An auditor at a contract lab in Ohio pulled three things at the start of a recent SW-846 audit: the method SOP, the certificate of analysis for the lot of 9 mm screw-thread vials in the cabinet, and the sample receipt log for a batch of 12 aqueous soil extracts. She did not ask to see the chromatograms first. The vial and container documentation is the entry point because it is the cheapest thing to verify and the most common place a method falls apart on paper. If you run EPA methods and your vial records are improvised, you will spend the audit explaining instead of working.
What follows is what EPA method requirements for HPLC autosampler vial selection and handling actually say, where the language is permissive versus mandatory, and the specific paperwork an auditor reaches for before they look at a single injection. I have sat on both sides of this table. The gap between what a method requires and what a lab documents is where most findings live.
The EPA SW-846 compendium is a set of test methods, not a product specification. Nowhere in Method 8000, 8260, 8270, or 8310 will you find a part number, a septum durometer, or a glass class for the autosampler vial. What you find instead is performance-based language: the container must not contaminate the sample, must not adsorb the analytes of interest, and must preserve the sample from receipt through analysis. The method tells you the outcome it needs. It leaves the container choice to you, and then it holds you to that choice through the QA/QC section.
This is why auditors do not check your vial against an EPA list. They check whether your documentation shows you controlled the container as a variable. A method that says “use appropriate containers” is a method that expects you to have defined “appropriate” somewhere in your SOP, with a rationale, and to have verified it. That verification is the document they want to see.
Where the method language becomes a container requirement
Three clauses in the SW-846 methods do most of the work when it comes to the vial:
- Holding time and preservation. Methods specify 4 °C storage and a maximum hold time (often 14 days for aqueous, 14 days from collection to extraction for many organics, 40 days to analysis for some). The vial is the container that has to hold that condition. A screw cap that vents or a septum that is permeable to solvent vapor over 14 days is a preservation failure, not a convenience issue.
- Contamination and interference. The method requires demonstration that the container does not introduce interference at the reporting limit. In practice this means a method blank prepared in the same container type, carried through the same handling.
- Sample integrity across the autosampler. The extract sits in the vial in the autosampler for the length of the run. A 40-injection sequence at 25 minutes each is over 16 hours on the tray. The method expects the analyte concentration to be stable over that window, and the vial is part of what determines stability.
None of these name a product. All of them convert into a specification you have to write down.
The container control section an auditor reads first
Before the chromatograms, before the calibration curve, the auditor opens your SOP and looks for a container control section. If it is missing, the audit is effectively over because every downstream result inherits an uncontrolled variable. What I put in that section, and what I have seen pass a second look, is a table that fixes the container as a specification.
| Parameter | What the SOP must state | Typical value that passes review |
|---|---|---|
| Vial format | Body dimensions and neck finish | 12 x 32 mm, 9 mm screw thread, GPI 9-425 neck |
| Glass class | Type and color | Type I borosilicate, clear (amber for photosensitive analytes) |
| Seal | Septum material and cap type | Silicone with PTFE facing, preassembled screw cap, PTFE-lined |
| Septum thickness | Nominal and tolerance | 1.0 mm nominal, +/- 0.1 mm |
| Puncture rating | Rated punctures before reseal failure | Stated by supplier, verified in-house at 3 punctures minimum |
| Fill volume | Working volume and headspace | 150 to 1000 µL working, consistent headspace per method |
| Cap torque | Applied torque for preassembled caps | Hand-tight, verified 0.4 to 0.6 N·m where a torque tool is used |
| Certification | What the CoA covers | Lot-level cleanliness, dimensional conformance |
This table is the spine of the audit response. Every column is a question the auditor can ask and you can answer with a document. The HPLC autosampler vial you buy is only as defensible as the specification you wrote around it, so start there.
Why the neck finish matters to the auditor, not just the instrument
The GPI 9-425 neck finish is a dimensional standard, and a cap that does not seat to it will not hold torque. I have watched a lab fail a leak check on a whole tray because they had mixed two cap lots from different suppliers on the same vial lot. The caps looked identical. The thread pitch did not match the vial thread closely enough, and the caps backed off under the autosampler needle’s downward force. The auditor’s question was not “did it leak.” It was “how do you know which cap lot was on which sample.” The answer has to be in the records.
What the method blank and the container blank actually prove
An auditor will ask for the method blank results and, if they are thorough, for a container blank. These are not the same document and the distinction matters. The method blank is carried through the entire preparation. The container blank isolates the vial. If your method blank is clean but you have never run a container blank, you cannot separate a clean method from a dirty vial, and the auditor knows it.
What I do: once per lot of vials received, run a container blank by filling three vials from the lot with the same final solvent used in the method, capping, and injecting. Acceptance is no peak above one-third of the reporting limit at the analyte retention times. Log the result against the lot number. That log is the single most persuasive container document in the binder because it ties a specific lot to a specific result.
The paperwork sequence an auditor follows
The order is predictable. First the SOP container section, then the lot CoA, then the container blank or method blank record, then the sample receipt log, then the holding-time calculation, and only after all of that the chromatograms. If any of the first five is missing, the chromatograms do not get a fair reading. Build the binder in that order and you will not be scrambling.
Field case: the 14-day hold that failed on a screw cap
Aqueous groundwater samples for a volatile-adjacent list were held 12 days at 4 °C before extraction. The method blank was clean. The audit finding came from a re-analysis: the same extracts, re-injected from the original vials 48 hours later, showed a 15 to 20 percent loss on the more volatile analytes. The measurement that identified it was a paired comparison of the first injection and the re-injection from the same vial, same septum, same cap. The loss tracked with headspace, not with the analyte.
