Title: ASTM F88 Peel Testing for Medical Packaging
Body: Two seals with the same peel-force number can pass and fail the same acceptance criteria. The difference isn't the seal — it's how the specimen was held in the grips.
That's the part of ASTM F88 testing that trips people up. The standard looks simple: cut a one-inch strip, pull it apart, record the force. But the support technique you choose changes the number you get, and if you don't fix and document that choice up front, your peel-force data won't mean what you think it means.
F88 is the principal tensile method for measuring seal strength of flexible barrier materials in sterile barrier systems — the force required to peel or break a one-inch-wide sealed specimen (ASTM F88/F88M-21). Those numbers feed material qualification and seal-process validation under ISO 11607-1:2019. Here's how to run the test so the data holds up.
What F88 Actually Measures — and What It Doesn't
What it does not tell you is whether your finished package maintains a sterile barrier. F88 is a destructive material- and seal-process test, not a whole-package integrity test (ASTM F88/F88M-21). I bring this up first because I've seen teams treat a passing F88 result as proof the package is good. It isn't. Seal strength is one input. You still need package integrity testing — dye penetration (ASTM F1929) or bubble emission (ASTM F2096) — to confirm the barrier itself.
Keep those two questions separate: how strong is the seal (F88) and does the barrier hold (integrity testing). They're different tests answering different questions.
Fixture Setup: Free-Tail vs. Supported
This is where the number you report gets decided.
- Technique A — unsupported (free-tail). The two tails hang free, unsupported by any backing. The seal is pulled at an angle that floats as the test runs.
- Technique B — supported at 90°. One tail is held against a support so the seal peels at a fixed 90° angle.
- Technique C — supported at 180° (aligned). Both tails are supported and aligned, so the specimen pulls in a straight line.
The free-tail technique is the most common default, and it usually produces the lowest, most conservative peel value because the peel angle isn't constrained. The supported techniques control the angle, reduce variability, and often read higher.
Here's the rule that matters: the support technique is not directly comparable across configurations, so you must report it with the result (ASTM F88/F88M-21). Pick one, fix it in your protocol, and run every specimen the same way. If you qualify a material with Technique A and then someone runs Technique C on the next lot, you're comparing two different measurements and calling them the same thing. That's a finding waiting to happen.
My advice: choose the technique during protocol design, document the rationale, and don't change it mid-program. If you have a reason to switch — say, the free-tail angle is causing erratic results on a stiff laminate — re-baseline and note it.
Grip Speed: The 12 in/min Convention
What I want to be clear about: F88 itself does not mandate a single fixed speed. The standard requires you to specify and report the speed you used (Instron, ASTM F88 methodology reference). The 12 in/min figure is a widely applied convention, not a regulatory requirement.
Why does it matter? Peel force is rate-sensitive. Pull faster and you'll generally read higher; pull slower and you'll read lower. If your protocol says 12 in/min and the technician runs at 20, your data isn't wrong exactly — it's just not the test you validated. Lock the speed in the protocol, set it on the frame, and verify it.
The Failure Modes That Invalidate Your Data
Not every test that produces a number produced a valid number. Two artifacts show up constantly, and both can quietly corrupt a data set if the technician isn't watching for them.
Jaw Breaks
A jaw break is when the specimen fails at the grip line instead of at the seal. The material tears where the jaw clamps it, and the force you recorded reflects the strength of the substrate failing at a stress concentration — not the strength of the seal.
When you see a jaw break, the result is suspect. You measured the wrong thing. The fix is usually grip-related: check the clamping pressure, look at whether the jaw faces are pinching a sharp edge into the film, and consider whether the specimen alignment is loading the material unevenly. Document the jaw break, exclude it per your protocol's rules, and re-run. What you can't do is report a jaw-break value as a seal-strength result and move on.
Specimen Slippage
Slippage is the opposite problem: the specimen slides in the grips before or during the peel. On the force curve, slippage shows up as a drop or a flat, non-physical region that doesn't represent the seal separating. You'll often catch it by watching the test — the tail creeps out of the jaw.
Slippage understates the peel force because some of the crosshead travel went into the specimen sliding, not the seal peeling. Causes are usually grip pressure too low, contaminated or worn jaw faces, or a slick film that needs serrated or rubber-faced grips. Fix the grip setup before you trust the data.
The practitioner's habit here is simple: watch every test, and treat the force curve and the visual failure together. A clean peel curve with a documented mode of failure at the seal is a valid result. A number with a jaw break or slippage behind it is noise wearing a data point's clothes.
Interpreting Peel Force Against ISO 11607-1
Now the part that connects the bench to the submission.
F88 requires you to calculate and report seal strength, with average seal strength and maximum seal force both commonly derived from the force-versus-extension curve (ASTM F88/F88M-21). The average tells you the sustained peel force across the seal; the maximum tells you the peak. Both matter, and your acceptance criteria should be explicit about which one you're judging against.
ISO 11607-1 requires that peelable seals demonstrate a minimum specified seal strength and that the sterile barrier system be validated — and it puts the burden on you to define and justify your seal-strength acceptance criteria (ISO 11607-1:2019). There is no universal numeric peel-force threshold in the standard. The threshold is product- and material-specific, and you have to show your reasoning.
That's the point people miss. ISO 11607-1 Annex B lists more than 100 recognized test methods and permits alternatives, but inclusion of F88 in that list does not establish your acceptance criteria for you (ISO 11607-1:2019, Annex B). The standard hands you the method; you still have to set and defend the number.
So a defensible interpretation looks like this:
- Report the support technique and crosshead speed alongside every result — without them the peel value isn't interpretable.
- State whether your criterion is average peel force, maximum force, or both.
- Justify the threshold against the seal's function: a peelable seal needs to be strong enough to maintain the barrier through distribution and aging, but weak enough to open without tearing the lid or shedding fibers into the sterile field.
- Report the distribution, not just the mean. The minimum value in your data set is what tells you whether a worst-case unit still clears the floor.
That last point is the one I push hardest in the lab. Treat your seal-strength spec as a floor that has to hold through distribution simulation and aging — not a number you hit once on fresh samples. Run F88 on conditioned specimens, not just pristine ones, and the average looks less reassuring once you see where the bottom of the distribution lands.
Where F88 Sits in the Validation Sequence
Seal-strength testing doesn't stand alone. Acceptance and process-control results from F88 feed the process validation and finished-device acceptance activities a compliant quality system requires (ISO 13485:2016 Clause 7.5.6 and Clause 8.2.6). Both ISO 11607-1 and ISO 11607-2 are FDA-recognized consensus standards and harmonized under EU MDR (FDA Recognized Consensus Standards Database) — so the data you generate here lands directly in your submission and your design history file.
In practice, F88 is part of seal-process qualification (your IQ/OQ/PQ on the sealer) and a recurring control test through the package's life — fresh, post-distribution, and post-aging. The number on the curve only earns its place in the file when the fixture, speed, and failure mode behind it are all documented and clean.
Get the setup right, watch every test, report the technique and speed, and justify your acceptance criteria against the seal's actual function. Do that and F88 becomes a strength in your validation package instead of a question for the reviewer.