery, reagent consumption, solution chemistry and project economics.

Highlights

·    ANSTO tested bulk composites prepared from samples across 26 intervals from 24 drillholes.

·    Recoveries from the Boland composite sample:

o  0.3 M AMSUL at pH 3 recovered 70% neodymium and praseodymium ("NdPr") and 67% dysprosium and terbium ("DyTb"), compared with 77% NdPr and 74% DyTb at 0.5 M AMSUL.

o  Acidified site groundwater with no added AMSUL recovered 51% NdPr, 46% DyTb and 50% heavy rare earth elements ("HREE").

·    Recoveries from the Head composite sample:

o  0.3 M AMSUL at pH 3 recovered 36% NdPr and 52 DyTb, compared with 40% NdPr and 57% DyTb at 0.5 M AMSUL.

o  Acidified site groundwater with no added AMSUL: 23% NdPr, 33% DyTb and 36% HREE recovery.

·    Reduced uranium mobilisation at lower molarities. At pH 3, reducing AMSUL from 0.5 M to 0.0 M reduced uranium recovery from 29% to 5% on the Boland composite.

·    The lower-reagent operating cases will be assessed for potential benefits including:

o  Lower AMSUL consumption and reagent cost.

o  Reduced exposure to AMSUL supply and pricing.

o  Lower additions of ammonium and sulphate to the confined aquifer further reducing environmental impacts.

·    Recovery, reagent consumption, solution chemistry and overall economics will be assessed together in the Scoping Study to define preferred commercial operating parameters.

Rupert Verco, Managing Director of Cobra, commented:

"Results demonstrate that Wudinna recoveries can be managed across a range of lixiviant strengths. At Boland, magnet rare earth recoveries were only reduced by 9% with a 40% reduction in ammonium sulphate reagent when compared to the industry standard 0.5M, while acidified site groundwater with no added ammonium sulphate recovered around half of the key magnet and heavy rare earths.

This operating flexibility matters because recovery cannot be assessed in isolation. Lower reagent concentrations may reduce cost, supply exposure and the salinity load introduced to the aquifer. The Scoping Study will test those potential benefits against the associated change in recovery.

The scale of the composites provides confidence that the response is representative of the mineralised material tested, and the consistent performance at both Boland and Head supports the next phase of optimisation.

We expect to announce the maiden Mineral Resource Estimate shortly, followed by further updates from the Scoping Study."

Note: Reported recoveries are calculated from analysed liquor concentrations adjusted for dilution and were compared with recoveries calculated from residue assays. Results from test CO-AcL-01, conducted on the Boland composite using 0.5 M AMSUL at a target pH of 4, have been excluded from reporting because reconciliation between the liquor and residue assays was outside acceptable limits.

About the Wudinna Rare Earths Project

The Wudinna Project comprises the Boland and Head prospects. Cobra has identified ionic rare earth mineralisation containing magnet and heavy rare earths within permeable palaeochannel sands confined by clay-rich units at Wudinna.

Bench-scale testing indicates that the mineralisation is amenable to ISR recovery techniques. ISR has been used for decades in geologically similar environments to recover uranium in South Australia, which has an established ISR regulatory framework.

Cobra's ongoing test work is evaluating the technical and economic potential for ISR recovery of critical rare earth elements at Wudinna. Engineering design, further test work and the Scoping Study will determine the potential operating configuration, recovery assumptions and development economics.

Follow this link to watch a short video of CEO Rupert Verco explaining the results released in this announcement: https://investors.cobraplc.com/link/PGR2or

Further information relating to the Wudinna Rare Earth Project is presented in the appendices.

Enquiries:

The person who arranged for the release of this announcement was Rupert Verco, Managing Director of the Company.

Information in this announcement relates to exploration results that have been reported in the following announcements:

·    Wudinna Rare Earth Purification Update: ANSTO purification test work results providecredible pathway to an upgraded rare earth product, dated 13^th September 2026

·    Exploration update: "Resource drilling completion", dated 5 May 2026

·    Metallurgical update: "Boland Delivers Industry-Leading Heavy Rare Earth Product" dated 2 March 2026

·    Metallurgical update: "Test work upgrades Boland liquor through 100% cerium removal resulting in a large increase in product value", dated 9th December 2025

·    Exploration update: "Successful first pass suppression of cerium to maximise valuable dysprosium and terbium", dated 20 November 2025

·    Exploration update: "Exceptional Results - Infield Permeability Study", dated 17 November 2025

·    Exploration update: "Metallurgical Optimisation Upside", dated 20 October 2025

·    Exploration update: "Exceptional Metallurgical Results from ISR Column", dated 14 October 2025

·    Exploration update: "Met Study Supports Even Lower-Cost Recoveries", dated 11^th September 2025

·    Exploration update: "Low-Cost Recoveries from Optimised Testing", dated 11^th August 2025

·    Exploration update: "Rare Earth ISR System beyond Boland", dated 4^th August 2025

·    Exploration update: "Favourable Boland Metallurgical Results", dated 21st July 2025

·    Exploration update: "Boland Project Update", dated 26^th June 2025

·    Wudinna Project Update: "Boland Aircore Drill Results", dated 25^th February 2025

·    Wudinna Project Update: "Further Positive Metallurgy Results from Boland Project", dated 16 December 2024

·    Wudinna Project Update: "2^nd Bench Scale ISR Study & £1.7M Placing", dated 26 November 2024

·    Wudinna Project Update: "ISR Bench Scale Study Completion", dated 4 November 2024

·    Wudinna Project Update: "ISR bench scale study delivers exceptional results", dated 1 October 2024

·    Wudinna Project Update: "ISR bench scale update - Exceptionally high recoveries with low impurities and low acid consumption; on path to disrupt global supply
of heavy rare earths", dated 28 August 2024

·    Wudinna Project Update: "ISR bench scale update -Further metallurgical success at world leading ISR rare earth project", dated 11 July 2024

·    Wudinna Project Update: "ISR bench scale update - Exceptional head grades revealed", dated 18 June 2024

·    Wudinna Project Update: "Re-Assay Results Confirm High Grades Over Exceptional Scale at Boland", dated 26 April 2024

Competent Persons Statement

The information in this report that relates to metallurgical results is based on information compiled by Cobra Resources and reviewed by Mr James Davidson who is Principal at Rendement and a Fellow of the Australian Institute of Mining and Metallurgy (FAusIMM). Mr Davidson has sufficient experience that is relevant to the metallurgical testing which was undertaken to qualify as a Competent Person as defined in the 2012 edition of the "Australasian Code for Reporting of Exploration Results, Mineral Resources and Ore Reserves". Mr Davidson consents to the inclusion in this report of the matters based on this information in the form and context in which it appears.

Information in this announcement has been assessed by Mr Rupert Verco, a Fellow of the Australasian Institute of Mining and Metallurgy. Mr Verco is an employee of Cobra and has more than 18 years' industry experience which is relevant to the style of mineralisation, deposit type, and activity which he is undertaking to qualify as a Competent Person as defined in the 2012 Edition of the Australasian Code for Reporting Exploration Results, Mineral Resources and Ore Reserves of JORC. This includes 13 years of Mining, Resource Estimation and Exploration.

About Cobra

Cobra Resources is advancing two South Australian critical minerals opportunities: the Wudinna ionic rare earth project (also referred to as the Boland project) and the Manna Hill copper project.

Cobra is advancing a rare earth discovery at Wudinna that is amenable to low-cost, low-disturbance controlled-aquifer in situ recovery (ISR), with the potential to progress towards small-scale production and bottom-quartile recovery costs, becoming the Western World's first rare earths ISR operation.

Cobra is also advancing the Manna Hill copper project in the Nackara Arc, where multiple underexplored prospects offer potential for large-scale copper-gold discoveries, including shallow skarn mineralisation and deeper porphyry potential.

Cobra also holds a significant shareholding in Barton Gold (ASX: BDG), following the 2025 sale of its non-core gold assets for A$15 million in cash and shares.

Regional map showing Cobra's tenements in South Australia

Follow us on social media:

LinkedIn: https://www.linkedin.com/company/cobraresourcesplc  

X: https://twitter.com/Cobra_Resources

Engage with us by asking questions, watching video summaries and seeing what other shareholders have to say. Navigate to our Interactive Investor hub here: https://investors.cobraplc.com/

Subscribe to our news alert service: https://investors.cobraplc.com/auth/signup

Appendix 1: Background information - the Wudinna Rare Earth Project and ISR

·    The Wudinna rare earth mineralisation was discovered by Cobra in 2023. Rare earth elements are associated with clays and organic matter within palaeochannel sands of the Narlaby Palaeochannel.

·    Mineralisation occurs within permeable sand in a hypersaline aquifer confined by low-permeability clay units.

·    ISR uses engineered injection and extraction wells to circulate a mildly acidic ammonium or magnesium sulphate lixiviant through mineralisation. The confining geology is intended to control solution flow between the wells.

·    Once REEs are mobilised into groundwater, the mineral-bearing solution can be recovered to surface through extraction wells without bulk excavation of the mineralised material.

·    ISR avoids bulk material movement and much of the conventional mining and processing infrastructure, because the REEs are recovered in solution.

·    Ionic mineralisation is highly desirable owing to its high weighting of valuable HREOs and the cost-effective method in which REEs can be desorbed

·    Ionic REE mineralisation in China is mined in an in-situ manner that relies on gravity to permeate mineralisation. The style of ISR process is unconfined and cannot be controlled, increasing the risk for environmental degradation. This low-cost process has enabled China to dominate mine supply of HREOs, supplying over 90% globally

·    Confined aquifer ISR is successfully executed globally within the uranium industry, accounting for more than 60% of the world's uranium production. This style of ISR has temporary ground disturbance, and the ground waters are regenerated over time

·    Cobra aims to demonstrate the economic and environmental performance of recovering ionic HREOs through controlled ISR within a confined aquifer.

Figure A1: Comparison between the Chinese and the proposed Boland process for ISR mining of REEs

Appendix 2: Results and Data Interpretation

ANSTO completed 24-hour desorption tests on bulk composite samples from the Boland and Head prospects. The programme compared 0.5 M, 0.3 M and, for Boland, 0.1 M AMSUL with acidified site groundwater containing no added AMSUL, across selected pH conditions.

At Boland, 0.3 M AMSUL at pH 3 achieved 70% NdPr, 67% DyTb and 67% HREE recovery. This retained approximately 91%, 91% and 93%, respectively, of the recovery achieved at 0.5 M while reducing AMSUL concentration by 40%. At 0.1 M, recoveries remained 58% NdPr, 53% DyTb and 56% HREE. Acidified groundwater with no added AMSUL recovered 51% NdPr, 46% DyTb and 50% HREE.

At Head, 0.3 M AMSUL at pH 3 achieved 36% NdPr, 52% DyTb and 53% HREE recovery, retaining approximately 90%, 91% and 91%, respectively, of the 0.5 M results. Acidified groundwater with no added AMSUL recovered 23% NdPr, 33% DyTb and 36% HREE. HREE and DyTb recoveries exceeded NdPr recovery under every Head test condition.

The results show that both AMSUL concentration and pH materially influence recovery and can be evaluated as operating trade-offs. Recovery from acidified groundwater without added AMSUL indicates that a portion of the ionically bound REEs can be desorbed without an external AMSUL addition. These tests do not establish commercial reagent consumption or the preferred operating condition; those conclusions require further column work, solution-management assessment and economic evaluation through the Scoping Study.

Table A1: Liquor recoveries from 24-hour desorption tests - Boland composite sample

Table A2: Liquor recoveries from 24-hour desorption tests - Head composite sample

Figure A2: Effect of AMSUL concentration on recovery from the Boland composite (605 ppm TREO) at pH 3

Figure A3: Effect of pH on recovery from the Boland composite at 0.5 M AMSUL

Figure A4: AMSUL concentration on recovery from the Head composite (706 ppm TREO) at pH3

Figure A5: Effect of pH on recovery from the Head composite at 0.5 M AMSUL

Table A3: Boland Composite Samples

Table A4: Head Composite Samples

Appendix 3: JORC Code, 2012 Edition - Table 1

Appendix 4: Section 2 reporting exploration results