Recycling and Circularity: What Comes Back and What Does Not
Because renewable infrastructure is a stock rather than a flow, what happens at the end of its life determines whether the critical mineral requirement is a one-time extraction or a recurring one. That question is being answered differently for each technology, and the answers are more advanced for some than the public conversation suggests.
Batteries: the Ore That Comes to You
A lithium-ion pack at end of life contains nickel, cobalt, copper, aluminium and lithium at concentrations far above any natural deposit. A nickel-rich cathode is roughly a third nickel by mass; a good nickel ore is one to two percent. The material has already been mined, refined and assembled, and it arrives at a gate rather than needing to be found.
Two routes dominate. Pyrometallurgy smelts the cells and recovers nickel, cobalt and copper as an alloy. It is robust, tolerates mixed feedstock and loses the lithium and aluminium into the slag. Hydrometallurgy dissolves the black mass in acid and precipitates metals selectively, recovering more including lithium, at the cost of a more complex process and sensitivity to what goes in.
Recovery rates in commercial plants now exceed 95 percent for nickel, cobalt and copper. Lithium was long uneconomic to recover because it was cheap relative to the effort; that changed when lithium prices rose, and modern hydrometallurgical plants recover it routinely.
Direct recycling is the emerging third route: recover the cathode material as a functioning compound rather than breaking it back to elements and rebuilding. It preserves the energy and cost embedded in manufacturing the cathode, which is the largest part of a cell's value, and it is the approach best suited to lithium iron phosphate, whose elements are individually worth too little to justify full separation.
Panels: Recoverable but Not Valuable
A crystalline silicon module is about 75 percent glass, 10 percent aluminium frame, 5 percent silicon and small quantities of silver, copper and polymer. By mass, roughly 95 percent is technically recoverable, and the frame and glass come off easily.
The difficulty is that glass and aluminium are cheap. A recycler spends money to disassemble a module and receives commodity prices for most of what comes out. The valuable fraction - silver in the cell contacts, high-purity silicon - is a few percent by mass and is bound into a laminate that must be separated thermally or chemically.
In consequence most panels currently go to bulk glass recycling, where the frame and glass are recovered and the cell laminate is discarded. Full-recovery processes exist and operate, notably in France and increasingly elsewhere, and they depend on the European WEEE directive making producers responsible for end-of-life rather than on the recovered material paying for the process.
The volume question is more interesting than the technical one. Panels last 25 to 30 years and degrade slowly, so the enormous deployment of the 2010s and 2020s produces almost no waste yet. Projections put global panel waste in the tens of millions of tonnes by 2050, which is large in absolute terms and modest beside the roughly 2 billion tonnes of municipal solid waste generated each year.
Blades: the Honest Failure
A wind turbine is around 85 to 90 percent recyclable by mass. Steel tower, copper windings, cast iron hub and concrete foundation all have established recovery routes. The blades do not, and they are the reason the figure is not higher.
A blade is glass or carbon fibre in a thermoset resin, usually epoxy. Thermoset means the polymer cross-links irreversibly during curing: it cannot be melted and reformed the way a thermoplastic can, and it does not dissolve. The property that makes a blade survive twenty years of storm loading is precisely the property that prevents taking it apart.
Current practice is mechanical grinding into filler for cement or construction boards, or co-processing in cement kilns where the resin burns as fuel and the glass fibre becomes feedstock for clinker. The second is genuinely useful and is not recycling in the sense of recovering the fibre; it is a controlled disposal that displaces some coal.
Two lines of work are changing this. Manufacturers have begun producing blades with thermoplastic or chemically cleavable resins that can be depolymerised at end of life, and the first commercial recyclable blades entered service in the mid-2020s. Solvolysis processes can recover glass fibre from existing blades, though the recovered fibre is shorter and weaker than virgin material and finds lower-value uses. Neither solves the roughly 2.5 million tonnes of blades already installed and due to retire in the coming two decades.
Why Circularity Is Different Here
Recycling a drinks can saves the energy of smelting aluminium. Recycling an energy system does something structurally different: it determines whether the mineral demand of the transition is a one-time withdrawal or a permanent one.
The arithmetic is favourable because the stock is large and the loss rate is low. Copper recycled at 95 percent recovery, cycled every thirty years, loses a few percent per generation. Over a century that is a manageable top-up rather than a continuous extraction. Compare that with coal, where every tonne is destroyed on use and the next tonne must be mined.
Two things determine whether this materialises. The first is design: a pack glued together for manufacturing convenience is far harder to recover than one bolted, and regulation has begun to require design for disassembly rather than merely requiring recycling. The second is timing, because a recycling industry needs volume to justify plants, and the volume arrives twenty years after the installation boom.
That lag is why current recycling capacity looks thin relative to deployment and why it is not evidence of failure. The first large cohort of electric vehicle batteries is reaching end of life now; the first large cohort of solar panels will do so in the 2040s. Building the capacity in advance of the material is the actual policy problem, and it is a financing question rather than a technical one.
Frequently asked questions
Can electric vehicle batteries be recycled?
Yes, and well. Commercial plants recover more than 95 percent of nickel, cobalt and copper, and modern hydrometallurgical processes recover lithium too. A spent pack contains metals at concentrations far above any natural ore, which is why the economics work without subsidy for nickel-rich chemistries.
Are solar panels recyclable?
About 95 percent by mass is technically recoverable - mostly glass, aluminium and silicon. The problem is value rather than technique: those materials are cheap, so recovery costs more than the output is worth. Full-recovery plants exist and depend on producer responsibility regulation rather than on the materials paying for the process.
Why can't wind turbine blades be recycled?
Because they are thermoset composites. The resin cross-links irreversibly during curing, so unlike a thermoplastic it cannot be melted and reformed, and it does not dissolve. The property that lets a blade survive twenty years of storms is the same one that prevents taking it apart. Recyclable resins now exist but do not help the blades already installed.
Where do blades go now?
Mostly they are cut up and either landfilled or co-processed in cement kilns, where the resin burns as fuel and the glass fibre becomes raw material for clinker. That is useful disposal rather than recycling. Solvolysis can recover fibre from existing blades, but the recovered fibre is shorter and weaker than virgin material.
Why is recycling capacity so small if this matters?
Because the material has not arrived yet. Panels last 25 to 30 years and batteries around fifteen, so the enormous installations of the 2010s and 2020s produce little waste so far. The first large cohort of vehicle batteries is reaching end of life now; panels will follow in the 2040s. Building capacity ahead of the volume is a financing problem, not a technical one.
