Search

Molecular dynamics (MD) simulations are capable of giving considerable insight into the polymorphism of organic molecules, a problem of major concern to the pharmaceutical and other speciality chemicals industries. We illustrate some of the challenges involved in small organic systems, which have complex solid-state phase behaviour, including characterizing rotationally disordered phases, modelling polymorphs with very different hydrogen bonding motifs and explaining the solvent dependence of a polymorphic system. Simulating the dynamics within the organic solid state can be very demanding of the model for the weak forces between the molecules. This has led to the development of DL_MULTI so that a distributed multipole electrostatic model can be used to describe the orientation dependence of hydrogen bonding and π–π stacking more realistically. Once a simulation is correctly reproducing the known crystal structures, there are also considerable system-specific challenges in extracting novel insights from t...

The DL_POLY package provides a set of classical molecular dynamics programs that have application over a wide range of atomic and molecular systems. Written for parallel computers they offer capabilities stretching from small systems consisting of a few hundred atoms running on a single processor, up to systems of several million atoms running on massively parallel computers with thousands of processors. In this article we describe the structure of the programs and some applications.

Elastic constants of a set of molecular organic crystals have been calculated within the crystal modeling program DMAREL, which was developed to allow the use of highly accurate, anisotropic atom−atom potentials. A set of six molecular crystals (durene, m-dinitrobenzene, the β form of resorcinol, pentaerythritol, urea, and hexamethylenetetramine) has been chosen to sample a range of intermolecular interactions and crystal symmetries. The sensitivity of such calculations to variations in empirical repulsion−dispersion parameters and the electrostatic model is determined and discussed relative to the other errors in the theoretical model and typical experimental uncertainties. We find that model potentials whose functional form has been simplified often reproduce crystal structures and lattice energies adequately but perform poorly when used to calculate elastic constants. This is because more theoretically justified potentials are required to satisfactorily model the curvature at the equilibrium geometries...

Organic molecules with high non-linear optic coefficients have polar groups often linked by delocalised π-electron systems. Thus the charge distribution around many atoms in such molecules will be highly non-spherical. Current methods of predicting the crystal structures of these molecules, and thus the macroscopic second harmonic generation capability, take no account of these features as they use fixed charges at each atomic site to model the electrostatic interaction. A more realistic representation of the electrostatic interaction is to use a set of multipole moments on each atom, which may be derived from a distributed multipole analysis of an ab-initio wavefunction for the isolated molecule

We are extending the use of the method of static simulation of ionic lattices to modelling phosphates and condensed phosphates, the mineral components of many biominerals whose structure, crystallinity and reactivity can be influenced by substitutions and other defects. Parameters for the interatomic potentials of some phosphates and pyrophosphates have been determined and we present our results for the modelling of α-magnesium pyrophosphate, a representative compound of moderate complexity. The approach has been to treat the compound as an ionic solid with Mg2+ as the cation and P2O7 4 as a predominately covalent anion which is, however, ionically bound to the metal. Good agreement was found between the calculated and experimental structures for this ionic model using formal or residual charges, a Born-Mayer term for non-bonded repulsive interactions, whose parameters were calculated using the electron gas technique, a van der Waals dispersion term and three body terms to account for the directi...

We describe the derivation of a potential model for ferroelectric lithium niobate, by means of fitting parameters in the potential to the experimentally determined structure and properties of the material. In doing so we highlight the strengths and limitations of the technique, compared to theoretical methods for calculating the parameters. We also see that it is important to include ionic polarisation directly in the fit, owing to the role of polarisation in stabilising the ferroelectric structure.

The technique of calculating lattice dissociation energies using static, minimum lattice energy, ionic models has been extended to allow for multiple occupancy of the ionic sites. A particular lattice site can have a fraction x of an ionic species A and a fraction y of an ionic species B, where the position of each can be relaxed separately along with the unit cell dimensions until an equilibrium is reached. Various degrees of long and short range order can be modelled. This technique has been applied to the mineral sillimanite, Al2SiO5, to calculate the effect on the lattice energy of (Al, Si) ordering over the tetrahedral sites. It is found using this method that (Al, Si) ordering with space group Pbmn stabilizes the material by 29.25 kcal/mol (Aliv-O-Aliv), with respect to the completely disordered material.

To investigate the effects of angle bending terms on the structure of amorphous silica, three different potential models have been developed for the molecular dynamics simulation of vitreous silica.

We review the current status of potential models for oxide materials, with emphasis on the applications of potentials in simulation studies. We pay special attention to transition-metal oxides, silicates and ternary metal oxides. Extensions of potentials beyond the two-body formulation are discussed.

We use an approach based upon the Born model of solids, in which potential functions represent the interactions between atoms in a structure, to calculate the phonon dispersion of forsterite and the lattice dynamical behaviour of the beta-phase and spinel polymorphs of Mg2SiO4. The potential used (THB1) was derived largely empirically using data from simple binary oxides, and has previously been successfully used to model the infrared and Raman behaviour of forsterite. It includes ‘bond bending’ terms, that model the directionality of the Si-O bond, in addition to the pair-wise additive Coulombic and short range terms. The phonon dispersion relationships of the Mg2SiO4 polymorphs predicted by THB1 were used to calculate the heat capacities, entropies, thermal expansion coefficients and Gruneisen parameters of these phases. The predicted heat capacities and entropies are in outstandingly good agreement with those determined experimentally. The predicted thermodynamic data of these phases were used to construct a phase diagram for this system, which has Clausius-Clapeyron slopes in very close agreement with those found by experiment, but which has predicted transformation pressures that show less close agreement with those inferred from experiment. The overall success, however, that we have in predicting the lattice dynamical and thermodynamic properties of the Mg2SiO4 polymorphs shows that our potential THB1 represents a significant step towards finding the elusive quantitative link between the microscopic or atomistic behaviour of minerals and their macroscopic properties.