Date:
Solar panels have a dirty secret: roughly 80 percent of the sunlight striking them never becomes electricity. It becomes heat, and that heat makes the panel worse at its only job. A Stanford University team decided to stop throwing that energy away. They bolted a copper tube cold plate to the back of a standard panel, cooled the module down, boosted its electrical output by 59.3 percent, and routed the captured waste heat into a reactor that pulls bagged fertilizer out of real human urine.
The result is not a lab curiosity. The research, published in Nature Water in August 2025, has since spawned a startup called Recovered Potential that is already pitching its system to slaughterhouses, backed by a U.S. Department of Energy program slot.
The core insight is straightforward. Photovoltaic cells lose voltage as they warm up, so a scorching summer afternoon can paradoxically produce less power than a cool, bright morning. The industry’s standard response has been to let that heat dissipate into the air. On a typical rooftop installation, every panel is shedding thermal energy right now, and nothing is catching it.
Stanford’s rig catches it. A copper tube plate sits flush against the back of the solar module. Coolant flows through the tubes, pulling heat away from the cells. The panel runs cooler, its voltage holds steadier, and electrical output climbs. Against earlier prototypes built without heat transfer or current control, the paired system delivered 59.3 percent more power, with a margin of error of 3.6 percentage points.
That alone would be unremarkable. Photovoltaic-thermal (PVT) systems, like hybrid solar converters that capture both heat and light, have existed for decades. The twist is where the warm coolant goes next: directly into an electrochemical reactor that needs warmth to function efficiently.
The reactor employs a process called electrochemical stripping, a technique originally developed by Will Tarpeh during his PhD at UC Berkeley. Urine flows into an electrochemical cell divided into chambers by ion-selective membranes. An electrical current, supplied by the same solar panel, drags ammonium ions across those membranes. The chemistry converts them to ammonia gas, which is then trapped on the far side in acid, producing ammonium sulfate.
Ammonium sulfate is a standard, widely used fertilizer. Nothing exotic comes out of this system. The output is the same bagged product stacked on farm supply store shelves.
The bottleneck in the process has always been the gas-escape step: ammonia must physically leave the liquid before it can be captured. Heat accelerates that transition, which is precisely why routing the panel’s waste heat into the reactor matters. The warming improved ammonia recovery efficiency by 22.4 percent compared to unheated runs, though the wider error margin of 7.4 percentage points is worth noting.
The team also installed charge controllers to prevent the panel from dumping excess current into the electrochemical cell. Pushing more amps than the cell can use wastes energy without moving additional nitrogen. The paper reports a saving of 2.24 kilojoules per gram of nitrogen for every excess milliamp per square centimeter avoided.
The study models net fertilizer revenues of up to $2.18 per kilogram of nitrogen in U.S. markets and up to $4.13 in African markets, where fertilizer costs are higher and grid infrastructure is thinner. Those are modeled ceilings under the paper’s assumptions, not sums anyone has been paid yet.
For context, DTN’s retail survey in mid-August 2026 placed urea at $678 per ton and anhydrous ammonia at $964 per ton. Converted to the paper’s units, an American farmer is currently paying roughly $1.63 per kilogram of nitrogen as urea and about $1.30 as anhydrous. The modeled ceiling clears both figures, even in a year when prices have swung sharply. Anhydrous ammonia hit $1,118 per ton in May before falling back, and urea ranged from $611 to $866 between February and April.
Those swings exist because industrial nitrogen fertilizer is fundamentally a natural gas product, a dependency that has pushed countries like India to lock in green ammonia supply agreements as an alternative. The International Energy Agency estimates ammonia production accounts for roughly 2 percent of global final energy consumption and about 450 million tonnes of direct CO2 emissions annually, making it nearly twice as emissions-intensive as crude steel.
The nitrogen already dissolved in human urine worldwide represents approximately 14 percent of annual global fertilizer demand. No one is claiming a solar panel and a copper plate can replace the Haber-Bosch process that dominates industrial ammonia production, though other projects are exploring solar-powered approaches to hydrogen and ammonia production at larger scales. The claim is narrower and more practical: a significant fraction of the nitrogen the world buys is being flushed away by the very populations that need it.
Lead author Orisa Coombs, a mechanical engineering PhD student at Stanford, captured the appeal simply: “You don’t need a giant chemical plant or even a wall socket.” Several of the experimental runs used synthetic urine for consistency, but others used the real thing, which matters because actual urine is chemically complex and full of compounds that foul membranes.
The technology has moved beyond the university lab. Recovered Potential, a startup based in Menlo Park, California, was founded by two researchers from Tarpeh’s Stanford group. CEO Kindle Williams, a former postdoc with a chemical engineering PhD from MIT, and CTO Jinyu Guo, who completed her Stanford PhD in the same lab, are leading the commercialization effort. Tarpeh serves as founding scientific advisor.
The company reports impressive technical achievements:
Two external markers support the timeline. In October 2025, the Department of Energy’s advanced research arm named ten winners under RECOVER, a nearly $25 million program targeting ammonia and critical mineral recovery from American wastewater. Recovered Potential is one of them, collaborating with Tarpeh’s Stanford lab and the Guest lab at the University of Illinois on recovering ammonium, phosphorus, and magnesium from anaerobic digestate. In 2026, both Williams and Guo were named to Activate’s fellowship cohort for hard-tech founders.
The company’s target customer list reveals the commercial logic. Recovered Potential is pursuing anaerobic digestate, meat and poultry processing waste, and fertilizer plant waste, all high-strength nitrogen streams where nitrogen arrives already concentrated, exactly what the electrochemical chemistry prefers. A flushed toilet, by contrast, is a dilute source that the system is not optimized for in its current form.
The business model follows naturally. Ammonia in wastewater is a compliance burden for the facility discharging it and a product for the farmer who needs it. The recovered fertilizer offsets part of the treatment cost, making the system a waste-management solution that happens to produce a saleable byproduct.
The solar-powered, off-grid version of the technology remains the longer-term vision. Coombs is building a follow-up prototype with triple the reactor capacity to push toward deployments where neither grid power nor centralized infrastructure exists.
One open question is longevity. Solar panels are famously durable, with modules still feeding the grid decades after installation even as their inverters have been swapped multiple times. A copper cold plate bonded to the back of a module is one more component that can fail, and the membranes, sulfuric acid supply, and urine collection logistics are where distributed nitrogen recovery has historically stalled.
What has changed in the past year is that the people who built the experiment now have a company, a DOE program number, and a prospect list that starts with a rendering plant rather than a toilet. Whether the economics hold at a commercial scale remains to be proven, but the combination of improved solar output and fertilizer production from waste streams that already exist represents a compelling dual-revenue model that neither technology could achieve alone.
Share post:
Popular

source