CRYSTAL-STRUCTURE SOLUTION FROM POWDER X-RAY-DIFFRACTION DATA - THE DEVELOPMENT OF MONTE-CARLO METHODS TO SOLVE THE CRYSTAL-STRUCTURE OF THE GAMMA-PHASE OF 3-CHLORO-TRANS-CINNAMIC ACID

Citation
Bm. Kariuki et al., CRYSTAL-STRUCTURE SOLUTION FROM POWDER X-RAY-DIFFRACTION DATA - THE DEVELOPMENT OF MONTE-CARLO METHODS TO SOLVE THE CRYSTAL-STRUCTURE OF THE GAMMA-PHASE OF 3-CHLORO-TRANS-CINNAMIC ACID, Chemistry of materials, 8(2), 1996, pp. 565-569
Citations number
11
Categorie Soggetti
Chemistry Physical","Material Science
Journal title
ISSN journal
08974756
Volume
8
Issue
2
Year of publication
1996
Pages
565 - 569
Database
ISI
SICI code
0897-4756(1996)8:2<565:CSFPXD>2.0.ZU;2-6
Abstract
Many important crystalline materials do not form single crystals of su fficient size and/or quality for single-crystal diffraction studies, a nd in such cases it is essential that the crystal structure can be sol ved from powder diffraction data; however, there are many difficulties associated with solving crystal structures, ab initio, from powder di ffraction data. In this paper, we report the successful application of a Monte Carlo technique to solve the (previously unknown) crystal str ucture of the gamma-phase of 3-chloro-trans-cinnamic acid from powder X-ray diffraction data. The ''structural fragment'' used in the Monte Carlo calculation comprised a rigid trans-cinnamic acid molecule (with the chlorine and hydrogen atoms omitted) with its oxygen atoms at a f ixed distance from the crystallographic center of symmetry, and with t he center of symmetry lying in the molecular plane. The structural fra gment was rotated by a random angular displacement around a random axi s constrained to pass through the center of symmetry. The ''correct'' position of this structural fragment was discriminated readily (on the basis of the agreement between experimental and calculated powder X-r ay diffractograms) from ''wrong'' positions sampled during the Monte C arlo calculation, and the ''correct'' position was then used as the in itial structural model in Rietveld refinement and difference Fourier c alculations. The paper concludes with a discussion of general consider ations relating to the application of the Monte Carlo method for cryst al structure solution from powder diffraction data.