Near-Infrared Diffuse Reflectance Spectroscopy of Azithromycin Dispersible Tablets.doc

 PAGE \* MERGEFORMAT 11 Near-Infrared Diffuse Reflectance Spectroscopy of Azithromycin Dispersible Tablets Author: Yang Xin He Lang-Chong Wang Na Wu Shaoping Xu Zheng [Abstract] Objective near-infrared diffuse reflectance techniques and chemometric methods for quantitative analysis of Azithromycin Dispersible Tablets. Methods The partial least squares method to establish a mathematical model to predict the set of projections, and the actual content of the samples were determined. The results of 40 samples by the internal cross-validation to establish prediction model, the internal cross-validation coefficient of determination R2 = 99.86, mean square error of internal cross-validation RMSECV = 0.504. With 8 samples for external validation, predicted values and the true value of the correlation coefficient of 0.9994. Predictive value of the average recovery was 100.2% (RSD = 0.85%, n = 8), precision of the method RSD = 0.89% (n = 7), method of stability, RSD = 0.90% (n = 7). Conclusion The method is simple, rapid, and accurate, suitable for the model involving pharmaceutical companies for rapid inspection or quality control. [Keywords:] Near-infrared; diffuse reflectance spectroscopy; Azithromycin Dispersible Tablets; quantitative analysis ABSTRACT Objective To apply near infrared diffuse reflectance spectrophotometry (NIRDRS) and chemometrics for the quantitative analysis of azithromycin dispersible tablets. Methods Partial least squares (PLS) regression models were set up using a calibration set (40 samples). The validation set was set up by 8 samples. Applying this method to predict the validation set and detect the samples. Results The determination coefficients R2 was 99.86 as indicated from the cross-validation, and the true mean predictive error of Root mean square error of cross validation (RMSECV) was 0.504. The correlation coefficient of the true value and prediction value from validation was 0.9994. The average recovery of the prediction set was 10