Rapid screening of critical process parameters based on near infrared spectroscopy: a case study of the ethanol precipitation process

Abstract

In process development, the design of experiments (DoE) approach has been widely used for screening the key influential factors and obtaining optimal operating conditions. However, measuring the quality attributes of many products produced in the DoE is time and cost intensive. Therefore, an innovative method using near infrared spectroscopy (NIRS) for rapid screening of CPPs is proposed in this work, taking the ethanol precipitation process of Danhong Injection as a case study. In this work, seven process parameters were studied with a fractional factorial DoE. The supernatants produced were analyzed by NIRS. The spectra were preprocessed and then decomposed by principal component analysis into a few scores and loadings. The scores were taken as the response variables in analysis of variance (ANOVA) to identify the CPPs. The results showed that water content in the concentrate (p < 0.0001), ethanol concentration (p < 0.0001) and ethanol consumption (p < 0.0001) were the three CPPs. To validate the proposed method, the intermediates were analyzed by HPLC. The concentrations of two representative compounds were taken as the response variables in ANOVA and the same three CPPs were identified. The proposed method was successfully applied and the advantages of integrating spectroscopic techniques into screening DoE were demonstrated.