Intra-sample vs. Inter-sample Variability in Architectural Paint

The aim of this project was to determine whether any intersample variations between architectural paint products of the same brand could be established. Different batches of the same product were analysed to test whether they could be distinguished. Samples analysed included batches of white paint manufactured over several months and red and blue colour samples. Samples were analysed by microscopy, microspectrophotometry (MSP), Fourier Transform Infra-Red (FTIR) and micro-XRF. Pyrolysis-GC/MS was also used to analyse a subset of the samples. The samples were compared by visual inspection of the data and by multivariate techniques for the complex data sets resulting from micro-XRF. The results showed that, at the brand level, the techniques tested were highly discriminatory. However, the variations within a single batch were equal to or greater than the variations between batches of a single product. There was no consistently successful technique for distinguishing between batches of the same product; nor was the overall analysis scheme sufficiently discriminating. These results suggest that when significant differences are observed between architectural paint samples it is strong evidence that the samples did not come from the same source. If no significant variations are found, they are probably of the same brand. In this case, the analyst will require national market data to assess the defence hypothesis and the strength of the evidence. Introduction Paint is ubiquitous in modern life and readily transferred by contact. It is commonly found at crime scenes. One of the questions which often arises is “what is the chance of a coincidental match with paint comparison evidence” (defence hypothesis)? Several topics need to be investigated to answer this question. Firstly, can the investigator discriminate between different brands of paint? Secondly, what is the type of paint and how common is it in the general population? If the paint is a common type, can the investigator discriminate within the brand (batch discrimination)? Finally, are there any post-manufacturing points of discrimination (eg: tinting, blending)?