Sealing it with Septa

While gas chromatographers may take their septa for granted, in fact these small and seemingly unremarkable polymer disks keep air out of the carrier-gas stream when used in an inlet and keep samples intact and uncontaminated when used in sample vials. Choosing the wrong septa can compromise method accuracy and repeatability as well as reduce column life in extreme cases. This installment addresses septa for inlets and sample vials.

In gas chromatography (GC), septa form part of the critical juncture between internal passages and the external ambient atmosphere with its oxygen and water. Unlike other sealing components such as ferrules, o-rings, or inlet ring seals, septa are mechanically challenged by a syringe needle once-or in the case of autosampling more than once-with every injection. To add insult to these injuries, inlet septa are subjected to continuous high temperatures and vial septa are exposed to solvent vapors, either of which can destroy the integrity of an inappropriate septum. Septum technology and chemistry have undergone extensive development and improvement during the more than half-century since septa first met gas chromatography. A number of septum-less solutions are also available.

Septa for Inlets

Arguably the most stressed-out parts of a GC system, inlet septa are tasked with maintaining a leak-tight seal between the atmosphere and carrier gas while not introducing significant contaminants into the carrier-gas stream-all of this while heated to upwards of 250 °C and being repeatedly punctured by sharp needles. I’ve only otherwise encountered this situation at the acupuncturist’s office, and it is an experience I’m not likely to repeat. While concentrating life-force energies may be good for us humans, in the case of septa it’s all about the chemistry.

Excluding gas-injection valving, which is not addressed here, a number of GC inlet systems do not have septa. Some on-column inlets have used various nonseptum sealing arrangements. Capsule-based inlets do not have septa, either. They work by sealing a liquid or solid sample into a metallic or glass capsule, placing it inside the inlet, purging with carrier gas while the capsule heats up, and then puncturing the capsule to release the vaporized sample. I had thought of capsule inlets as out-of-date, but a recent patent (1) describes a new type of capsule inlet that uses a heat-sealed polymeric capsule. One type of septumless inlet accessory works with heated inlets and has two seals that are activated by passage of the syringe needle.

Classical cold on-column inlets may have a septum but generally it is not heated, although in such cases it still is possible for pieces of septum to be displaced into heated areas and act as a contaminant source. The predominant inlet type today remains the heated inlet, and usually it is a split–splitless type. Packed column inlets remain in use as well and experience many of the same problems with septa.

The dichotomy of a heated-inlet septum is that it must simultaneously withstand high temperatures while providing a gas-tight seal over hundreds or more injections. Failure to do so can cause septum bleed, may create active adsorptive sites inside the inlet, and, in severe cases, can compromise the split ratio because of carrier-gas leakage.

Septum Bleed

Most gas chromatographers have heard of or experienced septum bleed-the appearance of extra peaks or an offset baseline because of septum materials entering the carrier gas stream and column. These unwanted peaks can then lead to quantitative errors as well as the misidentification of target analytes. Volatilization of the lighter fractions of a septum’s polymeric matrix and the deposition of septum particles in the inlet are the predominant sources of septum bleed.

Septa are composed of one or more layers of polymeric materials. Today, essentially all septa for heated inlets are made of various polysiloxane materials. The process of polymerization and cross-linking may leave behind some lower-molecular-weight prepolymeric molecules with fewer than 10 siloxane units in linear, cyclic, or branched configurations. Upon heating, the relatively volatile materials that reside on the surface of a septum will evaporate into the nearby space.

The septum does not get as hot as the bulk of an inlet. The septum nut acts as a heat sink that maintains the septum temperature between 75–100 °C below the inner inlet temperature. Septum temperatures vary a lot between different inlet systems, but typically for a 300 °C split–splitless inlet the septum will run at around 250 °C. Even so, this may be hot enough to cause detectable quantities of septum bleed to evaporate from the septum.

One of the primary functions of an inlet septum purge-the one for which it is named-is the removal of volatized contaminants from the septum area away from the active carrier-gas feed that leads to the column. The septum purge also removes any sample and solvent from injection that may enter the septum area because of inlet overloading. This secondary function prevents flash-back of sample during injection, especially in splitless mode.

Repeated inlet overloading with condensable materials may eventually deposit enough material in the cooler septum-purge exit tube, outside the inlet, to partially or completely block septum-purge flow. A primary symptom of this situation is the gradual appearance of septum bleed peaks. In extreme cases, sample and solvent may be entrained in the carrier-gas supply line to the inlet and then reappear in subsequent injections as baseline bleed or as discrete peaks carried over from previous injections. The continued presence of sufficient septum-purge flow can be assured by monitoring the flow as part of routine maintenance. And of course, avoid injecting excessively large sample volumes that could overload the inlet.

