Seal Integrity Testing: ASTM F1929, F88, F1140 and F2096 Against ISO 11607-1

Box of Class 4 self-sealing sterilization pouches beside a stack of 3.5 by 9 inch pouches with indicator strip

Four ASTM test methods are routinely cited for sterile barrier integrity, and they do not compete: each answers a different question about the same pouch. ASTM F1929 finds a channel through the seal, ASTM F88 measures how strong the seal is, ASTM F1140 measures when the pack bursts under pressure, and ASTM F2096 finds a gross leak anywhere in the pack. ISO 11607-1:2019 does not choose one of them — it sets the requirement that the sterile barrier system maintain sterility to the point of use, and it is the packaging material specification and your validated process that supply the numbers a test result is judged against.

That distinction decides how a clinic reads a test report. A "pass" from a 250 µm bubble test does not mean the seal is intact; it means no gross leak was detected at that sensitivity. A dye test that shows nothing does not prove strength; it proves no channel in the seal path the dye reached. The four methods below are laid out by what they actually detect, with the sensitivity each standard states in its own scope, so a report can be read for what it says rather than what it seems to say.

The packs in this article are the film-and-paper type: Class 4 self-sealing sterilization pouches with an external and an internal indicator, the 76 mm heat-seal roll for lengths a fixed pouch does not cover, and the constant-heat pouch sealer that closes roll stock. The test methods apply to that family far more than to rigid trays, and the difference matters below.

What does each of the four ASTM tests actually measure?

Each method is written around one failure mode. The table quotes the scope or significance clause of the current edition rather than a supplier summary.

Method (current edition) Question it answers What it detects Quantitative? Package types in scope Sensitivity or output stated in the standard
ASTM F1929-23 dye penetration Is there a channel through the seal? Channel defects in edge seals between transparent film and porous sheet; individual leaks only No — pass/fail, no leak size can be inferred Porous medical packaging (film-to-paper seals) Detects a leak equal to a channel formed by a 0.002 in (about 50 µm) wire, per SPMC guidance
ASTM F88/F88M-23 seal strength How strong is the seal, and is the process consistent? Weak, partial or drifting seals; process capability Yes — force per unit width, with maximum seal force reported Flexible barrier materials generally Not a leak method; a minimum level is a package requirement and an upper limit may be needed for peelability
ASTM F1140/F1140M-13(2025) internal pressurization Does the pack survive a pressure differential? Burst and creep failure of the whole pack Yes — failure pressure or time to fail Unrestrained packages; restraint plate options for porous packs Not a leak-detection limit; it evaluates failure under pressure differential from sterilization and transport
ASTM F2096-11(2019) bubble test Is there a gross leak anywhere in the pack? Gross leaks through a seal or through the material, in tray and pouch packs No — pass/fail against a stated sensitivity Trays and pouches 250 µm (0.010 in) with an 81 % probability; not evaluated for porous materials other than spunbonded polyolefin
ASTM F1886/F1886M-25 visual inspection Can a defect be seen? Channels, wrinkles, incomplete closure, delamination at the seal No — attribute inspection Transparent film seals Not applicable; it is a visual attribute standard

Two rows in that table are the ones most often misread. F2096 states its sensitivity as a probability, not a threshold: 250 µm at 81 % means roughly one gross leak in five at that size goes undetected in a single test. F1929 is explicitly not quantitative — the standard says no indication of leak size can be inferred from a dye result, and that packages showing any indication of leakage are normally rejected. A dye test therefore answers "channel or no channel", never "how big".

Which test detects which defect?

The defect decides the method, and choosing the other way round wastes either money or confidence. A clinic that has had three packs open in the drawer does not need a burst test; it needs to know whether the seal line has a channel, and whether the seal is strong enough to survive handling.

Defect found at inspection or in use Method that finds it Method that will not find it Why
Channel along the seal, visible only after the cycle F1929 dye penetration F2096 at 250 µm sensitivity; visual inspection if the channel is fine The dye travels the channel path; the bubble test needs a leak large enough to emit bubbles
Seal that is continuous but weak F88 seal strength F1929 (no channel), F2096 (no gross leak) Strength is a force measurement, not a leak test
Pinhole or puncture in the film F2096 (whole-pack gross leak), F1140 burst F1929, which is aimed at seals Dye penetration is applied to seal paths; a film pinhole may fall outside that path
Pack that fails when the chamber pressure changes F1140 burst or creep F88, which tests the seal alone F1140 loads the whole pack, as sterilization and transport do
Wrinkle, fold-over or scorched seal F1886 visual inspection All instrumented methods, which need a defect to measure Visual inspection is the only method that catches a defect with no leak and no strength loss

Where does ISO 11607-1 fit if it names no test method?

