To perform a packaging integrity test, choose the method that fits the package construction, fix the test conditions and the pass/fail rule before the first sample goes in, run it on packs pulled straight from the line, and record enough that someone else can repeat the result. The choice of method is what decides whether the result means anything: a vacuum bubble test on a laminate pouch will find a seal that is open, but not a pinhole the product has plugged, and it says nothing about whether the pack survives a pallet ride.
Match the test to three things: whether the package is porous or nonporous, whether the pack can be pierced and inflated or must survive the test, and whether you need the leak located or only a verdict.
| Packaging or test objective | Common method | How it works |
|---|---|---|
| Flexible bags or pouches containing headspace gas | Vacuum bubble-emission test — ASTM D3078 | Submerge the sealed pack in water inside a vacuum chamber and draw vacuum; gas leaving through a leak shows as a stream of bubbles. Gross leaks only — trapped air, or product pushed against a small opening by the pressure inside, can plug it. |
| Pouches or trays that can be pierced and inflated | Bubble test by internal pressurization — ASTM F2096 | Inject air through the wall, submerge, and watch for bubbles. Destructive, and the one method that takes packs too large or long for any chamber. Its sensitivity was established on spunbonded-polyolefin medical packs; it has not been evaluated on nonporous film. |
| Porous medical packaging: transparent film sealed to a porous sheet | Dye penetration — ASTM F1929 | Apply dye along the seal edge by injection, edge dip or eyedropper; a channel the size of a 50 µm wire or larger wicks it through. Locates the leak; pass/fail only. |
| Nonporous packaging: plastic-film or foil-laminate packs | Dye penetration — ASTM F3039 | Method A applies dye at the seal edge and checks for staining, preferably against an absorbent surface; it finds channels of 50 µm or larger. Method B finds 10 µm holes in flat sheet. Works on opaque film; destructive, go/no-go. |
| Rigid, semi-rigid or flexible packages that must not be destroyed | Vacuum decay — ASTM F2338 | Place the pack in an evacuated chamber and measure the rise in chamber pressure; leaking headspace gas raises it. Finds holes down to 5 µm in nonporous packs and is fast enough for 100 % on-line testing. Any porous surface must be masked off first. |
For a flexible pouch off a form-fill-seal line, the FDA’s Bacteriological Analytical Manual lists burst testing, seal tensile (peel) testing, incubation and visual inspection as the recommended routine checks; vacuum and dye testing are accepted alternatives, not the primary method.
Write down the atmosphere, the sampling, the test parameters and the pass/fail rule before the first pack is tested. The size of leak a test can find depends on the product inside, the packaging material and the parameters chosen, so a result without recorded conditions cannot be repeated or compared with last month’s.
Condition and test in the standard laboratory atmosphere of 23 ± 2 °C and 50 ± 5 % relative humidity; where that is not possible, record the actual temperature and humidity with the result. Pull representative samples from the running line, not from the setup packs. Inspect each pack visually first and mark any visible defect with a non-water-soluble marker, so a bubble or dye trace can be tied to a known spot afterwards.
Then fix the acceptance rule per test:
Put one sealed pack under water in a vacuum chamber, draw vacuum, and watch for a steady stream of bubbles from one point on the pack — that stream is a leak, while scattered bubbles that cling to the surface are not.
The chamber is a bell jar with a tight-fitting, greased lid, a vacuum pump, a gauge, a release valve, enough water to cover the pack, and a weighted fixture to hold the pack below the surface, since a pack with headspace floats.
Read the result four ways:
Run a distribution test when the pack has to reach the customer through a pallet or parcel network intact, and a shelf-life test when the label carries a date; neither is a leak test, and a pack that passed the vacuum chamber can fail both. Both are run on real packs with product inside, never on empties.
Distribution testing follows ASTM D4169: the shipping unit goes through vibration, compression and impact in the order given, at levels representative of the route, and stays unopened until the sequence is complete. For flexible packs in shipping cases moving by motor freight, the established plan is Distribution Cycle 6 at assurance level II, with a typical pallet load as the shipping unit and two acceptance criteria: no product damage, and every package in good condition.
Incubate all packs 14 days at 100 °F before the run, and afterwards incubate the survivors another 14 days at 100 °F before inspecting them. A pack that loses hermetic integrity in any phase, or during that incubation, has failed.
For individual packs shipped by parcel carrier at 70 kg (150 lb) or less, the procedure is ISTA 3A: two drop sequences around random vibration with and without top load, and an 8-hour leak test required for liquids.
For food, the integrity side of shelf life is incubation: hold sealed packs at 95 °F (35 °C) for 14 days (10 days for USDA-regulated products), or at warehouse temperature for 30 to 90 days depending on how warm it is. Swelling, off-odour, a pH shift or a change in appearance reads as a lost hermetic barrier.
Record each test in enough detail that it can be rerun by someone who was not in the room, then apply the rule written in step 2 without renegotiating it against the result: any indication of leakage on a bubble or dye test is a reject, vacuum decay is accept or reject against the baseline, and a shipping unit that failed any phase of its sequence has failed.
The record covers:
On a failure, examine the pack to find the location and cause of the leak — product in the seal area, a channel at a gusset fold, a seal narrower than the minimum — because the cause, not the verdict, is what goes back to the line.
An integrity test is only as good as the choices made before it: pick the method by package construction and by what you need to know, fix the conditions and the pass/fail rule first, and treat the vacuum bubble test as the fast screen it is. Distribution and shelf-life testing belong to the pack’s validation, when the claim needs them. Whatever the outcome, the record of conditions and cause is what lets the next test — and the line adjustment between the two — mean something.
