The baking soda soak is one of the few kitchen habits with a real peer reviewed study behind it. The study says it works. It also says something most people repeating the advice have never read: it was never a study about germs.
Baking soda removes pesticide residue well. Against bacteria it is close to the weakest thing under your sink. Against anything inside the grain no wash works at all: glyphosate after a pre-harvest spray stays in the seed coat for over a year of storage. How much that residue is worth worrying about is covered in our look at the Dirty Dozen list.
What the famous apple study actually did
Yang and He at the University of Massachusetts published it in the Journal of Agricultural and Food Chemistry in 2017. They took organic Gala apples (organic farming works with pesticides too, just from a different list) and deliberately dosed them with two pesticides: thiabendazole, a fungicide, and phosmet, an insecticide. Then they washed the apples three ways.
- Plain tap water.
- A 1 percent baking soda solution, roughly two teaspoons of soda in a liter of water.
- The EPA approved bleach solution used commercially on fruit after harvest.
Baking soda beat both of the others. Twelve minutes in the solution removed 80 percent of the thiabendazole from the surface. Fifteen minutes removed 96 percent of the phosmet.
The mechanism is the interesting part. Water dilutes and carries residue away. An alkaline solution chemically degrades these pesticides, breaking them into other compounds, so the soda is not rinsing so much as dismantling. That is why it outperformed the commercial bleach wash, which is formulated to kill microbes rather than to break down organophosphates.
The number that never makes the headline
The same paper reports where the rest of the pesticide went. Twenty percent of the applied thiabendazole and 4.4 percent of the phosmet had moved through the skin into the flesh of the apple.
Nothing you do at the sink touches that. Thiabendazole is a systemic fungicide, built to travel inside plant tissue, which is exactly what makes it useful in the orchard and unreachable in the kitchen. Peeling gets some of it, and costs you most of the fiber and a good share of the vitamin C.
So the honest summary of the strongest pro soda evidence is this: on the surface it works better than anything else you have, and the surface is not the whole problem. Whatever is dissolved inside the tissue, such as nitrate, no soak will reach.
Now the other question: does it do anything to bacteria?
Canada’s National Collaborating Centre for Environmental Health compiled the log reduction data for household agents in a 2014 review. Sodium bicarbonate comes out poorly, and it comes out poorly next to vinegar tested under identical conditions.
| Organism, suspension test | Baking soda, 8%, 5 min | Vinegar, 5% acetic acid |
|---|---|---|
| E. coli O157:H7 | 0.7 log | 2.4 log in 5 min |
| Staphylococcus aureus | 0.5 log | 0.3 to 2.3 log in 5 min |
| Salmonella choleraesuis | 2.3 log | over 6 log in 30 sec |
| Pseudomonas aeruginosa | 1.1 log | over 5.8 log in 30 sec |
Turning up the concentration does not rescue it. In a separate test, sodium bicarbonate solutions at 11, 33 and even 50 percent all produced less than 1 log reduction in E. coli O157:H7. Six times the soda, same result.
On food the picture is the same. Washing raw chicken breast for one minute with agitation gave 1.0 log with a 10 percent soda solution and 2.2 log with 5 percent vinegar. The review’s verdict is that baking soda is generally ineffective against E. coli, Pseudomonas aeruginosa, S. aureus and Salmonella. Why raw chicken is better left unwashed altogether, whatever the liquid, is covered in should you wash chicken.
There is a chemical reason it fails
Acids kill bacteria by driving protons across the cell membrane. Baking soda works in the other direction, and it does not go far enough to matter. A sodium bicarbonate solution in water equilibrates at a maximum pH near 8.34. Plenty of foodborne bacteria grow happily up to pH 9 or 10.
Baking soda never reaches a pH that inconveniences them. It is a mild alkali and a mild abrasive, and those are cleaning properties rather than antimicrobial ones.
One exception worth naming
Against viruses the result flips. Malik and Goyal reported in 2006 that a 5 percent sodium bicarbonate solution cut feline calicivirus, the standard laboratory stand in for norovirus, by 4 log in one minute on stainless steel disks.
Norovirus causes more foodborne illness than any bacterium, so that is not a small finding. Two cautions before anyone builds a routine on it. The test surface was steel, not a leaf or a peel, and a surrogate virus is not the real one. Vinegar did essentially nothing to viruses in the same body of work, cutting poliovirus by 0.32 log against baking soda’s 0.42, which is to say neither did anything at all.
Two questions, two answers
| What you are worried about | Is baking soda the right tool? |
|---|---|
| Pesticide residue on the surface | Yes, better than water or bleach, at 12 to 15 minutes |
| Pesticide absorbed into the flesh | No, and neither is anything else |
| Salmonella, E. coli, Listeria | No, it is among the weakest options tested |
| Norovirus on a hard surface | Promising, but not demonstrated on food |
| Dirt, grit, wax | Yes, the mild abrasion helps |
What I would actually do
If pesticide residue is your concern, the soak is defensible, and you have to give it the full twelve to fifteen minutes. A thirty second swirl is not the tested protocol and there is no evidence it does anything. Rinse afterward, because you are washing off a solution that now contains degraded pesticide.
If your concern is food poisoning, use something else. Firm produce responds to a brush under running water. Leafy greens respond to agitation and a spinner. Neither is a kill step, and for anyone pregnant, over 65 or immunocompromised the only reliable answer on high risk items is heat.
One safety note. Do not improvise by combining soda with vinegar. They neutralize each other into salt water and carbon dioxide, which is chemistry that makes a satisfying fizz and destroys whatever either one was contributing.
Bottom line
- The evidence behind the baking soda soak is about pesticides, and on that question it is genuinely the best household option tested.
- It needs 12 to 15 minutes. That is the protocol that produced 80 and 96 percent removal.
- Twenty percent of the thiabendazole had already crossed into the apple flesh, where no wash reaches.
- Against bacteria it is weak: 0.7 log on E. coli O157:H7, and raising the concentration to 50 percent changed nothing.
- Its pH tops out near 8.34, below the level that troubles foodborne bacteria.
- The one promising result is against a norovirus surrogate on steel, which has not been reproduced on produce.
The neighboring questions are covered separately: what the vinegar numbers really mean, where the famous 5 log figure comes apart on inspection, and whether commercial produce sprays beat tap water. The wider picture is in our food safety guide for greens and sprouts. Why an industrial line works at entirely different doses, and what it does with them, is covered in wash water sanitizers.
Sources
- Yang T, Doherty J, Zhao B, Kinchla AJ, Clark JM, He L. Effectiveness of commercial and homemade washing agents in removing pesticide residues on and in apples. J Agric Food Chem. 2017;65(44):9744-9752.
- Fong D, Gaulin C, Le ML, Shum M. Effectiveness of alternative antimicrobial agents. National Collaborating Centre for Environmental Health, August 2014, including Appendix A.
- Rutala WA et al., 2000; McKee LH et al., J Food Prot 2005;68(3):534-537; Malik YS, Goyal SM, 2006, as compiled in the NCCEH review appendix.
- US Food and Drug Administration, Selecting and Serving Produce Safely.
