An efficient near infrared spectroscopy based on aquaphotomics technique for rapid determining the level of Cadmium in aqueous solution

International Technical Postgraduate Conference

Putra, Alfian^1 ; Vassileva, Maria^2 ; Santo, Ryoko^2 ; Tsenkova, Roumina^2

Department of Chemical Engineering, Lhokseumawe Polytecnic State, Indonesia^1

Departement of Enviroment and Bioproduction Engineering, Graduate School of Agriculture Science, Kobe University, 1-1, Rokkodai, Kobe, Nada

657-8501, Japan^2

关键词: Correlation coefficient;    Drinking water samples;    Industrial pollutants;    Metal interactions;    Metal-organic complexes;    Partial least square regression analysis;    Pollution monitoring;    Regression coefficient;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/210/1/012014/pdf
DOI  :  10.1088/1757-899X/210/1/012014

来源: IOP

【 摘 要 】

Cadmium (Cd) is a common industrial pollutant with long biological half-life, which makes it as a cumulative toxicant. Near-infrared spectroscopy has been successfully used for quick and accurate assessment of Cd content in agricultural materials, but the development of a quick detection method for ground and drinking water samples is equal importance for pollution monitoring. Metals have no absorbance in the NIR spectral range, thus the methods developed so far have focused on detection of metal-organic complexes (move to intro). This study focuses on the use of Aquaphotomics technique to measure Cd in aqueous solutions by analyzing the changes in water spectra that occur due to water-metal interaction. Measurements were performed with Cd (II) in 0.1 M HNO3, in the 680-1090 nm (water second and third overtones) and 1110-1800 nm (water first overtone) spectral regions, and were subjected to partial least-square regression analysis. It was found/determined that A concentration of Cd from 1 mg L-1to 10 mg L-1could be predicted by this model with average prediction correlation coefficient of 0.897. The model was tested by perturbations with temperature and other metal presence in the solution. The regression coefficient showed consistent peaks at 728, 752, 770, 780, 1362, 1430,1444, 1472/1474 and 1484 nm under various perturbations, indicating that metal to influence the water spectra. The residual predictive deviation values (RPD) were greater than 2, indicating that the model is appropriate for practical use. The result suggested that this newly proposed approach is capable of detecting metal ion in a much simpler, rapid and reliable way.