A tamper-evident band is a ring moulded as part of the cap that snaps free on first opening and stays behind on the bottle neck; a shrink band is a separate film sleeve shrunk over the cap and shoulder that the consumer tears off and throws away. The ring adds no equipment to the filling line and leaves evidence you can inspect later, while the sleeve is more visible on a shelf and can carry print but needs a heat tunnel. Whichever you choose, approve it on four checks run on your own filled pack: survival through capping and transport, break behaviour across temperature, consistency of what is left after opening, and a failed attempt to put it back.
What a tamper-evident pack can and cannot promise
No closure on a glass bottle stops tampering. Someone with a heat gun, a stock of identical rings and thirty seconds can rebuild nearly any pack. What the pack can deliver is an event that is irreversible and visible: an ordinary consumer, in ordinary light, with no tools and no training, can see that the bottle has been opened.
That changes how the requirement should be written. "Tamper proof bottles" has nothing in it to test, since nothing can be shown to be tamper proof. "The closure shall incorporate a tamper-evident band" is barely an improvement, because it names a part and not a result. A specification that holds up in an audit states five things:
- which element has to break;
- what that element looks like before opening;
- what it looks like afterwards;
- where the broken piece ends up;
- the method used to confirm all of this on a sample.
Writing the outcome instead of a part number also protects you against quiet substitution. A ring geometry or sleeve film can look the same in a photograph and behave quite differently in a consumer's hand. If the specification describes what must be observed, incoming inspection catches the swap. If it does not, the first sign is usually a retailer or distributor raising a question months later, by which time the whole batch is in doubt.
How the two routes compare
Nearly every tamper-evident closure on glass comes down to one of these mechanisms, or to both on the same pack. They differ in where the permanent change happens, what the consumer feels, and what the line needs.
| Attribute | Tamper-evident (TE) band | Shrink band |
|---|---|---|
| What it is | A ring moulded in one piece with the cap shell and attached by thin bridges | A separate tube of heat-shrinkable film, typically PVC or PETG |
| How it is held | A raised bead on the neck (retaining bead or pilfer bead) at a set distance below the thread keeps the ring down | Hot air or a heat tunnel shrinks the film onto the cap and down over the shoulder |
| First opening | High resistance, a visible or audible release as the bridges snap, then a much lighter turn | No torque peak and no snap; the consumer finds the perforation and pulls the tear strip |
| Evidence afterwards | Residual: the broken ring remains on the neck | Transient: the torn film is discarded with the cap |
| Added line equipment | None beyond the closure | Heat tunnel, a temperature setting and a band-position check |
| Other points | Easier to pair with a safety closure | More visible in a shop and printable, hence its use in cosmetics and food |
Break torque and spin torque on a TE band
Engineers call the peak effort that snaps the bridges the break torque, and the light effort that follows the spin torque. The drop between them has to be designed. Set the break torque too low and bridges give way in transit, so the pack arrives looking opened. Set it too high and an older consumer cannot get in, which reaches you as a usability complaint, not a tamper complaint.
The value depends on the closure material, the cross-section of the bridges, the ambient temperature and how tightly the ring interferes with the neck bead. It therefore has to be measured on your finished pack. A catalogue figure is not a substitute.
A ring left on the neck is also the more useful kind of evidence. A residue is harder to explain away than a clean neck, and it is still there if the bottle comes back through a returns process or ends up in front of a regulator.
Film, tear strip and coverage on a shrink band
The sleeve works because shrunk film takes the shape of what is under it. Once it has conformed, it cannot be lifted away and replaced without a new sleeve and a heat source, or without an obvious cut. Most bands include a perforated tear strip so the pack opens by hand, and a band whose tear strip does not work is a usability defect.
Record the direction of assembly: does the sleeve cover the top of the cap, or only the sides and shoulder? A sleeve that stops below the top edge can sometimes be eased down and off without tearing. A pack that opens without destroying its band has no tamper evidence, however convincing it looks in a photograph.
The evidence a consumer and your QC team should see
Before specifying parts, write down the evidence chain: the observations a person can make without tools to decide whether a pack is intact. Doing this first avoids packs that are tamper evident on paper while the evidence is hidden, ambiguous, or easy to mistake for a moulding defect. A usable chain has four links.
- One irreversible break. Snapped bridges, torn film or a seal destroyed on removal all count. Friction fits, press fits and snap beads that re-seat do not, because a hand can restore them and they tell you nothing.
- A broken state that looks different from the intact one at arm's length in normal shop lighting. A ring with one of eight bridges broken still looks whole. So does film cut along its existing perforation, which is why the perforation must be placed so that any removal without tearing causes visible damage.
- The broken part stays where it can be seen: a ring on the neck, torn film still attached to the cap, a membrane retained inside the cap. Evidence that went in the bin has no audit value.
- Rebuilding the pack takes equipment a casual interferer lacks, in practice a capping head, a heat tunnel and matched rings or sleeves. This equipment requirement is the real barrier, more than any strength in the part.
One refinement is worth considering: print the lot code or expiry date on the tamper-evident element itself. A coded band or ring that must be destroyed to open the bottle cannot be moved to another bottle without carrying a code that will not match those contents. The evidence then doubles as a traceability control, which is the main reason the format persists in regulated categories.
Combining tamper evidence with a child-resistant closure
The two features are physically independent and can share a pack, but most avoidable failures occur where they meet. How a safety closure engages the neck, whether the outer shell spins freely and which protocol panels apply are covered in our guide to child-resistant closures on glass bottles. Here the concern is only how the tamper-evident element interacts with it.
A shrink band over a child-resistant cap leaves the safety mechanism working, but adds a step. The film bridges cap and neck, so the consumer must tear it before the push or squeeze action can start. If the senior-adult panel is run on the banded pack, tearing becomes part of the measured opening effort, and a band a little too thick or too tightly shrunk can take a marginal cap out of the pass band. If the panel is run without the band, the test does not describe the product the consumer receives. Either can be defended, provided the choice is documented and any later change of film or thickness is treated as a change to the tested object, not as a swap of consumables.
A ring is generally the easier partner, since it is part of the closure moulding and the safety mechanism was designed around it. The forces still need checking together. A push-turn cap depends on the outer shell passing downward force to the inner body, and a ring that is too stiff, or a neck bead at the wrong height, absorbs some of that force and alters the release.
Ask for the ring, shell and neck bead as one matched set with the cap, and have the finished filled pack tested instead of each component alone. Choosing the closure supplier and the bottle supplier separately, with the bead defined only by a drawing note, is how buyers end up with rings that will not break or that fall off whole. Never approve the tamper-evident element on a different pack from the one that carries the safety feature.
Four checks to run before approval
Buyers tend to leave verification to the supplier, and it is the part that most often comes back to them. These measurements can all be done on a modest bench. Where the product is regulated, repeat them on filled bottles, because filling changes the grip, the weight and the forces involved.
Retention through capping and transport
The element must get through capping at production speed, and through the shrink tunnel where one is used, with no mechanical damage. Control this with a visual standard and a first-article approval on the real line. For distribution, run a vibration and drop programme to a recognised method, then count the units whose evidence is already destroyed. Any unit that arrives looking opened is a fail, regardless of cause.
Break behaviour and opening effort
On a ring, measure the peak torque that breaks the bridges and confirm the fall to spin torque is clean. Take readings at both extremes of the temperature range the pack will meet. Closure plastics soften when warm and turn brittle when cold, so a system approved at room temperature can fail in an unheated container in winter or a hot warehouse in summer. Treat a supplier's figure as a reference range to confirm by measurement. How capping-head torque is set, how removal torque is measured and how a torque window goes into an acceptance standard are explained under cap torque testing for bottles; that test and this one meet at the same capping head but measure different things, and a single number asked to cover both usually describes neither.
Drop-off consistency
Open every unit in a defined sample by hand, the same way each time, and record how the evidence presents. Count the rings that stayed on the neck, those that came off with the cap, those left hanging by a single bridge, and the units that opened with no visible change. Consistency matters more than elegance. A pack that always fails in one observable way can be given a written acceptance criterion; a pack that behaves differently on one unit in fifty cannot.
The negative test
Finally, confirm that an opened pack cannot be made to pass as intact. Have the element removed and re-applied by hand, as someone at home might try, and look at the result. A ring that can be worked back onto the neck fails immediately. So does a sleeve that slides back over the cap with its tear turned to the rear.
What to ask a supplier, by tamper-evident approach
| Approach | Evidence to request | Deflection you may hear | Follow-up to ask | Accept when |
|---|---|---|---|---|
| TE band moulded with the closure | Closure drawing with bridge count and cross-section; neck bead diameter and distance below the thread; material grade; break and spin torque measured on a filled sample from the intended line | "It is a standard mould that every customer uses." True, and beside the point: bead height and bottle neck are not standard as a pair. | What is the break torque on my bottle, with my product in it, at the cold and hot ends of the shipping temperature range? | One clean break where intended, a consistent fall to spin torque, the ring kept on the neck, and no pre-broken units after the distribution test |
| Shrink band sleeve | Film type and thickness; layflat width against measured cap and shoulder circumference; shrink ratio; tunnel temperature and dwell; tear strip position; print proof; food-contact declaration for film and ink | "PETG and PVC perform the same." They differ in the tunnel, in print and in recycling discussions, and their declarations are not alike. | At your production tunnel setting, can the shrunk band be slid off by hand without tearing? Show me. | The band follows the cap contour, the tear strip opens by hand, no unit opens without visible film damage, and nothing lifts off untorn |
| TE band and shrink band together | Proof that both elements were on the samples used for child-resistance or usability testing, with the opening sequence written step by step | "The band does not affect the closure." It alters first-opening effort, and what is tested is the assembled pack. | Did the panel use banded or bare packs, and does the report name the film and its thickness as part of the tested object? | The tested object is the shipped pack, any substitution triggers requalification, and the written sequence matches consumer experience |
| Induction foil or pressure-sensitive inner seal | Seal specification; activation parameters; peel or puncture result on the filled pack; liner composition; declaration covering the seal layer and any adhesive | "The foil seal is the tamper evidence." Only if the membrane cannot be re-laid invisibly, and it tells you nothing about the cap. | Once peeled, can the membrane be replaced on the same neck so it looks unopened? Show the pack after trying. | The membrane is destroyed on first removal or leaves residue that cannot be hidden, and the seal passes a leak and integrity check on the filled pack |
Documents to collect with the first sample
Paper supports a tamper-evident claim as much as parts do. Request it alongside the first sample, while you still have leverage, and not with the first purchase order.
