Computer simulation interatomic potentials

Ten years ago when I attended a Faraday Discussion on Solid State Chemistry New Opportunities from Computer Simulations, interatomic potential methods were well developed and the use of ab initio methods starting to become widespread. In his Introductory Lecture Prof. C. R. A. Catlow asked With the continuing growth of the applicability of electronic structure techniques, can we see them as replacing interatomic potential based methods His reply then was there will be a continuing role for interatomic based potential based methods as the field moves to more complex systems. Over the last decade, ab initio electronic structure methods have progressed rapidly and for many applications plane-wave ab initio methods are now the first choice for calculations. Nevertheless that reply still holds true. 

Statistical mechanical theory and computer simulations provide a link between the equation of state and the interatomic potential energy functions. A fluid-solid transition at high density has been inferred from computer simulations of hard spheres. A vapour-liquid phase transition also appears when an attractive component is present hr the interatomic potential (e.g. atoms interacting tlirough a Leimard-Jones potential) provided the temperature lies below T, the critical temperature for this transition. This is illustrated in figure A2.3.2 where the critical point is a point of inflexion of tire critical isothemr in the P - Vplane. 

Ghrayeb R, Purushotham M, Hou M and Bauer E 1987 Estimate of repulsive interatomic pair potentials by low-energy alkalimetal-ion scattering and computer simulation Phys. Rev. B 36 7364-70... 

The overall scheme of ab initio molecular dynamics is similar to that of classical molecular dynamics described earlier but instead of using interatomic potentials, the Schrodinger equation is solved to provide the energy and the forces acting on the particles. The computational cost is huge and most studies are limited to small simulation cells (< 100 atoms) and time-scales of a few picoseconds. Within... 

Although it may be possible to use computation to simulate atomic motions and atomistic evolution, successful implementation of such a scheme would eliminate the need for much of this book if the computation could be performed in a reasonable amount of time. It is possible to construct interatomic potentials and forces between atoms that approximate real systems in a limited number of atomic configurations. Applying Newton s laws (or quantum mechanics, if required) to calculate the particle motions, the approximate behavior of large numbers of interacting par-... 

Jong K. Lee, Interatomic Potentials and Crystalline Defects, Proceedings of a symposium sponsored by the TMS-AIME Chemistry and Physics of Metals Committee and the MSD-ASM Computer Simulation Activity at the fall meeting of the Metallurgical Society of AIME, Pittsburgh, PA, October 6-7, 1980, Metallurgical Society AIME, Warrendale, PA, 1981. 

Madeleine Meyer and Vassilis Pontikis, Computer Simulation in Materials Science Interatomic Potentials, Simulation Techniques and Applications. Proceedings of the NATO Advanced Study Institute on Computer Simulation in Materials Science Interatomic Potentials, Simulation Techniques and Applications, in Aussols, France, 25 March—5 April 1991, in NATO ASI Series, Ser. E Applied Sciences, Vol. 205, Kluwer, Dordrecht, 1991. 

In spite of the great success of the computer simulation methods in the determination of the microscopic properties of the solutions, the capacity of the traditional MD and MC simulations is always limited by the choice of the suitable potential functions to describe the interatomic interactions. The potentials are most often checked by comparison of the structural properties calculated from the simulation with those determined experimentally. The reverse Monte Carlo (RMC) method, developed by McGreevy and Pusztai [41] does not rely upon knowledge of any interaction potential, instead it generates a large set of atomic configurations on the condition that the difference between the experimental and calculated structure functions (or pair-distribution functions) should be minimum. The same structural... 

Many of the earlier computations of thermodynamic parameters associated with hydrocarbon adsorption into zeolites entailed development of interatomic potentials so as to fit reasonably with one particular set of experimental data. As a result, although the correspondence between simulation and experiment was often reasonable [102], the transferability of the potential set from one zeolite composition to ano er or from one type of simulation to another was poor. In principle, if the parameterization truly describes the fundamental physics in an approximate way, it should be viable to develop a more generally applicable set of potentials. 

Computing the interatomic forces is the most time-consuming part in an MD simulation. The use of a cutoff radius is a standard trick of the trade that reduces computational cost by neglecting interactions between atoms separated by a distance larger than the specified cutoff. As described earlier, this truncation results in a discontinuity of both the potential and the force at the cutoff distance, but the drawback thus entailed can be avoided by implementation of either the shifted-force potential or a taper function. 

Finally, we refer to a quite recent paper where a first- principles molecular dynamics simulation of amorphous and liquid Si02 was performed [14]. This work confirmed that computer simulation based on the quantum-mechanical calculation of interatomic energy gives basically the same atomic structure of amorphous Si02 as mentioned above simulations based on semiempirical potential of Eq. (1). 

Tlie classical interatomic potential can be used to carry out MD simulations of fast film growth on a substrate. Although the MD growth rates are several orders of magnitude faster than the experimental rates, the MD-deposited films and their surfaces can be characterized in detail and compared with experimental measurements. The main aim of such MD simulations is a fundamental mechanistic understanding and comprehensive identiheation of chemical reactions that occur on the deposition surfaces, as well as analysis of surface diffusion and relaxation mechanisms. Reaction identification is a very important part of the computational hierarchy it is the key to interpretation of various experimental observations and construction of the list of reactions needed for KMC simulation of film growth. Tlie identified set of reactions can be analyzed further to contribute... 

Computationally efficient ab initio quantum mechanical calculations within the framework of DFT play a significant role in the study of plasma-surface interactions. First, they are used to parametrize classical force fields for MD simulations. Second, they provide the quantitative accuracy needed in the development of a chemical reaction database for KMC simulations over long time scales upon identification of a surface chemical reaction through MD simulation, DFT can be used to address in quantitative detail the reaction energetics and kinetics. Third, DFT-based chemical reaction analysis and comparison with the corresponding predictions of the empirical interatomic potential used in the MD simulations provides further... 

Despite the limitations of empirical potentials, for the last three decades computer simulations have improved the knowledge on physical properties of metals and alloys. In particular, due to the development of empirical interatomic potentials [18,19], it has become possible to describe by the MD technique a great number of solid properties such as recrystallization, structural relaxation, energetic barriers and mixing [23]. The EAM developed by Foiles et al. [18] has successfully described bulk properties of metal and alloys and some surface relaxation and reconstruction features [17,18,24,25], and the empirical potential developed by Ackland and Vitek has been applied successfully to investigate the structure of the noble metal alloys [19] and the deposition of Cu clusters on Au (111) [26] and of Cu and Au clusters on Cu (001) surfaces [27]. 

Molecular dynamics (MD) computations of coalescence have been made for silicon nanoparticles ranging in size from. 30 to 480 atoms, corresponding to a maximum diameter smaller than. 3 nm (Zachariah and Cairier, 1999), The compulations were based on an interatomic potential developed for silicon atoms with covalent bonding. The particle structure was assumed to be amorphous. The MD simulations indicate that the transition between solid and liquid-state behavior occurs over a wide temperature range significantly lower than the melting point of bulk silicon (1740 K), a well-known effect for nanoparticics (Chapter 9), The broadest transition occurred for the smallest particles studied (30 atoms), probably becau.se the surface atoms make up a large fraction of the particle mass. 

Note that other scenarios of glass formation are not excluded. For example, as computer simulations show [6.8, 10], the liquid with Lennard-Joncs potential of interatomic interactions can be vitrified under very high cooling rates ( > It)12 K/s) and rather low temperatures (about 50 K). This computer glass possesses a DRP-structure. Believably amorphous metallic films which can be obtained only by depositing the atom flux on a cryogenic substrate own the DRP-structure. 

In this chapter we review our recent efforts towards understanding many of the salient features of detonation using NEMD simulations. We will focus on large-scale NEMD simulations using a model interatomic potential (denoted REBO) to study generic, but complex, detonation phenomena and the use of a new, computationally more intensive, potential (denoted ReaxFF) that accurately describes a real nitramine energetic material. 

---