A datasheet value is either guaranteed, typical, or decorative, and the sheet--Aijiren Sample Vials

A customer sent me a COA last March for a lot of 9 mm short-thread vials that showed residual seal force at 18 N, comfortably above their internal 12 N floor, yet their autosampler needle was bending on roughly 1 in 300 injections. The number on the sheet was real. What the sheet did not say was that the 18 N was a mean of ten closures pulled on a single day, measured 24 hours after capping, with no confidence interval and no statement of the test standard. The oem vial datasheet specifications and what each value means is the difference between a purchasing decision and a guess. Most buyers read the sheet as a list of facts. It is actually a mix of tested values, calculated values, and marketing values, and the sheet rarely tells you which is which.

I have spent fifteen years on the manufacturing side of this, writing and reading these sheets, and I can tell you the format is not standardized. There is no ISO template for a vial datasheet. Two suppliers can publish what looks like the same document and mean completely different things by “capacity,” “CV,” or “residual seal force.” The job of the procurement manager is to read past the number and find the method behind it. This article is about how to do that, line by line, for the oem vial you are about to qualify.

A datasheet value is either guaranteed, typical, or decorative, and the sheet rarely tells you which

Before you read a single number, sort every line on the sheet into one of three buckets. This is the discipline that separates a useful datasheet from a glossy one.

Guaranteed (specification limit). The supplier commits that every unit in the lot falls inside the stated range, and there is a defined test method, sample size, and acceptance criterion. A failure is a nonconformance you can reject against.
Typical (nominal or representative). A value from a design calculation or a historical average. It describes what the product usually is, not what this lot is. It is not a rejectable limit.
Decorative. A number with no method, no sample size, and no tolerance, often rounded to a suspiciously clean figure. “Chemical resistance: excellent” and “inert” live here.

The trap is that guaranteed and typical values sit side by side in the same table, in the same font, with no marker. A dimensional tolerance of 12.00 ± 0.10 mm on the outside diameter is a guaranteed value with a real gauge behind it. A “residual volume” of 170 µL for a 2 mL vial is usually a calculated nominal, not a measured one, and it will vary by 20 to 40 µL depending on the bottom geometry and how the needle is positioned. If you treat the second like the first, you will design a method around a number that was never tested.

Ask for the control plan, not just the COA

The COA tells you what this lot measured. The control plan tells you what is measured on every lot, at what frequency, with what instrument, and against what limit. I have seen suppliers who measure residual seal force on every lot and suppliers who measure it once per molding tool per quarter. Both publish a number. Only one of them is telling you something about the vial in your hand. When you request qualification documents, ask for the control plan and the gauge R&R for the critical dimensions. If the supplier cannot produce a gauge repeatability study, the tolerance on the sheet is aspirational.

Which dimensional lines are actually gauged, and which are copied from the drawing

Dimensions are where buyers get most confident and most wrong. A drawing tolerance is a design intent. Whether the shipped part is measured against it depends on the inspection plan.

The critical dimensions on a 12 x 32 mm vial are the outside diameter, the height, the neck finish (thread form and thread start), the internal bore at the septa seat, and the bottom geometry. Of these, the outside diameter and height are almost always gauged, because they are cheap to check with an air gauge or a laser micrometer and they matter for autosampler grippers. The neck finish is the one that gets copied from the drawing and rarely measured, and it is the one that causes crimp failures and septa that do not seat.

Here is how I read the dimensional block on a sheet:

Spec sheet line	Usually guaranteed?	How it should be measured	What to ask
Outside diameter, e.g. 12.00 ± 0.10 mm	Yes	Laser micrometer or air gauge, 100% or AQL sample	Gauge R&R, sample size, sampling frequency
Height, e.g. 32.00 ± 0.20 mm	Yes	Optical comparator or height gauge	Same as above
Neck finish / thread form	Often no	Go/no-go thread gauge, profile projector	Is a thread gauge used at all?
Internal bore at septa seat	Sometimes	Pin gauge or CMM section	Tolerance and method
Bottom geometry / residual volume	Rarely	Gravimetric fill-and-withdraw test	Is 170 µL measured or calculated?
Wall thickness	Rarely	Sectioning and optical measurement	Nominal or limit?

If the sheet lists a neck finish tolerance but the supplier cannot tell you what gauge checks it, treat that line as nominal. The vial closures you pair with the vial depend on that thread form, and a closure that does not engage correctly will show up as a capping torque problem long before anyone blames the neck. GPI thread standards (the Glass Packaging Institute finish designations, e.g. 13-425, 15-425, 18-400) define the thread, but they do not define how tightly your supplier holds it. That is a separate question you have to ask.

