Comments on the validity of the direct phasing and Fourier methods in electron crystallography

Abstract

A recent simulation has attempted to evaluate the validity of direct phasing and Fourier techniques in electron crystallography. In response to this study, experimental electron diffraction data from copper perchlorophthalocyanine collected at 1200 kV were re-assessed to determine the most important deviation of these intensities from the single-scattering approximation. While n-beam dynamical scattering has indeed been observed for these electron diffraction intensities (in agreement with the simulation) and has been shown to be important for selection of data suitable for ab initio structure analysis, it is, however, not the major perturbation to data obtained at very high voltages. Rather, a simple correction for secondary scattering provides the best fit to the experimental data, an observation consistent with the analyses of other organic structures. Thus, in order to justify the use of electron diffraction intensities from any substance for an ab initio structure determination, it is, first of all, requisite that the actual conditions used for the diffraction experiment be closely modeled.