Resolution of opiate illicit drugs signals in the presence of some cutting agents with use of a voltammetric sensor array and machine learning strategies

In the present work, resolution and quantification of mixtures different opiate compounds in presence common cutting agents using an electronic tongue (ET) is evaluated. More specifically, ternary heroin, morphine codeine were resolved caffeine paracetamol. To this aim, array three carbon screen-printed electrodes modified with ink-like solutions graphite, cobalt (II) phthalocyanine palladium, their responses towards drugs characterized by means square wave voltammetry (SWV). Developed sensors showed a good performance linearity at µM level, LODs between 1.8 5.3 for 3 actual drugs, relative standard deviation (RSD) ca. 2% over 50 consecutive measurements. Next, quantitative model that allowed identification individual substances from overlapped voltammograms was built partial least squares regression (PLS) as modeling tool. With approach, achieved ?M total normalized root mean error (NRMSE) 0.084 test subset.