ISO 11607-1:2019 is a requirements standard, not a test-method standard. Its scope covers "requirements and test methods for materials, preformed sterile barrier systems, sterile barrier systems and packaging systems that are intended to maintain sterility of terminally sterilized medical devices until the point of use", and it applies to industry and to health care facilities alike (ISO 11607-1:2019, second edition 2019-02, amended 2023). It tells you the system has to work; the ASTM methods tell you how to demonstrate one property of it.

The interpretive layer is ISO/TS 16775:2021, which provides "guidance for the application of the requirements contained in ISO 11607-1 and ISO 11607-2" and states plainly that it "does not add to, or otherwise change, the requirements" (ISO/TS 16775:2021). The same document notes it is not a basis for regulatory inspection or certification assessment, and that using it is optional. In the United States the FDA recognises the consolidated edition — ISO 11607-1 second edition 2019-02 including Amendment 1:2023 — as a consensus standard (FDA recognition, ISO 11607-1), which is why suppliers there quote the amendment date next to the edition. That sentence is worth keeping in mind when a supplier presents a technical specification as if it were a mandatory rule.

For pouches and reels specifically, EN 868-5:2018 supplies the values the general standard does not: it "specifies test methods and values for sealable pouches and reels manufactured from porous materials", the materials are for single use, and the series "can be used to demonstrate compliance with one or more of the requirements specified in EN ISO 11607-1" (EVS-EN 868-5:2018). In practice, a pouch supplier's declaration of conformity to EN 868-5 is where a paper-film pouch's seal values come from, and the ASTM methods are how a clinic or a laboratory verifies a process against them.

How sensitive is each method, and what does a pass prove?

The sensitivity statements are short and easy to over-read.

F2096 states method sensitivity "down to 250 µm (0.010 in.) with an 81 % probability" and limits itself to tray and pouch packages; it also states the sensitivity "has not been evaluated for use with porous materials other than spunbonded polyolefin or with nonporous packag[ing]" (ASTM F2096-11(2019), scope 1.1–1.2). Paper-film pouches with medical-grade paper are not spunbonded polyolefin, so a bubble test result on that pack type is an indicator rather than a calibrated measurement of the smallest detectable leak.

F1929 is the more sensitive of the two on film-to-porous seals: SPMC's technical FAQ describes it as "significantly more sensitive than F2096" and able to detect a channel formed by a 0.002 in wire with 98–99 % confidence, while noting F1929 was not tested in tray/lid configurations and that the dye method applies only to individual leaks, not to a number of small leaks across a porous material (SPMC FAQ). The standard itself adds the wicking caution: dye wicks through porous material over time, which is why the immersion time is specified rather than left to judgement.

F88 is a different kind of number, and this is where a report can mislead. The standard says seal strength is "a quantitative measure for use in process validation, capability, and control", that a minimum level is a necessary package requirement, and that an upper limit is sometimes desirable so the pack can be peeled open (ASTM F88/F88M-23, significance 4.1). A clinic that asks only for "seal strength" will be handed a maximum force and no context; the useful record has the test speed, the specimen width, the minimum and maximum from the specification, and the number of specimens.

F1140 measures failure of the whole pack under internal pressure, which is the loading a pack sees during a vacuum cycle and during transport: the standard's significance clause frames it as evaluating "tendencies for package failure when the package is exposed to a pressure differential" from sterilization and transportation (ASTM F1140/F1140M-13(2025)). It also warns that correlating two pieces of equipment requires every parameter to match — package size, material, seal configuration, airflow rate and sensitivity among them.

What a pass proves, precisely:

Result What it establishes What it cannot establish
No dye penetration after the stated immersion No channel detected along the seal path tested, above the method's detection capability Seal strength; leaks in the film outside the seal path; performance after transport
Seal strength above the specified minimum The process produced a seal that holds under the test conditions used Absence of a channel; sterility of the contents
No bubbles at 250 µm sensitivity No gross leak detected at that size and probability Absence of leaks below 250 µm; sterility; seal strength
Burst above the specified pressure The pack survived that pressure without failing Seal quality at the edges; microbial barrier performance in storage

Who decides how many packs to test?

None of the four methods fixes a sample size. They fix procedure, specimen preparation, apparatus and — for F2096 — a sensitivity. The number of specimens, the frequency and the accept/reject rule come from the validation protocol and the quality system behind it, which is why two clinics can both claim "F88 tested" with one specimen each quarter and one specimen per shift, and neither is wrong on the standard's terms.

Three decisions belong in the written procedure rather than in a conversation:

  1. Sampling basis. A risk-based plan is normal: more specimens after a material change, a sealer repair or a run of failures, fewer once a process has a stable history. Where a formal sampling plan is used, it is usually an AQL plan such as ISO 2859-1:1999 applied to the seal attribute, not a number taken from a test method.
  2. Frequency. Visual inspection at every pack is the baseline; instrumented testing is periodic, and the period is a documented decision. SPMC's own FAQ points out that whether two integrity tests are redundant depends on what else is being verified — the question is never "which single test", it is "which property is unverified".
  3. Who performs it. A dye test needs a laboratory bench, dye solution at a stated concentration, controlled immersion and a reader who knows what wicking looks like. Most dental clinics do not run one in-house; they specify it with a packaging supplier or a test laboratory and keep the report, which is a legitimate way to satisfy the requirement that the property be verified.

What do these methods mean for a Canadian clinic's file?

The Canadian chain runs through reprocessing requirements rather than through ASTM. CSA Z314:23, the 2023 second edition of the Canadian reprocessing standard, extends coverage to professional offices and private practices and includes, among its revisions, the reprocessing of sterile barrier systems and consumables (CSA Z314:23). Public Health Ontario's provincial guidance document states the operational rule that ties the methods to daily work: at point of use, "the integrity of the packaging and the equipment/device must be checked", and a pack whose integrity is compromised is no longer sterile and must be reprocessed (Public Health Ontario, provincial guidance on cleaning, disinfection and sterilization of medical equipment and devices in all health care settings, recommendation 86). The same document describes shelf life as event-related rather than time-related, which is the reason a seal failure found in the drawer invalidates the pack regardless of its date.

Health Canada's own guidance puts the same content in the manufacturer's directions for use: the guidance for reprocessing and sterilization information for reusable medical devices exists to help manufacturers meet the requirement in paragraph 21(1)(i) of the Medical Devices Regulations that directions for use carry reprocessing information (Health Canada guidance).

For a clinic, that produces a three-level routine which is defensible in an audit without pretending to be a packaging laboratory:

Level What is done Evidence kept Method reference
Every pack Visual inspection of the seal and the indicator after the cycle and again at point of use Reject count and reason in the day's records F1886 attributes; PHO recommendation 86
Every sealer, every shift Start-of-shift seal on a test strip, checked for channel and closure Temperature or setting, result, initials Your validated parameter set
Periodically, or on trigger Seal strength and dye penetration on production packs Report with method, edition, specimens, values, accept/reject F88, F1929, F2096 as applicable

The trigger list is the part that keeps a file current: a new pouch supplier, a new material lot with a different film, a change from pouches to roll stock, a repaired or replaced heating element, or a run of failures in use. Any of those changes the material or the machine the seal is made with, and a test result from before the change describes a process that no longer exists.

What belongs on a seal integrity report?

A report is useful only if someone else can repeat the test from it. The fields below are the minimum that let a reviewer connect a result to a pack, a process and a standard edition; anything less turns the report into an assertion.

Field Why it is needed
Test method and edition (e.g. ASTM F1929-23) Versions differ; an unstated edition cannot be checked
Package construction and supplier lot Material drives both sensitivity and expected seal strength
Seal parameters in force (temperature setting, dwell, seal width) A result without the process cannot be reproduced or revalidated
Specimens tested and how they were selected Distinguishes a sample from a single pack
Apparatus or laboratory, with calibration status Comparative results require matched equipment and parameters
Result values (force, pressure, failure mode) or attribute outcome Pass/fail alone cannot be trended
Acceptance criteria and their source The value has to come from the specification, IFU or validated process — not from habit
Disposition of failures Rejected packs, rework, and what changed afterwards
Box of Class 4 self-sealing sterilization pouches beside a stack of 3.5 by 9 inch pouches with indicator strip Constant-heat pouch sealer loaded with a roll of sterilization film, used to close packs before integrity testing

Where the numbers come from before a test exists

A test only judges a number that already exists. For a paper-film pouch, three sources supply it: the material specification the pouch is declared against, which for pouches and reels in Europe is EN 868-5:2018 and elsewhere the supplier's own declaration; the pouch manufacturer's instructions for use, which state the sealing conditions the pouch was qualified for; and your own validated seal process, which is the only one of the three that describes your machine, your operator and your room.

