Forensic classification of lipsticks using attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy

摘要

Cosmetic products have been used extensively in human history for the purpose of beautification by women. However, due to globalisation, cosmetics have been increasingly used by people of both sexes. With increasing consumer awareness, it has become even more important to maintain the quality of cosmetics, which is often compromised in counterfeit and adulterated lipsticks. Moreover, cosmetic products are frequently encountered as trace evidence in forensic investigations. Cosmetics evidence are often encountered in cases of child abuse, child trafficking, kidnapping, murder, rape, sexual assault and also homicidal cases. In such cases, it becomes important to classify and predict samples based on their brands. All these cases, warrant the discrimination of lipstick samples of one brand from another. In the present work, potential utility of combined Attenuated Total Reflectance - Fourier Transform Infrared (ATR-FTIR) spectroscopy and chemometric techniques (PCA and LDA) have been evaluated to classify brown, pink, and red lipstick samples based on their brands. PCA accurately classified 93.61 %, 75.43 % and 87.27 % of brown, pink and red lipstick samples respectively. In contrast, LDA classified and predicted 100 % samples according to their brands. External validation was performed to predict the brands of brown and pink lipsticks with 100 % accuracy, whereas, brands of red lipstick samples were predicted with 96 % accuracy. These results are very significant and indicate that the developed model can be successfully used for the classification and prediction of lipstick samples. It is clear that ATR-FTIR is a non-destructive, rapid, cost effective, simple to use, efficient, and accurate technique which can be used effectively in combination with chemometric tools for the classification of lipstick samples.