Award

Synthetic opioids have been responsible for most drug overdoses in the US in the last several years. Specifically, the use of fentanyl and fentanyl analogues have led to an alarming 7.5X increase in the number of overdoses from 2015 to 2021. Developing a strategy for countering the proliferation of synthetic opioids and overdoses in the US is challenging for several reasons. First, the structure of synthetic opioids can be readily modified, making it easy to develop many new psychoactive analogues. Second, each new synthetic opioid has a different chemical structure, potency, and unique fingerprint, which makes it difficult to keep chemical detection equipment up to date. Finally, synthetic opioids are often present in low concentrations in mixtures, which further complicates the development of detection algorithms.In this proposal we describe the development of a synthetic opioid detection algorithm for infrared and Raman spectrometers that: 1) detects existing drugs, and non-targeted drugs that are likely to appear in the future, 2) detects low concentrations present in street drug mixtures, as well as high concentrations that are common in bulk transport, and 3) can be generalized to work on more than one vendor’s detection equipment. To demonstrate feasibility, we will: 1) use machine learning algorithms to create chemical structures for novel synthetic opioids, 2) use experimental infrared and Raman spectra and predicted theoretical spectra to develop synthetic opioid detection algorithms, and 3) establish limit of detection, probability of detection, and false alarm rate for several detection algorithms using real-world drug samples.