Residual seal force is the most abused number on the sheet

Residual seal force (RSF) is the force required to lift a septum off the vial after capping, measured on a tensile tester. It is the single best predictor of whether a closure will hold pressure and whether a needle will meet a consistent septa surface. It is also the number most often published without its method.

RSF depends on at least five variables: the closure material and hardness, the capping torque or crimp force, the time between capping and measurement, the temperature and humidity at measurement, and the tester’s crosshead speed. Change any one and the number moves. A sheet that says “RSF: 18 N” without stating the test standard, the crosshead speed, the dwell time, and the conditioning is telling you almost nothing.

Here is what a defensible RSF line looks like:

Method: referenced to a published standard, for example a USP <671> containment test or an internal SOP with a stated tensile protocol.
Crosshead speed: stated, typically 50 mm/min or 200 mm/min. The number changes with speed.
Sample size and statistic: n ≥ 10, mean and standard deviation, or a lower specification limit. An 18 N mean with a 4 N SD is a different product from an 18 N mean with a 1 N SD.
Dwell time: RSF measured at 1 hour and 24 hours after capping, because some elastomer blends relax.
Conditioning: 23 °C / 50% RH per ASTM D618 or equivalent.

The analytical testing lab that runs your qualification should be able to reproduce the RSF within a few newtons. If they cannot, the supplier’s method is not documented well enough to audit. I have walked into a supplier’s QC room and found RSF measured on a hand-pulled spring scale. The number was honest. The method was not.

Field case: the 18 N that was not 18 N

That March lot I mentioned at the top. The customer’s needle-bending problem traced back to RSF, but not the mean. We pulled 30 closures from the lot and measured RSF at 1 hour and 24 hours. The 24-hour mean was 17.6 N, close to the sheet. The 1-hour mean was 9 N, with a minimum of 5 N. The supplier had measured at 24 hours only, which is the correct protocol for the finished product, but the customer was capping and injecting within 90 minutes. In that window the closure was still relaxing, the septum was not fully seated, and the needle was meeting a soft surface at an angle. The fix was a supplier change to a closure with a faster-setting compound and a documented 1-hour RSF floor of 12 N. The datasheet now carries both time points.

Chemical resistance and “inert” are not values, they are claims

Every datasheet has a line like “chemically inert” or “compatible with common solvents.” Neither is a number, and neither is testable as written. What you want instead is a compatibility table with a stated exposure time and a stated failure criterion.

Ask for a table that lists, for each solvent class (aqueous, alcohol, ketone, chlorinated, aromatic), a pass/fail after a defined exposure, for example 24 hours at 40 °C, with the failure criterion being extractables above a stated threshold measured by GC-MS or LC-MS. If the supplier cannot produce that, the “inert” claim is decorative. For most buyers the practical move is to run your own extractables screen on the actual solvent you use, at the actual temperature, and treat the datasheet claim as a starting hypothesis. The U.S. EPA SW-846 methods give you a defensible extraction and analysis framework if you need one.

One more thing about the chemical block: it usually ignores the closure. The vial body may be borosilicate 1.1 (USP Type I) and genuinely inert, while the silicone or butyl septum is the source of every extractable you will see. Read the closure datasheet as carefully as the vial datasheet. The chromatography vial is a system, and the sheet that matters is the one for the part that touches your sample.

Extractables and leachables: which USP chapters the sheet should cite, and which it should not

A datasheet that cites USP <661> is telling you the glass passed the hydrolytic resistance test for its type. A datasheet that cites USP <381> is telling you something about the elastomer. A datasheet that cites USP <788> for particulates is telling you about a light obscuration or microscopic particle count. All three are real tests with real acceptance criteria, and all three are worth having.

What you should be suspicious of is a sheet that cites a USP chapter without stating the test conditions. USP <661> hydrolytic resistance depends on the autoclave cycle and the titration method. A “passes USP <661>” with no method is a claim, not a result. Ask for the actual report: the powder or surface test, the titration volume, the acceptance limit, and the date.

For leachables, the sheet often has nothing, and that is sometimes honest. Leachables depend on your drug or sample matrix, your contact time, and your storage temperature. No supplier can guarantee a leachable profile for a matrix they have never seen. What a good sheet gives you is the extractables data (a controlled extraction study, typically in water, ethanol, and a non-polar solvent, analyzed by GC-MS and LC-MS) so you can assess risk yourself. If the sheet has no extractables data at all, that is a gap you should price into your qualification budget.

Lot-to-lot variability is a number the datasheet usually omits

Every value on the sheet describes a typical unit or a specification limit. None of them describes how much lot A differs from lot B. That is the number that actually drives your process capability, and it is almost never published.

