The four dimensions that actually decide fit--HPLC Autosampler Vials Supplier,Manufacturer,and Factory

I have watched a lab buy 20,000 vials that would not seat in their own autosampler. The purchaser ordered 12 x 32 mm clear glass vials with 9 mm screw threads, which is the most common format on earth. The instrument was an older Shimadzu with a tray cut for 11 mm crimp necks. The vials fit the rack, sort of, and then the needle missed the septum on the third injection and bent. Nobody checked the neck finish against the autosampler spec sheet before the PO went out. That mistake cost about $4,000 in scrapped glass and a week of downtime. It is also entirely avoidable, because vial dimensions are published, standardized, and boring. The only hard part is matching them to the instrument you actually own.

This guide is about that match. Not what a vial does in the abstract, but the specific dimensional and material decisions you make when you sit down with an autosampler manual, a caliper, and a supplier catalog. I will give you the numbers, the failure modes that come from getting them wrong, and a decision checklist you can hand to procurement.

Every autosampler vial is defined by four measurements. Get all four right and the vial will work. Get any one wrong and you have a problem that ranges from annoying to instrument-destroying.

1. Body diameter

The two standards you will encounter are 12 mm and 12 x 32 mm nominal (the 32 mm is height). The 12 mm outside diameter is the de facto global standard for HPLC autosamplers and is referenced in the general chapters of USP <1058> on analytical instrument qualification. A few older or specialized systems use 8 mm (some micro-volume autosamplers) or 6 mm (rare, mostly microsampling). If your autosampler tray has round wells, the well diameter tells you the vial diameter. Measure it. A 12.0 mm well will not accept a 12.2 mm vial, and glass tolerances on cheap vials run ±0.2 mm easily.

2. Height

The standard is 32 mm from base to the top of the crimp or screw cap. Total height with cap installed is typically 32 mm for crimp and 32 mm for short-thread screw, but tall versions exist. A short screw vial is 32 mm; a tall screw vial is about 38 mm. If your autosampler has a fixed needle depth or a gripper that lifts vials by the cap, a tall vial in a short-vial tray will jam the gripper. I have seen a robotic arm stall and throw an error on a 38 mm vial in a 32 mm position. The arm did not break, but the run stopped at 2 a.m.

3. Neck finish

This is where most mismatches happen, and it is the dimension people ignore. The neck finish is the thread or crimp geometry at the top of the vial. There are four you will meet:

11 mm crimp: the classic. Aluminum seal crimped over a flange. Requires a crimper and decapper. Standard on older Agilent, Shimadzu, and Waters systems.
9 mm short-thread screw: the current default for most new HPLC autosamplers. Thread pitch and depth are standardized across major vendors so a 9 mm screw cap from one supplier fits a vial from another.
10-425: a wider screw thread used on some GC and headspace vials. Not interchangeable with 9 mm.
13-425: used on larger 15 x 45 mm and 15 x 85 mm vials, common in GC and some fraction collectors.

The thread callout (9 mm, 10-425, 13-425) is a GPI (Glass Packaging Institute) style designation. The first number is the nominal diameter in millimeters; the second is the thread finish code. A 9 mm cap will not seal a 10-425 vial, and forcing it will cross-thread the plastic and leak. If you are unsure, the vial spec sheet from your supplier should list the GPI finish explicitly.

4. Septum and cap thickness

The cap adds 1.5 to 2.5 mm to the total height depending on style. A crimp seal with a 1.0 mm PTFE/silicone septum sits lower than a screw cap with a 2.0 mm septum and a thick plastic overcap. If your autosampler needle depth is set to pierce to a fixed Z-height, a thicker cap means the needle goes shallower into the vial and may not reach the liquid. This is a real cause of “no sample injected” errors on systems with fixed needle penetration.

Matching to specific autosampler families

The published specs vary by vendor and generation. Here is what I have found in practice, with the caveat that you should always confirm against your own instrument manual or the vendor’s consumables page.

Agilent 1100/1200/1260/1290 Infinity

These use 12 x 32 mm vials. The standard tray accepts 9 mm screw or 11 mm crimp depending on the tray insert. The 1260 and 1290 autosamplers with the standard 100-vial tray are cut for 9 mm screw vials; the crimp tray is a separate part. Agilent publishes vial and cap compatibility in their chromatography consumables catalog, and it is worth downloading the PDF rather than trusting a distributor’s shorthand. Needle penetration on the 1290 is adjustable in the method; set it so the needle tip sits 1 to 2 mm above the vial bottom to avoid bending on the glass.

Thermo Scientific Vanquish and UltiMate 3000

Both use 12 x 32 mm vials with 9 mm screw caps. The Vanquish tray is molded for 9 mm; a crimp vial will physically fit but the cap height changes the gripper clearance. Thermo’s sample handling documentation, available through their chromatography and sample handling pages, lists the exact cap part numbers that are validated for each tray. I have seen a Vanquish reject a run because a third-party cap was 0.4 mm taller than the validated part and the gripper sensor flagged it. Stick to the validated cap height or measure and adjust.

Shimadzu Nexera and Prominence

Shimadzu uses 12 x 32 mm vials. The Nexera series is predominantly 9 mm screw; older Prominence units often shipped with 11 mm crimp trays. The tray is swappable, so the question is which tray is installed, not which instrument you bought. Look at the part number on the tray itself. This is the single most common source of the mismatch I described at the top.

Waters Alliance and ACQUITY

ACQUITY uses 12 x 32 mm with 9 mm screw in the standard sample manager. The Alliance e2695 historically used 11 mm crimp or 9 mm screw depending on the carousel. Waters publishes the carousel part numbers; a 9 mm carousel will not hold an 11 mm crimp vial securely, and the vial will tip during needle descent.

GC and headspace inlets

GC autosamplers are a different animal. Liquid injection GC vials are typically 11 mm crimp or 9 mm screw with a 2 mL nominal capacity, but the inlet itself (the injector port) does not care about the vial. What matters is the autosampler tower. A GC vial for a liquid autosampler is usually 11 mm crimp; check the tower’s vial rack. Headspace vials are larger: 20 mm crimp with a 10 mL or 20 mL capacity is the standard for headspace GC. The headspace vial must match the headspace sampler’s oven carousel, and the crimp diameter (20 mm) is not negotiable. A 20 mm vial in an 18 mm carousel will not seat and the sampler will fault.

A dimensional decision checklist with numbers

Hand this to whoever writes the PO. It takes five minutes and prevents the four-figure mistake.

Body diameter: measure the tray well. Accept only vials within ±0.15 mm of nominal (12.00 mm for standard). Reject lots that caliper outside that.
Height: confirm 32 mm base-to-cap-top for short vials, 38 mm for tall. Measure three vials per case with a caliper; glass height variation above 0.3 mm causes gripper faults.
Neck finish: read the GPI callout on the vial carton. Match it to the tray or carousel part number. Do not substitute 9 mm for 11 mm or 10-425 for 9 mm.
Cap height: total capped height should match the validated part within 0.5 mm. If you change cap suppliers, measure first.
Septum thickness: 1.0 mm for crimp, 1.5 to 2.0 mm for screw. Thicker septums reduce coring but raise the cap and change needle depth.
Needle penetration: set so the tip sits 1 to 2 mm above the bottom. On adjustable systems, verify with a dummy vial and a depth gauge.
Temperature rating: borosilicate glass is fine to 500 °C; the cap and septum are the limit. Silicone/PTFE septa are rated to about 200 °C; plain silicone to 180 °C. For headspace at 150 °C, use a PTFE-faced silicone septum and a crimp seal, not a screw cap.
Puncture count: a 1.0 mm silicone septum tolerates roughly 3 to 5 punctures before coring risk rises. A PTFE/silicone laminated septum handles 5 to 10 on a single injection port. Replace septa on a schedule, not on failure.
Crimp torque: a manual crimper should close an 11 mm aluminum seal to a consistent crimp height of 5.5 to 6.0 mm measured at the seal skirt. Too loose leaks; too tight cracks the neck. Use a crimp gauge, not feel.

That list is the whole game. The rest is confirming it against your instrument.

Where the standards fit (and where they do not)

Vial dimensions are not governed by a single ISO standard the way, say, a pipette tip is. The 12 x 32 mm format is a de facto industry convention, not an ISO number. The GPI thread finishes (9 mm, 10-425, 13-425) are the closest thing to a formal spec, and they are widely respected by major glass manufacturers. USP <1058> covers instrument qualification and touches on sample containers but does not fix dimensions. EPA methods under SW-846 specify sample preservation and holding times but leave the vial to the laboratory. That means the burden is on you to verify fit, because nobody is going to do it for you at the regulatory level.

What this means practically: buy from suppliers who publish a dimensional drawing with tolerances. If a catalog page does not show the neck finish, body OD, height, and septum thickness, ask for the drawing. A supplier who cannot produce it is a supplier who has not measured their own product. I have walked away from two vendors on that basis alone.

Material choices that interact with dimensions

Dimensions decide fit. Material decides whether the fit survives the run. The two interact more than people think.

Glass type

Type I borosilicate (USP Type I) is the default for HPLC and GC. It has low extractables and a low coefficient of expansion, so it survives autoclaving and temperature cycling without cracking at the neck. Type III soda-lime glass is cheaper and fine for aqueous samples at room temperature, but it leaches alkali and will raise the pH of unbuffered samples over hours. For trace metals or ion chromatography, use Type I and pre-rinse with 10% nitric acid. For a 12 x 32 mm vial, the price difference between Type I and Type III is a few cents per vial. Pay it.

Amber vs clear

Amber glass blocks UV below about 400 nm. If your analyte is photolabile (vitamins, some pesticides, many pharmaceuticals), amber is not optional. The dimensional specs are identical; only the glass color changes. Do not assume amber is available in every neck finish. Some 9 mm short-thread vials come only in clear; check before you standardize.

Septa

The septum is the most failure-prone component and the one most often specified by habit. A PTFE/silicone laminated septum has a PTFE face toward the sample (inert, low extractables) and silicone behind it (reseals after puncture). For HPLC with aqueous/organic mobile phases, that is the right default. For GC with a hot inlet, a silicone/PTFE septum with a higher durometer reduces bleed at 250 °C. For headspace, use a PTFE-faced silicone septum and crimp it; a screw cap will relax at 150 °C and leak.

Caps

Polypropylene screw caps are standard. They are chemically resistant and cheap, but they relax under heat and lose seal force. For methods that hold vials at 40 °C in the autosampler tray for hours (common in stability studies), a screw cap is fine. For anything above 80 °C, crimp. The crimp seal is a mechanical deformation, not a plastic memory, so it holds.

Operational precautions that prevent the mismatch from mattering

Even with the right vial, you can create a mismatch at the method level. These are the precautions I enforce in my own lab.

Verify tray part numbers at installation. Write the tray part number on a label inside the autosampler door. When someone swaps trays, the label tells the next person what vials fit.
Standardize on one neck finish per instrument. Do not run 9 mm and 11 mm on the same system unless you have separate trays and a written changeover procedure. Mixed formats are how needles bend.
Calibrate the crimper monthly. Measure crimp height on three seals with a gauge. A crimper that drifts 0.5 mm loose will leak on the third injection, not the first, so it passes a quick check and fails a long run.
Set needle depth with a dummy vial. Fill a vial with water, set the depth, and watch the needle tip through the glass. It should stop 1 to 2 mm above the bottom. Adjust and lock the setting.
Check cap height when you change suppliers. Measure ten caps with a caliper. If the mean differs by more than 0.5 mm from the validated part, re-validate the needle depth.
Store vials capped and upright. Dust in an open vial is a baseline artifact that looks like a ghost peak. I have chased a “contaminant” for two days that turned out to be cardboard fiber from a shipping box.
Use the right septa for the temperature. A silicone septum at 250 °C in a GC inlet bleeds siloxanes and shows up as a rising baseline. Match the septum rating to the inlet temperature, not the room temperature.

