Kombucha Bottles — Fermentation-Safe Guide

One of our guides on glass beverage bottles wholesale; start there for the overview.

Live kombucha keeps fermenting after capping and reaches 2–4 bar at room temperature, so it belongs in a pressure-rated glass bottle, by default a 330 ml longneck or stubby under a 26 mm crown cap, with 3–5% headspace. Screw caps only become an option once the product is pasteurized or sterile-filtered. Glass itself is unaffected by kombucha's pH of 2.5–3.5; the parts that need an acid specification are the cap liner and the cap's interior coating.

How much pressure the bottle has to hold

Residual sugar plus viable yeast means CO2 keeps building in a sealed bottle. The 2–4 bar figure applies at room temperature, and a 35 °C warehouse or a summer delivery van pushes it higher. That is the working range of a beer bottle, which is why the beer package has become the default for live kombucha.

Three controls keep that pressure inside safe limits:

1. The product. Ferment down to a target residual sugar, chill to slow the yeast, or stabilize by pasteurization or sterile filtration.
2. The package. Use bottles tested for internal pressure on an ISO 7458 sampling basis, with a documented rating and a certificate for each lot.
3. The headspace. At 3–5% the gas has room to collect. Pasteurized or sterile-filtered kombucha typically runs 2–3%.

A brand that chooses not to stabilize the product has no margin to relax the other two controls. Products that run at still higher pressures are a separate specification, described on our page about bottles for sparkling wine; beer-format pressure glass is covered under amber beer bottles.

Closure choice follows the kill-step

Decide first whether the product is alive in the bottle. Fermentation-active, refrigerated kombucha wants a crown or a swing-top. Once the product is pasteurized or filtered, ROPP and CT screw caps become workable and wide-mouth formats open up.

The crown is the safer default for a fermenting product. Its crimped seal tolerates pressure, its liner is made for carbonation, and co-packers everywhere have the capping equipment. Swing-tops seal just as well but complicate a high-speed line. Screw caps earn their place on multi-serve bottles, where the drinker needs to close the bottle again.

Acid resistance of glass, liners and coatings

Kombucha is among the most acidic mainstream drinks, harsher than most sodas. Its acetic and gluconic acids attack metals and can stress-crack some plastics or draw compounds out of them. Soda-lime glass does not corrode, leach or scalp flavor at beverage concentrations over any realistic shelf life, and it avoids the metallic notes that an acidic drink picks up from a can whose liner is less than perfect. Beyond food-contact compliance under EU 1935/2004, the glass needs no acid-specific requirement.

Everything else in contact with the liquid does. Ask for acid-resistant facings on crown liners, such as Saranex or food-grade PE, check the compound used in ROPP liners, confirm that metal caps carry an acid-resistant interior coating, and keep decoration away from the fill line. The bottle and closure are best sourced as one matched system, with annealing to ASTM C148 written into the bottle specification.

Sizes, weights and pressure ratings

The category sits in 250–330 ml for single-serve, which dominates convenience retail, and 500–750 ml for multi-serve in grocery and foodservice. The figures below are planning values; each is confirmed against the project drawing.

Flint glass suits most brands, and amber is available where a probiotic claim leans on light protection. Capacity tolerance to ISO 8106 is what keeps headspace consistent across a batch. A brand running both a single-serve and a multi-serve size can take them from one mold family to share decoration and simplify stock.

Hot fill and tunnel pasteurization

Glass copes with hot filling and in-bottle pasteurization as long as thermal shock is managed. The working rule is to keep temperature differentials below roughly 42 °C through filling and cooling, using bottles thermal-shock tested to ISO 7459. Tunnel pasteurization of filled, capped bottles is routine for shelf-stable kombucha. Tell the bottle supplier which thermal process you run, because annealing quality and per-lot thermal-shock testing are what make it safe at line speed.

Stock, modified stock or custom mold

Cost is quoted per project. It depends mainly on which of these routes you take, the decoration, the quantity and the freight terms.

Line settings to agree before the first commercial run

The bottle is half of the containment system. Filler, capper and chiller are the other half, and a good share of the "bottle problems" kombucha brands report start on the line.

• Residual sugar at fill. This is the biggest variable, and it is set in the brewery. A tight target caps the pressure the bottle can ever see. Write it on the batch record and divert runs that miss it.
• Filling temperature. Temperature drives both yeast activity and CO2 solubility, so the fill temperature determines the pressure that liquid carries later at 30 °C in a warm warehouse. Set a maximum at the valve and at the capper from your own pressure trials, log it per batch, and hold any batch that exceeds it.
• Filler type. An isobaric (counter-pressure) filler brings the bottle up to about the product's CO2 equilibrium before the valve opens, which keeps the liquid calm. Ambient and hot fillers work near atmospheric pressure and suit processes where stabilization comes after filling. Ambient filling of a live product into a cold bottle on a long line causes breakout and foam, and foam shows up later as uneven fill heights.
• Fill level. Headspace only buffers pressure if it repeats. Overfill and the same sugar gives more pressure; underfill and more oxygen stays in, which drives over-acidification and pellicle growth. If the drawing shows brimful capacity and a fill point, audit the filler against both.
• Time to chill. Every minute between capper and chiller is more CO2 in a live product. Measure temperature at the capper exit and specify the time allowed. Where the distribution temperature profile is unknown, assume the worst realistic case and prove the pack against it.
• Capping torque. Too little torque leaves a gas path and weeping caps. Too much deforms the liner and can damage the crown finish the cap seats on. Fix a torque window for the exact closure and finish, and check it with a torque tester at a set frequency each shift.
• Liner. The liner material inside the cap faces acetic acid on one side and sustained CO2 pressure on the other for the full shelf life. Closure and bottle suppliers should be working to the same drawing revision.
• Checks after capping. A checkweigher finds gross fill errors and a vision station finds missing or cocked caps. Neither helps unless the reject system for glass bottles removes the bottle without striking its neighbours or dropping pressurized glass on a hard floor. Keep a scheduled manual torque and fill-height check as well, since a checkweigher cannot see a liner that is failing slowly.

Where stabilization heats the capped bottle instead of the liquid before filling, the thermal profile of the filled pack becomes a packaging requirement. Our guide to a pasteurization line for glass bottles covers that decision.

Diagnosing gushers, leaks and breakage

With a live product the glass is usually the last link in the chain of causes. Check temperature and residual sugar first; they set the pressure and can be read straight from batch records. Check the closure and its torque second, because a bad seal accounts for weeping caps, flat product and violent gushers with no glass defect present. Look at the bottle third: the finish, the body, and whether the lot's pressure evidence meets specification.

Inspection evidence to require before shipment

Pressure failures come from the weak tail of a batch, not from its average. The sampling plan has to be wide enough to expose that tail while there is still time to act.

Lots, sampling and defect classes

Define a lot in advance, usually one production shift or pallet group, so any result traces to a machine and a time. Agree a sampling standard with sample sizes, accept and reject numbers and a defect classification. Anything that can affect pressure integrity is critical: cracks, a chipped finish, a body or base that fails the pressure check, signs of incomplete annealing. Samples should come from finished, packed stock, never from a rework pile, and the results should arrive with the calibration status of the test equipment.

Pressure and thermal-shock reports

A certificate that only says "pass" is of little use. Ask for the test method, the sampling basis, the measured values and the production date of the lot. Make sure you are getting routine lot testing, done per batch or shipment, and not a type test that was run once to prove the design; a sound design does not guarantee a sound lot. Where the bottle is hot-filled or tunnel pasteurized, the thermal-shock report should name the differential that was tested.

Glass defects that matter under pressure

• A pronounced or offset mold seam concentrates stress, most seriously on the body and near the base.
• Fine, scattered bubbles are normal. A large blister, a stone or a cluster near the inner surface thins the wall locally and is a real concern.
• A chipped or cracked finish is disqualifying by itself, because no closure seals on a broken edge.
• Cord or striae in the neck and an out-of-round finish reappear later as leaking caps.

A workable incoming rule: a visual and finish check on a sample from every lot, plus a pressure sample, with the whole lot held on any critical defect.

Transit tests, loading and claims

A drop or vibration result on a bare bottle tells you little about a loaded container. Request testing of the complete pack, meaning bottle, closure, divider, carton and pallet, under a protocol such as the ISTA series. The report should give drop height and orientation, whether vibration was random or swept-sine and its duration, a compression load that reflects the real stack height, and the sample count, conditioning and acceptance criteria for breakage, leakage and closure retention. Our page on transit testing for glass bottles explains why these have to be measured on your own finished configuration.

Most transit breakage comes from stacking and restraint. Agree a maximum stack height within the compression strength of the cartons and pallet, put corrugated dividers wherever bottles touch and layer pads between tiers, keep the load tight, and fill the void at the door with dunnage. Bottles should not be loaded warm, and a loaded container should not sit on a sunlit apron.

Put responsibility in writing before the first shipment. On ex-works or FOB terms, transit risk passes to the buyer at the agreed point; the supplier owes conforming product and packaging that meets the agreed test at handover. A claim needs a retained first-failure sample, photos taken at unpacking, the lot code and the carton identification, submitted within the agreed window. Some transit breakage is normal for glass, so settle the expected percentage range and how it is measured for each order.

What to include in a kombucha bottle brief

When you send us a kombucha bottle enquiry, we can match the project to a suitable plant faster if it states:

• whether the product is fermentation-active or pasteurized/stabilized;
• target capacity and style: longneck, stubby or swing-top;
• preferred closure and the capping equipment at your co-packer;
• thermal process: cold-fill, hot-fill or tunnel pasteurization;
• annual volume and the temperature profile of your distribution.

Kombucha bottle FAQ

Does every kombucha bottle need a pressure rating?

It does whenever fermentation can continue in the pack. A standard still-beverage bottle is not built for sustained internal pressure, so use glass that is rated and lot-tested, and make sure fermentation is effectively finished or the product is refrigerated before it ships.

Can a screw cap be used on kombucha?

Yes, on pasteurized or fully fermented, refrigerated product where pressure is under control. ROPP 28 mm and 38 mm CT caps reseal well on multi-serve bottles. Anything still fermenting should stay under a crown or swing-top.

Will the acid in kombucha harm the glass?

No. Soda-lime glass is essentially inert to organic acids at drink strength. Direct your checks at the liner facing and the coating inside metal caps.

Which formats are most common?

The 330 ml longneck with a crown finish, 250 ml and 330 ml stubbies, the 500 ml swing-top, and 750 ml for foodservice and grocery, where resealing matters.