The septum on the GC autosampler vial had been punctured 62 times over a three-day sequence, and by run 55 the octadecane internal standard area had drifted 11.4% low. The analyst blamed the inlet liner. The actual problem was a silicone septum sitting under a cap with 4.2 in·lb of torque, slowly extruding into the inlet and shedding a silicone film onto the column head. Switching to a PTFE-faced silicone septum at the same torque brought the RSD back to 1.8% across 100 injections. That is the whole argument of this article in miniature: which reagent bottle material you use for a given solvent and temperature decides whether your method holds or fails quietly.
Most labs I audit pick bottle material by habit. They buy borosilicate glass for everything aqueous, reach for PTFE when the solvent looks aggressive, and assume silicone is silicone. That works until it doesn’t, and the failure is usually a number drifting rather than a bottle cracking. What follows is how I actually map polymer grades to methods: PTFE, silicone, EPDM, and laminate, each with its solvent envelope, pH window, temperature ceiling, and the specific failure each wrong choice produces. Dimensions and pricing are covered elsewhere on the site; this is purely a materials decision.
Temperature gets all the attention, but the first filter is solvent class. A bottle that survives 150 °C in water can swell 3% in dichloromethane at room temperature. The reason is solubility parameter matching: a polymer with a similar Hildebrand solubility parameter to the solvent will absorb it, swell, and lose mechanical strength. PTFE has a solubility parameter around 6.2 (cal/cm³)^½, which is why it resists nearly every organic solvent. EPDM sits higher, and silicone higher still, so their solvent envelopes are narrower even though their temperature ceilings look respectable on a datasheet.
I sort solvents into four buckets before I touch a catalog. Aliphatic hydrocarbons (hexane, heptane, cyclohexane) are mild for all four polymers. Aromatic hydrocarbons (toluene, xylene) swell EPDM and silicone noticeably and do nothing to PTFE. Chlorinated solvents (DCM, chloroform) are the harshest test and will pull silicone apart within hours. Polar aprotic solvents (DMF, DMSO, NMP) attack EPDM and swell silicone, and DMF at 60 °C is where I have seen the most surprising failures. Acids and bases follow pH rules rather than solvent rules, and those are covered below.
Where each polymer actually sits
PTFE is the default for aggressive organics and for anything that must stay dimensionally stable. It handles the full pH range 0 to 14 and continuous service to 260 °C. The trade-off is mechanical: PTFE cold-flows under compression, so a PTFE-only cap liner will relax torque over time. That is why most cap liners are PTFE-faced rather than solid PTFE, with a silicone or EPDM backing that carries the compression load.
Silicone is the septum standard for a reason: it reseals after a needle punctures it, and it stays elastic from -60 °C to 200 °C. Its weakness is solvent compatibility and its tendency to shed particles if the needle gauge is too large. A 23-gauge needle (0.64 mm OD) in a 2 mm silicone septum leaves a hole that closes; a 16-gauge needle does not.
EPDM handles aqueous systems beautifully, including dilute acids and bases, and it holds up to 150 °C in steam. It is the right answer for pH 2 to 12 aqueous methods and the wrong answer for aromatic or chlorinated solvents.
Laminate (typically PTFE film bonded to silicone or EPDM) exists to combine the chemical resistance of the fluoropolymer face with the compression and reseal behavior of the elastomer backing. Get the bonding right and it is the best general-purpose liner; get it wrong and the PTFE film delaminates and ends up in your sample.
pH windows and temperature ceilings, with the numbers
A bottle selection guide that gives you “good acid resistance” is useless. Here are the working numbers I use, drawn from manufacturer chemical resistance charts (Agilent and Restek both publish usable tables for their sample handling lines) and from what I have seen fail in practice.
| Polymer | pH window (continuous) | Temperature ceiling (continuous) | Best solvent class | Failure mode when misapplied |
|---|---|---|---|---|
| PTFE | 0 to 14 | 260 °C | Chlorinated, aromatic, polar aprotic | Cold flow relaxes torque; solid PTFE liners leak after thermal cycling |
| Silicone | 3 to 10 | 200 °C (septum), 180 °C (liner) | Aliphatic, aqueous neutral | Swells and sheds in DCM, toluene; coring at large needle gauge |
| EPDM | 2 to 12 | 150 °C | Aqueous, dilute acid/base, steam | Hardens and cracks in aromatic or chlorinated solvent; loses reseal |
| PTFE/silicone laminate | 0 to 14 (PTFE face) | 200 °C (silicone backing) | Mixed aqueous/organic | Delamination if bond fails; silicone backing swells if film is pierced |
Read the table with two cautions. First, the pH window is for continuous contact, not for a one-hour extraction; a pH 13 method run for 24 hours will attack EPDM even though the chart says 12 is fine. Second, the temperature ceiling assumes no solvent. DMF at 60 °C is harder on EPDM than water at 120 °C, because the solvent softens the polymer before the heat gets a chance to age it.
What each wrong choice actually does to the method
This is the part that matters more than the datasheet. The failure you get tells you which polymer you should have used. I have grouped them by the four materials, because the symptom is diagnostic.
PTFE: torque relaxation, not chemical attack
PTFE does not dissolve, swell, or crack in any solvent you will put in a reagent bottle. What it does is cold-flow. A solid PTFE liner under a screw cap torqued to 4.5 in·lb will lose roughly 30% of its sealing force after 48 hours at 80 °C and after a thermal cycle back to room temperature. The symptom is a slow leak at the cap, visible as a ring of dried residue on the bottle shoulder or a falling headspace pressure in a sealed standard. The fix is a PTFE-faced laminate or a PTFE-lined cap with a silicone backing, torqued to 4 to 5 in·lb and re-checked after the first thermal cycle. Do not over-torque solid PTFE to compensate; you will crack the cap before you stop the leak.
Silicone: swelling, then shedding
Silicone in toluene or DCM swells by 15 to 25% within a few hours. Swollen silicone is soft, and soft silicone cores. The failure I see most is a septum that looks intact but has a ring of silicone particles sitting on the bottom of the vial, which then get injected onto the column. The tell is a sudden baseline rise on the GC or a cluster of late-eluting peaks that appear only in the last third of a sequence. The fix is to move aromatic and chlorinated methods to a PTFE-faced septum or a PTFE-lined cap, and to keep silicone for aliphatic and aqueous-neutral work. If you must use silicone with a harsh solvent, do not reuse the septum; the reseal property is gone after one puncture in a swelling solvent.
EPDM: hardening and cracking
EPDM is the workhorse for aqueous methods and the trap for organic ones. In toluene it does not visibly swell; it hardens, and a hardened EPDM septum stops resealing. The symptom is a slow leak at the cap that shows up as a loss of the more volatile analyte, usually the first peak in the run. I have seen a headspace method lose 40% of its methanol response over a 12-hour sequence because the EPDM liner had hardened and the cap was no longer gas-tight. The fix is a PTFE-faced silicone liner for any method with an organic solvent above 5%, and EPDM reserved for purely aqueous systems and steam-sterilized buffers.
Laminate: delamination and the pierced film
Laminate is the best general answer, but it has two specific failure modes. Delamination happens when the PTFE film separates from the elastomer backing, usually after repeated autoclaving or after a thermal cycle that exceeds the bond’s temperature rating. The symptom is a loose, wrinkled film inside the cap. The second failure is a pierced film: if the needle goes through the PTFE face and into the silicone backing, the solvent now contacts silicone directly and you get the swelling problem back. The fix is to inspect caps after autoclaving and to use a needle guide so punctures land in the same spot rather than tearing the film.
Temperature windows in practice: three cases
Three field cases, each with a symptom, the measurement that identified it, and the fix. These are the ones I use when training new analysts, because each one is a different polymer failing in its own way.
Case 1: the drifting internal standard
A GC method for residual solvents used silicone-septum vials at 4.2 in·lb cap torque. Over a 100-injection sequence the internal standard area fell 11.4%. The measurement that identified it was a septum coring check: we counted punctures and found the septum had been used for 62 injections, well past the 40-puncture limit I now recommend for silicone. The fix was a PTFE-faced silicone septum and a 40-injection replacement schedule, which brought the RSD to 1.8%. The lesson: silicone is fine for aliphatic solvents, but only if you respect the puncture count.
Case 2: the autoclave delamination
A microbiology lab autoclaved PTFE/silicone laminate-lined bottles at 121 °C for 30 minutes and reused them for 20 cycles. By cycle 15 the PTFE film had wrinkled and separated on 6 of 40 caps. The measurement was a simple visual and a leak test: we filled the bottles with a dye solution, capped them, and inverted them for 10 minutes. Six leaked at the cap. The fix was to move to a solid PTFE-lined cap for the autoclave step and reserve laminate for room-temperature storage. The lesson: laminate has a bond temperature, and autoclaving is a thermal cycle the bond was not designed for.
Case 3: the EPDM leak in a headspace method
A headspace method for methanol in aqueous matrix used EPDM-lined caps at 90 °C for 30 minutes. Methanol response fell 40% across a 12-hour sequence. The measurement was a weight-loss check: we weighed 10 capped bottles before and after the sequence and found a mean loss of 0.8 mg per bottle, consistent with a slow leak. The fix was a PTFE-faced silicone liner, which held the response within 3% over the same sequence. The lesson: EPDM is fine for water at room temperature, but a headspace method at 90 °C with even a trace of organic solvent needs a fluoropolymer face.
A decision checklist you can run before the method is validated
Before you commit a bottle to a validated method, run this checklist. It takes about 20 minutes and it will save you a failed sequence.
- Solvent class: list every solvent in the method and its percentage. If any chlorinated or aromatic solvent is above 5%, the liner must be PTFE-faced.
- pH range: state the continuous pH, not the adjusted pH. EPDM above pH 12 or below pH 2 is a no.
- Temperature: state the maximum continuous temperature and the number of thermal cycles. Above 150 °C, PTFE or glass; above 200 °C, PTFE only.
- Puncture count: silicone septa are good for about 40 punctures with a 23-gauge needle; PTFE-faced septa for about 100. Replace on schedule, not on symptom.
- Torque: 4 to 5 in·lb for screw caps with laminate liners; re-check after the first thermal cycle. Solid PTFE liners need a backing to hold torque.
- Autoclave: if the bottle is autoclaved, use solid PTFE or glass, not laminate. Laminate bonds fail above 121 °C over repeated cycles.
- Leak test: dye-and-invert for 10 minutes on a sample of 10 caps per lot. A 0.8 mg weight loss over a 12-hour sequence is a leak, not evaporation.
For the analytical side of this, the consumables you pair with the bottle matter as much as the bottle itself. The HPLC autosampler vial and its septum follow the same polymer rules, and the sample vial you choose should match the solvent class of the method. If you are sourcing across a full workflow, the reagent bottle range and the chromatography products catalog both publish material grades, and the Agilent GC accessories and Thermo Fisher sample handling pages carry chemical resistance tables you can cross-check against your solvent list. For regulatory context on container requirements in waste and environmental methods, EPA SW-846 is the reference, and Restek’s technical resources have useful septum and liner compatibility data.
Where the four materials land in a real lab
If I had to stock one liner for a general analytical lab, it would be the PTFE-faced silicone laminate, because it covers the widest solvent and temperature envelope. But I would keep solid PTFE-lined caps for autoclave and high-temperature work, silicone septa for aliphatic GC methods where reseal matters, and EPDM for aqueous buffers and steam. The decision rule is simple: match the polymer to the harshest condition in the method, not the average. A method that runs 95% aqueous and 5% DCM is a DCM method for material purposes. Pick the liner that survives the worst hour, then set the torque and the replacement schedule around it. That single rule has kept more sequences on the rails than any datasheet I have read.
References & further reading
Related resources from our lab consumables network:
- Reagent Bottle at reagentbottleforsale.com
- Hplc Autosampler Vial at hplcautosamplervials.com
- Sample Vial at samplevialprice.com
- Chromatography Products at chromatographyselling.com
External references and standards: