The four polymer families and the solvent classes each one tolerates

A 96-well plate run came back with 14% RSD on the internal standard for the last six columns of the plate, while the first six sat at 2.1%. Same method, same analyst, same lot of standards. The difference was a box of silicone-backed sealing mats pulled from a different carton. The silicone had absorbed about 4% by weight of the acetonitrile/water mobile phase over the 18-hour run, swelled, and lifted the well seals enough to let the more volatile analytes escape. That is a material selection failure, not an instrument failure. And it is the kind of thing this guide is about: which polymer belongs in which method, given the solvent, the pH, and the temperature you actually run at, and what breaks when you get it wrong.

I have spent fifteen years buying, specifying, and rejecting septa, caps, liners, tubing, and filter frits, and the single most common mistake I see from a chromatography supplier is a customer ordering “the usual PTFE/silicone” for a method that had no business using silicone. The second most common is buying EPDM because it is cheaper than PTFE-faced silicone and then wondering why the baseline drifts. Both are avoidable with a four-column decision table and a handful of numbers.

Almost every septum, cap liner, sealing mat, and length of lab tubing you will buy falls into one of four material families: PTFE (and its modified grades), silicone, EPDM, and a laminate (usually PTFE-faced silicone or PTFE-faced EPDM). Each has a characteristic solvent response, and the response is what determines whether the material belongs in your method.

PTFE: the near-universal barrier, with one real weakness

PTFE is chemically inert to essentially every solvent you will inject. It resists acetone, acetonitrile, methanol, dichloromethane, THF, DMSO, and concentrated acids and bases. Continuous service temperature runs to about 260 °C. It has no extractables worth worrying about at HPLC and GC detection limits. If you need one material that survives an unknown solvent list, it is PTFE.

The weakness is mechanical, not chemical. PTFE is soft and cold-flows. A plain PTFE septum under a crimp cap will deform and thin over time, and a needle puncture leaves a channel that does not reseal well. That is why plain PTFE is almost never used as a standalone septum in autosamplers. It is used as a facing layer bonded to an elastomer that provides the reseal. When a chromatography supplier quotes you “PTFE septum,” ask what the substrate is. A 0.25 mm PTFE film on a 1.5 mm silicone core is a very different part from a solid PTFE disc.

Silicone: excellent reseal, poor solvent resistance

Silicone reseals beautifully. A 2 mm silicone septum will take 50 to 100 punctures before coring becomes a problem, which is why it dominates autosampler septa. But silicone swells in nonpolar solvents (toluene, hexane, dichloromethane) and absorbs a measurable fraction of many mobile phases. In acetonitrile/water it can take up 3 to 5% by weight. That swelling is what lifted the well seals in the case above.

Silicone is also the material most likely to shed extractables into a sensitive method. If you are running LC-MS at low ng/mL and your blank shows siloxane peaks at m/z 281, 355, or 429, the silicone in your sample path is the first thing to suspect. Temperature ceiling for standard silicone is around 200 °C for continuous service; above that it degrades and off-gasses.

EPDM: the aqueous and polar-solvent workhorse

EPDM is a terpolymer with excellent resistance to water, aqueous buffers, and polar solvents. It handles pH from about 1 to 13 in continuous service and is the right choice for ion chromatography, aqueous SEC, and any method where the mobile phase is mostly water with a buffer. It is cheaper than PTFE-faced silicone and reseals well enough for most autosampler work.

EPDM fails in aromatic and chlorinated solvents. Toluene, xylene, dichloromethane, and chloroform will swell EPDM badly, sometimes 50% or more, and the part will fail mechanically within a run or two. It also has a lower temperature ceiling than silicone, around 150 °C, and it is not the material for high-temperature GC inlet septa.

Laminate: PTFE facing on an elastomer core

The laminate is the compromise most methods actually want. A thin PTFE film on the sample side gives you chemical resistance and low extractables; the silicone or EPDM core gives you reseal and mechanical strength. A PTFE/silicone septum at 0.25 mm PTFE on 1.5 mm silicone is the standard for HPLC autosamplers. A PTFE/EPDM laminate is better for aqueous methods where you want the EPDM core’s buffer resistance without sacrificing the PTFE barrier.

The failure mode unique to laminates is delamination. If the bond between the PTFE facing and the elastomer core is poor, the PTFE layer separates at the puncture site and you get a flap that either blocks the needle or sheds a fragment into the vial. I have seen this on cheap laminates from suppliers who bond the layers with an adhesive that softens above 60 °C. Ask your chromatography supplier whether the laminate is bonded or co-extruded, and whether the bond survives your autosampler needle temperature.

A decision table for solvent, pH, and temperature

The table below is what I would put on the wall next to the autosampler. It is not exhaustive, but it covers the methods that account for most of the volume in a typical analytical lab. Use it to pick the polymer family first, then work out the dimensions and fit (which are covered elsewhere).

Read the table as a filter, not a lookup. If your method uses acetonitrile and water at 40 °C, PTFE/silicone is fine. If it uses toluene at 60 °C, silicone is out and you need PTFE-faced EPDM or a solid PTFE part with a separate reseal mechanism. If it uses 0.1 M phosphate buffer at pH 7.4, EPDM is the right core. If it uses DMSO at 80 °C, you want PTFE facing and a core that will not swell, which usually means PTFE/EPDM rather than PTFE/silicone.

What the wrong choice actually does to a run

Material failures show up as a small number of recognizable symptoms. Learning to read them saves a lot of instrument time.

Symptom: baseline drift and ghost peaks in GC

If your GC baseline climbs over the course of a temperature program and you see unidentified peaks in the blank, the inlet septum is the usual suspect. A silicone septum running above 250 °C off-gasses siloxanes, and those siloxanes land on the column. The fix is a PTFE-faced silicone septum rated for the inlet temperature, or a lower-bleed silicone grade. Check the septum rating against your inlet temperature, not your oven temperature. An inlet at 280 °C with a 250 °C-rated septum will bleed every run.

Symptom: RSD creep across a plate or tray

This is the case I opened with. The measurement that identified it was a gravimetric swell test: weigh ten seals, soak them in mobile phase for 18 hours, weigh again. The silicone mats gained 4.1% by weight; the PTFE/EPDM mats gained 0.3%. The fix was a material change for that method, not a tighter injection protocol. If you see RSD creep that correlates with position on the plate, run the swell test before you touch the autosampler.

Symptom: needle coring and blocked lines

Coring is a punch operation, not a wear phenomenon, and the material determines how the punch behaves. A soft silicone septum cores more readily than a harder EPDM or a PTFE-faced laminate, because the needle tip pushes material ahead of it rather than cutting cleanly. If you are seeing septum fragments in the vial or on the column frit, the material is too soft for the needle gauge and the puncture count you are running. A 22-gauge needle through a 2 mm silicone septum at 100 punctures is a different risk profile from the same needle through a 1.5 mm PTFE/EPDM laminate at 40 punctures.

Field case: the buffer method that ate the caps

A lab running an ion chromatography method with a carbonate/bicarbonate eluent at pH 10.5 kept finding pinholes in the caps after about three weeks in the autosampler tray. The caps were silicone-backed. Silicone at pH 10.5 hydrolyzes slowly, and the pinholes were the visible end point of that hydrolysis. The measurement that confirmed it was a simple soak test: silicone discs in pH 10.5 buffer at 30 °C for 14 days lost 6% of their mass and developed surface crazing under a loupe. The fix was a switch to PTFE/EPDM caps, which are stable to pH 13. The lesson is that pH range in the table is a continuous-service number, not a “will it survive one injection” number. If your caps sit in the tray for a month, use the continuous number.

Field case: the DMSO method that swelled the seals

A screening lab running DMSO stocks at 70 °C on a plate-based assay saw the sealing mats lift off the plate after about four hours. The mats were silicone. DMSO at 70 °C swells silicone by 8 to 12% by weight, and the swelling lifted the mat off the well rims. The measurement that identified it was a dimensional check: the mat’s well pitch had grown from 9.0 mm to 9.6 mm after the run. The fix was a PTFE/EPDM laminate mat, which swelled less than 1% under the same conditions. This is the case where the PTFE facing does the chemical work and the EPDM core does the mechanical work, and neither alone would have held.

Selection criteria beyond the polymer name

The polymer family is the first decision, but it is not the only one. Four other criteria will separate a part that works from a part that works in your method.

• Facing thickness. A 0.25 mm PTFE facing is standard for HPLC; a 0.10 mm facing is enough for GC but may not survive a large-bore needle. Ask for the facing thickness in the datasheet, not just “PTFE-faced.”
• Durometer. Silicone septa are usually 40 to 50 Shore A; EPDM runs 60 to 70 Shore A. A softer septum reseals better but cores more. Match durometer to needle gauge.
• Extractable profile. For LC-MS, ask for the extractables data in the solvent you use, not a generic “low extractables” claim. A supplier who cannot give you a solvent-specific extractables number is telling you something.
• Certification and traceability. If your method falls under U.S. EPA SW-846 or a USP chapter, the material grade needs to be traceable to a lot. That is a procurement question as much as a technical one, and it is worth confirming with your supplier before you commit a method to a material.

For reference material on how these materials behave in real methods, the technical libraries at Restek and the consumables documentation from Agilent and Thermo Fisher are more useful than most supplier catalogs, because they publish the failure modes alongside the specs.

Troubleshooting checklist before you blame the instrument

When a run goes wrong and you suspect the consumable, work through this list in order. It takes about an hour and will save you a service call.

1. Confirm the solvent class against the material on the part. If the method uses toluene, DCM, or THF, silicone and EPDM are both suspect.
2. Confirm the pH against the continuous-service range, not the one-injection range. If the part sits in the tray for weeks, use the continuous number.
3. Confirm the temperature at the point of contact, not the oven setpoint. An inlet at 280 °C is not a 250 °C environment.
4. Run a gravimetric swell test on a spare part: weigh, soak 18 hours in mobile phase at method temperature, weigh again. More than 2% gain is a red flag.
5. Check the puncture count against the material rating. A silicone septum at 100 punctures is at the end of its life; an EPDM at 60 is past it.
6. Inspect the puncture site under a loupe for coring, delamination, or a flap. A flap means the laminate bond failed.
7. Check the blank for siloxane peaks (m/z 281, 355, 429) if you are on LC-MS. Those are silicone, not the column.

If you get through that list and the part passes every check, then you can look at the instrument. In my experience, about two-thirds of “instrument” problems in a method that has been running for months turn out to be a material change in a consumable lot. The vial closures and OEM vial suppliers are usually the first place to look, because that is where the material change is most likely to have happened quietly.

Where to spend the money

My recommendation, after fifteen years of buying this stuff: standardize on PTFE/silicone laminate for HPLC autosamplers running reversed-phase methods at or below 40 °C, and PTFE/EPDM laminate for anything aqueous, buffered, or above pH 8. Keep a small stock of solid PTFE parts for the odd method that uses DCM or THF. Do not buy silicone-backed parts for a method that runs above 50 °C or uses a nonpolar solvent, no matter how good the price is. And ask your supplier for the solvent-specific extractables data and the facing thickness before you commit a validated method to a material. If they cannot give you those two numbers, they are selling you a part, not a specification, and the difference will show up in your RSD eventually. The analytical testing labs I work with have all learned this the hard way at least once.

The decision rule is simple: pick the polymer by the solvent first, the pH second, and the temperature third, then check the facing thickness and the puncture count against the needle you actually use. If all four line up, the part will hold. If any one of them does not, the run will tell you, usually at the worst possible moment.

References & further reading

Related resources from our lab consumables network:

• Chromatography Supplier at chromatographysupplier.com
• Oem Vial at oemsamplevials.com
• Vial Closures at vialclosures.com
• Analytical Testing at labanalyticaltesting.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