Septum bleed may also stem from the mechanical breakdown of a septum, caused by abrasion of small septum pieces as the syringe needle passes through. Septum coring by the syringe needle is another less common source of septum pieces. Septum particles that migrate into the inlet liner will immediately experience a much greater degree of volatilization at the higher temperatures in the inlet’s active vaporization area. The particles may also selectively absorb higher-boiling sample components and cause sample discrimination effects, where the fraction of sample entering the column is a function of the individual components’ molecular weights. In some cases, the presence of septum pieces in the active sampling area may cause the partial decomposition of sensitive components and quantitative errors.

Septa with a fluoropolymer coating on the carrier-gas side may help alleviate septum bleed to some degree, but after a few injections the polymer coating will be compromised and the bulk septum material will be exposed to the carrier-gas flow.

Septum Leakage

A classic test for septum health and lifetime consists of sniffing the needle entry point of an inlet with a helium leak detector. This is a very sensitive check that is probably better performed at the low sensitivity setting of the leak detector. I once tried repeatedly puncturing a new septum with a syringe needle and then checking for the presence of helium. I was surprised to find that some helium leaked away for as much as 30 s after each injection, presumably because of a relatively slow-healing hole in the septum after each injection. Perhaps this effect was also related to the type of septum I used. I do recall that it was thought to be a high-quality silicone septum. After 10 or 20 tests with no differing results I gave up manually injecting and then later repeated the tests with an autosampler. The leakage was about the same, and the syringe needle itself in its standby position near the inlet nut seemed to be another momentary source of helium.

Leak-checking of a septum should be performed regularly, perhaps daily or weekly depending on the sample load. Wait a few minutes after injection to check for leaks. A small leak may not constitute an immediate problem, and it may not be practical to change a septum at the first appearance of any detectable leak. Some inlets’ septa are easier and quicker to change than others, and consider the possibility of air incursion when swapping in a new septum. Always ensure the column oven is close to room temperature and the inlet is sufficiently cool. For gas chromatography–mass spectrometry (GC–MS) systems, it may be necessary to vent the vacuum system as well, to avoid drawing air in through the column.

The type of syringe tip and size of the syringe needle also are crucial for the best septum performance. Much has been written on this topic about needle tip bevel angles, blunt needles, conical needles, side-hole needles, and so on. The best choice is to go with the syringe and inlet manufacturer’s recommendations for manual or autosampling injections. The best two styles seem to be the classical beveled or the conical blunt-tip styles in a 26-gauge size. However, these are guidelines with many exceptions, so follow the best available recommendations.

Prepunctured septa can yield increased life in some situations; again, follow the manufacturer’s recommendations. Multilayer septa are available as well. These have a softer layer of material on the outside for a good external seal to the inlet surfaces, and in the middle have a more robust material that is less prone to leakage after multiple syringe injections.

Generally speaking, an autosampler will inflict less damage on septa than will manual injection. The autosampler’s mechanism should hit the septum in a more consistent location and so can avoid multiple puncture points that could start to leak or to break down and pass particles into the inlet.

Septa for Sample Vials

Sample-vial septa have a different set of requirements than do inlet septa. Vial septa are not subject to high temperatures and are liable to be exposed to harsh solvents. Sample-vial septa must withstand multiple punctures during repeated syringe rinsing prior to injection, but there is no long-term requirement for a lasting seal. There are two remaining measures of vial septum performance: leakage and sample contamination.

Vial septa can be made of softer and better sealing materials than inlet septa. One of the crucial tricks for vial sealing is the quality of the mechanical aspects of the seal. A high performance crimper is highly recommended for consistent and reliable sealing. In a laboratory with any kind of normal to high sample load, the cost of a good crimper will be paid back in short order. Some laboratories prefer screw-top vials for ease of use and the possibility of cleaning and reusing the vial. This may be appropriate in some situations, but the crimped cap approach is ultimately more efficient because it avoids the necessity of ensuring that the vials and caps have been cleaned to the required degree.

At the same time, a softer vial septum may come with a different challenge: withstanding the solvent. Some solvents can leach septum polymers into the sample, with undesirable results similar to the effects of septum bleed. The addition of a fluoropolymer layer to the sample side of the septum largely eliminates this problem, but at the expense of making it more difficult to achieve a good seal. Here again, good quality vials, caps, and crimpers go a long way toward improved performance.

Finally, headspace vials present another challenge, that of slightly elevated temperatures and pressures. Sealing is also made more difficult in these vials by their larger diameters.

Overall, the best route for sample vial sealing is to follow the manufacturer’s recommendations and to use the vials that are provided by the manufacturer of the autosampling system. This approach may be a bit more expensive, but in the long run it is well worth it.

Reference

• A. Schleifer and C.A. Myerholtz, U.S. Patent 8,490,466 B2, 23 July 2013.

John V. Hinshaw “GC Connections” editor John V. Hinshaw is a Senior Scientist at Serveron Corporation in Beaverton, Oregon, and a member of LCGC’s editorial advisory board. Direct correspondence about this column to the author via e-mail: LCGCedit@ubm.com