- Drawings with critical dimensions, not just the outer profile: bridge geometry for a ring, film thickness and layflat width for a sleeve, and the neck bead position on the bottle drawing.
- The material grade by name. Moving from one polypropylene grade to another changes break behaviour while the drawing note stays the same.
- Food-contact documentation for every layer that touches the product, including film, ink and any adhesive, with a migration statement for the intended contents and temperature of use. A tub and a tincture are different exposures, so a declaration that does not name the contents cannot be used. What to request is set out in our notes on the food contact declaration for glass bottle packs.
- Controlled process parameters. For a sleeve: tunnel temperature, conveyor speed and the band-position check. For a ring: the neck bead dimension as an inspected characteristic, and the capping head setting that gives the intended break torque.
- Run records: a first-article report, an in-process inspection record, and a retained pair of intact and opened samples from the approved run.
The retained pair is worth more than any certificate. It gives a later complaint a physical reference from the day the pack was approved.
Keeping production the same as the approved sample
Most tamper-evidence failures are drift, not design. A film lot with a slightly different shrink ratio, a ring moulded with a different colour masterbatch, a tunnel running ten degrees hotter in the afternoon, a capping head with a worn clutch that deforms the bead on one side: the drawing is unchanged in every case, and the observable outcome is not.
The defence has three parts. First, a short list of controlled characteristics, each quick to inspect. Second, a change-notification clause that names them, so that a new film supplier, film thickness, closure material grade or ring mould requires written notice and requalification. Requalification means repeating the four checks above; a photograph does not count.
Third, keep a sealed, filled sample from every batch under trade conditions for the product's shelf life plus a margin. That is your reference when a complaint arrives a year later. In categories with a regulatory filing, retain the outer carton too. The carton sometimes carries the child-resistance or tamper-evidence feature, and then a bottle on its own proves nothing.
What to send when briefing a tamper-evident pack
Four details decide which routes are admissible, which film or closure materials are worth considering and which tests will be expected: the destination market, the contents with their approximate strength or pH, the fill volume in millilitres, and whether a child-resistant closure is also needed. Add the closure you have in mind if it is already fixed. A photograph of a competitor's pack is workable but slower, since bead and bridge geometry then have to be inferred from an image instead of confirmed on a drawing.
From that we can name a route, list the measurements to request from the closure and film suppliers, and describe what the approved sample should look like so later batches can be compared with it. Schedule and output depend on the neck bead, the film and the capping line, so they are settled per project once the specification is defined; a firm schedule offered before those are known is a guess.
For oils, aroma blends or extracts in small bottles, the closure and pack options that use these same formats are gathered under essential oil bottles. There the compatibility question is aroma and chemical attack on the liner. A sensible order for a first project is contents and closure first, evidence and torque second, paperwork last.
Questions buyers ask about tamper-evident glass bottles
Does an induction foil seal count as tamper evidence?
Only partly. A foil membrane heat-sealed to the neck is destroyed or leaves residue the first time it is removed, so it does show opening, but it cannot show whether the cap has been taken off and put back. Most buyers use a membrane for product integrity and a band or sleeve for the outside evidence. The membrane is a food-contact item itself, so its layers and adhesive need declaring.
Why do tamper-evident rings sometimes break before the customer opens the bottle?
There are three usual causes. The neck bead may be at the wrong height or diameter, loading the ring from the moment of capping until it creeps and the bridges fail. The bridges may be too thin for the material actually used, often after a change of grade or masterbatch weakened them. Or the capping head may have applied more torque than intended and deformed the bead or ring on the way down. Each of these shows up when production is compared with the retained sample pair.
Should the band carry the lot code and expiry date?
It suits regulated categories and products where counterfeiting is a concern. The trade-offs are that the print must survive shrinkage without distorting, and that a misprint is a reject, not a rework, because the band cannot be taken off and reused. If the cap or bottle shoulder already carries batch traceability, a plain band can be simpler.
Is a PVC shrink band acceptable for food or cosmetic packaging?
That depends on the market and the film formulation, and documentation should settle it, not habit. PVC films contain plasticisers, and both the plasticiser and any printing ink fall within the food-contact assessment for the intended contents and temperature. PETG avoids the plasticiser question and is often the easier choice for a new pack. Some markets now also treat the two films differently for recycling.