Applying Gas Chromatography to Environmental Geochemistry

In the second of a two-part Q&A The Column spoke to Paul A. Sutton, a research fellow in the Petroleum and Environmental Geochemistry Group at Plymouth University (Plymouth, UK) about his experience with high temperature gas chromatography (HTGC), and his best practices for analysts in the lab.

Q: Why is high temperature gas chromatography (HTGC) coupled with mass spectrometry your method of choice?

A. High temperature gas chromatography (HTGC) is routinely used in our laboratory for petroleum wax analysis and tetraacid (ester) screening. It is probably one of our busiest instruments and is used almost every day. Aside from petroleum analysis we have used HTGC for analysis of total organic extracts from sediment, shale and lignite, industrial feedstock formulations, microbial membrane lipids, insect cuticular extracts, and lipid extracts from marine invertebrates. Compared to conventional GC, which uses columns with an upper oven temperature limit of 300–350 °C, HTGC columns can be heated to 430–450 °C. In practical terms this limits conventional gas chromatography (GC) to the analysis of compounds with up to around 35–40 carbon atoms before thermal decomposition of the stationary phase becomes excessive and significant band broadening is observed during the isothermal stage. For HTGC the limit is nearer to 80–90+ carbon atoms before band broadening becomes an issue, so the analytical window is almost doubled. Very often we see peaks on the HTGC that are simply not eluted with a conventional set-up. Whilst some polar (derivatized) analytes subject to HTGC are amenable to identification using other mass spectrometry (MS) techniques (for example, liquid chromatography coupled to MS [LC–MS]) such techniques are not so useful for hydrocarbons and electron impact MS is more usual.

Photo Credit: Coneyl Jay/Getty Images

Coupling of HTGC with mass spectrometry is not a new idea and has been around for over 30 years. However, there are relatively few literature reports of HTGC–MS operated with oven temperatures above 380 °C, suggesting that such systems were difficult to operate routinely or it was considered that compounds would thermally degrade above this temperature. Some of these earlier issues have been ameliorated because of technological developments in column, ferrule, and mass spectrometer performance. I can give a few examples here. For consumables development, steel clad columns, as a result of their strong mechanical strength, have now largely replaced aluminum columns. Aluminium columns were hard to work with and required dissolution of the coating with sodium hydroxide before they could be fitted to prevent electrical continuity issues; they also readily became brittle and tended to snap, sometimes after only a few oven cycles. Supply was sometimes an issue and I once received a column a year after it had been ordered.

The availability of a choice of aluminium and steel ferrules for high temperature applications, even though they can only be used once, has ensured that column fitting can be made leak tight. With an extended analytical window the molecular weight range of compounds eluting from the HTGC can exceed the capability of most current bench top mass spectrometers, which at around 1000–1100 Da equates to about pentaheptacontane or nC₇₅. The mass spectrometer we use is a reflectron time-of-flight bench-top instrument that has been modified by the vendor to enable operation of the ion source up to 400 °C, data recording up to m/z 1850, and a transfer line capable of being heated to 450 °C. We don't need all of this capability all of the time so we can set each of these parameters independently according to the type of sample. Software development should also not be ignored. Unsurprisingly there is relatively high column bleed associated with cycling the chromatographic column to high temperature, but this can be largely removed using an algorithm that differentiates background ions from analyte ions and subtracts background from the chromatographic data. So, given the combination of the type of work undertaken in our laboratory and recent hardware, software, and consumables developments it seemed an appropriate time to revisit the coupling of HTGC with MS, especially in a bench-top format.

Q: What are the challenges associated with HTGC–MS?

A. The major challenge is keeping the system leak tight to air. Obviously when cycling the oven temperature there is the chance of connections opening up (this is mainly a result of the heating/cooling of the connector nuts rather than a ferrule issue), so in this respect it is perhaps not as routine as hoped. I leak check after every one or two analyses. The other susceptible point is the silica transfer line from the union with the end of the metal clad chromatographic column through the transfer interface into the mass spectrometer. I have had one of these snap whilst the GC oven was at 430 °C, but the mass spectrometer is remarkably robust and survived without damage. In order to minimize the amount of time that the silica is exposed to high temperature I usually operate a relatively fast oven ramp from 40–430 °C at 20 °C min^-1, with little loss in chromatographic resolution. The use of 0.18 mm high temperature silica for the transfer line instead of 0.25 mm improved things a great deal; its higher mechanical strength means I can usually do at least 40 analyses before needing to replace a broken transfer line. In terms of sample analysis, the most challenging aspect is optimizing the temperature of the transfer line and ion source. We have shown that operating at higher ion source temperatures gave more sensitive detection of "waxy" compounds like n-alkanes (for example, n-pentacontane, nC₅₀).

Q: Do you have any useful tips for analysts new to the technique?

A. In a practical sense, as with other GC and GC–MS techniques, column installation is important, except here you are dealing with a steel clad column. While the end of the column needs to be cut square, it is a bit more difficult with metal columns. I have tried using a steel column cutter but found that it pinched the tube so that the on-column syringe needle would not fit into the end. Now I score the column with the edge of a file and rotate (wiggle) the end of the column until it separates. The cut end of the column can be carefully squared off with a file before reaming it, so that the syringe needle fits snugly inside. I always check that once installed in the cool-on-column inlet the syringe needle can be smoothly inserted into the column through the inlet septum. If the column end is not squared off or reamed sufficiently to accommodate the syringe needle, the needle can get snagged in the end of the column. This is especially important if you are intending to inject using the autosampler. I have a couple of excellent examples of bent and broken syringe needles in the laboratory that bear testament to a poor column installation. When fitting steel ferrules to a steel column you may find that the external diameter of the column does not match the internal diameter of available ferrules. This can be overcome by carefully drilling out the ferrules to suit using modelling drill bits. Here it would be useful if manufacturers could produce steel or aluminium ferrules with an internal diameter matched to the external diameter of steel clad columns. High temperature string is a useful thing to have around and is ideal for keeping the 2 m transfer line tidy within the GC oven.

Don't forget about the nature of the sample matrix! For petroleum wax analysis it is necessary to make sure that the sample and syringe are heated to ensure that the waxes stay in solution until injected. I do this by dissolving the sample in cyclohexane and heating the sample and syringe at 70 °C for a period of time before doing the injection manually. Preparative sample separation can also be an extremely useful tool for MS. For example, in cases where the target analyte is present in low abundance compared to other chemicals in the sample it is necessary to separate the target analyte so that it can be concentrated to obtain a representative mass spectrum.

Q: Why did you apply HTGC–MS to the analysis of insect cuticles? What can the data generated provide?

A. This work started after my colleague Steve Rowland (Plymouth University) went to a presentation by Stephen Martin (University of Salford) about insect cuticular hydrocarbons and their importance as semiochemicals and water balance regulators. Up until this time there had been few reports of insect cuticular hydrocarbons with >35 carbon atoms and there is a lot of debate amongst entomologists about their role, so we thought we could help out using HTGC. To date, most of the work identifying >C35 cuticular hydrocarbons has involved the use of matrix assisted laser desorption ionization-mass spectrometry (MALDI–TOF-MS), which does not discriminate between different isobaric isomers. I screened approximately 100 insect cuticular lipids extracts, representing 11 different species, using HTGC, and a lot of these contained peaks beyond nC₃₅. Some of these peaks appeared in the chromatographic region we associate with triacylglycerides but many did not. When I analyzed these using HTGC–MS we found that besides C_50–54 triacylglycerides (which gave both molecular ion and characteristic fragment ions in a single analysis), sometimes C_40–50 wax esters were present, but more interestingly we identified C₃₇ and C₃₉n-alkanes and C_37–43 mono- to trienes. Following hydrogenation of the sample and re-analysis using HTGC–MS, the C₄₁ dienes were shown to be n-alkadienes. The coupling of HTGC with MS even enabled some of the isobaric isomers to be separated. This work supported the contention that hydrocarbons with >35 carbon atoms may be important in the role of insect cuticular compounds.

Q: What advantages did HTGC–MS offer over conventional GC–MS in this investigation?

A. The use of HTGC and HTGC-MS in this work was extremely beneficial as the HTGC techniques suffer from less discrimination between the responses of nC₂₀ to nC₆₀ than is seen for nC₃₅ in conventional GC work at a typical maximum oven operating temperature of 300 °C. Simply put, some of the compounds detected using HTGC–MS would not have been detected under conventional GC–MS conditions, and this is likely to be applicable to a wide range of sample types particularly organic solvent extracts.

Q: Anything else you would like to add?

A. Yes, whilst I have performed most of the practical work covered in this interview, none of this would have been possible without research and industrial sponsorship and support from the School of Geography, Earth, and Environmental Sciences and members of the research team within which I work.

Paul Sutton has been at Plymouth University since graduating as a mature student with a B.Sc. (Hons) degree in environmental science in 1995. After completing a PhD in organic geochemistry in 2000, he undertook a post-doctoral post investigating the nutritional status of soils in the Shimba Hills National Reserve, Kenya. This was followed by a three-year post-doctoral project characterizing chromatographically "unresolved complex mixtures" (UCMs) from crude oils using preparative-gas chromatography. Until 2011 Paul was employed as a scientific officer and was seconded onto a two-year Joint Industry Project to develop a method for the quantification of C₈₀ ("ARN") tetraacids in crude oils. In 2011 his role changed to a senior research fellow in the School of Geography, Earth, and Environmental Sciences. His current research interests include developing separation techniques for high-molecular-weight petroleum organic compounds and development of applications for HTGC.

E-mail: P.A.Sutton@plymouth.ac.uk

Website: http://www.research.plymouth.ac.uk/pegg/staff/sutton_pa/pasutton.htm

This article is from The Column. The full issue can be found here:http://images2.advanstar.com/PixelMags/lctc/digitaledition/October06-2014-uk.html#2