Global recycling market and timing
The installed base, its growth, regional distribution, and why feedstock timing follows installations from decades ago rather than last year.
Read more →We are an equipment manufacturer, not a market research house. Where a figure comes from an institution we name the institution and link to it. Where a figure would have to come from a paid report or from a project we cannot disclose, we do not quote one — we describe the mechanism instead and let you apply your own numbers.
That is why you will not find a headline market value in US dollars on this site. Market-size forecasts for solar panel recycling vary by an order of magnitude depending on assumed module lifetimes and on how much of the early-loss stream is counted. A figure from one of those forecasts would not improve a plant investment decision.
The installed base, its growth, regional distribution, and why feedstock timing follows installations from decades ago rather than last year.
Read more →What a module is made of by mass, where the recoverable value sits, and why concentration matters more than tonnage.
Read more →The thrifting trend, what has to be true for a silver route to pay, and the sensitivities that decide it.
Read more →What each route costs beyond the machine: abatement, effluent, assay and the working capital of a slower process.
Read more →Which markets have regulation, feedstock and downstream outlets — and which have only one of the three.
Read more →| Figure | Value | Source |
|---|---|---|
| Cumulative installed PV capacity, end 2025 | ≈ 2,973 GWp | IEA-PVPS, cited in the Fraunhofer ISE Photovoltaics Report |
| Cumulative capacity, alternative estimate | ≈ 2,392 GWp (on-grid plus off-grid) | IRENA, cited in the same report |
| Growth of cumulative installations, 2015–2025 | ≈ 27% CAGR | Fraunhofer ISE Photovoltaics Report |
| Share of cumulative capacity, 2025: China / Europe / North America | 51% / 17% / 9% | Fraunhofer ISE Photovoltaics Report |
| Module mass | 11.6 kg/m² | Fraunhofer ISE module composition data (2021) |
| Glass share of module mass | 67.45% | Fraunhofer ISE module composition data |
| Aluminium frame share of module mass | 12.73% | Fraunhofer ISE module composition data |
| Silicon usage per watt | ≈ 2.0 g/Wp in 2024, down from ≈ 16 g/Wp in 2004 | Fraunhofer ISE Photovoltaics Report |
| Mass of an old module that must be processed and recycled for reuse in the EU | at least 80% | EU WEEE Directive, as summarised by Fraunhofer ISE |
Figures are quoted as published and are checked when this page is revised. Always work from the current edition of the source.
The Fraunhofer ISE report states it plainly: it is technically feasible to recycle and reuse almost 100% of the materials in a PV module, but for economic reasons only the glass, the frame and the junction box are recycled today.
Read that twice, because it describes the entire commercial opportunity. The materials that are not recovered are not unrecoverable — they are simply too dilute, or too tightly bonded, to be worth recovering at the prices and scales that currently exist. Every deep-processing business case is therefore an argument about concentration and scale, not about whether the technology exists.
That is also the reason our own site is organised the way it is: dismantling first, separation second, and thermal or chemical processing only where a specific fraction has an outlet that justifies it.
Installed capacity and vintage within the distance you can economically collect from. The global figure is context; your catchment is the feedstock.
Module types, proportions, condition. Then the composition data on the material value page tells you what streams are theoretically present.
Who buys glass, aluminium, cell concentrate or metal in your region, and to what specification. This is the step most business plans get wrong.
Through-put follows from catchment and outlets, not from a target the business plan would like to hit.
A campaign-heavy feedstock and a steady one need different plants. Both appear in the same annual tonnage figure.
Tell us the country and the feedstock you expect. We will tell you which parts of the study bind hardest there and what equipment configuration they support.
Published forecasts vary by an order of magnitude because they rest on different assumptions about module lifetime and about how much of the early-loss stream is counted as waste. We do not quote a headline market value, because it would not improve a plant decision. What is stable and citable is the installed base — roughly 2,973 GWp cumulative at the end of 2025 — and the composition of the modules that base will return.
It depends on three market-specific variables: what you are paid to take modules in, what your local buyers pay for each output fraction, and your labour cost. The technology is not the constraint. Where gate fees are low and glass has no local outlet, the case is much harder than where both are favourable.
By mass, glass dominates at roughly 67% and aluminium is next at about 13%. By value per tonne, the cell fraction carrying silver and copper is the most valuable and the smallest. Aluminium is usually the value you can bank most easily, because the market for it is liquid almost everywhere.
Because a recovery rate is a property of a specific process, feedstock and market, not of solar panels in general. Quoting one would either overstate what a plant will achieve or understate what is technically possible. We publish the mechanisms and the composition data instead, so the calculation can be made with real numbers.
Tell us what arrives at your gate and what has to leave. The engineering team replies with the station configuration, the scope boundary and the information still missing before drawings start.
Prefer to talk first? WhatsApp +86 182 0399 3035 or email [email protected].
Three details are enough to start: what modules you receive, how fast you need to process them, and what output you have to produce.