Performance of new nanocomposites based on graphene-grafted-poly(itaconic acid-co-TRIM) via photoiniferter, thermal vinyl functionalization, and physical mixture as electrochemical sensing platforms for illicit drug determination

Abstract

Graphene-functional polymer nanocomposites-based sensors have been used as platforms for the voltammetric sensing of a wide range of molecules. However, different strategies can be used to stablish the integration/interface between the graphene and the functional polymer. Therefore, in this work, the performance of new nanocomposites based on graphene (GO and rGO) grafted with poly(itaconic acid-co-TRIM) prepared by controlled-radical polymerization via photoiniferter [named GO-poly(IA-co-TRIM)-iniferter and rGO-poly(IA-co-TRIM)-iniferter] and free-radical polymerization via thermal vinyl functionalization as platforms [named GO-VTMS-poly(IA-co-TRIM) and rGO-VTMS-poly(IA-co-TRIM)] for voltametric sensing of mephedrone (4-MMC) was evaluated. In addition, the physical mixtures of GO/poly(IA-co-TRIM) and rGO/poly(IA-co-TRIM) were also investigated for sensor preparation. Characterization of materials was performed by FT-IR, X-ray diffraction, Raman spectroscopy, SEM, TEM, TGA, and nitrogen adsorption/desorption measurements. For sensor preparation, nanocomposite suspensions were drop-casted on the surface of a glassy carbon electrode (GCE). No cathodic peak current for mephedrone was observed when using unmodified GCE, while a pronounced analytical signal was achieved when modifying the GCE with the physical mixture of GO and poly(IA-co-TRIM). The synergic effect of GO and the itaconic acid in the poly(IA-co-TRIM) in the physical mixture explains the higher adsorption of mephedrone on the surface of the sensor when compared to the other methods evaluated. A linear analytical curve (0.25 to 10.0 µmol L−1, R2 = 0.99) was obtained with limit of detection of 0.22 μmol L−1. 4-MMC determination was not influenced by the presence of caffeine, benzocaine, paracetamol, cocaine, and ethylone. The method was applied in simulated seized street samples, with accuracy assessed by HPLC-DAD as well as in synthetic urine samples, yielding recovery values from 95.8 to 103.9%.