Abstract

Digital image–based colorimetry (DIC) coupled with artificial neural network (ANN) was studied for the determination of phosphorus in natural rubber (NR) latex. The method is based on the RGB (Red, Green, and Blue) value of an image of phosphorus standard solution reacted with molybdenum blue solution. The picture of the complex solution was captured by using a complementary metal oxide semiconductor (CMOS) camera. Because of a colloidal property of NR latex, phosphorus residue could not be directly determined by this technique. Therefore, a UV digestion unit was designed and fabricated for the digestion of NR latex before color developing and processing with DIC-ANN. The digestion time needed for complete digestion was 50 min. The digestion capability of the UV digester to process 10 samples simultaneously led to an acceptable samples analysis frequency of 12 h^-1. Quantitative measurement of phosphorus was made in the range 0.1 to 1.0 mg L^-1. The proposed method was successfully applied to the determination of total phosphorus in NR latex and it was found to be simple, rapid, accurate, precise and low cost method.

Key words: Digital image-based colorimetry, artificial neural network, UV digestion, natural rubber latex, phosphorus.