Failure mode 1: septum coring and needle damage--Aijiren Vials for HPLC

I have watched a vial ruin a perfectly good sequence more times than I care to admit. The first time it happened, I was a junior analyst running a 96-well autosampler overnight and came back to a chromatogram that looked like a seismograph. The column was fine. The mobile phase was fresh. The problem was a batch of 2 mL vials with septum liners that had been sitting on the shelf for two years, and the silicone had hardened enough that the needle punched a core straight into the sample. That lesson cost me a re-run and a day of instrument time. Since then I have built a mental checklist of vial failure modes that I run through before any sequence, and I want to share it because most of these problems are preventable with a few concrete choices.

The vial is the cheapest component in the HPLC flow path and the one most likely to introduce variability. A 2 mL glass vial with a crimp cap costs somewhere between eight and forty cents depending on the closure and the glass grade. A column costs several hundred to several thousand dollars. Yet the vial touches the sample first, and its dimensional tolerance, chemical inertness, and closure integrity determine whether the data you collect is real. This article covers the failures I see most often in daily operation, what causes them, and the specific numbers and practices that prevent them.

Septum coring is the most common vial failure in autosampler workflows. The needle punches through the silicone or silicone/PTFE septum, and a small disc of material shears off and falls into the sample. That disc can block the needle, plug the injection port, or dissolve and contaminate the chromatogram with ghost peaks.

Root causes

Septum coring has three main drivers. The first is needle geometry. A standard 22-gauge blunt-tip autosampler needle has an outer diameter of about 0.72 mm, while a 26-gauge needle is about 0.47 mm. A larger needle punches a larger hole and stresses the septum more. The second driver is septum hardness. Standard silicone septa run around 40 to 50 Shore A, and pre-slit septa are softer and easier to penetrate but can leak under pressure. The third driver is repeated puncture. A septum rated for 100 punctures will core long before that if the needle is misaligned or the vial is not seated squarely in the tray.

I have also seen coring from vial height variation. If the vial is 0.5 mm taller than nominal, the needle does not fully retract between injections and drags through the septum on the way up, which tears rather than punches. This is why dimensional tolerance matters. A good 12 x 32 mm vial holds its height to within ±0.2 mm and its neck diameter to within ±0.1 mm.

Prevention

Use pre-slit septa only when the autosampler pressure rating allows it. Pre-slit silicone/PTFE caps reduce coring but are typically rated to about 1.5 bar, so check your system pressure.
Match needle gauge to septum. For 26-gauge needles, a standard silicone septum is fine. For 22-gauge or larger, use a PTFE-faced septum with a bonded liner.
Replace septa after 50 to 100 injections in high-throughput sequences, not after the rated maximum. The rating assumes perfect alignment.
Verify vial seating. A vial that sits 0.3 mm high in the tray will core even a fresh septum.

If you are sourcing vials across multiple product lines, check that the closure and septum are matched to the autosampler. The HPLC vial range at vialforhplc.com lists septum hardness and puncture ratings per SKU, which saves a lot of guesswork.

Failure mode 2: evaporation and sample concentration drift

Evaporation is silent. You inject the same sample at 9 a.m. and at 4 p.m. and the peak area has drifted 6 percent. The autosampler tray is at ambient, the septa are not sealing, and the solvent is slowly leaving the vial.

Root causes

The seal depends on the cap torque and the septum compression. A crimp cap that is under-crimped leaves a gap between the septum and the glass neck. An over-crimped cap deforms the septum and creates a channel. The target crimp is usually a cap height of 5.5 to 6.0 mm after crimping, measured with a crimp gauge. Hand crimpers rarely hit that window consistently; a pneumatic crimper set to 4 to 6 bar and checked against a gauge every 100 caps is more reliable.

Screw caps have a different failure mode. The thread is a 9 mm or 10-425 format, and the cap must be torqued to about 0.5 to 0.7 N·m. Under-torque and the PTFE-faced liner does not compress. Over-torque and the cap cracks or the liner extrudes. I have seen labs torque screw caps by feel and get a 15 percent evaporation loss over eight hours on a volatile solvent like dichloromethane.

Prevention

Use a calibrated crimper or torque wrench. Check cap height with a gauge on the first five caps of every batch.
For volatile solvents, use a cap with a bonded PTFE/silicone liner rather than a loose liner. Bonded liners do not shift during crimping.
Keep the autosampler tray at 4 to 10 °C if the method allows. Cold trays cut evaporation by roughly half compared with ambient.
Cap vials immediately after filling. A vial left open for two minutes loses measurable volume with a 50/50 water/methanol mix.

