Abstract
Synthesis, distribution and abuse of fentanyl, a synthetic opioid, has led to a critical worldwide epidemic. Mass spectral library searching for opioids remains unresolved despite being central to law-enforcement involving identification, monitoring and prosecution of opioid related crimes. In this article, two model problems are presented to illustrate difficulties associated with fentanyl identification. A collection of both currently-employed similarity measures and intuitive measures of dissimilarity are employed to simulate identifying fentanyl analogs with mass spectral library searching.