Anal. Chem. 1987, 59, 1882-1884
1882
indicative of telomeric products (23) that are inconsistent with a conventional depolymerization mechanism (24). The same sample with 252CfP D yields no positive ions, but its negative ion spectrum (peaks to m / z 2159) appears to represent the primary ionic products, RSOy, of the radiation degradation, suggesting that much of the Cs+ positive ion spectrum results from secondary ionization of neutral products of the radiation degradation (23).
(10) Settine, R . L.; Kinsinger, J. A.; Ghaderi, S. Eur. Spectrosc. News 1985, 58, 16-18. (11) Cody, R. B., Jr.; Kinsinger, J. A.; Ghaderi. S.; Amster, I. J.; McLafferty, F. W.; Brown, C. E . Anal. Chlm. Acta 1985, 178,43-66. (12) Laude, D. A.. Jr.; Johlman. C . L.; Brown, R . S.; Weil, D. A,; Wilkins, C. L. Mass Spectrom. Rev. 1988, 5 , 107-166. (13) Russell, D. H. Mass Spectrom. Rev. 1988, 5 , 167-189. (14) Hunt, D. F.; Shabanowitz, J.; Yates, J. R., 111; Zhu, N.-Z.; Russell, D. H.; Castro, M. E. Proc. Natl. Acad. Sci. U . S . A . 1987, 8 4 , 620-623. (15) Tabet, J. C.; Rapin, J.; Poreti, M.; Gumann, T. Chimia 1988, 4 0 , 169-1 71. (16) Viswanadham, S. K.; Weller, R . R.; Sheetz, M. A.; Hercules, D. M.; Giam, C. S. American Society of Mass Spectrometry Meeting, Cincinnati, OH, June 1986; paper MPE4. (17) Viswanadham. S. K.; Hercules, D. M.; Weller, R. R.; Giam, C. S. Biomed. Environ. Mass Spectrom, 1987, 14, 43-45. (18) Amster. I. J.; McLafferty, F. W.: Castro, M. E.; Russell, D. H.; Cody, R. B., Jr.; Ghaderi, S. Anal. Chem. 1988, 58,483-485. (19) Sichterman, W.; Benninghoven. A. I n t . J . Mass Spectrom. Ion Phys. 1981. 40. 177-184. (20) Amster, I . J.; Loo, J. A.; Furlong, J. J. P.; McLafferty, F W Anal Chem. 1987. 5 9 . 313-317. (21) Ens, W.; Standing, K . G.; Chait, B. T.; Field, F. H. Anal. Chem. 1981. 5 3 , 1241-1244. (22) Yang. Y . M.; Sokoloski, E. A,; Fales, H. M.; Pannell, L. K. Biomed. Environ . Mass Spectrom. 1986, 13, 489-492. (23) Loo, J. A.; Wang, B. W.; Wang, F. C.-Y.; McLafferty. F. W.; Klymko, P. Macromolecules 1987, 20, 698-700. (24) Bowmer, T. N.; Brown, J. R.; Grespos, E.; O'Donnell, J. H. Proc. IUPAC Cong. Macromolec. 1983, 28, 445. (25) Bowers, W. D.; Delbert. S.-S.; McIver, R. T., Jr. Anal. Chem. 1986, 5 4 , 969-972. (26) Hunt, D. F.; Shabanowitz. J.; Yates, J. R., I 1 1 J . Chem. Soc., Chem. Commun., in press.
CONCLUSIONS These preliminary results extend those of Tabet (15) in indicating that this unique ionization technique can be used with FTMS. To utilize the unusual 252CfPD capabilities for ionization of high molecular weight compounds, the present FTMS trapping efficiency of must be improved. Better methods for combining repeated spectral measurements could give resolution approaching the unusual FTMS capabilities demonstrated with higher intensity ionization methods. Although the much longer lifetimes required by FTMS than by TOF can reduce sensitivity, this make possible a variety of ion dissociation (e.g., by 193 nm photons) (25, 26) and reaction techniques for MS/MS, for which FTMS is far better suited than is TOF (2, 6).
ACKNOWLEDGMENT
Joseph A. Loo Evan R. Williams I. Jonathan Amster Jorge J. P. Furlong Bing H. Wang Fred W. McLafferty*
The authors are indebted to R. B. Cody, Jr., R. J. Cotter, T. Gaumann, S. Ghaderi, D. F. Hunt, D. P. Littlejohn, R. D. Macfarlane, P. Roepstorff, B. U. R. Sundqvist, and J. C. Tabet for helpful advice and discussions. Registry No. 252Cf,13981-17-4; bradykinin, 58-82-2;gramicidin S, 113-73-5;vitamin Biz, 68-19-9; poly(butene-1-sulfone), 25104-
Chemistry Department Baker Laboratory Cornel1 University Ithaca, New York 14853-1301
10-3.
LITERATURE CITED
Brian T. Chait Frank H. Field
Macfarlane, R . D. Anal. Chem. 1983, 55, 1247A-1264A. Chait. B. T.; Field, F. H. J . Am. Chem. SOC. 1984, 106,1931-1938. Pannell, L. K.; Sokoloski. E. A.; Fales, H. M.; Tate, R . L. Anal. Chem. 1985, 57, 1060-1067. Jonsson, G. P.; Hedin, A. B., Hakansson, P. L.: Sundqvist, B. U. R.; Save, 8 . G. S.; Nielsen, P. F.; Roepstorff, P.; Johansson, K.-E.; Kamensky, I.; Lindberg, M. S. L. Anal. Chem. 1988, 58, 1084-1087. Alai, M.: Demirev. P.; Fenselau, C.; Cotter, R. J. Anal. Chem. 1986, 5 8 , 1303-1307. Haddon, W. F.; McLafferty, F. W. Anal. Chem. 1989, 4 1 , 31-36. Comisarow, M. B.; Marshall, A. G. Chem. Phys. Lett. 1974, 25. 282-283. -. - -. .. Gross, M. L.; Rempel, D. L. Science 1984, 226,261-268 Marshall, A. G. Acc. Chem. Res. 1985, 18. 316-322.
Rockefeller University 1230 York Avenue New York, New York 10021
RECEIVED for review September 8, 1986. Accepted April 7 , 1987. Generous financial support was provided by the National Institutes of Health (Grant GM-16609) and by the Army Research Office (Grant DAAG-29-82-K-0179).
AIDS FOR ANALYTICAL CHEMISTS Comparison of Sample Loops Constructed of Several Different Materials for Gas Chromatographic Analysis of Parts-per-Billion-Level Organic Gas Mixtures John M. Allen and R. K. M. Jayanty* Research Triangle Institute, Center for Environmental Measurements, P.O. Box 12194, Research Triangle Park, N o r t h Carolina 27709 Darryl J. von Lehmden United States Environmental Protection Agency, Environmental Monitoring Systems Laboratory, Research Triangle Park, N o r t h Carolina 2771 1 The ability to accurately quantify very low concentration levels of volatile toxic organic compounds is of great impor0003-2700/87/0359-1882$01.50/0
tance in ambient air and source emission measurements. For this reason, 27 gaseous organic compounds at parts-per-billion G
1987 American Chemical Society
ANALYTICAL CHEMISTRY, VOL. 59, NO. 14, JULY 15, 1987
Table I. Part-per-Billion-Level Organic Gas Mixtures group 1 5 organics in
group 2 9 organics in
N2 carbon trichlorotetrachloride ethylene chloroform 1.2-dichloroN2
group 3
group 4
7 organics in N2
6 organics in N2
vinylidene chloride Freon 113
acrylonitrile
1.3-butadiene
ethane
perchlorethyl-
1,e-dibromo-
Freon 114
ethylene oxide
ene vinyl chloride
ethane Freon 11
acetone
benzene
Freon 12
1,4-dioxane
methylene chloride propylene oxide o-xylene
bromomethane toluene chlorobenzene methyl ethyl ketone
l,l,l-trichloroethane acetonitrile ranges of the cylinders currently
available
ranges of the cylinders currently available
7-90 ppb 7-90 ppb 90-430 ppb 90-430 ppb 430-10000 ppb
ranges of the cylinders currently available
ranges of the cylinders currently available
7-90 ppb 90-430 ppb
7-90 ppb 430-10000 ppb
(ppb) concentration levels in compressed gas cylinders have been selected by the United States Environmental Protection Agency (USEPA) for use in auditing ambient air and source emission measurement systems. Four different groups of cylinders have been prepared, which each contain a mixture of from five to nine of the 27 selected organic compounds (Table I). In order for the cylinder gases to be used as standards for audit purposes, they must be well characterized. Analyses of the cylinder components are performed by using standards that are specially prepared by the National Bureau of Standards (NBS) to assess the accuracy and long-term stability of the concentrations of the cylinder components. The method of analysis for the cylinder gases is gas chromatography (GC). Introduction of the gas-phase samples is accomplished by means of an automated gas sampling valve, which incorporates a sample loop having a typical internal volume of 3-10 cm3. A sampling valve was chosen as the method of sample introduction because it is the most accurate technique available due to the ability of the moment and volume of injection to be precisely controlled (1). It has been observed in our laboratory that the chromatographic peaks obtained from the analysis of certain oxygenated (methyl ethyl ketone, acetone, ethylene oxide, propylene oxide, and 1,4-dioxane) or polar (acetonitrile) compounds suffered from varying degrees of tailing. This peak tailing results in a significant reduction in the precision of the analysis. Other compounds included in the cylinders show no significant peak tailing. The source of the peak