As solar waste threatens to become a massive problem, one discovery reveals the difference between disposal and profit.
By Lab Report · Edited by Mal James
Published September 11, 2026 · How we edit
Follow Science Blog on Google
Solar panels do not last forever. Most are built to run for two or three decades, so the ones going up now are setting up a disposal problem for later. A decade ago, IRENA and the IEA put a number on it: retired solar panels, most of them glass, could pile up to as much as 78 million tonnes by 2050.
A South Korean team recently took the silicon out of a single dead panel, cleaned it up, and turned it into a hard ceramic worth far more than the shredded glass most panels become. The interesting part is not the tonnage. It is the choice between two very different things you can do with the same waste.
When a solar panel reaches the end of its life today, the easy parts come off first. The aluminum frame unbolts, the junction box detaches, and the wiring is simple to strip. What is left is a laminated sandwich of glass, plastic and a thin layer of silicon dotted with a little silver.
Much of that leftover gets shredded and sold as low-value glass, which critics call downcycling rather than genuine recycling. The materials people actually want back, the silver and the pure silicon, are the hardest and most expensive to reclaim cleanly. A 2024 review of solar recycling notes that these technical and cost hurdles keep panels flowing toward stockpiles, landfill and downcycling.
The economics are the sticking point. The same review puts the cost of recycling silicon panels at roughly $600 to $1,000 per tonne before any revenue from the recovered materials, and estimates the price would need to drop to around $300 to $400 per tonne for the numbers to work.
Arizona State’s Meng Tao put the gap plainly, telling MIT Climate that a panel costs about $20 to recycle while the recovered materials fetch maybe $10 to $12. When it costs more to recover something than it is worth, most of it ends up as cheap filler or in the ground.
The team from the Korea Institute of Energy Research and Chungnam National University tried a different angle. Instead of chasing every gram of silver or trying to feed silicon back into new solar cells, they set out to make the recovered silicon clean enough for a completely different use. Their study in Materials Today Sustainability describes recovering silicon from a real retired panel and purifying it to 99.95 percent.
Getting there took a sequence of unglamorous steps: grinding the material down, treating it in stages with acids to strip out impurities, and a simple water-settling step to float off a stubborn titanium impurity. One useful finding was that grinding harder made things worse, not better. Milling at 800 rpm left far more residual aluminum after the acid step than a gentler 400 rpm did. As corresponding author Jin-Seok Lee put it, “This clearly showed that optimization of a recycling process cannot be based simply on more intensive milling.” That reads as a lesson from one panel rather than a universal law, but it is a sensible warning against assuming more force means more purity.
The settling step only sounds trivial. Five minutes of settling removed 71.4 percent of the titanium impurity while keeping 92.3 percent of the silicon. Cheap, fast, and it held onto most of the material they actually wanted.
Purity was the setup, not the payoff. The team then reacted the recovered silicon to form silicon nitride, a hard, wear-resistant ceramic used in ball and roller bearings, cutting tools, and parts for aerospace and car engines. Its appeal there comes from its hardness and its ability to hold up under heat.
The cleaning step mattered a lot. Silicon cleaned to 99.95 percent yielded 93.1 percent of the useful form of the ceramic, compared with 54.7 percent when the extra cleaning was skipped. Leftover impurities do not just sit there; they change what the material becomes when it is cooked. Lee describes what the team says is likely a first: “To the best of our knowledge, this is the first demonstration of converting silicon recovered from actual EoL PV modules into Si₃N₄.” (EoL means end-of-life; Si₃N₄ is silicon nitride.)
Shredding a panel into glass recovers something, but the least valuable version of it.
The team says that avoiding this kind of downcycling was exactly the point Lee frames it this way: “Rather than simply recovering silicon as a secondary raw material, we aimed to demonstrate a practical pathway for giving recycled silicon a new, higher-value application.”
The logic is straightforward. If recycling only ever produces cheap filler, the numbers never work and panels keep heading to landfill. If the output is a ceramic worth real money, the cost of pulling the silicon out cleanly has a chance of paying for itself.
There is a useful twist in the choice, too. Silicon nitride does not need to be as flawless as solar-grade silicon. Feeding recycled silicon back into new solar cells demands extreme purity; a ceramic bearing tolerates more. Aiming at a market that forgives some imperfection may be a smarter target for material that started life as scrap.
A clean result on one panel is a long way from an industry. The study rests on a single retired Suntech panel, so it is best read as a proof of concept, a clue about what is possible, not evidence that it works at scale. Real waste streams are messier than one well-studied panel, and panels from different makers and different years will not all behave the same way under the same recipe.
Then there is the market. The world does not need 78 million tonnes of silicon nitride bearings, so even a wildly successful version of this process would only ever soak up a slice of the retired-panel mountain. Upcycling a fraction into something valuable and handling the rest more responsibly are not competing ideas; the ceramic route is one tool, not the whole answer.
The team is already thinking about logistics, which is often where recycling schemes quietly die. Lee says they are “currently working with Wonkwang S&T, a Korean PV recycling company, to develop mobile PV recycling technology that can process end-of-life PV modules closer to where they are generated.” Processing panels near where they retire, rather than trucking heavy glass across a country, is where the team hopes to cut transport costs by roughly 30 percent and emissions by more than 10 percent. Those are stated targets, not results in hand.
What we keep coming back to is the framing more than the chemistry. For years the default question about dead solar panels has been how to dispose of them cheaply. The more interesting question, and the one this work puts on the table, is whether the silicon inside them is worth treating as a raw material rather than a nuisance. A single purified panel cooked into a ceramic does not settle that but it does make the question harder to wave away.
Follow Science Blog on Google
Add us as a preferred source to see more Science Blog reporting in Google.
Standards
Articles on Science Blog are edited and fact-checked before publication. We use AI tools in the newsroom. See our editorial standards and masthead.
A Science Blog column
Lab Report covers new research as it publishes. Fresh studies, first findings, and the occasional retraction, drawn from the journals and the institutions doing the work.
Clear science for curious people
A Brown Brothers Media publication. All rights reserved © 2026.
Clear science for curious people — free PDF, then our science email.
Also available via RSS.

source