Q&A with Select Gas Chromatography Experts
This month Lab Manager poses four questions on GC usage, likes, and dislikes to a panel of five experts
Q: Describe your organization and how it uses GC.
A: Andrew Skroly: CDS is a specialty chemical company supplying the automotive, aerospace, refrigerant, and polymers industries. We use GC every day to characterize new products, support our R&D functions, troubleshoot customer complaints, and assay raw ingredients for antioxidants, polyol esters, solvents, urethane prepolymers, and others.
Gary Deger: We manufacture GC injection solutions and so run test samples for prospective customers and for studying new applications. Specifically, we make systems that eliminate GC sample prep. We generally analyze polymers, and our systems are used almost daily.
Daniel Fabry: Haverford is a small liberal arts college with a strong research-based chemistry program. We use GC-MS to identify or confirm molecular weights and GC-FID to monitor the progress of chemical reactions. At Haverford, we use GC within our organic and environmental research labs. The organic group studies natural products for potential medicinal applications; the environmental group analyzes samples from the Gulf of Mexico.
Philip Marriott: As an academic institution, we use GC to develop analytical methods in comprehensive two-dimensional gas chromatography (GCxGC) and multidimensional gas chromatography (MDGC), usually supported by MS detection. Our published research includes fundamental relationships in advanced GC methods, method developments for high-resolution chemical separations, and applications of GC, MDGC, GCxGC to demonstrate the scope and applicability of our methods to complex samples. Our samples include petrochemicals, pesticides, fatty acids, essential oils, aroma compounds, flavonoids and polyphenols, and illicit drugs.
William Terzaghi: My group primarily employs GC to analyze resveratrol content of various plant tissues and for characterizing fatty acids and lipids in plant tissues. Although we share the instrument with other groups, we utilize it at least one full day per week.
Q: What kinds of detectors do you use, and why?
A: Andrew Skroly: We use MS, FID, and TCD. Together, they provide a multitude of options for the many different samples we are asked to analyze.
Gary Deger: We use only MS detectors. Most of our samples are analyzed for unknown polymers and additives in plastics, rubbers, coatings, biofuel source material, and tobacco, among others.
Daniel Fabry: Haverford has three instruments: a Perkin Elmer Clarus GC-MS, an Agilent 7890 outfitted with a FID and MS detector, and a new Shimadzu-2014 GC with an FID. The Agilent 7890 is a dual inlet system with a PTV inlet; the Shimadzu GC is used to monitor reaction progress with mostly isothermal temperature programs. We employ the PerkinElmer GC-MS for identification and confirmation. We employ a variety of temperature programs depending on the column type and compound of interest.
Philip Marriott: We use a variety of detectors: a nitrogen-phosphorous detector for atmospheric samples, petrochemicals, and smoke; a flame photometric detector (FPD) for sulfur compounds from varied samples, FPD/P mode for organophosphate
pesticides, phosphate esters in flame retardants, and phosphorous in chemical weapons; electron capture for chlorinated pesticides and biphenyls; olfactometry (sniffing detection) for aroma compounds in wine, herbs, spices, and coffee; FID, quadrupole MS for general applications; and time-of-flight MS for GCxGC.
William Terzaghi: We use MS detection for all of our samples [in order] to obtain positive identification of every peak.
Q: What are the most significant bottlenecks in your GC workflows, and how do you overcome them?
A: Andrew Skroly: There are not many bottlenecks in our workflows. We have five GCs in the lab that we have set up for nonpolar, wax, and polar columns. Our injectors consist of on-column to split/splitless.
Gary Deger: We really do not have any workflow issues unless a system is down, which can result in samples backing up. We try to fix as many problems as we can ourselves, and if we cannot, then we call in a GC service engineer. Sometimes there is a small delay in getting them in, but never more than a few days. If anything, I would complain about their rates being too high.
Daniel Fabry: The most significant roadblocks arise from oven cooling after a high-temperature program ramp. At the forensic lab, a major bottleneck has been instrument and host computer communication. Recently, our Agilent GC-FID/MS has had some leak issues originating from our inlet setup. Of course, maintenance is always a bottleneck but that cannot be avoided. I perform instrument maintenance regularly to avoid any large issues from lack of care and cleaning. Instrument computers often generate problems when trying to connect to a network or upgrade software. Most instruments require a hardware upgrade before you can run them with a different operating system, which can be costly, and Microsoft Updates can be a source of problems if your computer is networked. Bottlenecks in workflow are most often remedied by good communication, proper care, and a good support system that can help you when things go wrong.
Philip Marriott: Since our work is exclusively basic GC research rather than routine applications involving classical workflow considerations, we are continually changing methods, changing columns, and reconfiguring our GCs for advanced, multicolumn methods. Our GC workflow normally includes activities such as Deans switch balancing, entering complex procedures for event control, then testing the method
for reliability and performance. We also require very fast performance of all our detectors, so the MS acquisition rate is critical.
William Terzaghi: Our most significant bottlenecks are sample throughput and instrument availability. Once we have acquired sufficient preliminary data, we hope to write a grant to purchase a GC-MS exclusively devoted to our projects [in order] to improve the availability, and we are always looking for ways to shorten the turnaround time between samples by altering programs.
Q: What can GC systems vendors do to improve their products and/or streamline your workflow?
A: Andrew Skroly: I’d like to see systems where you could change out detectors depending upon the analytes in question. Another improvement would be better communication with LIMSs, including improved ability to dump data and operate the GC.
Gary Deger: Most GC-MS systems perform similarly, but the operating software can be an issue. We have systems from all the major vendors, but some of the user interfaces crash or lock up more than others [do].
Daniel Fabry: There are two aspects to acquiring an instrument: the instrument itself and service. The major vendors should hire more service engineers who cover smaller areas, regardless of whether customers purchase service contracts. There is one major vendor, in particular, that has significant room for improvement in this regard. Agilent has started a YouTube channel that offers troubleshooting and maintenance tips that have been a huge help. Haverford College employs me to fix and maintain their instrumentation and does not purchase maintenance contracts. Every company treats their noncontract clients differently, but I feel that more companies are choosing to hire in-house instrument specialists to avoid the high cost of instrument contracts. I feel response time and willingness to divulge technical information varies for contract and noncontract customers. Another way GC vendors can improve my workflow is [by] offering a functional website for finding consumables. I cannot tell you how many times I enter a product number in a search and nothing comes up.
Philip Marriott: We have an interest in making MDGC methods setup more systematic, with method development guidelines or simulations that can provide accurate balancing conditions, and improvement in automated entries into events tables via a GUI interface. Generally, faster MS would be of advantage, especially for highresolution TOF-MS. In terms of GCxGC, there is a continuing need for innovation in the areas of software for data acquisition, presentation, and interpretation. These improvements would help GC and GC-MS users with new capabilities, especially if the future of GC becomes more directed to MDGC and GCxGC technology.
William Terzaghi: Vendors can make machines that shorten the run time for each sample. It will also be useful if they could develop kits for sample preparation that would shorten the sample prep time. Finally, columns that allow us to resolve resveratrol and its glycosylated forms without derivatization would be very helpful.