Failure mode 3: extractables and ghost peaks

Ghost peaks are the failure mode that sends chemists on wild goose chases. You see a peak at 8.2 minutes that is not in the standard, and it moves or disappears when you change vial lots. That is an extractable from the vial or the septum.

Root causes

Borosilicate glass is not inert to everything. Type I borosilicate (the standard for HPLC vials) has low alkali leaching, but at pH above 8 or below 3, glass surfaces can release sodium and borate ions that show up as baseline drift or as peaks in ion chromatography. The bigger culprit is the septum. Standard silicone contains plasticizers and low-molecular-weight siloxanes that leach into the sample, especially with nonpolar solvents like hexane or toluene.

I have measured siloxane extractables at 2 to 8 ppm from an unbonded silicone septum after 24 hours of contact with acetonitrile. That is enough to produce a shoulder on an early-eluting peak. The fix is a PTFE-faced septum, where the sample only contacts the PTFE film, which is essentially inert.

Prevention

For LC-MS or trace analysis, use vials with a PTFE/silicone bonded septum and a glass that meets USP <661> Type I requirements.
Run a vial blank: fill a vial with mobile phase, cap it, and inject after 24 hours. Any peak above the baseline is an extractable.
For pH extremes, use deactivated glass or polypropylene vials. Polypropylene has its own extractables, so run the blank test on the actual lot.
Store vials capped and in the original packaging. An open bag of vials picks up dust and airborne contaminants.

The lab vials catalog at labvials.com groups products by glass type and septum chemistry, which makes it easier to match the vial to the detection limit of your method.

Failure mode 4: autosampler jams and vial crashes

A vial crash stops the sequence and sometimes bends the needle. The autosampler reports a vial height error, and you find a vial that is 1 mm too tall or a cap that is not fully seated.

Root causes

Autosamplers like the Agilent 1260 series or the Thermo Scientific Vanquish have a vial height sensor that expects a 12 x 32 mm vial with a cap height of 5.5 to 6.0 mm. A cap that is crimped to 6.5 mm will trip the sensor. So will a vial from a lot with a height tolerance of ±0.5 mm instead of ±0.2 mm. I have also seen crashes from mixed vial types in the same tray: a 2 mL vial next to a 1.5 mL vial with a different neck height confuses the tray mapping.

Prevention

Use one vial format per tray. Do not mix crimp and screw vials in the same run.
Check cap height with a gauge after crimping. A 0.5 mm oversize cap is enough to crash.
Verify the vial lot against the autosampler specification. The Agilent chromatography consumables and columns documentation lists the dimensional tolerances for each autosampler model.
Run a tray map check before a long sequence. A five-minute check saves a four-hour re-run.

Failure mode 5: adsorption of analytes on glass

Some analytes stick to glass. Proteins, peptides, and certain basic drugs adsorb to the silanol groups on the glass surface, and the result is low recovery and poor reproducibility at low concentrations.

Root causes

Borosilicate glass has surface silanol groups that carry a negative charge above pH 3. Positively charged analytes bind electrostatically. The effect is worst at low concentrations, below 10 ng/mL, where the surface area of the vial is large relative to the amount of analyte. I have seen recovery drop to 60 percent for a basic drug in a plain glass vial versus 95 percent in a deactivated vial.

Prevention

Use deactivated glass vials for basic analytes and proteins. The deactivation covers the silanols with a hydrophobic or inert layer.
Add a small amount of organic modifier or a surfactant to the sample if the method allows. A 5 percent acetonitrile spike reduces adsorption significantly.
For proteins, use low-binding polypropylene vials. Glass is the wrong choice for peptide work at low concentration.
Run a recovery test: compare a fresh standard in a vial against a direct injection. A 10 percent loss is a red flag.

Failure mode 6: chemical incompatibility and vial deformation

Not every sample belongs in borosilicate glass. Strong bases attack glass, and some solvents soften the cap liner. I have seen a vial cap swell and pop off after contact with tetrahydrofuran because the liner was not rated for it.

Root causes

Glass is compatible with most solvents but not with pH above 12 or with hydrofluoric acid. Polypropylene vials handle strong bases but are attacked by some chlorinated solvents at elevated temperature. The cap liner is the weak point: silicone swells in toluene and hexane, and PTFE is compatible with almost everything but does not seal as well against aqueous samples.

Prevention

Check the chemical compatibility chart for the vial and the liner before the first run. The Thermo Fisher chromatography and sample handling resources include compatibility tables for common solvents.
For pH above 12, use polypropylene vials with a PTFE-lined cap.
For halogenated solvents, use a PTFE-faced septum and a cap rated for the solvent.
Do not store samples in vials long-term. Transfer to a compatible container if the hold time exceeds 24 hours.

Failure mode 7: particulates and needle blockage

A blocked needle stops the injection and the autosampler reports a pressure error. The cause is usually a particle from the sample, the septum, or the vial itself.

Root causes

Glass vials can shed particles from the neck or the rim if they are mishandled. Septum coring adds silicone particles. Samples that were not filtered add everything else. A 0.22 µm filter removes most of it, but I have seen labs skip filtration for “clean” samples and pay for it with a blocked needle.

Prevention

Filter samples through a 0.22 µm membrane before filling. For viscous samples, use a 0.45 µm filter.
Inspect vials before use. A vial with a chipped neck or a visible particle goes in the reject bin.
Use a needle with a larger internal diameter for particulate-prone samples, or add an inline filter to the autosampler.
Flush the needle with mobile phase between injections if the method allows.

Decision checklist: choosing the right vial for the job

Here is the checklist I use when specifying a vial for a new method. The numbers are the ones that matter in practice.

Format: 12 x 32 mm for standard autosamplers; 9 mm screw thread or 11 mm crimp. Confirm the autosampler model before ordering.
Glass type: Type I borosilicate for pH 3 to 8; deactivated glass for basic analytes; polypropylene for pH above 12.
Septum: PTFE/silicone bonded for LC-MS and trace work; silicone for routine work with 26-gauge needles; pre-slit only if system pressure is below 1.5 bar.
Puncture rating: 100 punctures for standard septa; replace at 50 to 100 in high-throughput sequences.
Crimp height: 5.5 to 6.0 mm, checked with a gauge every 100 caps.
Screw torque: 0.5 to 0.7 N·m for 9 mm screw caps.
Fill volume: 0.5 to 1.5 mL in a 2 mL vial; leave headspace for the needle and for expansion.
Temperature: ambient or 4 to 10 °C for volatile solvents; check the cap liner rating for the tray temperature.

For a side-by-side comparison of septum chemistries and glass grades, the chromatography consumables reference at chromatographyconsumables.com has a compatibility matrix that covers most common solvents and pH ranges.

Troubleshooting checklist for a failing sequence

When a sequence fails, I work through this list in order. It catches about 90 percent of vial-related problems in under an hour.

Check the chromatogram for ghost peaks. If a peak appears in the blank, it is an extractable. Swap to a PTFE-faced septum and re-run the blank.
Check peak area drift across the run. A downward drift suggests evaporation. Check cap height and tray temperature.
Check for a pressure spike or needle error. A blocked needle points to particulates or septum coring. Inspect the septum for a hole larger than the needle.
Check recovery against a direct injection. A 10 percent or greater loss suggests adsorption. Switch to a deactivated vial.
Check vial height and cap height. A crash or a height error points to dimensional tolerance. Measure five vials from the lot.
Check the lot number against the certificate. A new lot with a different glass supplier can change the extractable profile.

The U.S. EPA SW-846 test methods include sample handling and holding time requirements that assume a properly sealed vial. If your method follows SW-846, the vial is part of the method, not an accessory.

What I recommend

For routine HPLC work, I specify a 12 x 32 mm Type I borosilicate vial with a PTFE/silicone bonded septum and a 9 mm screw cap torqued to 0.5 to 0.7 N·m. That combination handles pH 3 to 8, most organic solvents, and the trace analysis that LC-MS demands. For basic analytes below 10 ng/mL, I switch to deactivated glass. For pH above 12, polypropylene. For volatile solvents, I keep the tray at 4 to 10 °C and check cap height every 100 caps.

The Aijiren brand line at aijirenvials.com is one I have used for high-throughput sequences because the dimensional tolerance is held to ±0.2 mm and the septum lots come with a puncture rating and an extractable certificate. That paperwork matters when a method is audited. The Restek technical resources also have good guidance on septum selection for trace analysis if you want a second opinion on liner chemistry.

Takeaway

Vial failures are not random. They come from septum coring, evaporation, extractables, dimensional tolerance, adsorption, chemical incompatibility, and particulates, and each one has a specific fix with a specific number attached. Check cap height at 5.5 to 6.0 mm, torque screw caps to 0.5 to 0.7 N·m, replace septa at 50 to 100 punctures, use PTFE-faced septa for trace work, and run a vial blank before every new lot. Those five habits will eliminate most of the re-runs I used to blame on the column.

References & further reading

Related resources from our lab consumables network:

External references and standards: