Predictive evaluation of multicomponent direct compress model tablets by integrating sphere UV-Vis spectroscopy and chemometrics.

Abstract

Control of the pharmaceutical manufacturing process and active pharmaceutical ingredients (API) is essential to product formulation and bioavailability. The aim of this study is to predict tablet surface API concentration by chemometrics using integrating sphere UV-Vis spectroscopy, a non-destructive and contact-free measurement method. Riboflavin, pyridoxine hydrochloride, dicalcium phosphate anhydrate, and magnesium stearate were mixed and ground with a mortar and pestle, and 100mg samples were subjected to direct compression at a compaction pressure of 6MPa at 7mm diameter. The flat surface tablets were then analyzed by integrating sphere UV-Vis spectrometry. Standard normal variate (SNV) normalization and principal component analysis were applied to evaluate the measured spectral dataset. The spectral ranges were prepared at 300-800nm and 500-700nm with SNV normalization. Partial least squares (PLS) regression models were constructed to predict the API concentrations based on two previous datasets. The regression vector of constructed PLS regression models for each API was evaluated. API concentration prediction depends on riboflavin absorbance at 550nm and the excipient dicalcium phosphate anhydrate. Integrating sphere UV-Vis spectrometry is a useful tool to process analytical technology.