I have replaced more crimp caps and septa than I care to count, and the single most expensive habit I see in headspace labs is blaming the septum for coring when the vial is the thing that broke the needle. Two years ago a customer running EPA method 8260 sent me a photo of a syringe plunger bent at maybe 15 degrees off axis. They had already switched septum suppliers twice. The real problem was a batch of vials with a 20.5 mm crimp finish where the neck bore was 0.4 mm off center, and the autosampler needle was clipping the glass on the way in. That is the story I want to tell here: how the vial itself, not the septum and not the needle, drives coring and needle damage, and how to find it before you lose a $600 syringe.
Coring is a specific failure. The needle punches a disc out of the septum and pushes it into the vial, where it either floats on the sample, sits in the headspace, or gets drawn back into the syringe and blocks it. Needle damage is a related but separate failure: the needle tip is bent, burred, or plugged, usually because it hit something hard (glass, a crimp cap edge, or a cored septum fragment that welded to the tip under heat). The two are often reported together because a cored septum fragment lodged in the needle will cause the next injection to miss, and an operator who sees a bad chromatogram will pull the syringe and find a bent tip.
What distinguishes vial-caused coring from septum-caused coring is where the damage appears. If the coring is a septum problem, you see it across every vial in the tray regardless of supplier. If it is a vial problem, you see it clustered in one lot, or in one position on the tray, or only on vials that were crimped by a particular operator. That clustering is the diagnostic. I have learned to ask for the tray map before I ask for the septum certificate.
The three vial features that govern needle entry
Three dimensions on a headspace vial control whether a needle enters cleanly or does damage:
- Neck bore diameter and concentricity. For a standard 20 mm crimp vial, the nominal neck ID is around 12.5 mm, but the tolerance that matters is concentricity relative to the outside diameter. A neck that is 0.3 mm off center will let a 0.5 mm OD needle through on one side and clip glass on the other. Most catalog sheets do not publish concentricity. Ask for it.
- Seat depth and flange flatness. The distance from the top of the crimp flange to the septum seat determines how far the needle has to travel before it reaches the septum. If the flange is warped by more than about 0.15 mm, the crimp cap sits at an angle and the needle enters at an angle.
- Wall thickness at the neck. Thin necks (under about 1.0 mm) deform under crimping force and pull the septum seat out of plane. I have measured 0.7 mm necks that ovalized by 0.2 mm after a 2.0 N·m crimp.
None of these numbers appear on a typical certificate of analysis. That is the root of the problem. You buy a headspace vial on the basis of volume, finish, and glass color, and the dimensions that actually govern needle entry are treated as manufacturing detail.
Why the vial, not the septum, is usually the culprit
Septum coring is well understood. A PTFE-faced silicone septum will core if the needle is dull, if the needle gauge is too large for the septum hardness, or if the same injection point is used more than about 100 times. Those are real failure modes and I do not want to dismiss them. But in a headspace autosampler, the needle does not enter the same point repeatedly. It enters a fresh vial every time. So the “same point” coring mechanism does not apply. What applies instead is entry angle and entry force, and both are set by the vial.
Entry angle
A headspace autosampler needle travels vertically. If the vial is not vertical in the tray, the needle enters at an angle. The vial is held vertical by the tray nest and by the crimp cap sitting flat on the transport mechanism. A warped flange or an off-center neck tilts the vial by 1 to 3 degrees. At a 2 degree tilt, a needle entering 8 mm into the septum produces a lateral load on the tip of about 0.28 times the insertion force. For a 23 gauge needle at 4 N insertion force, that is roughly 1.1 N of side load. That is enough to bend a 23 gauge tip over a few hundred injections, and enough to shear a septum disc rather than punch it cleanly.
Entry force
Entry force is the sum of septum penetration force and any friction from the needle rubbing the neck bore. A clean 20 mm PTFE/silicone septum at 23 gauge needs about 3 to 5 N to penetrate. If the needle is rubbing the neck because the bore is off center, add 1 to 2 N of friction, and the needle tip is now loaded laterally while it is trying to punch through silicone. That combination is what produces the characteristic crescent-shaped core rather than a clean round disc. If you see crescent cores under a magnifier, suspect the vial, not the septum.
Heat makes it worse
Headspace vials run hot. A 20 mL vial at 80 °C for 30 minutes softens the silicone septum (typical hardness drops from about 50 Shore A at 23 °C to roughly 40 Shore A at 80 °C) and expands the glass. The thermal expansion of borosilicate glass is about 3.3 × 10⁻⁶ per °C, which over a 12.5 mm neck bore at 80 °C is only about 0.003 mm, not enough to matter on its own. But if the neck was already at the low end of tolerance, the crimp cap relaxes slightly at temperature and the septum seat lifts by a few hundredths of a millimeter. That is enough to change the entry angle on the next injection. This is why vial-caused coring often appears on the second or third heating cycle of a re-used vial, not the first.
The vial features that prevent coring and needle damage
If you want to stop chasing this problem, buy vials on the basis of the following specifications, and demand them in writing.
Crimp finish and dimensional standard
Use a 20 mm crimp finish that conforms to the dimensional envelope in the Agilent chromatography consumables reference drawings, or the equivalent Thermo Fisher drawing for your autosampler. The key numbers to ask for:
- Neck ID: 12.5 mm ± 0.2 mm
- Neck concentricity to OD: ≤ 0.15 mm TIR
- Flange flatness: ≤ 0.10 mm
- Seat depth (flange top to septum seat): 2.4 mm ± 0.1 mm
- Neck wall thickness: ≥ 1.0 mm
These are tighter than what most catalog vials deliver. I have measured catalog vials at 0.35 mm TIR concentricity and 0.2 mm flange warp. Those vials will run fine on a gentle autosampler at low injection counts, and they will destroy a needle on a high-throughput system running 200 vials a day.
Glass type and annealing
Use Type I borosilicate glass (USP <731> / ASTM E438). Type I glass has a low coefficient of expansion and resists the thermal cycling of headspace. More importantly, annealed Type I glass holds its neck dimensions after crimping. I have seen Type III soda-lime vials where the neck ovalized by 0.25 mm after a single crimp because the glass was not properly annealed. The crimp tool is not the problem; the glass is.
Crimp cap and seal geometry
The crimp cap is part of the vial system. A 20 mm aluminum crimp cap with a PTFE-faced silicone septum (0.125 mm PTFE, 3.0 mm silicone) is the standard for headspace. The PTFE face is what the needle actually penetrates, and PTFE is soft enough to core cleanly at 23 gauge. The aluminum skirt must be 0.25 mm thick or the crimp will not hold torque. I have measured 0.20 mm skirts that relaxed after 20 minutes at 80 °C and let the septum lift. The result is a needle that enters at an angle and a chromatogram full of ghost peaks from a leaking seal.
For a deeper look at how the cap and septum interact with the vial headspace, the GC vial reference material is worth reading alongside this, because the same crimp mechanics apply to both 20 mm headspace and 11 mm GC vials.
A decision checklist with concrete numbers
Use this when you are specifying a new vial or auditing a supplier. Score each line pass or fail. Two fails means reject the lot.
| Check | Method | Accept | Reject |
|---|---|---|---|
| Neck ID | Pin gauge or caliper, 10 vials | 12.3 to 12.7 mm | < 12.2 or > 12.8 mm |
| Concentricity | Dial indicator on neck ID vs OD | ≤ 0.15 mm TIR | > 0.20 mm TIR |
| Flange flatness | Granite surface plate + feeler gauge | ≤ 0.10 mm | > 0.15 mm |
| Seat depth | Depth micrometer, 10 vials | 2.3 to 2.5 mm | < 2.2 or > 2.6 mm |
| Neck wall | Optical comparator | ≥ 1.0 mm | < 0.9 mm |
| Crimp retention | 2.0 N·m crimp, 80 °C 30 min, then pull test | Cap holds, no lift | Cap lifts or rotates |
| Septum penetration | 23 gauge needle, 5 vials, 10 injections each | ≤ 5 N, no crescent cores | > 6 N or any crescent core |
| Puncture count | Same vial, repeated injection at one point | ≥ 100 clean punctures | < 80 or coring before 100 |
The puncture count line matters because it tells you whether the septum and the vial are working together. If a fresh vial cores before 80 punctures at a single point, the septum is too soft or the needle is dull. If it cores before 80 punctures at multiple points, the vial neck is the problem. That distinction saves a lot of supplier arguments.
Operational precautions that actually prevent needle damage
Specification gets you a good vial. Operation keeps it good. These are the practices I insist on in my own lab and in customer labs I audit.
Crimp with a calibrated tool and a torque value
Use a crimp tool with a torque setting of 2.0 N·m ± 0.2 N·m for 20 mm caps. Under-crimping (below 1.5 N·m) lets the septum lift at temperature and the needle enters at an angle. Over-crimping (above 2.5 N·m) deforms the neck and cracks the glass at the flange. Both produce coring. I have seen labs crimp by feel for years and wonder why 3 percent of vials core. Calibrate the tool every 500 crimps and log it. The Thermo Fisher chromatography and sample handling documentation has a useful crimp tool calibration procedure if you need a template.
Inspect the needle every 500 injections
Pull the syringe, inspect the tip under 20× magnification, and check for a burr or a bend. A burred tip will core a good septum on a good vial. Replace the needle at the first sign of a burr, not after it starts missing injections. A 23 gauge needle on a headspace autosampler should last 2,000 to 5,000 injections if the vial is good. If you are replacing at 500, the vial is the problem.
