The strategic resource under your desk
Start with the dependency numbers, because they explain everything that follows.
The U.S. Geological Survey’s Mineral Commodity Summaries 2026, published in February, put the value of U.S. mineral production at $112 billion for 2025 — against $4.09 trillion in value added by the industries that depend on those minerals. The gap between what the country produces and what its technology economy consumes is filled by imports: USGS reported that net imports of processed mineral materials jumped from $77 billion in 2024 to $185 billion in 2025, and that China remained a major source for 14 of the 33 critical minerals the U.S. most depends on imports for — the same 14 minerals, from antimony to tungsten, that China now restricts from export to the United States.
Every one of those abstractions has a physical address inside an enterprise IT estate. The federal critical minerals list — 50 entries at its 2022 revision — reads like a teardown of a corporate laptop fleet: the rare earths in every hard drive’s magnets, the cobalt and graphite in every battery, the gallium in the chips, the tantalum in the capacitors, the palladium on the boards. When policymakers talk about “recoverable critical minerals and materials,” they are talking, in meaningful part, about the equipment enterprises retire every refresh cycle.
That is the frame this guide argues for: a retired device fleet is not just a data-security liability and a residual-value line — it is a small, already-collected ore body. The rest of the guide explains why that framing went from environmental talking point to national-security doctrine in three years, and what it changes for the people who manage disposition.
The U.S. technology economy runs on imported minerals; retired electronics are one of the few domestic sources that already exist at scale. That mismatch — not sentiment about recycling — is why end-of-life IT hardware became a policy surface.
The chokepoints, stage by stage
“Dependence on China and Taiwan” compresses two different problems into one phrase. Walk the supply chain from rock to rack and they separate cleanly.
A device reaches a desk through roughly five stages: minerals are mined, then refined and processed into usable materials, then formed into components — magnets, battery cells, capacitors — then fabricated into semiconductors, then assembled into finished hardware. The dependencies live at different stages, on different countries.
The minerals chokepoint is China — and it is mostly about refining
China’s position at the bottom of the stack is strongest not at the mine but at the refinery. Per USGS, China accounted for roughly 69% of world rare-earth mine production in 2025 (270,000 of 390,000 tons of rare-earth oxide equivalent) — but its share of processing runs far higher: analysis by CSIS notes that until 2023, China accounted for 99% of global heavy rare-earth processing, and reporting on its export controls puts its share of refined gallium supply near 98%. The U.S. imported 67% of the rare-earth compounds and metals it consumed in 2025, with 71% of those imports coming from China directly. Downstream, the imbalance compounds: China produced an estimated 300,000 tons of sintered NdFeB magnets in 2024, while U.S. production of the NdPr oxide that feeds magnet-making was measured in the low thousands of tons.
The chips chokepoint is Taiwan — and it is a different problem
Taiwan barely figures in the minerals story. Its chokepoint sits three stages up the stack: the fabrication of leading-edge logic. TSMC’s own reporting shows advanced nodes — 7-nanometer and below — accounted for 74% of its 2025 wafer revenue, and the world’s most advanced processors are overwhelmingly fabricated on the island. That concentration is why semiconductor policy (fab subsidies, export controls on chipmaking tools) exists as its own track. Conflating the two chokepoints leads to bad conclusions — recycling electronics does nothing for fab capacity, and building fabs does nothing for magnet supply. This guide’s subject is the first chokepoint: the minerals and materials layer, where recovery from end-of-life electronics is actually relevant.
Materials recovery addresses the bottom of the stack — the mined-and-refined layer China dominates. It is a genuine lever there, and no lever at all for chip fabrication. Precision about which dependency you’re discussing is the difference between a strategy and a slogan.
The export-control timeline: 2023–2026
The policy story is best read as an escalation ladder — each rung a Chinese export measure, a market reaction, or a U.S. response.
Gallium, germanium, graphite
China imposes export licensing on gallium and germanium (effective August 1), then extends controls to high-purity graphite — the first shots in the minerals campaign.
The know-how ban
China bans export of rare-earth extraction, separation, and smelting technology — targeting the processing layer where its advantage is deepest.
Antimony
Export limits on antimony. Chinese antimony exports fall roughly 97% and global prices surge about 200% — a live demonstration of chokepoint power.
The U.S.-specific ban
China prohibits exports of gallium, germanium, antimony, and superhard materials to the United States specifically — no longer licensing, but a named-country ban.
Rare earths join
Seven medium and heavy rare earths — samarium, gadolinium, terbium, dysprosium, lutetium, scandium, yttrium — go under export licensing; 16 U.S. defense and aerospace firms are placed on an export control list.
Extraterritorial reach
New rules extend Chinese controls to foreign-made products containing as little as 0.1% Chinese-origin rare earths — modeled on Washington’s own foreign direct product rule.
The conditional pause
Following late-October trade talks, China suspends the October measures through November 10, 2026, and the U.S.-specific mineral bans through November 27, 2026. The April rare-earth licensing and controls on tungsten, tellurium, bismuth, molybdenum, and indium stay in force.
The architecture hardens
China updates its export licensing catalogue in January and, in March, issues a supply-chain security framework integrating export controls with data-security and investment screening — the machinery becomes permanent.
Two features of this timeline matter more than any single entry. First, the controls concentrate on the processing layer — the December 2023 technology ban and the licensing regimes both protect China’s refining advantage rather than its mines. Second, the current calm is a pause, not a rollback: the suspensions are dated, the licensing infrastructure remains, and the extraterritorial rules are shelved, not repealed. Supply-chain planners — and the policymakers watching them — are behaving accordingly.
Between mid-2023 and late 2025, export restriction went from hypothetical risk to demonstrated practice across gallium, germanium, graphite, antimony, and the rare earths. The November 2025 suspensions bought time; they did not restore the old world.
Washington’s response: mine, make — and now recover
The U.S. counter-moves came in three waves — and the third wave is the one that reaches ITAD directly.
Wave one: unlock domestic production
Executive Order 14241, Immediate Measures to Increase American Mineral Production (March 20, 2025), invoked the Defense Production Act to accelerate domestic mining and processing — faster permitting, priority financing, and DPA authorities put behind new supply. It treats the problem at the mine.
Wave two: rebuild the middle of the chain
The most striking single move came on July 10, 2025, when the Department of Defense announced a partnership with MP Materials — operator of the flagship U.S. rare-earth mine at Mountain Pass, California — that made the Pentagon the company’s largest shareholder, at roughly 15% on an as-converted basis, via a $400 million preferred investment. The deal underwrites a second magnet plant (the “10X Facility,” expected to begin operations in 2028, lifting U.S. NdFeB capacity toward an estimated 10,000 metric tons), sets a ten-year price floor of $110 per kilogram for NdPr, commits DoD to purchase 100% of the new facility’s output for a decade, and finances expanded heavy rare-earth separation at the mine. Whatever one’s view of the state taking equity in a miner, the signal is unambiguous: magnet supply is now treated as defense infrastructure.
Wave three: keep the recoverable stream home
The third wave is the newest — and the reason this guide exists. On July 30, 2026, a Presidential Determination under Section 101 of the Defense Production Act declared that recoverable critical minerals and materials in end-of-life products and industrial waste — black mass, end-of-life rare-earth permanent magnets, swarf, and other waste and scrap containing critical minerals (copper scrap addressed separately) — are critical industrial resources essential to national defense, and delegated export-restriction authority to the Commerce Department. Commerce moved within a week: a Bureau of Industry and Security allocation order published August 6 and effective August 27, 2026 requires U.S. sellers of black mass and tungsten waste and scrap to allocate 100% of monthly sales to U.S. persons, with reporting of buyers, Schedule B codes, and quantities, an exception process through BIS, public comments due November 4, 2026, and a sunset currently set for August 27, 2027.
Read the three waves together and the shape of the policy is clear. Wave one grows new supply on a mining timescale — years to decades. Wave two rebuilds processing and components — also years. Wave three is the only lever that works on existing material: the minerals already above ground, already refined once, sitting in the country’s aging device fleets and scrap streams. That is the stream enterprise ITAD programs feed.
This section describes fast-moving policy as of publication (August 2026). The BIS order is open for comment until November 4, 2026 and currently expires August 27, 2027; the November 2025 Chinese suspensions carry their own 2026 expiry dates. Our companion brief tracks the developments as they land.
What’s actually inside retired IT hardware
Component by component, an enterprise device fleet maps onto the critical minerals list with uncomfortable precision.
| Component | Critical materials | What they do there |
|---|---|---|
| Hard disk drives | Neodymium, praseodymium, dysprosium; ruthenium, platinum | NdFeB permanent magnets drive the actuator and spindle motor, with dysprosium added for heat resistance; ruthenium and platinum appear in platter coatings |
| Batteries — laptops, UPS, mobile | Lithium, cobalt, nickel, graphite | Lithium-ion cells; when shredded, this stream becomes black mass — the material named first in the BIS allocation order |
| Printed circuit boards | Gold, silver, palladium, copper; tin | Contacts, connectors, plating, and multilayer ceramic capacitors; tin in solder throughout |
| Capacitors | Tantalum | High-reliability capacitors on servers, networking gear, and enterprise boards |
| Semiconductors & optics | Gallium, germanium, arsenic, silicon | Compound semiconductors, high-frequency chips, LEDs, and fiber-optic components |
| Chassis, cabling, power | Copper, aluminum | Conductors, heat sinks, and power delivery — the bulk metals that make recovery economics work |
The concentrations are the point. One metric ton of printed circuit boards contains at least 200 kilograms of copper, 0.4 kilograms of silver, and 0.09 kilograms of gold — precious-metal densities that can run ten times higher than naturally mined ores, which is why the industry calls board-stream recovery urban mining without irony. A data center decommissioning or a thousand-laptop refresh is, in materials terms, a concentrated deposit that somebody already dug, refined, and assembled once.
Two practical corollaries for IT and facilities teams. First, the value is unevenly distributed: drives, boards, and batteries carry most of the critical-minerals content, which is why serialized, stream-level handling beats weighing mixed scrap. Second, the same components that carry the minerals carry the data — the drive with the neodymium magnets is the drive with the customer records — so recovery and certified data destruction are not separate workflows but sequential steps in one chain of custody.
Drives, batteries, boards, capacitors, chips: the critical-minerals list is physically present in every rack and laptop bag. The materials Washington just declared essential to national defense are the ones your disposition program already touches.
How recovery actually works: from custody to refinery
Materials recovery is a staged industrial process, and each stage decides what the next one can achieve. It starts earlier than most people think — at collection.
Stage 1 — Collection and custody
Nothing is recovered that is never collected. Devices enter the stream through enterprise ITAD programs, takeback and mail-back channels, and municipal collection — and the chain of custody established here determines both data security and material traceability downstream. This is the stage enterprises control completely.
Stage 2 — Triage: reuse before recovery
Certified processors apply a hierarchy: equipment with remaining life is sanitized and redeployed or resold before anything is shredded. Reuse outranks materials recovery in the R2v3 hierarchy of responsible management strategies for good reason — a working device retains far more embodied value (and embodied minerals) than its separated commodities. Only what fails triage moves to recovery.
Stage 3 — Dismantling and depollution
Before mechanical processing, batteries come out — both as a fire hazard and as a distinct recovery stream — along with other components requiring special handling. Data-bearing media are destroyed or sanitized under the controls certification requires, with serialized certificates tying each drive to its destruction record.
Stage 4 — Shredding and mechanical separation
Shredded material passes through a cascade of physical separations: magnets pull ferrous metal, eddy-current separators kick out aluminum, density and optical sorting split copper, boards, and plastics. The outputs are commodity streams — ferrous, aluminum, copper, a precious-metal-rich board concentrate, and, from battery lines, black mass.
Stage 5 — Refining: pyrometallurgy and hydrometallurgy
Concentrates go to specialized refiners. Pyrometallurgy — smelting — recovers copper and precious metals from board concentrate at scale. Hydrometallurgy — chemical leaching and separation — is the route for battery black mass, pulling lithium, cobalt, and nickel back out as battery-grade salts. Which refiners, and in which country, is exactly what the new export rules now govern.
Stage 6 — The frontier: magnet-to-magnet recovery
The newest stage is the most strategically interesting. In a 2024 pilot, Western Digital, Microsoft, Critical Materials Recycling, and PedalPoint Recycling ran roughly 50,000 pounds of shredded end-of-life hard drives through an acid-free dissolution process developed at the Ames National Laboratory’s Critical Materials Innovation Hub — recovering about 90% of the rare-earth content and returning it to U.S. supply chains, with an estimated 95% reduction in greenhouse-gas emissions versus virgin mining. It is exactly the loop the July determination contemplates: American end-of-life drives becoming American magnet feedstock.
Recovery is a chain — custody, triage, dismantling, separation, refining — and its weakest link is almost always the first one: collection. Everything downstream, including the strategic outcomes, inherits from whether retired equipment enters a documented, certified stream at all.
The urban-mining math
The honest case for recovery runs through the UN’s Global E-waste Monitor 2024 — the standard census of the world’s e-waste — and the numbers cut both ways.
The scale side: the world generated 62 million metric tons of e-waste in 2022 — up 82% from 2010, on pace for 82 million tons by 2030. The metals embedded in that single year’s stream were valued at $91 billion, including $19 billion in copper, $15 billion in gold, and $16 billion in iron. Against that, documented formal collection and recycling captured just 22.3% of the total — about $28 billion in secondary raw materials recovered — and the monitor projects the rate falling toward 20% by 2030 as generation outruns collection infrastructure.
The concentration side is the part the mining comparison gets right: board-stream precious-metal densities running an order of magnitude above ore grades mean recovery per processed ton compares favorably with primary extraction — the 95% emissions advantage in the Ames Lab drive pilot is one measure of the same fact. And the monitor’s forward math is stark: lifting collection and recycling to 60% by 2030 would generate benefits exceeding costs by $38 billion.
Then the sobering side. For the minerals this guide is about, current recovery barely registers: no more than 1% of demand for essential rare earth elements is met by e-waste recycling. Recovery rates for the precious metals in e-waste run around 20%; even copper — valuable, easy to identify, infinitely recyclable — manages about 60%. Anyone selling recycling as a near-term replacement for mining is overselling it, and this guide won’t.
The material is real ($91 billion embedded in one year’s e-waste), the concentration is real (richer than ore for precious metals), and the shortfall is real (22.3% collected; under 1% of rare-earth demand met). The math says recovery is a serious secondary supply — if, and only if, collection improves.
Why the gap persists — and what closes it
If the minerals are valuable and the policy tailwind is real, why is so little recovered? Three structural reasons — each with a lever attached.
1. Collection is the bottleneck, not chemistry
The technology to recover most of these materials exists; the feedstock never arrives. Devices sit in drawers and storerooms, leave through informal channels, or ship abroad as untracked used equipment. The lever is unglamorous: organized takeback — enterprise ITAD programs, deployment-paired legacy retrieval, mail-back kits — that moves retired hardware into documented streams instead of gray ones. This is the lever enterprises hold personally.
2. The economics have been marginal
Recovery margins are thin, and for rare earths especially, recyclers competed against Chinese-subsidized primary supply that kept prices below viability — one reason so little magnet recovery existed before 2025. The lever here is policy, and it is now being pulled: the MP Materials price floor, DPA financing authorities, and the BIS order’s domestic-allocation requirement all change the revenue math for domestic recovery capacity.
3. The streams lose their identity
Mixed, undocumented scrap is worth its bulk metals at best. Recovering the critical fraction requires knowing what entered the shredder — drives segregated for magnet recovery, batteries for black mass, boards for precious-metal refining — which requires custody documentation from the first mile. The lever is the certified-processor infrastructure that already exists for data security reasons: serialized tracking, R2v3 downstream management, audited chains. The compliance apparatus and the strategic-recovery apparatus turn out to be the same apparatus.
Notice what all three levers have in common: they operate at the collection-and-custody layer, not the refinery. The refining capacity is being financed into existence by wave-two policy; the feedstock problem is solved — or not — one enterprise disposition program at a time.
The enterprise ITAD playbook
None of this requires an enterprise to become a minerals trader. It requires treating disposition as supply-chain participation — which mostly means asking existing questions with new seriousness.
- Get retired assets out of the drawer. Stored equipment is stranded feedstock and stranded risk in equal measure. A standing disposition cadence — refresh-paired takeback, mail-back kits for distributed sites, scheduled pickups — moves material into documented streams while it still has reuse value.
- Ask your vendor where materials physically go. Domestic certified downstream, or export? The BIS allocation order is pressure-testing export-dependent chains right now; a vendor with a documented domestic downstream is on the right side of where policy is moving. The due-diligence guide and scorecard turn this into concrete questions.
- Demand stream-level documentation. Serialized destruction certificates, downstream disclosures under R2v3 Appendix A, and weight-and-stream reporting are the custody layer that makes recovered material creditable — to your ESG reporting today, and to whatever documentation regime tomorrow’s rules bring.
- Let the hierarchy work. Reuse first: a redeployed or resold device keeps its minerals in service at full value. Recovery is the right answer for what genuinely has no life left — and value recovery on the rest funds the program.
- Pair deployment with retrieval. If new equipment is going out — a refresh, a build-out, a staged deployment — the legacy equipment should come back on the same truck. Collection is the bottleneck; same-visit retrieval is the cheapest fix for it that exists.
- Watch the register. The BIS comment window closes November 4, 2026; the order sunsets (or doesn’t) in August 2027; the Chinese suspensions expire on their own 2026 dates. Disposition contracts signed this year should assume the rules will move.
The through-line: enterprises don’t control refining capacity or trade policy, but they control collection, custody, and vendor selection — and those happen to be the three levers Section 08 identified as decisive. In a supply chain where the feedstock is the constraint, the disposition decision is the strategic decision.
Collect on a cadence, verify the downstream is domestic and documented, demand stream-level custody records, reuse first, retrieve at deployment, and assume the rules keep moving. That is the whole playbook — and most of it is what a well-run ITAD program does anyway.
Frequently asked questions
What critical minerals are in retired IT equipment?
More than most inventories capture. Hard drives carry neodymium-praseodymium magnets, often with dysprosium for heat resistance, plus ruthenium and platinum in platter coatings. Laptop and UPS batteries hold lithium, cobalt, nickel, and graphite — the stream that becomes black mass when shredded. Circuit boards concentrate gold, silver, palladium, and copper; capacitors use tantalum; and the chips themselves are built with gallium, germanium, and arsenic. Nearly all of these appear on the federal critical minerals list.
Can e-waste recycling really replace mining for critical minerals?
Not today, and honest advocates don't claim it can. The UN's Global E-waste Monitor 2024 found that no more than 1% of demand for essential rare earth elements is met by e-waste recycling, and only 22.3% of e-waste is formally collected and recycled at all. What recycling can do is grow a meaningful domestic secondary supply — the same report valued the metals embedded in one year's e-waste at $91 billion — and every point of improvement in collection rates moves recoverable material from landfills and export streams into domestic refining.
What did the July 2026 Defense Production Act determination change?
It formally declared that recoverable critical minerals and materials in end-of-life products and industrial waste — black mass, end-of-life rare-earth permanent magnets, swarf, and other scrap containing critical minerals — are critical industrial resources essential to national defense, and it delegated authority to the Commerce Department to restrict their export. The follow-on BIS allocation order, effective August 27, 2026, requires sellers of black mass and tungsten scrap to allocate 100% of monthly sales to U.S. persons. Retired IT hardware is upstream of exactly those streams.
Does the BIS allocation order apply to my company's retired IT assets?
For most enterprises, not directly. The order lands on the companies selling black mass and tungsten waste and scrap — recyclers and commodity traders — not on the organizations retiring laptops and servers. But it reshapes the downstream your ITAD vendor sells into: export-dependent disposition channels now face real restrictions, and reporting obligations follow the material. The practical enterprise question is where your vendor's downstream goes, and whether that path is documented.
Why does it matter where my ITAD vendor's materials go?
Because disposition is now supply-chain participation. A vendor processing through domestic, certified downstream channels keeps recovered materials in the U.S. supply chain and keeps your disposition clear of evolving export restrictions; an export-dependent chain is exposed to exactly the channels policy is tightening. Certification frameworks already audit this — R2v3's Appendix A governs downstream vendor qualification — so the evidence exists at any serious provider. Ask for it, and keep it with your disposition records.
Keep your retired fleet in the domestic supply chain
CyberCrunch processes retired IT assets through domestic, certified recycling channels — R2v3 downstream management, NAID AAA destruction security, serialized reporting, and full downstream transparency — so recovered materials stay in the U.S. supply chain and your disposition records stay audit-ready as the rules evolve.
This guide is informational only and reflects publicly available sources as of August 2026, including the USGS Mineral Commodity Summaries 2026, the UN Global E-waste Monitor 2024, CSIS and other policy analyses of Chinese export controls, the July 30, 2026 Presidential Determination and the August 6, 2026 BIS allocation order as published in the Federal Register, Executive Order 14241, and company and national-laboratory announcements of the MP Materials partnership and the Ames Laboratory hard-drive recovery pilot. It is not legal, trade-compliance, or investment advice; export-control obligations are fact-specific, and figures, shares, and policy status change quickly in this area — verify current requirements against the Federal Register and consult qualified counsel before acting.