What an autosampler vial actually has to do-- HPLC Autosampler Vials

I once watched a lab lose three weeks of stability data because a buyer saved $18 per case on vials. The septa were a generic silicone formulation, the autosampler needle punched through the same spot 40 times over a 48-hour run, and by the end of the sequence the injection volumes had drifted enough to fail system suitability. The vials were the cheapest line item on the purchase order and the most expensive decision on the project. That is the whole problem with autosampler vial selection: the cost of the vial is trivial, and the cost of getting it wrong is not.

This guide is what I wish someone had handed me when I started buying these things. It covers the glass, the closure, the septum chemistry, the dimensional tolerances that actually matter to your autosampler, and the failure modes I have personally debugged at 2 a.m. with a needle that would not seat. If you are a procurement manager comparing quotes, or an operator trying to figure out why your RSD jumped from 0.4% to 2.1%, this is for you.

Strip it back. An autosampler vial is a small, chemically inert container with a pierceable closure that must survive a robotic needle, hold a defined liquid volume, and not contaminate the sample. That is it. Everything else is engineering around those four jobs.

The tolerances are tighter than most people realize. A typical 2 mL HPLC vial for a modern autosampler has a body diameter of 12.0 mm ± 0.2 mm, a height of 32.0 mm ± 0.5 mm, and a neck finish that must mate with a crimp or screw cap to a defined torque. If the diameter runs to 12.4 mm, the vial may jam in the carousel or fail to seat in the gripper. If the height runs long, the needle may not reach the sample. If the neck thread pitch is off by a fraction, the cap will cross-thread and you will get evaporation or, worse, a needle strike on the septum edge.

The reference standard for the neck finish is the GPI 13-425 thread for 13 mm screw caps and the 8-425 for 8 mm caps on smaller-format vials. Crimp vials use a 11 mm aluminum seal with a defined crimp diameter (typically 11.0 mm for the seal, crimped to about 10.5 mm at the skirt). These are not arbitrary numbers. They are the interface between your vial and a $60,000 instrument, and they need to be right every time.

Glass: the first decision, and the one people skip

Almost every autosampler vial you will buy is borosilicate glass, usually Type I, Class A borosilicate per USP <660> and ASTM E438. That is the baseline. Type I glass has low alkali leaching, good hydrolytic resistance, and will not shed ions into your mobile phase. If a supplier cannot tell you the glass type, walk away.

Clear vs. amber

Clear glass is fine for most work. Amber glass is USP Type I with an ambering agent (usually iron oxide) and is specified when your analytes are photolabile: vitamins, certain antibiotics, catecholamines, anything with a conjugated double bond that will isomerize under lab lighting. I default to amber for any method that runs longer than four hours on the autosampler, because the sample sits in the tray under fluorescent light the entire time. The cost difference is a few cents per vial.

Glass vs. plastic

Polypropylene vials are cheaper and unbreakable, but they adsorb. Hydrophobic analytes, especially at low concentrations (sub-ppb), will stick to the polypropylene wall and you will see recovery drop. I have seen 15 to 30% loss of a lipophilic pesticide on PP versus glass in a 24-hour hold study. Use plastic only for aqueous, high-concentration work or where breakage is a real hazard (field sampling, high-throughput screening). For anything quantitation-critical, glass.

Deactivation and silanization

Bare borosilicate glass has surface silanol groups that will hydrogen-bond with basic analytes (amines, quaternary ammonium compounds) and cause peak tailing or outright adsorption. Deactivated glass (sometimes sold as “silanized” or “inert”) has the surface treated to cap those silanols. If you run basic drugs by reversed-phase HPLC and your tailing factor is 1.8 on a fresh column, try deactivated vials before you blame the column. It is a common and underdiagnosed cause of poor peak shape.

Closure systems: crimp, screw, and snap

This is where most buying decisions go wrong, because the closure is where the operator and the instrument actually interact.

Crimp caps

Crimp closures use an 11 mm aluminum seal with a septum, crimped onto the vial neck with a hand or pneumatic crimper. The advantages are a tight, consistent seal and a low profile that suits older autosamplers. The disadvantages are real: you need a crimper (and a decapper), the crimp force must be consistent (too loose and you leak, too tight and you crush the septum or crack the neck), and you cannot reseal a crimp vial once it is opened. Crimp consistency is a genuine source of RSD drift. If you go crimp, buy a pneumatic crimper and set it to the manufacturer’s spec (usually around 1.5 to 2.0 bar for a standard 11 mm seal), and check the crimp with a go/no-go gauge weekly.

Screw caps

Screw-thread closures (the 9 mm short-thread is the most common format on modern autosamplers) are what I recommend for most labs. They are resealable, need no tools, and the torque is controllable by hand or with a torque-limited driver. The spec that matters is torque: over-tightening deforms the septum and creates a coring risk when the needle punches; under-tightening allows evaporation. Aim for roughly 0.4 to 0.6 N·m on a 9 mm screw cap. A torque screwdriver costs about $80 and will save you a failed run. The 9 mm short-thread format is also the native format for most Agilent and Thermo autosamplers, so it drops in without adapters.

Snap caps

Snap or press-on caps are fast and tool-free, popular for high-throughput screening. The seal is less consistent than a torqued screw cap, and they are easier to dislodge. Fine for qualitative work, not my first choice for a validated method.

Septum chemistry: the part that touches your sample

The septum is the only component that contacts both the sample headspace and the needle, and it is where most contamination and coring problems originate. There are four common materials and you should know which one you are buying.

Silicone (pure): Excellent resealing, good chemical inertness, but high gas permeability and it can bleed siloxanes into GC. Cheap. Fine for short HPLC runs.
Silicone/PTFE laminated: PTFE faces the sample, silicone provides the reseal. The PTFE barrier stops the silicone from leaching into the sample and reduces gas permeability. This is the workhorse septum for HPLC. Specify it unless you have a reason not to.
PTFE/silicone/PTFE sandwich: PTFE on both faces, silicone core. Better chemical resistance, slightly harder to pierce. Good for aggressive solvents and long sequences.
Butyl rubber: Low gas permeability, used mostly for GC headspace and where oxygen ingress matters. Not common on HPLC.

Puncture count is the number you should ask your supplier for and most buyers never do. A good silicone/PTFE septum will take 50 to 100 punctures before coring becomes a problem; cheap septa core at 20 to 30. Coring is when the needle punches a plug of septum into the vial, which then sits in your sample and either blocks the needle or shows up as a phantom peak. If your method reinjects the same vial more than a few times (sequence reruns, calibration curves, long unattended runs), buy septa rated for high puncture counts and replace them on a schedule, not on failure. I change septa every 50 injections on critical methods regardless of what the supplier claims.

Pre-slit septa are available and reduce coring and needle friction, at the cost of slightly higher evaporation. For short runs they are a reasonable trade. For a 24-hour stability run, use a solid septum and a well-torqued cap.

Volume, format, and the insert question

The standard 2 mL (12 x 32 mm) vial holds about 1.5 mL comfortably with headspace. If your autosampler has a standard tray, this is the default and you should not overthink it. The complications come from small-volume work.

When you need an insert

If your sample volume is under about 300 µL, a standard 2 mL vial will not let the needle reach the liquid, and you will get air injections or partial-volume errors. The fix is a glass insert (often 150 µL or 250 µL) that sits inside the 2 mL vial and pools the sample at the bottom where the needle can reach it. Inserts come with or without a polymer foot (the foot centers the insert and cushions it). Use inserts when sample is precious (a single 200 µL prep from a microcentrifuge tube) or when you are running a method that specifies a small injection volume from a small total volume.

Two cautions on inserts. First, the insert must be matched to the vial and the autosampler needle depth; a mismatched insert can be missed by the needle entirely. Second, inserts add a handling step and a contamination risk. Do not use them unless you have to.

Small-format and high-throughput vials

For high-throughput work, short 9 mm vials and micro-vials (often 300 µL to 1 mL) let you fit more samples per tray. The trade-off is that not every autosampler accepts them without a tray adapter, and the smaller the vial, the more sensitive the needle-depth calibration. Check your instrument’s tray spec before you standardize on a small format. The Agilent consumables catalog and the Thermo Fisher sample handling pages both publish tray and vial compatibility tables that are worth bookmarking.

A selection checklist with real numbers

Here is the decision table I actually use when specifying a vial for a new method. Print it, put it in the SOP, and make the buyer sign off against it.

Glass type: USP Type I borosilicate, Class A. Amber if any analyte is photolabile or the run exceeds 4 hours.
Format: 12 x 32 mm, 2 mL, 9 mm short-thread screw top (default). Crimp 11 mm only if the autosampler requires it.
Neck finish: GPI 13-425 or 8-425 as applicable; verify against the instrument’s tray spec.
Septum: silicone/PTFE laminated, rated for ≥ 50 punctures, pre-slit only for short runs.
Cap torque: 0.4 to 0.6 N·m for 9 mm screw caps; pneumatic crimper at 1.5 to 2.0 bar for 11 mm crimp.
Volume: fill to no more than 75% of nominal (about 1.5 mL in a 2 mL vial) to leave headspace.
Insert: required below 300 µL sample volume; match insert to needle depth.
Certification: request a certificate of analysis with extractables data and a lot number you can trace.
Temperature: autosampler trays typically run 4 to 40 °C; confirm your septum’s rated range (most silicone/PTFE are good to 200 °C, but the cap and seal are the weak point at the low end).

Operational precautions that prevent most failures

Most vial problems are operator problems, and most operator problems are preventable with a few habits.

Do not overfill. A 2 mL vial filled to 1.9 mL has almost no headspace, and if the sample is volatile you will get pressure buildup, septum bulging, and injection volume error. Fill to about 1.5 mL.

Do not over-torque. Hand-tight is not a spec. Use a torque driver. Over-torqued caps deform the septum and cause coring and needle bending.

Check needle depth. After any change of vial format or insert, run a needle-depth verification (a dummy injection into a vial of water and a visual check) before you commit a sequence.

Mind the light and the temperature. A tray sitting under a window or next to a heat source will degrade photolabile and thermolabile samples. Use amber glass and, where the method allows, a cooled tray.

Track lot numbers. If you get a bad batch of septa, you want to know which lot and which runs it touched. A simple log of lot number, date opened, and runs covered has saved me more than once.

Troubleshooting checklist

When an autosampler sequence goes wrong, work through this list before you touch the instrument.

Rising RSD or drifting response: suspect septum coring or evaporation. Inspect the septum for a visible hole or plug; check cap torque.
Phantom peaks: septum bleed (siloxanes) or a cored plug in the sample. Switch to PTFE-faced septum and replace on schedule.
Needle will not pierce: over-torqued cap, wrong septum thickness, or a vial seated at an angle. Check the cap and reseat the vial.
Partial or air injections: sample volume too low for the needle depth; add an insert or increase volume.
Vial jammed in tray: body diameter out of tolerance or a damaged vial. Measure a sample of the lot with calipers against the 12.0 mm ± 0.2 mm spec.
Leak or evaporation: under-torqued cap, cross-threaded cap, or a damaged seal. Re-cap with a torque driver and check the thread.
Adsorption or low recovery: bare glass silanols or polypropylene walls. Switch to deactivated glass.

For method-specific guidance, the U.S. EPA SW-846 test methods are a good reference for sample handling and preservation requirements, and Restek’s technical resources have practical notes on septa and inlet consumables that apply directly to vial selection.

Buying: what to ask, what to pay, and where to source

When you go to market, ask every supplier the same five questions: glass type and class, septum material and puncture rating, neck finish standard, certificate of analysis availability, and lot traceability. A supplier who cannot answer all five is a supplier who is repackaging someone else’s product without knowing what is in it.

On price, expect roughly $0.10 to $0.25 per vial for standard clear 2 mL screw-top with silicone/PTFE septum at case quantities, and $0.25 to $0.60 for amber, deactivated, or certified vials. Inserts add $0.15 to $0.40 each. If a quote is dramatically below that range, ask what changed in the septum or the glass. It is almost always one of those two.

For a broad catalog of formats and closures, the HPLC autosampler vial range is a sensible starting point, and if you are comparing per-unit pricing across suppliers, Sample Vial is useful for benchmarking. For the wider consumables picture (columns, filters, syringes), Chromatography Products covers the categories you will end up buying alongside the vials. And if you want to talk to a manufacturer directly about custom formats or lot certification, Aijiren is one of the suppliers I have worked with on non-standard neck finishes.

My recommendation

For a typical reversed-phase HPLC method on a modern autosampler, specify a 2 mL clear or amber USP Type I borosilicate vial, 12 x 32 mm, 9 mm short-thread screw top, with a silicone/PTFE laminated septum rated for at least 50 punctures, torqued to 0.4 to 0.6 N·m. Buy a torque driver. Buy a lot-traceable product with a certificate of analysis. Replace septa every 50 injections on critical methods. Use deactivated glass if you run basic analytes, amber if you run photolabile ones, and inserts only when the sample volume is under 300 µL.

That specification will cost you maybe $0.20 a vial and will prevent the failure modes that cost you a week of data. The vial is the cheapest thing in the workflow and the one most likely to ruin it. Buy it like it matters.

References & further reading

Related resources from our lab consumables network:

External references and standards: