Environmental Analysis

Separation methods for environmental analysis range from extraction techniques such as Soxhlet, liquid–liquid, supercritical fluid, microwave, ultrasonic, solid-phase, and solid-phase microextraction to trace analysis techniques such as LC–MS, GC–MS, and GCxGC–MS.

Participants in this installment of the Technology Forum are Nick Bukowski of ALMSCO International, Jack Cochran of Restek Corporation, Mike Buchanan of Sigma-Aldrich/Supelco, Thomas Hartlein of Teledyne Tekmar, and Dipankar Ghosh of Thermo Fisher Scientific.

What are the current challenges related to separations in environmental analysis?

Bukowski: For nonregulatory environmental analyses the magnitude of the challenges in separations tend to be proportional to the cumulative neglect applied upstream; namely sample collection, sample handling, and preparation/extraction for analysis. These will include:
• Issues with matrix interferences. The unpredictable shifting of retention times of target compounds.
• The need to analyze smaller sample amounts. Through advances in mass spectrometry (MS) sensitivity it is now possible to introduce smaller samples. This leads to improved lab efficiency by increasing the interval between scheduled maintenance routines. An analytical system that is capable of delivering adequate quantitative precision from greatly reduced sample sizes is highly future-proofed against changes in regulatory MDLs. A suitably designed TOF mass spectrometer is best suited to performing this in contrast to scanning instruments.
• The requirement to perform existing quadrupole-based analyses in faster inject-to-inject cycles. This is facilitated by, for example, time-of-flight (TOF)-MS and compressed chromatography. A TOF system, with its high data acquisition rate and full spectral sensitivity, is the perfect complement to low thermal mass column heating devices enabling vastly accelerated gas chromatography (GC) separations.

Cochran:In the GC world, congener-specific separations are always important, probably the most easily recognized application being for chlorinated dioxins and furans, where you need to determine the 2378 substituted congeners in an unbiased way due to their toxicity. MS alone will not do it. Even today, the EPA methods specify a two-column approach for 2378 tetrachlorodibenzodioxin (TCDD) and 2378 tetrachlorodibenzofuran (TCDF) specificity. And one of the columns has extremely low thermal stability, the high cyano-content phase for 2378 TCDF.

So . . . we need selective GC columns for isomeric species separations encountered in environmental GC (and GC–MS) work: PAHs, PBDEs, PCBs, toxaphene, and so forth. And we’re not going to be fully happy unless those columns have high thermal stability, too.

Buchanan:The economy has forced some laboratories to do more with less staff. Contract required quantitation limits (CRQLs) seem to get lower and lower every year. The helium shortage and subsequent increased costs from a few years back may still have a lingering effect.



Hartlein:From our perspective the challenges have to do more with sample preparation than the actual separation. For us we have to first separate the compound of interest from the matrix. These matrices include air, water, and soil, which have been fairly exhaustively characterized. The detection limits of historical compounds of interest are one of the current challenges related to separation. The ability to remove some of these compounds at lower levels using older techniques is approaching the limit given the current method guidelines. We are continually challenged by customers to meet the newer sensitivities of the detector systems while not being able to modify the basic matrix separation technique.

The second challenge related to matrices is the new compounds that need to be detected in the environment are constantly being added to existing method lists.

Ghosh:More and more unsuspected contaminants are being found in our environmental water systems, from industrial pollutants, to pharmaceuticals, veterinary products, and pesticides. These molecules have different physico-chemical properties; thus developing one unified method to screen for these is difficult. Also detection limits are getting tighter according to regulatory guidelines, so any sample preparation techniques that will allow better detection limits will be valuable.

Have the methods specified by regulatory agencies for various analytes and matrixes kept up with advances in technique and instrumentation such as SPME, tandem MS, and GCxGC?

Bukowski:Solid-phase microextraction (SPME) is not a universal technique, and would not be the best choice for, say, VOC or SVOC soil analysis.

Tandem MS offers great potential for dealing with tough matrices. In Europe, many suites of compounds historically analyzed by GC–MS have migrated successfully to liquid chromatography (LC)–MS-MS, where the high capital cost of the analyzer is quickly offset by the elimination or reduction of sample prep. GC–MS-MS should offer advantages in both improved detection limits and the ability to deal with matrix interferences obviating GC–MS analysis.

There are four fundamental, but often overlooked, vulnerabilities of GC–MS-MS when applied to the analysis of dirty matrices:
• The problem of keeping inlets sufficiently inert for more than a few dirty sample injections. Some environmental labs have investigated automatic inlet liner exchange, but are challenged by validating methods that introduce such a fundamental performance change as a brand new liner mid-batch.
• The introduction of significantly more matrix into an analytical capillary column leads to problems in detecting low level analytes. Endogenous (nontarget) components at relatively high concentrations interfere with the retention times of some target compounds, causing time shifts of analytes falling outside the tight MRM ion group time windows.
• The delivery of significantly more matrix to a mass detector leads to an escalation in complexity of MS maintenance. Beyond mere ion source cleaning — familiar to most GC–MS labs as an occasional, nonroutine procedure — GC–MS-MS requires both frequently scheduled ion source cleaning and annual (typically) mass analyzer cleaning. In the case of quadrupole mass analyzers, this can seldom be performed by the user and may even require complete replacement of quadrupole assemblies. Tandem MS manufacturers, wary of this cost-prohibitive exercise, encourage users to adopt backflushing techniques with electronic pneumatically controlled (EPC) splitters placed after the analytical column, and immediately before the MS. This is usually sold as a productivity enhancement, to shave valuable minutes off the end of each separation. There are mixed reviews to the backflushing technique; the increased analytical complexity, concerns of dumping high boiling point column effluents in unheated inlet split lines, the finite leak introduced by the backflushing EPC delivering a very low flow rate of make-up gas during analysis.
• Tandem MS systems are excellent for quantifying target compounds in difficult matrices to low detection limits. They are not suitable for characterizing unknown samples, identifying artifact peaks in routine samples, or performing comprehensive QC of analytical standards and CRMs. In the GC world, the perfect complement to tandem MS is a benchtop TOF mass spectrometer.

European GC labs were initially quick to adopt the “dirty extracts with tandem MS” approach but for one or more of the reasons listed there now seems to be a return to cleaner sample extracts. This also eases overall lab workflow with fewer manual corrections for review staff and a corresponding reduction in data reprocessing.

Realistically, GCxGC coupled with specialist benchtop TOF-MS offers complementary analytical power to tandem MS. With a selectable spectral acquisition rate that exceeds 500 spectra/s, benchtop TOF coupled with a GC analytical system is capable of accommodating any GC or GCxGC separation.The primary performance advantages of GCxGC with TOF-MS are enhanced sensitivity and reduction of baselines. Compounds that would normally be eluted as peaks of a few seconds peak width are reinjected onto the second column in between two and four modulations whose eluted peaks have peak widths in the 50–200 ms. Some compounds deliver an inadequate number or abundance of product ions for MRM analysis by tandem MS. Retention time shifts — a major problem for MRM — are easily accommodated in GCxGC, and even badly shifted peaks can be automatically recognized from their mass spectral attributes. The TOF system acquires full range comprehensive spectra, all the time; there are no timed groups of ions to maintain. GCxGC–TOF-MS allows archived data files to be subsequently examined after new compounds are appended to evolving methods. GCxGC increases peak capacity with its vast resolving power.

Cochran:If you consider published methods to be a measure, the regulatory agencies are slow in keeping up with advances in technique and instrumentation, at least for GC. SPME with GC is probably a good example, since it has been in the analytical arena for almost 20 years, and to my knowledge there is only one EPA method (8272, PAHs in pore water) that uses it. Too bad, because it’s solventless (“green”) and easily automated.

Tandem MS for GC has been ignored, method-wise, by regulatory agencies in favor of high resolution MS for chlorinated dioxins and furans, PCBs, PBDEs, and so forth, even though some historical papers showed its selectivity was equivalent to high resolution MS. But GC–MS-MS may soon enjoy a renaissance given the success of tandem MS for LC environmental work. In fact, additional GC–MS-MS systems were recently introduced to the market. And you do see EPA methods that employ LC–MS-MS (for example, 535: acetamide herbicide degradates in drinking water; 1694: pharmaceuticals and personal care products in several matrices), so the regulatory agencies are keeping up with advances there.

