Development and Design of a Single-Stage ... - ACS Publications

Article pubs.acs.org/ac

Development and Design of a Single-Stage Cryogenic Modulator for Comprehensive Two-Dimensional Gas Chromatography Ahmed Mostafa†,‡ and Tadeusz Górecki*,† †

Department of Chemistry, University of Waterloo, 200 University Avenue West, Waterloo, Ontario N2L 3G1, Canada Department of Pharmaceutical Chemistry, College of Clinical Pharmacy, University of Dammam, 31441, Dammam, Eastern Province, Kingdom of Saudi Arabia

‡

S Supporting Information *

ABSTRACT: A new liquid nitrogen-based single-stage cryogenic modulator was developed and characterized. In addition, a dedicated liquid nitrogen delivery system was developed. A well-defined restriction placed inside a deactivated fused silica capillary was used to increase the cooling surface area and provide very efficient trapping. At the same time, it enabled modulation of the carrier gas flow owing to changes in gas viscosity with temperature. Gas flow is almost unimpeded at the trapping temperature but reduced to nearly zero at the desorption temperature, which prevents analyte breakthrough. Peak widths for n-alkanes of 30−40 ms at half height were obtained. Most importantly, even the solvent peak could be modulated, which is not feasible with any commercially available thermal modulator. Evaluation of the newly developed system in two-dimensional gas chromatography (GC × GC) separations of some real samples such as regular gasoline and diesel fuel showed that the analytical performance of this single-stage modulator is fully competitive to those of the more complicated dual-stage modulators.

C

This leads to peak shape irregularities and broad injection bands into the second dimension. Breakthrough is considered the major drawback of single-stage modulation. To solve this problem, dual-stage designs were introduced. In these designs, the primary column effluent collected in the first stage is thermally released with any potential breakthrough into a second trapping stage for additional focusing before injection into the second dimension column.22−24 All commercial cryogenic modulators use the dual-stage design principle. Currently available LN2 cryogenic modulators suffer from several drawbacks. For example, the quad-jet modulator design is quite complicated. The development of the delay loop modulator made the design simpler, but it introduced new issues. In particular, the length of the loop and the velocity of the carrier gas have to be carefully adjusted whenever the chromatographic conditions change. If the flow of the carrier gas is not adjusted properly, the band traveling through the delay loop might not reach the trapping spot at a time when it is cold, and therefore, it might not be refocused. Thus, there is a need for a high performance liquid N2-based cryogenic modulator with simpler design. Several research groups tried to develop single-stage cryogenic modulators. Adahchour et al. demonstrated satisfactory singlestage operation using liquid CO2 jet modulator by shortening the desorption time to a bare minimum.25 The disadvantage of this approach was that when using only one trapping zone, the timing of the jet and the tuning of the instrumental parameters

omprehensive two-dimensional gas chromatography (GC × GC) is currently one of the most effective techniques for the separation of volatile and semivolatile compounds.1−7 Two columns of different selectivities are coupled in series through a special interface called a modulator.8,9 This results in an enhanced sensitivity,10 increased peak capacity, and structured chromatograms. The technique has been successfully used in petrochemical, forensics, environmental, health, and food analyses. The fundamentals of GC × GC instrumentation and methodology have been described extensively (e.g., refs 9 and 11−15). The most critical component of any GC × GC system is the modulator. The main role of the modulator is to trap and/or sample the primary column effluent and inject it into the secondary column.16 GC × GC modulator designs can be classified into two main groups: thermal and flow modulators. Thermal modulators can be subdivided further into heaterbased (trapping at ambient temperature, including oven temperature) and cooling-based (trapping at subambient temperature). Most of the currently used flow modulators utilize pneumatic devices17 that use sample loops and valves to collect the primary column effluent and inject it into the secondary column. Description of flow modulators operation is beyond the scope of this work, and more details can be found in the literature, e.g., refs 18−21. In thermal modulation, analytes may be trapped and modulated at a single location and then reinjected directly into the secondary column (single-stage modulation). However, analyte breakthrough normally occurs in such systems because analytes eluting from the first dimension column during the desorption cycle pass through the modulator unimpeded. © 2016 American Chemical Society

Received: February 26, 2016 Accepted: April 13, 2016 Published: April 13, 2016 5414

DOI: 10.1021/acs.analchem.6b00767 Anal. Chem. 2016, 88, 5414−5423

Article

Analytical Chemistry

Figure 1. Liquid N2 supply system.

Liquid Nitrogen Delivery System. Figure 1 illustrates the design of the cryogen supply system. The custom-built cryogen system used a high-pressure (22 psi) liquid N2 Dewar (Praxair Canada Inc., Mississauga, ON, Canada) to supply liquid N2 through 1/4 in. O.D. copper tube. A two-way cryogenic solenoid valve (Asco Valve Canada, Brantford, Ontario, Canada) was used to turn on and off the flow of liquid N2. Liquid N2 was expelled out of the cryojet made of 10 cm, 1/4 in. O.D. copper tubing weldded into a 6 mm thick brass bracket. Two needle valves and a 1/4 in. T were mounted right after the solenoid valve to control the liquid N2 flow to the cryojet. The excess LN2 coming through the second needle valve was directed through 1/4 in. O.D. copper tubing wrapped tightly around the solenoid, needle valves, and connectors in the interface, thus continuously cooling the entire interface and carrying away the excess heat produced by the solenoid valve from the main LN2 delivery path. All lines were thermally insulated. Modulator Design. A detailed depiction of the modulator design is shown in Figure 2. The modulator capillary was a 10 cm × 0.32 mm segment of deactivated fused silica tubing (Agilent Technologies, Mississauga, ON, Canada). The capillary was held in place and stretched between two 1/16 in. Swagelok tees mounted on a custom-made plate with built-in clips. Nuts with graphite/vespel ferrules were used to accommodate the capillary and seal the ports of the Swagelok tees. The restriction in the trap was made in the form of a 3−4 mm plug of compressed fused silica wool (Restek Corp., Bellefonte, PA, USA) and quartz fiber filter (F & J Specialty Products Inc., Ocala, FL, USA) compressed in the middle of the deactivated fused silica capillary using a fused silica optical fiber (Polymicro Technologies, Phoenix, AZ, USA). The plug was kept in place by passing the portion of the capillary containing the plug through Ronson TechTorch flame (Ronson Corp., Mississauga, Ontario, Canada) twice. All column connections were made using SilTite mini unions (SGE, Austin, TX, USA). A warm jet heated with a rope heater (FGR-030, OMEGA Engineering Limited, Laval (Quebec), Canada) was mounted approximately 5 mm above the trapping capillary. A temperature controller (CN742, OMEGA Engineering Limited, Laval (Quebec), Canada) was used to control the warm air jet temperature. A temperature offset was controlled through connecting two K-type thermocouples (5TC-GG-K-20-36, OMEGA Engineering Limited, Laval, Quebec, Canada) differentially to the temperature controller with one of the thermocouples spot-welded to the warm air jet coil (kept inside the GC oven) and the other one connected to the GC oven itself. To help prevent the buildup of ice on the cryojet nozzle,

had to be done very carefully so as to minimize breakthrough from the first dimension while the trap was hot. Libardoni et al. developed a single-stage air-cooled modulator using a resistively heated stainless steel capillary with the center portion cooled for sample trapping.26 The system was used to analyze a sample of gasoline and a 40-component test mixture spanning a range of volatilities. However, this design also suffered from breakthrough due to the relatively high trapping temperature compared to cryogenic devices, as well as the slow modulator cooling rate. The authors later improved their design by using liquid ethylene glycol rather than air for cooling and programmable voltage for modulator heating to reduce sample breakthrough.27 Though the magnitude of the problem was reduced using the programmed voltage, breakthrough was not completely eliminated. This was related to the single-stage operation of the modulator.28 In this work, a new liquid nitrogen (LN2) single-stage, single jet cryogenic modulator was developed and characterized. The operating principle of the modulator is based on the change in carrier gas viscosity with temperature. As the temperature of the carrier gas decreases, its viscosity decreases as well, and vice versa. Thus, at the desorption temperature, the viscosity increases, slowing down the flow of the carrier gas and thus minimizing breakthrough. The new single-stage cryogenic modulator represents a significant advancement owing to its simplicity and trapping capacity exceeding that of any other cryogenic modulator, which vastly improves the dynamic range of modulation. In addition, the new LN2 delivery system design reduces the LN2 consumption. In the following sections, the development of the modulator will be described and the results of its characterization will be presented.

■

EXPERIMENTAL SECTION Apparatus. A model 6890 gas chromatograph (Agilent Technologies, Mississauga, ON, Canada) equipped with a split/ splitless injector and a flame ionization detector (FID) was used. The GC was controlled, and data were collected using LECO’s ChromaTOF software (LECO Corp., St. Joseph, MI, USA, version 3.25). Data were processed using ChromaTOF version 4.41. The GC oven door was replaced with a window made of heat-resistant glass attached to a piece of sheet metal to help observe the modulator behavior while the system was running. The door interlock was defeated with a small magnet. The primary column, 30 m × 0.25 mm × 1.0 μm VF1-MS (Varian, Mississauga, ON, Canada) was coupled to a 0.8 m × 0.25 mm × 0.25 μm SolGel-Wax phase second dimension column (SGE, Austin, TX, USA). 5415


Article


Figure 2. Schematic diagram of the modulator design.

Figure 3. Modulator design. (A) Front view of the cryojet mount. (B) Warm air supply system. Objects to the right of the dashed line are placed inside the GC oven.

another warm air jet was mounted to blow compressed air at the oven temperature across the face of the cryojet nozzle when the cold jet was off (Figure 3A). The flow of compressed air through the two warm air jets was controlled in such a way that the flow would turn on when the cryojet was off and then off for trapping when the cryojet was on. The warm air supply is shown in Figure 3B.

The LN2 spray was diverted outside of the GC oven through a 1/2 in. O.D. brass tube bent at a 90° angle mounted through the oven floor and facing the cryojet outlet. This was done to avoid any oven temperature fluctuations. Materials and Procedures. A linear n-alkane test mixture consisting of n-pentane through n-tetracosane in CS2 was prepared for modulator testing. The linear alkanes and CS2 were 5416


Article


Dimandja mixture; 0.2 μL split 50:1 for gasoline; and 1 μL split 50:1 for the Grob mixture and the diesel sample). The carrier gas was helium, delivered at a constant average velocity of 40 cm/s. The FID detector was operated at 320 °C and 100 Hz.

obtained from Sigma-Aldrich (St. Louis, MO, USA). Regular unleaded gasoline and diesel fuel samples were obtained from a local gas station. Diesel fuel was diluted 1:10 in CS2. A commercial Grob mixture (composed of 12 components including n-decane, n-undecane, 2,3-butanediol, dicyclohexylamine, 2,6-dimethylaniline, 2,6-dimethylphenol, 2-ethylhexanoic acid, nonanal, 1-octanol, methyl decanoate, methyl undecanoate, and methyl dodecanoate dissolved in methylene chloride) was obtained from Restek Corporation (Bellefonte, PA, USA). Dimandja mixture components29 (n-octane, n-nonane, n-decane, n-undecane, n-dodecane, 1-hexanol, 1-heptanol, 1-octanol, 2-heptanone, 2-octanone, 2-nonanone, heptanal, nonanal, octanal, and 2,6-dimethylaniline) were dissolved in n-hexane. They were all obtained from Sigma-Aldrich (St. Louis, MO, USA). For studies of the modulator performance (analyzing the n-alkane mixture), the second column was replaced by an 80 cm segment of 0.25 mm ID deactivated fused silica capillary. For the analysis of the alkane test mixture and diesel fuel sample, the temperature program started at 40 °C for 0.2 min and then was ramped to 280 °C at a rate of 6 °C/min, with a final hold time of 4 min. For gasoline, the final temperature of the oven program was changed to 220 °C; the rate was 4 °C/min, and the final hold time was 10 min. For the Grob mixture, the final temperature was changed to 225 °C; the rate was 10 °C/min, and the final hold time was 1 min. For the Dimandja mixture, the oven was ramped to 180 °C at a rate of 10 °C/min, with no hold time. The inlet temperature was 250 °C. It was operated in the split mode (1 μL split 100:1 for the alkane mixture and the

■

RESULTS AND DISCUSSION Development of the Cryogen Supply System. The new single-stage cryogenic modulator uses LN2 as the cryogen to trap the analytes in an uncoated fused silica capillary with a restriction in the middle. LN2 was used as the cryogen because LCO2 or cooled N2 gas are not capable of producing temperatures low enough to achieve trapping in uncoated capillaries. On the other hand, using LN2 as the cryogen has some drawbacks. The entire system must be kept at a very low temperature to keep the cryogen in the liquid state, which is not the case when using LCO2 that can be kept in the liquid state at room temperature under sufficient pressure. LN2 consumption was an important factor that was considered; thus, one of the main objectives was to develop a system that reduced the LN2 consumption while delivering LN2 to the jet reliably. The system developed is shown in Figure 1. Copper tubing with 1/4 in. O.D. was used because smaller diameter tubing had too large a surface area to volume ratio, which led to rapid boiling and evaporation of LN2. On the other hand, using larger diameter tubing caused the LN2 to be expelled from the croyjet at very high flow rate, which caused overcooling of the trap. Therefore, a needle valve was mounted in the path of the flow right after the solenoid valve, so that the flow could be reduced. Though the needle valve helped reduce the LN2 flow, the cooling was still excessive. Thus, a 1/4 in. T was connected to the needle valve to split the flow, and another needle valve was mounted to control the LN2 flow going to the cryojet (Figure 1). The configuration described above was capable of delivering LN2 to the cryojet efficiently. In addition, the consumption of LN2 was significantly reduced to ∼30 L per day versus 50 to 100 L per day for the commercially available cryogenic modulators.30 This consumption could most likely be reduced even further by the use of better thermal insulation for the connection lines and the valves.

Table 1. Effect of Temperature on Carrier Gas Flow through the 0.32 mm ID Deactivated Fused Silica Capillary with the Restrictor in the Form of a Fused Silica Wool Plug flow (mL/min) at 30 °C inlet pressure (psi)

LN2 jet off

LN2 jet on

9 12 25 33

0.2 0.2 0.9 1.3

1.4 1.6 5.9 8.4

Figure 4. Changes in carrier gas flow during modulation demonstrated using carrier gas doped with propane. The negative peaks (arrows) illustrate the significant drop in carrier gas flow during desorption. 5417


Article


Figure 5. Analysis of n-alkane mixture (n-C5 to n-C24 in CS2). 2D contour plot (A). Close up view of hexane (B) and pentane (C) peaks showing no breakthrough. n-C5, CS2, and n-C6 surface plot (D) and raw chromatogram of CS2 modulation (E).

Development of Warm Air Jets. Though the trapping capillary would heat up quite quickly in the oven on its own without any additional warm air jets when the cryogen flow was stopped, the desorption timing was not reproducible. Consequently, a warm jet was mounted approximately 5 mm above the trapping capillary to blow compressed air at the oven temperature toward the middle of the capillary. Though this helped somewhat, desorption was still irreproducible. Therefore, a rope heater was wrapped around the warm air jet coil to help heat the air inside. The temperature of the hot jet was kept at a constant offset with regard to the oven temperature with the help of a temperature controller and two thermocouples

connected differentially (one attached to the hot jet, the other to the oven wall). In this way, hot jet temperature tracked the oven temperature with the difference between the two kept constant. Trapping Capillary. Initially, a segment of deactivated fused silica capillary, 0.1 mm I.D., was used as the single-stage trap connected between the primary and the secondary columns. Even though using LN2 as the cryogen achieved a trapping temperature low enough to trap all of the analytes, breakthrough was observed (see Figure S-1: n-pentane peak modulation using 100 μm I.D. deactivated fused silica capillary). This was mainly attributed to the fact that, at typical carrier gas flow rates 5418


Article


Figure 6. GC × GC chromatogram of the Grob mixture. Compound identification: (1) 2,3-butanediol; (2) n-decane; (3) 1-octanol; (4) 2-ethylhexanoic acid; (5) nonanal; (6) n-undecane; (7) 2,6-dimethylphenol; (8) 2,6-dimethylaniline; (9) methyl decanoate; (10) dicyclohexylamine; (11) methyl undecanoate; (12) methyl dodecanoate.

a flame. Peak widths of 35, 40, and 51 ms at half height were obtained for n-C5, n-C12, and n-C19, respectively. Once the trapping capillary was constructed, the effect of the restriction created by the plug on the flow of the carrier gas was tested. One end of the trapping capillary was connected to the primary column outlet, and the other end was connected to a flow meter. Various inlet pressures were applied, and the flow through the capillary was measured when the LN2 was off. Then, the flow was measured again when the LN2 jet was on. When the LN2 jet was on, the flow of the carrier gas increased by a factor of ∼7 compared to the flow when the LN2 jet was off. Once the LN2 flow was reactivated, the carrier gas flow immediately increased again (Table 1). Modulator Performance. For initial testing of the interface, the carrier gas was doped with propane continuously bled from a disposable cylinder through a 10 μm I.D. fused silica capillary with the oven isothermal at 40 °C using a 4 s modulation period and 0.8 s hot pulse time. With propane continuously supplied to the FID, the detector acted effectively as a mass flow meter. Any decrease in the carrier gas flow (thus, the amount of propane delivered per unit time) would result in a drop in the detector signal and vice versa. The results are shown in Figure 4. The negative peaks in this figure illustrate the momentary decrease in the carrier gas flow due to increasing viscosity when the carrier gas temperature was increased during the desorption stage. The positive peaks correspond to the fraction of the propane that was trapped at the LN2 temperature and then released during the desorption step. To test the modulator with a broader range of compounds, a mixture of n-C5 to n-C24 in CS2 was analyzed with a modulation period of 4 s. The performance of the interface can be assessed through the peak shapes and widths. Figure 5A presents a 2D chromatogram of n-C5 to n-C24 alkanes in CS2 showing very sharp peaks. Peak widths of 60 and 65 ms at the base were obtained with no breakthrough evident (Figure 5B,C). Figure 5D,E shows a closer view of n-C5, CS2, and n-C6 peaks with efficient trapping and modulation of the volatile injection solvent (CS2) (discussed later).

(up to 1.5 mL/min), radial diffusion was too slow for analytes near the center of the column to reach the walls of the capillary while they traveled through the relatively short distance that was cooled by the jet (∼5−7 mm), and trapping could occur only when a collision of analyte molecule with the wall took place. For example, at 1.5 mL/min carrier gas flow rate, the time required to travel a distance of 5 mm in a 100 μm tube is only about 1 ms. More importantly, the trapping position did not constitute sufficient restriction to carrier gas flow. The idea to overcome this problem was to produce a significant restriction to the carrier gas flow through the incorporation of a small plug (∼3−4 mm long) composed of compressed fused silica wool and quartz fiber filter in the center of the capillary. This plug provided increased surface area for the analytes passing through the trap without significantly increasing its thermal mass. In addition, it provided a significant restriction to the carrier gas flow when the trapping capillary was at the oven temperature. While viscosity of liquids decreases with increasing temperature, viscosity of gases increases. This is due to the fact that, as gas is heated, the probability of collisions between gas molecules increases, which results in higher viscosity because the transfer of momentum between stationary and moving molecules is what causes gas viscosity. On the other hand, at the very low cryogenic trapping temperature (∼−196 °C), the viscosity of the carrier gas decreased to such an extent that the plug offered negligible resistance to the flow of the gas. Since the desorption time constituted only a small fraction of the modulation period, overall, the effect of the presence of the plug on the carrier gas flow rate was very small, and pressures required to reach the desired flow were only marginally higher than those required for a system without the restriction. Construction of the trap inside the capillary was challenging with regard to the packing procedure and immobilizing the plug so that it was not dislodged by the flow of the carrier gas. The packing was carried out with the help of a fused silica optical fiber (0.27 mm O.D.), which helped avoid scratching the inside of the 0.32 mm capillary. The plug was sintered in place by passing the portion of the capillary containing the plug through 5419


Article


Figure 7. GC × GC chromatogram of the Dimandja mixture (A). Compound identification: (1) n-C8; (2) 1-hexanol; (3) 2-heptanone; (4) heptanal; (5) n-C9; (6) 1-heptanol; (7) 2-octanone; (8) octanal; (9) n-C10; (10) 1-octanol; (11) 2-octanone; (12) octanal; (13) n-C11; (14) 2,6-dimethylaniline; (15) n-C12. (B) Surface plot, (C) contour plot, and (D) raw GC × GC trace of methylene chloride (Grob mixture solvent) trapping and modulation. 5420


Article


Figure 8. GC × GC contour plot chromatogram of regular gasoline (A); 2D contour plot of diesel fuel sample (B); raw GC × GC trace of diesel fuel sample (C).

breakthrough, was removed from the display. All 12 peaks were very sharp with no tailing or breakthrough even for analytes that are known to be problematic in their chromatographic separation, such as acids, amines, and diols. The Dimandja mixture was analyzed as well to test the performance of the modulator in trapping components with a wide range of polarities. As can be seen in Figure 7A, the 15 components of the Dimandja mixture were efficiently

