The role of solid state [formula omitted] NMR spectroscopy in studies of the nature of native celluloses

Abstract

Published spectroscopic observations pertaining to the crystal structure of native celluloses are reviewed for the purpose of defining our current level of understanding about crystalline polymorphism in these materials. Emphasis is placed on observations from solid state 13 C nuclear magnetic resonance (NMR), which first led to the postulate that most native, semicrystalline celluloses are composites of two crystalline allomorphs, labeled I α and I β. Historical background is presented, highlighting the structural controversies which mainly arose because different native celluloses were used, each one representing a different mixture of allomorphs. Input from Raman, infrared (IR) and electron diffraction data is included in the discussion of our current understanding of polymorphism in native celluloses. Also noted is the input from more recently studied celluloses (e.g., Halocynthia) as well as from newer processes that convert the I α to the I β form. On the basis of Raman and IR observations, it is argued that the I α and I β allomorphs differ in hydrogen bonding patterns only and that backbone conformations are nearly identical. Also, the point is made that the absence of correlation field splittings in the Raman spectra calls into question (although it does not disprove) whether the normal two-chain-per-unit-cell, monoclinic I β allomorph really possesses two equivalent chains. Considerable discussion is devoted to the allomorphic composition of cellulose crystallites in higher plants. Published methods of NMR lineshape analysis for the higher plant celluloses are reviewed and critiqued, both from the point of view of lineshape theory and from the point of view of self-consistency of inferences that are based on lineshape analyses for different carbons (particularly C1 and C4). It is concluded that higher plant celluloses most likely possess a minor amount of the I α allomorph where the I α/I β ratio is probably less than 0.25.