Aplicações de imagem digital usando smartphone em reações colorimétricas

Abstract

This dissertation proposes different applications of a digital image analysis on colorimetric reactions. In this work, the image was captured by using a smartphone model J3, aiming to determinate total iron content in water, and methanol in sugarcane spirit. The processing consisted of image treatment on RGB colors system by R software with RStudio graphical interface using EBImage package. This package does the digital image decomposition, where the pixel intensity is related to the concentration of the analyte. For the determination of total iron in water, the green channel showed a better grade of adjustment with a determination coefficient (R2 = 0.9985), with the concentration of analyte spiked with 5.0 µg. mL-1. For validation procedure, the results were well correlated with data obtained by UV spectrophotometry (78.91%). On the methanol determination, the blue channel provided a considerably strong correlation (89.09%) with data obtained by UV spectrophotometry. The obtained concentration was from 0.09 0.15% v/v, the limit of detection was from 0.002 0.011% v/v and the limit of quantification was from 0.006 0.034% v/v. The proposed methodology presented results with models statistically significative, with a better quality the adjustment about observed values and therefore it can be used as a fast and easy alternative methodology that needs relatively low cost