Crystallization
Tellurium is the rare case of a critical material that is nowhere mined on purpose.
Technology brief
Tellurium is the rare case of a critical material that is nowhere mined on purpose.
The challenge
Primary use cases: tellurium recovery from copper anode slime; tellurium recovery from copper-refinery effluent — decopperising liquors, wash solutions and the combined effluent stream, which in most plants is neutralised and discharged.
Outputs/uses: tellurium to product grade for CdTe thin-film photovoltaic absorbers and for Bi₂Te₃ thermoelectrics used in cooling for optics, sensors and spacecraft; selenium, copper, gold and silver recovered alongside from the same slime; ballast salts removed as a valorised fraction; water returned at 100 % by volume; refinery-effluent neutralisation eliminated.
Industries and users: copper refineries (Chile, Peru, USA, Poland, Zambia named in the source), thin-film solar manufacturers, space and optics.
Scale: container-class automated modules tied into the existing slime-processing and effluent circuits of an operating refinery.
ARBOK solution
Tellurium is the rare case of a critical material that is nowhere mined on purpose. World refined output is 500–750 t/year — less than gold in annual tonnage — and more than 90 % of it is recovered from copper anode slime, the by-product of electrolytic copper refining. Supply is therefore governed by copper production and by how completely each refinery chooses to extract tellurium; demand is governed by photovoltaics. CdTe thin-film modules consume ~40 % of annual tellurium output, CdTe manufacturing capacity grew from 2.8 GW in 2022 to 14 GW/year by 2026–2027, and the industry projects 100 GW by 2030. At the modern absorber intensity of 15.2 t/GW, the 100 GW target alone requires ~1,520 t of tellurium per year — two to three times present world production. The gap cannot be closed by exploration, because tellurium orebodies of economic grade do not exist. China accounted for ~75 % of refined output in 2024, up from 67 % in 2023. ARBOK applies ambient-temperature deep-vacuum phase separation to anode slime and, critically, to refinery effluent — the fraction that is presently not recovered at all.
> Core technology and architecture: see ARBOK-VC (Vacuum Cracking). This entry covers the tellurium-specific feedstock, market, and economics.
Feed enters a chamber held at approximately deep vacuum, at which pressure water evaporates at ambient temperature and condenses as clean water while the dissolved and suspended load concentrates in the residual liquor. The core process employs no membranes, no electrodes, no roasting and no added chemical reagents, and returns water to the circuit at 100 % by volume. Tellurium concentrates in the residual liquor as water is removed and is brought to product grade by a secondary pass through the same stage.
Three properties matter for tellurium-bearing feeds. First, separation is insensitive to feed complexity: slime liquors and refinery effluents are high-salinity, low-pH, multi-metal streams in which selective sorbents foul and selective precipitants are consumed by the matrix rather than by the target — a phase change is unaffected by that matrix. Second, ballast salts separate by density and are removed as a valorised fraction rather than accumulating on a sorbent, which is the failure mode governing operating cost in ion-exchange circuits applied to real refinery liquors. Third, because no roasting step is involved, the recovery stage produces neither SO₂ nor CO₂, whereas conventional slime treatment is a recognised emissions point within copper refining.
Where the tellurium is lost today, and what this addresses: conventional routes — roasting with soda, sulfation, pressure leaching, and the decopperising circuits preceding precious-metal recovery — are optimised for gold and silver, with tellurium treated as a secondary or tertiary product recovered only where plant-specific economics justify the extra circuit. That is the first loss. The second is the fraction mobilised into decopperising liquors, wash solutions and effluent, which is invisible in production statistics because it never becomes a product; a water-phase separation addresses it without touching the existing slime flowsheet.
Constraints carried from the source: the effluent tellurium fraction is not characterised in the public literature with comparable precision because it is not a product and is therefore not routinely assayed, so it must be measured at site rather than assumed. Slime grade is the dominant economic variable and spans more than an order of magnitude between operations. The tellurium market is thin, and volume additions interact with price.
Market and application
Price: $243.30/kg as of 7 August 2026 — +10.8 % year to date, +66.7 % since the start of 2025, −9.0 % against the August 2025 peak of $267/kg, +60.7 % over five years. The July 2026 preprint model used a working midpoint of $180/kg against a $120–245/kg range; that midpoint is now ~35 % below spot and every figure derived from it understates the case. Re-run any project model at prevailing price.
Supply: world output 500–750 t/year, less than gold; >90 % from copper anode slime; no primary mine anywhere; China ~75 % of refined output in 2024, up from 67 % in 2023 and rising.
