Gas Chromatography - Analytical Chemistry (ACS Publications)

---

Anal. Chem. 2002, 74, 2771-2780

Gas Chromatography Gary A. Eiceman*

Department of Chemistry and Biochemistry, New Mexico State University, Las Cruces, New Mexico 88003-0001 Jorge Gardea-Torresdey

Department of Chemistry, University of Texas, El Paso, El Paso, Texas 79968 Ed Overton and Kenneth Carney

Louisiana State University, Baton Rouge, Louisiana 70803 Frank Dorman

Restek Corporation, Bellefonte, Pennsylvania 16823 Review Contents Reviews, Books, and General Interest Columns Principles and Technology Stationary Phases Coating Techniques Models for Separations Fundamental Characterization Deactivation Chemistry Column Fabrication High-Speed and Portable Gas Chromatography Column Efficiencies Fast Temperature Programming Pressure-Flow-Based Techniques Multidimensional Techniques Fast GC/MS Portable GC Detectors Literature Cited

2771 2772 2772 2772 2773 2773 2773 2773 2774 2775 2775 2775 2775 2776 2776 2777 2778

This review of the fundamental developments in gas chromatography (GC) includes articles published from 2000 and 2001 with an occasional reference to work from early 2002. As with the prior recent reviews, emphasis has been given to discussion of selected developments considered particularly significant in the principles and concepts of gas chromatography. Consequently, not every article of merit could be included and fine works certainly have not been cited here for the sake of space. The structure of the review was again modified slightly to reflect those areas of research undergoing substantial interest and development. In an attempt to identify and critically evaluate these topics, a welcomed addition has been made with several investigators who have long-standing expertise in specific areas of chromatography. REVIEWS, BOOKS, AND GENERAL INTEREST A few general comments can be made regarding fundamental activities in gas chromatography during the last two years. As in past decades, the number of applications of gas chromatography far exceeds the number of works concerning basic developments. 10.1021/ac020210p CCC: $22.00 Published on Web 05/02/2002

© 2002 American Chemical Society

This reflects the value that GC continues to bring to chemical measurements as a mature technique. The growth of multidimensional gas chromatography, highlighted in previous reviews as an topic of growing interest, has continued to gain attention and refinement. The complexity of multidimensional GC, in both principles and practices, has undergone improvements through advances in technology. Similarly, high-speed or fast GC will now be available to an increasing number of investigators with commercially available columns. This may be expected to create further attention and fundamental refinements. A small but discernible trend has occurred in the preparation of highly polar stationary phases for capillary columns with high thermal stability. If the methods described are shown to be generally useful, the world of stationary phases, long dormant with the preeminence of highly stable commercial fused-silica columns, could be dramatically transformed. Opportunity and complexities in column selection and preparation could again become part of the landscape of GC separations. During this review cycle, four books on gas chromatography were released. A comprehensive treatment of multidimensional chromatography was published (A1) and as noted above is timely with respect to the growth in this topic. A general book on GC provides a current summary of the method (A2). Two monographs from multiple authors in an edited book appeared on the topic of gas chromatography/mass spectrometry (A3) and gas chromatography/olfactometry (A4). These are welcome additions and updates on two of the most important topics historically in gas chromatography. The same themes that appeared in books and generally throughout the GC literature also could be found in certain review articles. For example, the use of microcolumns for comprehensive two-dimensional separations was treated in an extensive manner (A5). The argument is made that multidimensional GC is clearly a superior method to attain extremely high peak capacity. The article was complete with a discussion of instrumentation, examples, and principles and can serve as an excellent starting paper in the topic. A particularly welcome review was that of chromatographic techniques for petroleum and related products (A6) where Analytical Chemistry, Vol. 74, No. 12, June 15, 2002 2771

GC has been a prominent and essential tool for decades. The role of GC in measurements with natural products was illustrated with review articles on the chromatography of sugars (A7) and optically active materials (A8). In this last review, a useful evaluation was given on molecular interactions, technology, and applications. The balance of treatment and evaluation of scope and limitations should be helpful to beginner and experienced investigator. A review of chiral separations spanning all chromatography methods including GC will be helpful in a comparative analysis of the differing practices of separation (A9). A large number of valuable articles were found of general interest or of value meriting special mention and only a few are included here. An update on statistical-overlap theory was used in support of the necessity of multidimensional GC (A10). Fortunately, progress was made one of the demanding facets of multidimensional GC as noted in prior reviews, namely, data processing (A11). In particular, a general method was provided to correct variations in retention time on both separation dimensions prior to chemometric data analysis algorithms. Such algorithms will further enhance the analysis capabilities of these methods and should provide a long needed systematic method of data handling. Other advances in multidimensional GC were made, and some of these are included in the section below on High-Speed and Portable Gas Chromatography. Clearly, this topic is a next stage in the development of GC and is gaining the foundation necessary to open new capabilities to GC on a generally available basis. Two articles appeared on the general subject of GC in the educational curriculum, and these are noteworthy for both sophistication and contemporary capabilities (A12, A13). In one (A12), the physical chemistry of GC was simulated to provide an experiment in the fundamentals of GC. In another, the connection is made between molecular modeling and GC performance (A13). From these types of experiments, perhaps a next generation of investigators will contribute advances in GC. Finally, an expansive look at GC and current trends in GC methods was given with 139 references (A14) and might be useful for both historic value and analysis of trends in recent past activities. COLUMN PRINCIPLES AND TECHNOLOGY In comparison to some facets of gas chromatography such as instrumental development, few fundamental advances have occurred in column technology for GC. However, considerable effort has been given to the development of specific methods of analysis with existing GC columns. The topic of fundamental advances in columns for GC can be categorized into a few subtopics including the following: stationary-phase development, coating techniques, application of computer modeling to the optimization and understanding of separation mechanisms, fundamental methods of characterization of existing stationary-phase chemistries, and deactivation chemistry. Each of these is given separate discussion below. Stationary Phases. In 1977, Grob (B1) delivered a challenge to the manufacturers and researchers of stationary phases for capillary columns saying that there was a need for “high-polarity” stationary phases other than poly(ethylene glycol)-based (PEG) phases. This statement was only the beginning of this topic since most stationary phases are similar in principles of selectivity. Since 2772

Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

most of the GC columns used analytically, other than the PEG phases, are either a 100% poly(dimethyl siloxane) (-1 phase) or a 5% diphenyl polydimethyl siloxane (-5 phase), stationary-phase chemistry is largely disregarded when optimization of separation is considered. In the period 2000-1, few actual advances have been reported in the design and implementation of other polysiloxanes. Consequently, advances in and exploration of the ultimate power of GC as a separation tool continues to be limited to the compounds that can be separated on the widely available stationary phases, with a few exceptions. Despite the lack of fundamental polysiloxane research for GC applications, several researchers have investigated some novel materials for use as gas chromatographic stationary phases and these publications concern either the use of liquid crystals or the application of cyclodextrins to achieve the separation of optically active compounds or isomers. The motivation is that liquid crystals or cyclodextrins may permit a specific separation that is limited on commercially available columns. For example, the difficult separation of 2,3,7,8-tetrachlorodibenzodioxin has been approached using a side-chain liquid crystal polysiloxane and separation was improved for some of the most toxic dioxin congeners as compared to the conventional “-5% phenyl” phases (B2). Another approach to add specificity to stationary phases is that of liquid crystals that impart a steric component to resolution. For example, a novel copper complex-containing siloxane polymer was demonstrated to improve separation of a series of phthalate esters as compared to conventional columns (B3). This material behaved as a liquid crystal and contributed a shape selectivity that conventional columns do not have. As with other liquid crystals, temperature is a limitation with comparatively low temperatures for maximum temperatures of operation. Coating Techniques. Possibly the most significant development over the past few years for capillary GC coating techniques is the introduction of sol-gel column manufacturing. A manufacturing process for sol-gel with capillary columns has been patented and demonstrated in several successful applications of this process (B4). A discussion of the development of sol-gel technology as applied to the manufacturing of capillary GC columns has been described (B5). The selectivity of retention can be modified to enhance the separation of target analytes by manipulating the gelation and processing conditions. Some of these materials have quite high thermal stability, so porous layers can be cast in stainless steel tubing for analysis of hydrocarbons well beyond the range of conventional PLOT columns. Success has occurred in forming gels of comparatively high polarity with the use of bridging structures. These columns can show waxlike selectivity, being able to separate p- and m-xylene. Other solgel-based stationary phases for capillary GC columns have been described (B6). One particular advantage of these new sol-gel phases has been their availability commercially as PDMS-like and waxlike columns with improved thermal stability over the conventional equivalents. Stabilizing stationary phases, once a major theme in packedcolumn technology, became nonexistant with nonpolar bonded phases. However, stability has remained an issue with polar phases and nanosized particles were employed in one approach to stabilize polar stationary phases (B7). This method may allow the use of more packed-column stationary phases for capillary applications

and has been demonstrated with some very polar stationary phases (e.g., SP2340, OV-275, and others). The technique was found to improve the stability of the stationary phase and yield high efficiency. If widely applicable, this relatively simple and inexpensive technique may permit an expansion of capillary separations to the more than 1000 or so packed-column stationary phases developed decades ago for packed-column technology. If this materializes, a dramatic change may be anticipated in both applied and fundamental developments in gas chromatography. Models for Separations. Even though mathematical descriptions of the separation process have long existed (B8), researchers have only recently had the computational ability to investigate the mechanisms of retention of a molecule by a stationary phase. Three approaches to such modeling exist, and all differ subtantially in general approach and details. These include the following: quantitative structure-activity relationship (QSAR), molecular dynamics (MD), and quantum mechanics (QM). In QSAR, initial success was attained with predictions for homologous series, but there has been little improvement of this technique for GC over the past few years. The MD approach has been successful in the prediction of retention of a molecule on a phase, without empirical input to adjust or correct models (B9). This approach could allow for the prediction and optimization of separations without needing to perform laboratory work and could also be used for determining the best functionalities required for specific separations. A first description of using quantum mechanical techniques to understand the selectivity for various analytes exhibited by PEG phases has been reported (B10). In this technique, the molecular orbitals of the analyte and the stationary phase are calculated, and the forces that determine selectivity of a stationary phase for a certain molecule can be understood. This research has demonstrated that this approach can be used to determine elution order of a series of aliphatic and alcohol molecules and can also be useful for the determination of novel stationary phases to achieve specific selectivities. Others have attempted to enhance separations with analysis not easily resolved through the use of several different commercially available GC columns in order to aid the choice of which stationary phase to use for particular separations. Notably, the relative abundances of the congeners in the Aroclor mixes closely follow the predictions from the theory of electrophilic chlorination of aromatic nuclei which has been demonstrated as the best separation of 209 PCB congeners (B11). The importance of molecular geometry has been illustrated in the separation of a series of N-TFA-O-alkyl amino acid derivatives using β- and γ-cyclodextrin GC phases (B12). Specifically, as separation was decreased on the β-cyclodextrin columns for increasing alkyl chain length, the separation on the γ-cyclodextrin phase was improved. In general, the separation of the enantiomers of N-TFA-O-Me amino acid esters was better on the larger-cavity γ--cyclodextrin phases, except for the esters of alanine, which were better separated on the β-cyclodextrin. Another fundamental advance in separations was an unexpected development where separation was enhanced through a stop-flow GC technique (B13). In this, two columns were placed in series with a connection at the column junction to allow the flow to be paused on the first column and accelerated on the second. While this is an instrumental solution, it is worth

mentioning in the GC columns section, as it allows for the optimization of separations by using two columns of different selectivity and then essentially flow programming the column ensemble to “tune” the selectivity of the column pair as the compounds elute. Fundamental Characterization. Characterization of commercially available stationary phases can lead to an understanding of the mechanisms of retention on these polymers. In analytical separations, the selection of the most selective stationary phase for difficult separations is best made when the physical interaction of the analyte with the stationary phase is understood and applied. During the past two years, a valuable assessment has been made on the selectivity of many commercially available stationary phases (B14-B16). Through fundamental characterization of these phases using a series of test analytes, the necessary data are provided to determine which stationary phase would be best for certain separations. Additional modes of selectivity for each stationary phase were identified, and the predominant selectivity for a phase was described. Deactivation Chemistry. Deactivation of capillary tubing for improved wetting is an essential step prior to coating the stationary phase for most GC columns. Two new chemistries have been reported and patented within the past few years that improve tubing surface deactivation for metal and fused-silica columns. Deactivation of metal tubing was accomplished by coating a layer of polypyrrone on the inside surface of the tubing prior to coating the stationary phase (B17). Since a metal tube does not have the limitation in thermal stability of polyimides found as protective layers on fused-silica columns, such tubing might be operated at temperatures higher than customarily expected with modern commercial columns. Another deactivation method, termed “Siltek”, was described and involves a chemical vapor deposition process (B18, B19) in contrast to the liquid-based methods found in contemporary chlorosilane or silazane coupling chemistry. Several commercially available stationary phases have been shown to wet the Siltek surface, and the material has also found use in deactivation of guard tubing; this approach to protecting analytical columns was effective for the analysis of chlorinated pesticides, semivolatiles, organic bases, and explosives. Column Fabrication. In this last section of Column Fundamentals and Technology, mention should be made of progress in the creation of columns on etched silicon for columns on a chip. The production of gas chromatography microdevices offers the prospect of an unprecedented change in the nature of GC, particularly in miniature GC and high-speed GC as desribed in the next section. The pocket GC has been an object of desire since the first micromachined chromatographic components were described in the early 1980s. Because of the difficulty in producing prototype or concept demonstrations of chromatographic microdevices, a great deal of work has been done modeling the performance of these systems prior to production of actual components. To address this issue, a process to facilitate the rapid design and production of masks was developed (B20). At present, the great majority of work on microscale separation devices revolves around the development of capillary electrophoresis or electrokinetic chromatography and other liquid-based devices. Nevertheless, the modeling and theoretical concepts related to microchannels can Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

2773

provide useful insights into the problems of scaling down GC systems. Two papers specifically addressing gas flow in microchannels (B21, B22) discussed the importance of compressibility in microchannels. Additionally, a good discussion of efficiency issues related to channels having rectangular cross sections was provided in the former article (B21). A number of articles concerned with incompressible fluids also are useful to developing an understanding of the scale-down issues related to separation processes in microdevices. One of the basic problems, dispersion that arises when the column is designed with very tight curves to put long channels on small microchips, was discussed (B23). The issue was discussed in terms of electrokinetic (electric fielddriven) flows, but other workers extended the discussion to include a comparison between electrokinetic and pressure-driven systems, albeit still for incompressible fluids (B24). One of the few working prototypes of this new generation of microdevices for gas chromatography was described (B25). While the chromatograms may look less than optimal to the chromatographer’s eyesthe peaks are broad and the peak capacity is lows the chromatograms can elute in only a couple of seconds, so the resolution rate (resolvable peaks per second) is high. This microGC takes advantage of all the multidimensional techniques such as using selective sample accumulation, chromatographic separation, and selective detection to produce what has been called the “microChemLab”. Two significant patents were issued within the last couple of years. One patent described certain microGC ensembles that could lead to a small device using multiple GCs operating in parallel with thermal conductivity detection (B26). It appears that the novel aspect is the potential for stacking a number of complete microfabricated GCs into a single package. Another patent for high aspect ratio columns described a column with the major axis of the cross section orthogonal to the curvature radius in microfabricated columns rather than parallel as has been the prior form (B27). When microfabricated columns are coiled onto a small microdevice measuring only a few centimeters, the radius of curvature is no longer much larger than the column diameter. Thus, molecules traversing the turn near the inner radius of the curve travel a shorter distance than molecules traversing the turn near the outer radius of the curve. This socalled “racetrack” effect leads to additional dispersion beyond that normally anticipated in GC. These “tall and thin” columns are thought to provide an advantage over “short and wide” microcolumns in reducing the dispersion associated with the differing path lengths between the inner and outer radii of the curves on the microdevice while retaining the resolution and sample capacity advantages of having relatively large stationary-phase surfaces with small column diameters. Another silicon wafer-based GC column that would be used for remote monitoring and other small-size, low-power applications has been described (B28). HIGH-SPEED AND PORTABLE GAS CHROMATOGRAPHY The trend in growing interest in high-speed gas chromatography over the past six to eight years was evident during this past review cycle with specific renewed interest in some of the fundamental relationships governing the transport, or rapid elution, of analytes through GC columns. These fundamental interests are driven by practical advantages of high-speed GC including the favorable economics of improved sample throughput 2774

Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

and a widening interest in portable GC applications; these find expression in chromatographic analyses with on-site, in-field, or extralaboratory venues. An essential question is “how fast is fast enough?” The answer to this question bears on all processes, direct and ancillary, related to a chromatographic analysis. These issues were discussed vis-a`-vis the concurrent development of high-speed sample preparation and data processing methods (C1). Though technical developments usually govern the level of inquiry into an instrumental experiment, the ramifacations for high-speed GC are exceptional and these components to a GC measurement, other than the column, will need advances for full use of the capabilities of available instruments. High-speed gas chromatography and portable gas chromatography are not synonymous though substantial areas of overlap can occur. The usefulness of portable GC applications derives largely from the short delay between sample collection and sample analysis. Additionally, elimination of sample transport and storage are attractions in portable GC; however, these separations need not be necessarily high-speed chromatography. Developments toward high-speed or fast gas chromatography are almost always consistent with the latter. The domain of high-speed or fast GC spans a range of topics from open tubular column separations that are fast only in comparison with packed-column counterparts to the other extreme, employing millisecond separations of a few compounds. Just as for the definition of a fast computer, the definition of fast GC is and will also continue to change. Separations on open tubular columns that were thought fast with the start of the replacement of packed columns by open tubular columns are now standard practice. The current convention for chromatographic speed is the benchtop GC performing separations with 15-30-m open tubular columns having inner diameters of roughly 0.25 mm and using average carrier gas velocities between 30 and 120 cm/s. Systems that perform somewhat equivalent separations in less than, say, 20% of the time of this conventional system can justifiably be considered a fast GC system. Finally, fast GC is not necessarily high-resolution GC. Some of the fastest chromatograms have peaks capacities on the order of 10, but when the separation occurs in tens to hundreds of milliseconds, they certainly qualify as fast GC. Fundamental advances in high-speed gas chromatography developments can be categorized as improved: mass transport efficiency in the column, column heating efficiency allowing higher temperature programming rates, sample introduction systems designed to load narrower bands onto the column, multidimensional techniques, and hyphenated techniques that use selective detection (notably MS). Often, compromises are made between resolution and resolution rate (peak capacity per unit time). The goal of fast GC development, then, is to produce the highest peak capacity per unit time, within the constraint that the resolution of the system is adequate to solve the analytical problem at hand. Substantial effort is also underway in the area of microfabricating chemical separation devices. While the vast majority of work on chemical separation microdevices is geared toward capillary electrophoresis applications, the work addresses flow and (size) scaling issues that are pertinent to microscale gas chromatography. It seems that to date none of the work on microfabrication has produced high-quality chromatography on a very short time scale. Apparently the most successful efforts in microfabricating

a gas chromatograph have been the relatively low resolution chromatograms obtained with high aspect ratio, rectangularly etched columns, and surface acoustic wave-based detectors (C2). Column Efficiencies. A fundamental limitation to the separation speed is imposed by the inherent efficiency of mass transport in the column. Because ultramicrobore columns yield more efficient mass transport than larger bore columns, equivalent separations can be achieved with shorter columns and in much less time. A well-known disadvantage of small-diameter columns is a smaller sample capacity that leads to column overloading and high demands on the detector. One attempt to overcome the small sample capacity of ultramicrobore columns has been to bundle them and essentially perform upward of 1000 separations side by side. In practice, the theoretical benefits of this approach have proven to be extremely difficult to achieve and some of the issues related to implementing this approach have been revisted (C3). Also, somewhat surprisingly, for simple mixtures of analytes with low capacity factors, efficient packed columns may be superior for high-speed GC than equivalent open tubular columns (C4). The concept of using solvating mobile phases with these packed columns was also presented, usefully blurring the distinction between gas and supercritical fluid chromatography (C4). Vacuum outlet GC has long been known to provide higher column efficiencies and has been discussed mostly in the context of GC/ MS. While the vacuum systems required in MS systems make vacuum outlet GC a natural fit for GC/MS, there is no reason to limit its use to GC/MS, though perhaps with less demanding vacuum requirements. Vacuum outlet GC has been investigated specifically in the context of high-speed GC. Peak widths of 100 ms were obtained with a 0.53-mm-i.d. column using low-pressure conditions in the column (C5). In order to maximize throughput, high-speed GC applications often sacrifice resolution. One must ensure, however, that the system is never underresolved. The thermodynamic origins of retention time distributions were modeled in the kind of fundamental study of retention behavior that will become more significant as specialized fast GC methods pare resolution to the minimum safe level in order to maximize sample throughput (C6). Fast Temperature Programming. All other things being equal, one of the most straightforward ways to increase the resolution rate is to use higher temperature programming rates. The history of oven development was summarized up to and including the ThermoOrion EZ-Flash high-speed GC accessory (C7). The EZ Flash is an excellent example of the divergence between the implementation of fast GC and the implementation of portable GC. An add-on to conventional benchtop instruments, the EZ Flash has an independent power supply for the column heater. Implementation of fast GC on conventional gas chromatographs (i.e., forced air ovens) is without exception a powerintensive operation. Additionally, the cooldown times for fast GC with conventional ovens can significantly exceed the time required to elute the chromatogram. Possible designs for heating techniques specific to fast GC were discussed (C8). Also a low thermal mass conductive heating system for fast temperature programmingsup to 250 °C/sshas been implemented for short capillary columns, thus reducing analysis time by an order of magnitude when compared with conventional analyses (C9).

