Looking for cobalt, finding uranium
UnsplashUnsplashUnsplash
· 13 min read
In July 2026, a paper published in Nature Communications estimated that between roughly 2,000 and 5,000 tonnes of natural uranium may have left the Democratic Republic of the Congo (DRC) over the past two decades without being recorded as uranium exports.
According to the researchers, the radioactive metal was not deliberately exported as a separate commodity. Instead, uranium was present as an impurity in cobalt hydroxide — an intermediate material subsequently processed into cathode materials for lithium-ion batteries.
The finding highlights an unusual blind spot in global mineral traceability. Yet uranium contamination is far from the only environmental and social challenge associated with cobalt production in the DRC.
If you are reading this on a smartphone, tablet or laptop, there is a good chance that cobalt is present in its battery.
One historically important cathode material is lithium cobalt oxide, LiCoO₂, which became central to the development of compact, high-energy-density lithium-ion batteries.
Lithium-ion battery research began decades ago, but an important breakthrough came in 1980, when John Goodenough and colleagues demonstrated the potential of LiCoO₂ as a cathode material. This helped make batteries sufficiently compact and energy-dense for portable electronics.
During charging, lithium ions leave the LiCoO₂ crystal structure and travel through the electrolyte toward the anode:
LiCoO₂ → Li₁₋ₓCoO₂ + xLi⁺ + xe⁻
At the same time, cobalt is oxidised from Co³⁺ toward Co⁴⁺, while electrons move through the external circuit.
When the battery discharges, the process reverses. Lithium ions return to the cathode and cobalt receives electrons, returning toward the +3 oxidation state.
A good cathode material must allow lithium ions to enter and leave relatively easily while maintaining structural stability over many charging cycles.
LiCoO₂ performs well because its crystal structure consists of layers of cobalt-oxygen octahedra separated by lithium layers. Lithium can move between these layers without severely disrupting the structure. The Co³⁺/Co⁴⁺ redox pair also produces a relatively high operating voltage of around 3.9 V, allowing substantial energy storage in a relatively small volume.
Cobalt is not unique in possessing useful electrochemical properties. Nickel, manganese and iron are all widely used in modern battery chemistries. Nickel can provide high capacity, manganese can improve stability and lower costs, while iron is abundant and inexpensive.
What made cobalt particularly valuable was the combination of high voltage, structural stability and reversibility.
Today, however, battery chemistry is rapidly becoming more diversified. Nickel-manganese-cobalt, or NMC, batteries reduce the amount of cobalt required per kilowatt-hour, while lithium-iron-phosphate, or LFP, batteries eliminate cobalt entirely.
Cobalt-bearing ores occur in several parts of the world, including Australia, Indonesia, Canada, New Caledonia and Cuba. In most cases, cobalt is not mined from dedicated cobalt deposits. It is recovered as a by-product of copper or nickel production.
The Democratic Republic of the Congo occupies an exceptional position because the Central African Copperbelt crosses the southern part of the country and continues into Zambia.
Its copper deposits are unusually rich in cobalt. As a result, large quantities of cobalt are produced alongside copper.
In 2024, the DRC accounted for approximately 76% of global mined cobalt production, making developments in the Congolese mining industry highly relevant to battery manufacturers throughout the world.
This concentration is particularly important for Europe.
The EU is trying simultaneously to expand electric mobility, stationary battery storage and renewable electricity while reducing excessive dependence on individual suppliers of critical raw materials. Cobalt therefore sits at the intersection of European industrial policy, battery manufacturing, supply-chain due diligence and strategic resource security.
Working conditions in parts of the Congolese cobalt industry remain highly problematic.
Human-rights organisations estimate that roughly 10–30% of cobalt production has historically been associated with artisanal and small-scale mining.
Artisanal miners, locally known as creuseurs, sometimes work in manually excavated shafts with inadequate structural support and ventilation. Children have also been documented working in parts of the sector.
Basic protective equipment such as masks, gloves and reinforced footwear is frequently insufficient.
Industrial mining is not automatically risk-free either.
Exposure to dust containing cobalt, copper and lead can contribute to serious occupational illnesses. Chronic cobalt exposure has been associated with pulmonary fibrosis — sometimes described as "hard-metal lung disease" — as well as asthma and other respiratory problems.
Research in mining areas of Katanga has also linked elevated exposure to heavy metals with adverse health outcomes among nearby populations, including higher risks of certain congenital abnormalities.
Human health is only one part of the environmental impact.
A 2026 Nature study led by David Edwards of the University of Cambridge analysed satellite observations of African forests between 2001 and 2020, land-use maps and data covering more than 16,000 mining locations.
Using deep-learning methods, the researchers compared areas surrounding mines with otherwise similar areas without mining activity.
They estimated that metal mining had directly caused the loss of at least 187,000 hectares of forest during the study period.
The indirect footprint can be much larger.
Research focusing on mining regions in the DRC has found that mines can act as centres from which deforestation spreads through road construction, settlements and other infrastructure.
One estimate suggests that for every hectare directly cleared by mining, the surrounding development associated with the operation can eventually contribute to around 34 additional hectares of forest loss.
Yet abandoning cobalt immediately is difficult.
Electrification requires enormous quantities of battery materials. Electric vehicles are the most visible source of demand, but stationary storage connected to solar and wind generation is another rapidly expanding market.
While researchers analysed satellite imagery, journalists documented unsafe mining conditions and initiatives such as the Fair Cobalt Alliance attempted to formalise artisanal mining and improve protective equipment, containers of cobalt hydroxide continued moving through international supply chains.
It now appears that uranium may have been travelling with some of them.
The presence of uranium in Congolese geology is not a new discovery.
Shinkolobwe in present-day Haut-Katanga Province was historically one of the world's most important high-grade uranium deposits and supplied material used by the Manhattan Project during the Second World War.
The mine has officially been closed for decades, and the DRC has not operated a conventional uranium-export industry comparable with major uranium-producing countries.
But closing a uranium mine does not remove uranium from the geology surrounding other copper and cobalt deposits.
Typical ores from parts of the Central African Copperbelt may contain around 10% iron, 2–5% copper, fractions of a percent cobalt and uranium concentrations measured in tens or hundreds of parts per million.
Those uranium concentrations appear small.
The problem is scale.
When millions of tonnes of ore are processed, trace concentrations can translate into very large quantities of uranium.
This is what Ryan A. Manzuk and Sébastien Philippe investigated. They attempted to reconstruct how uranium moves through the different stages of Congolese copper and cobalt processing.
The ore is first crushed and treated with sulphuric acid, dissolving the metals into solution.
Copper is typically removed first, often through solvent extraction followed by electrowinning.
After copper extraction, the remaining solution still contains cobalt, uranium and various impurities.
The next stage involves gradually increasing the solution's pH using lime or other alkaline materials.
This allows operators to precipitate unwanted substances before recovering cobalt as cobalt hydroxide — the intermediate product exported for further refining.
Iron is relatively easy to separate.
Iron hydroxides begin precipitating at approximately pH 3.5–4.5, while most cobalt remains dissolved.
Uranium creates a more difficult problem.
Uranium species begin precipitating at around pH 6.1, whereas cobalt hydroxide begins precipitating at approximately pH 6.8.
The difference is small.
As a result, significant quantities of uranium can precipitate alongside cobalt hydroxide.
The researchers noted that Congolese processing plants often conduct their final impurity-removal stage at pH 5.5 or below to avoid losing cobalt.
At those conditions, however, no more than roughly half of the uranium may be removed.
The remainder can continue into the cobalt product stream.
Uranium that is successfully removed does not simply disappear.
It generally enters processing residues and tailings.
Many tailings facilities associated with copper and cobalt mining were not originally designed as long-term storage systems for uranium-bearing material.
The researchers therefore identified these residues as another potential environmental risk because uranium may migrate from poorly managed waste deposits into surrounding soil or water systems.
This creates a difficult trade-off.
Removing uranium from cobalt products improves the purity and traceability of exported material, but unless the resulting uranium-containing waste is properly stabilised and stored, part of the environmental risk is simply transferred from the international supply chain back to the mining region.
There are established methods for separating uranium.
Two options discussed by the researchers are ion-exchange resins and the addition of phosphoric acid.
Phosphate treatment can convert uranium into poorly soluble minerals such as autunite. In this form, uranium is much less likely to remain in exported cobalt hydroxide and can potentially be retained in a more stable solid residue.
The problem is largely economic and infrastructural rather than scientific.
Neither the necessary ion-exchange materials nor sufficient quantities of specialised chemical reagents are produced domestically at the scale required by the Congolese industry.
Mining companies therefore need to import them.
The researchers examined customs records for chemical imports associated with individual mining operations.
Only three facilities appeared to import enough phosphoric acid to treat their entire cobalt-hydroxide output.
They found little evidence of large-scale use of ion-exchange resins for uranium removal.
By 2024, the authors estimated that only around 20% of cumulative cobalt exports had passed through processing routes capable of systematically removing uranium.
There is no comprehensive dataset directly recording uranium travelling as an impurity in cobalt products.
The researchers therefore constructed several scenarios using cobalt production volumes, uranium concentrations in ores and assumptions about the efficiency of uranium-removal processes.
Even under their conservative scenario — assuming relatively low uranium concentrations and comparatively efficient domestic removal — they estimated that more than 1,000 tonnes of uranium may have left the DRC over approximately two decades.
More representative scenarios produced estimates of around 1,500–2,500 tonnes.
If some deposits contained moderately higher uranium concentrations, cumulative exports could have exceeded 5,000 tonnes.
At the same time, another approximately 3,000–4,000 tonnes of uranium may have accumulated in potentially unstable mining and processing waste inside the DRC.
This does not mean that the country was secretly operating a uranium-export industry.
The mechanism is much more mundane.
Uranium behaved as an unintended by-product of copper and cobalt extraction — much as cobalt itself is frequently recovered as a by-product of copper production.
The difference is that international regulatory and traceability systems were not designed around this particular type of accidental material flow.
Almost certainly not.
Cobalt hydroxide is not placed directly into consumer batteries.
It undergoes additional purification and chemical processing before being converted into carefully controlled cathode materials such as LiCoO₂ or NMC compounds.
Battery manufacturers require extremely precise material specifications because impurities can substantially reduce performance, durability and safety.
Significant uranium contamination would therefore normally be removed during downstream refining or identified through quality-control procedures.
One documented example illustrates what may happen.
A cobalt-processing facility in Kokkola, Finland, which handled cobalt feedstock originating from the DRC between 2010 and 2017, recovered uranium during processing and declared the resulting material.
Other downstream processors may have done something similar.
The problem is that there is no global record showing exactly where all of the uranium eventually ended up.
A significant share probably entered industrial waste streams and tailings during subsequent refining.
The most important conclusion is therefore not that several thousand tonnes of uranium have mysteriously disappeared.
Rather, the case exposes a gap in the governance of critical-mineral supply chains.
International nuclear-material control systems are designed primarily around industries that intentionally mine, process or enrich uranium.
They work much less effectively when uranium appears unintentionally inside another commodity stream.
Cobalt ore is mined, dissolved, processed, converted into cobalt hydroxide, shipped abroad, refined again, mixed with material from other sources and eventually transformed into battery-grade chemicals.
After several such stages, reconstructing the original uranium flow becomes extremely difficult.
This matters particularly for Europe, which imports large quantities of processed critical raw materials while simultaneously introducing stricter requirements concerning responsible sourcing, batteries, waste, recycling and supply-chain due diligence.
For European refiners and battery manufacturers, traceability may increasingly need to cover not just carbon footprints, human-rights risks and recycled content, but also unexpected hazardous impurities moving through upstream mineral-processing chains.
A practical response would therefore start upstream:
Natural uranium consists primarily of two isotopes.
Approximately 99.3% is uranium-238, while around 0.7% is uranium-235.
Uranium-235 can sustain a nuclear fission chain reaction much more readily than uranium-238. Its proportion therefore has to be significantly increased through enrichment for most reactor applications and especially for use in nuclear weapons.
The uranium associated with cobalt processing is natural uranium.
It cannot simply be taken from cobalt hydroxide and used directly as weapons-grade material.
It would still require extraction, conversion and technologically demanding uranium-enrichment infrastructure.
The more immediate concern is therefore environmental and occupational rather than military.
Both uranium-238 and uranium-235 undergo radioactive decay and emit alpha radiation. Alpha particles do not penetrate deeply through external barriers such as clothing or the outer layer of skin.
The primary risk arises when uranium-containing particles enter the body through inhalation or ingestion.
Moreover, uranium is not only radiologically hazardous: like several other heavy metals, it can also produce chemical toxicity, particularly affecting the kidneys.
LiCoO₂ may have been essential to the early development of lithium-ion batteries, but it is no longer the only major cathode chemistry.
Many modern devices and electric vehicles use nickel-manganese-cobalt chemistries, including NMC variants with substantially reduced cobalt content.
At the same time, lithium-iron-phosphate batteries have expanded rapidly.
LFP batteries contain neither cobalt nor nickel.
Their energy density has historically been lower than that of some nickel-rich chemistries, but improvements in battery design have made them highly competitive for electric vehicles and particularly attractive for stationary energy-storage systems.
For Europe, this diversification has strategic significance.
Reducing cobalt intensity can decrease exposure to highly concentrated upstream supply chains, while recycling can increasingly return cobalt, nickel and lithium from end-of-life European batteries into production.
Under Europe's emerging circular battery economy, the long-term objective is therefore not simply to replace one imported mineral with another. It is to combine chemistry diversification, recycling, stronger traceability and more resilient sourcing.
The uranium discovery provides another reason why that transition matters.
For years, policymakers and industry have focused on the carbon footprint, labour conditions and geopolitical concentration of critical-mineral supply chains.
The cobalt case demonstrates that mineral supply chains can also transport substances that were never intended to become part of the traded commodity at all.
The lesson extends well beyond cobalt: as Europe electrifies transport and energy systems, understanding what accompanies critical raw materials through increasingly complex global supply chains may become almost as important as knowing where those materials came from.
illuminem Voices is a democratic space presenting the opinions of leading Sustainability Thought Leaders, their views do not necessarily represent those of illuminem.
The world needs sustainability knowledge. At illuminem, no interest group or shareholder can influence our work. Thank you for supporting our mission to make high-quality and independent sustainability information free for all. Every contribution helps. Thank you for donating today.
Yury Erofeev
Battery Metals · Labor Rights
Jonatan Núñez
Lithium · Battery Metals
Yury Erofeev
Mining Tech · Battery Metals
Financial Times
Battery Metals · Battery
Carbon Credits
Battery Metals · Battery
Politico
Battery Metals · Battery Tech