A July 22, 2026 GAO technology assessment finds that mature battery recycling technologies, including hydrometallurgical chemical leaching and pyrometallurgical smelting, offer a credible near-term pathway to reduce U.S. import dependence for copper, cobalt, lithium, and nickel within two to three years. The barrier is not the science: it is the near-total absence of domestic processing infrastructure, a gap made more urgent by recent recycler bankruptcies, a new federal black mass export ban, and China's tightening grip on refining capacity across most of the minerals on the U.S. critical list.
Introduction
The U.S. Government Accountability Office published report GAO-26-108687 on July 22, 2026, reaching a conclusion that is simultaneously encouraging and alarming: the technology to recover critical minerals from spent lithium-ion batteries is mature, commercially proven, and capable of meaningfully reducing American import dependence within two to three years. The problem is that the United States currently lacks the domestic infrastructure to deploy it at anything close to the necessary scale.
The 39-page technology assessment, submitted to the House Committee on Energy and Commerce and titled "Critical Minerals: Reducing U.S. Import Reliance with Substitution and Recycling Technologies," evaluated 11 minerals used in batteries and semiconductors from a universe of 60 designated critical minerals. Its central finding draws a sharp and consequential distinction: battery recycling and substitution technologies offer near-term import relief, while semiconductor recycling and substitution remain years, and in some cases a decade or more, away from maturity.
The timing of the report is not incidental. It arrived one week before President Trump signed an executive order on July 30 delegating Defense Production Act authority to Commerce Secretary Howard Lutnick to restrict exports of critical mineral-containing scrap, including battery black mass and used lithium-ion batteries. The Bureau of Industry and Security followed on August 5 and 6 with a temporary final rule requiring that, as of August 27, 2026, all U.S. sales of black mass and tungsten scrap be directed exclusively to domestic buyers for a period of one year. The GAO's assessment and the administration's emergency export controls are, in effect, two halves of the same diagnosis: the feedstock exists, the technology exists, but the processing capacity to turn either into domestic supply does not.
The Import Dependence Baseline: What the GAO Is Actually Measuring
Before assessing what recycling can accomplish, the GAO established precisely how severe the underlying problem is. For the 11 battery and semiconductor minerals it examined in depth, the United States is at minimum 41 percent reliant on imports, a figure that applies to nickel at the low end of the range. For five of the eleven minerals, arsenic, gallium, graphite, indium, and manganese, the United States sources its entire domestic supply from overseas.
The broader USGS Mineral Commodity Summaries 2026, released February 6, 2026, identified 16 critical minerals for which the U.S. has zero domestic production, up from a smaller group in prior years. China sits at the center of that dependence: it accounts for approximately 70 percent of U.S. rare earth imports, 55 percent of antimony, roughly half of arsenic and graphite imports, and controls 99 percent of global gallium production. For refined graphite, China accounts for 96 percent of global capacity; for manganese refining, 95 percent.
The GAO's framing of the risk is direct: "Dependence on imports poses a risk of serious, sustained and long-term supply chain disruptions that could impact the nation's security, defense readiness and economy." That language is not bureaucratic boilerplate. With cobalt at 76 percent import reliance as of 2024 and the USGS 2025 critical minerals list expanding from 50 to 60 designated minerals, the structural exposure is widening even as political pressure to address it intensifies.
Among the 11 minerals the GAO focused on, copper, cobalt, lithium, and nickel are the four where near-term domestic recovery via recycling is assessed as feasible. These four happen to be the primary value targets in a lithium-ion battery pack, which makes spent EV batteries and manufacturing scrap the most consequential recycling feedstock in the United States right now.
What the Technology Can Do: Hydrometallurgy, Pyrometallurgy, and the Recovery Rate Reality
The GAO's finding on technology maturity is worth stating plainly, because it differs from the way battery recycling is often discussed in policy circles. This is not an emerging or experimental sector. Hydrometallurgical processing, which uses chemical leaching with acids or other reagents to selectively dissolve and then precipitate target metals, and pyrometallurgical processing, which uses high-temperature smelting to separate metals by their physical and chemical properties, are both established at commercial scale. The question the GAO is asking is not whether they work; it is whether the United States has built the facilities to deploy them.
The recovery performance of modern hydrometallurgical systems is striking. Current commercial operations achieve recovery rates of approximately 95 percent for lithium, 95 percent for cobalt, and 97 percent for nickel. Hydrometallurgy accounts for roughly 70 percent of global battery recycling capacity and is generally preferred for its versatility and lower energy demand compared to high-temperature smelting alternatives. The GAO's report includes a figure depicting the typical process flow for both routes, from initial battery shredding and production of black mass through leaching, separation, and final metal or precursor cathode active material output.
For battery substitution, the GAO also found near-term potential, though with important caveats. Lithium iron phosphate batteries, commonly called LFP, avoid cobalt and nickel entirely and can be deployed for stationary grid energy storage within a two-to-three-year window, reducing import exposure for both of those minerals. Sodium-ion batteries, which can reduce or eliminate the need for lithium, graphite, and other materials present in conventional lithium-ion cells, carry similar near-term potential for certain applications. The GAO is careful, however, to note that LFP's lower energy density makes it unsuitable for long-haul electric vehicle applications, and that fully lithium-free battery chemistries are not yet commercially mature. The substitution opportunity is real but application-specific.
Taken together, the GAO concludes that expanded recycling could meaningfully reduce U.S. import needs for copper, cobalt, lithium, and nickel within two to three years, and that LFP and sodium-ion deployment could reduce cobalt, manganese, nickel, and graphite demand in the same window. Those are near-term, not speculative, timelines. They are conditioned, however, on one factor the technology itself cannot supply: infrastructure.
The Capacity Gap: Bankruptcies, Black Mass Exports, and the Infrastructure the U.S. Does Not Have
The GAO's core policy problem is not scientific; it is industrial. The United States currently processes approximately 9 percent of global battery black mass volume, based on first-quarter 2026 data. It exports nearly 33,000 metric tons per month of electronic waste and scrap containing recoverable critical minerals, much of it to Asia for processing. A lack of adequate domestic facilities means that manufacturing scrap and end-of-life battery material that could feed a domestic recovery loop is instead enriching foreign refining industries, primarily China's.
The bankruptcy record of the past 18 months is the most concrete evidence of how difficult it has been to close that gap. Li-Cycle, which had secured a $475 million Department of Energy loan commitment, entered creditor protection in the United States and Canada in May 2025 after cost overruns and operational failures made it unable to draw down the funds. Ascend Elements, which had raised over $1.1 billion in capital and held federal grants including a $316 million DOE commitment that was subsequently cancelled, filed for Chapter 11 protection on April 9, 2026 in the U.S. Bankruptcy Court for the Southern District of Texas. Ascend's CEO Linh Austin described the battery recycling industry as "capital-intensive and complex," and the company cited "insurmountable" financial challenges despite continuing operations through restructuring. Quebec-based Lithion Technologies entered creditor protection in November 2025.
The pattern across these failures is consistent: adequate technology, inadequate capital durability in a market where EV adoption timelines slipped, DOE grant conditions tightened, and the economic case for recycled materials was repeatedly undercut by price volatility in primary mineral markets. Domestic battery recycling capacity is projected to take two to four years to double from current levels even under favorable conditions, a timeline that aligns uncomfortably with the GAO's two-to-three-year window for import reduction.
The BIS black mass export restriction, effective August 27, 2026, is intended to force feedstock into domestic processing rather than letting it flow overseas. But industry analysts and at least one major industry voice, Tungco CEO Cliff Nance, who opposed a comparable tungsten scrap ban before the rule was finalized, have raised the concern that domestic plants simply may not have enough throughput capacity to absorb the volume that will be redirected. Securing the feedstock is a necessary condition for the GAO's scenario to work; it is not a sufficient one.
Semiconductors: A Decade Away and Structurally Different
The GAO's treatment of the semiconductor sector stands in deliberate contrast to its battery analysis and deserves careful attention, because policymakers and investors sometimes treat battery and semiconductor mineral exposure as symmetrical problems. They are not.
For semiconductor minerals, primarily gallium, germanium, arsenic, and indium, the GAO found that neither substitution nor recycling offers meaningful near-term import relief. The barriers are structural, not merely a matter of building more plants. Semiconductor devices contain only tiny quantities of these minerals per unit, and they are typically bonded with other materials in ways that make selective recovery extraordinarily difficult at scale. Recycling technologies for end-of-life electronics remain in pilot-stage development, and the market infrastructure to collect, sort, and route semiconductor scrap into recovery streams is underdeveloped to the point of being nearly nonexistent as a commercial proposition.
Substitution faces an equally steep barrier on the performance side. The GAO quotes industry experts directly: "Industry will not adopt substitutes until they can perform at the same level as current semiconductor materials in the intended application." For gallium nitride and gallium arsenide in high-frequency, high-power electronics, no substitute currently meets that bar in production environments. The GAO assesses that a viable substitute for these semiconductor minerals is at least a decade away, and that building a recycling infrastructure capable of making recovered semiconductor minerals a primary supply source would also require a ten-year horizon.
This matters for policy prioritization. China controls 99 percent of global gallium production and has had formal export licensing controls on gallium and germanium in place since mid-2023, with further restrictions implemented in 2025. Those are exactly the minerals the GAO says the U.S. cannot substitute or recycle its way out of in any near-term scenario. As noted in my coverage of the synchronized July 2026 rare earth and technology metal rally, germanium surged 27.8 percent in a single month to $3,417.36 per kilogram, a price signal reflecting exactly the kind of supply concentration risk the GAO is describing. For semiconductor minerals, the answer will have to come from mining, geopolitical diversification, and long-duration research programs, not from the recycling toolkit that works for batteries.
Four Policy Options and the Broader Federal Response
The GAO did not publish GAO-26-108687 as a pure diagnostic exercise. It identified four discrete policy options for Congress and the executive branch to consider. The first is establishing domestic manufacturing capacity for viable substitutes, including potentially repurposing existing industrial facilities to produce LFP cathode materials or sodium-ion cell components. The second is building domestic recycling infrastructure directly, through grants, loans, or public-private partnerships aimed at scaling hydrometallurgical and pyrometallurgical processing plants. The third is securing the inputs for recycling by supporting collection, sorting, transportation, and storage systems for spent batteries and electronic scrap, the upstream logistics chain without which even a well-capitalized recycling sector cannot operate. The fourth is sustaining research, development, and testing programs to advance the next generation of substitution and recycling technologies and identify the most promising pathways before committing large-scale capital.
The GAO also notes, without endorsing any particular approach, that policymakers could choose non-technological responses including expanded domestic mining, and that financial incentives for battery collection and storage could help, provided safety concerns around battery storage and the current insufficiency of demand signals can be resolved. National e-scrap collection programs are specifically identified as a mechanism for securing recycling feedstock at scale.
The Trump administration's July 30 executive order and the BIS final rule of early August represent the most aggressive federal action to date on the feedstock security dimension, effectively implementing option three through regulatory compulsion rather than incentive. The broader administration posture, including $100 billion in Pentagon loan authority through the Office of Strategic Capital's National Security Fund Finance program, direct equity investments in processing companies, and the February 2026 announcement of nearly $12 billion toward a strategic critical mineral reserve, reflects an approach that is reaching for multiple levers simultaneously.
USGS National Minerals Information Center Director Braden Harker offered a grounding observation in this context: "We're not likely to see changes in the mining sector because it just moves a lot slower in the interim." That observation reinforces the GAO's implicit argument: recycling and substitution are attractive precisely because they can, in principle, move faster than greenfield mine development. But only if the processing capacity is built. Building on my analysis of the rare earth transition deadline in August 2026, the pattern repeating itself here is familiar: the technology and the policy intent exist, but the physical industrial infrastructure required to translate either into reliable domestic supply consistently lags the timeline that strategic circumstances demand.
Conclusion: The Two-to-Three Year Window Is Real, But It Will Not Open Itself
GAO-26-108687 is a technology assessment, not a supply chain solution. What it establishes rigorously is that the pathway to reducing U.S. import dependence for copper, cobalt, lithium, and nickel through battery recycling is technically credible and achievable within two to three years, given the right policy environment. Hydrometallurgical recovery rates above 95 percent for the target minerals, combined with the proven commercial viability of the process route, mean the science is not the constraint. The constraint is capacity, feedstock logistics, and the financial durability of the companies attempting to operate in a market where primary mineral price swings can make secondary recovery economics unviable almost overnight.
The recent bankruptcies of Ascend Elements and Li-Cycle, together representing well over $1.5 billion in raised capital and federal commitments, demonstrate how brutal those economics can be even for well-capitalized operators. The BIS black mass export restriction attempts to address the feedstock dimension by administrative force rather than market incentive, redirecting material that would otherwise flow to Asian processors back into the domestic supply chain. Whether domestic processing capacity can absorb that redirected volume before it creates a bottleneck is the immediate operational question.
For the semiconductor mineral problem, no comparable near-term answer exists. Gallium, germanium, and indium will remain structurally exposed to Chinese supply concentration for at least a decade under the GAO's technology-readiness assessment, and the export control actions Beijing has taken since 2023 have already demonstrated that this exposure carries real price and availability consequences. The policy response for semiconductor minerals will need to emphasize diplomatic supply diversification and stockpiling rather than recycling.
The GAO's four policy options provide a coherent framework, but frameworks require implementation. With IDTechEx projecting 15.9 percent compound annual growth in critical battery material recovery through 2046, and with significant EV end-of-life volumes not arriving in volume until the mid-2030s, the two-to-three-year window the GAO identifies is narrow, time-sensitive, and unlikely to recur in the same form. The technology is mature. The question the United States now faces is whether it will build the industrial system to use it before that window closes.