GCxGC is still in the “early adopter” phase, to some extent. I know the regulatory agencies have instruments, are working with the technology, and have published scientific papers, but I’m not sure if they intend to specify methods. The biggest value for GCxGC, at least with time-of-flight mass spectrometry, seems to be its potential for emerging contaminants discovery, and for doing multiresidue quantification on one platform in place of multiple high resolution mass spectrometers. We are developing these types of applications in our lab using GCxGC–TOFMS.

Buchanan:Methods for the toxic chlorinated analytes (dioxins, furans, coplanar PCBs) include some of the more cutting edge MS instrumentation. The US EPA seems to have moved toward allowing laboratories to use results-based in-house modification to the “black-and-white” SW-846 methods. This is great as it allows the science of separation to be the driving force and allows laboratories to find less expensive processes. However, this puts the burden on the laboratory to show equivalency (or improvement) and may cause challenges if the regulatory agency is not scientifically savvy. I have not seen that the procedure for determining method detection limits (MDLs) (currently based on a statistically derived formula) was updated to a “real-world” process.

Hartlein:Generally the regulatory agencies have spoken of performance based methods for quite some time. We have seen some movement in the newly released drinking water method for VOCs, however the basic guidelines are still strictly enforced. From this perspective we still see them lagging behind the technology. General acceptance of a new technique for any given method can take years in which case the technology may have already exhibited another change.

Two more emerging applications are trace level detection using the TOF and the acceptance of a TOF, which gives classical electron ionization spectra enabling the use of current libraries for compound identification, sample preparation, and introduction of systems, such as thermal desorption.

Ghosh:Tandem MS is definitely the most widely used LC–MS technique in the environmental industry. However, high-resolution, accurate-mass techniques are proving immensely powerful both in terms of unequivocal identification of unknown contaminants, as well as quantification.

Which separations techniques have seen the greatest improvements in instrumentation for environmental analysis?

Bukowski:Ultrahigh-pressure LC (UHPLC) for tandem LC–MS. Modern TOF systems for GC, making ultrafast GC techniques (such as LTM and GCxGC) more amenable to MS.

Cochran:I hate to admit it since I’m a GC guy, but probably LC–MS-MS. The sensitivity of the newer platforms is just tremendous. And high-speed and programmable selected reaction monitoring capability allows multiresidue determinations even for narrow LC peaks.

Buchanan:GC–MS instruments that have the ability to operate with hydrogen (instead of helium) as the GC carrier gas. The benefits are twofold. First, hydrogen is a better GC carrier gas for temperature programmed analyses in terms of efficiency, resulting in faster analyses, sharper peaks, and increased sensitivity. Second, hydrogen can be generated on-site with a gas generator, resulting in a cost reduction compared to using helium in high-pressure cylinders. Higher sensitivity GC–MS systems have made it possible to achieve lower detection levels. In the case of GC–MS volatiles analysis, this has allowed for the use of shorter, narrow-bore columns to decrease run time.



Hartlein:There have been several changes over the years that are noteworthy. Capillary GC columns have made great strides in reducing separation time without sacrificing chromatography. This in turn has led the hardware manufacturers to improve on everything from detector sensitivity to pneumatic control. In our field, purge and trap has been challenged with better water removal systems as well as ultraclean materials that do not exhibit any background. In addition, more selective and reliable trapping materials must be employed to handle the ever-increasing compound additions as well as the automation that is currently available, which taxes the general life of the trap.

Ghosh:Online sample preparation and separation.

Did the recent acetonitrile shortage cause any movement toward “greener” environmental analysis techniques requiring less solvent usage?

Cochran:We saw environmental analysts wanting methanol-based liquid separations (for example, for PAHs) during the worst of the acetonitrile shortage. This actually had another benefit for some separations, as phenyl and biphenyl stationary phases can be more retentive and selective when using methanol versus acetonitrile. The acetonitrile shortage motivated other analysts to go to smaller inside diameter LC columns, or even make the jump to small particle UHPLC, where most of the solvent consumption savings is through much shorter analysis times.

