Bandaid, Medicinal & Plastic Off-Flavours

A beer that tastes of sticking plaster, TCP or hot plastic is one of the more demoralising faults, partly because it tends to turn up in beers that are otherwise well made. It is also one of the most misdiagnosed, because three chemically different compounds all get described as “phenolic” and they have almost nothing in common beyond the word.

Work out which one you have before you change anything. The fixes do not overlap.

The three culprits

Compound	Tastes like	Comes from	When it shows up
Chlorophenols	TCP, Savlon, antiseptic, sharp plastic	Chlorine or chloramine reacting with malt	In the very first bottle, then stable
4-ethylphenol	Bandaid, barnyard, horse blanket	Brettanomyces or wild yeast	Clean early, obvious by week four
4-vinylguaiacol	Clove, allspice, pepper	POF-positive yeast strains	From fermentation onward, stable

The quickest sensory shortcut: 4-VG is spicy, chlorophenol is antiseptic. If what you are tasting is “hospital” or “swimming pool” rather than “clove”, your yeast is probably not the problem.

Why chlorine is the first thing to check

Two facts put it at the top of the list.

The detection threshold is absurdly low. Chlorophenols are perceptible in the low parts per billion, and some of them down into parts per trillion. Very little else in beer operates at that concentration.

Put numbers on it. Tap water at 0.5 mg/L of chlorine, with two millilitres of rinse water left in a 750 mL bottle, puts about 1.3 µg/L of chlorine into the beer. Even if only one percent of that converts, you are on the threshold. A volume of water too small to see is enough to fault the batch.

It reacts with malt, not with yeast. This is the part that trips people up. The other half of the reaction is the phenolic compounds from grain husk and hops, which are present in vast quantity in every beer at every stage. So two things follow that are worth stating plainly:

“My beer is crystal clear, there’s no yeast left, so chlorine can’t be the issue” is not an argument. Clarity is irrelevant. Yeast is not involved.
The mash is the main reactor, not the packaging step. An hour at temperature with the whole grain bed in contact is a far better environment for the reaction than a bottle is. Chlorophenols formed in the mash sail through the boil, fermentation and packaging unchanged.

Two tests that tell you which one you have

The timing test. Bottle the beer and taste it at day three, then again at week four. Chlorophenol is at full strength immediately and does not change much — the reaction happened long ago. A biological phenol does the opposite: barely there early, unmistakable by week four, because something is growing.

The gravity test. Compare your actual final gravities against what the recipe predicted, and take a reading on a bottle that has been sitting a month. If gravity has drifted down, or bottles are over-carbonating relative to the keg, or anything has gushed, you have a live organism and no further debate. If gravity is rock stable and on target, it is chemical, not biological.

”But the keg tastes fine — it’s only the bottles”

This is a common and genuinely confusing pattern, and it usually does not mean the bottling step introduced anything. More often the fault is present in the keg too, sitting just under your perception threshold, and three things push it over the line once the beer is in a bottle.

Temperature. Keg beer pours at 2–4 °C. Bottles get drunk warmer, and volatile release climbs steeply with temperature. A phenol you genuinely cannot detect at 3 °C can be obvious at 12 °C.

Loss of cover. Fresh beer, especially anything hoppy, carries a big load of aroma sitting on top of a trace phenol. Over weeks in a bottle the hop aroma fades and the esters degrade. The phenol is not increasing — the thing hiding it is going away.

Palate adaptation. You have been drinking that keg for three weeks and you have acclimatised to it. Whoever opens the bottle has not.

The test takes two minutes and costs nothing: pour a sample from the keg you think is clean, cover the glass, let it come up to 12–15 °C, and smell it. If the bandaid is there, stop investigating your bottling gear. The problem is upstream.

Where the chlorine is actually getting in

Roughly in order of how often each turns out to be the answer.

