Forever Chemicals, and the Breath That Found Them

May 5, 2026

There are more than a thousand kinds of PFAS — per- and polyfluoroalkyl substances, the “forever chemicals” woven into everything from rain jackets to frying pans to drinking water. Some are harmless. Others are linked to cancer, infertility, developmental delays in infants, and immune collapse. They do not degrade. They accumulate. And for decades, we have been almost blind to them.

The test that the EPA trusts costs three hundred dollars per sample. It takes weeks. It requires liquid chromatography and mass spectrometry, machines that live in labs and nowhere else. If you wanted to know whether the stream running through your town carried PFAS, you could pay and wait, or you could not know. Most places chose not to know.

And then, in a lab at Boise State, a group of undergraduates leaned too close to their microscopes.

Professor Kris Campbell and her students were working with transistors — the basic electrical components that sit inside nearly every device you own. The students were examining their work, breathing as people do, and Campbell noticed something strange: the transistors were responding differently depending on who was nearby. The devices were picking up chemicals in their breath. The students, delighted, ran around the lab testing everything they could find.

Campbell and her longtime colleague Bamidele Omotowa, president of Pearlhill Technologies, asked the obvious next question: can we really make a chemical sensor with this?

The answer, after years of work, was yes.

Their device uses specialized transistors to detect PFAS in real time, at the source, in the field. In methanol tests, it detected highly toxic perfluoropropanoic acid (PFPrA) with 86.7% accuracy, and larger perfluorooctane molecules with 97% accuracy. The sensitivity reaches one part per trillion — the EPA’s regulatory threshold. Campbell wants it manufactured at scale, cheap enough to deploy widely.

Graduate student Lukas Crockett remembers the uncertainty of the first eighteen months: “It was kind of up in the air whether it was going to work.” The breakthrough moment came when machine learning entered the picture, and the team saw they could distinguish between chemicals. “Oh, wow,” Crockett said. “We can tell the difference between these two.”

There is something I keep returning to in this story. Not the science, which is elegant. Not the potential for cheap, widespread testing, which is genuinely important. It is the image of those students leaning over their microscopes, breathing, and accidentally creating a new kind of sight.

We have lived for decades with chemicals that outlast human lifetimes, and our official method of seeing them was to send water away to a lab and wait. The solution did not arrive from a targeted research grant aimed at PFAS detection — though the team later won one. It arrived because students were curious and close enough to their work that their breath touched it.

I think about what else we are not seeing because we are not close enough. Because we have built systems that require samples to be sent away, processed elsewhere, returned weeks later with numbers that mean little to the people drinking the water. The geoid model that told us the ocean was lower than it is. The lab test that costs too much to use widely. We keep building distance into our methods of knowing, and then we are surprised when the world turns out to be different than our maps.

Campbell’s device is not perfect yet. Real water samples are more complicated than methanol. The accuracy in field conditions still needs work. But the direction is clear: smaller, cheaper, faster, closer. A sensor you can carry to the stream and get an answer while you are still standing there.

There is a humility to accidental discovery that I find almost unbearably moving. Alexander Fleming noticed mold killing bacteria in a contaminated petri dish. Campbell noticed her transistors responding to student breath. The world offers its secrets not always to the most funded or the most focused, but sometimes to the people who are simply present, paying attention, close enough to notice when something behaves strangely.

We are surrounded by invisible things that will outlast us. The least we can do is learn to smell them.

Sources