Mining The Batteries We Already Own

AAnnouncements from the recycling industry tend to lead with a single reassuring number: the share of the metal inside a battery that a modern plant can pull back out, now quoted at above ninety per cent for nickel and cobalt. Taken on its own, that figure implies the loop is all but closed, and that a worn-out cell is merely a raw material waiting to be sent round again. The arithmetic becomes far less comfortable once a second number is set beside it. Recovery efficiency describes what happens to a pack after it reaches a licensed facility; it says nothing whatever about the much larger question of whether the pack arrives there at all. Multiplying the two rates together gives the quantity that genuinely returns to a factory, and by that measure performance in most countries remains modest. The distinction is easy to state and, judging by the way public targets are still written, easy to forget. Almost every disagreement traced below turns on which of the two rates is the binding constraint, and on whether that constraint is a matter of logistics or of price.

BThe case that collection is the real limit has been argued most persistently by Karin Voss of the Ostvedt Institute for Material Flows, whose group set vehicle registration records against the intake books of licensed processors in four northern European states. Of the 5,300 traction packs formally logged as retired between 2016 and 2021, only 3,100 could be followed as far as a permitted plant; a further 1,000 left the region inside second-hand cars bound for markets with no processing capacity, and the remaining 1,200 vanished from the paperwork altogether. Voss maintains that this loss is structural rather than accidental, since a workshop stripping a car for parts has no legal duty to say where the battery went, and a unit holding any residual charge is worth more sold quietly than surrendered. Smaller cells are worse again: a home with a drawer of dead tools and telephones is, in her phrase, a warehouse nobody manages. She notes that her figures cover passenger vehicles only, and that the four states she examined enforce among the strictest reporting rules in Europe.

CDaniel Oyelaran, an industrial economist at the Freeport Centre for Applied Economics, is sceptical that better paperwork would change a great deal. His objection begins with chemistry. Cells built around lithium iron phosphate, which hold neither cobalt nor nickel, climbed from a fifth of new production in 2018 to roughly half by 2024, and the metal recoverable from a tonne of them is worth a small fraction of that in the older nickel-rich designs. On his model of a mid-sized plant, a tonne of nickel-rich scrap yields a gross margin near 420 dollars, against 30 dollars for the cheaper chemistry, a sum that will not cover gathering, transport and safe discharge. Oyelaran therefore contends that the units going astray in the Voss audit are not so much being hidden from the system as declined by it, because nobody bids for material that costs more to assemble than it returns. His critics answer that the model prices metals at a single moment during an unusually depressed year, and that it leaves out buses and grid storage, where packs are large, few in number and simple to follow.

DTwo complications sit awkwardly across both accounts. The first is time. A pack that has dropped below the threshold for propulsion may still hold capacity enough for stationary storage, and an operator choosing that route withdraws the material from the recycling stream for perhaps another decade; forecasts of the volumes awaiting treatment in 2035 accordingly rest on an assumption about second life that nobody can yet test. The second is design. Where modules are bolted into a standard frame a technician can lift a battery out within the hour, but many manufacturers now bond the cells straight to the floor of the vehicle with adhesive, a choice that trims weight and cost while making dismantling slow and, in certain cases, unsafe. Rules due to take effect in 2031 will oblige producers to build a minimum share of recycled material into new cells, and it is this obligation, rather than any environmental preference, that has begun to shift behaviour, since a manufacturer short of certified recycled lithium must either buy it or forfeit access to the market.

EThe two explanations are less opposed than the exchange between their authors suggests. The disappearance Voss records and the economics Oyelaran models describe one failure caught at different moments: material goes astray because nobody is paid enough to chase it, and nobody is paid enough because what waits at the end of the chain is thinner than it used to be. Which limit dominates seems to depend less on principle than on the scale of the ownership involved. Where packs are large, uniform and held by a fleet operator, following them is straightforward and the real question is whether the metals justify the process; where they are small, dispersed and privately owned, the metals may be valuable and still never be gathered. That reading implies the recycled-content rules will bite hardest in precisely the segment Voss examined, while the household drawer stays untouched by any of it, much as it was left untouched by every earlier scheme. The urban mine, in short, is real, but it is not one deposit; it is several, and only some of them can be worked at a profit.