Theoretical crystal: Density functional theory reveals amino acid structure

Solid structure

Is it possible to obtain a structure using solid-state density functional theory on powder X-ray diffraction data of molecular crystals when those molecules are too large and complex for conventional refinement methods? The approach has now been applied to obtain the true crystal structure of the amino acid L-phenylalanine.

Matthew King and Timothy Korter of Syracuse University, New York and Thomas Blanton of Eastman Kodak Company in Rochester New York, explain that the determination of a crystal structures is often essential to pharmaceutical research as well as to improving our understanding of growth kinetics and energetics in crystal design and engineering. Obtaining good-quality crystals for many of the compounds that are of interest is often impossible for a variety of reasons. There is therefore an interest in finding alternatives ways to obtain structural information from limited data, such as powder-diffraction patterns. Of course, powder data usually only provide lattice dimensions and the possible space group symmetries rather than atomic-level information.

Refinement methods invariably fail to retrieve such data for larger and more complex molecules. The team suggests that density functional theory (DFT), a powerful predictive, analytical technique could now replace refinement methods for such molecules, given that procedures that once required supercomputers can now be carried out on desktop machines thanks to advances in software and hardware in recent years. Indeed, improvements and corrects in DFT has allowed chemists to accurately reproduce crystal structures for several organic molecules recently. The critical amino acid, phenylalanine, had remained an elusive target of crystallography, however.

Recent X-rays

Recent X-ray work suggests that the L-phenylalanine crystal is indeed monoclinic, contrary to earlier powder diffraction studies. Nevertheless, only poor-quality refinements have so far been achieved. Moreover, earlier studies have forced the phenyl rings of adjacent molecules into spaces they should not normally fit and so the DFT study by King, Korter and Blanton would reconcile the various problems seen in previous data and refinements.

"Elucidating the proper molecular geometry configurations in P2 symmetry within the large Z = 8 unit cell was possible only through the rigorous DFT-D optimization," the team says.

The researchers point out that robust DFT-D calculations are essential in the case of L-phenylalanine for obtaining a structure because even a thorough analysis of the PXRD data does not reveal the unit cell structure. They explain how their comparison of the calculated diffraction patterns of the P2 and C2 crystal structures show that the molecules would have to reorientate themselves within a large unit cell but that such movements would not actually have influences the PXRD pattern. As such, no standard refinement technique could yield a meaningful structure on this flexible basis, whereas the 92 atoms succumb to DFT-D quite readily, thus bridging the gap between X-ray data and conventional crystal refinement. Indeed, DFT-D could just as easily bridge the gap for smaller molecules offering a more realistic starting point for a plausible Rietveld refinement.

All in the detail

They add that, "This level of detail would not be possible using traditional refinement methods, such as Rietveld or simulating annealing." Essentially, the research offers an alternative approach to the determination of a full crystal structure of complex organic molecules that can only be produced as powders rather than substantial crystals. "In discovering the true crystal structure for L-phenylalanine, the results of this study set a benchmark for the application of DFT-D in complete structural determinations from limited crystallographic diffraction data," the researchers conclude.

Korter told SpectroscopyNOW that he and his colleagues have also now carried out a full prediction of a crystal structure of creatine starting from scratch.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.