The typical form for a molecular mechanics force field consists of a foundation of pair-wise terms to describe bonded and non-bonded atomic interactions, with multi-body correction terms to deal with the limitations of pair-wise terms. I present here the first attempts of a molecular mechanics model that is founded on multi-body terms, which are based on the Bond Valence Model (Brown, 2002) and recent developments in the Vectorial Bond Valence Model (Bickmore et al., 2013a; Harvey et al., 2006). I calibrated these models on pressure vs. energy curves for a set of SiO2 polymorphs. The average deviation for the best-fit iteration, with only six adjustable parameters was ±1.98 kcal/mol.

Here we describe the fitting and testing, via molecular dynamics simulation, of a bond-order potential for water with a unique force field parameterization. Most potentials for water, including some bond-order (reactive) potentials, are based on a traditional, many-body decomposition to describe water's structure with bond stretch, angle bend, electrostatics, and non-bonded terms. Our model uses an expanded version of the Bond Valence Model, the Valence Multipole Model, to describe all aspects of molecular structure using multibody, bond-order terms. Prior work successfully related these multibody, bond order terms to energy, provided the structures were close to equilibrium. The success of this equilibrium energy model demonstrated the plausibility of adapting its parameterization to a molecular dynamics force field. Further, we present extensive testing of ab initio methods to show that the ab initio data we obtained, using the CCSD(t)/cc-pwCVTZ level of theory, to augment the fitting set of our parameters is of the highest quality currently available for the OH system. While the force field is not yet finished, the model has demonstrated remarkable improvement since its initial testing. The test results and the insights gleaned from them have brought us significantly closer to adapting our unique parametrization to a fully functional molecular dynamics force field. Once the water potential is finished, it is our intent to develop and expand the Valence Multipole Model into a fully reactive alternative to CLAYFF, a non-reactive potential typically used to simulate fluid interfaces with clays and other minerals.

A large amount of information concerning interatomic distances in the solid state is available, but little has been done in recent times to comprehensively filter, summarize and analyze this information. Here, I examine the distribution of bond lengths for 135 ions bonded to oxygen, using 180,331 bond lengths extracted from 9367 refined crystal structures collected from the Inorganic Crystal Structure Database (ICSD). The data are used to evaluate the parameterization of the bond-length—bond-valence relation of the bond-valence model. Published bond-valence parameters for 135 cations bonded to oxygen, and the various methods used in their derivation, are evaluated. New equations to model the relation are tested and the common form of the equation is found to be satisfactory. A new method (the Generalized Reduced Gradient Method, GRG method) is used to derive new bond-valence parameters for 135 cations bonded to oxygen, leading to significant improvements in fit for many of the ions. The improved parameterization is used to gain crystal-chemical insight into the milarite structure. A literature review of 350+ published compositions is done to review the end-members of the milarite group and to identify compositions that should have been described as distinct minerals species. The a priori bond-valences are calculated for minerals of this structure, and are used to examine the controls of bond topology on site occupancy, notably by localizing the major source of strain of the structure (the B site). Examination of the compositions of all known milarite-group minerals shows that compositions with a fully occupied B site are less common than those with a vacant B site, in accord with the idea that the B site is a local region of high strain in the structure. The bond-length distributions for the ions of the alkali and alkaline-earth metal families are examined. Variations in mean bond-lengths are only partly explained by the distortion theorem of the bond-valence model. I have found that bond length also correlates with the amount of vibrational displacement of the constituent ions. The validity of some uncommon coordination numbers, e.g., [3]-coordinated Li+, [3]-coordinated Be2+, is confirmed. / October 2016