NEAR-INFRARED MASS MEDIAN PARTICLE-SIZE DETERMINATION OF LACTOSE MONOHYDRATE, EVALUATING SEVERAL CHEMOMETRIC APPROACHES

The influence of particle size on near-infra red (NIR) spectra is typi cally considered a 'nuisance factor' which many scatter correction met hods attempt to eliminate, e.g., multiplicative scatter correction. Ho wever, particle size is a key issue in the formulation of many pharmac eutical products and has a profound effect on the behaviour of both ra w materials and drug substances during formulation. NIR has already be en demonstrated as a potential alternative particle sizing technique t o current accepted methodology. This investigation assessed several ch emometric approaches that model this information, using lactcose monoh ydrate as the raw material. A variety of modelling techniques were app lied to both zero order and second derivative spectra namely multiple linear regression, partial least squares, principal component regressi on and artificial neural networks. One further data transformation eva luated was polar coordinates, although no statistical data were genera ted. Typically, cross-validation root mean square errors of calibratio n and cross-validation root mean square errors of prediction of approx imately 5 mu m were calculated for all of the modelling techniques. Th ese values are comparable to those associated with the reference techn ique (laser diffractometry). Correlation coefficients of approximately 0.98 for all techniques were also calculated. The predictive abilitie s for models generated using second derivative spectra were found to b e comparable to those obtained using zero order spectra.