|French, Alfred - Al|
|CSONKA, GABOR - Budapest University Of Technology|
Submitted to: Cellulose
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/25/2011
Publication Date: 4/11/2011
Citation: French, A.D., Csonka, G.I. 2011. Hydroxyl orientations in cellobiose and other polyhydroxy compounds – modeling versus experiment. Cellulose. 18:897-909, DOI 10.1007/s10570-011-9539-6.
Interpretive Summary: Recently a widely recognized cellulose scientist, Dr. Rajai Atalla, a spectroscopist retired from the USDA Forest Service, proclaimed that many of the issues of cellulose structure would only be solved by modeling work. These structural issues are widely thought to hold the keys to improved utilization of cellulosic materials such as biomass, wood pulp and cotton, as well as to better understand plant growth. This paper examines whether there is a relationship between the computer models of fragments of cellulose and their experimentally determined structures. It was concluded that the fine details of the experimental structures, which exist with neighboring molecules, cannot be predicted with models that do not include explicit neighbors. This information is of use to scientists trying to understand cellulose structures.
Technical Abstract: Theoretical and experimental gas-phase studies of carbohydrates show that their hydroxyl groups are located in homodromic partial rings that resemble cooperative hydrogen bonds, albeit with long H…O distances and small O-H…O angles. On the other hand, anecdotal experience with disaccharide crystal structures suggested that these clockwise ‘c’ or counter-clockwise (reverse ‘r’) sequences are not prevalent in the crystalline state. The situation was clarified with quantum mechanics calculations in vacuum and in continuum solvation, as well as Atoms-In-Molecules analyses. From the experimental side, the Cambridge Structural Database was searched. Geometric criteria for these sequences were developed. A criterion based on 120º ranges of hydroxyl orientations accepted 4% of sequences as having ‘c,c’ or ‘r,r’ orientations instead of the 7% expected based on chance. Criteria based on an O-H…O angle > 90º and a 90° lower limit of the absolute value of the H-O-C…H improper torsion accepted 7.0% of the 358 sequences as ‘c,c’ or ‘r,r’. Highly variable orientation of the hydroxyl groups in crystals was seen to depend mostly on strong inter-residue or intermolecular hydrogen bonds. That lack of specific orientation in general for the crystal structures was supported by the solvated calculations that showed very little variation in the energy when one of the hydroxyl groups in 1,2-dihydroxycyclohexane was rotated. The vacuum calculations found the energy to vary with rotation by more than 4 kcal/mol, confirming the gas-phase experiments and calculations on more complicated molecules. Molecules examined in some detail include scyllo inositol and native and methylated cellobiose.