At 14:42 on a Tuesday, an auditor asked my client for the lot disposition record on a box of 200 pre-slit PTFE/silicone septa. The lab manager produced a certificate of analysis, a purchase order, and an email from the vendor. None of it showed whether that specific lot had been accepted, rejected, or quarantined after incoming inspection. The audit finding took 20 minutes to write and six weeks to close. If you are preparing for hplc septa audit documentation and record retention, that is the gap I see most often: the lab has the vendor’s paper, but not its own decision record.

Auditors do not want a binder full of marketing datasheets. They want a closed loop from receipt to retirement. For HPLC septa, that loop has seven records. Miss one and the finding writes itself.

  • Incoming inspection log with lot number, date received, quantity, and the dimensional checks performed. If you run a Cpk on septum diameter, the raw measurement sheet belongs here.
  • Acceptance/rejection decision signed by the person who released the lot. This is the record my client was missing. A CoA from the vendor is not a release decision.
  • Specification sheet in force at the time of receipt. If you changed the hardness spec from Shore A 45 to Shore A 50 in March, the auditor needs to know which revision applied to the lot you received in February.
  • Usage log or batch trace linking septum lots to specific methods, instruments, or stability studies. This is what turns a consumable into a controlled material.
  • Failure or deviation records for any lot that leaked, cored, or produced carryover above the method limit.
  • Retention schedule showing how long each record type is kept and where.
  • Destruction or disposal record for expired or rejected lots.

That is the full set. An inspector from a notified body or a client auditing to ISO 17025 will sample maybe three of these. The problem is you do not know which three.

Retention periods: match the record to the rule, not the calendar

I have seen labs default to “keep everything seven years” and I have seen labs keep nothing past the current fiscal year. Both are wrong for different reasons. The correct retention period for HPLC septa records depends on what the septum was used for, not on the septum itself.

For routine analytical work

If your septa support release testing on a commercial product, the usage records and lot acceptance records should be retained for the shelf life of the product plus one year, or the period required by your quality agreement, whichever is longer. For a pharmaceutical product with a 36-month shelf life, that means at least 49 months from the date the septum lot was consumed. I have seen auditors accept less for non-GMP work, but never for anything that touches a batch record.

For stability studies

Septum lots used in stability-indicating methods follow the retention period of the study data. If a 24-month stability protocol produces data you must keep for 10 years under your regulatory filing, the septum records ride along. You cannot reconstruct a failed stability result without knowing which septum lot was on the autosampler vial.

For environmental methods

Labs running U.S. EPA SW-846 test methods under RCRA or CERCLA often face retention requirements tied to project records. A septum lot used on a groundwater monitoring project may need to be traceable for the life of the permit, which can outlast the septum’s own shelf life by a decade. Keep the incoming inspection record and the lot acceptance decision for the full project retention period, not just the vendor’s stated shelf life.

For method development and validation

This is where labs cut corners. A septum lot used during method validation becomes part of the validation package. If the method is later transferred to another site or audited by a client, the septum lot number, its dimensional data, and its material composition must be retrievable. I recommend retaining validation-related septum records for the life of the method plus five years. A method that runs for 12 years means 17 years of septum paperwork. That sounds excessive until you have to defend a method against a failing system suitability test and the auditor asks what changed.

The three gaps that show up most often

Every audit I have supported has surfaced at least one of these. The first one I already described: no internal release decision. The vendor’s CoA is a starting point, not an endpoint. The second and third are less obvious but just as common.

Gap 1: no dimensional verification record from incoming inspection

Most labs buy septa to a drawing or a GPI finish specification. The vendor ships a CoA that says the septum diameter is 9.00 mm ± 0.05 mm. The lab files the CoA and moves on. Six months later, an auditor asks how the lab verified that the lot actually met the diameter spec. The answer is usually silence.

You do not need a full metrology lab. A calibrated caliper and a sample of 30 septa from each lot, measured at three points around the circumference, will do. Record the mean, the range, and the individual values. If your acceptance limit is 9.00 mm ± 0.05 mm and your measured mean is 9.03 mm with a range of 0.04 mm, you have a defensible release decision. If you skip this, the auditor will assume you released on faith.

Gap 2: no link between septum lot and method performance data

A lab changes septum vendors, or the vendor changes a material lot, and nobody writes it down. Three weeks later, carryover on a trace-level assay drifts from 0.02% to 0.08%. The lab investigates the column, the mobile phase, the needle wash, and the detector. Nobody checks the septum because the septum is “just a consumable.”

The fix is a simple lot linkage. When you open a new box of septa, record the lot number in the instrument logbook or the run sequence file. Most CDS platforms have a field for vial and cap lot. Use it. Then when carryover shifts, you can pull the last five septum lots and look for a correlation. I have seen this single change cut investigation time from days to hours.

Gap 3: no destruction record for rejected or expired lots

A lab rejects a lot of septa because the Shore A hardness came in at 58 when the spec was 45 to 55. The box sits on a shelf for eight months. An auditor asks what happened to the rejected lot. The answer is “it is over there.” That is a finding.

Rejected lots need a disposition record: quarantine, return to vendor, or destruction. If you destroy them, document the date, the quantity, the method of destruction, and the person who did it. If you return them to the vendor, keep the RMA and the shipping record. The point is to show that rejected material cannot find its way back into use. A box on a shelf is not a disposition.

What a defensible septum record actually contains

I have reviewed hundreds of septum records from labs across pharma, environmental, and food testing. The ones that survive an audit share the same field set. The ones that fail are missing half of it.

Field What to record Why the auditor cares
Lot number Vendor lot and internal receiving lot Links the physical box to every downstream record
Date received Actual receipt date, not invoice date Establishes the start of the retention clock
Quantity Units received, units inspected, units released Shows sampling was proportional to lot size
Specification revision Revision number and date of the drawing or spec Proves you inspected against the correct version
Dimensional results Mean, range, and n for diameter and thickness Demonstrates the lot met the physical spec
Hardness results Shore A values, number of readings, durometer used Confirms the septum will seal under the cap torque you use
Release decision Accept, reject, or quarantine with signature and date This is the record most labs are missing
Expiry or retest date Vendor shelf life or internal retest interval Prevents use of degraded material
Disposition Consumed, returned, destroyed, or expired Closes the loop on every unit

If your current incoming inspection form has fewer than six of these fields, you have a gap. I would start by adding the release decision and the disposition. Those two are the difference between a record that survives an audit and a record that generates a finding.

Cap torque is a documentation problem, not just a hardware problem

I have written before about how seal loss happens. What I have not covered is how cap torque creates a documentation trail that auditors will follow. If your method specifies a cap torque of 18 N·cm for a 9 mm vial, the auditor wants to see that the septum you used can actually seal at that torque. That means the septum hardness and thickness must be on file.

A PTFE/silicone septum with a Shore A hardness of 50 will compress enough at 18 N·cm to seal a 9 mm opening. The same septum at Shore A 58 may not compress fully, leaving a microchannel for solvent vapor to escape. If your incoming inspection only checked diameter and skipped hardness, you have no way to defend the seal. The auditor will ask why you accepted a lot that was outside the hardness spec. The answer cannot be “we did not check.”

Record the torque value from your method in the septum specification. If you run multiple methods with different torques, the septum lot must meet the most demanding one. A septum that seals at 12 N·cm may not seal at 20 N·cm. Write the torque requirement into the spec sheet and verify it during incoming inspection with a calibrated torque driver.

Two field cases that fit this topic exactly

Case 1: the missing release decision that cost a contract lab its ISO 17025 scope

A contract lab running HPLC assays for a generic drug manufacturer was audited by the client. The auditor pulled the incoming inspection record for a septum lot used on a stability study. The lab had a CoA from the vendor showing the diameter and hardness met spec. What they did not have was any record of who reviewed that CoA, when, and what decision they made.

The auditor wrote a major nonconformity for lack of documented acceptance criteria and release. The lab argued that the CoA was the release record. The auditor disagreed: a vendor document is not a lab decision. The lab had to add a release signature block to its incoming inspection form, retrain all receiving staff, and re-audit every open lot. The finding delayed the client’s product filing by two months.

Case 2: the lot linkage that solved a carryover investigation in one afternoon

A food testing lab saw carryover on a pesticide residue method jump from 0.01% to 0.07% over a two-week period. The lab spent four days troubleshooting the column, the needle wash, and the detector. Nothing changed. Then a senior chemist noticed that the lab had switched to a new box of septa from the same vendor around the time the carryover started.

The lab had no record linking the septum lot to the runs. The chemist pulled the vendor’s lot history and found that the new lot was made from a different silicone batch. The septum was slightly softer, Shore A 42 instead of 48, and it was shedding more material under repeated needle punctures. The lab switched back to the previous lot and carryover returned to 0.01% within a day. The fix took 20 minutes once the lot was identified. The investigation took four days because the lot was not recorded.

That lab now records the septum lot number in every run sequence. The next time carryover shifts, they can pull the lot history in five minutes. That is the value of hplc septa audit documentation and record retention: it is not about passing an audit. It is about being able to answer the question “what changed” without a four-day investigation.

Retention schedules: write them down or they do not exist

An auditor will not accept an unwritten retention policy. If you tell an auditor you keep septum records for “as long as needed,” you have just admitted you have no retention schedule. The policy must be a controlled document with a revision number, an owner, and a defined retention period for each record type.

Here is a working retention schedule I have used with clients. Adjust the periods to your regulatory environment, but write them down.

  • Incoming inspection records: retain for the life of the septum lot plus 2 years, or the product shelf life plus 1 year, whichever is longer.
  • Lot acceptance decisions: same as incoming inspection records.
  • Usage logs linking septum lots to methods: retain for the life of the method plus 5 years, or the regulatory filing retention period, whichever is longer.
  • Failure and deviation records: retain permanently or for the life of the facility. These are the records that explain a bad result years later.
  • Destruction and disposal records: retain for 5 years after disposal, or the period required by your waste permit, whichever is longer.
  • Vendor CoAs and specification sheets: retain for the life of the septum lot plus 2 years. These are reference documents, not decision records.

The key distinction is between reference documents and decision records. A vendor CoA is a reference document. Your release decision is a decision record. Auditors weight decision records far more heavily because they show the lab exercised control. A file drawer full of vendor CoAs with no internal release decisions is a file drawer full of evidence that you delegated quality control to your supplier.

What to do with the records you already have

If you are reading this and realizing your current septum records are thin, do not panic. Start with a gap assessment. Pull the last five septum lots you received and ask three questions for each:

  1. Do I have a documented release decision for this lot?
  2. Can I link this lot to the methods or instruments where it was used?
  3. Do I have a disposition record showing what happened to any unused or rejected units?

For any lot where the answer is no, you cannot retroactively create the missing record. You can, however, write a retrospective note explaining what you know and when you learned it. That is not as strong as a contemporaneous record, but it is better than silence. The auditor will give you partial credit for acknowledging the gap and documenting a corrective action.

Then fix the process going forward. Update your incoming inspection form to include the release decision and disposition fields. Add a septum lot field to your run sequence template. Write the retention schedule into your quality manual. These are not expensive changes. They are mostly paperwork discipline.

Where the documentation burden actually falls

I have worked with labs that treat septa as office supplies. The procurement team buys them, the receiving clerk signs for the box, and the chemists open it when they need a new one. Nobody records the lot number, nobody checks the dimensions, and nobody files a release decision. That works until the first audit.

The burden falls on the lab manager or quality lead, not on procurement. Procurement can require a CoA from the vendor. That is a start. But the release decision, the lot linkage, and the retention schedule are lab functions. If you are the lab manager, own them. If you are the quality lead, audit them. If you are the chemist, record the lot number every time you open a new box. It takes 10 seconds and it saves days later.

For labs that buy septa through Hplc Septa, the vendor documentation is usually solid. The gap is almost always on the lab side. The same is true for labs that source Vial Closures or run methods on Hplc Vial systems. The vendor can give you a CoA. Only you can give yourself a release decision.

If your lab does any Analytical Testing under a quality system, the septum records are part of that system. An auditor will not treat them as optional. The question is whether you treat them as optional until the finding arrives.

Standards that reference septum documentation

You do not need to invent your own documentation framework. Several standards and guidance documents already point to what is expected.

ISO 17025:2017, clause 7.4, covers externally provided products and services. It requires the lab to ensure that externally provided products do not adversely affect the quality of the lab’s results. For septa, that means you need a specification, a verification process, and records of the verification. A CoA alone does not satisfy the requirement.

USP <621> on chromatography does not specifically address septa, but USP <1058> on analytical instrument qualification implies that consumables affecting method performance must be controlled. If your septum lot changes the seal and the seal changes the injection precision, the septum is a controlled consumable.

For environmental labs, U.S. EPA SW-846 methods require documentation of all materials that could affect data quality. Method 8000D, the determinative chromatographic separations method, is explicit about documenting the condition of consumables. An auditor working to SW-846 will expect to see septum lot numbers in the raw data package.

Vendor resources like Agilent chromatography consumables and Thermo Fisher chromatography and sample handling provide CoAs and lot traceability data. Restek technical resources publish guidance on septum selection and handling. Use those vendor documents as inputs. Do not mistake them for your own release decision.

The decision rule that closes the loop

Every septum lot that enters your lab gets one of three dispositions: accepted, rejected, or quarantined. The disposition is recorded, signed, dated, and linked to the lot number. The lot number is linked to every method or instrument where the septa were used. The records are retained for the life of the method or the product, whichever is longer. Rejected lots are destroyed or returned with a paper trail. If you can show an auditor that loop for the last 12 months, you will not get a septum finding. If you cannot, the finding is already written. The only question is when it arrives.

References & further reading

External references and standards: