FT-Raman Spectroscopy and PLS Regression for Quantitative Determination of Captopril in Tablet Powder; Effect of Different Preprocessing Methods

Abstract

Aims: This study was performed to develop a fast and simple method for the quantitative analysis of Captopril in tablet powder by FT-Raman Spectroscopy and PLS regression. The effect of different preprocessing methods on multivariate calibration was studied. Methodology: Four different preprocessing methods, namely mean center (MC), autoscale (AS), normalization (N1) and orthogonal signal correction (OSC) were applied to reduce the impact of sources of variability, such as instrument drift and sampling variability, correct for path length variation and increase the signal to noise ratio. PLS regression models were fitted and their predictive performance was evaluated by the root mean square error of prediction (RMSEP) and number of latent variables (LVs) Place and Duration of Study: School of Pharmacy, University of Maryland, between July to November 2012. Results: N1-AS combination (in this order) was found to give best results as it decreased the RMSEP by 15 % (from 0.92 to 0.77) with same number of latent variable used in the model (2 LVs). Other preprocessing methods have not yielded good results for this application. The results obtained by the proposed method were compared with those of the reported HPLC method. No significant difference has been observed regarding both accuracy and precision. Conclusion: FT-Raman Spectroscopy with PLS regression can be used as a reliable alternative for HPLC method, as it is faster and does not require sample pretreatment procedures. K