Page URL: /supply-chain/ Macro Context: The gallium supply chain as an integrated system - how mining, refining, production, and recycling connect from bauxite in the ground to 6N metal in a semiconductor fab, where the concentration points are at each stage, and what the emerging non-Chinese supply base looks like through 2028.
| Stage | Key Data Point | China’s Control |
|---|---|---|
| Mining (bauxite) | 57-59 ppm Ga in bauxite; Guinea produces most bauxite (123 Mt, 31%) | ~23% of global bauxite, but ~80% of gallium output |
| Recovery (Bayer process) | ~70% of ore Ga enters liquor; only 15% of refineries extract it | ~80% of global Ga recovery capacity |
| Refining (crude to 4N-6N) | 4 stages; zone refining 18 passes for 6N; Dowa (Japan) leads outside China | ~99% of 6N+ capacity |
| Production (named refiners) | Chalco: ~160 t/yr capacity; global output ~750-760 t (2024) | ~98-99% of output |
| Recycling (secondary) | New scrap: 27-47% recovery; end-of-life: ~0%; $6M US TRACE-Ga program | ~98% of secondary refining capacity |
| Emerging non-China supply | Metlen (Greece): 50 t/yr by 2028; Alcoa (Australia): 100 t/yr FID pending | Declining from 99% toward ~78% by 2028 if projects deliver |
The gallium supply chain runs through five sequential stages: bauxite mining, Bayer process recovery (where gallium is extracted from alumina refinery liquor), crude-to-high-purity refining, final production and distribution to industrial customers, and secondary recycling of manufacturing scrap. Each stage adds concentration - the 57 ppm gallium in raw bauxite becomes 4N-6N metal at 99.99-99.9999% purity by the time it reaches a GaAs wafer manufacturer. China controls every stage at 98-99%, meaning a disruption at any node - not just mining - cuts Western access to finished material.
| Stage | Input | Output | Concentration Added | Key Node |
|---|---|---|---|---|
| 1 - Bauxite mining | Raw ore (57-59 ppm Ga) | Crushed, washed bauxite | None - physical separation only | Guinea (most volume); China (most Ga recovery) |
| 2 - Bayer process recovery | Bauxite + NaOH at 150°C | Ga-enriched sodium aluminate liquor + crude Ga (~99%) | ~1,000x (ppm → %) | Chinese alumina refineries with recovery circuits |
| 3 - Refining (to 4N-6N) | Crude ~99% Ga | 4N (99.99%) to 6N (99.9999%) | ~10,000x impurity removal | China (99% of 6N); Dowa Japan (non-China leader) |
| 4 - Production / distribution | Refined Ga ingots | Gallium metal to wafer fabs, LED makers, solar producers | Logistics, certification, CoA | Dealer network; direct supply agreements |
| 5 - Recycling (new scrap) | GaAs, GaN, CIGS manufacturing waste | Recovered crude Ga (recycled back to stage 3) | Closes loop for ~27-47% of new scrap | Neo Performance Materials (Canada); Indium Corp (USA) |
China’s disproportionate gallium output is not geological - it is infrastructural. China built gallium recovery circuits into its alumina refineries systematically from the 1990s onward. Guinea produces more bauxite than China (123 Mt vs 93 Mt in 2023) but generates essentially no gallium because its bauxite is exported raw or processed into alumina without gallium recovery equipment installed. Australia’s Darling Range bauxite has the same ~57 ppm gallium content as Chinese deposits but produces a fraction of the world’s gallium for the same reason. The supply concentration problem can, in theory, be solved with capital investment - but that investment takes 5-10 years to deploy.
| Country | Bauxite (2023) | Bauxite Rank | Gallium Output | Gallium Rank | Recovery Circuits? |
|---|---|---|---|---|---|
| Guinea | 123 Mt | #1 | <1 t/yr | None | No - raw export focus |
| Australia | 104 Mt | #2 | <1 t/yr | None | Partial - Alcoa project pending |
| China | 93 Mt | #3 | ~735-745 t/yr | #1 (98-99%) | Yes - widespread installation |
| Brazil | 31 Mt | #4 | <1 t/yr | None | No |
| India | 30 Mt | #5 | <1 t/yr | None | No |
Approximately 30% of gallium in bauxite ore is lost to red mud (bauxite processing residue) at the Bayer leaching stage and cannot currently be recovered at commercial scale. Of the 70% that enters the sodium aluminate liquor, only refineries with installed solvent extraction circuits recover it - and only about 15% of global alumina refineries have such circuits. The result is that the effective sector-wide gallium recovery rate from bauxite is approximately 9-10% of the gallium theoretically present in the ore. The remaining 90% is either in red mud (unrecoverable today) or discarded with the liquor by refineries without recovery equipment.
| Loss Point | % of Original Ore Ga Lost | Recoverable? | Cumulative Ga Retained |
|---|---|---|---|
| Red mud separation (Bayer leaching) | ~30% | Not commercially (lab: 94.77%) | 70% remaining |
| Refineries without recovery circuits (~85% of sector) | ~63% of 70% = ~44% | Yes - requires capital investment | ~9-10% recovered overall |
| Refining losses (solvent extraction, electrorefining) | ~5-8% | Partially | ~55-63% of liquor Ga |
| End-of-life product disposal (LEDs, electronics) | ~95-100% of Ga in products | Not yet (no commercial infrastructure) | Near zero returned |
The largest recoverable opportunity: Installing gallium recovery circuits in the 85% of alumina refineries that currently lack them - not building new mines, not developing new recycling - would be the fastest path to expanding non-Chinese gallium supply.
Approximately 10-15% of global gallium production comes from zinc smelting rather than bauxite. Sphalerite (zinc sulfide ore) contains gallium at 50-2,137 ppm depending on deposit type - potentially far higher concentrations than bauxite per tonne of ore processed, though zinc volumes are much smaller than bauxite. Japan’s Dowa Holdings, the world’s leading high-purity gallium producer outside China, recovers gallium from zinc concentrates imported from Mexico. This makes Japan’s gallium supply chain dependent on uninterrupted zinc ore imports - a secondary supply chain dependency within what appears to be a non-Chinese supply route.
Two projects represent the most advanced non-Chinese gallium supply additions by 2028. Metlen Energy & Metals (Greece) took a final investment decision in January 2025 on a €295.5 million bauxite-alumina-gallium project backed by €90 million from the European Investment Bank - Europe’s first industrial gallium production facility, targeting 50 tonnes per year at full scale by 2028. Alcoa, Sojitz Corporation, and JOGMEC signed a strategic tri-party agreement to evaluate gallium recovery from Western Australian alumina refineries, with a final investment decision expected by end of 2025 and 100 tonnes per year targeted. Combined, these two projects would add approximately 150 tonnes per year - more than 10x current non-Chinese output of roughly 15 tonnes.
| Project | Country | Type | Ga Capacity Target | Start Date | Key Partner |
|---|---|---|---|---|---|
| Metlen bauxite-alumina-gallium | Greece | Primary (bauxite by-product) | 50 t/yr full scale | 2026 ramp; 2028 full | European Investment Bank (€90M) |
| Alcoa Western Australia | Australia | Primary (refinery addition) | ~100 t/yr | FID by end 2025; start 2026 | Sojitz + JOGMEC (Japan) |
| DOE TRACE-Ga program | USA | Recovery from Al/Zn streams | ~1 t/yr (prototype) | Awards early 2026 | US DOE ($6M) |
| RareX Cummins Range | Australia | Primary (carbonatite deposit) | TBD | Mining lease Jan 2026 | ASX-listed |
| Non-China total (2024) | Multiple | Various | ~15 t/yr | Current | - |
| Non-China projected (2028) | Multiple | Various | ~165-175 t/yr | If all projects deliver | - |
Critical gap: Both the Metlen and Alcoa projects target crude gallium (approximately 99-99.5% purity, below 4N) or 4N recovery - not the 6N purity required by semiconductor fabs. Japan’s Dowa Holdings remains the only non-Chinese refiner capable of producing 6N gallium at commercial scale.
The gallium supply chain has five distinct concentration points where a single actor or country controls a stage. The mining stage is relatively distributed - Guinea, Australia, China, and Brazil each contribute meaningfully to global bauxite. But every stage downstream from mining becomes progressively more concentrated in China. By the refining and production stages, concentration has reached 98-99%. The recycling stage adds a secondary loop but is also dominated by Chinese capacity. Each concentration point compounds the others: a disruption at any node from Bayer recovery onward cuts Western access to 98%+ of global supply.
| Stage | China Share | Next Largest Non-China | Single-Point Failure Risk |
|---|---|---|---|
| Bauxite (ore) | 23% | Guinea 31% | Low - distributed globally |
| Bayer recovery (liquor Ga extraction) | ~80% | Negligible | High - infrastructure concentrated |
| Crude refining (to 3N-4N) | ~95% | Japan/South Korea: ~3-4% | Very high |
| High-purity refining (5N-6N) | ~99% | Japan (Dowa): ~<1% | Extreme |
| Production / named refiners | ~98-99% | Japan: ~<1% | Extreme |
| Recycling (new scrap) | ~70-80% | Canada (Neo Perf. Materials): ~5-10% | High |
| End-of-life recycling | ~0% globally | ~0% | N/A - no infrastructure anywhere |
For the structural analysis of why this concentration creates supply vulnerability, see gallium supply chain risks.
Manufacturing scrap - reject wafers, epitaxial residues, and GaAs production floor waste - creates a secondary supply loop running parallel to primary mining. New scrap recycling currently recovers 27-47% of manufacturing waste gallium at specialist facilities including Neo Performance Materials (Peterborough, Ontario) and Indium Corporation (New York). End-of-life recycling - recovering gallium from discarded LEDs, solar panels, and electronics - runs at effectively zero because the per-unit gallium value is below the cost of manual disassembly. The US DOE TRACE-Ga program ($6 million, 2025) targets 1 tonne per year prototype recovery from industrial streams. At 50% new scrap recovery, secondary supply could increase 314% from current levels.
Where gallium exists geologically, why it has no primary ore deposit, how it concentrates in bauxite at 57-59 ppm and in sphalerite at up to 2,137 ppm, which countries produce the most bauxite, why China converts 23% of global bauxite into 80% of global gallium output, and what the emerging non-Chinese mining pipeline looks like - including RareX’s Cummins Range carbonatite deposit (6,826 g/t Ga₂O₃ peak grade, mining lease approved January 2026) and China’s 49,000-tonne confirmed gallium reserves in Inner Mongolia coal deposits.
The technical four-stage purification process from crude gallium at 98-99% to semiconductor-grade 6N at 99.9999% - covering solvent extraction with Kelex 100 and Aliquat 336, vacuum refining, electrorefining in NaOH electrolyte at 30-50°C, and zone refining at 7.3 mm/h over 18 passes. Includes purity grade specifications (4N through 7N), impurity limits by element, analytical verification methods (GDMS vs ICP-OES), application-specific purity requirements (6N for GaAs, 7N for GaN), and the critical finding that all Western new projects target 4N-5N while semiconductor fabs need 6N-7N.
Named companies, output capacities, and locations for every major global gallium producer - Chalco’s 160 t/yr four-branch network as the world’s largest single producer, Zhuzhou Keneng and Vital Materials as leading Chinese high-purity refiners, Dowa Holdings as the world’s number one non-Chinese high-purity producer, and the full non-Chinese pipeline through 2028. Includes the HHI concentration index (>5,000, vs the DOJ “high” threshold of 2,500), a table of historical producers that ceased (Germany 2016, Hungary 2015, Kazakhstan 2013, US 1987), and year-by-year Chinese export volume data from 2022 through the 2023-2025 export control period.
The two-stream recycling reality: manufacturing new scrap (27-47% recovery, commercially viable, operational at specialist facilities) vs end-of-life consumer product recovery (~0%, economically blocked by per-unit labor cost exceeding gallium value). Technical recovery processes by feedstock type - GaAs (97-99%), CIGS solar (96% at 99.49% purity), GaN LED (99% via HCl leaching but not economic at unit scale). Active policy programs including the US DOE TRACE-Ga ($6M), EU CRMA 2030 targets, and Metlen’s crude recovery project. Quantifies the 314% secondary supply increase possible if new scrap recovery hits 50%.
The gallium supply chain is secure for China and fragile for everyone else. China controls 98-99% of output across every downstream stage from Bayer recovery to 6N refining. The supply disruption from export controls since August 2023 created a 7-8x price split between Chinese domestic prices ($250/kg) and Western spot ($2,100/kg, March 2026) - not because global gallium is scarce, but because China’s refining capacity is inaccessible to Western buyers under current policy. The two largest Western supply projects (Metlen and Alcoa) will add approximately 150 tonnes per year of crude-to-4N material by 2028 - meaningful but not sufficient to close the 6N refining gap that semiconductor fabs actually need filled.
| Supply Chain Stage | Western Security Level | Time to Improve | What Would Improve It |
|---|---|---|---|
| Bauxite mining | Moderate (Guinea, Australia distributed) | 0-2 years | Add recovery circuits to existing refineries |
| Bayer recovery circuits | Very low (85% of refineries have none) | 3-5 years per project | Capital investment at Guinea, Australia, Brazil refineries |
| Crude refining (to 4N) | Very low | 3-5 years | Metlen + Alcoa projects (both in development) |
| High-purity refining (6N+) | Critically low | 7-10 years | Greenfield 6N refinery (no Western project announced) |
| Recycling (new scrap) | Low-moderate | 2-4 years | TRACE-Ga + improved industrial SX circuits |
| End-of-life recycling | None | 8-15 years | Automated collection + processing infrastructure |