Matcha and Pesticides: Why Eating the Leaf Is Nothing Like Drinking the Tea

Tea Science｜Author：祥興茶行｜Published：2026-07-15

Brewing tea acts as a natural filter that leaves most pesticide residues behind in the spent leaf — but matcha is the whole leaf, ground up and eaten. Here's what the research says about that missing filter, the fat effect in baking, and why the real risk lies in sourcing, not in matcha itself.

Matcha lovers, this one's for you. That jade-green pound cake in the pastry case, the matcha Basque cheesecake, the latte shaken into a full cup of foam — are they actually safe to eat? A thought has crossed many minds: don't we keep hearing about pesticide residues on tea leaves? So if you grind the entire leaf into powder and fold it straight into a cake, aren't you eating the pesticides along with it?

It's a genuinely good question, and the answer really is different from the one for drinking tea. Let's put the conclusion up front: the problem is not that "matcha is toxic." The problem is that eating the powder and drinking the brew are two completely different routes of exposure.

This piece picks up where our earlier column "Should You Really Discard the First Infusion?" left off. That article established a key premise: most pesticides lean lipophilic — they dissolve in fat, not water — so when you brew tea, the bulk of the residue never makes it into the liquor and stays behind in the leaf. Exposure from drinking tea is therefore genuinely low. We won't retell that story here. Instead, we turn the camera toward a new route entirely: what happens when you don't brew the leaf, but swallow it whole.

1. Eating the Powder ≠ Drinking the Tea: One Batch of Leaves, Two Fates

First, a clarification: matcha is simply green tea leaf ground into a fine powder. Take the same batch of fresh leaf, and you can either brew it and discard the spent leaves — or mill the whole thing and eat every last particle, stems, leaf and all. The amount of residue your body actually encounters differs enormously between the two.

The difference comes down to one word: the denominator.

Brewing is a natural filter When you steep tea, only a small fraction of the residue dissolves into the liquor; most of it stays locked in the leaf and gets thrown out with the spent leaves. A 2025 review of food processing concluded that only about 10 to 30 percent of tea-leaf residues transfer into the brew, while roughly 50 to 70 percent remains in the spent leaf [5]. What you drink is the small dissolved fraction. The spent leaf is the part that got filtered out.

2. The Vanishing Denominator: Matcha Takes the Filter Apart

When the European Food Safety Authority (EFSA) assessed propargite, a lipophilic acaricide, it produced a strikingly intuitive number. The maximum residue limit (MRL) for tea is set on the dry leaf, not on the brew — and the median processing factor from dry leaf to infusion for this pesticide was only about 0.02. In other words, someone drinking the tea encounters roughly 2 percent of the residue on the dry leaf [3].

Here's the catch: matcha is eaten whole. There is no "brew, then discard the leaf" step. The filter that slashed exposure to 2 percent has been dismantled entirely. Same leaf, same residue limit — the tea drinker gets about 2 percent, the whole-leaf eater gets close to 100 percent. In the propargite example, that's a gap of roughly an order of magnitude (about 50-fold in this case). (One caveat: 0.02 is the median value for this particular pesticide. Transfer rates vary widely with water solubility, so don't treat it as a universal constant.)

So which pesticides are the ones that "won't brew out"? A study devoted specifically to whole-leaf consumption of green tea and matcha ran the steeping experiment directly: representative pesticides were brewed in 80°C water, and essentially only the water-soluble organophosphates made it into the liquor (mostly monocrotophos, with a trace of mevinphos), peaking at about 37 μg/kg. The lipophilic allethrin (a pyrethroid) and several others were essentially undetectable in the brew — recorded as N/F, not found [1].

This dovetails neatly with the first-infusion article: lipophilic pesticides "won't rinse off and won't reach your cup" precisely because they stay put in the leaf. Good news for tea drinkers — but a critical reversal of risk for powder eaters. Because those residues remain in the leaf, eating the whole leaf carries every bit of them along.

〔圖〕One tea leaf, two fates: brewing versus eating whole

3. The Fat Effect: Butter and Cream Cheese Turn "Won't Wash Off" Into "Readily Absorbed"

There's another layer to this. When matcha goes into baking, it usually gets blended with butter, cream cheese, egg yolks — high-fat ingredients. And fat happens to be the best possible escort for getting lipophilic pesticides absorbed into the body.

An in-vitro study simulating human digestion measured bioaccessibility — the fraction of a compound available for uptake in the small intestine. Highly lipophilic chlorpyrifos was only about 2.8 percent bioaccessible in plain water, but raising the fat content of the matrix to 8 percent sent it climbing all the way to about 70.7 percent — a roughly 25-fold amplification. Moderately lipophilic parathion likewise rose from about 16 percent to about 72 percent. Water-soluble bendiocarb, by contrast, was well absorbed to begin with (about 93 percent), and adding fat made almost no difference [2].

In plain terms: lipophilic pesticides need the mixed micelles formed during fat digestion before they can dissolve and be absorbed. Fold tea powder into a buttery cake batter, and you are effectively handing over the key — converting pesticides your body could barely take up into ones it absorbs with ease. This is a mechanistically sound inference (it's a test-tube experiment, not a real cake, and no one has directly measured a matcha cake yet), but the direction is unmistakable.

💡 The point: for tea drinkers, lipophilic pesticides face two lines of defense — they don't dissolve out, and they aren't well absorbed. Eat the whole leaf with fat, and both defenses are bypassed at once.

What about high oven heat — doesn't baking burn the pesticides off? The answer: some go down, some go up; it depends on the compound. The same processing review examined bread baking (roughly 170–250°C) and found that some pesticides, such as chlorpyrifos, degrade under heat, dropping by 30 to 70 percent — while others, such as imidacloprid, actually become concentrated as moisture evaporates, rising by 10 to 80 percent [5]. So both "the heat burns it all off" and "baking makes it more toxic" are oversimplifications. And to date, no one has actually measured degradation rates for matcha powder inside a pound cake or a Basque cheesecake — this part remains a mechanistic extrapolation.

4. A Level-Headed Close: The Risk Lies in Sourcing and Grade, Not in Matcha Itself

Before you push that slice of matcha cake away, add the dose context back in. The picture balances out considerably.

The paper that built the whole-leaf consumption model for matcha actually tested 39 green tea samples (using green tea as a stand-in for matcha, rather than measuring commercial matcha powder directly), and assumed a daily intake of about 6.5 grams of matcha — roughly the amount in "one drink plus one pastry." Under this most conservative eat-every-leaf assumption, 13 of the 15 growing regions still came in with an aggregate hazard index (AHI) within the safety threshold of 1. Only a handful of regions exceeded 1, and almost entirely on the strength of a single pyrethroid (the paper's own naming of that pyrethroid is inconsistent in places, so we won't be more definitive than the authors themselves) [1].

1 (above threshold)"],"datasets":[{"label":"Growing regions (whole-leaf model, 15 total)","data":[13,2],"backgroundColor":["#3e6b4e","#9a4a2e"]}]},"options":{"plugins":{"legend":{"display":false},"title":{"display":true,"text":"Under the most conservative whole-leaf assumption, 13 of 15 regions stay within the safety threshold [1]"}},"scales":{"y":{"beginAtZero":true,"title":{"display":true,"text":"Number of growing regions"}}}}}'>

Another study was even more direct: even using the most conservative "dry leaf eaten whole" baseline, the chronic hazard quotients for water-soluble neonicotinoid pesticides all landed far below 1 — meaning that exposure from compliant tea mostly falls well within the margin of safety [4].

Three research gaps worth stating honestly ① There is no peer-reviewed survey of pesticide-violation rates in commercial matcha products (culinary versus ceremonial grade), which is why this article quotes no "X percent of matcha exceeds limits" figure; ② Regulators (EFSA, Japan, Taiwan) have not yet issued a dedicated exposure position on "tea products consumed as whole leaf"; ③ How matcha powder degrades and gets absorbed in real baked pastries has never been measured. These are blanks, not conclusions.

So the question that actually deserves your attention is not "is matcha toxic," but "where did this tin of powder come from, and what grade is it?" Choose a culinary-grade matcha backed by third-party test reports for pesticide residues and heavy metals, keep an eye on your total daily intake, and the risk stays within a manageable range.

The Takeaway, in One Cup

  1. Eating the powder ≠ drinking the tea: brewing acts as a natural filter, carrying over only 10 to 30 percent of residues; eating the whole leaf dismantles the filter, and exposure approaches the full dose.
  2. Fat amplifies: butter and cream cheese make lipophilic pesticides — the ones that "won't wash off, won't absorb" — readily absorbable, while oven heat cuts some compounds and concentrates others, depending on the pesticide.
  3. Don't panic — check the source: compliant tea mostly sits within the margin of safety. The risk lies in sourcing, grade, and processing, not in matcha itself. Pick a culinary-grade matcha with third-party testing, and enjoy that slice of cake in peace.

Further reading: "[Should You Really Discard the First Infusion?](/articles?slug=rinse-tea-first-infusion)" — why pesticide exposure from drinking tea is lower than you'd think.

References

  1. Tang, C., Ren, L., Wang, Y., Long, W., Wang, S., Wei, L., Lan, W., Chen, H., Fu, H.-Y., & She, Y. (2025). Detection and safety evaluation of pesticide residues in representative floral and foliage foods: Hangbaiju and green tea. _Microchemical Journal_, 220, 116414. DOI: 10.1016/j.microc.2025.116414
  2. Zhang, R., Zhang, Z., Li, R., Tan, Y., Lv, S., & McClements, D. J. (2020). Impact of Pesticide Type and Emulsion Fat Content on the Bioaccessibility of Pesticides in Natural Products. _Molecules_, 25(6), 1466. DOI: 10.3390/molecules25061466
  3. European Food Safety Authority. (2018). Setting of maximum residue limits for propargite in citrus fruits and tea. _EFSA Journal_, 16(2), e05193. DOI: 10.2903/j.efsa.2018.5193
  4. Fan, Y., Jin, H., Chen, J., Lin, K., Zhu, L., Guo, Y., Ji, J., & Chen, X. (2025). Neonicotinoid Residues in Tea Products from China: Contamination Patterns and Implications for Human Exposure. _Toxics_, 13(7), 550. DOI: 10.3390/toxics13070550
  5. Xiao, O., Li, M., Chen, J., Dai, X., & Kong, Z. (2025). A Comprehensive Analysis of the Fates of Pesticide Residues during Food Processing. _Journal of Agricultural and Food Chemistry_, 73, 18525–18544. DOI: 10.1021/acs.jafc.5c05285
  6. Gao, W., et al. (2019). Transfer of pesticide residues from tea to infusion. _Journal of Agricultural and Food Chemistry_. DOI: 10.1021/acs.jafc.8b04908