Chemometric discrimination of bleached and dyed human scalp hair using attenuated total reflectance infrared spectrocopy

Abstract

The range of products, different formulation, and variables used in cosmetic treatments hold out great potential for forensic identification of hair evidence although in reality little of that potential is realized due to preferences on DNA testing and lack of analytical chemistry expertise among forensic examiners. It is therefore of interest to produce a rapid data acquisition technique that can classify cosmetically treated human hair for forensic application. Six female donors with natural black hair underwent a series of cosmetic hair treatments namely bleaching and dyeing. The following hair strands were collected; natural (control), bleached, day-1 dyed hair/ week 0, week 2, week 4, week 6 and week 8. Statistical interpretation of the triplicate absorbance readings of 126 hair samples, determined using ATR-FTIR was used to classify the type of treatments, the two different brands and weekly intervals of collected hair samples. A wavenumber region of hair protein variability from 1750 to 800 cm-1 was selected for pattern recognition analysis, Principal Component Analysis (PCA) and Hierarchal Cluster Analysis (HCA). PCA provided a satisfying classification based on the types of cosmetic treatments, brands as well as weekly intervals of hair, and permitted up to more than 90% amount of variance, indicating the reliability and validity of the model. Results from HCA complemented the deduction. This present study sheds light in proposing the use of ATR-FTIR combined with chemometric analysis for a simple and accurate classification technique of cosmetically treated human scalp hair which can be incorporated into a forensic hair screening protocol.