Linear Regression Functions-Multivariate Calibration Methods for the Determination of Two-Component Mixture

Abstract

Two numerical methods based on the linear regression functions were proposed for the simultaneous determinations of hydrochlorothiazide (HCT) and moexipril hydrochloride (MEP) in commercial samples. Standard series of hydrochlorothiazide and moexipril hydrochloride in the range of 4.0-14.0,mu g/mL and 6.0-16.0 mu g/mL were prepared in methanol. The UV spectra of the related standard series were recorded between 210.0-300.0 nm. The same recording procedure was applied to the samples of hydrochlorothiazide and moexipril hydrochloride. For the application of the bivariate calibration, 60 linear regression functions for each compound were obtained with an interval of Delta lambda=1.0 nm in the wavelength range of 220-280 nm. Kaiser technique was applied to the sensitivity matrices corresponding to the slopes of the linear regression functions for each wavelength pair We found that 220.0 nm and 269.0 nm were reported to be the optimal ones for the bivariate calibration method. The second method, the linear regression-multivariate calibration, is based on the use of 60 linear regression functions, without finding an optimal wavelength-pair set [4]. Both methods were validated by analyzing the synthetic mixtures of two above mentioned drugs. Recovery results were found as 103.0 and 101.0% for hydrochlorothiazide and 104.0 % and 102.0% for moexipril hydrochloride using the bivariate calibration and the linear regression function-multivariate calibration. These proposed calibration methods were sucessfuly applied to the quantification of the related compounds in the commercial tablets.