The cause was a preassembled cap whose silicone layer was thinner than the lot before it. The septum compressed under the cap and did not reseal at the first puncture, so the vial slowly vented over the two-day gap between runs. The fix was threefold: a container blank per lot, a septum thickness check at receipt (caliper, three vials per lot, reject below 0.9 mm), and a rule that re-injection from a previously punctured vial is only valid within 24 hours. None of this is in the method. All of it is what the auditor wanted to see as evidence of control.
Septum and cap decisions that survive an audit
The septum is where most container findings originate, and the audit question is almost always the same: how do you know the septum is compatible with your solvent and your analyte list. The answer is a documented compatibility matrix, not a supplier claim. Silicone with a PTFE facing is the default for reversed-phase work because the PTFE face is what contacts the sample and the silicone provides the reseal. The facing has to be on the sample side. I have seen vials loaded upside down in a tray and the auditor caught it in the receipt photos.
For analytes that adsorb to silicone, the PTFE facing is doing the work and the silicone is only the spring. For highly volatile analytes, the reseal is doing the work and a thin septum is a liability. You cannot optimize both with one septum, which is why the SOP should name the septum by analyte class rather than by part number alone. The sample vial selection and the septum selection are separate decisions and should be documented separately.
Puncture count is a number you verify, not a number you trust
Suppliers quote puncture ratings, often 3 to 5 for a standard silicone/PTFE septum. That number is a lab-condition average, not a guarantee for your needle gauge and your injection volume. Verify it: take a fresh vial, puncture it repeatedly with your actual needle, and inject a standard after each puncture. The point where the peak area RSD across replicates exceeds 2 percent is your practical limit, and it is usually lower than the datasheet. Write that number in the SOP and use it to set your re-injection policy. An auditor who sees a verified puncture limit and a matching re-injection rule has nothing to challenge.
Holding time, temperature, and the vial as a preservation device
Methods set holding times and storage temperatures, and the vial is the device that has to deliver both. A 4 °C storage requirement is meaningless if the cap does not seal, because the headspace exchanges with the refrigerator air. The auditor’s holding-time calculation starts at collection and ends at extraction or analysis, and any gap in the chain is a finding. The vial records that support that chain are the receipt log (date, time, condition), the storage log (temperature, duration), and the extraction or analysis date. Three documents, one chain.
The failure mode I see most is not a broken chain. It is a chain that exists but does not name the container. A receipt log that says “12 samples received, stored at 4 °C” does not tell the auditor what container held them. Add the container lot to the receipt log and the chain closes. It is a one-column change that has saved more than one lab a finding.
What an auditor asks when the SOP and the bench disagree
The most uncomfortable audit moment is when the SOP says one thing and the bench does another. The auditor finds this by asking the analyst to walk through a sample from receipt to injection. If the SOP says preassembled caps torqued by hand and the bench is using a torque tool set to a different value, that is a finding even if the results are fine. The fix is to make the SOP match the bench, not the other way around, and then train to it. A realistic SOP that the analyst actually follows beats an idealized one every time.
Two questions I make sure the bench can answer cold: what lot number is on the vials currently in the autosampler, and what is the re-injection policy for a previously punctured vial. If the analyst hesitates on either, the container control is not real. The chromatography products you buy do not fix a bench that cannot answer those two questions.
A pre-audit container checklist
- SOP container section names vial format, glass class, septum, cap, torque, and fill volume with tolerances.
- Lot CoA on file for every vial lot in the cabinet, cross-referenced to the receipt log.
- Container blank result per lot, with acceptance criterion and analyst initials.
- Septum thickness check at receipt, three vials per lot, caliper reading recorded.
- Puncture limit verified in-house and written into the re-injection policy.
- Holding-time chain complete: receipt, storage, extraction or analysis dates, container lot on the receipt log.
- Bench walkthrough rehearsed: analyst can state current lot and re-injection policy without looking.
Run this checklist against your own binder before the auditor does. Every line you cannot answer is a finding you can still prevent. The Aijiren manufacturing documentation and the Restek technical resources on septum compatibility are both useful for filling the gaps, and the Agilent consumables documentation and Thermo Fisher sample handling references give you the instrument-side dimensional context when you write the specification.
Field case: the lot that passed the blank and failed the audit
A different lab had a clean method blank and a clean container blank for a vial lot, and still got a finding. The auditor asked for the container blank, saw it was run on a different lot than the vials in the cabinet, and the chain broke. The lab had received two lots in the same month, run the blank on the first, and never repeated it for the second. The measurement that identified the gap was the lot number on the CoA versus the lot number on the container blank log. Two numbers that did not match.
The fix was procedural, not analytical: container blank per lot, no exceptions, and the blank log keyed to the CoA lot number. The vials were fine. The documentation was not. This is the pattern I see most often. The product is rarely the problem. The record of the product is.
The decision rule
Treat the autosampler vial as a specified, lot-controlled component of the method, not a consumable you reorder by habit. Write the container section into the SOP with dimensions, glass class, septum material and thickness, torque, fill volume, and a verified puncture limit. Run a container blank per lot and key it to the CoA lot number. Put the container lot on the receipt log so the holding-time chain names the device that held the sample. If you do those four things, the auditor’s first three questions have answers before they are asked, and the chromatograms get read on their merits. If you do not, no vial on the market will save the audit.
References & further reading
Related resources from our lab consumables network:
- Hplc Autosampler Vial at hplcautosamplervials.com
- Sample Vial at samplevialprice.com
- Chromatography Products at chromatographyselling.com
- Aijiren at aijirentech.com
External references and standards:
