Macromol. Theory Simul. zyxwvut 3, zyxwvu 185-191 (1994) zyxwvu 185 Preliminary potential energy calculations of cellulose Ia crystal structure Alvo Aabloo Tartu University, Department of Solid State Physics, Tahe 4 Street, EE2400 Tartu, Estonia Alfred D. French* Southern Regional Research Center, U. zyxwvu S. Department of Agriculture, 1100 Robert E. Lee Blvd., P.O. Box 19687, New Orleans, LA 70179, U. S.A. (Received: January 25, 1993; revised manuscript of September 6, 1993) SUMMARY: Packing energy calculations were used to evaluate various models of Ia cellulose, based on unit cell dimensions proposed by Sugiyama et al. Both a rigid-ring method, PLMR, and a full- optimization molecular mechanics technique, MM3, were used. The model found to be best with both methods was packed “up” (the z coordinate of 05 is greater than that of C5); 0 6 atoms were in Cg positions, forming sheets of hydrogen bonded chains. With the PLMR program, the energy of the best model was almost 3 kcal/mol lower than the second best model. The MM3 studies also showed substantially higher energies for the alternative models. Also, some alternative PLMR models had substantially higher atomic movement during MM3 optimization. Several years ago, VanderHart and Atalla found that all crystalline native cellulose I is composed of two phases’). More recently, Sugiyama et al. characterized those two phases in Microdictyon tenuius with selected area electron diffraction techniques ’). Two unit cells were resolved, a monoclinic, two-chain cell and a triclinic, one-chain cell. When roughly equivalent amounts of both phases diffract simultaneously, the resulting diffraction pattern is identical to the pattern first indexed as an eight-chain cell by Honjo and Watanabe3). Thus, the eight-chain cell is apparently an artifact of the previous diffraction techniques. Previous structural studies were based on the assumption that cellulose I is only a single phase. Since those patterns contained information from two phases, new studies of cellulose I structure are needed. However, X-ray diffraction data are not yet available, so it is useful to propose a crystal structure based on modeling methods. The triclinic, one-chain unit cell is especially conducive to structural studies using packing energy because the variety of possible structures is minimal. All proposals must be based on parallel chains, and each chain must be identical with its neighbors. This communication reports the results of two different types of packing energy calculations using various models of Ia cellulose. All studies used the cell parameters u = 6,74 zyxw A, b = 5,93 A, zyx c = 10,36 A, zy a = 117 ’, p = 113 O and y = 81 O, as published by Sugiyama et al. The present work shows that reasonable models exist for those published unit cell dimensions. Our work is also part of a larger project to screen, with molecular mechanics, structures first proposed based on fiber diffraction studies. The first method used to select likely structures was based on the PLMR program developed by Pertsin and Kitaigorod~ky~,~). It kept the Arnott-Scott6) residues used in z 0 1994, Huthig & Wepf Verlag, Basel CCC 1022-1 344/94/$03.00