How to Cut an Onion Without Crying: What Lab Tests Show

Halfway through dicing an onion for soup, the sting arrives. Eyes water, the cook steps back from the board, and someone in the room suggests holding a slice of bread between the teeth. Kitchens collect fixes like this, and few of them have ever been tested. The chemistry behind the sting is well mapped, though, and a 2025 lab study from Cornell University tracked what sprays off the blade as it cuts. Together they split onion advice into a short list with measurements behind it and a longer list that rests on habit.

Two enzymes and a sulfur compound

An intact onion keeps the ingredients for tears apart. When a knife ruptures its cells, an enzyme called alliinase meets a sulfur-containing precursor, 1-PRENCSO (also called isoalliin), and splits it into sulfenic acids. A second enzyme, lachrymatory factor synthase, or LFS, converts one of them, 1-propenyl sulfenic acid, into the lachrymatory factor. A lachrymator is any substance that makes the eyes produce tears.

The LFS step was identified by Imai and colleagues in a 2002 paper in Nature, as summarized in a later study in Plant Physiology. Without LFS, the same sulfenic acid condenses with itself into a different compound, di-1-propenyl thiosulfinate, rather than the irritant.

The Cornell researchers name the irritant propanethial S-oxide and propose that droplets thrown off by the blade may carry it. That is the route their study followed; other routes were not tested. Put together, these steps show where any fix can act: how many cells the knife breaks, how the released irritant travels, and whether it reaches the eyes.

A cutting routine, ordered by evidence

Only the first steps below rest on measurements. The later ones follow from the chemistry but have not been put through a controlled trial.

Source: Cornell University, arXiv preprint 2505.06016 (May 2025).

1. Sharpen or hone the knife before starting. The Cornell team measured blade tip radius, meaning how rounded the cutting edge is at a microscopic scale, from about 0.91 µm to 13.3 µm. Droplet count rose up to fortyfold from the sharpest edge to the bluntest. The researchers link the first burst of droplets to pressure building inside the onion’s first layer as the blade presses in, and blunter blades produced more droplets, with more energy.
2. Cut at a steady, unhurried pace. Raising cutting speed from 0.44 to 2.03 m/s increased droplet count about fourfold, though those speeds came from a falling guillotine blade, not from tested hand chopping. Faster strokes also launched droplets with more energy, so a slower knife sends less liquid, less forcefully, into the air.
3. Treat chilling as optional. Onions stored at 1 °C for 12 hours showed no significant change in droplet velocity (p = 0.4; a p-value that high means the difference could easily be chance). They also ejected a larger volume. The authors said temperature effects were hard to isolate, so the test neither confirms nor rules out chilling.
4. Keep the eyes out of the path. Goggles that seal around the eyes, a running range hood or a fan pulling air away from the board all aim at the last link in the chain: the irritant has to reach the eye. These follow from the chemistry rather than from a measured trial, so treat them as backups to a sharp knife and a slower cut.

What the Cornell blade tests did and did not show

The study, posted to arXiv on 9 May 2025 by Jung’s laboratory, used a guillotine-style cutter: a falling blade rather than a hand-held knife. The onions were ordinary bulb onions (Allium cepa) from local vendors, used within two weeks. The team tracked hundreds to thousands of droplets in the first 0–20 ms of each cut, with starting speeds from 1.0 to 40 m/s.

Droplets formed in two stages. The pressure-driven burst came first. Then strands of liquid stretched into the air and broke apart into slower drops, a process the authors call ligament fragmentation. The direction of the cut relative to the onion made no significant difference to droplet velocity (p = 0.6).

The key limit is what the study did not count: tears. Tear induction was the motivation, not a measured outcome. The paper’s discussion says its results “provide experimental validation” for the belief that sharpening reduces tearing, but the data show fewer droplets, not fewer tears. The falling blade also stirred up airflow the team could not control, and tracking was error-prone where droplets were dense; about 80% of measured velocities fall within 20% error. Whether droplets carry germs from the board was not tested either.

Bread, candles, gum and soaking: no test behind them

Several popular tricks have no controlled study behind them in the published work cited here. Holding bread in the mouth, chewing gum and lighting a candle near the board are passed around as ways to absorb or burn off the fumes. No study cited here measured whether any of them changes how much irritant reaches the eyes. Cutting the root end last is another widespread tip with no published measurement in these sources.

Soaking a peeled onion in water before cutting is also often suggested. None of the studies above measured it, so its effect on droplets or on tears is unknown.

Can an onion be made not to sting?

In the lab, yes. Eady and colleagues used RNA interference, a technique that switches off a chosen gene, to silence LFS in onion. Their 2008 paper in Plant Physiology reports LFS activity reduced “by up to 1,544-fold,” and wounded onions produced significantly less of the tear-inducing compound.

A February 2008 release from Crop & Food Research, the New Zealand institute later succeeded by Plant & Food Research, said researchers were “unable to induce tearing by crushing” the model onions. It also described them as “in the developmental stages.” These were experimental plants, not a product on shop shelves. Onions sold under “tearless” labels are a separate matter; the 2008 papers say nothing about them.

Where the evidence runs out

No study cited here has counted tears in people cutting onions by hand, so the step from fewer droplets to fewer tears remains an inference. Whether a hand-held knife at kitchen speeds behaves like the falling guillotine blade is also untested. The studies used ordinary bulb onions and did not compare varieties, so they cannot say whether yellow, white or red onions, shallots or leeks differ in how much they sting.