Export control. On 4 February 2025, by Announcement No. 10 of 2025 of the Ministry of Commerce and the General Administration of Customs, China imposed export controls on items related to tungsten, tellurium, bismuth, molybdenum and indium. This is not a ban but a licensing regime: added administrative requirement and reduced flexibility for buyers outside established channels. Coming from a supplier holding ~75 % of refined output, it converts the tellurium question from a price question into an access question, and it is the single fact on which the strategic case rests. Note that bismuth falls under the same notice — the two metals that make Bi₂Te₃ were restricted in one document on the same day.
Demand: CdTe consumes ~40 % of annual output; capacity 2.8 GW in 2022 → 14 GW/year by 2026–2027 → 100 GW planned by 2030. At 15.2 t/GW the 100 GW build-out needs ~1,520 t/year against 500–750 t produced. Early-generation modules at ~93 t/GW would have required ~9,300 t, so absorber thinning has already absorbed most of the available efficiency gain and further intensity reduction cannot substitute for supply. Second demand channel: Bi₂Te₃ thermoelectrics for optics, sensors, spacecraft.
Consequence: the deficit resolves either through demand destruction in thin-film photovoltaics or through higher recovery from copper. Geographies: copper refineries in Chile, Peru, USA, Poland, Zambia. Every copper refinery outside China holds a quarter of the world's tellurium and the independence of the solar industry.
The model is presented per tonne of anode slime, since slime rather than water volume is the natural basis for this feedstock. Recalculated at $243.30/kg (7 August 2026). The July 2026 preprint figures at $180/kg are retained in the right-hand column for comparison.
| Slime grade | Te per tonne | Value at $243.30/kg | Preprint model at $180/kg |
|---|---|---|---|
| 0.5 % Te (low end of range) | ~5 kg | ~$1,220 | ~$900 |
| 2 % Te | ~20 kg | ~$4,870 | ~$3,600 |
| 4 % Te | ~40 kg | ~$9,730 | ~$7,200 |
| 8–9 % Te (high-grade sites) | 80–90 kg | ~$19,500–21,900 | ~$14,400–16,200 |
| 22 % Te (top of reported spread) | ~220 kg | ~$53,500 | ~$39,600 |
The price move alone adds ~35 % to every line above without any change to the process, the feedstock or the recovery assumption. Selenium, copper, gold and silver recovered from the same tonne stack on top of this and are addressed in the corresponding case entries; elimination of refinery-effluent neutralisation across the treated volume is a further credit. Model presented at July 2026 prices and re-run against prevailing prices for any given project. CAPEX per module, OPEX and payback: (the source offers purchase or build-own-operate-maintain without stating figures).
Use cases
Primary use cases: tellurium recovery from copper anode slime; tellurium recovery from copper-refinery effluent — decopperising liquors, wash solutions and the combined effluent stream, which in most plants is neutralised and discharged.
Outputs/uses: tellurium to product grade for CdTe thin-film photovoltaic absorbers and for Bi₂Te₃ thermoelectrics used in cooling for optics, sensors and spacecraft; selenium, copper, gold and silver recovered alongside from the same slime; ballast salts removed as a valorised fraction; water returned at 100 % by volume; refinery-effluent neutralisation eliminated.
Industries and users: copper refineries (Chile, Peru, USA, Poland, Zambia named in the source), thin-film solar manufacturers, space and optics.
Scale: container-class automated modules tied into the existing slime-processing and effluent circuits of an operating refinery.
Implementation sequence: assay of slime and of the effluent fraction → selection of the recovery and secondary-pass configuration → module sizing → installation on the existing slime-processing and effluent circuits → commissioning. Modules are container-class and automated, operating at ambient temperature and pressure. Because the tie-in is to streams the refinery already handles, installation does not disturb the precious-metals flowsheet on which the plant's primary economics depend — an important consideration for operators for whom gold and silver recovery, not tellurium, governs the business case.
Commercial structures: outright purchase, or a build-own-operate-maintain arrangement under which the operator commits no capital and pays against recovered product and eliminated treatment cost.
Same platform and same feedstock as ARBOK-GOLD — the gold case on copper anode slime is documented separately, and gold and silver are recovered from the same tonne treated for tellurium. Same platform applied to mine waters in ARBOK-Rhenium and ARBOK-Molybdenum (the post names rhenium from mine waters as the sister case). Sits alongside copper-refinery water treatment covered by ARBOK-Copper-Waters. Platform: ARBOK-VC (Vacuum Cracking).
