The auditor pulled 14 vials from the prep bench and asked for the lot number on the septa. We had it, because the shipping carton was still in the stockroom. What we did not have was a written statement that the septa were the same lot the method had been validated with. That single gap turned a two-hour walkthrough into a four-hour finding. If you handle environmental samples under EPA methods, the septum is not a consumable you buy by the case and forget. It is part of the documented sample container, and the paperwork trail around it is what gets checked first.
This article is about EPA method requirements for vial septa and what auditors actually ask to see. It is not about how a septum works or why they fail. It is about the paper, the container specification, and the handful of numbers an auditor will compare against your records before they ever look at a chromatogram.
SW-846 puts the container in the method, and the septum is part of the container
Start with the text. SW-846 is the EPA’s test methods for evaluating solid waste, and it is the document most environmental labs live inside. Chapter Three, “Inorganic Analytes,” and Chapter Four, “Organic Analytes,” both spend real space on sample containers and holding times. The volatile organic methods, 5030 and 5035, are the strictest. Method 5035A is explicit that samples for volatile analysis go into sealed containers with minimal headspace, and that the container closure must not introduce contamination or allow loss.
Read that again. The closure is named. The septum is the closure. For a 40 mL VOA vial, the septum is a PTFE-faced silicone disc, typically 0.125 in thick, 22 mm diameter, seated in a screw cap with a bonded or loose liner. Method 5035A tells you the sample must be preserved and sealed, and it tells you the holding time (14 days for most volatiles when preserved and cooled to 4 °C). What it does not do is hand you a septum part number. That is where the auditor starts asking questions.
What the method text actually constrains
The methods constrain the septum indirectly, through four requirements that a septum can violate:
- No contamination introduced. The septum must not leach plasticizers, curing agents, or sulfur compounds into the sample. A general-purpose silicone septum with a sulfur-cured elastomer can throw a sulfur peak into a GC-FPD run. That is a method failure, not an instrument problem.
- No loss of analyte. Volatiles diffuse through silicone. A septum that sits on a vial for 14 days at 4 °C will lose a measurable fraction of a light analyte like vinyl chloride if the face is not PTFE-barrier. The method’s holding time assumes the container holds.
- No headspace change from the closure. A septum that absorbs solvent and swells changes the headspace volume. For a 20 mL headspace vial, that shifts the phase ratio and the calibration.
- Physical integrity through handling. The container gets inverted, packed in a cooler, and shipped. A septum that cores on the first needle puncture is a container that failed.
None of those four appear as a septum specification in the method text. They appear as performance requirements on the container, and the auditor will ask you to show that your septum meets them.
The paperwork an auditor asks for first, in the order they ask
I have been through enough audits to know the sequence. It is almost always the same, and it starts with a document, not a product.
1. The container specification in your SOP
Your standard operating procedure for sample collection has to name the container. “40 mL amber VOA vial, PTFE-lined silicone septum, open-top screw cap” is a specification. “VOA vial” is not. The auditor will read your SOP against what is actually on the bench. If the SOP says PTFE-lined and the bench has a silicone-only septum, that is a finding regardless of whether the data is good. I have watched a lab get written up for a septum material mismatch on a method that had been running clean for three years. The data was fine. The paper was not.
2. The certificate of analysis or certificate of conformance
Every lot of septa you put into service should have a CoA or CoC on file. What matters to the auditor is not the logo on the letterhead. It is whether the certificate states the material, the dimensions, and the lot number that matches the carton. A CoC that says “silicone septum, 22 mm” with no lot number is a piece of paper that proves nothing. Ask your supplier for lot-specific documentation. If they cannot produce it, that is a supply-chain finding waiting to happen.
3. The lot traceability record
The auditor will pick a sample from the logbook, find the collection date, and ask which septum lot was in use that week. You need to be able to answer. The simplest system I have seen work is a stockroom log that records the carton lot number against the date it was opened. It takes 30 seconds per carton and it has saved more audits than any instrument qualification I have ever done.
4. The holding-time and preservation record
This is where the septum meets the clock. The method gives a holding time. The container has to hold the analyte for that entire window. If your septum is not PTFE-faced, you cannot defend a 14-day holding time for volatiles. The auditor knows this. The question is usually phrased as “what is the basis for your holding time?” and the answer has to include the container, not just the preservation.
5. The training record
Whoever caps the vial has to know which side of the septum faces the sample. I am not joking. A PTFE-faced septum installed upside down is a silicone septum as far as the sample is concerned. The auditor will ask whether the technician was trained on container assembly. Have the training record.
What the auditor compares, number by number
Once the paper is on the table, the auditor moves to the numbers. They compare what your SOP says against what the method requires and what the product actually is. The table below is the comparison I keep in my own audit-prep folder. It is the checklist I run before any external audit touches the bench.
| Item | What the method or SOP should say | What the auditor checks | Typical failure |
|---|---|---|---|
| Septa material | PTFE-faced silicone, sample side PTFE | Physical inspection, CoA material statement | Silicone-only septum on volatile method |
| Dimensions | 22 mm OD, 0.125 in thick for 40 mL VOA | Calipers against SOP | 20 mm septum in 22 mm cap, poor seal |
| Lot traceability | Carton lot recorded against open date | Logbook cross-reference to sample date | No lot on CoC |
| Holding time basis | Container + preservation + temperature | SOP justification statement | Holding time cited with no container basis |
| Puncture count | Single puncture per vial for volatiles | Injection log vs vial count | Multiple injections from one VOA vial |
| Cap torque | Hand-tight, consistent, no septum distortion | Visual, occasional torque check | Over-torqued cap, septum deformed, leak path |
| Temperature | 4 °C ± 2 °C, septum rated for it | Cooler log | Silicone stiffening at low temp, seal loss |
Two of those rows deserve their own treatment, because they are where I have seen the most findings.
Puncture count is a method requirement, not a convenience
For volatile methods, the vial is sampled once. The septum is punctured once. A septum rated for 50 punctures on an HPLC autosampler is not the same product as a VOA septum that gets one needle and then sits in a cooler for two weeks. If your lab is running a volatile method and pulling multiple injections from a single 40 mL vial, the auditor will find it in the injection log. The septum will show it too: a second puncture through a PTFE face that has already been deformed is a leak path. I have seen a lab defend a re-injection practice on the grounds that “the septum is good for 100 punctures.” That number came from an HPLC datasheet. It has nothing to do with a 40 mL VOA vial.
Torque is the number nobody records until it fails
Screw caps on VOA vials are hand-tightened. There is no torque specification in the method. But over-torquing distorts the septum, and a distorted septum does not seal. The practical rule I use is hand-tight plus a quarter turn, and I check it by looking at the septum after capping. If the PTFE face shows a ring impression deeper than about 0.5 mm, the cap is too tight. That is not a method number. It is a bench number that keeps the method number honest.
Field case 1: the vinyl chloride that drifted inside the holding time
A lab running Method 5035 for a landfill site kept getting low recoveries on vinyl chloride from trip blanks and field samples alike. The instrument was clean. The calibration was fresh. The preservation was verified. The recoveries were running 68 to 74 percent against a 75 percent lower control limit, so it was a marginal fail that showed up on about one run in four.
The measurement that identified it: they ran a set of fortified vials held for 0, 7, and 14 days, and the 14-day vials had lost 22 percent of the vinyl chloride relative to the 0-day set. The septa on those vials were silicone-only, no PTFE face. The stockroom had substituted them during a backorder six weeks earlier. Nobody had updated the SOP, and nobody had flagged the material change.
The fix was two-part. They went back to a PTFE-faced silicone septum on the 40 mL vials, and they added a line to the receiving procedure that any septum material change requires a holding-time re-verification before the lot goes into service. The recovery came back to 88 to 94 percent at 14 days. The auditor never saw the bad lot, because the re-verification record now exists for every material change.
Field case 2: the coring that showed up as a carryover peak
A different lab, running a headspace method on a HPLC vial line that had been repurposed for GC headspace, started seeing a ghost peak at the retention time of their internal standard on every tenth or twelfth injection. The pattern was regular enough that they suspected the autosampler. They replaced the syringe, the needle, and the loop. The peak stayed.
The measurement that identified it: they pulled the septa from the vials that produced the ghost peak and inspected them under a loupe. Every one had a cored plug of silicone sitting in the puncture, and the plug had been pushed into the vial on the next injection. The septa were 0.040 in thick silicone, which is too thin for the needle gauge on that autosampler. The needle was 23 gauge, and the septum was rated for a finer needle.
The fix was a thicker septum (0.060 in) with a PTFE face on the sample side, and a needle gauge check written into the method. The ghost peak disappeared. The lesson is that the method’s container specification has to match the autosampler’s needle, and that match belongs in the SOP. An autosampler vial septum and a VOA septum are different products for different jobs, and the auditor will ask why you chose the one you chose.
How to build the septum record before the auditor asks
The labs that pass cleanly are not the ones with the best instruments. They are the ones with a septum record that answers the question before it is asked. Here is the sequence I recommend, and it is the one I would want to see if I were auditing.
Write the container spec into the SOP, with numbers
Name the material, the face orientation, the dimensions, and the temperature rating. For a 40 mL VOA vial that reads: PTFE-faced silicone septum, 22 mm OD, 0.125 in thick, PTFE face to sample, rated -40 °C to 200 °C. For a 20 mL headspace vial: PTFE-faced silicone, 20 mm, 0.125 in, magnetic cap. The numbers are what the auditor compares against the carton.
Keep the carton lot with the CoC
When a carton is opened, record the lot number, the open date, and the CoC reference in the stockroom log. When the carton is empty, record the close date. That single log answers the lot-traceability question in under a minute.
Re-verify the holding time on any material change
Any time the septum material, thickness, or face changes, run a fortified holding-time study at 0, 7, and 14 days before the new lot goes into service. Keep the data. This is the record that turns a backorder substitution from a finding into a documented decision.
Match the septum to the needle and the injection count
Put the needle gauge and the maximum puncture count in the method. For volatile methods, that count is one. For autosampler methods, pick a septum rated for at least twice the injections you actually run, and check the coring at the end of the sequence. A vial that has been punctured past its rating is a vial that has already failed, whether or not the chromatogram shows it yet.
Buy from a supplier who can give you lot-specific documentation
This is the part I am most opinionated about. A septum is a small part with a large paper trail, and the supplier either supports that trail or they do not. Ask for lot-specific CoAs before you place the first order. Ask what happens when a material changes. Ask whether the septa are qualified against a method or just against a dimension. The suppliers who can answer those questions are the ones whose cartons survive an audit. The ones who cannot are the ones who create findings. If you are sourcing septa for environmental work, start with a supplier who treats the septa as a documented component rather than a line item. The chromatography consumables catalogs from Agilent, Thermo Fisher, and the technical notes at Restek are worth reading for the container and septum specifications they publish, because those specifications are the language the auditor speaks.
Troubleshooting checklist for the week before an audit
Run this against your own bench. Every item is something an auditor has asked me for at least once.
- Does the SOP name the septum material and face orientation? If it says “VOA vial” and nothing else, fix it now.
- Is there a CoC on file for every lot currently in service? Pull the cartons and check the lot numbers against the certificates.
- Does the stockroom log link lot numbers to open dates? If not, start one today. It takes an afternoon to backfill a month.
- Is the holding-time justification written down, including the container? A holding time cited without a container basis is a finding.
- Is the puncture count in the method? For volatiles, it should be one. For autosamplers, it should be at least twice your actual injection count.
- Is the cap torque practice documented? Hand-tight plus a quarter turn, septum face flat, no ring impression deeper than 0.5 mm.
- Is the septum rated for the storage temperature? A silicone septum that stiffens at 4 °C is a seal that fails in the cooler, not on the bench.
- Is the needle gauge matched to the septum thickness? A 23-gauge needle through a 0.040 in septum cores. Go thicker or go finer.
- Is there a re-verification record for the last material change? If the supplier substituted a lot, you need the study.
- Is the technician training record current? Face orientation, torque, and puncture count, all three.
If you can answer all ten, the septum part of the audit is a short conversation. If you cannot, the auditor will find the gap, and it will cost you more time than the checklist would have.
The decision rule
If your lab runs an EPA method that specifies a sealed container with a holding time, treat the septum as a documented component of that container: write the material, face orientation, dimensions, temperature rating, and puncture count into the SOP, keep lot-specific certificates and a carton log, and re-verify the holding time on any material change. If you cannot produce that record for the lot in service, the septum is not qualified for the method, no matter how clean the chromatogram looks.
References & further reading
Related resources from our lab consumables network:
- Vial Septa at vialsepta.com
- Hplc Vial at hplcgc.com
- Autosampler Vial at autosamplerhplc.com
- Vial at vials-filters.com
External references and standards:
