Artificial Neural Network Processing of Stripping Analysis Responses for Identifying and Quantifying Heavy Metals in the Presence of Intermetallic Compound Formation

Abstract

Feed-forward neural networks have been trained to identify and quantify heavy metals in mixtures under conditions where there were significant complications due to intermetallic compound formation. The networks were shown to be capable of (i) correlating voltammetric responses with individual heavy metals in complex mixtures, (ii) determining the relationship between responses and concentrations (including nonlinear relationships due to overlapping peaks and intermetallic compound formation), and (iii) rapidly determining concentrations of individual components from mixtures once trained. Using simulated data, modeled after complex interactions experimentally observed in samples containing Cu and Zn, it has been demonstrated that networks containing two layers of neurons (a nonlinear hidden layer and a linear output layer) can be trained to calculate concentrations under a variety of complicated situations. These include, but are not limited to, cases where the response of the intermetallic compound formed is observed as a shoulder of one of the pure metals and cases where the response of the intermetallic compound formed is not observed in the potential window. In addition, the network described above was trained to simultaneously determine concentrations of four metals (Cu, Pb, Cd, and Zn) in a concentration range where all responses were complicated by intermetallic compound formation (1-500 ppb).

Nickel tungstate–graphene nanocomposite for simultaneous electrochemical detection of heavy metal ions with application to complex aqueous media

“…These results are encouraging as there is a good separation between Cu(II) and Hg(II) ions peaks ruling out the formation of intermetallic compounds from these metals which are the case with many electrode systems. 15,[30][31][32] This result is attributed to the highly porous structure of RGO/NiWO 4 nanocomposite which lowered oxidation overpotential of each metal ions. 29,31,33 Further, simultaneous analysis of three ions namely Cd(II), Cu(II) and Hg(II) ions was performed and results indicated existence of individual peaks for each ion in DPASV response as demonstrated in Fig.…”

Resolution of Heavy Metal Mixtures from Highly Overlapped ASV Voltammograms Employing a Wavelet Neural Network

“…Thallium and Indium cannot be precisely allocated, although its redox activity is confirmed in the same range of potentials studied (among À 0.65 V for Tl þ and À 0.75 V for In 3þ ). Despite the high sensitivity of ASV in the conditions studied, the selectivity shown is poor as the overlapped signal corresponds to the contribution of each metal under study as well as to certain intermetallic species formed [34]. In this way, the obtained signals present noticeable overlapping and interference with the components in the sample, which makes it difficult to directly extract the concentration of each component.…”

The Study on Novel Microelectrode Array Chips for the Detection of Heavy Metals in Water Pollution

“…Besides, the concentration of zinc is several times that of copper in the natural water, so the problem rising from the copperÀzinc intermetallic compounds can be neglected in the actual detection. Otherwise, this effect can be reduced by spiking the analyte solution with gallium ions, 21 and regression modeling methods such as partial least square regression or artificial neural network method 22 can also be used for calibration. Therefore, decrease in the stripping peak currents may be mostly caused by the inaccurate addition and the shift of the baseline.…”