Dr. Wilson's research interests focus on the development and understanding of computational chemistry methodology and the application of this methodology to examine interesting problems in a number of areas including environmental chemistry, transition metal chemistry, and materials science. Particular emphasis is on the development of ab initio and density functional methods and basis sets, though presently, there are over 30 active ongoing projects in the Wilson group which range from method and basis set development to a wide variety of chemical problems, many of which are collaborative projects with experimentalists that certainly span the periodic table.
Computational chemistry plays a vital role in understanding chemical processes at both qualitative and quantitative levels, and has proven particularly useful in ascertaining properties that are either difficult or cost-prohibitive to measure by experiment. In fact, ab initio methodology has now reached the level where it is an invaluable help in quantitative predictions of properties such as bond energies and reaction barriers. Using these advanced approaches, it is now possible to address convincingly the chemistry of vital issues such as acid rain and ozone depletion, and aid in areas such as in the prediction and understanding of spectroscopic properties of luminescent materials.
While qualitative studies can be useful for elucidating aspects of such problems, a complete understanding of many chemical systems requires calculations of high accuracy. Advances in ab initio chemistry now make it possible to compute the energetics and properties of small molecules (2-4 non-hydrogen atoms) to an accuracy that rivals, sometimes even surpasses that of experiment. Unfortunately, despite the advances in computing resources and innovative computational methods, severe limitations remain as to the size of chemical systems that can be studied, particularly at a quantitative level. Therefore, a tremendous challenge facing computational chemistry today is how to achieve quantitative results for medium- to large-sized molecules. Adequately addressing this issue will then allow ab initio calculations to reach a new level of usefulness to exciting areas such as environmental, transition metal, and materials chemistry. Dr. Wilson's research includes developing computational approaches that will allow quantitative results to be obtained for medium- to large-sized molecules.
Over the past decade, the use of systematic series of basis sets in conjunction with ab initio methods has proven to be a valuable aid in understanding the successes and failures of ab initio methods, in assessing intrinsic errors in these methods, and in developing a now well-established hierarchy of methods. The key to this work was the development of the correlation consistent basis sets, first introduced by Thom H. Dunning, Jr. in 1989. Since this initial development, families of correlation consistent basis sets have emerged which include sets for first-, second-, and third-row (main group) atoms, as well as diffuse, core-valence, tight d, and weighted core-valence sets. (Wilson has lead the development for some of these sets.) David Feller, and later others, noting the systematic behavior of this family of sets upon increasing the size of the basis set, introduced a procedure which has enabled extrapolation to the complete basis set (CBS) limit. At the CBS limit, all errors resulting from basis set truncation have been eliminated, and, thus, the remaining error, the intrinsic error, is the result of choice of approximation (method). From this approach, a hierarchy of ab initio methods has evolved, and has provided a powerful means to achieve highly accurate descriptions for a wide range of molecular properties and energetics. Unfortunately, such a hierarchy has not yet emerged for density functional theory (DFT), which is a much more computationally efficient (and, therefore, less time consuming) computational approach than ab initio methods. Dr. Wilson's research includes achieving a better understanding, and developing approaches to better understand, the performance and reliability of density functionals.