SW-846 is a living document, and chapter 3 is where containers live

An auditor walked into a client lab last spring, pulled a 40 mL VOC vial out of the refrigerator, and asked to see the method blank record for the lot it came from. The lab had the analytical results. They had the calibration curve. What they did not have was a document tying that specific vial lot to a method blank run under the same conditions, and the finding landed on the corrective action log within the hour. That is the shape of EPA method requirements for chromatography products in practice: not whether the column separates, but whether the paper trail proves the container and the consumable did not contribute to the result.

I have sat on both sides of that table, as the supplier whose certificate was being examined and as the operator who had to explain why a QC check failed. What follows is the audit-facing view of SW-846 and the container and handling clauses it inherits, plus the documents an auditor reaches for before they ever look at a chromatogram.

Most chemists know SW-846 by its method numbers: 8260 for volatiles, 8270 for semivolatiles, 8081 for pesticides, 1311 for TCLP. Fewer have read SW-846 Chapter Three end to end, and that chapter is where the container requirements actually sit. Chapter Three covers sample handling, preservation, and holding times for the whole compendium. The individual methods then add their own container language on top.

Method 5030 and 5035 handle the purge-and-trap side. Method 5035 is the one that catches people, because it requires field preservation of volatile samples in sealed vials, either by freezing or by chemical preservation with sodium bisulfate or methanol, and it specifies the vial type and headspace condition. Method 3510 and 3520 cover separatory funnel and continuous liquid-liquid extraction and specify the glassware. Method 1311 (TCLP) specifies zero-headspace extraction vessels and a 2 L or 1 L container depending on the waste code.

What the container clauses actually require

• Amber glass for semivolatile and pesticide samples, to block UV degradation of analytes like the organochlorines.
• Zero headspace for volatiles, filled so the meniscus sits at the shoulder of the vial, typically with a 40 mL vial filled to within 5 mm of the septum.
• PTFE-lined septa as the default closure, because silicone alone absorbs and re-releases volatile organics and skews the recovery.
• Preservation chemistry that matches the method: HCl to pH under 2 for metals and many semivolatiles, sodium bisulfate for the 5035 volatile protocol, no headspace for the VOC vial.
• Holding times that start at collection, not at receipt in the lab: 14 days for most semivolatiles, 7 days for mercury, 48 hours for the sulfide and cyanide fractions, and 14 days for volatiles when preserved per 5035.

None of these clauses name a brand. They describe a function. The auditor’s job is to confirm the container you used met the function, and the supplier’s job is to give you a document that lets the auditor confirm it without calling you.

The paperwork auditor asks for first is not the COA

I have watched three different EPA contractors run a container audit, and the first request is always the same: the lot traceability sheet, not the certificate of analysis. The COA is the second request. The reason is simple. The COA tells you what the manufacturer tested. The lot traceability sheet tells you which physical vials are in the box in front of you, and whether the COA on file belongs to that box.

The four documents, in the order they get pulled

1. Lot traceability linking the carton label to the manufacturing batch and the certificate. If the carton has a lot number that does not match the COA header, you have a finding before the auditor opens the COA.
2. Certificate of analysis with the specific tests run: dimensional check, septum durometer or thickness, extractable background by GC-MS at the method’s reporting limit, and closure torque or seal integrity.
3. Method blank and continuing calibration verification records for the analytical batch that used those vials, so the auditor can see the container did not contribute a peak above the reporting limit.
4. Storage and handling log showing the vials were kept in the original packaging, away from solvent vapors, and used within the shelf life stated on the carton.

A COA that says “meets specifications” without a number is worse than no COA, because it invites the auditor to ask what the specification is. A COA that lists the extractable background at a stated reporting limit, with the lot number and the test date, closes the question.

A method blank is the only container test that matters to an auditor

Here is the part suppliers rarely say out loud. The extractable background test on a COA is a screening test run by the manufacturer. The test the auditor trusts is your method blank, run on your instrument, with your method, on the day you used the vials. If the blank comes back at or below the method’s reporting limit, the container is cleared for that batch. If it does not, the container is the first suspect.

For 8260 the reporting limit is often 5 µg/L for the target list. For 8270 it can be 10 µg/L or lower for the early eluters. A vial that contributes a phthalate peak at 8 µg/L will fail an 8270 blank even though the COA said the vial passed. That is not a supplier failure; it is a screening limit that was set for a different method. The fix is to buy vials with a stated extractable background below the reporting limit of the method you actually run, and to ask for the test method on the COA, not just the number. The Restek technical library publishes extractable background data by vial lot for exactly this reason, and it is worth comparing against whatever your supplier sends.

Field case: the phthalate that came from the cap

A semivolatiles lab in the Midwest kept seeing di-n-butyl phthalate at 6 to 9 µg/L in method blanks, always in the third or fourth injection of a batch, never the first. They swapped vials and the peak moved with the vial lot, not the instrument. The COA for that lot listed extractable background at 10 µg/L, which passed the supplier’s screen but sat right at the lab’s 8270 reporting limit. The instrument was fine. The vials were within spec. The specification was wrong for the method. The fix was a change to a lot screened at 2 µg/L, and the blanks went to non-detect for six months. The lesson: the number on the COA has to be below your reporting limit, not below the supplier’s.

Septum and closure specs are where handling audits actually bite

Sample handling is not just the vial. It is the closure, the septum, and the way the sample was introduced. An auditor reviewing a 5035 batch will ask how the sample was preserved and whether the vial was filled to zero headspace. If you answer “we filled it to the shoulder” and the vial is a 40 mL with a 22 mm crimp neck, the auditor may ask for the fill volume record. The 40 mL vial filled to 38 mL leaves a headspace that will lose about 3 to 5 percent of the volatile analyte to the gas phase over 14 days. That is a real number, and it is the kind of thing a data reviewer flags.

Closure specs that show up in audit findings

• Septum thickness of 3 mm for 22 mm crimp closures is the 5035 default. A 1.5 mm septum on a 20 mm opening deforms under the crimp and leaks.
• PTFE-faced silicone for volatiles, with the PTFE side facing the sample. If the silicone faces the sample, the analyte partitions into the silicone and the recovery drifts low.
• Crimp torque on a 20 mm aluminum seal of roughly 0.6 to 0.8 N·m, enough to deform the seal skirt but not the septum. Hand crimpers vary by 30 percent operator to operator, which is why a torque tester is worth the $400.
• Screw thread of 9 mm or 13 mm depending on the autosampler, with a GPI 18-400 or GPI 24-400 thread spec where the closure is a screw cap. A mismatched thread cross-threads and the seal fails on the second use.

The vials with caps catalog lists the thread spec and septum thickness per SKU for this reason. If your purchase order does not state the septum thickness and the thread spec, you are buying a vial by picture, and the auditor will find the gap on the next visit.

Holding time and preservation records are the second-most-common finding

The auditor’s second pass is the preservation and holding time record. This is where chromatography products intersect with the sample log. The vial was collected on day one, preserved with the right chemistry, and extracted on day ten. The auditor wants to see the collection date, the preservation date, the extraction date, and the analysis date, all tied to the vial lot. If the vial lot cannot be tied to the sample, the holding time is unverifiable and the batch is flagged.

What the record has to contain

That table is the whole audit in one page. Every finding I have seen on a container audit maps to one of those six rows, and the fix is a form field, not a new instrument.

Choosing the vial and closure so the audit never finds the gap

The decision rule is to buy the container to the method, not to the catalog page. Start with the method number. 5035 for volatiles, 3510 or 3520 for aqueous extraction, 1311 for TCLP. Read the container clause in the method. Then match the vial to that clause on four parameters: material, volume, closure, and extractable background.

The four-parameter checklist

• Material: amber Type I borosilicate for semivolatiles and pesticides, clear for metals, and never soda-lime for anything with a pH swing.
• Volume: 40 mL for 5035 volatiles, 1 L amber for 8270 aqueous, 250 mL for the smaller semivolatile fractions, and 2 L for TCLP when the waste code requires it.
• Closure: 22 mm crimp with 3 mm PTFE-faced septum for 5035, 24 mm screw with PTFE liner for 8270, and a torque spec in the SOP for every closure type.
• Extractable background: stated on the COA at or below your method reporting limit, with the test method named.

The vial supplier side of this is where the paperwork starts. Ask for a sample COA before you place the first order, and read the extractable background line. If it is blank or says “pass,” ask what the pass threshold is. A supplier who cannot answer that in writing is a supplier whose COA will not survive the audit.

Field case: the cross-threaded cap that failed the second use

A metals lab was reusing 15 mL screw-cap digestion tubes for a 3010A prep. The caps were 24-400 thread, the tubes were 24-400, and they should have matched. The problem was a lot of caps with a thread pitch that ran 0.2 mm short, so the cap seated on the first turn and cross-threaded on the third. The symptom was a pressure leak during the 95 °C digestion, which showed up as low recovery on the check standard, not as a visible leak. The measurement that found it was a torque test: the cap backed off to zero at 0.3 N·m instead of holding 0.6. The fix was a cap lot with a verified thread gauge and a torque check at receipt. The lab now checks ten caps per lot with a torque wrench before releasing the lot to the bench.

What the auditor asks when the COA and the blank disagree

The hardest moment in an audit is when the COA says the vials are clean and your method blank says they are not. The auditor will not accept “the supplier says it is fine.” They will ask which test the supplier ran and at what reporting limit. If the answer is a screening limit above your method reporting limit, the finding is on your specification, not on the supplier. The corrective action is to tighten the purchase specification and requalify the lot.

This is where the Agilent consumables documentation and the Thermo Fisher sample handling references are useful. Both publish the test method behind their extractable background claims, so you can compare like with like. A COA that cites the test method and the reporting limit is auditable. A COA that cites a number with no method is a conversation, not a record.

Troubleshooting checklist for a failed container audit

• Pull the carton label and the COA. Confirm the lot numbers match.
• Check the COA extractable background against the method reporting limit. If the COA limit is higher, the specification is the finding.
• Run a fresh method blank with a new lot of the same SKU. If the blank clears, the lot is the suspect.
• Check the septum thickness and the PTFE orientation on a sample of ten closures from the lot.
• Check the crimp or screw torque on ten closures with a torque tester. Record the value.
• Pull the holding time record and confirm the clock started at collection, not receipt.
• Confirm the preservation chemistry and the final pH are recorded per sample.
• Write the corrective action against the specification, not the supplier, if the supplier met the spec you wrote.

That checklist is the same one an auditor runs, in the same order. Run it yourself before they do and the finding never makes the log.

The decision rule for the next purchase order

Write the method number and the reporting limit on the purchase order, and require the COA to state the extractable background at or below that reporting limit with the test method named. Require the lot number on the carton to match the COA. Require the septum thickness, the thread spec, and the closure torque to be stated on the SKU sheet. If the supplier cannot put those four things in writing, the vials are not auditable, and the next EPA method audit will find the gap whether or not the chromatography was clean. The chromatography products catalog and the Aijiren manufacturing documentation are built around that four-line spec, because the audit does not care how good the separation was if the container cannot be tied to the blank.

References & further reading

Related resources from our lab consumables network:

• Chromatography Products at chromatographyselling.com
• Aijiren at aijirentech.com
• Vials With Caps at vialswithcaps.com
• Vial Supplier at vialsupplier.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