We lost a 96-position sequence last March because the septa in a lot of 2 mL vials were silicone, and the autosampler was running 100% ethyl acetate. By injection 40 the peak areas for the internal standard had dropped 18% against the bracketing calibrators, and the GC-MS source was reading siloxane fragments at m/z 73, 147, 207, 281. The vials were fine. The closures were wrong. That is what this guide is about: which polymer belongs in which method, and what happens when you get it wrong. If you are choosing a 2ml vial material for a given solvent and temperature window, the decision is mostly about the septum and the liner, not the glass. The glass is almost always borosilicate 1.1 or Type I; the polymer stack above it is where methods live or die.

Most buyers pick a closure by price or by what is already on the shelf, then discover the incompatibility at validation. I prefer to work from the solvent first. Build a short list of every solvent and additive in the method, including the wash solvents and the needle rinse, because the autosampler wets the septum with those too. Then map each one against the polymer table below. If any single solvent in the list is a “no” for a polymer, that polymer is out, even if the main mobile phase is fine.

The reason is diffusion, not dissolution. A septum does not need to dissolve in a solvent to fail. It needs the solvent to swell it, extract low-molecular-weight species, and carry those species into the headspace. Silicone in ethyl acetate is the classic case. PTFE in nothing is the other extreme. The useful question is not “is this polymer compatible” but “how much of it moves into my sample at this temperature over 24 hours, and does that matter at my detection limit.”

What each polymer actually does against solvents, pH and heat

PTFE and PTFE-faced silicone

Pure PTFE is the most chemically inert septum material you can buy for a 2 mL vial. It is compatible with essentially every common HPLC and GC solvent, tolerates pH from 0 to 14, and holds up to 260 °C short-term. The problem is mechanical. Pure PTFE is hard, does not reseal well after a needle puncture, and will leak after roughly 3 to 5 punctures on a standard 8 mm septum unless it is backed by something softer. That is why most “PTFE” septa sold for 2 mL vials are actually a laminate: a thin PTFE face bonded to silicone or EPDM.

The PTFE face is the barrier. It stops the sample from touching the silicone. It works well until the needle tears the face or the bond fails. When the bond fails you get a flap of PTFE lifting at the puncture, and the silicone underneath is now exposed to the solvent. Failure mode: siloxane bleed appearing mid-run, not at the start, because the first 20 injections are fine and the 40th is not.

Silicone

Silicone has the best reseal performance of the common septum materials. A good 8 mm silicone septum will hold a seal for 50 to 100 punctures at room temperature, which is why it is the default for autosamplers that pierce the same cap repeatedly. It is also the worst choice for nonpolar and chlorinated solvents. Toluene, hexane, dichloromethane, ethyl acetate and THF all swell silicone and extract siloxanes. At 40 °C the extraction rate roughly doubles versus room temperature. Above 200 °C silicone starts to degrade on its own, and you will see cyclic siloxanes in a GC-MS blank even with no sample present.

Use silicone when the method is aqueous or methanolic, the oven stays below 200 °C, and the autosampler is doing many injections per vial. Do not use it for headspace analysis of nonpolar analytes, and do not use it for trace work at ppb levels in DCM.

EPDM

EPDM is the workhorse for aqueous and polar methods where silicone would bleed. It handles pH from about 1 to 12, tolerates steam and autoclave cycles, and is compatible with methanol, acetonitrile, water, and dilute acids and bases. It is not compatible with aromatic or chlorinated hydrocarbons, and it swells badly in hexane and toluene. Temperature ceiling is lower than silicone, around 130 to 150 °C continuous.

Where EPDM earns its place is in ion chromatography and in aqueous HPLC methods where you cannot tolerate siloxane background. The failure mode when you push it into a nonpolar solvent is swelling, which shows up as a cap that will not seat to the same torque and a septum that extrudes into the vial neck after a few hours. If your capper torque was set at 1.5 in-lb and suddenly caps are backing off, check whether the septum has swollen.

Laminate septa

The laminate is the practical answer for most mixed methods, but the construction matters more than the label. A PTFE/silicone laminate gives you chemical resistance on the sample side and reseal on the needle side. A PTFE/EPDM laminate gives you chemical resistance plus better aqueous performance. The failure mode unique to laminates is delamination: the PTFE face separates from the backing at the puncture, usually because the needle is dull or the capper is over-torquing and crushing the bond. I have seen a lot of 500 caps where the PTFE face lifted on 12% of them after 30 punctures, which is enough to poison a sequence.

The temperature window is where most methods quietly fail

Solvent compatibility gets all the attention. Temperature is where I see more silent failures, because a septum that is fine at 25 °C can outgas at 150 °C in a headspace oven. The rule I use: derate the septum’s rated ceiling by 30 °C for any method that holds the vial at temperature for more than 30 minutes. A silicone septum rated at 200 °C is a reasonable choice at 170 °C for a 20-minute headspace run. At 190 °C for an hour, it is not.

Headspace is the harsh test. The vial is sealed, heated, and the septum is the only thing between your sample and the carrier gas. Any volatile species in the septum itself, plasticizers, antioxidants, residual monomers, end up in the headspace above your sample. That is why headspace methods should use a PTFE-faced septum with an EPDM or low-bleed silicone backing, and why you should run a blank vial at method temperature before you trust a new lot. A blank that shows siloxane at m/z 73 above 1% of your analyte response is a fail.

For HPLC at ambient temperature, the temperature question mostly disappears and the solvent question dominates. For GC with a hot injector port at 250 to 280 °C, the septum sees the port temperature, not the oven temperature, and that is where a cheap silicone septum bleeds. If you are running split/splitless injection at 280 °C, buy a PTFE/silicone laminate and change it on a schedule tied to injection count, not calendar.

What the wrong choice looks like in the data

Field case one. A contract lab running EPA SW-846 Method 8260 by purge-and-trap kept failing the internal standard recovery criterion on a specific lot of vials. The symptom was a 15 to 20% low recovery for fluorobenzene and a rising baseline at the front of the chromatogram. The measurement that identified it: they ran the same water standard in vials from the failing lot and from a known-good lot, side by side, and the only difference was the closure. The failing lot had an EPDM septum; the good lot had a PTFE/silicone laminate. The EPDM was absorbing the aromatic internal standard from the aqueous sample over the 24-hour hold before analysis. The fix was a laminate septum and a shorter hold time. Recovery came back to 98 to 102%.

Field case two. A pharmaceutical QC lab running a residual solvent method by headspace GC at 140 °C for 45 minutes saw an unidentified peak at retention time 4.2 minutes that appeared only in sample vials, never in standards. They assumed contamination. The measurement that identified it: a blank vial with only water, heated under the same method, produced the same peak at about 0.3 ppm equivalent. The septum was a silicone-backed laminate with a thin PTFE face, and the face had been compromised by a dull autosampler needle that was coring rather than piercing. The fix was a needle change and a switch to a thicker PTFE face laminate. The peak disappeared.

Both cases share a pattern. The vial was not the problem. The closure was, and the failure only showed up when the method combined a specific solvent, a specific temperature, and enough time for diffusion to matter. That is why I tell buyers to test the closure in the actual method, not on a compatibility chart alone. A compatibility chart tells you what will not dissolve. It does not tell you what will extract.

A selection checklist you can run before you buy

Here is the decision sequence I use. It takes about 20 minutes per method and saves months of troubleshooting.

• List every solvent and additive in the sample, the diluent, the wash, and the needle rinse. Include water and buffer salts.
• Note the maximum temperature and hold time the vial sees, including the autosampler tray if it is heated, and the injector port if the septum is in the gas path.
• Count punctures per vial in the method. If it is more than 5, reseal performance matters as much as chemical resistance.
• Check the detection limit. At ppb levels, septum bleed becomes a real constraint even for a “compatible” polymer.
• Match to the table above, then derate the temperature ceiling by 30 °C for long holds.
• Run a blank at method temperature on a sample of the candidate lot before release to production.
• Confirm the capper torque is within the cap manufacturer’s spec, typically 1.2 to 1.8 in-lb for a 9 mm screw cap and 0.8 to 1.2 in-lb for an 11 mm crimp. Over-torque crushes laminate bonds.

If you are sourcing across multiple methods, it is usually cheaper to standardize on one laminate for everything and accept that it is not optimal for any single method, than to stock five closure types and risk the wrong one reaching the bench. That is a supply-chain decision as much as a technical one, and it is the kind of thing worth discussing with a chromatography supplier who can hold lot traceability across a single part number.

Where the glass and the liner fit in

The focus here is the polymer, but two other components interact with the material choice and are worth a sentence each. First, the glass. Borosilicate Type I at 1.1 mm wall is standard and is not the variable. What varies is the neck finish and how it mates with the cap; a mismatched neck and cap will leak regardless of septum material. Second, the liner in the cap. Most 9 mm screw caps for 2 mL vials come with a bonded liner, and if you are buying caps and septa separately you own the assembly quality. I prefer bonded closures for exactly this reason: the septum cannot shift or wrinkle during capping, which is one of the main causes of the delamination failure above.

If you are specifying an OEM vial program, put the septum material and the cap torque spec on the drawing. Do not leave it to “standard.” Standard is how you end up with silicone in an ethyl acetate method.

The recommendation, by method class

For reversed-phase HPLC with aqueous or methanolic mobile phases at ambient temperature, an EPDM septum or a PTFE/EPDM laminate is the right default. It is clean, cheap, and reseals well enough for most autosamplers.

For normal-phase, GPC, or any method using toluene, DCM, THF, or hexane, use a PTFE-faced laminate with a silicone backing, and change closures on an injection-count schedule. Pure PTFE is the fallback only if you are doing single-injection work.

For headspace GC, use a PTFE/silicone laminate with a face thickness of at least 0.010 in, verify with a blank at method temperature, and keep the hold time as short as the method allows. For split/splitless GC with a hot injector, the same laminate is the right answer, replaced on a count basis.

For ion chromatography and any method where siloxane background is fatal, EPDM is the answer, and you should not push it above 130 °C or into aromatic solvents.

If you are buying across a mixed portfolio and want one part number, the PTFE/silicone laminate is the least-bad compromise, provided you control needle condition and capper torque. Most of the failures I have dealt with were not the polymer choice itself but a dull needle or an over-torqued capper turning a good laminate into a bad one. Check those two things before you blame the material.

The decision rule is simple: pick the polymer by the most aggressive solvent in the method, then check it against the maximum temperature and hold time, then verify with a blank at method conditions before you release the lot. If the blank is clean and the reseal count matches your injection schedule, you have the right closure. If either fails, change the material, not the method. For a wider view of the consumables that sit around the vial, the HPLC consumables catalog and the technical notes at Restek and Agilent are worth keeping on the bench, and Thermo Fisher publishes septum bleed data that is useful when you are comparing lots. The 2ml vial itself is rarely the variable. The polymer stack above it is.

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