tailing for the oxygenated compounds is clearly an interaction between the compounds and one or more components of the gas chromatographic system (i.e. sampling valve, sample loop, transfer lines, gas chromatographic column, and/or detector). The inner surfaces are usually heterogeneous and contain sites with high adsorption energies from which desorption into the gas phase is slow (2). Interaction of the compounds with the sample loop was considered to be a likely source of the observed peak tailing. Commercially available sample loops are most often constructed of stainless steel. Frequently, Hastalloy or stainless steel are cited as the material of choice when metal parts are
1883
Table 11. Physical Characteristics of Evaluated Sample Loops sample loop construction material aluminum tubing Teflon tubing stainless steel tubing copper tubing
inside outside diameter, diameter, cm cm
length, cm
vol, cm3
0.17 0.24
0.32 0.32
35.1 16.4
3.0
0.22
0.32
20.5
0.17
0.32
35.1
3.0 3.0
3.0
exposed to the sample (3). The severe peak tailing observed for oxygenated compounds in our laboratory occurred when a stainless steel sample loop was used. Literature references on the selection of construction materials for gas chromatographic sample loops are limited. Graven et al. reported that Teflon and glass were both found to be superior to stainless steel for sample loop construction for the analysis of certain oxygenated compounds ( 4 ) . In contrast, literature references on comparison of construction materials for gas chromatographic columns are more prevalent (5-8). Some researchers have found that materials such as nickel and Teflon perform better than stainless steel for column construction. Columns constructed of nickel or glass are widely available commercially. Due to reports of improved chromatographic performance with columns constructed of these different materials, a comparison of sample loops constructed of several different materials seemed appropriate. Therefore, a study was undertaken to compare the performance of sample loops constructed of several different materials.
EXPERIMENTAL SECTION Apparatus. A Hewlett-Packard 5880A gas chromatograph equipped with a flame ionization detector (FID) and two Valco six-port sampling valves was used for the sample analysis. The cylinder gases were introduced into the system via a Hoke needle valve attached to the cylinder. The gas chromatographic column used was 20-ft X l/s-in. stainless steel tubing packed with 10% SP-lo00on sO/lOO Supelcoport. The carrier gas used was nitrogen at 35 mL/min. the FID fuel gases used were hydrogen at 40 mL/min and air at 400 mL/min. The oven temperature was kept at 120 "C. Procedure. Sample loops having the same internal volume (3.0 cm3) were constructed of several different materials. The materials of construction along with structural dimensions are listed in Table 11. The stainless steel sample loop was installed in the valve oven of the gas chromatograph. A group 2 cylinder containing a multicomponent mixture, which included acetonitrile and methyl ethyl ketone, was attached to the system via the needle valve and allowed to purge the sample loop. The contents of the sample loop were allowed to come to equilibrium with ambient pressure by switching the sample valve so that the sample loop was in common with a vent line and then were swept by the carrier gas onto the head of the chromatographic column. This procedure was repeated until five injections were made. A group 3 cylinder containing acetone and 1,4-dioxaneand a group 4 cylinder containing ethylene oxide and propylene oxide were then each injected five times in the same manner. The stainless steel sample loop was then removed and replaced with each of the sample loops described in Table 11. The same procedure was repeated for each sample loop. RESULTS AND DISCUSSION A tabulation of the data obtained for the compounds which exhibited peak tailing is shown in Table 111. Table I11 lists the percent relative standard deviation of the integrated area counts for each of the compounds investigated on each of the four sample loops. An examination of the data indicates that the sample loop constructed of aluminum exhibited the best precision for the area counts while the sample loop constructed of stainless steel gave the poorest precision. The peak shapes
Anal. Chem. 1987, 59, 1884-1886
1884
Table 111. Percent Relative Standard Deviation for Area Counts Obtained with Evaluated Sample Loops
compd evaluated
copper
acetone 1,4-dioxane methyl ethyl ketone acetonitrile ethylene oxide propylene oxide
sample loops evaluated stainless aluminum steel 1.37 0.97
1.90 1.27
3.42
1.90
17.20 3.54
4.35 1.62 1.12
3.49
4.41
3.26 3.31 20.87 5.76 17.87
Teflon 1.72 1.47 3.11
20.93 2.89 1.97
to drive off possible contaminants. However, this procedure had no effect upon the data obtained when the samples were reinjected with the stainless steel sample loop. Limited work with a sample loop constructed from nickel indicates that nickel is also superior to stainless steel for sample loop construction. The nickel sample loop gave peak shapes similar to the aluminum sample loop. Other construction materials for sample loops, which are considered likely to give good results, are glass or glass-lined stainless steel. However, glass sample loops are difficult to work with because they are easily broken when being installed or removed from the valve oven of a gas chromatograph. In addition, glass-lined stainless steel is relatively expensive and is difficult to handle. As a result of these findings, a 3-cm3aluminum sample loop is now being routinely used for all of the compounds shown in Table I. There have been no observable differences between the results obtained for the aluminum sample loop and the stainless steel sample loop when the nonoxygenated compounds listed in Table I are analyzed. This study has shown that a significant improvement in peak shape and quantitative precision can be obtained for the oxygenated compounds when the aluminum sample loop is used.
ACKNOWLEDGMENT The authors thank Maurice Jackson, Gary Howe, and Alex Gholson for their insightful and constructive suggestions. Registry No. Al, 7429-90-5;Cu, 7440-50-8;H,CCOCH, 6764-1; H3CCOCH2CH,78-93-3; H,CCN, 75-05-8; Teflon, 9002-84-0; stainless steel, 12597-68-1;l,Cdioxane, 123-91-1;ethylene oxide, 75-21-8; propylene oxide, 75-56-9.
LITERATURE CITED
0
2
4
6
Time (rnin) Figure 1. Comparison of peak shapes obtained with (a) stainless steel and (b) aluminum sample loops at 10 ppb concentration level.
obtained from sample loops constructed of aluminum, copper, and Teflon were observed to be similar to one another while the peak shapes obtained with the stainless steel sample loop were far less symmetrical. Figure 1 contrasts typical peak shapes obtained for acetone and 1,4-dioxane between the aluminum and stainless steel sample loops. The stainless steel loop was heated with a hot-air gun for several minutes while being purged with dry nitrogen in order
(1) Laub, R. J.; Pecsok, R. L. physicochemical Applications of Gas Chromatography; Wiiey: New York, 1978. (2) Conder, J. R.; Young, C. L. Physicochemical Measurement by Gas Chromatography; Wiley: New York, 1979. (3) McNair, H. M.; Bonelli, E. J. Basic Gas Chromatography; Varlan: Palo Alto, CA, 1969. (4) Graven, W. H.; Harrison, H. R. Anal. Chem. 1965, 37, 1626. (5) Desty, D. H.; Goldup, A,; Whyman, B. H. J. Inst. Pet. 1959, 45, 287. (6) Bertsch, W.; Shumbo, F.; Chang, R. E.; Zlatkis, A. Chromatographla 1974, 7, 128. (7) Petitjean, D. J.; Hiftoult, C. J. J. Gas Chromatogr. 1963, 3, 18. (8) Fenimore, D. C.; Whitford, J. H.; Davis, C. M.; Ziatkis, A. J. Chromatogr. 1977, 10, 141.
RECEIVED for review January 16, 1987. Accepted March 23, 1987. This project was conducted by Research Triangle Institute, Research Triangle Park, NC, under Contract No. 68-02-4125for the Quality Assurance Division, Environmental Monitoring Systems Laboratory, of the United States Environmental Protection Agency. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
Preparation of Biological Tissue for Determination of Arsenic and Selenium by Graphite Furnace Atomic Absorption Spectrometry Alexander J. Krynitsky U.S. Fish and Wildlife Service, Patuxent Wildlife Research Center, Laurel, Maryland 20708
A great deal of interest has been generated in the measurement of selenium in biological tissue because of the high toxic levels of selenium that were found in the agricultural
drainwater a t the Kesterson National Wildlife Refuge (KNWR), Merced County, California ( I ) . Consequently, a method was needed for the digestion of large numbers of tissue
This article not subject to US. Copyright. Published 1987 by the American Chemical Society
