What coring actually is, and why it is not the same as wear--Aijiren HPLC Septa

Coring is the failure I get called about most often, and it is almost always misdiagnosed. A lab reports “ghost peaks” or “carryover that won’t wash out,” replaces the column, replaces the mobile phase, replaces the autosampler needle, and three weeks later finds a neat little plug of silicone sitting in the bottom of a vial or wedged in the injection port. The septum was the problem the whole time. This article is about how coring happens, how to measure it, and how to stop it, with the numbers I actually use on the bench.

Coring is a mechanical failure where the injection needle punches a disc of septum material out of the sheet instead of parting it. The disc, usually 0.5 to 1.5 mm across and 0.5 to 3 mm thick depending on the septum, ends up in one of three places: in the sample vial, in the needle bore, or sitting on the inlet liner. Wear, by contrast, is the gradual enlargement of the puncture hole as the same needle goes in and out hundreds of times. Wear produces leaks. Coring produces contamination and, occasionally, a blocked needle. They have different causes and different fixes, and confusing them wastes a lot of time.

The key distinction is that coring is a shear event. The needle edge cuts through the material rather than displacing it. That happens when the needle is dull, the septum is too hard or too thin, the needle diameter is too large for the septum’s mechanical support, or the needle is misaligned so it enters at an angle. Any one of those will do it. Two together will do it on the first injection.

Why silicone and silicone/PTFE septa are the usual suspects

Most HPLC septa are a silicone rubber disc laminated to a PTFE film on the sample side. The silicone provides the reseal force. The PTFE provides chemical inertness so the sample never touches silicone. That construction is excellent for chemical resistance and terrible for shear resistance if the PTFE is thin or the silicone is under-cured. A 50 µm PTFE film will coring-punch cleanly at the first injection with a dull needle. A 125 µm PTFE film will resist it. I have measured both. The difference in puncture count before first core is roughly tenfold.

Pure silicone septa (no PTFE) core less often but extract more. You are trading contamination risk for particle risk. If your method is LC-MS at low ng/mL, you want the PTFE. If your method is a rugged UV assay at µg/mL and your needles are old, silicone may be the pragmatic choice. There is no universal answer, which is why the septum spec sheet matters more than the marketing.

The four mechanical causes, ranked by how often I see them

1. Dull or burred needles (the number one cause)

A properly bevelled autosampler needle has a sharp, polished cutting edge. After 2,000 to 5,000 injections the bevel starts to burr, and a burr is a coring tool. It does not cut cleanly. It tears. I have pulled needles off instruments where the bevel looked fine under a loupe and terrible under a 10x stereo microscope. Buy the microscope. It costs less than one wasted week of troubleshooting.

The practical rule I give labs: replace the needle at the same interval as the septum, or at most every second septum change. If you change septa every 500 injections and never change the needle, you are coring by design and blaming the septum.

2. Needle diameter versus septum support

A 0.5 mm needle through a 1.0 mm thick septum with a proper crimp cap is fine. A 0.8 mm needle through the same septum, especially with a loosely crimped cap, will core. The cap matters because the aluminium crimp compresses the septum against the vial neck and provides radial support. A cap crimped to the wrong torque, or a snap cap that does not compress, lets the septum flex and the needle punches through unsupported material.

For 11 mm crimp caps on 2 mL vials, I target a crimp that just dimples the cap skirt without crushing the septum. If you can spin the cap by hand, it is loose. If the septum bulges into the vial, it is over-crimped. Both cause coring. The vial closures you use matter as much as the septum itself, and I would not buy them from different suppliers without checking the fit.

3. Septum hardness and thickness

Shore A hardness for HPLC septa typically runs 45 to 60. Softer septa reseal better and core less but deform more under autosampler pressure and can be pushed into the vial. Harder septa resist deformation but core more readily. Thickness runs 1.0 to 3.0 mm. A 1.0 mm septum in a 2 mL vial is standard. A 3.0 mm septum in the same vial reduces headspace and can cause the needle to bottom out on the septum rather than the vial, which is its own coring mechanism.

If you are running a short injection cycle with a fast autosampler and a 0.8 mm needle, use a 1.5 mm septum with a 125 µm PTFE film and a Shore A around 50. That combination has given me the best puncture-count-to-contamination ratio on standard 2 mL vials.

4. Misalignment and injection depth

Needle-to-septum alignment should be within 0.1 mm laterally. Anything more and the needle enters at an angle, which turns a clean puncture into a tearing cut. This is a service issue, not a consumable issue, but it shows up as a consumable complaint. If the same septum cores in the same position on the same tray, check alignment before you buy different septa. The Agilent consumables documentation has alignment tolerances for their autosamplers, and Thermo publishes similar figures for their sample handling systems. Read them before you blame the septum.

How coring shows up in the data (and how to confirm it)

Coring rarely announces itself. It presents as one of these:

Ghost peaks at a consistent retention time, often late in the run, that do not match any standard.
Carryover that will not wash out even with a strong needle wash, because the plug is in the needle bore, not on the needle exterior.
Rising baseline noise as silicone particles accumulate on the column inlet frit.
Pressure spikes when a particle lodges in the inlet frit or the needle.
Injection-to-injection RSD% climbing from a normal 0.5% to 3 to 8% without any change in method.

The confirmation test is simple. Take the septum out of a used vial, examine it under a 10x to 20x stereo microscope, and look for a clean round hole versus a torn, ragged opening with missing material. A cored septum will show a punched disc missing from the PTFE face. Photograph it. Keep the photo in your method file. It is the fastest way to settle the argument about whether the septum or the needle is at fault.

If you cannot see the septum, look at the vial bottom. A silicone disc 0.5 to 1.5 mm across is visible to the naked eye against a white background. Shine a torch through the vial. If you see a disc, you have coring. If you do not, you have wear or a leak, and you should look elsewhere.

Puncture counts: what the numbers actually mean

Manufacturers quote puncture counts, and those numbers are useful only if you know the test conditions. A septum rated for 1,000 punctures was tested with a specific needle diameter, at a specific injection depth, at a specific speed, on a specific cap. Change any of those and the number is meaningless. I have seen septa rated at 1,000 puncture core at 150 because the lab used a 0.8 mm needle instead of the 0.5 mm the rating assumed.

Here is the comparison I use when selecting septa for a new method. These are working numbers from my own bench, not marketing figures.

Septum selection checklist with working numbers

Standard 2 mL HPLC, 0.5 mm needle, crimp cap: 1.0 mm silicone/PTFE, 125 µm PTFE film, Shore A 50. Expect 500 to 800 punctures before first core. Change at 500.
Standard 2 mL HPLC, 0.8 mm needle, crimp cap: 1.5 mm silicone/PTFE, 125 µm PTFE film, Shore A 45. Expect 200 to 400 punctures. Change at 200. Yes, that is a lot of septa. It is cheaper than a new column.
LC-MS at low ng/mL, 0.5 mm needle: 1.0 mm silicone/PTFE, 125 µm PTFE, pre-cleaned and lot-tested for extractables. Expect 300 to 500 punctures, but change at 300 because the contamination risk dominates the puncture count.
Headspace or GC, 0.5 mm needle, high temperature: 3.0 mm silicone/PTFE, temperature rated to 250 °C continuous. Expect 100 to 200 punctures at 200 °C. Change at 100. Heat softens the silicone and coring accelerates.
Snap-cap vials, 0.5 mm needle: 1.0 mm silicone/PTFE, but check the cap compression. Snap caps that do not compress the septum will core at 50 to 100 punctures regardless of septum quality.

Notice that the puncture count is a function of the whole system, not the septum alone. That is the point. If you buy a septum rated at 1,000 and get 150, the septum is not lying. Your system is asking more of it than the rating assumed.

Temperature and coring: the interaction most labs miss

Heat softens silicone. A septum at 40 °C in a thermostatted autosampler tray cores at roughly the same rate as at ambient. At 60 °C the coring rate roughly doubles. At 80 °C it triples. If you are running a heated autosampler for solubility reasons, your puncture count is not the ambient number. Halve it.

For GC inlet septa the number is worse. A septum at 250 °C inlet temperature is soft, and the needle is hot. Coring at the inlet is common and shows up as a bleed peak or a ghost peak from the previous injection. The Restek technical library has good data on inlet septum bleed versus temperature, and it is worth reading before you pick a septum for a high-temperature method.

The fix is either a higher-temperature-rated septum (usually a different silicone formulation, sometimes with a PTFE film on both sides) or a lower inlet temperature. If the method allows the lower temperature, take it. If not, budget for more frequent septum changes and log them.

Needle damage: the other half of the problem

Coring damages the needle as well as the septum. A cored septum leaves a burr on the needle bevel, and that burr cores the next septum faster. It is a feedback loop. This is why a lab that changes septa but not needles sees accelerating coring. The first septum cores because the needle is slightly dull. The coring burrs the needle. The next septum cores faster. Within a month the lab is changing septa every 50 injections and cannot work out why.

Break the loop by changing the needle whenever you see the first cored septum. Not the next time. The same time. The needle is usually cheaper than the septum in volume terms, and it is certainly cheaper than the column you will replace if you do not.

How to inspect a needle for coring damage

Remove the needle from the autosampler and inspect the bevel under a 10x stereo microscope.
Look for a rolled edge, a burr on the long bevel, or a flattened tip. Any of those is a coring tool.
Check the bore for a silicone plug by flushing with methanol and watching the flow. A restricted flow means a plug.
Check the needle seat and the ferrule for wear. A loose needle moves laterally and cores by misalignment.
Log the injection count. If you cannot tell me how many injections the needle has done, you cannot diagnose coring.

The same inspection applies to the sample needle in the autosampler and the injection needle in the manual injector valve. Manual injectors core less often because the operator feels the resistance, but they do core, especially with a Rheodyne-style valve where the needle guide wears.

A field checklist for coring and needle damage

This is the sequence I run when a lab calls me about ghost peaks or carryover. It takes about two hours and it finds the problem nine times out of ten.

Step 1. Confirm the symptom. Run a blank injection with a new septum and a new needle. If the ghost peak disappears, you have a septum or needle problem. If it persists, you have a column or mobile phase problem and you should stop here.
Step 2. Inspect the used septum. 10x microscope. Look for a punched disc, a torn hole, or missing PTFE. Photograph it.
Step 3. Inspect the needle. 10x microscope. Look for a burr. Flush the bore. If either is bad, replace the needle.
Step 4. Check the cap crimp. Spin test. If the cap turns by hand, re-crimp. Check the crimper anvil for wear.
Step 5. Check alignment. Run a tray of vials and see if coring is position-dependent. If it is, the autosampler needs service.
Step 6. Check the temperature. If the tray or inlet is above 40 °C, halve your septum change interval.
Step 7. Check the septum spec. Confirm the PTFE film thickness (125 µm minimum for coring resistance), the Shore A (45 to 55), and the temperature rating against your method.
Step 8. Log it. Record the injection count at septum change and needle change. If you do not log, you cannot trend, and you will be back here in three months.

What to buy, and what to stop buying

My recommendation, having run this loop too many times: standardise on a 1.0 mm silicone/PTFE septum with a 125 µm PTFE film, Shore A 50, temperature rated to at least 60 °C, for all ambient HPLC work. For heated trays, go to a 1.5 mm version rated to 80 °C. For GC inlets, use a 3.0 mm high-temperature septum rated to 250 °C and change it every 100 injections. Buy the needle and the septum from the same supplier where you can, because the puncture count rating assumes a matched pair.

Stop buying the cheapest septum on the catalogue. The price difference between a $0.15 septum and a $0.40 septum is $0.25 per injection. A cored septum that costs you one column, one day of troubleshooting, and one failed batch costs hundreds. Do the arithmetic. It is not close.

If you are setting up a new method and want a second opinion on the septum spec against your needle diameter and injection volume, the analytical testing side of our network runs method development work and can check the whole sample path, not just the septum. For the septum itself, the HPLC septa range on this site covers the standard dimensions and the PTFE film options, and the HPLC vial page has the matching vial and cap dimensions so you can check the fit before you order.

One more thing: the septa you cannot see

The septum in the vial is the obvious one. The septum in the injection port of a manual injector, the septum in the solvent reservoir cap, and the septum in the sample loop are the ones that get forgotten. A cored solvent reservoir septum drops a silicone disc into the mobile phase, and you spend a week chasing a ghost peak that came from the bottle. Check every septum in the flow path, not just the one in the autosampler tray. The EPA SW-846 methods are a good reminder of how many sample handling steps sit between the bottle and the detector, and each one is a potential coring site.

Takeaway

Coring is a system failure, not a septum failure. The septum is the part that shows the damage, but the cause is almost always a dull needle, a loose cap, a needle that is too large for the septum, or a septum that is too thin or too hard for the job. Fix the needle and the cap first. Then check the septum spec against the numbers: 125 µm PTFE minimum, Shore A 45 to 55, 1.0 to 1.5 mm thick for ambient HPLC, 3.0 mm and 250 °C rated for GC inlets. Change the needle when you change the first cored septum, log the injection counts, and inspect under a microscope before you replace anything expensive. The lab that does those four things stops having coring problems. The lab that does not keeps buying septa and wondering why the ghost peaks come back.

References & further reading

Related resources from our lab consumables network:

External references and standards: