1993
Anal. Chem. 1980, 52, 1993-1994
n-alkyl bonded phases. By contrast, rapid base line separation is obtained in 70/30 methanol/ water on the perfluorocarbon bonded phase.
of Organic Chemistry, Rijksuniversiteit-Gent (Belgium), for a sample of 4-tert-butyl-2-methoxy(N-trifluoroacetyl)piperidine.
CONCLUSION T h e (heptadecafluorodecy1)dimethylsilylbonded phase shows specific fluorine-fluorine interaction and enhanced retention for fluorine-containing compounds. The retention increases with the number of fluorine atoms in the solute. This behavior can be used successfully for the separation of fluorine-containing compounds from their non-fluorine-containing analogues.
ACKNOWLEDGMENT Thanks are due to H. Vierdag from Shell Nederland Chemie (CSC, Rotterdam) for a generous gift of Barnon, Mataven, and Suffix and to V. Asher and R. Callens from the Laboratory
LITERATURE CITED (1) Berendsen, G. E.; De Galan, L. J. Chtomatogr. 1980, 796, 21. (2) Berendsen. G. E.; Pikaart. K. A,; De Galan, L. J. Liq. Chromatogr., In
press. (3) Berendsen, G. E.; Pikaart, K. A,; De Galan, L. Proceedings of the International Liquid Chromatography Symposium V, Amsterdam, April 1980, in press. (4) Schwarzenbach, R. J . Chromatogr. 1978, 117, 206. (5) Langer, S. H.; Hein, D. T.; Bolme, M. W. Anal. Chem. 1979, 51,1091. (6) Vigh, Gy. J. Chromatogr. 1978, 117, 424. (7) Berendsen, G. E.; Regouw, R.; De Gaian, L. Anal. Chem. 1979, 57, 1091. (8) Berendsen, G. E.; De Gaian, L. J. Liq. Chromatogr. 1978, 7, 561.
RECEIVED for review May 5, 1980. Accepted July 15, 1980.
On-Column Cryogenic Trapping of Sorbed Organics for Determlnatlon by Capillary Gas Chromatography Davld Kalman, Russell Dills, Cherlll Perera, and Foppe DeWalle Trace Organics Analysis Center, Department of Environmental Health, University of Washington, Seattle, Washington 98 195
Various devices for low-temperature trapping and thermal focusing of samples for gas chromatographic analysis have been reported. These include both on-column ( 2 , 2) and extra-column (3, 4 ) devices. In this report, we describe a simple on-column cryogenic trap suitable for capillary analysis that can be hand fabricated from common materials. This trap was developed for the purge-and-trap (5) analysis of wastewater and sludge samples by capillary column gas chromatography, where the resolution provided by capillary columns is considered essential in view of the complexity of the sample. Three different cold trap designs were evaluated, as depicted by Figure l. Design A consists of a notched tube of in. 0.d. copper, prebent to a 5.5-cm radius to conform to the curvature of the capillary column. Liquid nitrogen is introduced under modest pressure. (In this and the other designs, the throttled headpressure from the liquid nitrogen reservoir was used to deliver up to 25 mL/min of coolant.) Design B consists of two machined aluminum plates sandwiched together around the column to form the trap and held in place with metal clips. Liquid nitrogen is delivered to one side of the trap, and warm air can be introduced from the other. Design C is a thin wall (1/8-in.0.d.) stainless steel tube bent to fit the capillary column. It is lightweight enough to be threaded onto the column prior to end straightening and can be installed with the column. (Newly introduced fusedsilica capillary columns (6) do not require straightening and will conform to the shape of the cryotrap, simplifying its construction.) All three designs can be mounted from a central support or supported from the liquid nitrogen supply tube, which was bulkhead mounted (as depicted schematically in Figure 2). When installed, each of the three designs covers approximately 10 cm of the column. This provides a residence time of about 0.2 s for normal flow rates between 1 and 2 mL/min, which is sufficient at liquid nitrogen trapping temperatures to capture even the most volatile organic compounds, such as chloromethane. Typical operation of the device as part of an analysis of aqueous samples for purgable organics is as follows: The sample (a standard containing all Environmental Protection 0003-2700/80/0352-1993$0 1 .OO/O
Agency (EPA) priority purgables except acrolein and acrylonitrile a t 4 ppb) is purged with organic-free helium for 15 min at 20 mL/min. The purged organics are trapped on conditioned Tenax-GC. During the last 5 min of Tenax trapping, the cryogenic trap is precooled. (When the trap is fully cooled by liquid nitrogen flow, air will condense on the outside, adding to whatever liquid nitrogen overflows from the trap.) The coolant flow is continued throughout the cryogenic trapping, which takes place for 10 min during which time the sorbed organics are thermally desorbed into the carrier gas stream (at 2 mL/min flow rate). During this period, the Tenax is heated to 200 "C. Upon warming of the cold trap by a vigorous flow (in excess of 500 mL/min) of gas supplied from a compressed air tank and heated to 40 "C by passage through a heated copper coil, the gas chromatographic separation begins. Normally, this heating method produces elution of nonretained components from the cryogenic trap (such as carbon dioxide) within 5 s of the column dead time as determined by methane injection. This indicates that heating up of the cryogenic trap is approximately that rapid. When device C was used in the manner described, the result