Four polymers, four failure modes, and no universal winner

A 12-hour sequence on a 6890 GC kept throwing a ghost peak at 11.2 minutes, RSD on the internal standard climbing from 1.8% to 6.4%. The column was fine, the inlet liner was new, the standards were fresh. The culprit was a silicone septum that had been fine for hexane extracts but was sitting in a tray of dichloromethane. The silicone had swollen 22% by volume, the puncture channel no longer re-sealed, and solvent vapour was bleeding into the carrier stream between injections. Changing to a PTFE-faced silicone laminate dropped the ghost peak to below the detection limit on the next run. That is the whole game with which vial supplier material to use for a given solvent and temperature: the polymer is not a passive container, it is a reactive component that fails in a specific, predictable way when you put the wrong solvent or the wrong temperature against it.

Every septum and cap liner you buy from a vial supplier is one of four base chemistries, sometimes bonded into a laminate. They do not fail the same way, and they do not fail at the same solvent. If you hold one rule in your head, hold this: match the polymer to the worst solvent in the sample, not the average one, and then check the temperature against the same polymer.

PTFE: the inert workhorse and its one weakness

PTFE is the most chemically inert material in routine chromatography. It resists essentially every organic solvent, concentrated acids and bases, and it does not extract into your sample. Temperature window is roughly -100 °C to +260 °C continuous. The weakness is mechanical, not chemical: virgin PTFE is soft, cold-flows under compression, and does not re-seal well after a needle puncture. A 2 mm PTFE-only liner will leak on the third or fourth injection because the puncture channel stays open. That is why you almost never buy bare PTFE as a septum. You buy it as a facing on a laminate.

Silicone: excellent re-seal, poor solvent resistance

Silicone has the best puncture memory of the four. A good 10 mm silicone septum takes 50 to 100 punctures at 23-gauge before leakage becomes measurable, and it re-seals instantly. Temperature range is about -60 °C to +200 °C. The problem is swelling. Silicone absorbs nonpolar solvents, chlorinated solvents, and many aromatics. Swell by 15 to 25% and the septum no longer sits flat in the cap, the puncture channel widens, and you get both leakage and extractables. Against hexane, toluene, or DCM, bare silicone is the wrong answer.

EPDM: good for aqueous, poor for hydrocarbons

EPDM (ethylene propylene diene monomer) is the aqueous-phase polymer. It handles water, dilute acids, dilute bases, and polar solvents like methanol and acetonitrile well, and its temperature ceiling is around +130 °C to +150 °C depending on grade. Against nonpolar solvents and chlorinated hydrocarbons it swells badly and extracts. EPDM is the right choice for ion chromatography, aqueous pharmaceutical diluents, and any method where the sample is 90% water. It is the wrong choice for a DCM extract or a toluene standard.

Laminate: PTFE facing plus silicone or EPDM backing

The laminate is the compromise that wins most methods. A PTFE-faced silicone septum gives you the chemical resistance of PTFE on the sample side and the re-seal of silicone on the needle side. A PTFE-faced EPDM does the same for aqueous methods where you want a slightly stiffer backing. The catch is delamination: a poorly bonded laminate will separate at the PTFE/silicone interface after 20 to 30 autoclave cycles or after prolonged exposure to hot solvent vapour. Buy laminates with a documented bond and check them at incoming inspection.

The solvent-to-polymer matrix, with the numbers that matter

Here is the working table I keep on the wall. It is not exhaustive, but it covers the solvents that show up in 90% of the methods I see. The “failure” column is what you will actually observe when the pairing is wrong.

Two numbers to remember beyond the table. First, puncture count: a silicone-backed septum is good for 50 to 100 punctures at 23-gauge, but a PTFE-faced laminate drops to 30 to 50 because the facing work-hardens. Second, torque: crimp caps should be set to 1.5 to 2.0 in-lb (roughly 0.17 to 0.23 N·m) on a 11 mm crimp. Over-torque and you cold-flow even a good laminate into a leak. Under-torque and the septum lifts under headspace pressure.

Temperature windows are narrower than the datasheet says

Every polymer datasheet gives you a continuous service temperature. What it does not give you is the temperature at which the polymer starts to outgas into a headspace vial, or the temperature at which the seal relaxes under pressure. Those are the numbers that matter for a GC or HPLC method.

Silicone is rated to +200 °C, but above +150 °C it begins to release low-molecular-weight siloxanes that show up as siloxane peaks on a mass spec. If you are running headspace at 140 °C, a standard silicone septum will contaminate your scan. Use a high-temperature laminate with a documented low-bleed backing, and verify with a blank run at your method temperature before you trust it.

EPDM is rated to +150 °C, but its compression set climbs sharply above +120 °C. At 130 °C for 30 minutes, a standard EPDM septum loses about 30% of its sealing force. On a headspace vial that is a leak. On a crimp vial it is a slow vapour loss you will not see until your recovery drops 10%.

PTFE has no outgassing problem at any temperature you will use, but its cold-flow behaviour means the seal relaxes over time. A PTFE-only cap liner on a screw cap will lose torque after a week at room temperature. If you are storing extracts, use a PTFE-faced laminate with a silicone or EPDM backing that maintains compression.

For a deeper treatment of how the septum interacts with the needle and the inlet, the resources at Lab Chromatography and the technical library at Restek are worth bookmarking. The general principle from both: the septum is the first component the sample vapour meets after it leaves the vial, and it will contaminate or leak before the column does.

pH range is a real limit, not a marketing bullet

pH is the second axis, and it is where aqueous methods fail. Silicone is stable across roughly pH 3 to 10. Below pH 3, the siloxane backbone hydrolyses slowly; above pH 10, it degrades faster. EPDM is stable from pH 1 to 13 for short exposures, but concentrated bases at elevated temperature will attack it. PTFE is stable across the entire pH range, which is why it is the only polymer I trust for a 1 M NaOH extract.

The failure mode is not a dramatic rupture. It is a slow change in the septum’s durometer. A silicone septum sitting in a pH 2 buffer for a week will harden by 5 to 10 Shore A, and a hardened septum no longer re-seals. You will see it as a rising leak rate on the autosampler, or as a drift in the internal standard area count over a long sequence. If your method involves an acidic or basic diluent and the vials sit in the autosampler for hours, use PTFE-faced laminate and check the septum durometer at incoming inspection.

Field case: the RSD drift that looked like an instrument problem

A contract lab running a 96-injection pesticide panel on an Agilent LC-MS system saw the internal standard area count drop 14% from the first injection to the last. They replaced the column, the guard, and the needle, and the drift persisted. The measurement that identified it: they ran a blank vial with the same septum lot and saw the same drop, with a small peak at the solvent front. The septum was a standard silicone, the diluent was 80:20 acetonitrile:water with 0.1% formic acid. The acetonitrile had swollen the silicone enough that the puncture channel was weeping, and the formic acid had started to hydrolyse the backing. The fix was a PTFE-faced silicone laminate and a fresh lot of caps. The drift went to 2.1% across 96 injections. That is the pattern: a slow drift with a solvent-front peak is almost always the septum, not the column.

Field case: the headspace leak that only showed at 130 °C

A food-flavour lab was running headspace GC at 130 °C for 30 minutes with EPDM septa. Recovery on a spiked standard was 88% and drifting. They checked the transfer line, the loop, and the vial pressure, and found nothing. The measurement that identified it: they weighed a set of vials before and after the headspace run. The EPDM-sealed vials had lost 0.8 mg each, the PTFE-faced silicone vials had lost 0.1 mg. The EPDM had taken a compression set at 130 °C and the crimp had relaxed. The fix was a high-temperature PTFE-faced silicone septum and a re-check of the crimp torque at 1.5 in-lb. Recovery went to 98%. The lesson: at headspace temperatures, the septum is a pressure vessel component, and EPDM is not the right pressure vessel material.

What the wrong choice does to your chromatography, by symptom

This is the troubleshooting checklist I use when a method goes sideways and the instrument looks clean.

• Ghost peaks at a fixed retention time, RSD rising over the sequence: solvent swelling of a silicone or EPDM septum. Check the diluent against the table. Swap to PTFE-faced laminate.
• Baseline rise or a broad hump late in the run: extractables from the septum backing. Confirm with a blank vial and a solvent blank. If the hump appears only with the sample septum, it is the polymer.
• Siloxane peaks on MS at m/z 73, 147, 207, 281: silicone outgassing, usually above 150 °C. Move to a high-temperature laminate or lower the headspace temperature.
• Recovery drift on headspace, worse at higher temperature: compression set or leak at the crimp. Weigh vials before and after. Re-torque to 1.5 to 2.0 in-lb.
• Septum lifts off the cap or bulges: over-pressure from a volatile matrix plus a soft polymer. Use a stiffer laminate and check fill volume.
• Delamination visible as a white edge on the septum: bond failure in a cheap laminate. Reject the lot at incoming inspection.
• Hardened septum after storage: pH or solvent attack changing durometer. Check with a Shore A gauge; a 5-point shift is a reject.

If you are filtering the sample before it reaches the vial, the filter material has the same class of problem, and the guidance at HPLC Filter and PTFE Membrane Filter covers the membrane side of the same solvent-compatibility question. A PTFE membrane and a PTFE-faced septum are chosen for the same reason: nothing else is inert enough for the widest solvent range.

How to buy: the spec you write, and the questions you ask

When you specify a septum to a vial supplier, give them the polymer, the facing, the backing, the temperature, the solvent, the pH, and the puncture count. Then ask four questions. First, what is the documented service temperature for this laminate, not the base polymer? Second, what is the bond strength between the facing and the backing, and how is it tested? Third, what is the extractables profile in the solvent you are using, at your method temperature? Fourth, what is the lot-to-lot durometer tolerance?

A supplier who cannot answer the third question is selling you a commodity. A supplier who can give you a solvent-specific extractables test at your method temperature is selling you a component. The difference shows up in your RSD, not in the price. For reference on how the regulatory side frames container and extractables requirements, the U.S. EPA SW-846 test methods set the container expectations for waste and environmental samples, and the major instrument vendors publish compatibility matrices that are worth cross-checking against your own data: Agilent’s GC consumables documentation and Thermo Fisher’s chromatography and sample handling pages both list septum recommendations by method type.

For incoming inspection, set a simple acceptance test. Take five septa per lot, measure durometer with a Shore A gauge (accept within ±3 of the specified value), puncture each 30 times with a 23-gauge needle and check for leakage under 5 psi, and run a blank vial at your method temperature to confirm no ghost peaks above your detection limit. Reject the lot if any of the five fails. That test takes 40 minutes and catches the delamination and hardening failures before they reach a sequence.

The decision rule

Choose the polymer by the worst solvent in the sample and the highest temperature in the method, not by the average. If the sample contains a chlorinated solvent, an aromatic, or a nonpolar hydrocarbon, use a PTFE-faced laminate and verify the bond. If the sample is aqueous or polar and the temperature stays below 120 °C, EPDM is the economical and reliable choice. If the temperature exceeds 150 °C, use a documented low-bleed high-temperature laminate and confirm with a blank run. Never use bare silicone against a swelling solvent, and never trust a laminate without a bond test at incoming inspection. The septum is the cheapest component in the workflow and the one most likely to cost you a sequence, so specify it with the same rigor you apply to the column.

References & further reading

Related resources from our lab consumables network:

• Vial Supplier at vialsupplier.com
• Lab Chromatography at labchromatography.com
• Hplc Filter at hplcfilters.com
• Ptfe Membrane Filter at ptfemembranefilter.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