Skip to main content
ARS Home » Southeast Area » New Orleans, Louisiana » Research » Publications at this Location » Publication #326682

Title: Calculating cellulose diffraction patterns

item French, Alfred - Al
item Santiago Cintron, Michael
item Nam, Sunghyun
item YUE, YIYING - Louisiana State University
item WU, QINGLIN - Louisiana State University
item AGARWAL, UMESH - Forest Products Laboratory
item SIMKOVIC, IVAN - Slovak Academy Of Sciences

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 2/21/2016
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Although powder diffraction of cellulose is a common experiment, the patterns are not widely understood. The theory is mathematical, there are numerous different crystal forms, and the conventions are not standardized. Experience with IR spectroscopy is not directly transferable. An awful error, that there is a peak for amorphous material in a normal cellulose I pattern at 15° 2', is repeated in the literature. One paper [ ] tries to standardize conventions for axis labels and peak Miller Indices for the various crystal forms. Thus, the main reflections of native cellulose I' have Miller indices of (1-10), (110) and (200), with a=7.78Å, b=8.20Å, c=10.38Å and a monoclinic angle '=96.5°. Other points included simply converting the crystal structure coordinates (e.g. Ref. ) into idealized powder patterns with the Mercury program. [ ] The calculated pattern for cellulose I shows that the peaks near 15° 2''are not for amorphous material; they arise from the coordinates of the crystal structure. No information relating to amorphous material goes into Mercury. Another work shows how crystallite size affects the diffraction pattern and changes the Segal Crystallinity Index [ ], casting doubt on the validity of the Segal method. Instead, it was suggested that researchers should calculate patterns that looked as much as possible like their experimental ones and then describe the inputs to the calculated patterns in a trial-and-error approach. Calculated patterns of cellulose I and II could be mixed in different proportions on a spreadsheet. Then it was found that the calculated pattern for cellulose II with a very wide peak width at half height of 9° mimicked the pattern for ball-milled, or amorphous, cellulose, [ ] and that the Solver routine in Excel could be used to minimize the calculated differences between the experimental and calculated patterns based on various amounts of cellulose I, II, and amorphous material. However, this was still a trial-and-error approach, needing numerous different calculated patterns for different samples. The Rietveld method compares observed and calculated patterns based on possible optimization of all variables, even atomic coordinates when the data are good enough. We have joined others who are using it. The Maud program, a free download, has many desired capabilities, [ ] with on-line tutorial videos and schools on its use available. It can interpret the experimental data in terms of the amount of each phase (e.g., cellulose I, II, and amorphous) and the isotropic size of the different crystallites.