The first number on the sheet is usually a dimension, and it is rarely the

A lab manager I worked with once signed off on a 10,000-piece septa order because the datasheet said “recommended for GC headspace, 200°C.” Three weeks later, every run above 160°C showed ghost peaks. The batch was unusable, the procurement file was closed, and the datasheet was technically not wrong. It just never said what “recommended” was tested against, or whether 200°C was a measured value or a marketing number. If you buy septa by reading only the bolded claims, you are buying someone else’s optimism. This guide walks through a vial septa datasheet line by line, separating what is actually tested from what is nominal, and how to spot a sheet that hides its method.

The first number on the sheet is usually a dimension, and it is rarely the one that matters

Most septa datasheets open with diameter and thickness. A typical entry reads “11 mm septa, 1.3 mm thick, PTFE/silicone.” Those are real numbers, but they are nominal dimensions, not tested specifications. A molded septum comes out of a cavity that has a tolerance. The datasheet rarely tells you what that tolerance is. I have measured septa from three different suppliers all listed as “11 mm” and found diameters ranging from 10.82 mm to 11.14 mm. All of them fit an 11 mm crimp cap, but the 10.82 mm septum seated differently under compression and leaked after three injections on an autosampler vial run at 40 psi.

The dimension line that matters more is thickness, specifically the PTFE layer thickness when you are buying laminated septa. A sheet that says “PTFE/silicone” without a layer thickness is giving you a material stack, not a specification. I have seen PTFE layers range from 0.05 mm to 0.25 mm on septa sold as equivalent. The 0.05 mm layer punctures cleanly but offers almost no chemical barrier. The 0.25 mm layer resists solvents better but cores more easily with a blunt needle. When I specify septa for HPLC work, I want the PTFE layer called out as a tested value, typically 0.10 mm to 0.13 mm for a balance between barrier performance and puncture behavior. If the sheet does not list it, I ask for the incoming QC record. If they cannot produce it, I move on.

Durometer is a tested number, but only if the method is stated

Durometer, usually Shore A, tells you how hard the elastomer layer is. A typical silicone septum runs Shore A 45 to 55. A butyl septum runs Shore A 50 to 60. These numbers are measured with a durometer gauge per ASTM D2240 or ISO 7619-1. The problem is that durometer readings depend on sample thickness, temperature, and the operator’s timing. A 1.0 mm thick sample stacked to 6 mm reads differently than a solid 6 mm puck. A sheet that prints “Shore A 50” without the test method and sample prep is giving you a nominal value dressed as a measured one.

I once rejected a lot of silicone septa that measured Shore A 62 on incoming inspection. The datasheet said 50. The supplier’s response was that their reading was taken on a 6 mm compression-molded slab, not on the finished 1.3 mm septum. That is a legitimate difference, but it means the datasheet value was not a specification for the part I was buying. It was a property of their raw material. For a vial septa buyer, the durometer of the finished septum is what predicts reseal performance and needle force. A septum that is 10 points harder will require noticeably more needle force and may not reseal after multiple punctures.

Puncture count and resealability: the line that separates tested from invented

A datasheet line like “resealable after multiple injections” is not a specification. It is a hope. A real puncture specification gives you three things: the needle gauge, the number of punctures, and the pass criterion. For example: “reseals after 5 punctures with a 23-gauge needle, no visible leak at 15 psi.” That is a test. I have seen sheets that say “multi-puncture” with no needle size, no count, no pressure. That sheet is hiding its method because it does not have one.

The puncture test itself is straightforward. A septum is mounted in a cap on a vial, pressurized to a set value, punctured with a specified needle, the needle is withdrawn, and the vial is checked for pressure retention or leak rate. The pass criterion is usually a pressure drop threshold over a set time. For GC headspace work, I want a septum that holds 20 psi for 10 minutes after 3 punctures with a 23-gauge needle with less than 10% pressure loss. For HPLC autosampler use, the criterion is different: no visible coring particles and no leak at 5 punctures with a 22-gauge needle at 100 bar vial pressure. If the sheet does not state the gauge and the count, the puncture claim is nominal.

The needle gauge matters more than most buyers realize. A 26-gauge needle punctures a silicone septum with far less tearing than a 22-gauge needle. A septum that easily reseals after 10 punctures with a 26-gauge needle may fail after 2 punctures with a 22-gauge needle. When you compare two datasheets, you cannot compare puncture counts unless the needle gauge is identical. I have seen a supplier claim “10 punctures” while their competitor claimed “3 punctures,” and the 3-puncture septum was actually more robust because it was tested with a larger needle.

Temperature ratings: the most abused line on the sheet

Every septum datasheet has a temperature line. “Maximum temperature: 200°C” or “operating range: -40°C to 250°C.” What almost none of them tell you is whether that number is a continuous-use rating, a short-exposure rating, or a thermogravimetric analysis (TGA) onset temperature. A TGA onset temperature is the point where the material starts losing mass under a controlled heating ramp. It is not an operating temperature. A silicone septum with a TGA onset of 300°C will still outgas volatile siloxanes at 200°C over a 30-minute headspace equilibration. Those siloxanes become ghost peaks in your chromatogram.

For GC work, the relevant number is not the decomposition temperature. It is the septum bleed temperature, measured by conditioning a septum in a GC inlet at a set temperature and measuring the baseline rise or the appearance of specific ions. Restek and Agilent publish bleed data for their high-temperature septa, and that data is the result of an actual test: install septum, ramp inlet, monitor baseline. A sheet that says “low bleed” without a temperature, a time, and a detection method is not giving you a specification. It is giving you an adjective. I have had “low bleed” septa produce baseline humps at 180°C that a properly specified septum did not produce until 250°C. The difference was not the material. It was that one supplier tested and one supplier guessed.

For headspace and SPME work, the temperature rating also interacts with the sample matrix. A septum rated to 200°C in dry air may fail at 180°C in a vial headspace saturated with DMSO or DMF vapor. The solvent attacks the elastomer, softening it and accelerating outgassing. A datasheet temperature rating is almost always measured in clean air, not in your sample matrix. If your matrix is aggressive, you need to derate. I typically derate a septum’s stated maximum by 20°C to 30°C for polar aprotic solvents and by 10°C to 15°C for chlorinated solvents.

Extractables and leachables: the line that costs real money to verify

Some septa datasheets include a line about extractables, often phrased as “low extractables” or “meets USP <661> requirements.” USP <661> is a real standard with defined extraction conditions: the septum is extracted in a set of solvents at a set temperature for a set time, and the extract is analyzed for non-volatile residue, heavy metals, and UV-absorbing substances. A septum that passes USP <661> has been tested. A septum that says “USP <661> compliant” may have been tested, or the supplier may be asserting compliance based on material composition. The difference matters.

I have seen a supplier change their silicone cure system, and the new formulation passed USP <661> for non-volatile residue but produced a new extractable peak in an HPLC gradient run. The USP test did not catch it because the test conditions did not match the analytical method. USP <661> is a minimum bar, not a guarantee of chromatographic cleanliness. If your method is sensitive to specific extractables, the only way to know is to run a septum extraction in your mobile phase and look at the chromatogram. Some suppliers will provide extractables data on request, typically a GC-MS or LC-MS scan of a solvent extract. That data is tested, not nominal, and it is worth requesting before you commit to a large order.

For regulated work, the relevant standards are USP <661.1> and <661.2>, which replaced the older <661> monograph. The newer chapters require more detailed extractables profiling. If a datasheet references USP <661> without the decimal, it may be referencing an outdated test. Ask which chapter and which version. The Thermo Fisher chromatography technical library has useful reference material on extractables testing for sample handling components, and it is worth reading before you accept a “compliant” claim at face value.

Hardness and compression set: two numbers that predict reseal failure

Compression set is the permanent deformation a septum retains after being compressed for a set time at a set temperature. It is expressed as a percentage. A silicone septum with a compression set of 15% at 70°C for 22 hours will recover its shape better than one with a 30% compression set under the same conditions. Compression set is a tested value per ASTM D395 or ISO 815. The problem is that many datasheets omit it entirely, even though it is the single best predictor of whether a septum will reseal after the cap is crimped and the vial sits on an autosampler tray for 48 hours.

I ran a stability study where vials were crimped and stored at 40°C for two weeks before injection. The septa from one supplier showed visible permanent indentation from the crimp, and 12% of the vials leaked during autosampler piercing. The datasheet had no compression set value. The supplier’s technical team eventually provided a compression set of 38% at 70°C. The replacement septum, with a compression set of 18%, showed no failures in the same study. The difference was not visible on the original datasheet. It was hidden in a number that was not printed.

Hardness and compression set work together. A harder septum resists compression set better but requires more needle force and may core more easily. A softer septum seals with less crimp force but takes a set more readily. For a HPLC vial septum that must survive multiple piercings over a 100-injection sequence, I look for a Shore A of 45 to 50 and a compression set below 20% at 70°C. Both numbers need to be tested, not nominal.

How to spot a sheet that hides its test method

After 15 years of reading these sheets, I have a short list of red flags. A datasheet that uses the word “typical” for a critical property is telling you the number is not guaranteed. “Typical” means the supplier measured it once, on one lot, under one set of conditions. It is a data point, not a specification. A sheet that says “low” or “excellent” without a number and a method is giving you an opinion. A sheet that lists a test standard but not the test conditions is giving you a partial specification. ASTM D2240 without the sample thickness and conditioning time is not a durometer specification. It is a reference to a standard that has options.

The most reliable sheets I have seen list, for each critical property: the value, the test method, the test conditions, and the acceptance criterion. For example: “Hardness: Shore A 50 ±5, ASTM D2240, 6 mm sample, 23°C, 3 s reading.” That is a specification. Anything less is a conversation starter. When I evaluate a new supplier, I ask for their certificate of analysis for the last three lots of the septum I am buying. A supplier that can produce lot-specific test data for hardness, thickness, and PTFE layer thickness is one that actually controls their process. A supplier that sends me the same datasheet I already have is one that is reselling someone else’s product.

Comparison table: what a real datasheet tells you versus what a nominal sheet tells you

Field case: the headspace run that failed at 160°C

A contract lab running residual solvents by headspace GC switched septa suppliers to save 18% on unit cost. The new datasheet said “200°C maximum, low bleed, PTFE/silicone.” The method equilibrated vials at 160°C for 20 minutes. Within the first sequence, the analyst saw a broad hump eluting between 3 and 6 minutes that was not present in the blank runs using the previous septum. The hump grew over the sequence. The lab pulled the septum and ran a bleed test: install septum, ramp inlet to 160°C, hold 30 minutes, monitor baseline. The new septum produced a baseline rise of 4.2 pA. The previous septum produced 0.8 pA under identical conditions. The datasheet’s “low bleed” claim was not tied to a temperature or a detection method. The fix was to return to the previous septum and add a bleed specification to the purchasing document: baseline rise less than 1 pA at 160°C for 30 minutes by GC-FID. The new supplier could not meet it. The 18% savings disappeared the first time a client questioned a residual solvent result.

Field case: the crimp that looked fine and leaked at hour 48

An environmental lab running semi-volatiles by GC-MS per U.S. EPA SW-846 methods noticed a sudden increase in internal standard response variability. The vials were crimped, stored at 4°C for 48 hours, then loaded onto the autosampler. The variability started after a septum lot change. The lab measured crimp height and found it consistent. The problem was not the crimp. It was the septum compression set. The new lot had a compression set of 34% at 70°C, measured after the fact. The previous lot had been 19%. After 48 hours under crimp compression, the new septa had permanently deformed enough that the needle did not seal during piercing, allowing a small pressure leak that changed the injection volume. The fix was to add a compression set specification to the purchase order and to require lot-specific test data. The lab also shortened its maximum storage time for crimped vials from 72 hours to 48 hours as a precaution. The variability disappeared.

Troubleshooting checklist: when the datasheet and the bench disagree

• Ghost peaks or baseline rise above 150°C: measure septum bleed at your method temperature with your detector. If the sheet does not state a bleed temperature, the sheet is not the spec. The measurement is.
• Leaks after multiple injections: check the puncture specification. If the sheet does not state needle gauge and count, run your own reseal test at your operating pressure. A 23-gauge needle is not a 22-gauge needle.
• Coring or particles in the vial: measure the PTFE layer thickness and the needle bevel condition. A PTFE layer over 0.15 mm with a blunt needle is a coring combination. The sheet may not warn you about this.
• Variability after storage: request compression set data. If the supplier cannot provide it, run a storage test: crimp 20 vials, store at your temperature for your maximum time, then check for leaks and reseal.
• Extractables in the chromatogram: run a septum extraction in your mobile phase or solvent. The USP test is a minimum bar, not a method-specific guarantee. Your method is the final arbiter.
• Dimensional mismatch: measure the actual diameter and thickness of the lot you received. A nominal 11 mm septum that measures 10.85 mm will seat differently under crimp. The tolerance is the specification, not the printed number.

What to write in the purchase order so the datasheet stops lying to you

The fix for a misleading datasheet is not to read it more carefully. It is to write your own specification. When I buy septa for a regulated method, the purchase order includes the following lines, and the supplier must confirm each one in writing:

• Diameter and thickness with tolerance, measured on finished part
• PTFE layer thickness with tolerance, measured by cross-section microscopy
• Shore A hardness with ASTM D2240 method and sample prep stated
• Puncture reseal: needle gauge, puncture count, pressure, and pass criterion
• Bleed temperature: temperature, time, detector, and maximum baseline rise
• Compression set: ASTM D395 or ISO 815, temperature, time, maximum value
• Extractables: USP <661.1> or <661.2> tested, extractables profile available
• Certificate of analysis for each lot, with actual measured values

This is not overkill. It is the difference between buying a component and buying a promise. A vial septum is a $0.10 part that can invalidate a $500 analytical run. The datasheet is the only document between you and that failure. If the sheet does not tell you what was tested, under what conditions, and what the pass criterion was, it is not a datasheet. It is a brochure.

For GC method development and inlet consumables, the Agilent GC columns and accessories resource includes application notes on septum selection and bleed testing. For troubleshooting guides and reference chromatograms, the Restek technical library is the most practical free resource I have used. Both are worth consulting before you accept a supplier’s temperature or bleed claim.

The decision rule is simple: if a critical property on the sheet lacks a number, a method, and an acceptance criterion, treat it as absent. Buy on tested values, not on adjectives. The septum that costs $0.08 more and comes with a real certificate of analysis is cheaper than the one that costs $0.08 less and costs you a sequence, a client, or a batch.

References & further reading

Related resources from our lab consumables network:

• Vial Septa at vialsepta.com
• Hplc Vial at hplcgc.com
• Autosampler Vial at autosamplerhplc.com
• Vial at vials-filters.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