Sizing an HPLC vial to real sample volume and injection count--Aijiren Vials for HPLC

I spent the better part of a decade watching analysts pick a vial by grabbing whatever case was closest to the balance, and then I spent another decade fielding the phone calls when those vials ran dry at 3 a.m. mid-sequence. Sizing a vial to your sample volume and injection count is not a glamorous calculation, but it is one of the few places in an analytical method where a two-minute arithmetic check saves you a re-run, a lost plate, and a conversation with a client about why their 96 samples turned into 71 results. This guide is the calculation I actually use, with the numbers that matter.

The core question is simple: given how much liquid you have, how much the autosampler needs to pull per injection, and how many injections you must deliver, what vial and insert format fits without starving the needle or wasting sample? Everything else (glass grade, closure, septum) is downstream of that arithmetic. Get the arithmetic wrong and no amount of premium glass will save you.

The three numbers you need before you open a catalog

You cannot size a vial without three figures, and most labs have never written them down in one place. Write them down now.

1. Usable sample volume

This is not the volume you pipetted. It is the volume you pipetted minus what the transfer lost, minus what the autosampler cannot reach. If you reconstituted a 10 mg extract into 2.0 mL of diluent in a 2 mL vial, your nominal volume is 2.0 mL. But a standard 12 x 32 mm vial with a 9 mm screw thread holds a maximum fill of about 1.5 mL in practice. Fill it to 1.8 mL and the autosampler needle, which typically descends to a fixed depth measured from the vial base, will hit liquid reliably only if your method’s needle depth is set correctly. Fill it to the shoulder and you risk the needle drawing air because the liquid surface drops below the needle tip as the vial empties.

Rule of thumb I use: never fill a 12 x 32 mm vial past 80% of its nominal internal capacity. For a 2 mL vial that is 1.6 mL. For a 1.5 mL vial, 1.2 mL. This leaves headroom for the needle, for thermal expansion in a cooled autosampler tray, and for the vortex you will inevitably run.

2. Injection volume per run

This is set by your method, not by the vial. A typical reversed-phase HPLC injection is 1 to 20 microliters; a GC injection is often 0.5 to 2 microliters. Note the units. The autosampler pulls microliters, not milliliters. This is where the arithmetic usually surprises people.

3. Injection count, including standards and QC

Count every injection the vial must survive: the sample itself, any duplicate, the bracketing standards if you use the same vial, and the QC injections. A 96-well plate run with a single standard, a blank, and a duplicate per sample is not 96 injections. It is closer to 200. If you are sizing individual vials for a small batch, count them honestly.

The arithmetic that decides your format

Here is the calculation, worked through with a real example. Suppose you have 1.2 mL of reconstituted sample, you inject 10 microliters per run, and you need 6 injections from that vial (sample, duplicate, and four QC bracketing injections).

Total volume consumed by injections: 6 x 10 microliters = 60 microliters
Autosampler dead volume (needle depth set to 2 mm above the base): for a 12 x 32 mm vial, roughly 100 to 200 microliters remains unreachable at the bottom
Evaporation loss over a 24-hour run in a tray at 4 degrees C: negligible; at 25 degrees C with a loose cap, up to 50 microliters per day for volatile diluents like methanol
Total needed: 60 + 200 + 50 = 310 microliters

You have 1,200 microliters. You are comfortable in a standard 12 x 32 mm vial. No insert required. This is the case for the majority of routine work, and it is why the 2 mL screw vial remains the workhorse format.

Now change one number. Suppose you have 150 microliters of sample (a micro-extraction, a precious clinical specimen, a stability timepoint pulled from a single tablet) and you need 6 injections at 10 microliters. Total consumed is 60 microliters. Dead volume in a standard vial is 100 to 200 microliters. You cannot do it in a standard vial. You need a micro insert, typically a 150 to 250 microliter glass insert with a tapered or conical bottom that fits inside a 12 x 32 mm vial. The insert’s dead volume is on the order of 5 to 15 microliters, so 150 microliters gives you 60 for injections plus 15 dead plus margin. It fits, barely. If you needed 20 injections at 10 microliters (200 microliters consumed), you would be short. That is when you step up to a 300 microliter insert or a dedicated short-form vial.

A decision checklist with real numbers

Use this table. It is the one I have taped inside the consumables cabinet.

Sample volume 1.0 to 1.5 mL, injections up to 50 at 10 microliters: standard 12 x 32 mm, 2 mL, 9 mm screw. Dead volume 100 to 200 microliters. No insert.
Sample volume 300 to 800 microliters, injections up to 30: 12 x 32 mm vial with a 300 microliter insert. Dead volume 10 to 20 microliters.
Sample volume 100 to 250 microliters, injections up to 15: 12 x 32 mm vial with a 150 microliter insert. Dead volume 5 to 15 microliters.
Sample volume under 100 microliters, injections up to 5: dedicated micro-vial, 6 x 32 mm or 8 x 30 mm, with a 100 microliter capacity. Requires an autosampler tray adapter.
Sample volume under 50 microliters: consider a sealed micro-insert or a direct-injection capillary system. Standard autosamplers will struggle.

Notice the pattern. The insert is not a luxury. It is the format that lets you run a precious sample without diluting it into a volume you do not have. The 9 mm screw thread geometry is standardized across most autosamplers, which is why the 12 x 32 mm format dominates. If you are sourcing beyond a single vendor, the Hplc Vial catalog is where I point buyers who need the full format matrix in one place.

Why fill level and needle depth are a coupled problem

Most injection failures I have investigated were not septum coring or a bad seal. They were a needle depth set to the wrong value for the vial actually loaded. Autosamplers like the Agilent 1260 series and the Thermo Vanquish let you set needle depth in millimeters from the vial base. A standard 12 x 32 mm vial has an internal depth of about 28 mm. If your method’s needle depth is set to 2 mm and you swap in a 150 microliter insert, the insert’s internal bottom sits higher, and the needle may not reach the liquid. Conversely, if you set depth to 10 mm for an insert and then load a full-size vial, the needle sits above the liquid surface once the vial is half empty.

The fix is procedural: when you change format, change the needle depth in the method and verify with a dye test. Fill a test vial with water plus a drop of methylene blue, run one injection, and look at the needle. If it is wet above the expected depth, your setting is wrong. This is a five-minute check that catches the majority of “the autosampler missed the vial” calls.

Temperature, evaporation, and the long-run problem

Sample volume and injection count are not static. They change over the run. A 96-vial tray at 4 degrees C loses very little to evaporation, but a tray at ambient with a volatile diluent (methanol, acetonitrile, hexane) loses volume from every vial over a 12-hour sequence. I have seen a 1.0 mL fill in a 2 mL vial drop by 40 to 60 microliters over 18 hours with a loose screw cap and a methanol diluent. That sounds small until your last injections are drawing from a vial whose liquid level has fallen below a needle depth set for the first injection.

Two mitigations. First, use a pre-slit silicone/PTFE septum with a properly torqued screw cap (I torque 9 mm screw caps to about 0.5 to 0.7 N·m; over-torquing deforms the septum and creates a leak path, under-torquing lets vapor escape). Second, set your needle depth conservatively (deeper than you think you need) and validate that the needle does not contact the vial bottom. A needle that touches glass will blunt and core the septum on the next stroke.

If you are running long sequences with volatile diluents, the honest answer is a cooled tray and a capped vial. If you cannot cool the tray, size up one format so the evaporation loss is a smaller fraction of the starting volume. A 1.5 mL fill in a 2 mL vial tolerates more loss than a 300 microliter fill in an insert.

Matching vial format to autosampler geometry

The autosampler tray is the constraint. Standard 12 x 32 mm vials fit the common 9 mm trays from Agilent, Waters, Shimadzu, and Thermo. Micro-vials (6 x 32 mm, 8 x 30 mm) require tray adapters, and those adapters are format-specific. Before you commit to a micro-vial for a precious sample, confirm the adapter exists for your tray. I have seen labs buy a case of 6 x 32 mm vials only to discover the adapter was discontinued.

For reference on tray and format compatibility, the Agilent chromatography consumables and columns pages and the Thermo Fisher chromatography and sample handling documentation both list the vial formats each autosampler accepts. Check the tray, not the vial catalog, when you are unsure.

The insert decision: when a micro insert earns its place

An insert is worth the cost when the sample is limited and the injection count is high enough that dilution is not an option. The arithmetic earlier showed the break-even. But there are secondary reasons to use an insert even when you have volume to spare.

Reduced dead volume means less sample wasted per vial. If you are running a method development study with 20 conditions and only 500 microliters of each, the insert lets you run all 20 without re-extracting.
Inserts protect the autosampler from particulates. A conical insert settles debris below the needle path. I have used 150 microliter inserts specifically to keep a dirty sample from clogging a needle.
Inserts let you run small-volume standards without a dilution error. A 100 microliter standard prepared directly in an insert has one transfer, not two.

The cost is handling time and the risk of insert misalignment. A 150 microliter insert that is not seated fully in the vial will sit high, and the needle will miss. Seat it with a firm push and verify by eye. If you are running a plate of 96 inserts, that is 96 chances to seat one wrong. This is why many labs reserve inserts for the precious samples and use plain vials for the rest.

Glass grade, closure, and the numbers that go with them

Once the format is settled, the material choices follow. For HPLC, Type I borosilicate glass (USP Type I) is the default. It has low extractables and low alkali leaching. Type II (soda-lime treated) is acceptable for some aqueous work but I would not use it for trace analysis or for anything going into an LC-MS. The U.S. EPA SW-846 test methods and USP <661> both give you the material acceptance criteria; cite them in your SOP rather than relying on a supplier’s marketing sheet.

Closure choice matters for the injection count problem too. A silicone/PTFE septum tolerates more punctures than a plain silicone septum, and a pre-slit septum reduces coring. Realistic puncture counts: a standard silicone/PTFE septum handles 50 to 100 punctures before coring risk rises; a pre-slit septum is good for 100 to 200. If your method calls for 6 injections from a vial, a standard septum is fine. If you are running 96 injections from a single vial (some autosamplers do this for long equilibration sequences), use a pre-slit septum and inspect it.

For a broader view of closure and septum options, the Chromatography Consumables catalog groups them by format and material, which saves a lot of back-and-forth with procurement.

Troubleshooting checklist for volume and injection-count problems

When a run fails and the symptom is “the autosampler missed the vial” or “the result is low,” work through this list in order. It is ordered by how often each cause is the actual culprit in my experience.

Needle depth mismatch. Confirm the method’s needle depth matches the vial or insert actually loaded. Dye test.
Fill level below needle reach. Check the starting volume against the 80% rule. Check evaporation over the run length.
Insert not seated. Pull the vial and look. A high insert is a missed injection.
Septum cored. Inspect the septum for a plug of silicone. Replace and re-run.
Cap over-torqued. A deformed septum creates a leak path and a coring site. Re-torque to 0.5 to 0.7 N·m.
Wrong vial format for the tray. A micro-vial in a standard tray without an adapter will sit loose and the needle will miss.
Sample volume too small for the injection count. Re-run the arithmetic. If total consumption plus dead volume exceeds the fill, you need a smaller format or a larger fill.

Restek’s technical resources are useful here for the septum and closure side of the checklist, particularly their notes on coring and puncture life.

What I recommend, with the parameters that matter

For routine HPLC work with adequate sample volume, use a 12 x 32 mm, 2 mL, 9 mm screw-thread vial with a silicone/PTFE pre-slit septum, filled to no more than 1.5 mL, torqued to 0.5 to 0.7 N·m, with needle depth set to 2 mm above the base and verified by dye test. This handles up to about 50 injections at 10 microliters with margin for dead volume and evaporation.

For limited sample volume, use a 150 or 300 microliter conical insert seated fully in the same 12 x 32 mm vial, with needle depth adjusted to the insert bottom and verified. Use a pre-slit septum if the injection count exceeds 50. Do not over-torque.

For very small volumes (under 100 microliters), step to a 6 x 32 mm or 8 x 30 mm micro-vial with a tray adapter confirmed against your autosampler’s documentation before purchase. Check the Lab Vials format pages for the adapter compatibility matrix, and if you are sourcing a full consumables program, the Aijiren Brand range covers the 9 mm screw format and inserts with the dimensions above.

Takeaway

Sizing an HPLC vial to your sample volume and injection count is a three-number calculation (usable volume, injection volume, injection count) followed by one comparison against the vial’s dead volume and fill limit. Do the arithmetic before you open the catalog, set needle depth to match the format you actually loaded, and verify with a dye test whenever you change format. The vial is cheap; the re-run is not.

References & further reading

Related resources from our lab consumables network:

External references and standards: