Semester

Summer

Date of Graduation

2022

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Forensic and Investigative Science

Committee Chair

Luis E. Arroyo

Committee Co-Chair

Tatiana Trejos

Committee Member

Tatiana Trejos

Committee Member

Keith B. Morris

Committee Member

M. Julia Arcos-Martinez

Committee Member

John N. Richardson

Abstract

Forensic science relies on the use of multiple techniques in the assessment of evidence to increase the accuracy and reliability of the results. However, with the rapidly changing drug landscape due to the introduction of novel psychoactive substances, many traditional screening methods are no longer sensitive or selective enough for use. Additionally, many screening methods such as chemical color tests are prone to false positive and negative results and are subjective. Therefore, the goal of this dissertation was to develop a novel analytical scheme that can provide a more efficient, rapid, and sensitive method that will facilitate adoption in the laboratory and onsite in the field. To this end, electrochemistry and Raman spectroscopy were assessed independently for their use in orthogonal testing scenarios, and in tandem via spectroelectrochemical methods to improve both the sensitivity and selectivity of seized drug screening. A panel of 15 drug analytes and 15 diluent compounds comprising some of the most encountered analytes was considered. This panel was characterized via electrochemistry for use as the first tier in investigating a seized substance. Then normal Raman spectroscopy was explored as the second tier in an orthogonal analytical scheme. Both 785 nm and 1064 nm Raman systems were tested with the panel of analytes. The pure compounds were characterized prior to analysis of simulated mixture samples with binary mixtures prepared at 1:4, 1:7, 1:10, and 1:20 ratios. Finally, electrochemistry and Raman spectroscopy were combined into a third-tier technique: time-resolved spectroelectrochemistry. Both targeted and nontargeted electrochemical-surface enhanced Raman spectroscopy method (EC-SERS) was developed. The targeted EC-SERS method provided an analysis approach for fentanyl and fentanyl analogs, while also allowing for in situ generation of a SERS substrate with a silver electrode and simultaneous analysis of the sample. This method demonstrated excellent selectivity and sensitivity with a limit of detection for the most sensitive compound (4-ANPP) of 10 ng/mL. Additionally, differentiation of fentanyl analogs was demonstrated to be achievable. Various potentially interfering compounds were assessed including heroin, cocaine, methamphetamine, caffeine, quinine, and others. Quinine was the only compound observed to have significant interference with fentanyl detection. To demonstrate the fit-for-purpose of this method, authentic seized samples were analyzed from a forensic laboratory including both true fentanyl positive and true fentanyl negative samples. Using GC-MS and LC-MS/MS as ground-truth, an overall accuracy of ~88% was achieved for the detection of fentanyl within the seized samples. More importantly, the power of this EC-SERS screening approach was demonstrated by the accurate identification of fentanyl even in mixtures containing 5 or more compounds.

Beyond the problems associated with drugs of abuse and opioid epidemic, another critical area of increasing interest deals with firearms-related crimes. Gun violence in our country results in thousands of deaths each year. However, the discipline of gunshot residue analysis is one of the few areas in forensics that does not have a reliable screening approach. Although the gold standard method utilizes scanning electron microscopy-energy dispersive X-ray spectrometry (SEM-EDS), providing morphological and elemental analysis, the method is time-consuming, expensive, and cannot analyze organic gunshot residues. Therefore, this dissertation also looked at the usefulness of electrochemistry to provide a simple, inexpensive, and portable screening method for the simultaneous analysis of inorganic and organic gunshot residues. Due to the rapid nature of electrochemistry, an extremely large population of over 1,000 authentic sample sets was collected and analyzed. These samples comprised several different subpopulations: leaded shooter, lead-free shooter, mixed leaded and lead-free shooter, low-risk background non-shooter, high-risk background non-shooter, and leaded shooter with post-shooting activities. The low-risk background allowed for the assessment of the presence of gunshot residue analytes in the background population not associated with the discharge of a firearm and provided a statistical analysis of background levels. These critical thresholds were used to define a positive and negative sample and resulted in accuracy values ranging between 75% and 82%. Additionally, due to the large number of samples and large data sets, the usefulness of machine learning approaches, specifically neural networks were investigated for their ability to make use of the data trends and provide more accurate classification of samples. The neural network utilized in this approach provided an overall accuracy of 98%. The use of electrochemistry for the screening of gunshot residues from the hands of individuals provides a rapid, inexpensive, simple, and portable screening approach that can improve investigative leads, improve case triage, and lower backlogs and costs associated with gunshot residues. Additionally, the speed of analysis provides the opportunity for investigators to collect more samples and from other surfaces that may not have been done in the past due to the disadvantages of the current methodologies.

This dissertation demonstrates the advantages of employing simple, fast, and portable technologies in the screening of common forensic evidence. Electrochemistry, spectroscopy, and spectroelectrochemistry are powerful techniques that can revolutionize analytical schemes used in forensic science, and this research provides the foundations and a path forward.

Download

Share

COinS