Fermentation science
Kombucha microorganisms: which yeasts and bacteria do the work
How kombucha's yeasts and acetic acid bacteria divide the fermentation between them, why the pellicle forms at the surface, and which species claims remain unverified.
8 min read ·
What microorganisms are in kombucha?
Kombucha is fermented by two functional groups of microorganisms working in the same vessel: yeasts, which ferment sugar to ethanol and carbon dioxide, and acetic acid bacteria, which oxidise that ethanol to acetic acid. The FRMNT catalogue entry for kombucha describes the culture in exactly these functional terms — a living SCOBY carried batch to batch, with the fermentation proceeding sequentially rather than in a single step. It does not name individual species, and this article follows that restraint.
That restraint is deliberate rather than lazy. The published microbiology of kombucha is genuinely variable: the composition of a culture depends on where it came from, what it has been fed, the tea used, ambient temperature and how many generations it has been propagated. Genus-level names that recur in the wider literature — Komagataeibacter among the acetic acid bacteria, Brettanomyces, Saccharomyces and Zygosaccharomyces among the yeasts — are plausible and widely reported, but they are not present in the FRMNT grounding and are recorded here at low confidence, pending verification against a primary source. A named species on a jar in a domestic kitchen is a guess unless someone has sequenced it.
Lactic acid bacteria are sometimes reported as part of kombucha communities and sometimes not. Their presence appears to be inconsistent between cultures rather than definitional, and the FRMNT entry does not include them in its description of the mechanism. Treat the yeast-plus-acetic-acid-bacteria pairing as the load-bearing claim, and anything beyond it as UNKNOWN — pending verification.
How do the yeasts and bacteria work together?
They work in sequence, and the sequence is the whole point: the yeasts act first, converting the sucrose in the sweetened tea to ethanol and carbon dioxide, and the acetic acid bacteria then oxidise that ethanol to acetic acid. Neither group could produce kombucha alone. The yeasts cannot generate the sourness that characterises the finished drink; the acetic acid bacteria have no substrate to oxidise until the yeasts have made ethanol for them.
This is a cross-feeding relationship, and it explains why kombucha's flavour changes direction over the course of a ferment rather than simply intensifying. Early on, sugar is abundant and the drink is sweet with a faint yeasty character. As the yeasts work through the sugar, ethanol accumulates and the bacteria have more to convert; acidity rises and the sweetness recedes. FRMNT's entry frames the endpoint as a judgement rather than a measurement — taste from around day 7 and stop when the balance of sweet and sour suits you, typically somewhere in the 1–3 week window. Refrigeration slows further souring; it does not stop it.
Because ethanol is an intermediate rather than an end product, finished kombucha normally contains only a small residual amount — but not zero, and the quantity depends on how the two populations are balanced and how long the ferment ran. FRMNT does not publish a figure for residual alcohol in kombucha and this article will not supply one. Anyone who needs a number for regulatory or personal reasons should measure the specific batch rather than rely on a typical value.
Why does a SCOBY form at the surface of the liquid?
The pellicle forms at the surface because the bacterial half of the culture needs oxygen. Oxidising ethanol to acetic acid is an aerobic process, and the FRMNT entry places it explicitly at the liquid/air interface — which is where the acetic acid bacteria are most active and where the cellulose mat they produce accumulates. This is also why the vessel is covered with tightly-woven cloth secured at the rim: the ferment requires air exchange, but it must be protected from insects, and it must never be sealed during the open ferment.
A useful correction to a common assumption: the pellicle is a by-product of the culture, not the culture itself. The majority of the microbial population — both the yeasts and a large share of the bacteria — lives in the liquid. This is why the starter tea carried forward matters as much as the disc of cellulose, and why a new pellicle typically forms on each batch even when the old one is buried mid-jar or discarded. The claim that acidic starter liquid alone can carry a culture forward is widely reported in home-fermentation practice and is consistent with the mechanism, but the FRMNT entry lists both a SCOBY and starter tea in its ingredients, so treat the liquid-only route as an inference rather than as catalogue guidance.
Two practical consequences follow from the aerobic mechanism. Cooling the sweet tea to room temperature before adding the culture is not fussiness — hot liquid kills the organisms you are relying on. And a wide-mouthed jar gives a larger interface than a narrow one, which changes the rate at which ethanol is oxidised. FRMNT specifies a clean wide-mouthed vessel without prescribing dimensions; the mechanism explains why the shape is worth thinking about at all.
Are the microbes in jun, water kefir and ginger beer the same as kombucha's?
No — they are related in strategy but distinct as cultures, and only jun is described in similar SCOBY terms. Jun uses green tea sweetened with raw honey in place of refined cane sugar, ferments at ambient temperature in a loosely covered vessel, and is propagated from a living jun SCOBY plus acidic back-slop liquid. Its primary ferment is commonly given as roughly 3–7 days, an illustrative range. The frequent claim that jun is a historically or functionally distinct culture — because honey carries its own yeast and enzyme content — is noted in the FRMNT catalogue as commonly cited but not independently confirmed, and the related claim that jun ferments faster than standard kombucha is likewise unsourced. Both are recorded here as unverified.
Water kefir, or tibicos, runs on a visibly different starter: water-kefir grains, a living culture propagated in sugar water, typically for roughly 1–3 days at ambient temperature. The grains are a self-perpetuating matrix rather than a surface pellicle, and they are the same culture carried forward batch to batch rather than a new structure grown each time. Ginger beer is different again, in that its starter is spontaneously cultivated: the ginger bug is built over roughly 3–7 days of daily feeding, drawing on microbiota carried on the fresh ginger root itself, and only becomes a maintained culture once established.
The common thread is that all four are ambient, open-fermented, sugar-driven beverages whose primary stage leans aerobic and whose optional secondary carbonation stage happens in a sealed vessel for a limited, closely monitored period before refrigeration. What differs is the identity and the physical form of the culture. Substituting one starter for another is a change of organism, not a change of recipe, and the fermentation behaviour and timing should be expected to change with it.
What does kombucha's microbiology mean for safety?
The acidity generated by the culture is the principal protective factor, so any sign that acidification has failed is a reason to discard the batch rather than to taste it. That includes visible mould — fuzzy, dry-looking or coloured growth, as distinct from the smooth new pellicle — an unpleasant, solventy or rotten smell, or a batch that has simply not soured over the expected period. Where the evidence is unclear, discard: a litre of sweet tea is not worth the ambiguity.
The starter tea is the mechanism that makes the early hours of a batch defensible. FRMNT records it as a safety control precisely because it drops the pH of the fresh sweet tea immediately, before the transferred culture has had time to establish itself and produce acid of its own. Sweetened tea with no acidic starter is a hospitable medium for organisms other than the ones you want. The catalogue does not publish a target pH value for kombucha, and this article does not supply one; for context, pH 4.6 is the general reference threshold below which Clostridium botulinum will not grow, but that is a broader food-safety benchmark rather than a kombucha specification, and a home pH target for this ferment is UNKNOWN — pending verification.
Two further points follow from the mechanism rather than from any single figure. Because the ferment needs air exchange but must exclude insects, the cloth cover is doing safety work as well as fermentation work; a sealed vessel during the open stage removes the oxygen the acetic acid bacteria need and changes the ferment into something else. And because yeasts continue producing carbon dioxide during any sealed secondary stage, that stage should be short, monitored and followed by refrigeration. Finished kombucha is best treated as a living product with a short refrigerated life, not as a shelf-stable one. None of the above is medical advice; it concerns the handling of the ferment only.