The new interface was also tested in the analysis of a commercial Grob mixture sample, which contained various classes of organic components including hydrocarbons, esters, aldehydes, acids, bases, and alcohols to test the performance of the modulator in trapping a wide range of different polarity components. Figure 6 shows a 2D contour plot chromatogram of the mixture. For clarity of presentation, the solvent peak (methylene chloride), which was modulated without 5421



■

modulated showing sharp peaks with no tailing or breakthrough. The n-hexane solvent peak was again removed from the display for clarity. Solvent Trapping and Modulation. One of the most interesting features of the new interface was its very high trapping capacity, making it possible to trap and modulate the injection solvents. As can be seen in Figures 5D,E, the solvent peak of CS2 was perfectly trapped and modulated without any tailing or breakthrough. Figures 7B−D show the trapping and modulation of methylene chloride peak (Grob mixture solvent). To the best of our knowledge, none of the commercially available or commonly used thermal modulators is capable of efficiently trapping and modulating volatile injection solvents. This is very important when some of the analytes of interest elute close to the solvent peak, as it permits their detection and precise quantification. The very high trapping capacity, and hence wide dynamic range of modulation, would also be very important for samples containing a few major components and a large number of minor ones (e.g., some essential oils), where the major component breakthrough in the modulator leads to obscuring of minor components eluting nearby. No breakthrough was observed for the solvent peak or any of the analytes in the entire range of carrier gas flow rates that are typical in GC × GC/MS (up to ∼1.5 mL/min). Performance Testing with Real Samples. The new interface was tested in the analysis of regular unleaded gasoline. Numerous chromatograms have been published in the literature using different GC × GC interfaces.24,31−33 Gasoline is a useful sample to compare interface performance to that of other interfaces. Figure 8A shows a GC × GC chromatogram of regular gasoline obtained with the new single-stage modulator. The chromatogram shows very close resemblance to those reported for other thermal modulators. Good peak shapes with no tailing or breakthrough for any peak, especially those eluting at the beginning of the chromatogram, were observed. Figure 8B shows the GC × GC chromatogram of a diesel fuel sample. The chromatogram presents sharp peaks with no tailing or breakthrough and shows distinct bands of analytes grouped by specific chemical characteristics.

AUTHOR INFORMATION

Corresponding Author

*Phone: +1 519 888 4567 ext. 35374. Fax: 519 746 0435. E-mail: [email protected]. Notes

The authors declare no competing financial interest. Presented in parts at the 38th ISCC and 11th GC × GC symposium, Riva del Garda, Italy, May 18−23, 2014 and 39th ISCC and 12th GC × GC symposium, Fort Worth, TX, May 16−21, 2015.

■

ACKNOWLEDGMENTS The Natural Sciences and Engineering Research Council (NSERC) of Canada and Ontario Graduate Student Scholarship program (OGS) are gratefully acknowledged for the financial support of this work. Pak Hin Law provided technical assistance in the project. The authors thank Polymicro Technologies for supplying the fused silica tubing.

■

REFERENCES

(1) Seeley, J. V.; Kramp, F. J.; Sharpe, K. S.; Seeley, S. K. J. Sep. Sci. 2002, 25, 53−59. (2) Vestner, J.; Malherbe, S.; Du Toit, M.; Nieuwoudt, H. H.; Mostafa, A.; Górecki, T.; Tredoux, A. G. J.; de Villiers, A. J. Agric. Food Chem. 2011, 59, 12732−12744. (3) Weldegergis, B. T.; Crouch, A. M.; Górecki, T.; de Villiers, A. Anal. Chim. Acta 2011, 701, 98−111. (4) Weldegergis, B. T.; de Villiers, A.; McNeish, C.; Seethapathy, S.; Mostafa, A.; Górecki, T.; Crouch, A. M. Food Chem. 2011, 129, 188− 199. (5) Dymerski, T.; Chmiel, T.; Mostafa, A.; Śliwinska, M.; Wiśniewska, P.; Wardencki, W.; Namieśnik, J.; Górecki, T. Curr. Org. Chem. 2013, 17, 853−870. (6) Janoskova, N.; Vyviurska, O.; Spanik, I. J. Food Nutr. Res. 2014, 53, 353−362. (7) Purcaro, G.; Barp, L.; Beccaria, M.; Conte, L. S. Anal. Bioanal. Chem. 2015, 407, 309−319. (8) Górecki, T.; Harynuk, J.; Panic, O. J. Sep. Sci. 2004, 27, 359−379. (9) Edwards, M.; Mostafa, A.; Górecki, T. Anal. Bioanal. Chem. 2011, 401, 2335−2349. (10) Mostafa, A.; Górecki, T. LC GC Eur. 2013, 26, 672−679. (11) Marriott, P.; Shellie, R. TrAC, Trends Anal. Chem. 2002, 21, 573−583. (12) Pursch, M.; Sun, K.; Winniford, B.; Cortes, H.; Weber, A.; McCabe, T.; Luong, J. Anal. Bioanal. Chem. 2002, 373, 356−367. (13) Dalluge, J.; Beens, J.; Brinkman, U. A. T. J. Chromatogr. A 2003, 1000, 69−108. (14) Górecki, T.; Panic, O.; Oldridge, N. J. Liq. Chromatogr. Relat. Technol. 2006, 29, 1077−1104. (15) Mostafa, A.; Górecki, T.; Tranchida, P. Q.; Mondello, L. In Comprehensive Chromatography in Combination with Mass Spectrometry, 1st ed.; John Wiley & Sons, Inc.: New York, 2011; pp 93−144. (16) Harynuk, J.; Górecki, T. Am. Lab. 2007, 39, 36−39. (17) Seeley, J. V.; Kramp, F.; Hicks, C. J. Anal. Chem. 2000, 72, 4346−4352. (18) Bruckner, C. A.; Prazen, B. J.; Synovec, R. E. Anal. Chem. 1998, 70, 2796−2804. (19) Bueno, P. A.; Seeley, J. V. J. Chromatogr. A 2004, 1027, 3−10. (20) LaClair, R. W.; Bueno, P. A.; Seeley, J. V. J. Sep. Sci. 2004, 27, 389−396. (21) Micyus, N. J.; McCurry, J. D.; Seeley, J. V. J. Chromatogr. A 2005, 1086, 115−121. (22) Kinghorn, R. M.; Marriott, P. J. J. High Resolut. Chromatogr. 1998, 21, 620−622. (23) Beens, J.; Adahchour, M.; Vreuls, R. J. J.; van Altena, K.; Brinkman, U. A. T. J. Chromatogr. A 2001, 919, 127−132.

■

CONCLUSIONS Single-stage modulation is a viable alternative to more complicated dual-stage designs. Band breakthrough during injection can be prevented by using changes in carrier gas viscosity with temperature to reduce the carrier gas flow during desorption. In addition, the newly developed LN2 delivery system reduces the LN2 consumption. Very sharp peaks with no tailing or breakthrough were obtained, including the injection solvents. The new modulator simplifies the design of cryogenic modulators and greatly increases the dynamic range of concentrations that can be efficiently modulated without breakthrough (as illustrated by solvent modulation) and, as such, should be of great interest to the GC × GC community.

■

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b00767. n-Pentane peak modulation (PDF) 5422


Article

Analytical Chemistry (24) Harynuk, J.; Górecki, T. J. Chromatogr. A 2003, 1019, 53−63. (25) Adahchour, M.; Beens, J.; Brinkman, U. A. T. Analyst 2003, 128, 213−216. (26) Libardoni, M.; Waite, J. H.; Sacks, R. Anal. Chem. 2005, 77, 2786−2794. (27) Libardoni, M.; Hasselbrink, E.; Waite, J. H.; Sacks, R. J. Sep. Sci. 2006, 29, 1001−1008. (28) Mostafa, A.; Edwards, M.; Górecki, T. J. Chromatogr. A 2012, 1255, 38−55. (29) Dimandja, J. M. D.; Clouden, G. C.; Colon, I.; Focant, J. F.; Cabey, W. V.; Parry, R. C. J. Chromatogr. A 2003, 1019, 261−272. (30) Ledford, E. B.; Billesbach, C.; Termaat, J. Pittcon 2002, March 17−22, 2002, New Orleans, LA; contribution #2262P. (31) Frysinger, G. S.; Gaines, R. B.; Ledford, E. B. J. High Resolut. Chromatogr. 1999, 22, 195−200. (32) Frysinger, G. S.; Gaines, R. B. J. High Resolut. Chromatogr. 2000, 23, 197−201. (33) Frysinger, G. S.; Gaines, R. B. J. Sep. Sci. 2001, 24, 87−96.

5423


Development and Design of a Single-Stage ... - ACS Publications

Recommend Documents