Troubleshooting checklist: symptom to dimensional cause

When a run fails, work through this list before you blame the column. Most of these are dimensional.

Needle bends or misses the septum: vial height or neck finish mismatch. Check tray part number against vial GPI finish. Check body OD against well diameter.
“No sample” or low injection volume: needle depth too shallow because the cap is thicker than validated. Measure capped height; reset depth.
Gripper fault or arm stall: vial too tall for the tray. Confirm 32 mm vs 38 mm. Check that the cap is fully seated.
Leak or evaporation over a long run: crimp too loose (measure crimp height, target 5.5 to 6.0 mm) or screw cap relaxed at temperature. Switch to crimp above 80 °C.
Ghost peaks or rising baseline in GC: septum bleed. Check septum temperature rating against inlet temperature. Replace with a higher-durometer septum.
Vial tips in the tray: body OD undersized or tray cut for a different finish. Caliper the vial; reject lots outside ±0.15 mm.
Coring (septum fragments in the sample): too many punctures on one septum, or a blunt needle. Replace septa every 5 to 10 injections on a single-port system; inspect the needle bevel.
pH drift in an unbuffered sample: Type III glass leaching. Switch to Type I borosilicate.

For method-specific guidance on sample handling and injection, the technical resources at Restek are a good cross-check, particularly their notes on septum selection and inlet maintenance. They publish bleed profiles for common septa that are more useful than any catalog claim.

What I recommend

Standardize on 12 x 32 mm, 9 mm short-thread screw vials in Type I borosilicate glass with PTFE/silicone laminated septa, 1.0 mm, and polypropylene screw caps. That combination fits the majority of current HPLC autosamplers (Agilent 1260/1290, Thermo Vanquish, Shimadzu Nexera, Waters ACQUITY), seals reliably at room temperature and up to about 60 °C in the tray, and is available in amber for photolabile work. Buy from a supplier who publishes a dimensional drawing with tolerances and who can quote the GPI finish on the carton.

For older instruments with 11 mm crimp trays, keep a separate stock of 11 mm crimp vials and a calibrated crimper. Do not try to convert a crimp tray to screw vials with an adapter unless the instrument vendor sells one; the adapter changes the cap height and the gripper clearance, and you are back to a dimensional mismatch.

For headspace, use 20 mm crimp vials, 10 or 20 mL, with PTFE-faced silicone septa and aluminum seals, and confirm the carousel diameter. For GC liquid injection, 11 mm crimp or 9 mm screw, 2 mL, matched to the tower rack.

If you are buying a new autosampler, ask the vendor for the tray part number and the validated vial and cap part numbers in writing before you sign. That single document prevents the mistake I described at the top, and it costs nothing to request. You can cross-check the vial side of that document against the format listings at autosampler vial, which covers the common formats and finishes in one place.

The takeaway is simple. Vial dimensions are not a detail to delegate to purchasing without a spec. They are four numbers (body OD, height, neck finish, cap height) plus two material choices (glass type, septum). Match those to the tray part number on your instrument, measure a sample from each lot, and set the needle depth with a dummy vial. Do that and the vial stops being a variable in your method. Skip it and you will eventually buy 20,000 vials that do not fit, and you will learn the tray part number the hard way, at 2 a.m., with a bent needle in your hand.

References & further reading

Related resources from our lab consumables network:

External references and standards: