Desenvolvimento de uma metodologia para a determinação da granulometria de excipientes da Olanzapina utilizando espectroscopia de infravermelho próximo (NIRS) e calibração multivariada

• Autor(a) principal: Belém, Andressa Elna Mesquita
• Data de Publicação: 2017
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UFRN
• Texto Completo: https://repositorio.ufrn.br/handle/123456789/38284
• topic: Tamanho de partícula; Espectroscopia NIR; PLS; Excipientes

In this research, the influence of particle size (distribution) of some excipients used in the manufacture of Olanzapine tablets was investigated, with the near infrared spectra, in order to use this technique as an alternative to the method currently employed. To this end, 3 samples of microcrystalline cellulose 102 and 3 samples of crospovidone were processed using a sieve shaker, and the 18 fractions obtained from each excipient were analyzed in the NIR spectrophotometer. The spectral data acquired were used to create calibration models, applying preprocessing and partial least squares regression (PLSR) chemometric technique, which verify the linear relationship between the variables. For both excipients, the X-7SB model was selected, in which the data were pre-processed using a combination of the Savitzky-Golay smoothing with a 7-point window and the baseline. For microcrystalline cellulose 102, this model was selected based on the RMSEC 0.0035, RMSEP 0.0077, R2Cal 0.998 and R2Pred 0.993, while for crospovidone the values obtained for square root of the mean square error and for the coefficients of determination in the X-7SB model were RMSEC 0.0122; RMSEP 0.0160; R2Cal 0.974 and R2Pred 0.956. In addition, significance tests were performed to verify if there were significant differences between the results of the reference method and the alternative method within a 95% confidence interval. For the paired t-test, the values obtained were 1.1463 for microcrystalline cellulose and 0.8715 for crospovidone (ttabel = 2.571), while in the EJCR the two excipients presented the ideal point within the confidence region. Thus, by the results of the t test and the confidence ellipse, it was possible to observe that the method developed using near infrared and multivariate calibration presented results comparable to the reference method, showing an alternative also of low cost and faster of analysis. Finally, the calculation of the figures of merit linearity, accuracy, precision, limit of detection, limit of quantification, sensitivity and selectivity were also performed. Linearity was assessed by R2 and residual plot and accuracy by RMSE and EJCR. For cellulose an accuracy of 1.9072, LD 0.0133, LQ 0.0445, sensitivity 0.8711 and selectivity 0.0098 were obtained. For crospovidone the values were precision of 1.1453, LD 0.0321, LQ 0.1069, sensitivity 1.2522 and selectivity 0.0176.