Statistical approach for solid-state NMR spectra of cellulose derived from a series of variable parameters

Principle component analysis (PCA) was used to extract components from the solid-state nuclear magnetic resonance (NMR) spectra of bacterial cellulose (BC). Polymers such as cellulose have several domain structures, and their structure and dynamics are reflected in the variety of solid-state spectra derived from different parameters. The complexity of the obtained spectra makes the analysis of spectra from relaxation measurements difficult. Multivariate analysis, such as PCA, has been suggested as an option to improve NMR spectral analyses. In this study, we attempted to extract peak components from cross polarization (CP) experiment data from the variable contact time spectra of BC by using PCA. The extracted peaks were annotated according to previous reports, and the existence of crystalline I alpha was clearly recognized. The important components of the BC structure (that is, crystalline Ia, amorphous form and mobility) were separated in the PCA-loading plot and CP curve. Polymer Journal (2012) 44, 895-900; doi:10.1038/pj.2012.82; published online 16 May 2012