Structure and crystallization of sub-elementary fibrils of bacterial cellulose isolated by using a fluorescent brightening agent

The structure and crystallization of carefully isolated sub-elementary fibrils (SEFs) of bacterial cellulose have been investigated using TEM, WAXD, and high-resolution solid-state C-13 NMR. The addition of a suitable amount of fluorescent brightener (FB) to the incubation medium of Acetobacter xylinum effectively suppressed the aggregation of the SEFs into the microfibrils, as previously reported. However, this study confirmed for the first time that serious structural change in the SEFs occurs during the removal of excess FB by washing with buffer solutions having pH values higher than 6 or with the alkaline aqueous solution that was frequently used in previous studies. In contrast, the isolation of unmodified SEFs was successfully performed by utilizing a washing protocol employing pH 7 citrate-phosphate buffer solution containing 1% sodium dodecyl sulfate. High-resolution solid-state C-13 NMR and WAXD measurements revealed that the SEFs thus isolated are in the noncrystalline state in which the pyranose rings of the almost parallel cellulose chains appear to be stacked on each other. The respective CH2OH groups of the SEFs adopt the gt conformation instead of the tg conformation found in cellulose I (alpha) and I (beta) crystals, and undergo significantly enhanced molecular motion in the absence of intermolecular hydrogen bonding associated with these groups. The main chains are also subject to rapid motional fluctuations while maintaining the parallel orientation of the respective chains, indicating that the SEFs have a liquid crystal-like structure with high molecular mobility. Moreover, the SEFs crystallize into cellulose I (beta) when the FB molecules that may adhere to the surface of the SEFs are removed by extraction with boiling 70 v/v% ethanol and 0.1N NaOH aqueous solution. On the basis of these results, the crystallization of the SEFs into the I (alpha) and I (beta) forms is discussed, including the possible formation of the crystalline-noncrystalline periodic structure in native cellulose.