Novel Spectrophotometric Method for Robust Detection of Trace Copper and Cobalt in High-Concentration Zinc Solution.

In the purification process of zinc hydrometallurgy, the spectra of copper and cobalt seriously overlap in the whole band and are interfered with by the spectra of zinc and nickel, which seriously affects the detection results of copper and cobalt in zinc solutions. Aiming to address the problems of low resolution, serious overlap, and narrow characteristic wavelengths, a novel spectrophotometric method for the robust detection of trace copper and cobalt is proposed. First, the Haar, Db4, Coif3, and Sym3 wavelets are used to carry out the second-order continuous wavelet transform on the spectral signals of copper and cobalt, which improves the resolution of copper and cobalt and eliminates the background interference caused by matrix zinc signals and reagents. Then, the information ratio and separation degree are defined as optimization indexes, a multi-objective optimization model is established with the wavelet decomposition scale as a variable, and the non-inferior solution of multi-objective optimization is solved by the state transition algorithm. Finally, the optimal second-derivative spectra combined with the fine zero-crossing technique are used to establish calibration curves at zero-crossing points for the simultaneous detection of copper and cobalt. The experimental results show that the detection performance of the proposed method is far superior to the partial least squares and Kalman filtering methods. The RMSEPs of copper and cobalt are 0.098 and 0.063, the correlation coefficients are 0.9953 and 0.9971, and the average relative errors of copper and cobalt are 3.77% and 2.85%, making this method suitable for the simultaneous detection of trace copper and cobalt in high-concentration zinc solutions.