New Developments and exciting results at the MSC

William A. Goddard III

Over the last year, the Materials Process and Simulation Center (MSC) has made significant progress in a number of areas of science and technology. I will indicate here some of the more exciting or significant results and advances, most of which will be discussed over the next two days.

Since its founding in 1990, the vision of the MSC has been to connect the fundamental description of materials, chemical, and biological systems at the level of accurate quantum mechanics to the practical applications of science to modern industrial technologies. The strategy is a hierarchy of scales where the parameters and concepts at each level are based on averaging or coarsening from the more fundamental and finer scale. With quantum mechanics, we can, in principle, solve the Schrodinger equation to calculate the properties of new materials prior to experimental synthesis and characterization. Indeed, there has been dramatic progress in recent years in extending these methods to large systems, However so most systems of interest are too large. Thus for prototype systems we calculate the electronic wavefunctions and then average over the electrons to obtain a force field (FF), allowing us to describe the forces in terms of atom positions. With the FF we can carry our Molecular Dynamics calculations (solving the Newton's equations rather than the Schrodinger equation) to describe systems with up to ~ one million atoms=2E This may seem very large but for polyethylene this would be a cube of ~ 25 nm on a side. Thus for simulating larger systems we need to coarsen from atoms to segments or psuedoatoms. Here again the strategy is to choose the parameters in the mesoscale so that the dynamics match the detailed atomistic dynamics. The procedures for doing this are just beginning to be worked out. There might be more that one mesoscale level, but eventually we will get to a continuum description using finite grids to describe the phenomena. Our philosophy is that going up the hierarchy all averaging is based on the deeper level from theory. Thus, we eschew empirical fits to data. In this way, it will become possible to predict new materials, new processes, and design new systems all prior to synthesis and characterization. It is at this point that theory and simulation become the core of new developments in materials, chemistry, and biology. We refer to this as de novo Simulation.

Taking the long view, we see three overriding grand challenge problems that we consider our major goal over the next decade:

We are involved in many activities that do not fit neatly into one of these three categories, but these are the critical problems facing theorists and experimentalists, professors and industrial technologists.

This morning our focus will be on catalysis. Here the big advance has been on the energetics of chemisorbed intermediates on group VIII metals (talks A2 and A3). This work is leading to a new level of understanding about the trends in bonding and substituent effects on metal surfaces. We continue to focus attention on single-site homogenous or heterogeneous catalysts and have made some significant progress on a variety of polymerization catalysts (Talks A4, A5, and A6). In addition, we are continuing our efforts to learn how to activate small alkanes such as methane (talk A7). With catalysts most of the theory is quantum mechanics. Advances here are described in A8 and P1. Additional catalysis projects are described in P2 and P7. Other QM projects involve interpreting photochemical processes (P3 and P13). In addition, P11 describes the use of QM to predict solvation of Al ions.

The afternoon will focus more on ceramics and molecular dynamics. Here there have been advances in the FF, allowing phase transitions to be studied with MD. Thus we have been able to simulate metal oxides and metal alloys and to describe transformations induced by temperature, shocks, and rapid straining. The FF and applications are described in Talks B1-B8.

Our efforts in semiconductors focus on issues in growing high quality GaN (P4, P5) and in characterizing the Si-SiO2 interface.

A major new focus in the MSC over the last year or so is in understanding the chemistry of combustion and explosion of high-energy materials such as HMX. Some of the progress is described in B9 and B10. The polymers important for use with these systems are described in E4 and E5.

Friday morning the focus is more on biological systems. There have been dramatic advances in using advanced Monte Carlo techniques (CCBB) to predict protein structures. This is being used for ab initio protein folding predictions (D2, P9, and P8). In addition, we are using these techniques to predict structures for membrane bound proteins such as olfactory receptors (D3) and endothelial cells (D1). Other projects involve predicting chelators for cancer radiotherapy (D4) and the changes in secondary structure induced by particular amino acid substitutions (D5). For biological systems, it is particularly important to properly include solvation effects in the quantum mechanics and molecular dynamics while keeping the costs practical. Significant progress is been made in this line (D6).

We believe that theory and simulation can play an important role in developing strategies for protecting and improving the environment. This led to developing a program in molecularly based environmental technology. Talk D7 will provide an overview of this program and some of the current progress.

The problems faced by chemical engineers in optimizing chemical manufacturing processes designing new have become even more difficult with the pressure of reducing pollution. This requires research in the basic thermodynamics and rates but also requires predictions of phase equilibria for complex systems. Talk E1 will outline how theory and simulation can provide new data and relationships. P12 shows a practical implementation of empirical fitting to obtain mesoscale parameters. Other talks tackling practical industrial issues discuss pressure sensitive adhesives (E2), the mechanism of wax inhibitors and scale dissolvers (E3), wear inhibitors (B8), and the equation of state of Estain (E5) and Kel-F (E4), polymers used to stabilize explosives. Our main focus in polymers has been on dendrimers, a project in collaboration with several experimental groups around the country. Some of this work is summarized in E6-E8.

Of particular interest are the opportunities and challenges facing the petroleum industry over the next 20 years. We have pursued a number of projects using theory and simulation to address these problems. However, in many systems it is essential to carry out experimental studies in parallel with the theory. Talk E9 will summarize the issues and opportunities. One new project in the MSC, chemical modeling of oil basins is just beginning.

Posters P7, P6, and P12 demonstrate the development of simple PC-based software that incorporate sophisticated theory in ways that are useful for applications scientists. P10 shows the computers and workstations of the MSC and how they are networked.