The successful use of high heating rates involves more than adding extra power for the heating element. Excessive heating rates can waste the downstream portions of the column for early eluters and can push elution temperatures to the upper temperature limits of the column for late eluters, thus losing the advantages of temperature programming. An empirical description of the effect of high heating rates on peak capacity, resolution rates, and resolution was given (C10), and a theoretical basis was discussed in terms of dimensionless heating rates for high-speed temperature programming (C11, C12). In this concept of a dimensionless heating rate, the column dead time, to, is the fundamental time unit. In other words, heating rates are expressed in degrees per to rather than in seconds or minutes. Normalizing heating rates in this way facilitates the comparison of and translation of methods between conventional systems that use longer columns and standard carrier flow rates and high-speed systems that typically use shorter columns and higher carrier flow rates. Pressure-Flow-Based Techniques. One of the earliest methods of increasing chromatographic throughput involved the use of carrier gas flow rates higher than that suggested by the mimimums in Golay plots of plate height versus carrier flow. This concept was clarified in the context of high-speed temperature programming and varying column lengths (C11, C12). Pressure and flow have been used in a unique fashion to fine-tune the selectivity of sequentially connected columns. By independently controlling the flow rate through each of two sequential columns of differing polarity, one is able to control the relative residence time of analytes in each of the two columns. Varying the relative flows through the two columns alters the effective polarity of the ensemble. Furthermore, the flow rates can be varied in real time during a chromatographic analysis, allowing the polarity of the ensemble to be fine-tuned while the separation is in progress. Despite the seeming technical difficulty of applying this technique, these so-called “tunable ensembles” have been used with vacuum outlet GC using air as a carrier (C13), with air carrier systems and surface acoustic wave (SAW) detectors (C14), with time-offlight mass spectrometric (TOFMS) detection (C15), and with temperature programming (C16). Multidimensional Techniques. The tunable ensemble (C12C15) is in essence two-dimensional chromatography with detection in only one dimension, a composite of the two separation dimensions that is continually variable. The technique of “thermal modulation” has subsequently come to be referred to as “comprehensive two-dimensional chromatography”. Comprehensive chromatography amounts to nearly continuous heart cutting. The effluent from a more or less conventional primary column is thermally modulated such that it is refocused at the head of a very short secondary column which then is heated at very high rates to produce “microchromatograms”. The thermal modulation cycle is very fast compared to the elution from the primary column; five or more microchromatograms are collected over the time that a typical peak would elute from the primary column. The result is a two-dimensional separation with one separation axis having a relatively long time scale (minutes) and one having a very short time scale (milliseconds). Even though the total time for “comprehensive” analyses is often not greatly shorter than for conventional analysis, the peak capacity per unit time can be Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

2775

enormous. A notable feature is that homologous series tend to display linear patterns in the two-dimensional output. This twodimensional techniques has been extended to three dimensions by using two thermal modulators instead of one (C17). An implementation of comprehensive two-dimensional chromatography using a multiport valve rather than thermal modulation has been developed (C18). This technique generated secondary chromatograms 1 s in length with peak widths (at halfheight) on the order of 50 ms, comparable to thermal modulation. Comprehensive two-dimensional chromatography via thermal modulation is, in practice, difficult to implement, although considerable development efforts of Zoex Corp. seem to have had some success. Implementation via multiport valves seems simpler, and more straightforward, given the availability of sufficiently fast and reliable multiport valves. The recent advent of a commercially available multiport diaphragm valve may signal more widespread access to this technique. The direct implementation of comprehensive multidimensional chromatography requires detector systems that are capable of providing the high data sampling rates (200 points/s or higher) to accurately characterize the short secondary chromatograms. Most commercially available benchtop GCs lack this capability. A potential problem with comprehensive chromatography is common to many fast GC implementations, namely, the smaller sample capacity of the small-diameter, thinfilm columns often used for fast GC. The small sample capacity of these columns makes column overload, and hence nonlinear partitioning behavior, more likely. The effect of nonlinear partitioning on resolution in comprehensive chromatography was discussed and modeled (C19). The focusing of bands eluting from the primary column as they are collected by thermal modulation on the secondary column was modeled (C20). The developed model provides a guide to understanding and predicting detection limit enhancements and column overload potential in comprehensive GC. One of the most significant changes brought about by fast GC techniques is that the benefits of using advanced data processing techniques may be more compelling when chromatograms are acquired in seconds rather than minutes. The use chemometric techniques to extract information from multidimensional fast GC data was explored in several contexts (C21-C23). The techniques were applied to actual output from comprehensive two-dimensional GC analyses. It is noteworthy that these two applications consider problems such as retention time variations that are a normal part of any chromatographic and are often more significant when the time scales are very short. Fast GC/MS. The use of selective detection provides an additional dimension to increase chromatographic throughput. By shifting some of the load of resolving components from the separation system to the detector, the speed of the separation step can be increased simply by sacrificing resolution. This is one of the oldest tradoffs in chromatography, speed versus resolution. While various detectors such as coated and uncoated SAW devices have been presented in the past, the detector currently creating the greatest interest is the TOFMS. The wealth of information in GC/MS data is clear. Until recently, the state of the art in massselective detectors was represented by quadrupole instruments (both mass filters and ion traps). The speed of the quadrupole instruments was marginal for use in high-speed GC applications. 2776

Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

Quadrupoles can acquire mass spectra at rates on the order of 10 spectra/s even under the most favorable conditons. New TOF instruments can acquire wide-range spectra (>500 mass units) at speeds of up to several hundred spectra per second. Thus, these new instruments can deliver mass spectra at a rate sufficient to characterize analyte bands eluting from fast GC columns. As a result, a number of workers investigated the use of TOFMS as a detector system for fast GC applications (C24-C26). Portable GC. The benefits of the above developments in fast GC have yet to reach most of the users of portable gas chromatography. Some of the most widely used portable GC’s are the Photovac models which, though very fine units, can still require on the order of 15-30 min for even modestly difficult separations. A recent comparison of the Photovac technology with a nonchromatographic sensor technology showed that the GC technology still has advantages (C27). The 1980s microchip GC is now represented, in modestly updated form, by portable GCs from Agilent and Varian Corp.. Though the relatively high detection limits offered by these TCD-based instruments pose difficulties now as ever, they still find effective use for portable applications as does a field-portable GC/MS for volatile organic compounds that has been available for some time (C28). New developments in portable GC include new developments in field-portable GC/MS. The overhead associated with mass spectrometers generally stretches the meaning of the term portable, but developments in MS hardware continue to make field use of mass spectrometry more viable. A review of the use of fieldportable GC/MS (C29) included mobile mounted instruments and a discussion of ruggedness requirements. A development related to portability more than fast GC is the cylinder-free GC (C30). The system comprises a GC-FID approach modified with a simple onboard electrolysis system. A unique feature of this approach is that the hydrogen and oxygen are not separated and the mixture is used as both carrier and FID fuel/oxidant. The major logistical difficulty with portable gas chromatography is providing a gas supply for the column and detectors, a difficulty that is greatly aggravated when one of the gases is flammable. This has led to efforts by others to use air as a carrier gas, and while at first glance it may seem impractical, using the hydrogen-oxygen mixture is probably not much harder on the column. As has been reported for air carrier systems, the column temperature limits were reduced when the oxygen-containing carrier gas was used. Among the air carrier systems under development is a system that uses sample adsorbent-based sample preconcentration as an inlet to a pressure-tuned column ensemble and SAW array detection (C31). The system has demonstrated parts per billion detection limits for 20 common volatile organics having vapor pressures ranging from 8 to 231 Torr. In terms of complete field-ready instrumentation, a portable GC developed over the past several years is just now becoming available (C32). This portable dual column GCFID has many of the features of laboratory GCs: heated inlets, temperature programming, and electronic pressure control. It also has features desirable in a portable instrument, small size, low gas and power consumption, and high sample throughput. Rapid sample preparation techniques such as the SPME techniques to complement fast, portable GC instrumentation received continued development (C33). A more recent alternative to SPME is the use of sol-gel-formed sorbents in micropacked

traps (C34). Fast sample preparation methods will continue to be an important component in the growth of portable and high-speed GC techniques. In some cases, some or all of the sample preparation may be incorporated into the instrument (C32). In other cases, external, but rapid and facile sample preparation techniques such as the SPME or sol-gel adsorbents will provide the proper approach. As more and more complex sample preparation issues are addressed, the utility of high-speed GC applications will continue to expand. This purpose of this section has been to provide a brief overview of the current work in portable and high-speed GC and not an exhaustive literature review. After a frenzy of developmental activity over the past two years, the field has settled into a relatively more focused period with significant effort by many of the major researchers to move the developed technologies from the laboratory/developmental stages into more finished and widely available products. DETECTORS Detectors provide integral value in the analytical GC instrument, affecting strongly the information gleaned from a GC experiment and the required performance for the entire system. As in past review cycles, the number of applications highlighting the merits of detectors is massively greater than those involving fundamental advances either in detector technology and principles or in the integration of the detector with the remaining chromatographic components. A listing of the main detectors used in the recent past is instructive and discloses the breath of activity in GC detectors; these include the following: flame ionization detector (FID), electron capture detector (ECD), photoionization detector (PID), Fourier transform infrared absorption (FT-IR), helium pulsed-discharge photoionization detector (He-PDPID), flame photometric detector (FPD), atomic emission detector (AED), nitrogen- and phosphorus-specific thermionic detector (TID), glow discharge detector (GDD), microwave-induced plasma detector (MIPD), quartz crystal microbalance detector (QCMD), electroconductivity detector (ELCD), thermal conductivity detector (TCD), electroantennographic detector (EAD), mass spectrometry (MS), ion mobility spectrometry (IMS), atomic fluorescence spectrometry (AFS), inductively coupled plasma spectrometry (ICP), atomic absorption spectrometry (AAS), and ICP-mass spectrometry (ICPMS). One of the trends of the past review cycle is the combination of two or more detectors for enhanced analytical specificity or selectivity. These include reports on PID/ FID (D1), ECD/TID (D2-D4), ECD/AED/MS (D5), and MS/ FPD (D6). Quenching of luminescing species by coeluting hydrocarbons is a known complication in the FPD, and a novel method of investigating the chemical behavior of both analyte and quencher molecules in the FPD has been described (D7). In this report, collection of the effluent demonstrated that 2-82% of various organic compounds may survive passage through the diffusion flame and be recovered intact; the recovery unchanged of several model hydrocarbons was found to decrease with increasing carbon number. Heteroatoms such as sulfur, nitrogen, or oxygen greatly decrease the recovery of molecules relative to their pure hydrocarbon analogues. In another advance in detector principles, the response mechanism of a thermionic detector with enhanced

nitrogen selectivity was described (D8). According to the accepted theory, the analyte has to contain electronegative functional groups in order for negative ions to be formed by the extraction of electrons from the thermionic source. This leads to a selective detector response for compounds containing nitro groups or multiple halogens. However, in the tests described in this report, polycyclic aromatic nitrogen hydrocarbons, acridines, and carbazoles were used as reference chemicals for response and these compounds contain no electroactive functional groups. None of investigated acridines exhibited any response from the detector, but carbazoles generated a strong structure-related detector response. It was demonstrated that the acidic hydrogen atom attached to the nitrogen heteroatom of the carbazoles has a strong influence on the detector response. Ionization of carbazoles may occur by dissociation of the nitrogen-hydrogen bond during contact with the thermionic surface. Support of this theory was provided by the linear relationship between the relative detector response and the deprotonation energy of the carbazoles. Further, there appeared to be no linear relationship between the detector response and electron affinity of the carbazoles. Thus, the mechanism involved in ionization of the carbazoles is probably not direct electron transfer from the thermionic surface to the carbazoles. In addition, principal component analysis showed that the thermal conductivity of the chemically inert detector gases has an influence on the detector response. It was found that the thermal conductivity can be used to rank the detector response of the carbazoles, and there was no discernible response when helium, which has the highest thermal conductivity, was used as the detector gas. In another report, photon yields were determined for several elements in the flame photometric detector (D9). Photon yieldss the number of photons generated per analyte atomsare of great mechanistic importance in flame chemiluminescence. This study showed that sulfur and phosphorus were the strong emitters, followed by manganese and ruthenium. Iron and selenium produced emitters of intermediate strength. These results provided a database for kinetic purposes, and the determined photon yields serve at least as mechanistic limits. The flame ionization detector has long been the means by which the percent composition of a hydrocarbon mixture has been determined since it was previously established as a “carbon counting device”. However, its ionization mechanisms are not well understood. A report appeared that compared the FID with a He-PDPID, and thus, a mechanistic ionization behavior of the FID could be achieved (D10). Although no fundamental ionization explanation was reached, it was shown that the He-PDPID is more accurate than the FID in determining the percent composition of a hydrocarbon mixture. This detector is also more effective for determining the percent composition of mixtures containing organic compounds with a variety of other functional groups. Some novel detector designs were reported during the past two years including a chlorine-selective pulsed discharge emission detector (CI-PDED) for GC (D11, D12). This detector is based on a reaction of krypton with chlorine and a unique detector design. A krypton ion produced in the krypton-doped helium pulsed discharge reacts with chlorinated compounds within the pulsed discharge to produce an excited species of KrCl* which emits at 221-222 nm. The reaction has the following advantages Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

