58 Research Article Received: 10 September 2008 Accepted: 19 September 2008 Published online in Wiley Interscience: 14 November 2008 (www.interscience.com) DOI 10.1002/xrs.1116 X-ray microanalysis of glass for forensic purposes – research on the persistence of glass fragments on clothing Zuzanna Bro ˙ zek-Mucha * Thirty-five experiments were carried out, breaking various kinds of multiple-glazed windows (ordinary, toughened and laminated) simulating a break-in or an act of vandalism, in order to obtain a set of data on the persistence of glass micro-traces on the clothing of people taking part in such an activity. The recovered glass fragments were analysed by means of scanning electron microscopy coupled with energy-dispersive X-ray spectrometry. The number of glass fragments recovered from the clothing of the breaker and the accompanying persons depended on the distance of a person from the window pane, the time between the incident and securing the clothing for examination, the retentive properties of the clothing, and the kind of window panes. An empirical study on background levels of glass on clothing, as well as secondary transfer, was also carried out. Results of the study of glass found at random showed that most of the clothing was free from glass fragments. The experiments simulating break-ins or vandalism provided important information that will contribute to a reliable interpretation of the analytical results of examinations of glass as evidence. Copyright c  2008 John Wiley & Sons, Ltd. Introduction Fragments of broken glass are well established evidential material in a variety of crimes such as car accidents, burglaries, assaults and acts of vandalism. The recovery of glass fragments as evidential material (without a sample for comparison) enables the examiner to identify the type of glass and to classify it. [1] However, as soon as a comparison sample is provided, the task of comparison or discrimination can be performed. In the case of a positive match it is still a group identification, but the size of the group is being narrowed down and it may make a significant contribution to the solution of a forensic problem. [2] Differences in physical and chemical properties of various types of glass are used by forensic examiners for discrimination and classification. One of the most popular and efficient methods for discrimination among glass micro-fragments is the thermo- immersion measurement of their refractive index by means of the Glass Refractive Index Measurement (GRIM) method. [3–7] However, this method is not suitable for classification of the majority of popular types of glass since the ranges of their refractive indices overlap. A simultaneous classification and discrimination can be achieved by elemental analysis, such as scanning electron microscopy coupled with energy-dispersive X-ray (SEM-EDX) spectrometry, [8 – 10] micro-X-ray fluorescence (μ- XRF) [11] and inductively coupled plasma-mass spectrometry (ICP- MS) itself, or in combination with laser ablation (LA-ICP-MS). [12 – 14] The problems of both discrimination and classification of glass micro-traces cannot be solved without taking into account a possible lack of homogeneity of the properties of glass within an object. One of the examples can be anisotropy of the refractive index of float glass, as well as the presence of tin in the surface of the glass which was in contact with a bed of molten tin during the float process. [4] This enables a forensic scientist to recognise float glass and use this information for classification purposes. However, in order to make comparisons between glass samples, the bulk rather than the surface fragments are taken into account as they are more representative of a glass object. [6] The administration of justice raises new challenges for forensic examiners. Thus, an important part of an expert’s report becomes an evaluation of the probability of the evidence if the suspect did or did not commit the crime. It requires taking into account the objective data, i.e. analytical data obtained from the analysis of the evidential material such as glass fragments, data obtained from population studies (e.g. on the recovery of traces from clothing), as well as subjective data, i.e. information on the crime. Much work has already been done on this kind of evaluation of the evidence in the field of glass traces examined by means of GRIM, allowing the description of the evidence with one variable. However, mathematical bases for the likelihood ratio method of evaluation of glass fragments described by one or more variables have recently been described. [15 – 17] The conclusions concerning problems of primary transfer, persistence, glass found by chance and secondary transfer were drawn from results of experiments in which a single glass pane was broken. The influence of particular factors such as the time elapsed between the event and the collection of the breaker’s clothing, the distance of a person from the broken window, and the retentive proprieties of clothing were analysed by authors in various countries and discussed by Curran et al. [4] They were a valuable contribution to forensic knowledge with regard to general trends. However, the experience gathered in one laboratory in one part of the world cannot be directly used by another one. The properties of glass, and the results of its analysis for forensic purposes, depend on the climate, natural resources, people’s habit of dressing, and the technology of both glass production ∗ Correspondence to: Zuzanna Bro˙ zek-Mucha, Institute of Forensic Research, Westerplatte 9, 31-033 Krakow, Poland. E-mail: zbrozek@ies.krakow.pl Institute of Forensic Research, Westerplatte 9, 31-033 Krakow, Poland X-Ray Spectrom. 2009, 38, 58–67 Copyright c  2008 John Wiley & Sons, Ltd.