SOLUTION AND REFINEMENT OF DRUG STRUCTURES FROM POWDER DATA. Kenneth Shankland, ISIS Facility, Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OX11 0QX, U.K.

The vast majority of pharmaceuticals used in everyday life are moderately sized organic compounds. Usually, their molecular structures are known from single crystal X-ray experiments. However, polymorphism (which is frequently found in pharmaceuticals) often makes the determination of a particular crystal form very difficult. In such cases, obtaining a structure solution from powder data is an attractive option, but one which presents a considerable challenge given that most of the compounds of interest will crystallise in low symmetry space groups with large unit cells, leading to complex diffraction patterns with lots of peak overlap. Model building is usually precluded due to conformational flexibility in the molecules, so ab initio methods offer the best hope of a solution.

Computational strategies such as standard direct methods, combined maximum entropy / log-likelihood gain and optimal extraction of structure factors will be discussed, but a particular emphasis will be placed on using sample preparation and data collection strategies to maximise the chance of obtaining a structure solution.