Computational Thermodynamics and Kinetics: Diffusion and Kinetics II
Sponsored by: TMS: Chemistry and Physics of Materials Committee, TMS: Computational Materials Science and Engineering Committee
Program Organizers: Niaz Abdolrahim, University of Rochester; Stephen Foiles, Sandia National Laboratories; James Morris, Oak Ridge National Laboratory; Raymundo Arroyave, Texas A & M University

Thursday 2:00 PM
March 2, 2017
Room: 11A
Location: San Diego Convention Ctr

Session Chair: Ebrahim Asadi, University of Memphis; Hesam Askari, University of Rochester

2:00 PM  Invited
A Molecular Simulation Study of the Effect of Composition Gradients on Intermetallic Nucleation: Peng Yi1; Michael Falk1; Timothy Weihs1; 1Johns Hopkins University
    DSC and TEM experiments in the Ni/Al multilayer system have suggested that the formation of certain intermetallic phases can be suppressed by a sharp composition gradient across the Ni/Al interface, giving other intermetallic phases the opportunity to form instead. We studied the effect of a composition gradient on the nucleation of NiAl intermetallic phase from amorphous solutions in the Ni/Al system using molecular dynamics simulations and a semi-empirical EAM potential. We observed that crystallization of NiAl intermetallic was deferred by the presence of a composition gradient. Important thermodynamic quantities including the melting point, interfacial free energy and critical nucleus size were calculated to estimate the thermodynamic driving force as a function of concentration gradient. These calculations were used to validate and develop a classical nucleation theory formulation that incorporates the composition gradient effect, originally proposed by Desre and Yavari (1990).

2:30 PM  
Defect Migration Using Atomistic-continuum Coupling: Liam Huber1; Raheleh Hadian1; Blazej Grabowski1; Jrg Neugebauer1; 1Max-Planck-Institut fr Eisenforschung GmbH
    Atomistic simulations are a powerful tool for understanding underlying mechanisms which control material behaviour and are a critical component of knowledge-based design for new alloys. However, even with modern computational power and highly efficient algorithms, molecular dynamics (MD) simulations still have trouble reaching experimental time- and length-scales. For the problem of length-scales one approach is to couple MD to a less expensive continuum model, often represented using finite elements (FE). We present improvements to the FE-atomic (FEAt) coupling method to make it fully adaptive with respect to which parts of the system are treated atomistically and which are treated by the continuum model. This allows the MD-domain to track the migration of planar and line defects while maintaining the cost-advantage of the coupled method. We apply this method to the study of grain boundary migration in Al and the nucleation and motion of dislocations during Al nano-pillar compression.

2:50 PM  Cancelled
Diffusion Mechanisms of 'Fast Diffusers' in Ti Alloys: Alessandro Mottura1; Lucia Scotti1; 1University of Birmingham
    Different mechanisms such as precipitation and high-temperature properties are governed by the diffusion of solute atoms. In Ti alloys, diffusion behaviour of Fe, Co and Ni has been a long-standing conundrum. Their diffusivity in hcp-Ti is too fast for simple vacancy-mediated diffusion, yet these solute atoms are expected to sit substitutionally within hcp-Ti. In this work, we used density functional theory to calculate the energy of Fe, Co and Ni defects in alpha-Ti. Using the nudged elastic band method, we calculated energy barriers for possible diffusion pathways, computing diffusivities with kinetic Monte Carlo. We find that these solute atoms can occupy severely distorted interstitial sites, making interstitial diffusion possible. In addition, these solute atoms may be able to swap with host Ti atoms without the need for vacancies when sitting at lattice sites, making host diffusion faster.

3:10 PM  
Measurement of Diffusion Coefficients and Investigation on Precipitation in Mg-based Systems Using Diffusion Experiments: Wei Zhong1; Ji-Cheng Zhao1; 1The Ohio State University
    An integrated approach involving liquid-solid diffusion couples (LSDCs) and forward simulations is employed to reliably obtain diffusion coefficients of several important alloying elements such as Al, Zn, Sn, Ca, and Y in Mg. The Ca diffusion in Mg is the first experimental measurement. The experimental data are essential input to reliable diffusion (mobility) databases for simulation of kinetic processes in Mg alloys. Slices of LSDCs are annealed at lower temperatures to induce precipitation at various supersaturations formed in the LSDCs to effectively study the precipitation as a function of alloy composition. Such results can be used to aid alloy design and to validate model predictions.

3:30 PM Break

3:50 PM  
Quasiparticle Approach to Diffusional Atomic-scale Kinetics in Complex Structures: Helena Zapolsky1; Mykola Lavrskyi1; Gilles Demange1; Armen Khachaturyan2; Renaud Patte1; 1University of Normandy, Rouen; 2University of California and Rutgers University
    With the rapid development of nanotechnology, materials science is confronted with the need for an increasingly precise control of final product at nanoscale. It can be met efficiently only by achieving a thorough understanding of physical phenomena at atomic scale. Our goal was to develop a new theory that provides a computationally effective approach to this problem. In this new theory, called the quasiparticle approach, two novelties have been introduced, a concept of quasiparticles, fratons, used for a description of dynamic degrees of freedom and model Hamiltonian taking into account a directionality, length and strength of interatomic bonds. In this paper the results of modelling of different challenging phenomena as solute segregation at grain boundaries, growth of graphene on Ni surface or self-assembly of hexagonal structure with four different kinds of molecules will be discussed.

4:10 PM  
Dissimilar Solid-Liquid Interface Free Energy and Anisotropy of Metals Using Molecular Dynamics Simulations: Seyed Alireza Etesami1; Ebrahim Asadi1; 1University of Memphis
    Understanding interface of dissimilar solid-liquid coexisting structures (at the melting point of one of the elements) is an important problem for liquid phase sintering, liquid infiltration after metal 3D-printing, etc. While the solid-liquid coexisting of similar metals (solidification) has been widely studied using molecular dynamics (MD) simulations, there is no MD work to study the interface properties of dissimilar solid-liquid coexisting structures. We investigate two-phase dissimilar solid (Fe) –liquid (Cu) coexisting structures by MD simulations using MEAM potential. MEAM parameters for binary Fe-Cu alloy is developed by considering high- and low-temperature properties of the alloy. Capillary fluctuation method is used to determine the interface free energies and anisotropy as well as the diffusion rate of liquid Cu into Fe. These data have been used to determine the dihedral angle for Fe as a function of grain boundary angle. The effect of point defects on these quantities have been also investigated.

4:30 PM  
Kinetic Monte Carlo Simulations of the Growth of Gold Thin Films: Michele Fullarton1; Darnel Allen2; Aleksandr Chernatynskiy3; Simon Phillpot1; 1University of Florida; 2University of Wyoming; 3Missouri University of Science and Technology
    Thin films are of technological relevance to many fields such as semiconductors, memory storage, energy generation and other electronics. Inhomogeneity of a thin film can have a detrimental impact on the desired materials properties and cause poor adhesion to underlying ceramic substrates. Gold deposited on aluminum oxide substrates grows in isolated islands which nucleate and form thin films with ill-formed structures with high porosity. Surface diffusion, bulk diffusion and shadowing effects all contribute to the homogeneity of thin film growth. In this work kinetic Monte Carlo simulation in conjunction with density functional theory calculations have been utilized to simulate deposition of gold onto both an existing gold surface and an aluminum oxide surface. By characterizing and controlling the relative importance of surface and bulk diffusion and shadowing, fabrication of gold films can be guided to achieve minimal porosity and improved device performance.

4:50 PM  
Theory and Simulation of Quantum Dot Formation in Heteroepitaxialy Grown Thin Films under External Forces: Nur Seda Aydin1; Ersin Emre Oren1; 1Bionanodesign Laboratory, Department of Biomedical Engineering, TOBB University of Economics and Technology, Ankara, Turkey
    Capillary-driven shape and microstructural evolution under external force fields has been observed in various phenomena (e.g. tissue growth, planetary surface formation and development of quantum dots (QDs)). Despite the extensive research, this subject is still a challenging theoretical problem. Here, a well-posed moving boundary-value problem, describing the dynamics of curved interfaces/surfaces under electric and stress fields, is obtained via irreversible thermodynamics treatment of surfaces and interfaces with singularities (i.e. tripple junctions). We modeled the formation of QDs during heteroepitaxial growth, and carried out extensive simulations to understand the interplay between stable QDs and material properties including crystallographic orientation and initial thickness of the film, diffusion and surface stiffness anisotropies, surface and interfacial energies, wetting contact angle, mismatch stresses and external electric and stress fields applied. The simulations revealed the ternary diagrams that show the stable QD configurations for a given set of material properties. Supported by TUBITAK (grant no 315M222).