2777

in respect to the detection of chlorinated compounds: (1) the reaction is an ion-molecule reaction that is 100-1000 times faster than a reaction of neutrals, which greatly enhances the sensitivity; (2) the KrCl* emission wavelength is far separated from interfering carbon emissions at 193 and 247.3 nm; (3) the KrCl* emission is transparent to air and can be recorded without a helium purge of the monochromator. This new CI-PDED is the most sensitive chlorine-selective detector with a minimum detectability of ∼50 fg of Cl/s. Another advance in detectors for GC was seen in a PDED that was coupled directly with a vacuum UV monochromator so that the vacuum UV atomic emissions from Cl, Br, I, and S could be observed (D13). In a related development, a halogen-specific detection method (XSD) for the determination of chlorinated fatty acids was reported (D14). Throughout the development of AED for GC, high performance has been attained for many elements but nitrogen detection has been characterized by high background levels; thus, research in this area is still active. Progress toward this purpose was accomplished using the cyanogen (CN) molecular band at 388 nm, instead of the 174-nm atomic nitrogen emission line (D15). Examination of the response for the CN 388-nm line showed it to be more than 100 times more sensitive than for the atomic 174-nm line and 100 times more selective. Others reported the use of the change in the oscillation frequency of the current of a novel atmospheric pressure helium glow discharge detector for capillary GC (D16). In this study, the effluent of a capillary column was directed into the glow discharge cell perpendicular to the axis of the glow discharge that existed between a platinum anode and cathode. A stable discharge was obtained when several hundred volts were applied between the 0.2-mm gap between the anode and cathode. The results of this study showed that the discharge current and discharge gap have a strong influence on the detector response. The discharge current showed positive peaks; however, frequency peaks are positive or negative depending on the discharge conditions. In another report, a high electric field, radio frequency ion mobility spectrometry (RF-IMS) analyzer was used as a small detector in GC separations of mixtures of volatile organic compounds including alcohols, aldehydes, esters, ethers, pheromones, and other chemical attractants for insects (D17). In this study, they found that the RFIMS scans were characteristic of a compound and provided a second dimension of chemical identity to chromatographic retention adding specificity in instances of coelution. The separations of pheromones and chemical attractants for insects illustrated the distinct patterns obtained for GC with RF-IMS scans in real time and suggest an analytical utility of the RF-IMS as a small, advanced detector for on-site GCs. A report appeared during this review cycle where a micromachined plasma chip is coupled to a conventional GC to investigate its performance as an optical emission detector (D18). This device employs a 180-nL plasma chamber in which an atmospheric pressure dc glow discharge is generated in helium. A number of organic compounds were detected in the column effluent by recording the emission at 519 nm. However, the detector signal showed a marked peak broadening and tailing when compared with the signal of a flame ionization detector. This was mainly attributed to dead volume and chromatographic processes introduced by the connecting tubing and the chip glass channels. 2778 Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

Gary A. Eiceman is Professor of Chemistry at New Mexico State University where he joined the faculty in 1980 after a postdoctoral fellowship at University of Waterloo (Canada) and graduate studies at University of Colorado. His research interests include detectors for gas chromatography such as the ion mobility spectrometer, which has been a research interest for over two decades. He is author of over 135 research articles and book chapters on gas chromatography, ion mobility spectrometry, and environmental measurements and is coauthor with Zeev Karpas of Ion Mobility Spectrometry (CRC Press). He lectures in undergraduate and graduate courses in general, quantitative, and instrumentation chemistry. Jorge Gardea-Torresdey is the Richard M. & Frances M. Dudley Professor of Chemistry in the Department of Chemistry at The University of Texas at El Paso in El Paso, TX. He received his Ph.D. in 1988 at New Mexico State University in Las Cruces, NM. His research interests include environmental chemistry of hazardous heavy metals and organic compounds, gas chromatography, gas chromatography/mass spectroscopy, atomic absorption and emission spectroscopy, inductively coupled plasma/ mass spectroscopy, X-ray absorption spectroscopy, and investigation of metal binding to biological systems for remediation of contaminated waters and soils (e.g., phytoremediation). Last year, he received the 2001 University of Texas-El Paso’s Distinguished Award for Research. He has authored or coauthored over 130 research articles and book chapters. He has taught analytical chemistry and instrumental analysis at the undergraduate level and advanced analytical chemistry and environmental chemistry at the graduate level. Ed Overton is an analytical environmental chemist, Professor in the Department of Environmental Studies, and Adjunct Professor in the Department of Chemistry at Louisiana State University. His research interests include studying the movement and impact of dangerous chemicals in the environment and is currently focused primarily on the detection of these types of compounds at the point of sample collection. He has developed and patented several technologies that allow small, fast gas chromatographic instruments to perform analyses on site and is actively involved in converting these technologies into commercially available instruments for use in numerous applications benefiting from small size and short analysis times. Kenneth Carney is a senior research associate at the Department of Environmental Studies at Louisiana State University. He has worked in the area of instrument development for over 17 years, more than 10 years focused on high-speed gas chromatography techniques. His research interests include the development of instrumentation and methodologies for chemical analysis in unconventional settings, the physical chemistry of chemical analysis, and the application of systems analysis to analytical chemistry. Frank Dorman received his B.S. degree in chemistry from Juniata College in Huntingdon, PA, in 1987 and his Ph.D. in analytical chemistry from the University of Vermont in Burlington, VT, in 1992. He was then employed by Inchcape Testing Services-Environmental Laboratories in several roles, eventually becoming Senior Chemist for methods development. Since leaving the Environmental Laboratory in December 1996, he has worked for Restek Corporation in several capacities, most recently as the Director of Technical Development. His research has been in the area of the development of new capillary column stationary phases through the use of computer modeling and development of applications for emerging separation needs.

LITERATURE CITED REVIEWS, BOOKS, AND GENERAL INTEREST (A1) Mondello, L.; Lewis, A.; Bartle, K. D. Multidimensional Chromatography; John Wiley and Sons: New York, 2002. (A2) Handley, A. J., Adlard, E.dward R., Eds. Gas Chromatographic Techniques and Applications; CRC Press: Boca Raton, FL, 2001. (A3) Niessen, W. M. A., Ed. Current practice of gas chromatographymass spectrometry; Chromatographic Sciences Series 86; Marcel Dekker: New York, 2001. (A4) Leland, J. V., Ed. Gas chromatography-olfactometry: the state of the art; ACS Symposium Series 782; American Chemical Society: Washington, DC, 2001. (A5) Liu, Z. Y.; Lee, M. L. J. Microcolumn Sep. 2000, 12 (4), 241254. (A6) Barman, B. N.; Cebolla, V. L.; Membrado, L. Crit. Rev. Anal. Chem. 2000, 30 (2-3), 75-120. (A7) MolnarPerl, I. J. Chromatogr., A 2000, 891 (1), 1-32. (A8) Schurig, V. J. Chromatogr., A 2001, 906 (1-2), 275-299 (A9) Gubitz, G.; Schmid, M. G. Biopharm. Drug Dispos. 2001, 22 (7-8), 291-336. (A10) Davis, J. M.; Samuel, C. HRC-J. High Resolut. Chromatogr. 2000, 23 (3), 235-244. (A11) Fraga, C. G.; Prazen, B. J.; Synovec, R. E. Anal. Chem. 2001, 73 (24), 5833-5840. (A12) Haigh, J.; Lord, J.; R. J. Chem. Educ. 2000, 77 (11), 15281528.

(A13) Simpson, J. M.; Rivera, O. J. Chem. Educ. 2001, 78 (7), 942943. (A14) P. Yashin YI, Yashin AY J. Anal. Chem. 2001, 56 (3), 200213. COLUMN PRINCIPLES AND TECHNOLOGY (B1) Grob, K. Chromatographia 1977, 10, 625. (B2) Chang-Chien, G.-P.; Lee, W.-S.; Tsai, J.-L.; Jeng, S.-H. J. Chromatogr., A 2001, 932, 97-105. (B3) Chuen-Ying, L.; Yang, S.-H.; Chau, M.-H.; Shiue, C.hia-C. J. Chromatogr., A 2001, 933, 117-128. (B4) Malik, A.; Wang, D., World patent 2000011463, 2001. (B5) Lautamo, R. M.; Archer, J. O. Sol-Gel Derived Capillary Columns for Gas Chromatography: Extending the Range of PLOT Column Technology. Presented at The Pittsburgh Conference, New Orleans, LA, 2000. (B6) Cervini, R.; Day, G.; Hibberd, A.; Sharp, G. Chem. Aust. 2001, 68 (4), 16-18. (B7) Tolnai, G.; Alexander, G.; Horvolgyi, Z.; Juvancz, Z.; Dallos, A. Chromatographia 2001, 53, 69-75. (B8) Giddings, J. C. Unified Separation Science; John Wiley & Sons: New York, 1991. (B9) Dorman, F. L.; Schettler, P. D.; English, C. M. Development of Application Specific Stationary Phases for HRGC Using Computer Modeling. Presented at the 24th International Symposium on Capillary Chromatography, Las Vegas, NV, May 20-24, 2001. (B10) Holmes, H.; Trease, N. M.; Pratt, M.ichael E.; Horvath, T. D. Modeling Chromatographic Retention Using Ab Initio Molecular Orbital Calculations to Determine Nonbonded Interaction Energies. Presented at the 24th International Symposium on Capillary Chromatography, Las Vegas, NV, May 20-24, 2001. (B11) Frame, G. M. Optimization of Columns and Detectors for Profiling Aroclor PCB Congener Distributions by GC-MS with a New High-Temperature, Low-Bleed Stationary Phase. In PCBs: Recent Advances in Environmental Toxicology and Health Effects; Robertson, L. W., Hansen, L.arry G., Eds.;University Press of Kentucky, Lexington, KY, 2001; pp 67-71. (B12) Spanik, I.; Oswald, P.; Krupcik, J.; Benica, E.; Sandra, P.; Armstrong, D. W. J. Sep. Sci. 2002, 25 (1/2), 45-52. (B13) Veriotti, T.; Sacks, R. Anal. Chem. 2001, 73, 4395-4402. (B14) Poole, C. F.; Li, Q.; Kiridena, W.; Koziol, W. W. J. Chromatogr., A 2000, 898, 211-226. (B15) Kiridena, W.; Koziol, W. W.; Poole, C. F. J. Chromatogr., A 2001, 932, 171-177. (B16) Poole, C. F.; Li, Qinglin; K., Waruna; Koziol, W. W. J. Chromatogr., A 2001, 912, 107-117. (B17) Yang, H.; Chen, W. International Patent C08L079-04, 2001. (B18) Smith, D. A.; Salabsky, D.; Feeney, M. J. A Unique Deactivation Process for Chromatographic Surfaces. Presented at the Eastern Analytical Symposium, Somerset, NJ, November 2000. (B19) U.S. patent application 09/388,868, 2000. (B20) Hoettges, K. F.; Gwilliam, R. M.; Homewood, K. P.; Stevenson, D. Chromatographia 2001, 53 (Suppl.), S424-S426. (B21) Spangler, G. E. J. Microcolumn Sep. 2001, 13 (7), 285-292. (B22) Wong, C. C.; Zoeller, T. L.; Adkins, D. R.; Shadid, J. N. Proc. ASME Fluid Eng. Div. 2000, 251, 1703-1710. (B23) Griffiths, S. K.; Nilson, R. H. Anal. Chem. 2000, 72 (21), 54735482. (B24) Tallarek, U.; Rapp, E.; Scheenen, T.; Bayer, E.; Van As, H. Anal. Chem. 2000, 72 (10), 2292-2301. (B25) Frye-Mason, G.; Kottenstette, R.; Mowry, C.; Morgan, C.; Manginell, R.; Lewis, P.; Matzke, C.; Dulleck, G.; Anderson, L.; Adkins, D. In Micro Total Analysis Systems; Ramsey, J., van den Berg, A., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2001; pp 658-660. (B26) Yu, C. M. U.S. Patent 6,306,200, 2001. (B27) Overton, E. U.S. Patent 6,068,684, 2001. (B28) Sacks, R.; Whiting, J.; Zellers, E. Emerging Technologies for the Development of an Autonomous Micro Gas Chromatograph. Presented at the 24th International Symposium on Capillary Chromatography, Las Vegas, NV, May 20-24, 2001. HIGH-SPEED AND PORTABLE GAS CHROMATOGRAPHY (C1) Hinshaw, J. W. LC-GC 2001, 19 (2), 1527-5949. (C2) Frye-Mason, G.; Kottenstette, R.; Lewis, P.; Heller, E.; Manginell, R.; Adkins, D.; Dulleck, G.; Martinez, D.; Sasaki, D.; Mowry, C.; Matzke, C.; Anderson, L. In Proceedings, 4th Micro Total Analysis Systems; Van den Berg, A., Olthius, W., Bergveld, P., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2000; pp 229-232. (C3) Zhdanov, V. P.; Sidelnikov, V. N.; Vlasov, A. A. J. Chromatogr., A 2001, 928 (2), 201-207. (C4) Wu, N.; Medina, J. C.; Lee, M. L. J. Chromatogr., A 2000, 892, 3-13. (C5) vanDeursen, M.; Janssen, H.; Beens, J.; Lipman, P.; Reinierkens, R.; Rutten, G.; Cramers, C. J. Microcolumn Sep. 2000, 12 (12), 613-622. (C6) Davis, J. M.; Pompe, M.; Samuel, C. Anal. Chem. 2000, 72 (22), 5700-5713.

