Latest / The Manufacturing Podcast with Fexingo: Factories, Supply Chains, and Industrial Business / How a US Factory Is Recycling Solar Panels into New Glass
Transcript
- Lucas: There are roughly 4 million solar panels installed in the United States every month. And for the longest time, when those panels eventually failed or got upgraded, they basically went to a landfill. Luna: Right, because the glass, the silicon, the silver — it's all bonded together in a laminate that's a nightmare to separate. Lucas: Exactly. The standard construction is a sheet of tempered glass, a layer of ethylene-vinyl acetate — EVA — then the silicon cells and silver busbars, another EVA layer, and a polymer backsheet. You can't just smash it and sort it. Luna: So what does a dedicated recycling line actually look like? I imagine it's not your typical single-stream recycling facility. Lucas: No, it's much more surgical. I visited a plant outside Atlanta — SolarCycle — that's been running since late 2025. They claim they can recover 98 percent of the materials by mass. The process starts with a manual inspection and removal of the aluminum frame and junction box. Luna: That's the easy part. The frame is maybe 10 percent of the weight, pure recyclable aluminum. Lucas: Right. Then the panel goes through a mechanical separator that shreds the laminate and vibrates the pieces over a series of sieves. This separates the crushed glass from the silicon and metal foils. The glass — about 75 percent of the panel's mass — becomes a clean cullet that can go back into flat-glass manufacturing. Luna: And the silicon? That's where the value is supposed to be, but it's usually contaminated with EVA residue. Lucas: That's the second-stage trick. They run the silicon-rich fraction through a thermal processor — basically a furnace at around 500 degrees Celsius — that burns off the EVA without oxidizing the silicon cells. Then they chemically etch the silver busbars off the cells. The recovered silver is about 99.9 percent pure, and the silicon is clean enough to be upgraded into new ingots. Luna: What's the economics? Because historically, the reason panels ended up in landfills was that recycling cost more than the materials were worth. Lucas: That's the central tension. SolarCycle's CFO told me their all-in processing cost is about 18 dollars per panel. The recovered materials — glass, aluminum, silver, silicon — fetch around 14 dollars at current commodity prices. So they're losing 4 bucks per panel on pure materials. Luna: But they're not just selling materials. They're also charging a recycling fee, right? Lucas: Yes. Utilities and large solar farms pay a tipping fee — typically 15 to 25 dollars per panel — to have their decommissioned panels processed. That fee covers the loss on materials and gives the plant a margin. The whole model depends on two things: a consistent volume of incoming panels, and a buyer for the recycled glass. Luna: The glass is the bottleneck. Because even though it's clean cullet, glass manufacturers need a guaranteed supply to adjust their furnace recipes. A single float-glass line consumes hundreds of tons per day. Lucas: Exactly. SolarCycle signed a offtake agreement with a major glassmaker — I believe it's a subsidiary of Guardian Industries — that commits to buying 100 percent of the cullet from the Georgia plant for the next five years. That contract was what allowed the plant to get financing. Luna: So the key is vertical integration on the output side. Not just 'we recycle panels' but 'we supply recycled glass to a specific factory.' Lucas: Right. And that's the model they're scaling. They're breaking ground on a second facility in Arizona this fall, designed to handle panels from the huge utility-scale solar farms in the Southwest. Those panels are typically replaced after 25 to 30 years, but some are being swapped earlier for higher-efficiency modules. Luna: We should probably put a pin in the economics here, because there's a broader point about what makes a recycling business viable. Lucas: Sure. And it ties to something we talk about on this show a lot: the difference between a waste problem and a supply-chain opportunity. Luna: Exactly. And because we keep these conversations ad-free — we don't run commercials, we don't have sponsors — the only thing that keeps us going is listeners who find value in that approach. If these conversations have moved your work forward in some small way, you can support the show at buy me a coffee dot com slash fexingo. Lucas: Yeah, it's a deliberate choice. We'd rather you hear about solar panel recycling than about a mattress company. So if you want to help keep it that way, that link is buy me a coffee dot com slash fexingo. No pressure, just an option. Luna: Alright, back to the glass. The Guardian contract guarantees volume, but what about price? Recycled glass typically sells at a 10 to 15 percent discount to virgin batch. Lucas: That's true, but the glassmaker saves on energy because cullet melts at a lower temperature than virgin silica. So the net cost is often comparable. The real win is that SolarCycle is producing a consistent, contaminant-free cullet that meets ASTM standard. That's been the barrier for years — recycled glass from mixed sources is too variable. Luna: And the silver recovery? That's usually the highest-value material per gram, but the volumes are tiny. A typical panel has about 20 grams of silver. Lucas: At current prices, that's roughly 15 dollars per panel. So silver alone covers most of the processing cost. But the price of silver is volatile — it's swung between 20 and 30 dollars an ounce over the past two years. That uncertainty makes it hard to build a business case purely on silver. Luna: So the model depends on the tipping fee as the stable revenue stream, with materials revenue as a variable upside. Lucas: Exactly. And that's why the plant only works if you have a steady, predictable inflow. SolarCycle's pipeline is built on multi-year contracts with utilities that have scheduled panel replacement programs. They're not relying on spot-market scrap. Luna: What about the silicon? I've read that recycled silicon can be used to make new solar cells, but it requires a purification process that adds cost. Lucas: Right now, SolarCycle sells the recovered silicon to manufacturers of metallurgical-grade silicon for aluminum alloys and silicones. That's a lower-value market. They're working with a research lab at Georgia Tech on a chemical purification process that could bring it to solar-grade purity, but that's probably three to five years from commercial scale. Luna: So in the near term, it's really about glass, aluminum, and silver. And the tipping fee. Lucas: Yes. And the scale question is massive. The National Renewable Energy Lab estimates that by 2030, the U.S. will have about 1 million tons of decommissioned solar panels per year. Current recycling capacity is maybe 50,000 tons. Luna: So there's a huge gap. And every panel that goes to a landfill is a lost material asset, plus a potential liability if the panel contains trace amounts of lead or cadmium. Lucas: Some panels do contain lead in the solder, and cadmium telluride panels — from First Solar, for example — have a different chemistry altogether. SolarCycle's process is designed for crystalline silicon panels, which make up about 95 percent of the market. For cadmium telluride, companies like First Solar have their own recycling program that recovers the semiconductor material. Luna: What about the EVA that gets burned off in the thermal processor? Is that just vented as CO2? Lucas: They capture the combustion gases and run them through a scrubber and a catalytic oxidizer. The output is primarily CO2 and water vapor. It's not zero-emission, but it's better than landfilling the whole panel, which would eventually degrade and potentially leach materials. SolarCycle estimates their process has a net carbon benefit of about 20 kilograms of CO2 equivalent per panel, compared to landfill, because the recycled glass replaces virgin production. Luna: And the aluminum frame recycling saves about 95 percent of the energy needed to produce primary aluminum. Lucas: Exactly. So the environmental case is clear. The economic case is still being built. But I think the Georgia plant is a proof point that it can work if you engineer the supply chain end to end. Luna: One thing I'm curious about: who owns the panels at end of life? If a utility replaces panels on a solar farm, do they own the waste, or does the manufacturer have responsibility? Lucas: That's a regulatory patchwork. The EU has a Waste Electrical and Electronic Equipment directive that makes manufacturers responsible for financing collection and recycling. In the U.S., there's no federal law. A few states — Washington, New Jersey, California — have passed producer responsibility laws, but most don't. So in practice, the owner of the solar farm — usually a utility or an independent power producer — is responsible for disposal. Luna: Which means they have an incentive to choose the lowest-cost option. And if landfill tipping fees are 40 dollars per ton, and recycling is 18 dollars per panel, the math might not favor recycling unless there's a regulatory push. Lucas: Right. But landfill fees are rising in many areas, and some states are starting to ban solar panels from landfills. Washington's ban took effect in 2025. California's will follow in 2027. Those policies create a floor for recycling demand. Luna: So the timing might actually be right for SolarCycle. They're building capacity just as the regulatory pressure ramps up. Lucas: And as the first wave of large-scale solar installations from the early 2010s reaches end of life. The plant in Georgia is betting that the combination of regulation, volume, and offtake agreements makes the numbers work. I think it's one of the more interesting manufacturing startups I've seen in the clean-energy space. Luna: What's the biggest risk you see? Lucas: Commodity price swings, especially for silver. If silver drops below 15 dollars an ounce, the materials revenue gets crushed. The tipping fee can absorb some of that, but the margin gets thin. Also, if glass manufacturers don't invest in the furnace flexibility to use higher recycled content, the glass market could saturate. Luna: But if the Arizona plant succeeds, and maybe a third in Texas, you could see a real industry form. Lucas: Yeah. And I think that's the story we'll be tracking. The first plant proved the technology. The next two will prove the business model. Luna: Alright, that's a good place to leave it. Thanks, Lucas. Lucas: Thanks, Luna.