American moist snuff is fermented because fermentation builds flavour, and the bacteria that drive it also convert nitrate into nitrite, which turns tobacco alkaloids into nitrosamines; Swedish snus is heat-treated instead because heat suppresses that bacterial activity. FDA sets out both halves of that in its own voice. But the popular story overstates it: heat treatment is one of about five changes the Swedish industry made, the leaf itself may matter more, and it is not pasteurisation in the food-science sense.
This is the single most repeated claim in snus writing, and most of it is folklore. I went to the regulators’ own documents, and the chemistry is exactly as advertised — while the causal story is noticeably messier, and more interesting, than “pasteurised, so no nitrosamines”.
What fermentation actually does
The clearest regulatory statement of the mechanism is in FDA’s 2017 proposed product standard for NNN in smokeless tobacco. FDA describes the chain in its own words:
“The primary nitrosating agent is nitrite. Reduction of nitrate by bacteria such as halotolerant micrococci, Coryneforms, and halophilic rods during the fermentation process is the primary source of nitrite in smokeless tobacco products.”
And the step before it:
“NNN is formed either by the nitrosation of nicotine with the loss of a methyl group or by nitrosation of nornicotine, primarily during the curing of tobacco.”
Read in order, that is the whole story. Tobacco leaf contains nitrate. Bacteria reduce nitrate to nitrite. Nitrite nitrosates the alkaloids in the leaf. The products are NNN and NNK, the two tobacco-specific nitrosamines that matter most. FDA puts the consequence plainly: “Tobacco and smokeless tobacco products with low nitrite concentrations have low levels of NNN, while products high in nitrite contain higher concentrations of NNN.”
Then the sentence that names the American process:
“Fermentation is commonly used in the production of U.S. smokeless tobacco products. Fermentation imparts flavor and contributes to higher nitrite and NNN levels.”
Note what that says. Fermentation is not an accident or a legacy inefficiency. It is there for the flavour, and the nitrosamines come with it. That trade is the reason the two industries diverged.
What heat treatment does instead — and at what temperature
FDA describes the Swedish alternative in the same document, with numbers:
“One of the methods used to limit microbial activity is pasteurization. In this process, the leaf tobacco is ground and subjected to heat treatment. The heating is achieved by combining the tobacco with water and salt, placed in closed process blenders, and using steam to achieve temperatures up to 80 to 100 °C for several hours.”
FDA’s own causal claim is narrower than the one you usually read. It says heat treatment limits microbial activity. It does not say heat prevents nitrosation. The chain runs heat → fewer bacteria → less nitrate reduction → less nitrite → less nitrosation, and every link is worth keeping in that order, because the folklore version collapses it into a single step that the regulator does not claim.
FDA’s product descriptions line the two up neatly. American moist snuff: “It is typically made of fire-cured or air-cured tobacco that has been finely ground or shredded and fermented.” Swedish snus: “It typically consists of low-nitrosamine tobacco that has been air-cured, moistened, ground, and heat treated.” Fermentation appears in the first list and not the second. And fermentation is the American default rather than a moist-snuff quirk — FDA describes dry snuff as “fire-cured, fermented, and finely ground”, and chewing tobacco as “fermented or aged”.
It is not pasteurisation, and both WHO bodies say so carefully
Everyone calls it pasteurisation, including FDA. The two international bodies that have looked at it both hedge the word. IARC Monograph 89:
“the tobacco is processed in a heated closed system that resembles pasteurization of milk, which eliminates bacteria that may be indirectly responsible for the formation of the nitrosamines”
“Resembles”. And “indirectly responsible” — IARC is being precise about the causal distance. The WHO Study Group on Tobacco Product Regulation calls it “a pasteurization-like process”.
The hedging is justified. Milk pasteurisation has a legal definition, and in the United States 21 CFR 1240.61 sets it out as heating every particle of milk to a tabulated temperature for a corresponding time: 30 minutes at the low end of the table, 15 seconds at 72 °C, and hundredths of a second at 100 °C. Snus heat treatment runs at 80–100 °C for several hours. In time-at-temperature terms that is orders of magnitude more severe than any legal milk pasteurisation at the same temperature. The shared property is the goal — reduce the microbial load without sterilising — not the process. That comparison is mine, but both halves of it are quoted regulator text.
Pasteurisation is one lever of five, and probably not the biggest
This is where the popular story goes furthest wrong. WHO lists the Swedish changes in a single sentence, and heat treatment is one item on it:
“The procedures include a short ageing process, use of tobacco with a low nitrate content, a pasteurization-like process that destroys bacteria implicated in the formation of nitrosamines and changes in the methods for curing tobacco.”
WHO adds a fifth: “Refrigeration of the product also helps limit the formation of nitrosamines during storage.” That one exists because, as WHO puts it, “The concentrations of TSNA increase in smokeless tobacco during storage at room temperature due to microbial action.” The cold chain in Swedish snus retail is not about freshness in the ordinary sense.
IARC puts the leaf first and credits it independently:
“Instead of the dark fire-cured tobacco commonly used in US snuff, Swedish Match uses tobacco with a low nitrate content, which itself reduces TSNA levels.”
Think about the chemistry for a moment and that ordering makes sense. If the nitrate is not in the leaf, there is very little for bacteria to reduce to nitrite in the first place, and the heat step is downstream of a problem that has already been made small. FDA’s own description of the Swedish input is “low-nitrosamine tobacco”. There is a curing contrast too: American moist and dry snuff are typically fire-cured, over wood smoke, while Swedish snus is air-cured.
My read: an article that puts pasteurisation at the centre of this story is naming the most visible lever rather than the most load-bearing one. Nobody has run the experiment that would separate them — I could find no study comparing a fermented and a heat-treated product made from the same leaf, which is the only design that would isolate the process from the raw material.
What Swedish law actually requires
The received version is that snus is legally a food in Sweden and so must be made to food standards. The regulator is more careful than that. Livsmedelsverket’s own guidance states that snus, snus-like products and chewing tobacco “är inte livsmedel enligt definitionen av livsmedel” — are not food under the EU definition — and that they have been “jämställts med livsmedel som en lagteknisk lösning”, equated with food as a legal-technical solution, in order to apply selected parts of food legislation to them.
What those selected parts do is more useful than the label. Sweden sets binding limits on contaminants, and Livsmedelsverket notes that “När det gäller snus och tuggtobak finns även gränsvärden för tobaksspecifika nitrosaminer” — for snus and chewing tobacco there are also limits for tobacco-specific nitrosamines. The current limit is 2 mg/kg dry weight for NNN and NNK combined. On top of that sit selected EU food-hygiene rules, a HACCP duty where necessary, registration of manufacturing premises and inspection by a food authority.
I checked whether Swedish law requires heat treatment. It does not: there is no pasteurisation requirement, no heat-treatment requirement and no microbiological limit anywhere in that regime. What the law creates is an enforceable nitrosamine ceiling and an enforceable hygiene duty. Heat treatment is how the industry meets them. That is a weaker claim than “Swedish law requires pasteurisation”, and it is the true one.
The same goes for GothiaTek, Swedish Match’s own standard. It sets constituent limits — for NNN and NNK combined, a limit of 0.95 mg/kg “as is”, against a stated 2025 average of 0.35 — and it says nothing about the manufacturing process at all. The word usually attached to it is “voluntary”, and that word is worth sourcing properly: it is FDA’s, describing “the GothiaTek voluntary standard”, not the company’s own.
What the numbers show
Units are where this subject goes wrong. Dry weight runs roughly double “as is” weight for a product around half water, so a figure quoted on the wrong basis is out by a factor of two.
| Sample | Total TSNA | Basis | Source and funding |
|---|---|---|---|
| US traditional moist snuff brands | 7.42 µg/g | dry | Stepanov et al. 2008, NIH-funded, authors declare no conflicts |
| Skoal Dry | 4.54 µg/g | dry | same |
| New US “snus” products (Taboka, Marlboro Snus, Camel Snus) | 1.97 µg/g | dry | same |
| US market, moist snuff, NNN alone (2015 sampling) | mean 3.01 µg/g | dry | FDA, NNN proposed rule |
| GothiaTek limit for NNN + NNK | 0.95 mg/kg | “as is” | Swedish Match — manufacturer’s own standard |
| Swedish statutory limit, NNN + NNK | 2 mg/kg | dry | Swedish food regulation — binding |
| WHO recommended limit, NNN + NNK | 2 µg/g | dry | WHO Study Group, 2009 |
The widely repeated “five times lower” figure for Swedish snus comes from an FDA comparison table in which the snus column is drawn from Swedish Match’s own submitted data. The cleanest independent comparison I found — same laboratory, NIH funding, no declared conflicts — puts the gap nearer two and a half times. I would quote the smaller, cleaner number and describe the other for what it is.
There is a regulatory punchline. WHO recommended a 2 µg/g dry-weight ceiling for NNN plus NNK in 2009 and observed that no US moist snuff then on the market met it. FDA proposed a 1.0 µg/g NNN standard in January 2017, saying it “would bring the NNN level in U.S. smokeless tobacco products in line with those of Swedish snus”. Nine years later it is still a proposal. The gap was identified, quantified and proposed against, and nothing was enacted.
The American industry is optimising fermentation, not abandoning it
If heat treatment were simply better, you would expect American makers to adopt it. A patent granted in February 2024 to American Snuff Co — the Grizzly and Kodiak maker — suggests otherwise. It describes the conventional process in the company’s own words: tobacco “packed for multiple years” in hogsheads, then moistened and stored warm in large tubs so that it is “exposed to an environment highly conducive to bacterial growth, promoting fermentation”, with the mass turned repeatedly to stop the fermentation running away. Then:
“This fermentation process is typically allowed to continue for 60 to 90 days, and in some cases, until the manufacturer is satisfied that a desirable flavor level has been achieved.”
Sixty to ninety days, and the stopping condition is flavour. What the patent proposes is moving the fermentation into the sealed consumer can and deleting the tub stage — relocating fermentation, not replacing it with heat. Two caveats: a patent application is an account of practice and intention, not evidence about what is in any can on a shelf, and a granted patent does not mean the method is in use.
For the historical scale of this, IARC’s description of older American snuff manufacture records that the chopped leaves “underwent further fermentation for about 2 months”. The duration has not changed much in a century.
What FDA authorised, and what it did not
Swedish snus is the only smokeless product to hold a US modified-risk authorisation, and the scope is a closed list. FDA’s 2019 order authorises exactly one sentence of modified-risk information:
“Using General Snus instead of cigarettes puts you at a lower risk of mouth cancer, heart disease, lung cancer, stroke, emphysema, and chronic bronchitis.”
Six named conditions, conditional on use instead of cigarettes. Nothing in that order says these products are safe, and FDA’s technical review is explicit about the limit:
“since, according to the available toxicological evidence, there is no established threshold level for NNN carcinogenicity, the products pose an increased risk of mouth cancer compared to non-use”
FDA declined to remove the mouth-cancer warning from these products, and wrote that “long-term use of Swedish snus is not without health risks”. In the NNN rule it added that a product meeting the proposed standard is still not being described as “acceptable” or “negligible” from a public health perspective. Lower is not low.
The honest limits
Three things are worth knowing about the evidence behind this page.
First, no regulator appears to have independently characterised the Swedish manufacturing process. FDA’s pasteurisation passage, IARC’s “resembles pasteurization of milk” and WHO’s “pasteurization-like process” look like three independent official descriptions; each of those documents cites the manufacturer’s own material for it. The chemistry is independently established many times over. The process description is not.
Second, the manufacturer does not make the strong claim the internet attributes to it. Swedish Match’s published account of heat treatment describes it as aiming to improve taste and reduce microbial activity for shelf life — nitrosamines unmentioned. “Pasteurised so no nitrosamines form” is not the owner’s claim about its own product.
Third, the measured decline in Swedish snus nitrosamines over the 1980s and 1990s is well documented by Sweden’s own food authority laboratory. What that laboratory measured was the change; it did not identify which process change caused it. The attribution to pasteurisation comes from the industry.
Frequently asked questions
Is Swedish snus really pasteurised? It is heat-treated, at 80–100 °C for several hours, in a way IARC describes as resembling milk pasteurisation. It does not meet the legal definition of pasteurisation for any food.
Does pasteurisation remove nitrosamines? No. It reduces the bacterial activity that generates the nitrite that forms them. Nitrosamines already present in the leaf are not destroyed by it, and Swedish snus still contains NNN and NNK.
Why does American snuff get fermented at all, if it raises nitrosamines? For flavour. FDA says so directly: “Fermentation imparts flavor and contributes to higher nitrite and NNN levels.”
So is Swedish snus safer? It carries lower nitrosamine levels, and FDA has authorised a narrow comparative claim against cigarettes. FDA also says there is no established threshold for NNN carcinogenicity and that these products pose an increased risk of mouth cancer compared with not using them. No nicotine product is safe.
Why is snus sold refrigerated? Because nitrosamine levels rise during storage at room temperature through microbial action — WHO recommends refrigerated storage and sale for smokeless tobacco for exactly that reason.
Do nicotine pouches have this problem? Tobacco-free pouches contain no tobacco leaf, so the nitrate-to-nitrite-to-nitrosamine route does not apply in the same way. That is a different subject with its own evidence base.
If you want the process itself rather than the reasoning behind it, see how Swedish snus is made, from leaf to can and the GothiaTek standard explained. For the basics, start with what is snus, and for how the two product families compare on nicotine, nicotine content in snus versus nicotine pouches. The packaging side of the same shelf is covered in how many pouches are in a can, and the legal strength ceilings in strength caps by country.
Sources
- FDA, Tobacco Product Standard for N-Nitrosonornicotine Level in Finished Smokeless Tobacco Products, proposed rule, 82 FR 8004, 23 January 2017 — the nitrosation chain, the fermentation and heat-treatment passages, the product descriptions and the proposed 1.0 µg/g limit. Still a proposal; no final rule.
- IARC, Monograph 89, Smokeless Tobacco and Some Tobacco-specific N-Nitrosamines (2007) — “resembles pasteurization of milk”; low-nitrate leaf; the two-month American fermentation.
- WHO Study Group on Tobacco Product Regulation, third report, WHO Technical Report Series 955 (2009) — the five procedures, the storage warning, and the recommended 2 µg/g dry-weight limit for NNN plus NNK.
- Livsmedelsverket, Kontrollwiki article 887, snus, snusliknande produkter och tuggtobak — snus is not food but is equated with it; the contaminant and nitrosamine limits; the hygiene and HACCP duties.
- 21 CFR 1240.61 — the legal definition of milk pasteurisation and its time-temperature table.
- FDA, modified risk tobacco product order, General Snus, 22 October 2019, and the accompanying technical project lead review — the authorised claim and the mouth-cancer finding.
- Stepanov I, Jensen J, Hatsukami D, Hecht SS, “New and traditional smokeless tobacco: comparison of toxicant and carcinogen levels”, Nicotine & Tobacco Research 2008;10(12):1773–82, PMID 19023828 — NIH-funded, authors declare no conflicts.
- US Patent 11,903,406 B2, “Method for fermenting tobacco”, American Snuff Co LLC, granted 20 February 2024 — the conventional American process in the manufacturer’s own words.
- Swedish Match, the GOTHIATEK standard — manufacturer’s own constituent limits, stated “as is”.
Last verified 19 October 2026.