Buchanan:Not directly as acetonitrile is not widely used in environmental laboratories. However, to control costs for bulk solvents (dichloromethane, hexane, acetone) and bulk drying agents (sodium sulfate), supercritical fluid extraction (SFE) is making a comeback as an extraction technique and Gore-Tex-like drying membranes are now on the market. Both situations, while driven by the desire to reduce operating expenses, result in less waste (less solvent evaporated up fume hoods, less used sodium sulfate in the trash) as a positive consequence.



Hartlein:As we are primarily a gas extraction sample preparation company we have not felt any of the effects of the acetonitrile situation. If we were to offer an example it would concern the uses of alternative gases such as nitrogen or hydrogen to helium. We are always being requested to produce more technical data on the alternative gases as the price of helium can be up to three times more expensive than nitrogen, for example.

Ghosh:The recent acetonitrile shortage caused a stronger movement of the separation market toward UHPLC. UHPLC provides dramatically fast analysis times thereby decreasing the amount solvent per sample used for method development and sample analysis. This has also produced a tremendous cost savings on the spending of waste disposal because many solvents have a very high disposal cost.

What is the future of environmental analysis with respect to separations techniques?

Bukowski:A revolution in fast GC column techniques — as seen in the world of liquid chromatography with emergence of UHPLC pumps and column technologies. Miniaturization of the separation system allowing faster analyses and lower detection limits. Ever-reducing LODs, demanding higher instrumental selectivity and sensitivity. Tandem MS is one route; comprehensive chromatography with TOF-MS is another. These technologies complement each other.

Cochran:Ouch, I think I have to answer LC–MS-MS again. The newer systems are so sensitive they open the door for easier sample preparation, sometimes even “dilute and shoot.” And many of the latest environmental contaminants of concern, including some of the brominated flame retardants like hexabromocyclododecane that are thermally sensitive, are best analyzed using liquid separations.

However, there is hope for the GC environmental analyst, as many major contaminants, like PCBs, PAHs, chlorinated dioxins and furans, many pesticides, and so forth, are best done with GC and electron ionization MS. They don’t ionize well in electrospray or atmospheric pressure chemical ionization sources used for LC and you need a bit more peak capacity for congener separations than LC can currently offer. That takes us back to the future of more selective GC columns or GCxGC to do these analyses.

Buchanan:I believe there are many possibilities for the future:
• GC–MS — The migration of analyses from GC to GC-MS should continue. As MS sensitivity increases, the combination of low detection limits and confirmatory identification is hard to resist.
• Fast GC —The mind-set of fast GC will be adopted by the masses.
• Capillary columns with new, alternative selectivity— Looking at the resolution equation, selectivity has the greatest influence.
• MS-MS —The development of GC-MS-MS and LC-MS-MS methodologies for confirmation and background elimination.
• Congeners — The requirement for congener specific identification for polychlorinated biphenyls (PCBs), polybrominated diphenylethers (PBDEs), chlordane, and toxaphene coupled with the need for columns to do these separations.
• GCxGC — As the software catches up to the hardware, this technique should become adopted.
• Chiral chromatography — As the scientific community learns more about the toxicity of chiral pesticides, the requirement to identify and quantify enantiomers should be implemented.

Hartlein:The future of separation techniques for environmental analysis is difficult to forecast. Current methods are constantly being refined to provide fast results to help environmental laboratories approach profitability. In addition the advances in automation have allowed laboratories to utilize single platforms for multiple analyses. I believe this trend will continue in the future and expand to cover myriad techniques beyond the traditional ones currently employed.

Ghosh:Environmental analysis laboratories will be able to increase productivity and efficiency with regard to separation techniques as UHPLC, which is more broadly integrated into their workflows. The ability to run a larger number of samples in a given time dramatically increases productivity, revenue generation, and sample capacity. The actual technique of UHPLC, which utilizes less solvent, enables the laboratories to work more efficiently at lower costs per sample due to savings in the solvent purchasing and disposal.

If you are interested in participating in any upcoming Technology Forums please contact Group Technical Editor Steve Brown or Associate Editor Meg Evans for more information. Next month’s forums will focus on the LC–MS and Homeland Security markets.