You can estimate it from the COA history. Ask for the last 20 COAs and plot the RSF, the outside diameter, and the height. If the outside diameter moves by 0.05 mm between lots and your tolerance is ±0.10 mm, you have half your window consumed by supplier variation before your own process adds anything. If the RSF mean moves by 3 N between lots, your capping process has to absorb that. I have seen a supplier whose RSF mean drifted from 16 N to 22 N over six months as they changed elastomer compound lots, with every individual COA passing. The datasheet never changed. The process did.

Field case: the drifting diameter that passed every COA

A pharma customer running a 12 x 32 mm vial on a robotic gripper started seeing pick failures at about 1 in 500. Every COA showed outside diameter within 12.00 ± 0.10 mm. We pulled retained samples from six consecutive lots and measured 100 vials each on a laser micrometer. The lot means were 11.94, 11.96, 11.97, 11.99, 12.01, and 12.03 mm. Every lot passed. The trend was real, and it tracked a change in mold cooling time at the supplier. The gripper was set for 11.97 mm nominal and had a 0.04 mm grip window. The fix was two-part: the supplier agreed to a tighter lot-mean target of 12.00 ± 0.03 mm with the ±0.10 mm limit retained as the individual-unit tolerance, and the customer widened the gripper window to 0.08 mm. The datasheet now carries both the individual limit and the lot-mean target. That second number is the one that prevents the failure.

Temperature ratings: the number on the sheet is a material limit, not a use limit

A sheet that says “maximum temperature: 200 °C” is telling you the glass did not deform in a test. It is not telling you the closure survived, the septum held, or the sample did not degrade. Borosilicate 1.1 glass will take 200 °C and higher without complaint. The silicone septum may be rated to 200 °C but will outgas and lose reseal performance well before that. The butyl closure may be rated to 120 °C and will flow above it.

Read the temperature line as a system question. If your method requires heating to 150 °C, ask for the RSF and the reseal data at 150 °C, not the material’s rated maximum. Ask for the autoclave cycle data if you sterilize. Ask for the reseal count after the thermal exposure, not before. A “rated to 200 °C” with no post-exposure reseal data is a decorative number.

Puncture count and reseal: the closure sheet, not the vial sheet, owns this

Puncture count (how many times a septum can be pierced before it leaks) and reseal (how well it closes after withdrawal) are closure properties, but they appear on vial datasheets often enough that buyers get confused. The vial body has no puncture property. If a vial datasheet quotes a puncture count, it is quoting the closure it was tested with, and you need to know which closure.

The number itself is method-dependent. A septum rated for 10 punctures was tested with a specific needle gauge, at a specific speed, with a specific leak criterion (usually a pressure decay or a dye ingress test). Change the needle from 22 gauge to 26 gauge and the count changes. Change the leak criterion and it changes again. Ask for the needle gauge, the puncture pattern (same point or distributed), the leak test method, and the acceptance criterion. For GC methods with a single injection per vial, puncture count is irrelevant. For a headspace method that re-samples, it is the number that decides your septum.

A reading checklist you can run in fifteen minutes

When a new datasheet lands on your desk, run this sequence before you send it to anyone else.

Find the method column. If there is no method column, the sheet is a brochure. Ask for the controlled document.
Sort every line into guaranteed, typical, or decorative. Mark the typical and decorative lines and do not design around them.
Check the critical dimensions for gauge and sampling. Outside diameter and height should have a stated gauge and an AQL or 100% inspection. Neck finish should have a go/no-go gauge.
Check RSF for method, speed, dwell, sample size, and SD. If any of the five is missing, request it.
Check the temperature line for post-exposure reseal data. Material maximum is not a use limit.
Check the chemical block for exposure time and failure criterion. “Inert” is not a value.
Request 20 COAs and plot the critical values. Look for lot-to-lot drift, not just pass/fail.
Ask for the control plan and one gauge R&R. This is the difference between a sheet and a system.

If the supplier can answer all eight, you have a datasheet you can qualify against. If they can answer five, you have a starting point and a list of questions. If they can answer two, you are buying on price and hoping. That is a legitimate business decision, but make it consciously.

For reference on what a well-documented consumable program looks like, the technical libraries from Agilent, Thermo Fisher, and Restek are worth reading, not for their vial specs but for the way they document test methods and acceptance criteria on their own consumables. The format is the lesson.

The decision rule is simple. A number without a method is a marketing claim, and a claim without a sample size and a standard deviation is not a specification. Buy only against values that carry a method, a sample size, and a rejectable limit, and treat every other number on the sheet as a description of the product rather than a promise about the lot in the box. When two suppliers look identical on paper, the one who can show you the control plan and the gauge study is the one whose sheet you can actually use.

References & further reading

Related resources from our lab consumables network:

External references and standards: