Chemometrics resolution and quantification power evaluation: Application on pharmaceutical quaternary mixture of Paracetamol, Guaifenesin, Phenylephrine and p-aminophenol

• Advanced chemometric models used for the analysis of challenging quaternary mixture.
• The analytical performance of CRACLS, MCR-ALS and PCA-ANN was compared and benchmarked to that of classical PLS-Calibration.
• Individual calibration provides better result with lower RMSEC than the general calibration model.
• Using PCA-ANN instead of ANN allows higher prediction concentrations with lower RMSEC and RMSEP.
• The proposed methods effectively detected the cited drugs in their pharmaceutical dosage form.

Three advanced chemmometric-assisted spectrophotometric methods namely; Concentration Residuals Augmented Classical Least Squares (CRACLS), Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) and Principal Component Analysis-Artificial Neural Networks (PCA-ANN) were developed, validated and benchmarked to PLS calibration; to resolve the severely overlapped spectra and simultaneously determine; Paracetamol (PAR), Guaifenesin (GUA) and Phenylephrine (PHE) in their ternary mixture and in presence of p-aminophenol (AP) the main degradation product and synthesis impurity of Paracetamol. The analytical performance of the proposed methods was described by percentage recoveries, root mean square error of calibration and standard error of prediction. The four multivariate calibration methods could be directly used without any preliminary separation step and successfully applied for pharmaceutical formulation analysis, showing no excipients’ interference.

Figure optionsDownload as PowerPoint slide