Tellurium is the rare case of a critical material that is nowhere mined on purpose. World refined output is 500–750 t/year — less than gold in annual tonnage — and more than 90 % of it is recovered from copper anode slime, the by-product of electrolytic copper refining. Supply is therefore governed by copper production and by how completely each refinery chooses to extract tellurium; demand is governed by photovoltaics. CdTe thin-film modules consume ~40 % of annual tellurium output, CdTe manufacturing capacity grew from 2.8 GW in 2022 to 14 GW/year by 2026–2027, and the industry projects 100 GW by 2030. At the modern absorber intensity of 15.2 t/GW, the 100 GW target alone requires ~1,520 t of tellurium per year — two to three times present world production. The gap cannot be closed by exploration, because tellurium orebodies of economic grade do not exist. China accounted for ~75 % of refined output in 2024, up from 67 % in 2023. ARBOK applies ambient-temperature deep-vacuum phase separation to anode slime and, critically, to refinery effluent — the fraction that is presently not recovered at all.
> Core technology and architecture: see ARBOK-VC (Vacuum Cracking). This entry covers the tellurium-specific feedstock, market, and economics.
Primary use cases: tellurium recovery from copper anode slime; tellurium recovery from copper-refinery effluent — decopperising liquors, wash solutions and the combined effluent stream, which in most plants is neutralised and discharged.
Outputs/uses: tellurium to product grade for CdTe thin-film photovoltaic absorbers and for Bi₂Te₃ thermoelectrics used in cooling for optics, sensors and spacecraft; selenium, copper, gold and silver recovered alongside from the same slime; ballast salts removed as a valorised fraction; water returned at 100 % by volume; refinery-effluent neutralisation eliminated.
Industries and users: copper refineries (Chile, Peru, USA, Poland, Zambia named in the source), thin-film solar manufacturers, space and optics.
Scale: container-class automated modules tied into the existing slime-processing and effluent circuits of an operating refinery.
Feed enters a chamber held at approximately deep vacuum, at which pressure water evaporates at ambient temperature and condenses as clean water while the dissolved and suspended load concentrates in the residual liquor. The core process employs no membranes, no electrodes, no roasting and no added chemical reagents, and returns water to the circuit at 100 % by volume. Tellurium concentrates in the residual liquor as water is removed and is brought to product grade by a secondary pass through the same stage.
Three properties matter for tellurium-bearing feeds. First, separation is insensitive to feed complexity: slime liquors and refinery effluents are high-salinity, low-pH, multi-metal streams in which selective sorbents foul and selective precipitants are consumed by the matrix rather than by the target — a phase change is unaffected by that matrix. Second, ballast salts separate by density and are removed as a valorised fraction rather than accumulating on a sorbent, which is the failure mode governing operating cost in ion-exchange circuits applied to real refinery liquors. Third, because no roasting step is involved, the recovery stage produces neither SO₂ nor CO₂, whereas conventional slime treatment is a recognised emissions point within copper refining.
Where the tellurium is lost today, and what this addresses: conventional routes — roasting with soda, sulfation, pressure leaching, and the decopperising circuits preceding precious-metal recovery — are optimised for gold and silver, with tellurium treated as a secondary or tertiary product recovered only where plant-specific economics justify the extra circuit. That is the first loss. The second is the fraction mobilised into decopperising liquors, wash solutions and effluent, which is invisible in production statistics because it never becomes a product; a water-phase separation addresses it without touching the existing slime flowsheet.
Constraints carried from the source: the effluent tellurium fraction is not characterised in the public literature with comparable precision because it is not a product and is therefore not routinely assayed, so it must be measured at site rather than assumed. Slime grade is the dominant economic variable and spans more than an order of magnitude between operations. The tellurium market is thin, and volume additions interact with price.
| Parameter | Value |
|---|---|
| Process temperature | Ambient temperature (water evaporates at deep vacuum without heat input) |
| Water return | 100 % by volume |
| Consumables / reagents | None — no membranes, no electrodes, no roasting, no added chemical reagents |
| Emissions from recovery step | No SO₂, no CO₂ (no roasting) |
| Tellurium finishing | Secondary pass through the same stage to product grade |
| Feed — anode slime, typical grade | 1–4 % Te by mass |
| Feed — anode slime, high-grade sites | 8–9 % Te |
| Feed — anode slime, full reported spread | 0.5–22 % Te |
| Feed — refinery effluent | Site-specific, unassayed in most plants; must be measured [требует уточнения из базы] |
| Co-recovered elements | Selenium, copper, gold, silver from the same tonne of slime |
| Ballast handling | Salts leave by density as a valorised fraction; sorbent not fouled |
| Module format | Container-class, automated; ambient temperature and pressure |
| Module throughput / energy consumption | [требует уточнения из базы] |
Deep-vacuum phase-separation chamber operating at deep vacuum, in which water evaporates at ambient temperature and condenses as clean water while the load concentrates in the residual liquor; a secondary pass through the same stage brings tellurium to product grade. Density-based removal of ballast salts as a valorised fraction. Parallel recovery of selenium, copper, gold and silver from the same slime, so a single installation addresses the full value of the feedstock rather than one element of it. Modules are container-class and automated, tied into both the existing slime-processing line and the effluent circuit. Base train detail: see the platform entry. The gold case on the same feedstock is documented separately — see ARBOK-GOLD.
Technical: separation is driven by phase change rather than by roasting, leaching chemistry or sorption, so it is insensitive to the high-salinity, low-pH, multi-metal matrix that fouls sorbents and consumes precipitants; ballast salts leave by density instead of accumulating on a sorbent; recovery extends to the effluent fraction that is presently not recovered at all.
Economic: anode slime is the richest concentrate on a copper refinery and is already collected, already handled as a valuable stream and already moving through a processing line — what is at issue is completeness of extraction, not access to the resource. Across the feedstocks examined in the same preprint series (mine waters carrying rhenium, molybdenum and cobalt at fractions of a gram per litre; phosphogypsum at 0.3 % REO; spent catalysts at 3 % rare-earth oxides), copper anode slime at 1–4 % tellurium is the richest.
Environmental: no roasting, therefore no SO₂ and no CO₂ from the recovery step; 100 % water return by volume; refinery-effluent neutralisation eliminated; no fines.
Strategic: every non-Chinese copper refinery is a supply point on which the independence of non-Chinese thin-film solar rests; full recovery at existing plants costs less than any new project, and no new mine can substitute because tellurium mines do not exist.
Same platform and same feedstock as ARBOK-GOLD — the gold case on copper anode slime is documented separately, and gold and silver are recovered from the same tonne treated for tellurium. Same platform applied to mine waters in ARBOK-Rhenium and ARBOK-Molybdenum (the post names rhenium from mine waters as the sister case). Sits alongside copper-refinery water treatment covered by ARBOK-Copper-Waters. Platform: ARBOK-VC (Vacuum Cracking).
Implementation sequence: assay of slime and of the effluent fraction → selection of the recovery and secondary-pass configuration → module sizing → installation on the existing slime-processing and effluent circuits → commissioning. Modules are container-class and automated, operating at ambient temperature and pressure. Because the tie-in is to streams the refinery already handles, installation does not disturb the precious-metals flowsheet on which the plant's primary economics depend — an important consideration for operators for whom gold and silver recovery, not tellurium, governs the business case.
Commercial structures: outright purchase, or a build-own-operate-maintain arrangement under which the operator commits no capital and pays against recovered product and eliminated treatment cost.
TRL 9 — проставлен Михаилом 2026-08-06.
The platform is deployed in container-class modules and is industrially validated at TRL 9 (stated in the preprint). No separate TRL is stated for the tellurium configuration: the figures reported constitute a feedstock-level model built on platform parameters, with project-level output established from site assay — in particular the effluent tellurium fraction, which must be measured rather than assumed. TRL of the tellurium-specific configuration: .
Price: $243.30/kg as of 7 August 2026 — +10.8 % year to date, +66.7 % since the start of 2025, −9.0 % against the August 2025 peak of $267/kg, +60.7 % over five years. The July 2026 preprint model used a working midpoint of $180/kg against a $120–245/kg range; that midpoint is now ~35 % below spot and every figure derived from it understates the case. Re-run any project model at prevailing price.
Supply: world output 500–750 t/year, less than gold; >90 % from copper anode slime; no primary mine anywhere; China ~75 % of refined output in 2024, up from 67 % in 2023 and rising.
Export control. On 4 February 2025, by Announcement No. 10 of 2025 of the Ministry of Commerce and the General Administration of Customs, China imposed export controls on items related to tungsten, tellurium, bismuth, molybdenum and indium. This is not a ban but a licensing regime: added administrative requirement and reduced flexibility for buyers outside established channels. Coming from a supplier holding ~75 % of refined output, it converts the tellurium question from a price question into an access question, and it is the single fact on which the strategic case rests. Note that bismuth falls under the same notice — the two metals that make Bi₂Te₃ were restricted in one document on the same day.
Demand: CdTe consumes ~40 % of annual output; capacity 2.8 GW in 2022 → 14 GW/year by 2026–2027 → 100 GW planned by 2030. At 15.2 t/GW the 100 GW build-out needs ~1,520 t/year against 500–750 t produced. Early-generation modules at ~93 t/GW would have required ~9,300 t, so absorber thinning has already absorbed most of the available efficiency gain and further intensity reduction cannot substitute for supply. Second demand channel: Bi₂Te₃ thermoelectrics for optics, sensors, spacecraft.
Consequence: the deficit resolves either through demand destruction in thin-film photovoltaics or through higher recovery from copper. Geographies: copper refineries in Chile, Peru, USA, Poland, Zambia. Every copper refinery outside China holds a quarter of the world's tellurium and the independence of the solar industry.
The model is presented per tonne of anode slime, since slime rather than water volume is the natural basis for this feedstock. Recalculated at $243.30/kg (7 August 2026). The July 2026 preprint figures at $180/kg are retained in the right-hand column for comparison.
| Slime grade | Te per tonne | Value at $243.30/kg | Preprint model at $180/kg |
|---|---|---|---|
| 0.5 % Te (low end of range) | ~5 kg | ~$1,220 | ~$900 |
| 2 % Te | ~20 kg | ~$4,870 | ~$3,600 |
| 4 % Te | ~40 kg | ~$9,730 | ~$7,200 |
| 8–9 % Te (high-grade sites) | 80–90 kg | ~$19,500–21,900 | ~$14,400–16,200 |
| 22 % Te (top of reported spread) | ~220 kg | ~$53,500 | ~$39,600 |
The price move alone adds ~35 % to every line above without any change to the process, the feedstock or the recovery assumption. Selenium, copper, gold and silver recovered from the same tonne stack on top of this and are addressed in the corresponding case entries; elimination of refinery-effluent neutralisation across the treated volume is a further credit. Model presented at July 2026 prices and re-run against prevailing prices for any given project. CAPEX per module, OPEX and payback: (the source offers purchase or build-own-operate-maintain without stating figures).
Feedstock grade variability: slime grade is the dominant variable and spans 0.5–22 % across reported sites, more than an order of magnitude, so project output cannot be generalised from the 2–4 % model case.
Unmeasured effluent fraction: the tellurium content of decopperising liquors, wash waters and combined effluent is not characterised in the public literature because it is not a product and is not routinely assayed — it must be established by site assay, and this is precisely the fraction on which the incremental case rests.
Thin market: tellurium volume additions interact with price, which favours deployment aimed at supply security for identified offtake rather than at maximum tonnage.
Price exposure: $243.30/kg at 7 August 2026, against a $120–245/kg historical range by purity and market. The metal is already near the top of its own range and 9 % below the August 2025 peak, so the model should not be run at spot without a downside case at $180/kg.
Operator priorities: gold and silver recovery, not tellurium, governs the refinery business case — tellurium has historically been a secondary or tertiary product recovered only where plant-specific economics justified the extra circuit, and the same logic applies to adopting a new one.
Concentration of refining: China at ~75 % of refined output in 2024 and rising shapes both pricing and the strategic case, and since 4 February 2025 exercises it directly through export licensing under Announcement No. 10 of 2025.
Structural: even complete recovery from copper is bounded by copper production, since the supply function is inelastic to tellurium demand by construction.
TRL of the tellurium configuration, module throughput, energy consumption and CAPEX are not documented: [требует уточнения из базы].
ARBOK-GOLD · ARBOK-Rhenium · ARBOK-Molybdenum · ARBOK-Copper-Waters · ARBOK-VC (Vacuum Cracking) · ARBOK-Bismuth
Related technologies
Sludge from mine water and landfill leachate is normally paid for twice: once to generate it through neutralization or filtration, and again to haul it…
TEG (ТРГ, thermally expanded graphite) is the graphene-like carbon material ARBOK produces in-house by its own process.
Arbok-Potassium converts agricultural pruning waste and beached sargassum into potassium carbonate (K₂CO₃) and metallic potassium (K⁰) using the ARBOK vacuum desalinator and a proprietary engineered-carbon…
Partnership pathway