Where a film has no published seal values, laboratory seals made under ASTM F2029-16(2021) and peeled under ASTM F88 establish that starting relationship between temperature, dwell and strength; the practice's own scope says its purpose is to help establish "starting relationships for sealing flexible barrier materials" and that it does not cover validation of production equipment.

ISO 11607-2:2019 is the part of the pair that deals with that third source — validation of forming, sealing and assembly — and it applies "to industry, to health care facilities, and to wherever medical devices are packaged and sterilized" (ISO 11607-2:2019). A clinic that seals its own packs is inside that scope, which is why the seal parameters belong in a procedure with a revision date rather than in a note taped to the bench.

CliniEco Medical supplies the pouches, roll stock and sealers described above, and distributes them under Health Canada MDEL #35334. The standard editions and methods cited here apply to any supplier's packaging; what a buyer should compare is the material declaration and the instructions for use, not the brand on the box.

Clinics that buy packaging and sealer supplies for several sites can put the range on one line through wholesale and multi-site ordering, and the sterilization pouches collection lists the formats stocked for clinics. Reselling? become a distributor.

Recording what you already test? The compliance log centre carries a printable seal check sheet and a browser-based log generator for seal parameters, test results and reject counts, with no account required.

Have a question about your own packs? Send the pouch construction, the sealer setting, the test method you were asked for and the failure you saw, and you will get a written answer that cites the standard or regulation it rests on. Ask your compliance question.

Related reading

Frequently Asked Questions

Is a bubble test a substitute for a dye penetration test?

No. ASTM F2096 finds gross leaks anywhere in a tray or pouch pack, with a stated sensitivity of 250 µm at 81 % probability. ASTM F1929 finds a channel along a seal between film and porous sheet and is markedly more sensitive on that specific defect. A pack can pass the bubble test and still carry a channel that the dye test would show, and a pack can pass both and still have a weak seal that only ASTM F88 measures.

Do I need all four tests on my pouches?

Almost never, and the requirement is not to run every method. The requirement is that the sterile barrier system maintain sterility to the point of use, with the properties verified where they are at risk. In a dental clinic that means visual inspection of every pack, a documented seal check at the start of each shift, and instrumented testing — F88 seal strength, plus F1929 when a channel is suspected — on a stated frequency or after a change of material, machine or process.

What sample size does ASTM F88 require?

None. ASTM F88 defines how to make and measure a seal-strength specimen, not how many to test. The number of specimens per run, the frequency and the accept/reject rule belong to your validation protocol or quality system, and they should be written down with the reason. A procedure that says "ten specimens per shift, reject the run if any falls below the specification minimum" is auditable; "we test now and then" is not.

Can I run a dye penetration test in the clinic?

It is possible but rarely sensible. ASTM F1929 requires a dye solution of stated concentration, controlled immersion time, a reader who can distinguish a channel from wicking through the porous material, and a disposal route for the dye. The wicking caution in the method is exactly what an untrained reader mistakes for a failure. Most clinics specify the test with the packaging supplier or an accredited laboratory and keep the report in the sealer file.

Why does the standard say the bubble test sensitivity is a probability?

Because detection at the limit is statistical. ASTM F2096 states sensitivity down to 250 µm with an 81 % probability, meaning a leak at the limit is missed in roughly one test in five. That does not make the method unusable — it makes it a gross-leak screen, and it explains why a single pass at the limit is not evidence that no leak exists anywhere in the pack.

Does ISO 11607-1 require one particular seal test?

No. ISO 11607-1:2019 sets requirements for materials, sterile barrier systems and packaging systems, and it references test methods for specific properties rather than mandating a single integrity test. The method is chosen by package type and by the property at risk, and the acceptance values come from the material specification, the manufacturer's instructions for use and the validated process. ISO/TS 16775:2021 explains how to apply the requirements and explicitly adds no new ones.

Vous travaillez déjà avec des cliniques ? Devenez représentant régional.