BIOLOGICAL MASS SPECTROMETRY, VOL. 20, 493-497 (1991)

Determination of Kerosene and Light Oil Components in Blood K. Kimura, T. Nagata,t K. Kudo and T. Imamura Department of Forensic Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812, Japan

K. Hara Department of Legal Medicine, School of Medicine, Fukuoka University, 7-45-1 Nanakuma, Jonan-ku, Fukuoka 814-01, Japan

A sensitive and rapid method to analyse fuel components in blood from rats exposed to kerosene or light oil vapour was developed by making use of capillary gas chromatography/mass spectrometry. The aliphatic hydrocarbons with carbon numbers 8-10 and aromatics such as toluene, xylene, 3- and kthyltoluene and trimethylbenzenes were clearly detected in blood from rats exposed to kerosene or light oil vapour, using the head-space method c o m b i d with the salting-out technique. The concentration ratio of pseudocumene to toluene in blood exposed to tight oil was higher than that in the case of exposure to kerosene. The lower limits of detection were 50 pg and 1 ng in toluene and pseudocumene, respectively. Our suggestion is that this method is useful in forensic investigations to detect fuel components in blood and for the purposes of differentiating kerosene and light oil in blood tissues.

INTRODUCTION From the aspect of forensic medicine, burning of a body before death has been confirmed based on factors of erythema, bulla, soot in the airway and a high concentration of carboxyhaemoglobin (CO-Hb) in blood. However, these vital signs may be ambiguous when fuel substances used are inflammable materials. For practical purposes, the detection of fuel components from biological tissues exposed to fuels is extremely significant to elucidate not only the cause of death, but also conditions before death. Although there are reports on analyses of fuel components in body tissues or in arson residues, based on gas chromatography (GC)'-'' or gas chromatography/mass spectrometry (GC/MS),' '-I3 these represent general screenings and detailed studies remained to be done. In our ongoing study on analysing fuels, a reliable method involving use of GC/MS made feasible detection of fuel components in blood exposed to gasoline or kerosene vapour, and the two fuel components could be differentiated.14*1 In recent years, light oil has been commonly used for fuel in diesel engines. This fuel is refined from crude oil under conditions of much the same temperature used in the case of kerosene. We attempted to analyse oil present in blood, and to differentiate between kerosene and light oil exposure. Capillary GC/MS combined with the head-space method and the salting-out technique was used for these purposes.

t Author to whom correspondence should be addressed. 1052-9306/9 1/08049345 $05.00

0 1991 by John Wiley 8~Sons, Ltd.

MATERIALS AND METHOD

Reagents n-Paraffins, branched parafins and olefin kits were provided by Gasukuro Kogyo Inc., Tokyo, Japan. Toluene and pseudocumene were obtained from Ishizu Seiyaku Co. Ltd., Osaka, Japan. Toluene-d8 was purchased from Aldrich Chemical Co., Milwaukee, USA. All reagents used were of guaranteed grade. Kerosene and light oil obtained from three different oil companies were mixed and used as standard fuels. Preparation of standard solutions Standard solution for quantitative analysis was prepared by dissolving 1 g each of toluene and pseudocumene in 100 ml ethyl alcohol. This solution was further diluted with ethyl alcohol to give concentrations of 0.01, 0.05,0.1,0.5 and 1.0 pg pl-'. Toluene-d8 used as an internal standard (IS) was prepared at a concentration of 1.0 pg pl-' in ethyl alcohol. Experimental animals Eight male Wistar rats weighing 250-300 g were used. Four rats were independently exposed to 1800 p.p.m. kerosene vapour for 1 h and the other four to lo00 p.p.m. light oil vapour, under equal conditions. The temperature of the water bath and the evaporating chamber was set at 45 "C. The concentration of fuel Received 6 January 1991 Revised 25 March 1991

494

K. KIMURA ET AL.

vapour was measured directly by the attached gas detector, model SP 237H, owing to the relatively low concentration of kerosene and light oil vapour, while the gasoline vapour was diluted 10 or 20 times so as to allow determination within the detection range of the meter used in previous work.16 Blood samples were collected from the heart immediately after pithing the rats and all samples were kept at -20°C until analysis to prevent the evaporation of fuel components and to ensure equal conditions of preservation. Human whole blood was used as the control sample.

