A buyer sent me a 2 mL vial datasheet last March with a line that read “septum puncture resistance: 5 punctures.” I asked which needle gauge, which septum durometer, and whether that was a mean or a minimum. Nobody at the supplier could answer. The lot we eventually bought averaged 3.2 punctures before coring on a 22-gauge needle, and we lost a 96-position autosampler run at 4 a.m. because of it. That datasheet was not lying. It was reporting a nominal number with no test method attached, and the buyer had no way to know. Reading a 2ml vial datasheet is a skill, and most procurement teams have never been taught it.

This is not a piece about which vial to buy. It is a piece about how to read the numbers on the sheet in front of you, so you can tell which ones are guaranteed by a test method and which ones are just a drawing dimension or a nominal callout. If you already know your vial works in your instrument, the question left is whether the next lot will. That answer lives on the datasheet, but only if you can tell the tested values from the typed ones.

Every number on a 2 mL vial datasheet is one of three things: a tested value, a nominal value, or a derived value. The supplier rarely labels them. You have to infer the category from the language and from what the number physically is.

A tested value comes from a measurement performed on a sample from the lot. It has a method, a sample size, an instrument, and a distribution. “Crimp seal integrity: 100% of samples pass dye ingress at 1.5 bar for 30 s” is a tested value. “Inside diameter: 12.0 mm” is a nominal value, because it comes from the mold cavity drawing and is only checked at incoming inspection or during a first-article run. “Headspace volume: 1.2 mL” is a derived value, calculated from a nominal ID and a nominal fill height, and it is only as good as the two nominals that fed it.

The trap is that nominal values look just as precise as tested ones. A drawing callout of 12.00 ± 0.15 mm is not a promise. It is a target with a tolerance band that the forming station is supposed to hold closed-loop. Whether the actual parts sit at 12.00 or at 11.86 depends on the mold, the glass lot, and the day. If the sheet gives you a tolerance but no Cpk or process capability figure, you are looking at an intent, not an outcome.

My rule of thumb: if the datasheet does not tell you how many pieces were tested, treat the number as nominal. If it tells you how many but not the method, treat it as typical. If it gives you both, you can treat it as a guarantee and hold the supplier to it.

Why “typical” and “minimum” are not the same word

A “typical” value is a mean or a median from a qualification run. A “minimum” or “guaranteed” value is a lower bound you can reject against. On a crimp vial, “seal integrity typical: 2.0 bar” means the average of the qualification batch held 2.0 bar, and half the batch held less. “Seal integrity minimum: 1.5 bar” means every piece in the qualification batch held 1.5 bar. If your method needs 1.8 bar to survive a hot autosampler tray, the typical number is useless to you and the minimum number is the one to compare against.

I have watched a lab switch suppliers because the old sheet said “typical 2.0 bar” and the new sheet said “minimum 1.8 bar.” The new vial was actually the safer choice, but the buyer read “2.0 vs 1.8” as a downgrade. That is the cost of not knowing the difference between the two words.

Dimensional lines are nominal unless a Cpk or a first-article report backs them

The dimension block on a 2 mL vial datasheet usually lists five values: outside diameter, inside diameter, height, neck OD, and neck ID (or crimp ring diameter). These come from the mold drawing. They are checked at first article and then sampled at a rate the supplier chooses, often 5 pieces per 8-hour shift or less. That is a process check, not a guarantee.

What you actually need is the critical dimension for your instrument, and the datasheet does not know which one that is. On a standard 12 × 32 mm vial for a 2 mL autosampler position, the numbers that matter to the instrument are usually the outside diameter (for gripper clearance), the height (for needle depth), and the neck geometry (for crimp or screw closure engagement). The 2 mL vial family on our site lists these as nominal with tolerances, and I tell buyers to ask for the first-article dimensional report before they qualify a new lot.

A first-article report is the only document that shows you the actual measured distribution. Ask for 30 pieces minimum, measured with a calibrated optical comparator or a CMM, and ask for the mean and the range on each critical dimension. If the supplier cannot produce one, the tolerance on the sheet is a hope.

The neck geometry line is where most “compatible” claims fall apart

The neck OD and crimp ring diameter determine whether your crimper will seal the vial and whether your autosampler gripper will pick it up. A datasheet that lists “neck OD: 11.0 mm” with no tolerance is telling you nothing. The crimp seal needs the ring diameter held within about ±0.10 mm for a standard 11 mm aluminum seal to deform correctly. If the sheet is silent on that tolerance, ask.

I have seen a lot where the neck OD drifted to 11.15 mm nominal-plus and the crimper started producing seals that passed dye ingress at 1.0 bar but failed at 1.5 bar. The datasheet tolerance was ±0.15 mm, so the parts were “in spec.” The method needed 1.5 bar. The fix was a tighter incoming inspection on neck OD (we added a go/no-go gauge at 11.10 mm) and a conversation with the supplier about their forming station setpoint.

Chemical and thermal lines are almost always typical, not guaranteed

When a datasheet says “chemical resistance: suitable for HPLC solvents” or “maximum temperature: 120 °C,” it is almost never giving you a tested value. It is giving you a category judgment from a compatibility chart, usually borrowed from the resin supplier’s data, not from a test on the finished vial.

The two lines that matter most here are the maximum continuous service temperature and the extractable/leachable profile. On temperature, a borosilicate 2 mL vial rated to 120 °C continuous is fine for a hot autosampler tray at 40 °C and fine for a 100 °C derivatization step. It is not fine for repeated autoclave cycles at 134 °C, even if the sheet’s “maximum” line implies otherwise. The maximum is a service temperature, not a sterilization rating. If you need autoclave cycles, ask for the actual cycle test data, and expect it to be a separate qualification document.

On extractables, a datasheet that says “low extractables” is telling you nothing testable. You need the actual USP <661> or USP <1663> data, or a study to ICH Q3D for elemental impurities, with the test conditions (solvent, time, temperature) stated. If the sheet hides the method behind “low,” treat the line as marketing. The HPLC consumables side of our business sees this constantly: a vial that is “low extractable” in water at 25 °C may be a different animal in acetonitrile at 60 °C.

Septum lines: durometer, thickness, and puncture count need three separate numbers

Septum specs are where datasheets are most often incomplete. A proper septum line has at least three values: durometer (Shore A), thickness (mm), and puncture count (with needle gauge and injection volume stated). If any of the three is missing, the puncture count is not comparable to any other supplier’s number.

A 45 Shore A silicone septum at 1.0 mm thickness on a 22-gauge needle might give you 8 to 12 punctures before coring. A 55 Shore A silicone/PTFE laminate at 1.5 mm might give you 4 to 6 on the same needle, because the laminate is stiffer and the PTFE face is what the needle actually cuts. Neither number is better. They are different products for different methods. The datasheet needs to tell you which one you are reading.

For the instrument side, Agilent publishes needle and septum guidance in their GC columns and accessories documentation that is worth reading alongside any vial datasheet, because it tells you what the autosampler expects from the septum geometry.

A side-by-side reading of two datasheets for the same nominal part

Here is the exercise I run with new buyers. Two suppliers, same nominal 2 mL crimp vial, same 11 mm aluminum seal. Read the lines side by side and see which ones are tested and which are typed.

Both sheets describe a vial that will probably work. Only one of them lets you reject a lot that does not. That is the entire difference between a datasheet you can buy against and a datasheet you can only hope against.

Field case: the puncture count that was really a needle gauge problem

A contract lab running a 96-well autosampler started seeing coring at position 47 almost every run, and only at that position. The vial datasheet said “puncture count: 10” with no needle spec. The lab was using a 20-gauge needle on a 10 µL injection, which is heavier than the 22-gauge the septum was qualified on. The septum was not the problem. The needle was.

The measurement that identified it: we ran the same lot of septa on a 22-gauge needle and got 9.4 punctures average, and on the 20-gauge we got 5.1. The datasheet number was real, but it belonged to a different needle. The fix was to switch that method to a 22-gauge needle and to add the needle gauge to the incoming inspection record. The vial was fine. The reading of the datasheet was not.

This is the most common failure I see in vial procurement: a tested value that is true but not applicable, because the test conditions were never printed. Any supplier that gives you a puncture count without a needle gauge is giving you half a number.

Field case: the “12.0 mm” that was 12.0 mm on average

A pharma QC lab qualified a new lot of 2 mL vials against a datasheet that said “outside diameter 12.0 mm.” The autosampler gripper started dropping vials about once every 200 picks. The datasheet had no tolerance and no Cpk. We measured 50 pieces with a calibrated comparator and got a mean of 12.04 mm and a range of 11.91 to 12.18 mm. The gripper on that instrument needed the OD under 12.10 mm to close reliably. About 15% of the lot was over.

