A 47 mm crimp vial from a new supplier failed USP <621> on the third injection of a 20-vial qualification run: peak area RSD came back at 4.8% against a method limit of 2.0%, and the injection port septum showed a 1.1 mm core punched through the center. The vendor’s certificate of analysis said “pass.” That gap between what a CoA claims and what a lot actually does is the reason a sample vial supplier audit checklist has to be built around evidence you can score, not paperwork you can file. I have run these audits from both sides of the table, as a buyer and as the person whose factory was being walked. The difference between an audit that predicts a bad lot and one that just produces a signed form is almost entirely in what you request before anyone shows up.
A 40-page quality manual tells you the vendor has a document control clerk. The index tells you which procedures exist and which are missing. Ask for the controlled-document list with revision numbers and effective dates. Then ask for three specific procedures by number: incoming glass inspection, septum lot release, and nonconforming material disposition.
What I look for in the index:
- An incoming inspection procedure that names the instrument. If it says “visual inspection per AQL” and nothing else, the vendor is checking for chips and nothing more. Dimensional verification of the 12 x 32 mm body, the 6 mm neck OD, and the 11 mm crimp finish needs calipers or an optical comparator with a stated sampling plan.
- A septum release procedure with a hardness and a durometer standard. Chlorobutyl and butyl septa are typically specified at Shore A 45 to 55. If the procedure has no durometer callout, the vendor is not measuring what causes coring.
- A nonconformance procedure with a disposition authority named by role. “Quality manager” is fine. “Sales” is a red flag, and I have seen it.
If the index is dated more than 24 months ago with no revisions, the system is dormant. Dormant systems ship good lots until they do not.
The three records that predict a bad lot before it ships
You cannot audit your way to lot quality. You can, however, ask for records that expose whether the process is in control. Three records do most of the work.
Control charts on neck ID and crimp finish diameter
Ask for the last six months of X-bar/R charts on neck inner diameter and on the 11 mm crimp finish. The spec for a 12 x 32 mm vial neck ID is typically 6.00 ± 0.10 mm; the crimp finish OD is 11.00 ± 0.20 mm against the GPI 11 mm dimensional standard. What matters is not that the points are in spec. It is whether the control limits are calculated from the process or pasted in from the spec sheet. A chart whose UCL and LCL exactly equal the print tolerance is a chart nobody is using. Look for a Cpk above 1.33 on neck ID. Below 1.0, expect seating problems on your crimper.
Septum compression-set data
Compression set is the property that decides whether a septum reseals after the needle withdraws. Ask for compression set at 100°C for 22 hours per ASTM D395 Method B. A butyl septum that sets more than 30% under that protocol will leak on a hot autosampler tray. Vendors who cannot produce this data are buying septum sheet from a converter and slitting it, which means they do not control the compound.
Extractable and leachable screening
For LC-MS work, ask for the extractable profile in the relevant solvent. A generic “USP <661>” statement is not enough. You want the actual GC-MS or LC-MS scan of a solvent blank stored in the vial for 24 hours at 40°C, with the peak list. I have rejected a lot on the strength of a single siloxane peak at m/z 281 that showed up at three times the level of the incumbent supplier’s vial. The vendor’s response was that the peak was “below the reporting threshold.” It was, for their threshold. Not for mine.
If the vendor cannot supply extractable data, that is itself the finding. It tells you the compound is purchased, not formulated, and the audit scores accordingly.
What to ask for in advance so the audit day is short
The pre-audit document request is where most of the value sits. Send it 10 working days ahead and require responses in a single indexed folder. The items that matter for a sample vial supplier audit checklist:
- ISO 9001 certificate with the certification body named, plus the scope statement. A certificate scoped to “distribution of laboratory consumables” does not cover molding. Check the scope against what you are buying.
- Glass supplier identity and the Type I borosilicate specification. ASTM E438 Type I, Class A or B. Ask for the hydrolytic resistance result at the glass supplier level, not just the vial level.
- Septum compound data sheet with the base polymer, filler, and cure system. Chlorobutyl versus bromobutyl matters for temperature rating: chlorobutyl is good to about 100°C continuous, bromobutyl to roughly 110°C.
- Dimensional capability study (Pp/Ppk) on the three critical dimensions, run on at least 30 consecutive parts.
- Change notification procedure with a stated notice period. Anything under 30 days is not a procedure.
- Retention sample policy with the retention period and storage condition.
Score each item on arrival: 2 for complete and specific, 1 for partial, 0 for absent. A vendor that scores below 8 out of 12 on the document set is not worth a site visit yet. That pre-screen alone has saved me two trips a year.
How to score a vendor without arguing about it
Opinion-based audits fail because two auditors score the same vendor differently. Build a weighted rubric before the first call and agree the weights internally. Here is the one I use, with the weights that have held up across roughly 20 vial supplier audits.
| Category | Weight | What earns full marks | Typical failure |
|---|---|---|---|
| Dimensional capability | 25 | Cpk ≥ 1.33 on neck ID, crimp OD, and body length; 30-part study supplied | Charts plotted to spec limits, no Cpk |
| Septum control | 20 | Own compound or named compound supplier; durometer and compression set data; puncture test protocol | Slit sheet from a converter, no durometer record |
| Chemical cleanliness | 20 | Extractable scan in the relevant solvent; Class 10,000 or better wash and dry area | “USP <661> compliant” with no data |
| Change control | 15 | Written notification ≥ 30 days; re-qualification trigger list | Silent glass source switch |
| Traceability | 10 | Lot-to-mold-cavity-to-glass-batch traceability; 12-month retention | Lot number traces to a shift, not a cavity |
| Documentation responsiveness | 10 | Complete pre-audit pack within 10 days | Partial pack, items “available on site” |
Score each category 0 to 5, multiply by weight, divide by 5. A vendor below 70 does not get a trial order. Between 70 and 85, trial order with a tightened incoming inspection. Above 85, standard qualification. The rubric is not the point. The point is that the vendor sees the rubric before the audit and knows which numbers you will ask for.
Field case: the crimp finish that was round on average
Symptom. A 2 mL crimp vial lot leaked intermittently on a GC headspace tray. Roughly 1 in 200 vials showed a pressure drop in the leak test, and the pattern moved between trays.
Measurement. We measured crimp finish OD on 100 vials with a micrometer at two orientations 90° apart. The mean was 11.02 mm, well within the 11.00 ± 0.20 mm spec. But the difference between the two readings (the ovality) ran to 0.18 mm on the failing vials against 0.04 mm on the passing ones. The finish was round on average and oval part by part, so the crimper sealed on the long axis and left a gap on the short one.
Fix. The vendor’s mold had a cooling imbalance on one cavity side. We added an ovality check to incoming inspection: two-orientation measurement on 30 vials per lot, reject above 0.10 mm difference. The vendor reworked the mold cooling. Leak failures dropped to zero over the next four lots.
Field case: the septum that passed the puncture count and failed the reseal
Symptom. An autosampler method with a 26-gauge needle and 150 injections per vial started showing area drift after about 60 injections. The septum passed the vendor’s puncture test, which specified 100 punctures without coring.
Measurement. We ran a reseal test: pierce the septum, hold 10 seconds, withdraw, then pressurize the vial to 1.5 bar and measure leak-down over 60 seconds. At puncture 60 the leak-down rate climbed from 0.02 bar to 0.11 bar. The septum was not coring. It was losing compression set and not closing the needle track. The vendor’s test measured coring only.
Fix. We rewrote the septum specification to require a reseal test at punctures 1, 50, and 100, with leak-down under 0.05 bar at each. The incumbent passed. The alternate supplier’s chlorobutyl compound did not, and we found out before the lot entered the lab rather than after a failed run. If you buy through a distributor, put that requirement in the purchase spec and make the distributor carry it back to the molder; vials with caps assembled by a third party are exactly where septum provenance gets lost.
The audit items that are specific to vials and get skipped
Generic supplier audits check calibration and training records. Those matter, but they do not predict vial failures. These do.
Wash water conductivity and final rinse resistivity
Ask for the last 30 days of final rinse resistivity logs. Type I glass vials for HPLC should be rinsed in water at ≥ 18.2 MΩ·cm. I have seen a vendor rinse in 1 MΩ·cm water from a tired deionizer and call it “DI water.” The residue shows up as a sodium peak in your blank.
Depyrogenation or oven validation if the vial is sold as low-particle
If the vendor claims a low-particle or low-endotoxin vial, ask for the oven or tunnel validation with the 3-log endotoxin reduction data and the temperature profile. A claim without a validation is marketing.
Septum lot traceability to compound batch
Ask the vendor to trace one shipped lot backward to the septum compound batch and the glass batch. If the trace stops at “septum, 1000 pcs, received March,” the vendor is an assembler. Assemblers are fine if they control incoming septum release. They are not fine if they do not.
Packaging and cleanroom classification
Vials packed in a Class 100,000 area pick up particles. Ask for the packing room classification and the double-bagging procedure. For LC-MS work I want the inner bag sealed in a Class 10,000 or better area. This is a one-line question with a binary answer, and it separates the serious molders from the repackagers.
What a good vendor will push back on, and why that is a good sign
The best supplier I audit pushes back on two things every cycle. First, they refuse to guarantee a puncture count without knowing the needle gauge and bevel, because a 22-gauge blunt needle and a 26-gauge beveled needle do not wear a septum the same way. Second, they refuse to sign a leak spec tighter than their reseal test can resolve. A vendor that agrees to every number you put in front of it has not read the numbers. A vendor that pushes back on two of them has a process it understands.
Ask for their internal scrap rate on the molding line. Anything above 3% on a 12 mm body means the process is fighting the tool. Below 1.5% is a healthy line. This number is hard to fake because it has to reconcile with the yield implied by their shipping records, and a vendor that has never calculated it will not have an answer ready.
The pre-qualification sequence, in order
Run the steps in this order. Each one is a gate.
- Document pack requested 10 working days ahead. Score out of 12. Below 8, stop.
- Rubric scored from documents and a 30-minute technical call. Below 70, stop.
- Trial order of 5 lots, each from a different molding run. Test dimensional capability, reseal at punctures 1/50/100, and extractable scan in your solvent.
- Lot-to-lot RSD comparison against your incumbent on a real method, 20 vials per lot. Accept if the new vendor’s RSD is within 0.5 percentage points of the incumbent’s.
- Change control letter signed before the first production order.
Steps 1 and 2 cost a few hours of desk work. Step 3 costs a few hundred dollars and two weeks. Skipping them costs a failed run and a root-cause investigation, which is the expensive part. If you buy vials as part of a broader consumables program, the same rubric applies to the rest of the chromatography products you source, and the document pack can be combined into one request per vendor.
Where the checklist meets the market
Most sample vial vendors on the open market are one of three types: molders who make the glass, assemblers who buy glass and septum and wash and pack, and repackagers who buy finished vials and relabel. The audit checklist above is calibrated for the first two. For a repackager, the only meaningful questions are incoming release testing and traceability, because everything upstream is out of their control. Know which type you are auditing before you send the request, or you will spend a day asking a repackager for control charts they have never seen.
When you compare quotes, normalize on the audit score. A vial at $0.11 from a vendor scoring 60 is more expensive than one at $0.14 from a vendor scoring 88, once you price in the failed run. The sample vial price sheet does not carry that number, which is why procurement keeps buying on unit price and the lab keeps investigating failures. Suppliers that publish real technical data, like the reference material at Aijiren, make the document pack step faster because half the evidence is already on the record. For method-specific container requirements, the EPA SW-846 method text and the technical notes at Restek and Agilent tell you what the vial has to do; the Thermo Fisher chromatography sample handling documentation is a useful cross-check on autosampler compatibility when you set the trial order.
The decision rule is this: do not qualify a sample vial vendor on a CoA and a price. Qualify on a scored document pack, a 5-lot trial with a reseal test at punctures 1, 50, and 100, and a lot-to-lot RSD within 0.5 percentage points of your incumbent. If a vendor will not supply the evidence for those three gates, the audit is over, and you have saved yourself the site visit.
References & further reading
Related resources from our lab consumables network:
- Sample Vial at samplevialprice.com
- Chromatography Products at chromatographyselling.com
- Aijiren at aijirentech.com
- Vials With Caps at vialswithcaps.com
External references and standards: