Determination of Coconut Fatty Acid Diethanol Amid in Liquid Detergent by Spectrophotometry and Multivariate Calibration

Abstract

This paper describes a procedure for the determination of coconut fatty acid diethanol amid in liquid detergent by UV-vis spectrophotometry. A partial least squares (PLS) regression was performed to resolve highly overlapping spectrophotometric signals obtained from coconut and other components in solution. Inputs of model were taken from absorbance spectral data which recorded in wavelength region 200-600 nm. In other steps, first and second order derivative spectra were considered as inputs of model. The proposed method has found to be precise and accuracy. The experimental calibration was designed with 16 standard samples. The concentrations for coconut were varied between 0.0025-0.0040 g/mL. The statistical parameters such as square correlation coefficient (R²), root mean squares error crossvalidation (RMSECV) has been evaluated and number of factors and spectral have been optimized. The optimized model, obtained using PLS-1 was applied to the determination of compound in liquid detergent. The results for coconut in zero, first and second order derivative spectra were found to be satisfied. The error obtained in this method was ± 2.50%.