The fix was two-part: the supplier tightened their forming station setpoint to bring the mean down to 11.98 mm, and the lab added a go/no-go gauge at 12.08 mm to incoming inspection. The datasheet line was not a lie. It was a nominal with no distribution attached, and the lab had assumed a nominal was a guarantee. If you want to see what a datasheet with the distribution attached looks like, the OEM vial documentation we produce for private-label customers includes the Cpk on the critical dimensions as standard.

Spotting a datasheet that hides its test method

There are a handful of tells. Once you know them, you can triage a datasheet in about 90 seconds.

• No sample size (n). A tested value without an n is a typical value dressed up as a guarantee.
• No method reference. “Passes dye ingress” with no pressure, time, or standard is not a test. “USP <661>” or “ASTM D2240” is.
• No tolerance on a dimension. A single number on a dimensional line is a target, not a spec.
• “Typical” or “nominal” used as the value. Fine for a drawing, useless for a purchase order.
• Vague adjectives in place of numbers. “High,” “low,” “excellent,” “superior” are all tells that the supplier did not test.
• Temperature with no duration. “120 °C” means nothing without “for 24 h” or “continuous.”
• Extractables with no conditions. Solvent, time, and temperature must all be stated, or the number is not comparable.
• No revision date or lot traceability. A datasheet without a revision is a datasheet you cannot hold anyone to.

I keep a one-page checklist taped inside the procurement binder. When a new supplier sends a sheet, I run it against the list before I run it against the price. A sheet that fails three or more of these items goes back with a request for the missing data. Half the time the supplier has the data and just did not print it. The other half, they do not have it, and that is the answer you needed.

What to put in a purchase order so the datasheet becomes enforceable

The datasheet only protects you if the PO references it. Here is the language I use, and it has saved us two lot rejections in the last three years.

• Reference the datasheet revision by date and number. “Per datasheet rev. 2024-03, lines 4 through 9.”
• Name the tested values as acceptance criteria. “Crimp seal integrity minimum 1.5 bar, dye ingress, n=50 per lot, 30 s hold.”
• State the critical dimensions and their tolerances. “OD 12.00 ± 0.10 mm, height 32.0 ± 0.20 mm, neck OD 11.00 ± 0.10 mm.”
• Require a certificate of analysis per lot with the measured values, not just a “pass” stamp.
• Require a first-article dimensional report on the first lot and on any mold or forming-station change.
• State the sampling plan. ANSI/ASQ Z1.4 general inspection level II, AQL 1.0 on critical dimensions, is a reasonable default.

If you are buying through a distributor, ask them to pass the same language upstream. The chromatography supplier network we work with will do this on request, but you have to ask. A distributor’s own datasheet is often a reformatted copy of the manufacturer’s, and the tested values sometimes get dropped in the reformatting.

A quick note on standards to cite

If you are writing acceptance criteria, cite the standard that matches the test. For glass vial dimensional work, the relevant references are the GPI 12 mm finish conventions and the instrument maker’s own dimensional guidance. For extractables, USP <661> and USP <1663> are the chapters that reach the vial itself. For elemental impurities, ICH Q3D. For environmental methods, the U.S. EPA SW-846 series names the vial in several sample-handling sections, and the Thermo Fisher chromatography and sample handling documentation and Restek technical resources both publish instrument-side guidance on what the autosampler expects. None of these tell you what a specific supplier’s datasheet means. They tell you what to ask for.

The decision rule I give every new buyer

If a datasheet line has a test method, a sample size, and a tolerance or minimum, treat it as a guarantee and write it into the PO. If it has any two of the three, treat it as typical and ask for the third. If it has fewer than two, treat it as a drawing note and do not buy against it. That rule alone will catch about 80% of the misleading lines on a 2 mL vial datasheet, and it will tell you in one pass whether the supplier in front of you is selling a tested product or a typed one.

References & further reading

Related resources from our lab consumables network:

• 2Ml Vial at 2mlsamplevial.com
• Hplc Consumables at hplcconsumable.com
• Chromatography Supplier at chromatographysupplier.com
• Oem Vial at oemsamplevials.com

External references and standards:

• Agilent chromatography consumables and columns
• Thermo Fisher chromatography and sample handling
• U.S. EPA SW-846 test methods for waste and materials
• Restek technical resources for chromatography