Fault
Geranium taint
A sharp, unmistakable crushed-geranium-leaf smell produced when lactic acid bacteria metabolise sorbic acid added as a preservative.
Also called geranium off-odour, sorbate taint.
- Severity
- Serious — the batch may not be salvageable
- Where it starts
- Stabilisation, Blending
- Can it be fixed?
- Yes, if caught in time
- When you notice it
- Stabilisation, Packaging, Storage
- Signs
- 5 recorded
What it is
Geranium taint is what happens when potassium sorbate is added to a cider that still contains viable lactic acid bacteria. Sorbate prevents yeast from refermenting sugar but does nothing to lactic acid bacteria, and those bacteria convert sorbic acid into a compound that smells overwhelmingly of crushed geranium leaves. It is one of the very few cider faults with a single, avoidable cause, and one of the most destructive: the compound has an extremely low odour threshold and the character is unmistakable.
What you notice
Grouped by sense, because that is how the fault presents itself rather than how it works.
On the nose
- A sharp, green, crushed-geranium-leaf smell — Distinctive enough that anyone who has smelled a geranium leaf recognises it immediately.
- A harsh, pungent edge that dominates everything else in the glass
On the palate
- A metallic, slightly bitter finish
How the batch behaves
- The character appearing days or weeks after sorbate was added, not immediately
- A cider that was sweetened and preserved rather than filtered or pasteurised
The words for it: Geranium. Each links to what produces it.
When it appears. After sorbate has been added to a cider that still carries lactic acid bacteria. It appears within weeks and does not resolve.
It is entirely preventable and entirely irreversible, which makes it the clearest argument on this site for the position that additions are decisions rather than routine: sorbate added to a cider carrying live lactic bacteria produces it reliably.
Andrew Lea, Peter Mitchell / Cider and Perry Academy
What it is mistaken for, and how to tell
One observation per rival, chosen because it separates them rather than because it describes either. Several of these are not faults at all — a style feature, a normal outcome, or the fruit behaving as it does.
- Sorbate off-flavour — Sorbate on its own gives a slightly synthetic, plastic impression; the geranium note is specifically the product of lactic bacteria acting on it. The second requires the first and is worse.
- Genuine floral or geranium aroma — Whether sorbate was used. The descriptor exists in the aroma lexicon as a legitimate floral term, and the taint is the same word arriving in a cider that was chemically stabilised.
What is actually happening
The chemistry or microbiology behind it. Understanding the mechanism is what makes the prevention make sense rather than being a list of rules.
Potassium sorbate dissociates in cider to sorbic acid, which inhibits yeast by interfering with membrane transport and with several enzymes. It does not inhibit lactic acid bacteria. Where those bacteria are present and active, they reduce sorbic acid to sorbinol — 2,4-hexadien-1-ol.
Under the acidic conditions of cider, sorbinol undergoes rearrangement and reacts with ethanol to give 2-ethoxyhexa-3,5-diene. That is the compound responsible, and its odour threshold is exceptionally low, so the quantity needed to ruin a batch is minute relative to the sorbate that was added.
The reaction requires all three components: sorbate, lactic acid bacteria, and ethanol. Sorbate in a non-alcoholic juice does not produce it. This is why the fault belongs specifically to sweetened, sorbated, unfiltered cider and is not a general property of the preservative.
There is also a separate, milder consideration. Sorbic acid itself, at higher concentrations, carries a faint plastic or candle-wax note independent of any bacterial action, which is a different fault and is covered separately.
How it happens
| Cause | Stage | How often |
|---|---|---|
| Sorbate added to a cider carrying viable lactic acid bacteriaThe sole essential cause; everything else is a precondition for it. | Stabilisation | common |
| Sorbate used without any accompanying sulphite, which would have restrained the bacteria | Stabilisation | common |
| Back-sweetening an unfiltered cider and preserving it with sorbate alone | Blending | common |
| A high-pH cider in which any sulphite present is largely ineffective | Stabilisation | occasional |
| Sorbate added before a spontaneous malolactic fermentation had run | Stabilisation | occasional |
Can it be put right?
Most cider faults cannot be reversed. Where that is the answer it is given plainly — an honest 'this batch is what it is' is more useful than a procedure that will not work.
| Option | Effectiveness | What it involves |
|---|---|---|
| Do not use sorbate on unfiltered cider again | Reliable | The only genuinely effective response. This is a preventable fault with a known single cause, and the lesson transfers completely to the next batch. |
| Add sulphite to stop the bacteria before the reaction is complete | Unlikely to work | Only conceivable if the taint has just begun and the sorbate is largely unconverted, and even then the character present does not go away. |
| Remove the compound from affected cider | Not possible | The threshold is far too low and the compound cannot be selectively removed. A batch with geranium taint is finished. |
| Mask it with sweetening, carbonation or blending | Not possible | The aroma is too potent and too alien. Blending simply produces a larger volume of tainted cider. |
CiderHQ gives no additive dosage figures. An addition is a food-safety decision that depends on legal limits, on the juice in front of you and on what you are protecting against, and the right figure comes from your supplier’s or regulator’s own guidance rather than from a general reference.
Preventing it
- Never use sorbate on its own. If it is used at all, it should accompany an effective sulphite level, chosen from supplier guidance for the measured pH.
- Allow any malolactic fermentation to complete before considering sorbate, so there is no active bacterial population left to work on it.
- Consider whether sorbate is needed at all — filtration or pasteurisation stabilises a sweetened cider without introducing this risk.
- Take the sorbate addition rate from the product supplier and from the legal maximum that applies, not from a remembered figure.
- Where the cider will be sweetened, decide the whole stabilisation strategy before adding anything, rather than reaching for a preservative at the last moment.
Geranium taint is the clearest example in cidermaking of a fault created entirely by a well-meant intervention. A maker sweetens a cider, worries correctly about refermentation, adds sorbate to prevent it, and produces something far worse than the problem they were solving.
The reasoning error is a specific one: treating sorbate as a general preservative. It is not. It inhibits yeast and only yeast, and it is only ever a partial answer to microbial stability. Used alongside an effective sulphite level in a cider at a sensible pH, it does the job it is for. Used alone in an unfiltered cider, it is an invitation.
The alternative routes are worth stating because they are more reliable. Sterile filtration removes the organisms rather than inhibiting them. Pasteurisation kills them. Either of those stabilises a sweetened cider without adding anything that a surviving bacterium can convert into something unpleasant.
It also belongs in a general lesson about additives at small scale. Every substance added to cider is a substrate for something, and a preservative that stops one organism can hand a different one an opportunity. Deciding what to add is a question about the whole microbial population, not about the one thing that is currently worrying.
The compounds involved
Sorbic acid
A preservative that stops yeast but not bacteria, and whose effectiveness — like that of sulphur dioxide — depends on pH, because only the undissociated molecule works.
Potassium sorbate
The soluble salt in which sorbic acid is actually added, used to hold sweetness in a bottled cider, and dangerous to use without sulphite because of what lactic bacteria do to it.
Ethanol
The alcohol yeast makes from fruit sugar, which converts a perishable juice into a keepable drink and carries most of its aroma to the nose.
Lactic acid
The softer acid that replaces malic when malolactic fermentation runs, halving the acid a cider carries and changing its texture as much as its sharpness.
The organisms involved
Oenococcus oeni
The acid-tolerant lactic acid bacterium that carries out most deliberate malolactic fermentation, converting malic acid to lactic acid after the yeast has finished.
Lactiplantibacillus plantarum
A versatile lactic acid bacterium, renamed out of *Lactobacillus* in 2020, capable of malolactic conversion and of a range of faults depending on conditions.
Lactobacillus brevis
An obligately heterofermentative lactic acid bacterium, renamed *Levilactobacillus* in 2020, associated with volatile acidity, biogenic amines and mousiness rather than with clean malolactic conversion.
Lactobacillus collinoides
A lactic acid bacterium first described from cider, and the organism most closely associated with acrolein bitterness through its conversion of glycerol.
Where in the process it arises
Sorbate stabilisation
Adding potassium sorbate to prevent yeast from restarting a fermentation in the package — an inhibitor rather than a killer, and one that must never be used without sulphite.
Back-sweetening
Adding sugar, juice or concentrate to a cider that has fermented dry — a straightforward adjustment that leaves the drink microbiologically unstable until something is done about it.
Sulphite stabilisation
Managing free sulphur dioxide through maturation and at packaging so that a useful antimicrobial and antioxidant fraction survives into the bottle.
Sterile filtration
Membrane filtration at a rating that retains viable yeast and bacteria, which is what allows a sweet cider to be shelf-stable without heat or preservative — provided the filling that follows is aseptic.
Pasteurisation
Heating cider enough to inactivate the organisms that would spoil it, either in the sealed package or in-line before filling, at a measurable cost in fresh aroma.
Faults it is confused with
These present similarly. What separates them is set out on each page.
Sorbate off-flavour
A faint plastic, candle-wax or celery-like note from sorbic acid used as a preservative, distinct from the geranium taint that bacteria produce from it.
Lactic off-flavours
Sauerkraut, sour milk, silage or cheesy notes from lactic acid bacteria working on sugars and other substrates rather than on malic acid alone.
Unwanted refermentation
Fermentation restarting in a sealed container, generating carbon dioxide that was not planned for and pressure that a bottle may not withstand.
Where this comes up in a guide
The link lands on the step where the fault arises rather than at the top of the pathway, because that is where the decision that causes it is taken.
What people ask next
Questions readers ask about the things this page mentions. Each one goes to the section that answers it rather than to a page written to receive the question.
- Why did my sweetened cider start fermenting again — Live yeast met the sugar that was added back. Sweetening is only stable if the yeast has been removed by sterile filtration, killed by pasteurisation, or held in check by sorbate together with sufficient sulphite.
- How many calories are in cider — Roughly 40 to 60 kcal per 100 ml for most ciders, so a UK pint falls somewhere around 200 to 250 kcal. Alcohol contributes about 7 kcal per gram and residual sugar about 4, so both strength and sweetness matter.
- How do i make my cider sweeter
- How do i sweeten cider without it fermenting again — Either remove or kill the yeast — sterile filtration or pasteurisation — or use a non-fermentable sweetener. Sorbate alone is not enough: it prevents yeast multiplying but does not kill what is already there, and needs sulphite alongside it.
- Which bacterium carries out malolactic fermentation in cider — Oenococcus oeni, in cider as in wine. Genome studies find that the cider strains are genetically distinguishable from the wine ones and that the strain most basal to the whole species was isolated from cider.
- Why does cider contain lactic acid — Because malolactic bacteria converted the malic acid into it. Malic acid has two acid groups and lactic acid has one, so roughly half the titratable acidity disappears and the cider tastes softer.
Where to go next
- The fault finder — Describe what you can smell and see, and narrow it down from the signs.
- All faults — Grouped by where they come from and how serious they are.
- Cider science — The chemistry and microbiology these faults come out of.
Sources
What this page rests on. Where a source is marked as registered rather than read, CiderHQ is recording that the body is authoritative on the subject without claiming to have worked through the document itself. See our evidence policy for what each state means.
The Science of Cidermaking and associated technical writing
Andrew Lea · reference work · passage verified 2026-08-24
Written by a food chemist who worked at Long Ashton on apple phenolics. Unusual among specialist cider writing in that it is primary-research-adjacent: the author is describing work he did, and cites the literature. This is why it is registered at tier 1 for chemistry while a general cider book is not.
Peer-reviewed literature on cider fermentation microbiology
Various journals · peer-reviewed literature · registered as competent for this subject
Covers the microbial succession of spontaneous cider fermentation, the role of non-Saccharomyces yeasts in the early stages, malolactic conversion by Oenococcus and Lactobacillus species, and the organisms behind the principal spoilage faults.
Cider and perry production technical training material
Peter Mitchell / Cider and Perry Academy · reference work · registered as competent for this subject
Searched again on 2026-08-25 and it remains the one printed source here that a bibliographic check cannot fix, because it is not a published work with an edition. It is course material issued to participants of a training programme, so there is no catalogue record to verify, no ISBN, and no lawful public copy. Its state stays `registered` for that reason rather than through inattention.