Do not re-use vials
A headspace vial is a single-use consumable. Re-crimping a used vial puts a second crimp on a neck that has already been deformed once. The second crimp is never as flat as the first, and the needle entry angle shifts. I have measured 0.18 mm of additional flange warp on a re-crimped vial. That is enough to core. If your method requires re-use for cost reasons, at least discard any vial that shows a visible crimp mark on the neck.
Match the needle gauge to the septum
Use a 23 gauge (0.64 mm OD) needle for standard 20 mm PTFE/silicone septa. A 21 gauge needle (0.81 mm OD) will core a soft septum at the first injection. A 25 gauge needle (0.51 mm OD) is too flexible and will bend if the vial neck is off center. The 23 gauge is the sweet spot, and it is what most autosampler manufacturers specify.
Control the tray and the nest
Check the autosampler tray nest for wear every quarter. A worn nest lets the vial sit at a tilt of up to 1 degree, which is enough to produce the lateral loading I described earlier. I have seen a tray with 0.5 mm of wear on the nest shoulder produce a 40 percent increase in coring rate across the whole tray. The fix was a $90 nest replacement.
Troubleshooting checklist: is it the vial, the septum, or the needle?
Work through this in order. Stop at the first fail and fix that before moving on.
- Map the failures. Are cored vials clustered in one tray position, one lot, or one operator’s crimps? If yes, suspect the vial or the crimp. If no, suspect the septum or the needle.
- Inspect the core shape. Clean round disc = normal septum wear. Crescent or half-moon = entry angle problem, which is a vial or crimp problem. Ragged edges = dull needle.
- Measure 10 vials from the suspect lot. Neck ID, concentricity, flange flatness, seat depth. Compare to the numbers in the checklist above.
- Check the crimp torque. Calibrate the tool. Re-crimp 10 vials at 2.0 N·m and run them. If coring stops, the tool was the problem.
- Inspect the needle tip. 20× magnification. Burr, bend, or plug = replace.
- Run a septum control. Same lot of septum on a known-good vial lot. If coring stops, the vial lot was the problem. If it continues, the septum or the needle is the problem.
- Check the tray nest. Look for wear, tilt, or loose fixtures.
- Review the temperature program. If coring appears only on the second or third heating cycle, suspect thermal relaxation of the crimp, which points back to the cap and the crimp torque.
Step 6 is the one most labs skip, and it is the one that settles the argument. It takes an hour and it saves weeks of supplier back-and-forth.
What I recommend, with numbers
Buy 20 mm crimp headspace vials that meet the dimensional envelope I listed: 12.5 mm ± 0.2 mm neck ID, ≤ 0.15 mm concentricity, ≤ 0.10 mm flange flatness, 2.4 mm ± 0.1 mm seat depth, ≥ 1.0 mm neck wall, Type I borosilicate glass. Pair them with 20 mm aluminum crimp caps, 0.25 mm skirt, PTFE-faced silicone septum (0.125 mm PTFE, 3.0 mm silicone). Crimp at 2.0 N·m ± 0.2 N·m with a calibrated tool. Use a 23 gauge needle. Replace the needle at the first burr, and expect 2,000 to 5,000 injections per needle on a good vial. Do not re-use vials.
If you are running a high-throughput lab, ask your supplier for the concentricity and flange flatness data per lot. A supplier who cannot provide it is selling you a vial that will core. The Restek technical resources library has good reference material on septum selection and needle compatibility that is worth reading alongside the vial specs.
One more point on the sample matrix. If you are running volatile organics in a heavy matrix, the vial is not the only consumable that matters. The reagent bottle you use for standards and the HPLC autosampler vial you use for the liquid-side confirmations both feed into the same method, and a dimensional problem in any of them shows up as a bad injection. Keep the whole sample path to the same dimensional discipline.
For labs running EPA SW-846 methods, the method documents specify the vial finish and septum type but not the concentricity. That is a gap the method leaves to the lab. Fill it with your own incoming inspection. Ten vials and a dial indicator will tell you more about your coring rate than any certificate of analysis.
Takeaway
Coring and needle damage in headspace autosamplers are usually blamed on the septum and usually caused by the vial. The dimensions that matter (neck concentricity, flange flatness, seat depth, neck wall thickness) are not on the catalog sheet, so you have to ask for them and measure them. Buy to the numbers in this article, crimp at 2.0 N·m with a calibrated tool, use a 23 gauge needle, do not re-use vials, and run the septum control before you call the supplier. The clustering of failures, the shape of the core, and the puncture count will tell you which component is at fault in under an hour. That is a better use of your time than another septum supplier switch.
References & further reading
Related resources from our lab consumables network:
- Headspace Vial at headspacevialsupplier.com
- Gc Vial at gcvialfactory.com
- Reagent Bottle at reagentbottleforsale.com
- Hplc Autosampler Vial at hplcautosamplervials.com
External references and standards: