Electrochemical Determination of Fentanyl Using Carbon Nanofiber-Modified Electrodes

In this work, we report the direct electrochemical oxidation of fentanyl using commercial screen-printed carbon electrodes (SPCEs) modified with carboxyl-functionalized carbon nanofibers (fCNFs). CNFs have surface chemistry and reactivity similar to carbon nanotubes (CNTs), yet they are easier to produce and are of a lower cost than CNTs. By monitoring the current produced during the electrochemical oxidation of fentanyl, variables such as fCNF loading, fentanyl accumulation time, electrolyte pH, and differential pulse voltammetry parameters were optimized. Under an optimized set of conditions, the fCNF/SPCEs responded linearly to fentanyl in the concentration range of 0.125–10 μM, with a limit of detection of 75 nM. The fCNF/SPCEs also demonstrated excellent selectivity against common cutting agents found in illicit drugs (e.g., glucose, sucrose, caffeine, acetaminophen, and theophylline) and interferents found in biological samples (e.g., ascorbic acid, NaCl, urea, creatinine, and uric acid). The performance of the sensor was also successfully tested using fentanyl spiked into an artificial urine sample. The straightforward electrode assembly process, low cost, ease of use, and rapid response make the fCNF/SPCEs prime candidates for the detection of fentanyl in both physiological samples and street drugs.


Figure S1 .
Figure S1.Characterization of unmodified and fCNF-modified SPCEs.(a) Raman spectra of pristine and acid functionalized carbon nanofiber, (b) sample CVs for a bare SPCE and fCNF/SPCE electrodes prepared with 0.25, 0.5, 1, 2, and 3 mg mL -1 fCNFs in 1 mM [Fe(CN 6 )] 3-/4-, and their corresponding ΔE p .(c) The Bode plots from EIS measurements of the electrodes presented in (b), the Randles circuit fit, and corresponding best-fit R CT values.(d) The anodic and cathodic peak currents of the electrodes presented in (a) at scan rates ranging from 10 to 400 mV s -1 , and corresponding ECSAs.(e) The SEM images of (i) a bare SPCE electrode, (ii) a 1 mg/mL fCNF-modified SPCE, and (iii) the expanded view of a fCNF showing its hollow core.

Figure S2 .
Figure S2.Cyclic voltammograms of (a and b) norfentanyl and (c and d) phenylacetaldehyde with an SPCE electrode modified with a 1 mg mL -1 fCNFsuspension.(a) A 75 µM norfentanyl solution tested at 10-200 mV s -1 scan rates between -0.4-1.2V. (b) A close up of (a) showing the OX3 and R2 redox peaks.(c) A 75 µM phenylacetaldehyde solution tested at 10-200 mV s -1 scan rates between -0.4-1.2V. (d) A close up of (c) showing the absence of peaks.The electrolyte in all cases was 0.1 M PB pH 8.0 buffer.

Figure S5 .
Figure S5.DPV parameter optimization.(a) Scan rates from 10 to 150 mV/s in increments of 10 mV/s.(b) Close up of selected scan rates.100 mV/s was selected as optimal.(c)Step size and sample period optimization for scan rates of 100 mV/s.20 mV step size and 0.2 s sample period were selected as optimal.(d) Pulse size optimization while keeping the pulse time constant.100 mV pulse size was selected as optimal.(e) Pulse time optimization while keeping the pulse size constant.0.02 s was selected as optimal.(f) Testing the effect of pulse size on the pulse time.No clear advantage was observed by increasing pulse size, and thus optimized parameters were based on the data in (d) and (e).

Figure S6 .
Figure S6.The OX1 current responses via DPV of several 1 mg/mL fCNF-modified SPCE electrodes exposed to 10 µM and 20 µM fentanyl during a single-use weekly study.All electrodes were prepared during week 1 and stored under vacuum, while retrieving 1 modified electrode per week for these studies.The gray horizontal lines indicate the average currents, while the dotted lines show the standard deviation values.

Table S1 .
Composition of artificial urine.