SEPARATION AND IDENTIFICATION OF BLOOD POLLUTANTS EBERHARDT KUHN, MEERA DATTA, and JASON ELLIS Agilent Technologies Inc. 91 Blue Ravine Rd. Folsom, CA 95630
INTRODUCTION With the high level of alcohol consumption in the US and the growing enforcement of DUI laws, blood alcohol analysis is one of the most common clinical analyses performed today. Headspace Gas Chromatography (GC) is commonly used to determine blood alcohol levels. It is a convenient method because it can be automated and biological products that can cause interferences are not directly injected into the GC system. Dedicated columns have been developed for this analysis and now blood alcohol analysis can be completed in less than 2 minutes (1). However, there are sometimes other pollutants in the blood. Not only is it important to identify those compounds coming from inhaling or ingesting dangerous and controlled substances, but they could potentially interfere with the quantitation of the typical blood alcohols. A list of some of the more common analytes of concern was determined by talking to several analysts working in forensic labs. Separation and identification of several important inhalants are demonstrated on the DB-ALC1 and DB-ALC2 columns. Those columns, although specifically developed for the effective and fast analysis of blood alcohol analytes, allow for the analysis of a wide variety of other potentially harmful substances in blood. The mix of potential pollutants is also run on several different stationary phases. Additionally, window diagramming is used to determine if coupled columns would be capable of resolving each of the blood alcohol analytes from the other pollutants. INHALANTS Inhalants are chemical substances that are quickly inhaled ( huffed ) in order to produce mind-altering effects. The substances used as inhalants are usually materials commonly found in the household and workplace. 2
Inhalants fall into the following categories: Solvents industrial or household solvents or solvent-containing products, including paint thinners or solvents, degreasers (dry-cleaning fluids), gasoline, and glues art or office supply solvents, including correction fluids, felt-tip-marker fluid, and electronic contact cleaners Gases gases used in household or commercial products, including butane lighters and propane tanks, whipping cream aerosols or dispensers (whippets), and refrigerant gases household aerosol propellants and associated solvents in items such as spray paints, hair or deodorant sprays, and fabric protector sprays medical anesthetic gases, such as ether, chloroform, halothane, and nitrous oxide (laughing gas) Nitrites aliphatic nitrites, including cyclohexyl nitrite, which is available to the general public; amyl nitrite, which is available only by prescription; and butyl nitrite, which is now an illegal substance. (2) Inhaling the above substances can have a variety of health effects ranging from reversible effects (liver and kidney damage, blood oxygen depletion) to irreversible effects (hearing loss, peripheral neuropathies, central nervous system or brain damage, bone marrow damage). Additionally many people die from inhalants due to suffocation caused by the displacement of oxygen in the lungs and then the central nervous system until breathing ceases (2). Inhalants are most commonly used by teenagers. Approximately 12% of 8 th graders, 9% of 10 th graders, and 7% of 12 th graders reported using inhalants in the in the 1999 survey on drug use conducted by the University of Michigan s Institute for Social Research. The good news is that inhalant use by adolescents appears to be declining over the last few years (3). 3
WINDOW DIAGRAMMING When dealing with complex sample mixtures it is sometimes difficult to find a single analytical column that will be capable of resolving all of the analytes in the mixture. Window diagramming is a technique that can be used to determine if a combination of stationary phases can be used to resolve all of the components of a mixture that a single phase can not. The analytes of interest are run on two phases under the same conditions. The distribution constant (K c ) is calculated for each analyte on both phases: K c = kβ k = (t r -t m )/ t m β = r c /2d f t r = retention time of the analyte t m = retention time of a non-retained analyte r c = radius of the capillary column (µm) d f = film thickness (µm) Figure 1 shows an example of a simple window diagram. The distribution constants for phase A (DB-ALC2) are plotted on the left y axis and the distribution constants for phase B (DB-1) are plotted on the right y axis. Lines are drawn to connect the distribution constants for each analyte on both phases. The diagram is then evaluated to try to find points along the x axis where the lines are not crossing these points indicates the percentage of each phase that should be used to resolve the analytes. (4,5) 4
Figure 1. Example of Windows Diagramming K c (DB-ALC2) Acetonitrile IPA Ethyl formate Ethanol Methanol Acetone n-propanol t-butanol K c (DB-1) Acetaldehyde 100% DB-ALC2 35% DB-ALC2 65% DB-1 100% DB-1 Percent Phase Composition EXPERIMENTAL The GC used in this study was a Agilent Technologies 5890 Series II (Wilmington, DE). The capillary columns used were from Agilent Technologies (Folsom, CA): DB-ALC1: 0.53mm, 30m, 3.00µm, proprietary stationary phase. DB-ALC2: 0.53mm, 30m, 2.00µm., proprietary stationary phase. DB-WAX.etr: 0.53mm, 30m, 2.00µm, 100% PEG stationary phase. DB-MTBE: 0.53mm, 30m, 3.00µm,, proprietary stationary phase. The conditions specific to the analysis are listed on the chromatograms in the results section. The chemicals were from Aldrich Chemical Company (Milwaukee, WI). Individual compound standards of some of the more common blood pollutants were made up in CS 2 with ethanol as the internal standard. The individual component standards were analyzed to determine elution order on each column. A mixture of all components (minus the 5
isopropylamine due to chemical interactions) was also run on each column. t-butyl alcohol was included in the mixture because it is commonly used as the internal standard for the blood alcohol analysis in Europe. Distribution constants from earlier work were used for the following columns (6): DB-624: 0.53mm, 30m, 3.0µm, (6%cyanopropyl-phenyl)- methyl polysiloxane stationary phase. DB-1: 0.53mm, 30m, 3.0µm, 100% Dimethylpolysiloxane stationary phase. DB-WAX: 0.53mm, 30m, 1.0µm, 100% polyethylene glycol stationary phase. The appropriate lengths of columns determined by window diagramming were coupled using press-fit unions with polyimide sealing resin. The mixture of the components and individual standards were analyzed on the coupled columns. RESULTS AND DISCUSSION The list of analytes used in this are listed in Table 1. The analytes that are part of the blood alcohol analysis are noted with asterisks. t-butyl alcohol is also indicated with an asterisk as it is the internal standard for the European blood alcohol analysis. Figures 2 and 3 show the mixture of possible pollutants run on the DB-ALC1 and DB-ALC2 respectively. Between the two columns most of the analytes are resolved. However, there are a few pairs of analytes that co-elute on both columns: acetone/ethyl formate, acetonitrile/t-butyl alcohol, chloroform/sec-butyl alcohol, and 1,1,1-trichloroethane/carbon tetrachloride. It should also be noted that t-butyl alcohol and acetone co-elute on the DB-ALC1, which is a concern for the European blood alcohol method. Figure 4 shows the mixture on the DB-MTBE. The DB- MTBE is not able to resolve methanol and acetaldehyde 6
two of the analytes in the blood alcohol analysis. However, this column in combination with the DB-ALC1 or the DB- ALC2 resolves all the components used in this study and would be useful as a confirmation column. Figure 5 is the combination column determined by window diagramming the retention data of the DB-624 and the DB-MTBE. It is made up of 37% DB-624 and 63% DB-MTBE. Once again there are a few co-elutions but the combination of the DB-ALC1 or the DB-ALC2 and the 624/MTBE mixture is able to resolve all of the analytes. The resolution of the methanol and acetaldehyde is too close for this column to be used as a single column in the analysis of blood alcohols. Figures 6 and 7 are two other combination columns (Combination A and B respectively) that were put together based on window diagramming of other stationary phases. Combination A would make a good confirmation column with either DB-ALC1 or DB-ALC2 but once again the methanol/acetaldehyde separation is too close for it to make a valid stand alone blood alcohol column. Combination B does an excellent job of resolving the 7 blood alcohol components and could be used for a stand alone blood alcohol column as well as a confirmation column with either the DB-ALC1 or DB- ALC2. CONCLUSION The DB-ALC1 and DB-ALC2 columns work very well for what they were designed for the fast analysis of blood alcohols. However, there are a couple of co-elutions on both columns that limit their applicability to an extended analysis of other potential blood pollutants. The DB-MTBE, DB-624/DB-MTBE, and Combination A Column could all be used as confirmation columns with the DB-ALC1 or DB-ALC2 but are unable to stand alone as blood alcohol columns due to the limited resolution of methanol and acetaldehyde. 7
The combination B column shows promise as a possible third blood alcohol column. Further study needs to be completed. ACKNOWLEDGEMENTS A special thanks to Katherine Warniment of the FDLE for her technical advice. REFERENCES 1. Gay Sun. The Effect of GC Column Inner Diameter on Blood Alcohol Analysis. Poster #1617, Pittsburgh Conference 1999. Reprints available through J&W Scientific, 91 Blue Ravine Rd. Folsom, CA 95630. 2. NIDA INFOFAX, Inhalants 13549 (on-line). Available www.nida.nih.gov; accessed 02/16/00. 3. LD Johnston, PM O Malley, JG Bachman. (Dec. 1999). Drug trends in 1999 are mixed. University of Michigan News and Information Services: Ann Arbor, MI. [On-line]. Available at: www.monitoringthefuture.org; accessed 02/16/00. 4. Dean Rood. Optimizing Separations Using Coupled Columns and Window Diagramming. LC GC Vol. 17 #10, p. 914-22 (Oct. 1999). 5. Walt Jennings, E. Mittlefehldt, P. Stremple. Analytical Gas Chromatography, 2 nd Edition. Academic Press, 1997, p. 4-7. 6. Solvent Retention Data. J&W Scientific Application Note. J&W Scientific, Folsom, CA. 8
Table 1. Compound list for Figures 2-7 1. Methanol* 2. Acetaldehyde* 3. Ethanol* 4. Diethyl ether 5. Isopropyl alcohol* 6. Methylene chloride 7. Acetone* 8. Acetonitrile 9. Ethyl formate 10. t-butyl alcohol* 11. 1-Propanol* 12. MTBE 13. Hexane 14. Chloroform 15. sec-butyl alcohol 16. 2-Chlorobutane 17. MEK (2-Butanone) 18. Ethyl acetate 19. 1,1,-Trichloroethane 20. Carbon tetrachloride 21. 1-Chlorobutane 22. Benzene 23. 1-Butanol 24. Heptane 25. Ethylene glycol. Isoamyl alcohol. Toluene 28. Isopropyl amine (not shown) 29. Ethylbenzene 30. m,p-xylene 31. o-xylene 32. DMSO Figure 2. DB-ALC1 30 m x 0.53 mm I.D., 2.0 µm 30 7*,8,9,10* 31 28,29 1* 5* 21, 22 23 3*, 4 11* 13 15 19, 20 17,18 24 32 2* 6 12 14 16 25 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 Figure 3. DB-ALC2 30 m x 0.53 mm I.D., 2.0 µm 30 5*,12,13 8,10* 31 29 4 1* 3* 14,21,24,15 22 17,19,20 23 2* 7*,9 11* 16 18 32 6 25 28 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 9
Figure 4. DB-MTBE 30 m x 0.53 mm I.D., 3.0 µm 30 31 23,21,25 29 1*,2* 8 3* 10* 4,9 7* 5* 6 11* 14,16 13 15 17 18 12 19 20,22 24 32 28 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 Figure 5. 37% DB-624/63% DB-MTBE 35 m x 0.53 mm I.D., 3.0 µm 30 18,15,16 31 1* 4,5* 29 6,10* 12,11* 22 7* 3* 8 23 24 13 2* 9 17 21 32 14 19 20 25 28 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 Figure 6. Combination A 30 m x 0.53 mm I.D., 3.0 µm 29 30 31 6,10* 8 20,22 1* 7* 3* 5* 4 12,11* 15,16 13 21 23 24 33 9 17 18 2* 19 14 25 28 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 10
Figure 7. Combination B 30 m x 0.53 mm I.D., 3.0 µm 30,32 31 29 20,22 10* 1* 8 4,9 5* 7* 18,15 13,16 21 23 24 3* 11* 17 2* 6 12 14 19 28 25 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 11