Sparge water wasn’t treated. The classic half-fix. Sparge water is about half your total volume and it hits the grain bed at peak extraction. Treating only the mash water feels like water treatment but isn’t. Campden goes in all of it.
Boiling used instead of treating. A rolling boil drives off free chlorine reasonably well, but does essentially nothing to chloramine. If your supply is chloraminated, boiling is not a treatment.
A carbon filter running at full tap pressure. Fine for free chlorine, but chloramine needs slow flow and long contact. And an exhausted cartridge is worse than no filter — ineffective, and a bacterial reservoir on top.
Chlorine-based steriliser tablets. Milton, baby-bottle steriliser and most generic “steriliser” tablets are hypochlorite or NaDCC. They are everywhere in New Zealand homebrewing and they are a direct chlorophenol source. Sodium percarbonate (“oxy”) products are fine — that is oxygen bleach, no chlorine in it.
Dishwasher detergent. Most powders are chlorine-based. Keep bottles and beer glasses out of the dishwasher entirely; the rinse aid will wreck your head retention as a bonus.
Making your sanitiser up with chlorinated tap water. The one people miss — see the next section, because “no-rinse” means you are deliberately leaving that solution in contact with the beer.
Rinsing with tap water after sanitising. StarSan is no-rinse. Rinsing it off with chlorinated water reintroduces exactly what the process exists to avoid. Drain and go.
Filling from a garden hose. Chlorinated water plus the hose’s own plasticiser. Worse if it has been sitting in the sun.
Old campden tablets. Metabisulphite degrades, particularly in a damp shed. An ageing tin under-doses without telling you.
Tank water that has been shock-chlorinated. Roof and tank supplies carry no residual — right up until someone doses the tank with bleach, which is normal practice and easy to forget about.

What about the water you make your sanitiser with?

Worth its own section, because the answer surprises people.

StarSan at 1.5 mL per litre is 99.85% water. Whatever is in that water is in your sanitiser. And the phosphoric acid does not neutralise chlorine — only a reducing agent does that, which is exactly why campden works and acid doesn’t.

It’s actually slightly worse than neutral. Dropping the pH to 2–3 works against you on both disinfectants: free chlorine shifts entirely to hypochlorous acid, the more reactive form and the one that attacks phenol rings, and chloramine hydrolyses toward free chlorine at low pH. So acidified chlorinated water isn’t chlorine-free water — the chlorine in it is arguably more available than it was in the tap.

Then the no-rinse instruction means you leave it there on purpose.

How much actually reaches the beer

This is the useful part, because the exposure scales inversely with vessel size — residual doesn’t shrink in proportion when the container does. Starting from tap water at 0.5 mg/L of chlorine:

Where the residual sits	Residual	Beer	Chlorine in the beer
Drained 30 L fermenter	~10 mL of film and foam	23 L	~0.2 µg/L
CO₂-purged 2 m fill line	~1 mL of drips and wall film	first 750 mL bottle	~0.7 µg/L
Drained 750 mL bottle	~2 mL	0.75 L	~1.3 µg/L
Un-purged 2 m fill line	~40 mL	first 750 mL bottle	~25 µg/L

A hundred-fold spread. How much of that chlorine becomes chlorophenol isn’t precisely predictable, so treat these as ballpark rather than gospel — but measured against a threshold in the nanograms per litre, the fermenter is marginal, the bottle is over the line, and the un-purged line isn’t remotely close.

If you already purge your line

Good — but it doesn’t go to zero, and the leftover is worth knowing about.

A 2 m length of 5 mm bore line has about 314 cm² of internal wall. Once the slug is pushed out you’re left with a drained film roughly 0.05 mm thick plus a few visible drips, which works out around 0.5–2 mL. That’s a 35-fold reduction — genuinely worth doing, and it takes the line from far and away your worst source down to roughly the same order as the residual sitting in the bottle itself.

Which is the useful conclusion: purging removes the line as the main suspect but doesn’t eliminate it, and these contributions add up. A chlorinated bottle plus a chlorinated purged line is around 2 µg/L before your brewing water is even in the picture. If you purge religiously and still get the fault, stop looking at the line and go after the make-up water instead — switching to RO zeroes every row in that table at once, which is why it’s the better lever than chasing residuals one at a time.

Two places a purge reliably misses:

Any sag or low point. Gas takes the easy path along the top of a partly filled horizontal tube instead of shoving the liquid ahead of it, so a dip can still be holding several millilitres after a purge that looked and sounded successful. Run the line downhill, no loops on the floor.
Dead volume in the fittings. Quick-disconnects and the internals of a bottling gun or counter-pressure filler hold liquid that the gas path never touches. Pull them apart and drain them rather than trusting the purge.

The un-purged case is also why this fault can be intermittent within a single session — that residue goes into the first bottles and is gone once beer has displaced it, so one run gives you some faulty bottles and some clean ones.

This isn’t a contradiction of “don’t fear the foam”

That advice is about the surfactant. StarSan’s foam won’t harm your yeast and won’t flatten your head retention the way detergent residue does — that part is completely true and still stands. It was never a statement about whatever else is dissolved in the water you mixed it with.

Don’t fear the foam. Fear the water you made the foam from.

What to do about it

Make your sanitiser with RO or distilled water. It also lasts much longer that way — hard or alkaline water buffers the acid and the solution goes cloudy sooner.
Or keep a campden-treated bucket for everything cold-side: sanitiser make-up, rinsing, topping up. A whole tablet in a 10–20 L bucket is a deliberate over-dose and that’s fine here, because the water barely touches the beer. Don’t over-dose your actual brewing liquor the same way.
Purge the fill line with CO₂ after sanitising and before the first bottle. Costs nothing and removes the biggest number in that table.
Drain properly. Invert things and let them drip instead of filling a wet vessel. It reduces the residual rather than removing it, but it’s free.

Why it comes and goes

If the fault is intermittent, this is usually why: the chlorine in your tap is not constant. Residual levels rise after mains flushing, reservoir work, heavy rain and turbidity events, and generally through summer, and they vary depending where you sit in the network. A brewer running a marginal dose is fine most of the year and gets caught when the network shifts.

There is a New Zealand angle worth knowing. Since the Havelock North inquiry and the Water Services Act 2021, residual disinfection has become close to universal across reticulated supplies under Taumata Arowai. Some supplies that were historically untreated now carry a residual. It is entirely possible for a process that worked fine for years to start producing faulty beer with nothing changing at your end.

The biological routes

Brettanomyces makes 4-ethylphenol, and lives in the places cleaning does not reach: beer lines, tap shanks, keg posts and dip tubes, bottling wands and beer guns, and scratched plastic. The tell is the timeline — clean at day three, faulty at week four.

Diastaticus is more likely than Brett in most home breweries, because it does not arrive by accident. It comes in ordinary commercial saison and Belgian yeast packets. Strictly, being diastaticus (the STA1 gene, which lets it eat dextrins) and being POF-positive (which makes the phenol) are two separate genes, but they travel together in saison and Belgian strains. So cross-contamination from one of those batches brings you over-attenuation and phenol together. If you have brewed a saison, a Belgian or a hefeweizen recently, or you share equipment with someone who has, look hard at the fermenter tap, any scratched plastic, and your yeast-harvesting jars.

The ones people miss

Peated or smoked malt through a shared mill. Peated malt is loaded with guaiacol and cresols, and in trace amounts — with none of the smoky context to explain it — it reads as flat-out medicinal TCP. If you mill at a homebrew shop you have no idea whose rauchbier grist went through the rollers before yours. Same risk with shared grain bins, or your own mill if you have ever done a smoked beer.

Plasticiser from the wrong tubing. Anything touching wort above pitching temperature has to be food-grade and heat-rated. PVC and unrated vinyl are cold-side only. Hot wort through the wrong hose gives you a distinct plastic note.

Perished rubber. Old keg lid O-rings, tap seals, bungs and grommets throw a rubbery-phenolic character as they break down.

Iodophor above concentration, or not drained properly. Straightforwardly medicinal.

”But phenolics come from hot fermentation and stressed yeast”

You’ll hear this constantly, and it’s correct often enough to be genuinely misleading. It’s true for one class of yeast and false for another, and which one you’re holding decides whether temperature control fixes your problem or wastes your money.

POF is a switch. Temperature is a volume knob. The switch is genetic — two genes, PAD1 and FDC1, which together convert ferulic acid from the malt into 4-vinylguaiacol. Most clean ale and lager strains carry broken copies, which is why manufacturers publish POF as a flat yes or no rather than a scale.

Where the rule holds. For a POF-positive strain it isn’t just true, it’s your main control. Push Abbaye, Farmhouse, Belle Saison, Munich Classic or WB-06 warm and you get more clove and pepper. Under-pitch and you get more again. Add wheat, or a ferulic acid rest at 45 °C, and more still. That’s exactly how the character is made on purpose in a hefeweizen or a saison.

Where it fails. None of those levers do anything in a POF-negative strain, because there’s no enzyme for the heat to speed up. US-05, S-04, Verdant, London Ale III, Conan and every standard lager strain cannot make 4-vinylguaiacol at any temperature, at any pitch rate, on any grist. Carrying the Belgian experience across to a hazy is the trap.

How to find out which you’ve got. Lallemand states POF status on most of its technical data sheets, and Wyeast has published a proper characterisation study banding 21 German and Belgian strains as POF-negative, weak, moderate or strong — it’s on their site and it’s the best single reference available. White Labs and Mangrove Jack’s don’t publish a POF field at all, so for those you’re reading the strain description and its style role, which is reliable at the extremes and vague in the middle.

The misdiagnosis worth knowing about

Here’s the bit that keeps the folk rule alive, and it’s the most useful thing on this page if you brew clean styles.

A warm, under-pitched fermentation on a clean strain produces fusel alcohols and ethyl acetate — solvent, nail polish, hot, harsh. A lot of people taste that and reach for “chemical”, or even “medicinal”, and log it as phenolic. It isn’t. There is no phenol in that beer at all.

The rule then appears to work, because the prescribed fix — get the temperature under control, pitch properly — is the right fix for what’s actually in the glass. Right action, wrong reason. It only falls over when someone with real chlorophenol buys a fermentation fridge and the fault sails straight through it.

So: if your “bandaid” turned up in a beer that fermented hot or was under-pitched on a clean strain, suspect fusels before you suspect anything on this page. Solvent and nail polish are the honest descriptors for that fault, and it needs a completely different fix.

The one case where heat really does phenolate a clean beer

Temperature can’t create a POF pathway, but it will happily amplify a POF-positive contaminant. A diastaticus or wild Saccharomyces carryover grows faster and throws more phenol in a warm ferment, so “we let it free-rise and the hazy came out phenolic” can be literally true — with heat as the amplifier on a contamination problem rather than the cause of anything.

Hazy IPA is the worst possible substrate for this, for a reason worth spelling out. Wheat carries more ferulic acid than barley, and around 30% wheat is the documented sweet spot for maximum 4-vinylguaiacol. A hazy at 20–40% wheat and oats is sitting right in that window with the precursor fully loaded — while every yeast anyone actually uses for the style is POF-negative and can’t touch it. The fuel is in the tank and your yeast has no matches. Anything POF-positive that gets in has a very good run.

What it definitely isn’t

Yeast left in the beer. Covered above — chlorine reacts with malt compounds, not yeast, so clarity has nothing to do with it.

Also worth ruling out before you commit: a harsh, tangy, mineral edge from too much sulfate (above roughly 250 ppm) is not phenolic at all, and neither is the drying astringency you get from a hot, high-pH sparge. Both get called “chemical” and neither is fixed by anything on this page.

Tests, cheapest first

Chlorine test strips on your brewing tap. A few dollars from a pool or aquarium shop. Read free and total chlorine; total minus free gives you chloramine. Test on several different days and after rain. This is the cheapest decisive test for the biggest suspect and almost nobody bothers.
The warm keg sample. Free, two minutes, and it tells you whether this is a systemic fault or a packaging one.
Read the label on every cleaning product in the brewery. Anything listing hypochlorite or dichloroisocyanurate is your answer until proven otherwise.
Final gravity against prediction, across your last few batches. Free, and it settles the diastaticus question on the spot.
One control batch on bought RO or distilled water, with campden added anyway. Twenty dollars of water buys a single-variable answer instead of another six months of guessing.
A split bottling run, if the fault really does track packaging. From one keg in one session: (A) your current process as the control; (B) sanitiser made with RO water and no rinse; (C) as B, plus the fill line blown clear with CO₂ before the first bottle; (D) filled straight off the keg tap with no bottling line at all. That isolates water, line and gear in a single sitting.

Prevention checklist

One hard truth

None of these can be fixed after the fact. Chlorophenols, 4-ethylphenol and 4-vinylguaiacol cannot be removed from finished beer, and no amount of conditioning will age them out. Everything on this page is prevention for the next batch. The good news is that the most likely cause is also the cheapest to eliminate: half a campden tablet and a change of rinse water.

See also: Brewing Water Basics for dechlorination in the context of the rest of your water treatment, and Cleaning & Sanitation for what to use instead of anything with chlorine in it.