The x-ray and neutron diffraction pattern of crystalline materials is usually dominated by the Bragg scattering contribution. It provides information about the content of average crystalline unitcell. Moreover, frequently a less intense diffuse scattering intensity is also present. Analyzing it, differences from the average structure, e.g. short range order correlation between atoms, might be revealed. These two kinds of scattering types form the total scattering powder diffraction pattern.
Recently, modeling of the total scattering powder diffraction pattern of crystalline materials has become popular (see for instance [Keen 2015]). Several methods exist for this computation. One of them is the Reverse Monte Carlo (RMC [McGreevy 1988]) method, which does it on a data-driven way. During its application, the configuration containing atomic coordinates is varied by changing the positions of atoms in order that the total diffraction pattern calculated from configuration gradually approaches the measured diffraction pattern. At the end of this procedure an atomic configuration is obtained that is consistent with the measured data. Then, investigations of correlations and short-range order can be performed using the atomic coordinates of the configuration(s).
However, the RMC method might result in such kind of configurations that are not optimal from the point of view of minmal energy of the system. To achieve a more preferable situation, applying simple geometrical contraints is the usual procedure; exploiting interaction potencials migh result in even more realistic structures (for liquids, see. [Gereben 2012]). We expect that introducing potentials will facilitate the study of total diffraction patterns of molecular crystals, where molecules consist of some tens of (or even more) atoms.
The task of the PhD student would be the integration of two-body and bonded kind of potentials (which latter realizes the connection between atoms in the molecules) into the RMC procedure using the already available RMCPOW simulation code ([Mellergård 1999]). She/he would demonstrate the applicability of this method starting from molecular dynamics simulations for recently studied (e.g. ice Ih, CBr4 [Temleitner 2010]) and new materials, as well.
[Gereben 2012] Gereben O., Pusztai, L.: "RMC_POT, a computer code for Reverse Monte Carlo modeling the structure of disordered systems containing molecules of arbitrary complexity", Journal of Computational Chemistry 33(2012), 2285, DOI: 10.1002/jcc.23058.
[Keen 2015] Keen, D. A., Goodwin, A. L.: "The crystallography of correlated disorder", Nature 521(2015), 303. DOI: 10.1038/nature14453.
[McGreevy 1988] McGreevy, R. L., Pusztai, L.: "Reverse Monte Carlo simulation: A new technique for the determination of disordered structures", Molecular Simulation 1(1988), 359, DOI: 10.1080/08927028808080958.
[Mellergård 1999] Mellergård, A., McGreevy, R. L.:"Reverse Monte Carlo modelling of neutron powder diffraction data", Acta Crystallographica A 55(1999), 783. DOI: 10.1107/S0108767399000197.
[Temleitner 2010] Temleitner, L., Pusztai, L., "Local order and orientational correlations in liquid and crystalline phases of carbon tetrabromide from neutron powder diffraction measurements", Physical Review B 81(2010) 134101. DOI: 10.1103/PhysRevB.81.134101.
Above the general expectation, programming skills in C++ and/or Fortran2003/2008 languages and skills in molecular dynamics would be also helpful.