(C7) Hinshaw, J. V. LC-GC 2000, 18 (11), 1142-1147. (C8) van Deursen, M.; Janssen, H. G.; Beens, J.; Rutten, D.; Cramers, C. J. Microcolumn Sep. 2001, 13 (8), 337-345. (C9) Munari, F.; Cadoppi1, A.; Overton, E.; Roques, N.; Carney, K.; Proceedings of the 23rd International Symposium on Capillary Chromatography, Riva del Garda, Italy, June 2000. (C10) Grall, A.; Leonard, C.; Sacks, R. Anal. Chem. 2000, 72 (3), 591-598. (C11) Blumberg, L. M.; Klee, M. S. Anal. Chem. 2000, 72 (17), 4080-4089. (C12) Blumberg, L. M.; Klee, M. S. Anal. Chem. 2001, 73 (3), 684685. (C13) Whiting, J.; Sacks, R. Anal. Chem. 2002, 74 (1), 246-252 (C14) Whiting, J. Lu, C.; Zellers, E. T.; Sack, R. Anal. Chem. 2001, 73 (19), 4668-4675. (C15) Veriotti, T.; Sacks, R. Anal. Chem. 2001, 73 (13), 3045-3050. (C16) Coutant, C.; Sacks, R. Anal. Chem. 2000, 72 (21), 5450-5458. (C17) Ledford, E. B.; Billesbach, C. A.; Zju, Q. J. High Resolut. Chromatogr. 2000, 23 (3), 205-207. (C18) Seeley, J. V.; Kramp, F.; Hicks, C. J. Anal. Chem. 2000, 72 (18), 4346-4352. (C19) Ong, R.; Shellie, R.; Marriott, P. J. Sep. Sci. 2001, 24 (5), 367377. (C20) Lee, A. L.; Bartle, K. D., Lewis, A. C. Anal. Chem. 2001, 73 (6), 1330-1335. (C21) Prazen, B. J.; Johnson, K. J.; Weber, A.; Synovec, R. E. Anal. Chem. 2001, 73 (23), 5677-5682. (C22) Fraga, C. G.; Prazen, B. J.; Synovec, R. E. Anal. Chem. 2001, 73 (24), 5833-5840. (C23) Fraga, C. G.; Bruckner, C. A.; Synovec, R. E. Anal. Chem. 2001, 73 (3), 675-683. (C24) van Deurson, M. M.; Beens, J.; Janssen, H.; LeClerg, P. A.; Cramers, C. A. J. Chromatogr., A 2000, 878 (2), 205-213 (C25) Vreuls, R. J.; Dalluge, J.; Beens, J. Spectra Anal. 2001, 220 (30), 37-40. (C26) Syage, J. A.; Nies, B. J.; Evans, M. D.; Hanold, K. A. J. Am. Soc. Mass Spectrom. 12 (6), 648-655 (C27) Verma, D. K.; Saunders: G. A.; cheng, W. K. Appl. Occup. Environ. Hyg. 2001, 16 (8), 832-840. (C28) Crume, C. Environ. Test. Anal. 2001, 10 (3), 22-26. (C29) Eckenrode, B. A. J. Am. Soc. Mass Spectrom. 2001, 12 (6), 683-693. (C30) Frishman, G.; Tzanani, N.; Amirav, A. Field Anal. Chem. Technol. 2001, 5 (3), 107-115. (C31) Lu, C.; Zellers, E. T. Anal. Chem. 2001, 73 (14), 3449-3457. (C32) Overton, E. B.; Carney, K. R.; Roques, N.; Dharmasena, H. P. Field Anal. Chem. Technol. 2001, 5 (1-2), 97-105. (C33) Jia, M.; Koziel, J.; Pawliszyn, J. Field Anal. Chem. Technol. 2000, 4 (2-3), 73-84. (C34) Bingham, S.; Medlar, J.; Kabir, A.; Shende, C.; Alli, A.; Malik, A. Anal. Chem. 2002, 74 (4), 752-761. DETECTORS (D1) Dojahn, J. G.; Wentworth, W. E.; Stearns, S. D. J. Chromatogr. Sci. 2001, 39, 54-58. (D2) Mendas, G.; Drevenkar, V.; Zupancic-Kralj, L. J. Chromatogr., A 2001, 918, 351-359. (D3) Sampedro, M. C.; Martin, O.; Lopez de Armentia, C.; Goicolea, M. A.; Rodriguez, E.; Gomez de Balugera, Z.; Costa-Moreria, J.; Barrio, R. J. J. Chromatogr., A 2000, 893, 347-358. (D4) Mendas, G.; Tkalcevic, B.; Drevenkar, V. Anal. Chim. Acta 2000, 424, 7-18. (D5) Sousa, S. R.; Bialkowski, S. E. Anal. Chim. Acta 2001, 433, 181-186. (D6) Otake, T.; Yoshinaga, J.; Yanagisawa, Y. Environ. Sci. Technol. 2001, 35, 3099-3102. (D7) Thurbide. K. B.; Aue, W. A. J. Chromatogr., A 2001, 905, 241250. (D8) Carlsson, H.; Robertsson, G.; Colmsjo, A. Anal. Chem. 2001, 73, 5698-5703. (D9) Aue, W. A.; Singh, H. Spectrochim. Acta, Part B 2001, 56, 517-525. (D10) Dojahn, J. G.; Wentworth, W. E.; Deming, S. N.; Stearns, S. D. J. Chromatogr., A 2001, 917, 187-204. (D11) Sun, K.; Wentworth, W. E. Stearns, S. D. J. Chromatogr., A 2001, 872, 141-165. (D12) Sun, K.; Wentworth, W. E.; Stearns, S. D. J. Chromatogr., A 2001, 872, 167-178. (D13) Wentworth, W. E.; Sun, K.; Zhang, D.; Madabushi, J.; Stearns, S. D. J. Chromatogr., A 2000, 872, 119-140.

Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

2779

(D14) Akesson-Nilsson, G.; Nilsson, O.; Odenbrand, I.; Wesen, C. J. Chromatogr., A 2001, 912, 99-106. (D15) Gonzales, A. M.; Uden, P. C. J. Chromatogr., A 2000, 898, 201-210. (D16) Kim, H. J.; Woo, Y. A.; Kang, J. S.; Anderson, S. S.; Piepmeier, E. H. Mikrochim. Acta 2000, 134, 1-7.

2780

Analytical Chemistry, Vol. 74, No. 12, June 15, 2002

(D17) Eiceman, G. A.; Tadjikov, B.; Krylov, E.; Nazarov, E. G.; Miller, R. A.; Westbrook, J. J. Chromatogr., A 2001, 917, 205-217. (D18) Eijkel, J. C. T.; Stoeri, H.; Manz, A. Anal. Chem. 2000, 72, 2547-2552.

AC020210P