Analysis of atomic pair distribution function (PDF) has been widely applied and many wonderful results were obtained. However, studies on organic materials haven’t been practically feasible. The major difficulty is to maintain the molecular geometry during refinement. A method towards the solution and refinement of organic crystal structure are developed. By considering constraints on molecular geometries, such as bond lengths and bond angles, and allowing different atomic displacements (ADP) for intramolecular and intermolecular movement, structures to a series of organic materials are obtained.
Fittings are carried out on complex modeling platform DiffPy-CMI. Fitting strategy is described as following: firstly, refinement on a given structure is performed then random variables on molecular positions and orientations, along with restrains on molecular geometries are applied. Analysis on Quinacrione, naphthalene and allopurinol are presented and all results show excellent agreements with single-crystal data. Correct structure always yields the best goodness of fit (Rw) value during fitting processes with random structure variables and reproducibility is excellent. For the case of allopurinol, we are even able to solve and refine the structure in P1 with four independent molecules.