Model Formalisms and Refinement. T. Koritsanszky, Free University of Berlin, Germany

The nature of the static electron density[1] derived from X-ray diffraction data, the modelling procedure[2] involved - with special emphasis on non-trivial ambiguities and their detectation[3], the design and optimization of the refinement[4] and the tests on the physical significance of the results will be discussed. Not only the well known experimental problems but also the difficulties arising from inadequate interpretation of the data, limit the reliability of the electronic properties to be derived. It will be demonstrated how XD[5], the new computer program package developed for such applications, provides,at the same time, a high degree of freedom in modelling and sophisticated tools for monitoring and controlling the refinement.

[1] Stewart, R.F. and Feil, D. Acta Cryst. (1980) A36, 503.

[2] Hansen, N.K. and Coppens, P. Acta Cryst. (1978) A34, 909.

[3] Sen, A. and Srivastava, M., regression Analysis (1990), Springer-Verlag, NY, USA.

[4] Prince, E. and Nicholson, W.L., Structure and Statistics in Crystallography (1985), p. 183, Ed. Wilson, A.J.C., Adenine Press, Guilderland, NY, USA.

[5] Koritsanszky, T.; Howard, S.; Mallinson, P.R.; Su, Z.; Richter, T. and Hansen, N.K.: XD a Computer Program Package for Multipole Refinement and Analysis of Electron Densities from Diffraction Data (1994).