Cheese leaves behind a pale, watery liquid that most dairies treat as a disposal headache. A lab in Zurich turned some of it into a gold nugget.
Not metaphorically. An actual nugget, in three pieces, the largest about five millimetres wide.
Working with metal parts stripped from 20 discarded computer motherboards, researchers at ETH Zurich dissolved the lot in acid and then fished the gold back out using a sponge made from whey protein. Final haul was 450 milligrams, at 91 per cent gold with copper making up the remainder. That works out to 22 carats, which is wedding-ring territory.
What whey is doing in a metals lab
Whey is the liquid left over when milk curdles and the solids get strained off for cheese. A fraction becomes protein powder, another portion goes into animal feed, and a great deal goes nowhere useful at all. Raffaele Mezzenga, the ETH Zurich materials scientist who led the work, told IEEE Spectrum that much of the protein simply ends up wasted.
His group heated whey protein under acidic conditions, which unravelled the compact protein globules into long strands known as amyloid nanofibrils. Freeze-drying the resulting gel produced an aerogel, a puck of solid material that is mostly empty space. Chemical engineer Mohammad Peydayesh told Chemical & Engineering News that the pucks are light enough to sit on top of a flower without crushing it.
All that emptiness is the point, because adsorption happens on surfaces and an aerogel is almost nothing but surface.
Telling one metal from another
Gold ends up inside electronics because it conducts electricity well and refuses to corrode, so manufacturers plate it in thin films onto connector pins, edge contacts and the little metal frames that chips sit on. Tiny amounts, spread across an enormous number of boards.
Dissolving a motherboard is the easy part. Aqua regia, a mixture of nitric and hydrochloric acid, reduces the metal components to a soup of ions: copper, nickel, lead, iron, tin and a small quantity of gold. Getting one of those back out of the soup on its own is where recycling costs start to pile up.
Writing in Advanced Materials, Peydayesh, Enrico Boschi, Felix Donat and Mezzenga reported that their sponge captured 93 per cent of the gold from a test solution while taking up less than 10 per cent of any other metal present at the same concentration. Capacity came in at roughly 167 milligrams of gold per gram of aerogel.
Heat finishes the job. Combusting the loaded sponge reduces the trapped gold ions to metallic flakes, which are then melted down with borax into a single lump.
Cheaper than burnt wood
Gold recovery already leans on a workhorse adsorbent: activated carbon, made from coal or timber. Per the ETH figures, recovering one gram of gold called for 5.26 grams of the protein aerogel against 15.15 grams of activated carbon, cutting the carbon footprint of that step by around a quarter.
Mezzenga reckons the wider economics are comfortable too, putting the combined material and energy costs at about one fiftieth of the value of the gold recovered. He told Anthropocene Magazine that leaning on food by-products keeps the input materials unusually cheap. That costing comes from the researchers themselves, and it describes a bench process rather than an audited plant.
The size of the pile
Scale is why any of this matters. UNITAR and the International Telecommunication Union counted 62 million tonnes of electronic waste generated worldwide in 2022, of which 22.3 per cent was documented as formally collected and recycled in an environmentally sound way. Gold embedded in that year’s discarded electronics was valued at US$15 billion, and most of it went to landfill or into informal backyard operations that run on cyanide and open flame. Annual generation is rising by roughly 2.6 million tonnes a year, heading for 82 million tonnes by 2030, while formal recycling falls further behind.
Where the caveats sit
One paper, one lab, gram quantities of gold.
Nothing here has been demonstrated at industrial scale, and the aqua regia bath remains a thoroughly unpleasant piece of chemistry that any facility would have to contain properly. Impressive as the selectivity result is, it remains singular, and has yet to be reproduced by anyone outside the group that reported it. Mustafa Guler, a biomaterials scientist at the University of Chicago with no involvement in the study, described it as a strong example of building sustainable materials from food waste.
There is an awkwardness at the front of the supply chain as well. Whey is abundant and cheap because dairy farming produces so much of it, and dairy farming carries a heavy greenhouse footprint, so the green credentials thin out somewhat once the accounting goes back a step. Mezzenga’s group has since built a version of the sponge from plant proteins, though whether it grabs gold as effectively is still untested.
Mezzenga says he is “bombarded by requests” from people who want to try the process in their own kitchens, which he does not recommend, and aqua regia is reason enough to take his word for it. The more interesting queue is the technical one. Protein sponges of this kind will latch onto platinum, palladium and silver as readily as gold, and if that holds up outside a Swiss laboratory, the nugget will turn out to be the least valuable thing this experiment produced.