ATR-FTIR spectroscopy is a fast and accessible, minimally or non-destructive technique which provides information on physiochemical characteristics of analyzed materials. In forensic and archaeological sciences, it is commonly used for answering numerous questions, including the archaeological or forensic context of the human skeletal remains. In this research, the accuracy of ATR-FTIR-obtained spectra for separation between forensic, WWII, and archaeological human skeletal remains was investigated. Building from the previously proposed methodological procedures, various ratio-based and whole spectra separation procedures were applied, carefully analyzed, and evaluated. Results showed that employing whole spectral domains works best for the separation of archaeological, WWII, and forensic samples, even with samples of highly variable origin. Principal component analysis (PCA) further highlighted the necessity of acknowledging all the major components in the remains: amides, phosphates, and carbonates for the separation. Most influential proved to be amide I, namely its secondary structure, which presented well-preserved and organized collagen structure in forensic and WWII samples, while highly degraded in archaeological samples. Using the whole spectral domain for separation between samples from different contexts proved to be fast and simple, with no manipulation beyond baseline correction and normalization of spectra necessary. However, a dataset with samples of known origin is required for the learning model and predictions. A less accurate alternative is separation based on combining ratios of peaks correlating to organics and minerals in the bone, which eliminated overlapping and managed to classify the majority of the samples correctly as archaeological, WWII, or forensic.
Keywords: ATR-FTIR spectroscopy; Forensic; Human skeletal remains; Physiochemical composition of bone; WWII archaeological remains.