In this study, 162 samples from different sexual lubricant manufacturing types were analyzed and classified based on the components identified in the sample.
The increased use of condoms may be one reason why there is a reduced potential of finding DNA evidence in sexual assault cases. This dilemma has led to research in other trace evidence, such as sexual lubricants. In the absence of DNA, the analysis of trace lubricant residue may aid in connecting the suspect to the victim and/or crime scene. The current study developed FTIR, GC–MS, and DART-HRMS methods and protocols that can be used to analyze samples and identify lubricant components for forensic purposes. Neat and solvent extracts of samples were analyzed in triplicate via the DART-HRMS and FTIR; however, only the extracts were analyzed by GC–MS in triplicate. Multivariate statistical techniques included hierarchical cluster analysis, principal component analysis, linear discriminant analysis, Fisher's ratio, and least absolute shrinkage and selection operator (LASSO) regression were used to characterize the lubricants to develop a comprehensive classification scheme for the lubricants. The classification scheme was tested using an internal dataset and an external independent test set to evaluate the accuracy of the developed model(s) for unknown samples. The classification scheme developed from this study and provided along with the database will allow analysts to classify unknown lubricants and lubricant residues found at crime scenes using analytical results from the aforementioned instruments. This research will aid in classifying unknown sexual lubricants based on components indicative of a particular class. (Publisher Abstract)