How “trapped” is measured — and why a metal is trapped
Recycling = end-of-life recycling input rate (EU CRM 2023 EOL-RIR): the share of a material’s supply met by recycled end-of-life scrap. We cross it with companionality into a trapped index = companionality × (1 − recycling): high only when supply can neither be mined afresh (by-product) nor recovered from scrap. Then we split the trapped metals by why they aren’t recycled: dissipated (used dispersively or physically lost — helium vents to air, arsenic and phosphate disperse into chemicals and soil — so no technology recovers them: trapped forever) vs recoverable-but-uncollected (locked in products that could be recycled but currently aren’t — trapped for now, fixable by collection and policy).
Sources & caveats: recycling metric = EU CRM 2023 EOL-RIR, one of several the field uses (the UNEP / International Resource Panel 2011 status report also defines EOL-RR, recycled content and old-scrap ratio; current benchmark is the IEA 2024 recycling report). It is EU-centric and omits in-process/new scrap and future capacity — today’s functional secondary supply, not a ceiling. The dissipative vs recoverable split is classified from each metal’s dominant end-use per Ciacci, Reck & Graedel, “Lost by Design” (ES&T 2015) and “Losses and lifetimes of metals” (Nature Sustainability 2022) — a literature-based judgement, like companionality. Inputs: data.json × companionality.json → recycling.json.
The no-way-out map
Up = more of it is an unavoidable by-product (no new mine). Left = less is recovered from scrap (no secondary buffer). The top-left is the trap: no primary response and no recycling.
Every material, ranked by how trapped it is
| Material | supply type | by-prod % | EOL recycling | why not recycled | trapped |
|---|
Trapped forever, or trapped for now?
“Barely recycled” hides two very different fates. Some metals are dissipated — used dispersively or physically lost, so no collection scheme can recover them. Others are simply uncollected — sitting in products that could be recycled if anyone bothered. The first is a wall; the second is a to-do list.
What this closes, and what it opens
This completes the supply-structure arc that began with the satellite attribution gap: primary geography (mines), supply elasticity (companionality), systemic exposure (host shock), and now the secondary buffer (recycling). Read together they say the same thing three ways — for a handful of metals, the usual mitigations simply aren’t available, and the only real levers are host-side recovery yield, stockpiles, and substitution. The open branch from here is demand: which technologies pull hardest on exactly these trapped metals? That is the next arm of the atlas to grow.