Performance evaluation of handheld Raman spectroscopy for cocaine detection in forensic case samples

Area of Science:

• Forensic Science
• Analytical Chemistry
• Spectroscopy

Background:

• Handheld Raman spectroscopy is an emerging technology for on-site drug detection.
• Current devices often lack extensive validation and user insight into raw spectral data and identification criteria.

Purpose of the Study:

• To evaluate the performance of a commercial handheld Raman spectrometer for cocaine detection.
• To assess its reliability for both presumptive identification and legally admissible evidence.

Main Methods:

• Performance evaluation using binary cocaine mixtures (0-100 wt%).
• Retrospective analysis of 3,168 case samples compared with gas chromatography-mass spectrometry (GC-MS).
• Assessment of spectral selectivity and comparison of instrument results with chemometric models (PLS-R and PLS-DA).

Main Results:

• Limit of detection varied (10-40 wt%), but street samples averaged above this threshold.
• Achieved a 97.5% true positive rate for cocaine in case samples with no false positives.
• 12.5% of negative samples were initially inconclusive; spectral analysis confirmed high selectivity for key cocaine peaks.

Conclusions:

• The commercial handheld Raman spectrometer demonstrates high performance for cocaine detection in real-world forensic samples.
• Chemometric models further enhance the instrument's accuracy and reliability.
• The device is suitable for presumptive detection and providing legally valid evidence of cocaine presence.