Self-modeling curve resolution (SMCR) kernel analysis of time-dependent near-infrared (NIR) spectra of water and cellulose mixtures

The relationship between the crystalline structure of cellulose and its property to absorb water molecules were explored by self-modeling curve resolution (SMCR) kernel analysis. Sets of time-dependent near-infrared (NIR) transmittance spectra of water and cellulose mixtures were collected during the evaporation of water from ground cellulose. The dissimilarity in the volatility between free and bound water molecules were described in the matrix representation by SMCR kernel analysis. Asynchronous correlation between the water species, which essentially represents the difference in the rate for cellulose to release the absorbed water molecules, reflects the change in the crystallinity of cellulose samples induced by different degree of grinding. Thus, it was demonstrated that the amorphous component of cellulose can be a key factor to control physical properties, such as absorption and retention of water, of pharmaceutical tablets containing cellulose as an excipient. The result supports the fact that a certain level of control for pertinent physical properties of cellulose excipient becomes possible by the appropriate grinding process.