Chemical fingerprinting of gasoline: 2. Comparison of unevaporated and evaporated automotive gasoline samples
Introduction
Conditions at a crime scene before, during and after a fire are uncontrolled. It is, therefore, likely that the gasoline sample recovered from a fire scene will have a different level of evaporation than the gasoline sample seized from a suspect. Furthermore, control samples of gasoline obtained from local service stations by arson investigators will not be evaporated at all (if collected and stored appropriately). Having gasoline samples at different levels of evaporation will increase the complexity of the forensic comparison of those samples.
There are a number of studies in the scientific literature that have attempted to address the problem of comparing samples of unevaporated and evaporated gasoline [1], [2], [3], [4], [5], [6]. Mann [1] used the chromatographic profiles of some of the more volatile compounds to differentiate unevaporated samples of gasoline. In a study of evaporated gasoline, Mann [2] observed that evaporated samples could be compared to less evaporated or unevaporated samples using the relative ratios of compounds that had similar vapour pressures. Although able to make limited comparisons between evaporated and unevaporated gasoline samples, the author concluded that the method was most useful in “eliminating the possibility of common origin between two samples” [2]. Hirz [3] compared chromatographic profiles of high boiling components of evaporated gasoline samples obtained from nine different refineries in Europe and was able to classify, in a limited way, which samples came from a particular refinery. Potter [4] used principal component analysis (PCA) to explore how different isomeric groups could be used to compare a small number of evaporated and unevaporated samples of gasoline. Potter [4] was unable to satisfactorily compare a 50% evaporated gasoline sample with the original, unevaporated sample. It has been shown that three-dimensional fluorescence spectroscopy could not only distinguish different samples of gasoline, [5] but could also match a sample evaporated up to 50% back to the original, unevaporated gasoline sample [6]. Recently, we reported the discrimination of 35 unevaporated gasoline samples into 32 unique groups using the C0- to C2-naphthalene profile obtained from GC–MS (SIM) data [7]. These higher boiling compounds were selected for the characterisation of gasoline samples, in part, to facilitate the comparison of evaporated samples of gasoline [7].
Environmental regulations have resulted in similar comparisons being attempted between evaporated and unevaporated crude oils. For example, Wang and Fingas [8] have discussed the use of isomer ratios to characterise and compare crude oil spills. Wang and Fingas [9] also studied and quantitated the changes that occur when a light crude oil is evaporated and even devised a “Weathering Index” to allow the degree of evaporation of a crude oil to be estimated. More recently, Smallwood et al. [10] used carbon isotope ratios for selected compounds in three gasoline samples to successfully compare an evaporated sample to its original, unevaporated sample.
The significance of a forensic comparison between two gasoline samples is dependent on an understanding of how gasoline composition changes with time. The degree to which gasoline samples, collected from a single service station, change with time can provide a court of law with important information that will allow the court to assess the significance of the gasoline comparison. If, for example, it can be demonstrated that gasoline samples taken from a single service station can be differentiated from week to week then the court may place greater evidential weight on a comparison between two samples found to be similar. This fact was recognised by Mann [1]. Mann collected six samples from a single service station over a period of 28 days. The results of Mann’s work showed that, as long as the distribution terminal received new shipments of gasoline from the refinery, differences between deliveries to a single service station did exist [1]. To our knowledge, no other published work has examined the change in gasoline composition with time at a single service station.
The use of the C0- to C2-naphthalene profile obtained from gas chromatography-mass spectrometry with selected ion monitoring (GC–MS (SIM)) data to discriminate samples of gasoline has been recently reported [7]. The aim of the first part of this study is to apply this method to 35 samples of gasoline, each at five different levels of evaporation (0, 25, 50, 75 and 90%), in order to assess the relative degree of change in the C0- to C2-naphthalene profile and to examine the ability of the method to differentiate between different samples of gasoline irrespective of their level of evaporation. The aim of the second part of this study is to expand Mann’s original work [1] from one service station to three (each one representing a different brand of station), and to extend the collection time from 4 to 16 weeks. This information will provide a better understanding of how the C0- to C2-naphthalene composition in gasoline changes with time at a service station and how it differs between stations over time.
Section snippets
Unevaporated and evaporated gasoline samples
Thirty-five samples of unleaded gasoline (regular, premium and lead replacement) were obtained from 24 service stations in metropolitan Sydney, Australia (Table 1) over a 7-month period (March to September 2001). The majority of the samples (29) were collected over a period of seven weeks (March to April 2001). Samples were collected in 125 ml amber, Boston round glass bottles sealed with polypropylene caps fitted with Teflon-faced foamed polyethylene liners (Wheaton) and stored in the dark at
Overview of results for unevaporated and evaporated gasoline samples
For each of the 35 samples, the level of evaporation was brought as close to the target level of 25, 50, 75 or 90% evaporated by weight as possible. The level of evaporation achieved for the 35 samples at these four evaporation levels is summarised in Table 2. It was observed that at each level of evaporation not all samples evaporated at the same rate. It also became progressively easier to monitor and control evaporation rates as the target evaporation levels increased. The rate of
Unevaporated and evaporated gasoline samples
The fact that grouping of samples by LDA was found to be fairly consistent for the 25, 50 and 75% evaporated samples (Table 4) is a significant finding. The significance of this result is that one would be more likely to perform a forensic comparison on samples at these levels of evaporation than on samples that are either unevaporated or 90% evaporated. Evaporation of forensic samples to between 25 and 75% by weight is realistic because liquid gasoline samples recovered from the scene or
Conclusions
This study has shown that the C0- to C2-naphthalene profile obtained from GC–MS (SIM) data can be used to classify and compare different samples of gasoline, irrespective of their level of evaporation. The division of gasoline samples into unique groups is possible even when the samples span a wide range of evaporation levels (0, 25, 50, 75 and 90% by weight). The results presented in this article show that a forensic comparison of gasoline samples collected on the same day can be made even if
Acknowledgements
The authors wish to acknowledge the contribution of the Centre for Forensic Science, University of Technology, Sydney, Australia where this research was conducted. The authors gratefully acknowledge the co-operation of the owners and employees of the three service stations involved in the 16-week collection over time study. The authors wish to thank ExxonMobil (USA), the Natural Sciences and Engineering Research Council (Canada), and the Australian Technology Network via the University of
References (19)
Gasoline brand identification and individualization of gasoline lots
J. Forensic Sci. Soc.
(1989)- et al.
Chemical fingerprinting of unevaporated automotive gasoline samples
Forensic Sci. Int.
(2003) - et al.
Differentiation of the source of spilled oil and monitoring of the oil weathering process using gas chromatography–mass spectrometry
J. Chromatogr. A
(1995) - et al.
Stable carbon isotope composition of gasolines determined by isotope ratio monitoring gas chromatography mass spectrometry
Org. Geochem.
(2002) Reduction of error propagation due to normalization: effect of error propagation and closure on spurious correlations
Anal. Chem.
(1995)The determination of tetraalkyl lead compounds in petrol using combined gas chromatography atomic absorption spectrometry
Forensic Sci. Int.
(1981)Volatiles from carpet: a source of frequent misinterpretation in arson analysis
J. Chromatogr. A
(1994)Comparison of automotive gasolines using capillary gas chromatography. I. Comparison methodology
J. Forensic Sci.
(1987)Comparison of automotive gasolines using capillary gas chromatography. II. Limitations of automotive gasoline comparisons in casework
J. Forensic Sci.
(1987)
Cited by (77)
Review of contemporary chemometric strategies applied on preparing GC–MS data in forensic analysis
2022, Microchemical JournalPreservation of vapor samples on adsorbent alumina capillaries and implications for field sampling
2021, Journal of Chromatography AA method for forensic gasoline comparison in fire debris samples: A numerical likelihood ratio system
2020, Science and JusticeCitation Excerpt :Already in 1987 [2,9], the scientific literature identifies as key issues for a ‘gasoline comparison’: evaporation, microbial degradation and pyrolysis effects. Concerning classification of evaporated gasoline samples to a limited number of neat gasoline samples, a number of papers address evaporation under laboratory conditions [10–13]. In these studies, samples of gasoline are evaporated to a desired percentage of weight loss, and it is tried to link the evaporated gasoline to the original gasoline and discriminate it from a number of other gasoline samples.