Semester

Fall

Date of Graduation

2019

Document Type

Thesis

Degree Type

MS

College

Eberly College of Arts and Sciences

Department

Forensic and Investigative Science

Committee Chair

Luis Arroyo

Committee Co-Chair

Tatiana Trejos

Committee Member

Nianqiang Wu

Committee Member

John Richardson

Abstract

Utilizing screen-printed carbon electrodes (SPCEs), a fast, simple, and sensitive approach toward the detection, identification, and quasi-quantitation of fentanyl was achieved both in an electrochemical cell and as a drop on the electrode surface. Electro-oxidation of fentanyl at the electrode was demonstrated using adsorptive stripping square-wave voltammetry between -0.5 V and +1.6 V with 100 mM Tris-HCl buffer at pH 8.5 as supporting electrolyte. Parameter optimization was conducted during method development to include supporting electrolyte and pH, electrochemical technique, pre-treatment and equilibration time, and various surface modifications. The simplest method utilizing an unmodified SPCE was determined to be appropriate for the identification of fentanyl. Electro-oxidation of the fentanyl compound was observed to occur as an irreversible process due to both diffusion to the electrode surface and oxidation of adsorbed species on the working electrode.

The resulting voltammograms demonstrated the presence of two oxidation peaks at 750 mV (peak I) and 880 mV (peak II) versus a pseudo-Ag/AgCl reference. Fentanyl oxidation was observed at concentrations of ~76 ng/mL in cell and ~300 ng/mL in a 100 mL drop. Statistical limits of detection were determined to be slightly above the observable oxidation peaks with limits of detection of 145 ng/mL for the cell method and 530 ng/mL for the drop method. Reproducibility between electrodes, assessed as the average relative standard deviation (RSD), for peak I and peak II in the cell was 12% and 18%, respectively. RSD in the drop was 13% and 15% for peaks I and II.

Accuracy of the detection method was determined in the cell by analyzing single-blind samples prepared in the laboratory and demonstrated better accuracy in lower concentrations of fentanyl versus higher concentrations. The effects of interfering compounds were considered due to the likelihood of fentanyl being found in mixtures. Quinine and cocaine were found to interfere with peak II, while peak I remained identifiable except when present with large concentrations of interferent. Methamphetamine was observed to have a similar effect although drastically reduced in comparison to both quinine and cocaine. Acetaminophen and caffeine did not produce interfering signals. Analysis at various ratios of the compounds demonstrated that the identification of fentanyl could still be achieved through the presence of peak I. The oxidative mechanism of fentanyl was proposed based on the literature available for the oxidation of amines and voltammetric data present for fentanyl and related compounds. The proposed mechanism rejects some previously hypothesized oxidation mechanisms of tertiary amines where the presence of two peaks was observed. It was suggested that a two-step oxidation process of the tertiary amine followed by the oxidation of the newly formed secondary amine product resulted in the two observable peaks. However, this work agrees with literature supporting the effect of adsorption of the tertiary diamine to the electrode surface. This mechanism is presented herein, whereby the observed oxidation peaks result from the adsorbed species and the diffusion of the species to the electrode surface, owing to the difference in peak potentials for peak I and peak II.

A confirmatory LC/MS/MS method for the analysis of fentanyl and fentanyl analogs in oral fluid was developed and validated. Optimization of fragmentor voltage, collision energy, and fragmentation ions was achieved and used in the construction of a dynamic multiple reaction monitoring (dMRM) method. Chromatographic separation demonstrated resolution between 13 fentanyl-related compounds along with 7 internal standards. The calibration model used was linear with a weighting of 1/x between the range of 0.1 ng/mL to 50 ng/mL. The limit of detection for the majority of drugs was determined to be 0.01 ng/mL with the limit of quantitation at the lowest calibrator of 0.1 ng/mL with correlation coefficients between 0.9992-0.9999. Bias, precision, matrix effects, recovery, and process efficiency were assessed and were within the guideline range for acceptability for the majority of analytes assessed using a solid-phase extraction procedure with spiked oral fluid. Twelve commonly encountered illicit drugs were used to assess selectivity. No interferences were found for fentanyl or its analogs. Stability was assessed for processed samples kept at room temperature in auto-sampler and the freezer, as well as, for freeze/thaw stability. The majority of analytes were considered stable under all conditions for up to 72 hours.

Together these two methods demonstrate the identification and quasi-quantitation of fentanyl through electrochemical oxidation and confirmatory analysis via liquid chromatography-tandem mass spectrometry (LC/MS/MS). The combined use of these techniques seeks to emulate the SWGDRUG requirement, although electrochemistry has, to this point, not been included in the list of acceptable techniques.

Other work contained herein demonstrates assessment of various electrode modification techniques to improve the signal of fentanyl, attempts at enzymatic detection of codeine and fentanyl utilizing cytochrome P450 isozymes 2D6 and 3A4, and electrochemical detection of the synthetic cannabinoid PB-22.

Share

COinS