Development and validation of chemometric-assisted spectrophotometric models for efficient quantitation of a binary mixture of supportive treatments in COVID-19 in the presence of its toxic impurities: a comparative study for eco-friendly assessment.

The use of sustainable solvents has increased significantly in recent years due to advancements in green analytical methods. The number of impurities in the drug substance determines how safe the finished product is. Therefore, during the whole medication planning process, contaminants need to be closely watched. Using chemometric models, the concentrations of hyoscine N-butyl bromide (HYO) and paracetamol (PAR) were determined in the presence of three PAR impurities [P-nitrophenol (PNP), P-aminophenol (PAP), and P-chloroacetanilide (PCA), as well as DL-tropic acid (TRO) as a HYO impurity]. It was possible to isolate and measure these dangerous impurities. Fever and spasms associated with COVID-19 are reported to be considerably reduced when PAR and HYO are taken together. Artificial neural networks, principal component regression, multivariate curve resolution-alternating least squares, and partial least squares are the four chemometric-assisted spectrophotometric models that were created and verified. All of the proposed methods' quantitative analytical potency was assessed using recoveries%, root mean square error of prediction, and standard error of prediction. For PAR, HYO, PNP, PCA, TRO, and PAP, respectively, the indicated approaches were used in the ranges of 4.00-8.00, 16.00-24.00, 1.00-5.00, 0.40-0.80, 4.00-12.00, and 2.00-6.00 µg/mL. They are able to get around difficulties like collinearity and spectral overlaps. After statistical testing, there was no discernible difference between the recommended methods and the published one. The degree of greenness of the established models was evaluated using three different green assessment methods. In the presence of their harmful impurities, PAR and HYO could be identified using the recommended methods.