An auditor asked me for the extractable carbon data on a lot of 40 mL VOA vials and I had 11 minutes to find it. That is what an EPA method audit feels like. Not a review of the instrument log. Not a walk through the calibration curve. A stopwatch on the paperwork for the container that holds the sample, because under SW-846 the container is part of the method. If you buy vials, septa, or closures for environmental work, the EPA method requirements for a chromatography supplier and the auditor expectations that come with them apply to you, not just to the lab that runs the analysis. I have been on both sides of that table, selling the consumables and sitting in the audit room, and the gap between what a supplier assumes is enough and what an auditor actually opens first is wider than most people think.
Most suppliers read SW-846 as a set of analytical procedures. Read it again as a procurement document. Method 5030 (purge and trap), 5035 (closed-system purge and trap for solids), 8260 (VOCs by GC-MS), and 8270 (SVOCs) all specify the vessel, the volume, the headspace, and the preservation. The container is written into the method the same way the column temperature program is. When you ship a case of vials, you are shipping a component of a regulated method. The U.S. EPA SW-846 test methods page is where that language lives, and any supplier who has not read the container sections of 5030 and 5035 line by line is guessing.
What the methods actually pin down, and what you should be able to state without looking it up:
- 40 mL nominal for aqueous VOC samples in 5030 and 5035, filled to zero headspace (a meniscus that domes above the rim).
- Amber glass for anything light-sensitive, clear for the rest, borosilicate or type I glass, not soda-lime.
- PTFE-lined silicone septa as the default closure, with the liner facing the sample.
- HCl to pH < 2 for most aqueous VOC preservation, with the caveat that some analytes (the method lists them) get no acid.
- 4 °C ± 2 °C storage, and a holding time of 14 days for most aqueous VOC samples, 7 days for some.
None of that is exotic. The point is that a supplier who cannot tie each of those numbers to a lot and a certificate is a supplier who will cost you a finding. The method does not care that the vial looked fine. It cares that you can prove the vial met the specification on the date it was made.
What the auditor opens before the instrument log
Here is the sequence I have watched play out in three different contract labs. The auditor does not start with the chromatograms. The auditor starts with the sample receipt log and works backward into the container.
The certificate of analysis, and whether it is lot-specific
The first document requested is the CoA for the specific lot number on the box in the stockroom. Not a generic spec sheet. Not a “meets or exceeds” statement. A CoA with the lot number, the manufacturing date, the test methods used, and the measured values. If your supplier sends one CoA per year and applies it to every lot, that is the first finding. Auditors have seen that trick and they check the lot number against the receipt date.
The extractable and volatile residue data
This is where suppliers get caught. A vial that leaches trace organics or a septum that off-gasses into the headspace will show up as a ghost peak, and the auditor knows the ghost peak has a source. The relevant test is a solvent blank run against the container: fill with reagent water or the method’s blank matrix, run it through the method, and show the target analyte list is clean at the reporting limit. For VOC work the reporting limit is often 0.5 to 5 µg/L, and a septum that has not been baked and conditioned will blow that. Ask for the blank data on the lot. If the supplier cannot produce it, you are the one who has to generate it before you can defend the lot.
The dimensional and closure data
Auditors who have been burned ask for crimp or thread dimensions and septum thickness because a closure that does not seal to the method’s tolerance is a headspace leak waiting to happen. A 20 mm crimp finish has a defined neck geometry; a 24-414 thread has a defined thread pitch. If the vial body and the closure come from different suppliers with no dimensional cross-check, that is a documented risk. The vial closures you pair with a body need a stated dimensional match, not a hopeful one.
The paperwork an auditor asks for first, in order
I keep this as a literal checklist because it is the order the requests arrive in. If a supplier cannot answer items 1 through 4 within a day, the lot is at risk regardless of how good the glass looks.
| # | Document | What the auditor checks | Acceptable answer |
|---|---|---|---|
| 1 | Lot-specific CoA | Lot number matches the box; test methods named; measured values present | One CoA per lot, dated, method-referenced |
| 2 | Extractable/volatile residue blank | Target list clean at the reporting limit | Blank chromatogram at ≤ 5 µg/L for VOC list |
| 3 | Dimensional spec | Body OD, neck finish, septum thickness, liner thickness | Stated tolerances in mm, e.g. 22.5 ± 0.5 mm OD, 20 mm crimp |
| 4 | Seal integrity data | Closure torque or crimp force, leak test result | Crimp at ~1.5 to 2.0 in-lb equivalent, leak-tested |
| 5 | Material and liner grade | PTFE facing thickness, silicone durometer | PTFE 0.05 to 0.13 mm, silicone 45 to 55 Shore A |
| 6 | Cleanliness/packaging | Whether vials ship pre-cleaned and in what | Class-appropriate packaging, no cardboard dust contact |
That table is the audit, compressed. Every row is a question I have watched a supplier fumble. The fix is not more marketing. It is a document trail that survives a stopwatch.
Why “pre-cleaned” is the claim auditors test hardest
Pre-cleaned is the most abused phrase in this category. The method does not define it. An auditor will, and the definition they use is empirical: run the blank and show me the result. A vial sold as pre-cleaned that has not been through a validated cleaning process (solvent rinse, then a bake, then a verified blank) is just a vial that was handled carefully. The difference shows up at 0.5 µg/L, not at the parts-per-million level where everything looks clean.
I have seen a lot of 40 mL vials fail a method blank at 2.1 µg/L for a common solvent that was never in the sample. The supplier’s answer was “they are pre-cleaned.” The measurement that identified it was a reagent-water blank run on three vials from three positions in the case. Two were clean. One was not. That is a cleaning-process control problem, not a random defect, and it means the lot is unusable for the method regardless of the CoA. The fix was switching to a supplier who could show a per-lot blank and a bake temperature on the certificate. If you want to see how the container side of this is specified, the OEM vial documentation is a useful reference for what a complete specification actually contains.
The septum is the part the auditor will not forgive
Glass is inert and easy to defend. The septum is where the method meets the physical world, and it is where I have seen the most findings. Two numbers matter more than any datasheet adjective: the liner facing and the conditioning.
A PTFE-faced silicone septum has a thin PTFE film on the sample side. That film is the chemical barrier. If it is too thin (below roughly 0.05 mm) it will corrugate or puncture through during crimping or sampling, and the silicone underneath contacts the sample. If the septum has not been conditioned (baked at an elevated temperature to drive off residual volatiles), it will off-gas into the headspace and show up as a ghost peak, usually early in the run. The method’s blank will catch it. The auditor will ask why.
The physical failure mode is just as real. A septum that is over-crimped or under-crimped leaks. Over-crimp and the PTFE film splits at the neck; under-crimp and the seal is incomplete. The crimp force that works is a narrow window, and it depends on the neck finish and the septum thickness together. This is why I insist on buying body and closure as a matched set with a stated dimensional cross-check, and why I tell people to read the closure spec before the vial spec. For a deeper look at how the seal actually behaves under load, the chromatography supplier reference material covers the mechanics.
Field case: the ghost peak that was not the column
Symptom: a contract lab running Method 8260 saw a consistent early-eluting peak at about 1.8 µg/L in method blanks, present in every batch, absent when the same water was run in a different vial type.
Measurement that identified it: they ran the blank in three container types (their standard 40 mL vial, a competitor’s, and a baked control) and the peak tracked the septum lot, not the glass. The septum lot had a PTFE facing measured at 0.04 mm, below the supplier’s own printed spec of 0.05 mm minimum. The film was thin enough that the crimp corrugated it and the silicone contacted the water.
Fix: reject the lot, move to a septum with a verified 0.10 mm PTFE facing and a documented bake, and add a per-lot blank to the incoming inspection. The ghost peak disappeared. The finding was avoided because the lab could show the corrective action. If they had not measured the facing thickness, the audit would have ended differently.
Field case: the 14-day holding time that was really a 9-day holding time
Symptom: a lab’s aqueous VOC results drifted low on a set of samples that were within the stated 14-day holding time. The auditor flagged it as a preservation or container problem.
Measurement that identified it: the sample receipt log showed the vials arrived with a visible headspace bubble of about 3 mm. Zero headspace is a method requirement, and a 3 mm bubble in a 40 mL vial is roughly 0.2 mL of air, enough to partition volatile analytes out of the aqueous phase. The bubble was traced to a closure that sealed initially but relaxed over transit because the crimp was under-torqued on that lot.
Fix: the lab moved to a closure with a verified crimp setting and added a headspace check at receipt. The holding time did not change. The effective holding time did, because the container was not holding the sample the way the method assumed. This is the part suppliers miss: the method’s holding time is conditional on the container doing its job. A closure that leaks does not extend the holding time, it shortens it.
What a compliant supplier can hand you on day one
The suppliers who survive an EPA audit are the ones who treat the documentation as a product, not an afterthought. Here is the decision rule I use when qualifying a new supplier for environmental work. If any of these is missing, the supplier is not ready for method-driven customers.
- Per-lot CoA with lot number, manufacture date, test methods, and measured values, not a generic spec sheet.
- Blank data for the relevant method matrix (aqueous VOC, SVOC, or both) at the method reporting limit.
- Dimensional tolerances in millimeters for body OD, neck finish, septum thickness, and liner thickness.
- Seal integrity data (leak test or crimp/torque specification) tied to the closure, not just the vial.
- Material grade for the liner: PTFE facing thickness and silicone durometer, stated as numbers.
- Packaging and cleanliness statement that matches how the vials are actually handled before they reach you.
That list is also the audit checklist. The auditor is not asking for anything a serious supplier does not already have. The Thermo Fisher chromatography and sample handling resources and the Restek technical resources both publish the kind of container and septum data that a method-driven buyer should expect, and comparing a candidate supplier’s paperwork against those references is a fast way to spot a thin file. For the instrument side, the Agilent GC columns and accessories documentation shows how the consumable spec is written when the manufacturer takes the method seriously.
The one question that separates a compliant lot from a finding
Ask the supplier this, in writing, before you place the order: “For the lot you will ship, what is the measured blank result at the method reporting limit, and what is the PTFE facing thickness on the septum?” A supplier who can answer both with numbers is a supplier whose paperwork will survive the stopwatch. A supplier who answers with adjectives is a supplier who will cost you a finding. The EPA method requirements for a chromatography supplier are not about the glass. They are about whether the supplier can prove, on paper and per lot, that the container meets the method on the day it was made. Buy the documentation as deliberately as you buy the vial, and the audit becomes a records check instead of a scramble.
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: