Computer-vision based second-order (kinetic-color) data generation: arsenic quantitation in natural waters

Abstract

Computer vision-based analytical methods have gained popularity in the literature, digital images and/or movies have been used to build univariate (traditional analytical line) and multivariate models. This paper describes, for the first time to the best of our knowledge, second-order data processing obtained with a computer vision-based analytical device. Therefore, a flow batch assembly coupled whit a drop system to determining arsenic in water samples without chemical/external pretreatment was employed. Arsenic is extracted from the water samples as arsine to react with a drop of silver diethyldithiocarbamate producing a colored complex. The entire reaction is recorded with a digital microscope to obtain videos as a function of time, generating second order data that is subsequently treated with Multivariate Curve Resolution Alternating Least Squares (MCR-ALS). The proposed low-cost method exhibits good performance, satisfactory detection limit (0.07 µg L−1) and linear response from 0.05 to 1.00 µg L−1 of As in water samples.