was as depicted in Figure 3. This particular analysis utilized mass spectrometry detection, which revealed no breakthrough of organics during cryogenic trapping. The analysis was performed on a Hewlett-Packard 5840-A gas chromatograph and a 30-m SE-54 (0.25-88 i.d., J&W Scientific) glass capillary column. The gas chromatograph was coupled by a direct glass-lined steel tube interface to a Finnegan 4023 mass spectrometer data system. Helium carrier gas a t 40 cm/s linear velocity (at 30 "C) was used. The gas chromatograph oven temperature was increased from 30 to 50 "C a t 4O/min and from 50 to 280 "C at 8"/min. The results obtained by using design A showed poor peak shape and occasional split peaks for compounds more volatile than dichloromethane; this was apparently due to insufficiently rapid heat up of the cryogenic trap. This shortcoming was eliminated in design B, which permits the use of a warm air back-flush of the cryogenic trap as a means of initiating the chromatographic analysis. However, design B was cumbersome to install due to its weight and rigidity. Design C, 0 1980 American Chemical Society
1994
Anal. Chem. 1980. 52. 1994-1998 Cross section
b
a
0 34
1 d!\ 30 00 TIME
W.H
Figure 3. Reconstructed gas chromatogram of EPA priority purgables and deuterated standard compounds: (1) carbon dioxide; (2) chloromethane; (3) chloroethane; (4) bromomethane; (5) chloroethane-d,; (6) trichlorofluoromethane; (7) acrylonitrile-d,; (8) dichloromethane; (9) Z-1,2-dichloroethane; (10) 1,l-dichloroethene; (1 1) 2,2-dichloropropane-d,; (12) chloroform-d,; (13) trichloromethane; (14) l,l,ltrichloroethane-d,; (15) 1,l ,l-trichloroethane; (16) 1,2dichloroethane; (17) benzene-d,; (18) tetrachloromethane; (19) benzene; (20) 1.2dichloropropane; (21) bromodichloromethane; (22) Z-1,3dichloro-lpropene; (23) toluene-d,; (24) methylbenzene; (25) 1,d-dichloro-lpropane (E); (26) l11,2-trichloroethane; (27) dibromochloromethane; (28) tetrachloroethane; (29) chlorobenzene; (30) ethylbenzene; (31) bromoform-d,; (32) tribromomethane; (33) lI1.2,2-tetrachloroethane; (34) 1,4-dichlorobenzene-d,.
Figure 1. Jacket designs for oncolumn cryogenic trapping: (a) capillary column (from injector); (b) capillary column (to detector); (c) warm air inlet; (d) coolant inlet (from liquid nitrogen reservoir).
Itransfer line glass capillary column
I
M 3
I I I
GC oven
coolant reservoir
cryotrap ( - 10 cm)
[q,,,)
~
Figure 2. Schematic of purge and trap-GC/MS system
which gave equivalent performance to design B, is preferable in that it is mechanically easier to install and to remove. In the use of design B or C, the directions from which coolant and warming air are introduced are optimum as depicted in Figure l. As Kaiser (7) has shown, the direction of thermal gradients established during the cryogenic trapping and thermal desorption plays a critical role in the quality of the subsequent gas chromatography. For this reason, i t is preferable that the coolant enter the cryogenic trap along the downstream side of the column and exit nearest the injector and that the warming air enter nearest the injector and follow
the direction of column flow. The flows of coolant and air are readily automated, especially when a microprocessorcontrolled gas chromatograph is used. As with the extracolumn device reported by Rijks ( 4 ) ,this cryogenic trap can also be used for headspace analysis and for the concentration of organics from large gas samples. Potential users are cautioned that some consideration should be given to the amount of water retained in adsorption and its effect during cryogenic trapping. If other sorbents are used in addition to or in place of Tenax-GC or if the water sample is heated during sparging, moisture can be trapped in sufficient quantities to plug the column during cooling. Further, the effect of unavoidable traces of water on capillary column stability somewhat limits the choice of stationary phase that can be used. During the evaluation of these traps, excellent results were obtained by purging room-temperature samples onto Tenax-GC alone and analyzing with 0.25 mm i.d. SE-54 glass capillary columns. Under these conditions, plugging of the column by ice occurred only once in more than 100 analyses.
LITERATURE CITED (1) Rushneck, D. R . J . Gas Chromatogr. 1965, 3, 318. (2) Willis. D. E. Anal. Chem. 1968, 4 0 , 1597. (3) Zlatkis, A,; Lichtenstein, H. A,; Tishbee, A. Chromatographia 1973, 6 , 67. (4) Rijks, J. A,; Drozd, J.; Novak, J. J . Chromatogr. 1979, 152, 195. (5) Bellar, T. A.; Lichtenberg,J. J. J.-Am. Water Works Assoc. 1974, 66, 739. (6) Dandenau, R.;Beute, P.; Rooney, T.; Hiskes, R. Am Lab (FairfieM, Conn.) 1979, 1 1 , 61. (7) Kaiser, R. E. Anal. Chern. 1973, 45, 965.
RECEIVED for review May 28, 1980. Accepted July 21, 1980.
Dual Coulometric-Amperometric Cells for Increasing the Selectivity of Electrochemical Detection in High-Performance Liquid Chromatography Gary W. Schieffer Pharmaceutical Research Division, Norwich-Eaton Pharmaceuticals, Division of Morton-Norwich Products, Inc., Norwich, New York 138 15
Electrochemical cells, especially those with some form of carbon as the working electrode, have been widely used as 0003-2700/80/0352-1994$01 .OO/O
inexpensive, sensitive, and highly specific detectors for high-performance liquid chromatography (HPLC) (1). The @ 1980 American Chemical Society
