Tellurium

Tellurium's defense relevance is concentrated in two technology families: mercury-cadmium-telluride (HgCdTe, also called MCT) infrared detector materials, and bismuth-telluride (BiTe) thermoelectric devices used to cool those same detectors. Both are explicitly named in China's February 2025 control list, alongside the civilian CdTe solar application (Fastmarkets, Feb 2025).

1. Mercury cadmium telluride (HgCdTe/MCT) in thermal imaging and missile seekers

HgCdTe is a tunable-bandgap semiconductor alloy that can be engineered to detect infrared radiation across the short-wave, mid-wave and long-wave IR bands by varying its cadmium-to-mercury ratio — a flexibility no competing detector material matches (Mercury cadmium telluride, technical reference). This makes it the material of choice for forward-looking infrared (FLIR) imaging pods, night-vision thermal weapon sights, and infrared-homing missile seekers, alongside indium antimonide (InSb) in the mid-wave band (Infrared homing, technical reference). The U.S. Army continues to fund next-generation dual-band HgCdTe focal-plane-array detector development through its Small Business Innovation Research program (Army SBIR, FLIR Dual-Band Focal Plane Array topic).

2. Bismuth telluride (BiTe) thermoelectric coolers

HgCdTe detectors require cryogenic or near-cryogenic cooling to suppress thermal noise; bismuth-telluride thermoelectric (Peltier) coolers are a standard solid-state solution for compact IR detector packages where bulk cryocoolers are impractical (Laser Focus World, thermoelectrically cooled IR detectors). Per USGS, thermoelectric devices consume roughly 15% of global tellurium demand, the second-largest end use after solar cells (USGS MCS 2026).

3. Metallurgical and steel-alloy additive uses

Roughly 10% of global tellurium consumption goes into metallurgy — primarily as a free-machining additive to steel and copper alloys used in precision-machined components, including for defense and aerospace manufacturing (USGS MCS 2026). USGS notes several substitutes exist for this application — bismuth, calcium, lead, phosphorus, selenium and sulfur can replace tellurium in free-machining steels, "usually with losses in efficiency or product characteristics."

Global refined tellurium trade is small in absolute tonnage — total world refinery output was only about 980–998 t in 2024 and roughly 1,000 t in 2025 (USGS MCS 2026) — which makes China's 750–800 t/yr refining share the dominant variable in every downstream market.

World refinery production by country, 2024–2025 (tonnes, tellurium content)

Source: USGS Mineral Commodity Summaries 2026, world refinery production table.

U.S. import reliance and sourcing

The United States imports the large majority of the tellurium it consumes; U.S. imports for consumption were 42 t in 2021, 37 t (2022), just 8 t (2023), 6 t (2024), and an estimated 14 t in 2025 (USGS MCS 2026). Over 2021–2024, import sources were led overwhelmingly by Canada (64%), followed by the Philippines (14%), Japan (8%), Germany (5%) and other sources (9%) — China accounted for only a marginal share of direct U.S. tellurium imports even before the 2025 controls, reflecting the existing North American (Rio Tinto–5N Plus–First Solar) supply relationship. U.S. exports remained minimal at 1–15 t/yr over the same period, underscoring that the U.S. is a net importer processing material domestically rather than a raw-material exporter.

China's CdTe compound exports — the downstream flow that matters most

Because China both refines the most tellurium metal and converts a large share into cadmium telluride (CdTe) compound domestically, the more consequential trade statistic for the solar industry is CdTe compound exports, not tellurium metal. Chinese customs data compiled by Fastmarkets show China exported approximately 1,215 t of cadmium telluride in 2024, roughly flat with 2023, and up 75% from 685 t in 2022; Malaysia, Vietnam and India are the primary destinations (Fastmarkets, Feb 2025). A February 2025 Dongguan Securities research note, cited by PV Magazine, calculated that China's 2023 CdTe exports of roughly 1,200 t supported approximately 12 GW of downstream thin-film solar production capacity outside China (PV Magazine, 7 Feb 2025).

Canada's role as the North American conduit

Canada is both a primary tellurium refiner (27–28 t/yr) and the site of 5N Plus's semiconductor conversion facility that processes Rio Tinto Kennecott's output before it reaches First Solar. Natural Resources Canada identifies tellurium alongside uranium and niobium as a critical mineral where Canada plays "a crucial role as a supplier ... to the United States" (Natural Resources Canada, Mineral Trade), consistent with Canada supplying 64% of direct U.S. tellurium metal imports.

A compact chronology of the events that moved tellurium from a niche copper-refining byproduct to a formally designated U.S. and EU critical mineral at the center of the solar-energy and infrared-defense supply chains.

More than 90% of the world's tellurium is recovered as a byproduct of electrolytic copper refining, specifically from the anode slimes (sludge) that accumulate at the bottom of copper electrorefining cells alongside gold, silver, selenium, and platinum-group residues (USGS MCS 2026). Because tellurium is never the primary economic driver of a copper smelter's investment decisions, recovery is added or dropped based on copper-refining economics, anode-slime composition, and the smelter's own appetite for building a dedicated Kroll- or soda-roast-based tellurium circuit — not based on the tellurium price itself.

1. The Nassar/USGS anode-slime study: quantifying the recovery gap

A peer-reviewed 2022 study co-authored by USGS researchers, published in Resources, Conservation & Recycling, modeled the tellurium content of anode slimes generated by electrolytic copper refineries worldwide between 1986 and 2018. For 2018, the study estimated global anode slimes contained 1,930 (range 1,500–2,700, 95% confidence interval) metric tons of tellurium — "nearly quadruple the reported tellurium production for that year" (Nassar, Kim, Frenzel, Moats & Hayes, Resources, Conservation & Recycling, 2022). The study further found that China holds the greatest potential to expand tellurium supply from its existing copper-refining base, but cautioned that "most of the tellurium potentially recoverable by Chinese refineries appears to come from copper mined elsewhere" — meaning China's byproduct leverage over the tellurium market stems from its smelting capacity, not from domestic copper reserves.

2. Why recovery lags: economics, not geology, is the binding constraint

A 2025 technical review of copper-anode-slime tellurium extraction technologies concluded that "anode slime generated from copper electrolytic refining accounts for 90% of the global supply" of tellurium, but that recovery economics — not the physical scarcity of the element — limit how much smelters actually extract, since dedicated tellurium-recovery circuits require capital investment that competes against a smelter's core copper, gold and silver recovery priorities (ScienceDirect, comprehensive review of tellurium extraction and recycling from copper anode slime, 2025). This is the structural reason USGS states world tellurium "resources and reserves...are generally not reported at the mine or country level and cannot be reliably quantified" — unlike primary metals such as copper or zinc, there is no tellurium ore body to delineate; there is only a fraction of a byproduct stream that a smelter chooses, year to year, to process into metal.

3. Boliden Rönnskär (Sweden): Scandinavia's byproduct base

Boliden's Rönnskär smelter in northern Sweden recovers tellurium (via copper telluride) as one of a broad slate of byproducts extracted from copper-refining residues, alongside copper sulfate, zinc clinker and iron sand from smelting slag; Boliden separately recovers copper telluride at its Harjavalta operation in Finland (Boliden, By-products). USGS credits Sweden with roughly 46–48 tonnes per year of tellurium-bearing concentrate production in 2024–2025, making it the fourth-largest source globally after China, Japan and Russia (USGS MCS 2026). Boliden's own sustainability disclosures describe tellurium as one of several "by-products" whose recovery it treats as a resource-efficiency and circular-economy objective rather than a primary revenue driver (Boliden Group Annual and Sustainability Report).

4. Teck Trail Operations (Canada) and Yunnan Chihong (China): integrated multi-metal byproduct recovery

Teck Resources explicitly lists tellurium among the eight critical minerals its Canadian operations produce or process — copper, zinc, germanium, molybdenum, indium, antimony, tellurium and bismuth — recovered through the integrated lead-zinc-copper circuit at its Trail Operations smelter and refinery complex in British Columbia, one of the world's largest fully integrated zinc and lead smelting complexes (Teck Resources, submission to House of Commons Finance Committee, Aug 2023; Teck Resources, Trail Operations). Teck's own regulatory filings describe its principal products as copper and zinc, with tellurium among the "various specialty and other metals, chemicals and fertilizers" recovered as byproducts (Teck Resources, SEC Annual Information Form). In China, Yunnan Chihong Zn & Ge operates on the same integrated-byproduct model at far larger scale: the company reported 2024 output of 289,800 tonnes (metal content) of lead-zinc concentrate and 651,400 tonnes of smelted lead-zinc products, alongside more than 1,000 tonnes per year of combined precious and rare byproduct metals including gold, silver, cadmium, bismuth and antimony, plus a dedicated 60 tonnes/year germanium-products capacity — illustrating the scale advantage Chinese integrated smelters hold when recovering minor byproduct metals like tellurium alongside base metals (Metal.com, Yunnan Chihong 2024 production summary).

USGS's 2025 end-use breakdown for tellurium shows an application base far more concentrated in a single technology than almost any other critical mineral: solar power cells, 70%; thermoelectric devices, 15%; metallurgy, 10%; other applications, 5% (USGS MCS 2026). That solar share has risen sharply over the past decade as First Solar's CdTe manufacturing base has scaled, making tellurium's demand profile unusually exposed to a single company's capacity decisions compared with other byproduct-sourced minor metals.

1. CdTe photovoltaics vs. the much larger crystalline-silicon market

First Solar's cadmium-telluride thin-film technology remains commercially dominant within thin-film PV, but thin-film as a category is a small fraction of the overall solar market, which is roughly 95% crystalline-silicon (c-Si) globally. The U.S. Department of Energy's Solar Futures Study frames c-Si's overwhelming market share as the central substitution risk for tellurium demand: any acceleration of c-Si cost declines or supply-chain buildout directly displaces addressable CdTe volume, even without a drop-in material substitute for tellurium itself (U.S. Department of Energy, Solar Futures Study, 2021). Within thin-film specifically, USGS names amorphous silicon and copper indium gallium diselenide (CIGS) as the two principal competing chemistries to CdTe (USGS MCS 2026, Substitutes), but First Solar's scale, bankability track record, and lower carbon footprint per watt have kept CdTe the dominant thin-film chemistry through 2026, while CIGS producers have struggled to match First Solar's manufacturing cost curve at gigawatt scale.

2. Thermoelectrics: Peltier coolers and NASA's radioisotope thermoelectric generators

Bismuth telluride (Bi₂Te₃) is the workhorse thermoelectric material for solid-state Peltier coolers used in IR-detector cooling, portable refrigeration, and precision temperature control, and accounts for the bulk of tellurium's 15% thermoelectric end-use share (USGS MCS 2026). At the extreme high-reliability end of the same technology family, NASA's radioisotope thermoelectric generators (RTGs) — including the units that have powered the Voyager 1 and Voyager 2 spacecraft for more than four decades — historically used lead telluride (PbTe) and silicon-germanium thermoelectric couples to convert the heat from decaying plutonium-238 directly into electricity with no moving parts, a design NASA describes as relying on "today, we use lead telluride and silicon germanium" as the semiconductor materials of choice for deep-space power (NASA Science, How Does an RTG Work? The Seebeck Effect). Voyager's own multi-hundred-watt RTGs remain the longest continuously operating thermoelectric power systems ever built, a durability record that keeps tellurium-based thermoelectrics a NASA design default for any deep-space mission beyond useful solar-array range (NASA Science, Voyager's RTG).

3. EU classification: absent from CRMA Annexes I & II, but flagged in earlier EU critical-materials work

Unlike bismuth, antimony, gallium, germanium and tungsten, tellurium does not appear on either Annex I (Strategic Raw Materials) or Annex II (Critical Raw Materials) of the European Commission's Critical Raw Materials Act, COM(2023) 160 final — the EU's own current 34-material critical list omits tellurium entirely from its binding legal classification (European Commission, COM(2023) 160 final, Annexes I & II). This is a notable divergence from the United States, where USGS added tellurium to the Final 2025 List of Critical Minerals. The European Court of Auditors' 2026 special report on critical raw materials for the energy transition specifically flagged this gap, noting that when the EU's list was benchmarked against similar lists from Australia, India, Japan, South Korea, the UK and the United States, "tellurium and indium, deemed highly critical for the energy transition" by at least one of those other jurisdictions, are conspicuously excluded from the EU's own binding CRMA annexes (European Court of Auditors, Special Report 04/2026). Tellurium did appear as a candidate in the European Commission's earlier, non-binding 2023 Study on Critical Raw Materials for the EU, which listed it among 34 candidate critical materials reviewed before the final CRMA annexes were legislated (Institute of Geologists of Ireland, EU CRMA factsheet, citing the 2023 EU Study). In practice, this means tellurium enjoys none of the CRMA's strategic-project fast-track permitting, stockpiling coordination, or diversification targets that apply to its co-listed byproduct cousins bismuth and gallium.

4. Recycling: First Solar's closed-loop take-back program is the sector's structural outlier

Tellurium has no meaningful economy-wide recycling stream comparable to lead-acid battery recycling or copper scrap; USGS states plainly that tellurium "was recycled from CdTe solar cells in the United States, but the quantity recycled was limited because most of these cells were relatively new and had not reached the end of their useful life" (USGS MCS 2026), consistent with the brief's end-of-life PV recycling rate of under 5% industry-wide. First Solar is the clear exception and industry leader: the company established the solar industry's first global panel take-back and recycling program in 2005, pre-funding collection and recycling costs at the point of module sale, and now operates recycling facilities in the U.S., Germany, Malaysia, Vietnam and India (First Solar, Recycling). First Solar states more than 90% of module materials can be recovered for reuse, that its India facility recovers over 90% of processed-module materials, and that cadmium and tellurium separation and refining — the two regulated, highest-value elements in a CdTe module — are conducted by a dedicated third-party processor before being returned to First Solar's closed-loop semiconductor supply chain; the company states one kilogram of its recycled semiconductor material can be reused up to 41 times, a reuse cycle it calculates as more than 1,200 years of effective material life (First Solar, Recycling). Because First Solar's take-back program is prefunded and mandatory only for its own modules, it does not create an industry-wide tellurium recycling infrastructure — a structural gap that leaves the broader CdTe end-of-life stream, and any future competing thin-film manufacturer's modules, without a comparable recovery pathway.