Unique Terahertz Features of Fentanyl and its Analogs Revealed via High Performance Computations

Fentanyl is one of the most potent synthetic opioid pain relievers. However, the abuse of opioids may lead to death. Rapid identification methods of both known and emerging opioid substances in fentanyl family would help to prevent this abuse. Terahertz (THz) spectra provide the identification information without damaging the sample. In this study, we predict THz spectra of fentanyl and 20 of its analogs in solid state by density-functional theory method PBEh-3c to investigate the frequencies of rotational-vibrational normal modes in THz region (0.1~100 cm
^-1
). Substitution of single bond or aromatic group of fentanyl leads to significant change of crystal packing and is reflected in its THz spectrum. We discuss structure-property relationships of these synthetic opioids such as ohmefentanyl and sufentanil that are more potent than fentanyl. Since new drugs come up in black market frequently, we applied machine learning methods to predict THz spectrum of yet unknown synthetic opioids. Other “white powders” such as sucrose and lactose will be tested to validate our prediction.