Crushed solar cells went into a tank of nothing but water and air, and 90 minutes later almost all the silver in them was riding the froth on top in a scoop weighing an eightieth of the feed

A retired solar panel does not look like an ore body.

Dark blue squares, an aluminum frame, a sheet of glass, and a junction box on the back.

Printed across the face of every one of those squares is a grid of fine silver lines.

Silver is the highest value material in the whole assembly.

Almost all of it goes into a landfill.

Not because it cannot be recovered, but because recovering it costs more than throwing it away.

What water and air do that acid cannot

The standard route to silver in a dead panel is acid leaching, which means chemicals going in and hazardous liquid waste coming out.

Both cost money, and the disposal side of it costs money twice.

Flotation works on a different property entirely. Some particle surfaces repel water and some do not, and a bubble will stick to the ones that repel it.

Grind the material fine, add water, blow air through it, and put in small amounts of reagents that make the target surfaces shed water.

The particles you want attach to bubbles and ride to the surface as froth, which gets scraped off. Everything else stays in the tank.

None of that is new. Mines have separated ore this way for a century, which is exactly why the interesting question was whether it works on a panel.

What went into the tank

The feed came from about 1,000 pounds of end of life panels, roughly 23 residential modules.

Stripping those down yielded about 49 pounds of ground solar cells, and that is what the pilot ran on.

The run went for about 90 minutes, continuously, which is the part that had not been done before.

A batch test stops and restarts. It can hold a recovery rate for a few minutes because nothing is moving through it.

A continuous circuit has feed entering while product leaves, and it has to hold that rate while everything moves.

That gap between a beaker and a working circuit is where most promising recycling chemistry quietly dies.

What came off the top

Recovery came out at close to 100 percent of the silver in the feed.

What it went into was a concentrate weighing 1.25 percent of the material that entered, about one eightieth of the feed.

Concentrating the same amount of metal into an eightieth of the mass means the product carries more than 80 times the silver content of what went in.

That is the number that matters commercially, because a refiner buys grade rather than tonnage.

An earlier round of the same work, run in batches, had already cleared 97 percent in a matter of minutes.

The team sits at the University of Newcastle in Australia, in a center set up for critical minerals and urban mining, which is where the flotation expertise already lived.

The team’s own preliminary costing puts the flotation route at three to five times cheaper than acid leaching.

What 49 pounds cannot tell you

The work is out as a preprint and has not been through peer review, which is the first thing to say about it.

The second is the scale. Forty nine pounds is a demonstration, and the distance from there to a plant handling thousands of tons a year is measured in feed systems, reagent handling and offtake agreements rather than in chemistry.

A concentrate also is not metal. Somebody still has to refine it, and the pilot does not include that step or its cost.

Silver is also the top of the value stack and a very small share of the mass. Glass and silicon are the bulk, and the plastic layer holding them together is still the expensive problem nobody has solved.

Work on the silicon side is moving too, with retired wafers coming out of a mild acid clean enough for battery anodes.

None of that is settled either, and none of it is in this pilot.

Why the cost gap is the whole story

Recycling a panel in Australia currently runs somewhere around seven to ten dollars.

Sending it to a landfill costs a few dollars. That difference is the reason most panels never reach a recycler at all.

The country is expecting more than a million metric tons of retired panels by the middle of the next decade, holding an estimated 300 to 500 metric tons of silver.

At today’s prices that is a serious pile of metal sitting in a waste stream, and it is not a rounding error.

Nothing here makes recycling cheap. It moves one line in the ledger, and that line happens to be the most valuable one.

Which is what the pilot claims and no more, and it is enough to matter if the next run is a hundred times bigger.