Molecular conformations, celluloses

It is evident that the solid-state NMR data combined with vibration spectra call for a reasessment of earlier views of cellulose I and II. Furthermore, earlier interpretations of diffractometric data that ignore differences in molecular conformations will, according to Atalla 191 be insensitive to subtle structural variations which are central to the phenomena of polymorphy and so clearly indicated by the NMR... 

Fig. 2 Schematic representation of cellulose structures in solution Part A shows the fringed micellar structure. Parts B and C show possible chain conformations of celluloses of different DP. For high molecular weight cellulose, C, intra-molecular hydrogen bonding is possible...

Recent studies of the Raman spectra of Celluloses I, II, and IV have indicated that the different polymorphic forms involve two basically different molecular conformations in addition to the differences in crystalline packing (7,8,9). The conformation variations suggested by the Raman spectra are such that they could play an important role in determining the susceptibility of glycosidic linkages to attack by hydrolytic agents. The questions raised by this possibility will be addressed in this chapter. 

Both in theory and in practice there exist eight gluco-pyranose homopolymers, and some of the molecular conformations of three of these, i.e. cellulose and amylose (l.,2., 3,4), and (1+3)-8-D-glucan (5.,6.,.7) have been established by x-ray analysis. Although (1+3)-a-D-glucan is among the five homopolymers previously unsolved by x-ray diffraction, possible chain conformations were predicted with computers to be an extended ribbon (8.,9.) a single helix (9.), or a double or triple helix (10). 

In the analyses of the spectra of model compounds, changes of the magnitude indicated in Figure 8 were associated exclusively with the occurrence of differences in conformations. It seemed very probable therefore that the differences between the spectra of celluloses I, II, and III reflect changes in the skeletal molecular conformation accompanying the transition from one form to the other. Since the basic ring structure is not... 

In view of the considerable variation observed in the Raman spectra of celluloses as a result of changes in molecular conformations, there can be little question that the spectra in Figure 9 represent evidence that the conformations of the cellulose molecules in Vahnia and Halocynthia are essentially identical. It is also important to note that the Raman spectra of the celluloses from V. macrophysa and V. ventricosa, both of which have been used in different studies as representatives of the Iq, form, are effectively indistinguishable in all regions of the spectra. This is also true of the Raman spectra of celluloses from the algae C. glomerata and Rhizoclonium heirglyphicum, which have also been used in many studies as representative of celluloses that are predominantly of the Iq, form. 

Cellulose, which Is one of the most abundant organic substances found In nature, has been extensively studied by various techniques such as x-ray scattering, electron microscopy, IR and Raman spectroscopy, NMR spectroscopy etc. However, the crystal structure and noncrystalline state are not yet solved for cotton, ramie, bacterial and valonla celluloses which can be easily obtained in pure form. Cross-polarization/magic angle spinning(CP/MAS) C NMR spectroscopy is a promising new method to study these unsolved problems of cellulose, because this method is very sensitive to local molecular conformations and dynamics. 

Figure 1 shows a 50 MHz CP/MAS C NMR spectrum of ramie cellulose and a stick-type nmr spectrum of low molecular weight cellulose( DP <10) In deuterated dimethyl sulfoxide solution(DMSO) (8 ) (The broken and solid lines In the CP/MAS spectrum will be explained below.). As already reported(9,10), the assignments for the Cl, C4 and C6 carbons are relatively easy, based on analogies with the solution state spectrum. However, It should be noted that these resonance lines shift downfleld by 2.3-9.6 ppm In the solid state compared to the solution state. The cause of such large downfleld shlfts(to be explained In the next section) Is attributed to the different conformations about the P-l,4-glycosldlc linkage and the exo-cyclic C5-C6 bond in which these carbons are Involved. 

In studies of the Raman spectra of different native celluloses, Atalla (32) concluded that the two forms 1 and Ig consist of molecular chains which have the same molecular conformation. In the chapter by Wiley and Atalla in the present volume, evidence is presented to suggest that though the molecular conformations are the same, the hydrogen-bonding patterns differ in the two forms. 

The key conclusion that is relevant here is that the native celluloses are composites of more than one crystaline form, but that the difference between the two forms lies not in the molecular conformation but in the hydrogen bonding patterns. Thus, it is possible that the native celluloses have unit cells with very similar atomic coordinates for the heavy atoms, but with different coordinates for the hydrogens. The similarities in the heavy atom locations could account for the many comonalities in the diffraction patterns, while the differences in the coordinates of the hydrogen atoms could be responsible for the differences between the patterns. This would account for the greater incidence of nonallowed reflections in the electron diffraction patterns. 

With respect to the con arison between celluloses I and II, the spectral data leave little question that the molecular conformations are indeed different. The chapter by Wiley and Atalla sets forth some of the evidence based on Raman spectroscopy. The validity of the theoretical arguments developed in support of the hypothesis that two distinct conformations do indeed occur has been demonstrated through its application in studies of model compounds. The most con rehensive is a study of the vibrational spectra of the inositols (47), wherein spectra of seven of the isomers were investigated and the effects of conformational differences accounted for. 

Figure 11 shows the spectra of the noncrystalline components of rayon fibers with the water contents of 0% and 158%( ). In contrast to the case of cotton, the linewidths of the respective resonances do not remarkably decrease by the addition of water for examples, the half-value width of the Cl resonance line is 205 Hz for the hydrated sample, whereas it is 256 Hz for the dry sample. As pointed out above, this fact implies that the molecular mobility of the noncrystalline chains does not greatly increase with the increase of water content. Moreover, the noncrystalline component of rayon does not undergo such a significant change of distributions in torsion angles 4> vjj as observed for cotton cellulose, possibly because the molecular conformation of this component is rather random in the dry state. In other words, such a disordered conformation may hardly allow marked distortion of the noncrystalline chains to be produced upon drying cupra rayon. 

The differences between the spectra of ramie and Valonia are quite small compared to the differences between native cellulose and cellulose II (see Figure 7). In the spectra of ramie and Valonia, the different peak widths and relative intensities can be attributed to the difference in the crystallite sizes. In the spectrum of cellulose II, however, the frequency and number of peaks is significantly different. In previous publications, the differences between the spectra of celluloses I and II have been interpreted as evidence for different conformations in celluloses I and II (40-41). The spectral differences which are indicative of conformational change are not observed in the spectra of ramie and Valonia. Since ramie and Valonia have different I to Ip ratios, it would appear that celluloses 1 and Ip must have similar molecular conformations. 

On the molecular level, cellulose is a linear polymer of P-(l 4)-D-gluco-pyranose units in " Ci conformation. The fully equatorial conformation of 3-linked glucopyranose residues stabilizes the chair structure, minimizing its flexibility. 

Meader D., Atkins E.D.T., and Happey F. 1978. Cellulose trinitrate. Molecular conformation and packing considerations. Polymer 19 1371-1374. 

Studies of the molecular conformation of cellulose sulphate in gels indicated that there are no double helices present. Stereo views of the cellulose sulphate molecule are shown in Figure 1. 

---