Analytical method Qualitative analysis. One millilitre of distilled water and

1 g of sodium chloride-salt for the salting-out technique-were put into a 17ml volume glass bottle. This preparation was then cooled in a refrigerator. After adding 2 g of whole rat blood, the bottle was tightly stoppered with a silicone plug and was placed in a water bath at 55 "C for 20 min with occasional shaking. One millilitre of the head-space vapour was then aspirated off and injected into a GC/MS instrument. Qualitative analysis was made by mass chromatography (MC). Fuel components detected in MC were identified, based on comparisons of retention times and mass spectra of the peaks with those of standard substances.

Quantitative analysis. Each of 0.5 g of whole blood and 1

p1 of IS solution were put into a glass bottle treated in the same manner as for the qualitative analysis. The sample was equilibrated and 0.5 ml of vapour phase was subjected to GC/MS analysis. Quantitative analysis of toluene and pseudocumene was made by selected ion monitoring (SIM). The monitoring ions were set at m/z 92, 100 and 105. The m/z 92 and 100 ions were the molecular ions of toluene and toluene-d8, respectively. The m/z 105 ion ([M-CH,]+) was selected to monitor the base peak ion of pseudocumene, so as to prevent any decrease in sensitivity.

Conditions of GC/MS The instrument used was a Shimadzu QP-lo00 gas chromatograph/mass spectrometer run in electron impact (EI) mode and controlled through a data system. The fused-silica capillary column (10m x 0.53mm i.d., 2.65 pm film thickness) was coated with Hewlett Packard HP- 1 cross-linked methylsilicone gum phase. The column oven temperature was programmed from 50°C (2 min) to 200°C at the rate of 10°C min-'. Other operating temperatures were as follows: injection port, 200 "C; separator, 250 "C; ion source, 250 "C. The carrier gas was helium at a flow rate of 15 ml min-'. The ionization energy was set 20 eV in MC and 70 eV in SIM.

RESULTS AND DISCUSSION Qualitative determination Human whole blood (control blood; 2 g) containing 200 nl standard kerosene (158 pg) or light oil (165 pg) was vaporized and analysed by MC. The obtained mass chromatograms are shown in Fig. 1. Nearly 40 peaks appeared on the total ion intensity chromatogram (TII), and the m/z values ranged from 57 to 120 in both samples. The components were identified as aliphatic and aromatic hydrocarbons with carbon numbers of 7-17 in kerosene (Fig. l(a)) and 7-18 in light oil (Fig. l(b)). Although most of the components were common in kerosene and light oil, as seen in Fig. 1, n-octadecane was detected only in light oil vapour. When the fuel vapour was inhaled by the rats, the number of peaks on the chromatograms markedly decreased to one-fourth of fuel-spiked blood, as shown in Fig. 2. Low volatile compounds with a carbon number over 11 were not observed. Aliphatic hydrocarbons with carbon numbers 8-10 and aromatics such as toluene, xylene, n- and isopropylbenzene, 3- and 4-ethyitoluene and trimethylbenzenes were detected in both blood samples. Despite the close chromatographic patterns of kerosene and light oil components, findings which differentiated these components were: extremely high peak of toluene in the case of kerosene exposure, and markedly higher peaks of 3- and 4- ethyltoluene and pseudocumene in blood in the case of exposure to light oil. Among these aromatic hydrocarbons, toluene and pseudocumene were clearly separated on the mass chromatogram, hence these two aromatics were selected for the quantitative analysis by SIM. Quantitative determination Each of the calibration curves obtained by plotting the peak area ratio of toluene and pseudocumene to the IS solution showed a straight line, in concentrations ranging from 0.01 to 1.0 pg per 0.5 g sample, as shown in Fig. 3. The correlation coefficients of the calibration curves for toluene and pseudocumene were 0.998 and 0.986, respectively. The lower limits of detection were 50 pg for toluene and 1 ng for pseudocumene. The mean values of the concentration of toluene and pseudocumene in blood of rats exposed to kerosene vapour were 0.51 and 0.56 pg g-' (n = 8); those of light oil exposed blood were 0.40 and 0.90 pg g-' (n = 8). The concentration ratio of pseudocumene to toluene in each sample was 1.13 & 0.19 for kerosene exposure and 2.30 & 0.15 for light oil exposure. This concentration ratio was significant for a differential determination between kerosene and light oil. CONCLUSION Our capillary GC/MS analysis combined with the headspace method and the salting-out technique provides a

KEROSENE AND LIGHT OIL COMPONENTS IN BLOOD

495

a

!y: ."

1

fi3

92 106

-8

120 ~ ' ' ' ' 1 ' ' " 1 ' ' ' ~ 1 ' ' ' ' 1 1 ' ' ' ~ ' ' ' ~ ~

50

100

150

200

250

300

350

SCAN NO.

b

TII c9

50

c12

100

150

200 SCAN NO.

250

300

35v

Figure 1. Mass chromatograms of kerosene and light oil: (a) vaporized standard kerosene; (b) vaporized standard light oil. C8,n-octane; C9, n-nonane; C10,n-decane; Cll,n-undecane; C12;n-dodecane; C13;n-tridecane; C14,n-tetradecane; C15,n-pentadecane; C16, n-hexadecane; C17, n-heptadecane; C18, n-octadecane; 1, toluene, 2, m- and p-xylene; 3, o-xylene; 4, isopropylbenzene; 5, npropylbenzene; 6,3-and 4-ethyltoluene; 7,mesitylene; 8,pseudocumene; 9, 1,2,3-trirnethylbenzene.

K. KIMURA ET AL.

496

a

1

92

I

I

A

105 1O G

I

I

I

I

50

I

, , I I 100

,

,

,

,

,

,

,

I

,

150

,

200

,

300

250

120

, , ) ,

, , , (

)

350

SCAN NO.

b

c9

57 85

92 105

1OG 120 #

I

,

,

1 ' ' ' ' 1

50

100

. " ' I " ' '

150 SCAN

I " "

200

I "

250

I

.

I

300

,

,

.

1

350

NO.

Figure 2. Mass chromatograms of kerosene and light oil: (a) from the blood of a rat exposed to 1800 ppm kerosene vapour; (b) from the blood of a rat exposed to 1000 ppm light oil vapour. Abbreviations as Fig. 1.

KEROSENE AND LIGHT OIL COMPONENTS IN BLOOD

0

0.1

Figure 3. Calibration curves: -@ .,

497

0.5

calibration curve of toluene; A-A, calibration curve of pseudocumene.

reliable qualitative and quantitative determination of small amounts of kerosene and light oil compounds in the blood. The differentiation between exposure to kerosene and to light oil was feasible, based on the concentration ratio of pseudocumene to toluene, in addition to the peak pattern of toluene, 3- and

4-ethyltoluene and pseudocumene on the mass chromatogram.

Acknowledgement We thank M. Ohara for helpful comments.

REFERENCES 1. U. P. Schlunegger,Arch. Toxicol, 22,252 (1967). 2. V. E. Bruce and F. A. Mark, J. Forensic Sci. 13, 76 (1968). 3. R. J. Nelms, R. L. Davis and J. Bond, Am. J. Clin. Pathol. 53, 641 (1970). 4. R. W. Clodfelter and E. E. Hueske, J. Forensic Sci. 22, 116 (1977). 5. P. J. Loscalzo, P. R. DeForest and J. M. Chao, J. Forensic Sci. 25,162 (1980). 6. B. Shankles, S. 8. Weinberg and L. A. Del Cortivo, J. Anal. Toxicol. 6, 241 (1982). 7. A. Carnevale, M. Chiarotti and N. de Giovanni, Am. J. Forensic Med. Pathol. 4, 153 (1983). 8. E. S. Case, A. Poklis and A. Mackell, J. Forensic Sci. 30.208 (1985).

9. A. A. Tony and 0. Mary, J. Forensic Sci. 31,666 (1986). 10. H. J. Kobus, K. P. Kirkbride and A. Maely, J. Forensic Sci. SOC.27,307 (1987). 11. M. H. Mach, J. Forensic Sci. 22, 348 (1977). 12. R. M. Smith, J. Forensic Sci. 28, 318 (1983). 13. J. Ikebuchi, S. Kotoku, M. Yashiki, T. Kojima and K. Okada, Am. J. Forensic Med. Pathol. 7, 146 (1986). 14. T. Nagata, M. Kageura, K. Hara and K. Totoki, Jpn. J. Leg. Med. 31,136 (1977). 15. K. Kimura, T. Nagata, K. Hara and M. Kageura, Hum. Toxicol. 7,299 (1988). 16. K. Kimura, K. Hara and T. Nagata, Forensic Sci. Int. 40,57 (1989).

Determination of kerosene and light oil components in blood.

A sensitive and rapid method to analyse fuel components in blood from rats exposed to kerosene or light oil vapour was developed by making use of capi...
301KB Sizes 0 Downloads 0 Views