I spent a while working on this sort of thing a few years ago. Right now there's basically no market for this sort of thing. The characterization technique(s) n

I spent a while working on this sort of thing a few years ago. Right now there's basically no market for this sort of thing. The characterization technique(s) need to be really sensitive, really specific, really fast, and generalized to the whole domain possible contaminants, which means expensive. Consumers don't care enough to pay for it right now. Pretty much the only practical tech right now is a card with a shitload of immuno assays on it, like those pan-drug test strips. There are a pain to validate and expensive/no fun to develop. The business model here would be spending a few million buying antibodies from commercial suppliers, validating tests, and selling cards for <<$1 per test. Test cards would likely be e.g. "artificial hormones", "pesticides I/I/III" etc Education is also an issue, when the domain of possible contaminants is so large. The sample prep in this case is still a problem, though. Easy for drinking water, harder when you're trying to take a standard sample of e.g. apple skin There is a little bit more 'political will' in China because there is greater awareness of food contamination. But the market only supports water conductivity meters and gimmicks like that, unless something's changed in the past year. The closest thing I think is the "Consumer Physics SCiO," aka the Theranos of chemical characterization. Their device is both too expensive (for a mainstream consumer product) and useless for any real world contaminant test, although it is fine at testing water/sugar/protein content for e.g. bad baby formula