Linezolid (LIN) is determined in the presence of its alkaline (ALK) and oxidative (OXD) degradation products without preliminary separation based on ultraviolet spectrophotometry using two-way chemometric methods; principal component regression (PCR) and partial least-squares (PLS), and three-way chemometric methods; parallel factor analysis (PARAFAC) and multi-way partial least squares (N-PLS). A training set of mixtures containing LIN, ALK and OXD; was prepared in the concentration ranges of 12-18, 2.4-3.6 and 1.2-1.8 jig mL(-1), respectively according to a multilevel multifactor experimental design. The multivariate calibrations were obtained by measuring the zero-order absorbance from 220 to 320 nm using the training set. The validation of the multivariate methods was realized by analyzing their synthetic mixtures. The capabilities of the chemometric analysis methods for the analysis of real samples were evaluated by determination of LIN in its pharmaceutical preparation with satisfactory results. The accuracy of the methods, evaluated through the root mean square error of prediction (RMSEP), was 0.058, 0.026, 0.101 and 0.026 for LIN using PCR, PLS, PARAFAC and N-PLS, respectively. Protolytic equilibria of LIN and its degradation products were evaluated using the corresponding absorption spectra-pH data obtained with PARAFAC. The obtained pK(a) values of LIN, ALK and OXD are 5.70, 8.90 and 6.15, respectively. The results obtained were statistically compared to that of a reported HPLC method, and there was no significant difference between the proposed methods and the reported method regarding both accuracy and precision. (C) 2014 Elsevier B.V. All rights reserved.
