Mold and Mycotoxins in Coffee, and Why We Send Ours to a Lab
A mom stood at our counter in Big Lake this spring holding her phone. She'd read something about mold in coffee, and she wanted to know whether the pot she'd been making her husband every morning for nine years was a problem.
A mom stood at our counter in Big Lake this spring holding her phone. She'd read something about mold in coffee, and she wanted to know whether the pot she'd been making her husband every morning for nine years was a problem.
She wasn't being silly. She'd read the same things everyone reads.
What are mycotoxins in coffee?
Coffee is a crop. It's picked wet, dried in the sun, bagged, shipped across an ocean and stored, and every one of those steps is a chance for moisture to sit where it shouldn't. When it does, mold can grow. Some molds leave behind compounds called mycotoxins, and the mycotoxin in coffee that people mean is almost always ochratoxin A.
That much is real. Mold in coffee is a quality-control problem, and it's been studied for decades.
Is there mold in coffee?
The number you've probably seen
If you've gone looking, you've run into a figure. Ninety-one point seven percent of samples contaminated. Ninety percent of store-bought coffee. Half of everything.
Those numbers come from real studies, and they get used in a way the studies don't support. The 91.7% figure traces back to a survey of green coffee, the raw unroasted beans before anyone has done anything to them, and it counted mold presence rather than toxin in a finished cup. Detectable and dangerous are different words. Almost every alarming version of this you'll read online quietly swaps one for the other.
Pooled surveillance of coffee people actually drink finds ochratoxin A detectable in roughly 53% of samples, often at parts per billion. Samples that exceed the legal safety limit are a small minority of those.
Where mold in coffee comes from
Mold needs moisture. Below about 0.65 water activity, which measures how much moisture is free for something to live on rather than how damp a thing feels in your hand, mold doesn't get going at all. The particular molds that produce ochratoxin A want conditions considerably wetter than that, and they want warmth with it. Coffee that's dried on schedule and held dry afterwards gives them very little to work with. Coffee that sat wet in a heap for three days gives them a great deal.
What predicts contamination better than anything else is defects. When researchers separated defective beans out of Brazilian arabica lots and tested them on their own, the black beans averaged 25.7 micrograms per kilogram and the sour beans 11.3, against low single digits for the lots they'd been sitting in. The problem concentrates in the beans you can see are wrong.
Every bag we roast carries a roast date on the front. See what's fresh this week.
Arabica vs robusta
Which one has more mycotoxins?
You'll read that robusta is the problem and arabica is safe. Surveillance does find ochratoxin A somewhat more often in robusta, roughly 37% of samples against 26% for arabica in one comparison. The gap is real, it's smaller than the framing suggests, and the reason for it has little to do with the bean itself.
Robusta is more often dried with the fruit still attached, grown at lower and more humid elevations, and strip-picked, which sweeps ripe, unripe and overripe cherries into one bag. Every one of those practices raises risk in arabica too. Brazilian arabica that's strip-picked and dried in the fruit behaves much the same way. How a coffee was grown and handled tells you more than which species is printed on the bag.
Does roasting destroy mycotoxins?
Roasting destroys most of them. Published reductions of ochratoxin A run from about half to essentially all of it, depending on how dark the roast goes and where the coffee came from, and darker roasts destroy more.
You'll still see it claimed that roasting does nothing to mycotoxins in coffee. That one's easy to check, and it doesn't hold up.
Is coffee mold dangerous?
What the limits say
The European Union caps ochratoxin A in roasted coffee at 3 micrograms per kilogram. The United States sets no limit at all, which sounds alarming until you look at why. The FDA tested 514 coffee samples between 2008 and 2022. One of them came back above even the internal threshold the agency uses to decide whether a sample needs a closer look.
We're not going to pretend the science sits still, either. European regulators moved to a more cautious framework in 2020, stepping away from a single safe-intake number and toward a method that pays closer attention to people at the high end of exposure. Coffee is one of several everyday foods that contribute, alongside grains, wine, dried fruit and spices, and in most studies it isn't the largest one. Anyone telling you the matter is entirely settled in either direction is ahead of the evidence.
Why we test for mycotoxins
We've found that the flavor arrives before the risk does. The Specialty Coffee Association's green grading standard counts a single fungus-damaged bean as a full primary defect, the same weight as a full black or a full sour bean, and enough on its own to keep a lot out of specialty grade. The tasting notes for it read like a warning label: fermented, moldy, earthy, dirty, phenolic.
Those are the same beans the researchers found carrying the highest readings. Buying to a specialty grade throws them out before anyone gets to the question of toxins, which is the plainest reason we buy the way we do. Coffee with a mold problem tastes wrong long before it becomes a health question.
We also send samples out. Ours go to IEH Analytical Laboratories in Seattle, an ISO 17025 accredited food-safety lab that runs ochratoxin A and aflatoxins by HPLC and LC-MS/MS. We name them because a testing claim you can't check isn't worth much.
If you want coffee bought to a specialty grade and sent out for testing, ours is here. Browse our coffee, or set up a subscription so it arrives fresh instead of sitting on a shelf.
What we're not going to tell you
We won't tell you that anything we sell is certified mold-free. No such certification exists. The phrase isn't defined by the FDA, the USDA or anyone else, which means any company can print it on a bag.
We won't tell you that your coffee is the reason you're tired.
A lot of people tell us Ember sits easier on their stomach, and we hear it often enough to pass it along without dressing it up as a medical finding. Air roasting lowers acidity. That part we can stand behind.
Keeping mold out at home
The useful moves here are unglamorous. Buy whole bean and grind it as you go, since pre-ground sits around longer. Keep it sealed somewhere dry and out of the fridge, where it picks up moisture every time the door opens. Drink it within a few weeks of the roast date, and buy from anyone willing to print that date on the bag.
Then stop worrying about it.
Back to the counter
She asked one more thing before she left. She wanted to know whether we were telling her all this because we wanted her to buy our coffee.
Partly, yes. We'd have told her the same thing if she'd walked out with nothing.
Ready to taste the difference? Shop our air-roasted coffee. Every bag is roasted in our own roastery in Big Lake, Minnesota, and dated the day it comes off the roaster.