Brian Laird of the University of Kansas is supported by an award from the Theory, Models and Computational Methods program. The project focuses on understanding the role that atomic and molecular interactions play in shaping thermodynamic and kinetic properties of solid-liquid and solid-solid interfaces. This effort requires development, evaluation and application of new computational methods. The proposed methodology is a mixture of fundamental studies of interfaces using model potentials with studies of realistic materials. Reliable experimental results of interfacial phenomena are challenging and rare. Thus, atomistic simulation is an important means for determining the thermodynamic phenomenology of such processes.
The thermodynamics and growth kinetics of interfaces in condensed matter systems are primarily controlled by two quantities: the interfacial free energy and the kinetic coefficient. The interfacial free energy between two coexisting phases is the amount of work required to reversibly form a unit area of interface. The kinetic coefficient is the constant of proportionality between the growth velocity of an advancing interface and the undercooling drop in temperature. Accurate values of both are necessary for the full understanding of phenomena such as dendritic crystal growth, crystal nucleation, wetting, liquid-metal embrittlement, and others. Specific tasks that will be addressed are: (i) new method to quantify the dependence of the interfacial free energy on atomic and molecular interactions, (ii) direct method development to calculate grain boundary free energies, (iii) method adaptation to study the interface thermodynamics and structure of chemically heterogeneous solid-liquid interfaces; (iv) method for detecting the dependence of the kinetic coefficient on intermolecular forces, and (v) method to calculate the interfacial free energy of molecular systems, including binary mixtures.
Broader impacts include continued participation in a cooperative research team on Dynamics and Cohesion of Materials Interfaces encompassing semiannual collaborative meetings between materials scientists, chemists and physicists. Efforts for training of students in the techniques of computational chemistry, programming and modeling will be undertaken within the Chemistry NSF-funded REU program at Kansas University, which has a strong history of participation by underrepresented groups.
Understanding the interface between a solid and a liquid is crucial to many technologically important material phenomena, including, for example, dendritic growth of crystals (such snowflake-like formations can degrade the strength of cast metal), crystal nucleation and growth, wetting (does a liquid bead up on a solid surface, or does it spread evenly over the surface? – for example putting wax on a car changes the properties of the surface causing water to bead up instead of forming a smooth coating), or the casting/growth of metals and semiconductors. Central to these phenomena are the molecular-level liquid structure at the interface and the work required to form the solid-liquid interface (SLI) – known as the interfacial free energy. Because many SLI’s of technological importance (such as metal-metal SLI’s) are sandwiched between two dense materials, the experimental study of such interfaces – especially at the molecular level - is difficult and definitive measurements are rare. This lack of experimental data has increased the importance of molecular modeling and computer simulation in determining the basic phenomenology. In the Laird group, we have over the last 15 years developed a number of computational methods to study solid-liquid interfaces (SLI) – especially with the calculation of the solid-liquid structure, interfacial free energy and crystal growth kinetics. Thanks this NSF grant we have made advances in the study of the SLIs of chemically inhomogeneous systems (these are systems in which the solid and liquid are compositionally very different). In addition, this grant has contributed to the training of one undergraduate, two graduate students and a highly skilled postdoctoral student. Examples of some of the specific results from the current grant period include: Premelting at a solid-liquid interface: Premelting is the melting of a thin layer of solid at the interface between the solid and another material below the normal melting point of the solid. This has been observed to occur at solid-vapor and solid-solid interfaces – premelting at the surface of ice is believed to facilitate ice skating and premelting at the ice/rock surface has a major role in frost heave. Our simulations on the solid aluminum and liquid lead SLI predict a premelting transition – this is the first prediction of this phenomenon at a SLI. The image provided shows several snapshots from our simulations on the the solid Al (blue)/liquid Pb (red) SLI at various temperatures. Just below the melting point of Al (for this model at 922.4K), one can see the predicted formation of a liquid layer of Al (green). This work was published in 2013 in the prestigious journal Physical Review Letters. Prefreezing at the copper (Cu)-lead(Pb) solid liquid interface: We have calculated the properties of the Cu-Pb SLI near the melting point of Pb. For one specifical crystal orientation, we have observed a prefreezing transition, in which a thin layer of crytalline Pb forms between the solid Cu and liquid Pb. This phenomenon was not seen in the other interfacial orientations – we have demonstrated that this prefreezing layer greatly facilitates the freezing of this crystal orientation relative to the others as the temperature is decreased below the melting point of Pb. This phenomena will help to understand homogeneous nucleation (crystal formation at a surface) for many chemically heterogeneous materials. This work was publised in Acta Materialia – a premier journal in Materials Science. Interfacial free energy for solid-liquid mixture interfaces: A model material called hard spheres is a model in which the atoms and molecules interact much the same way as billiard balls. This model, while quite simple, was instrumental in forming our current understanding of the generic properties of simple liquids. During this grant period we use simulation techology developed in our group to understand the interfacial free energy of a mixture of hard-spheres at a hard wall – this is a simple model for a solution a surface. Through a comprehensive simulation study of this system, we were able to test recent theories of structure and thermodynamics - this work was published in the Journal of Chemical Physics – a premier journal at the border between chemistry and physics. Studies of the time dependence of crystal growth from a solution mixture: In this work, we examined the problem of rapid solidification from a liquid mixture – it is known that if the solidification is rapid enough, some components of the mixture can become trapped and the resulting crystal developes a different composition than for slow crystal growth. There are a variety of different theories of this process, but experiemental data has not been precise enough to compare and evaluate these theories. Our simulations on a model liquid were the most precise and complete of any effort to date and were able to succefully validate one of the several competing theories. This work was published in Physical Review Letters.