Overview

The purpose of our research is to gain a better understanding of the physical world through chemical means. Specifically, we have used primarily computational chemistry as well as some synthetic organic chemistry and spectroscopic methods to address a variety of chemical problems. Our primary interest has been to better understand the structure and reactivity of organic molecules. Recent projects include analysis of solvation effects in the acetylation of phenols, ring-opening decomposition of thiazolothiazoles, and basicity of highly conjugated polyenes. The latter two projects are collaborations with the Materials Lab at Wright Patterson Air Force Base that have been funded by the Computational Science Committee at Wittenberg University.

If you are interested in learning more about our research or joining us please continue reading or stop by my office (Science 245 E).

Research in the Chemistry Department

The Chemistry Department considers research an integral component of an undergraduate education in the sciences. Most Wittenberg chemistry students conduct a research project before graduation. This can be done at Wittenberg or off-campus, during the school year or over the summer. Research done during the academic year is typically awarded academic credit, though it is possible with sufficient planning to be paid an hourly rate instead. Research done over the summer is typically supported by a $3-5,000 stipend. There are three sources of funding for the summer stipends: the Faculty Research Fund Board, the Computational Science Committee, and the Chemistry Department. Funding from all three sources is competitive and requires a research proposal written in collaboration with a research mentor. Please see below for research proposal suggestions.

Student Collaborators - Past and Present

Steven Koppenhafer ('09)

Adam Jara ('08)

Timothy Verrilli ('08)

David Mowrey ('08)

Ryan Weiss ('08)

Computational Chemistry

Computational chemistry is essentially the development, implementation, and use of mathematical models to describe chemical systems. The development and implementation of the models used in computational chemistry is a highly specialized process that utilizes the expertise of chemists, mathematicians, and computer scientists. Though a highly important endeavor, most undergraduates would find model development and/or implementation difficult. The use of computational chemistry models, however, requires only basic computer skills, some understanding of the model being used, and some understanding of the chemistry being studied. I encourage any students interested in computational chemistry to pursue the Computational Science minor available at Wittenberg, but this is not necessary to pursue research in computational chemistry.

There are two major types of mathematical models used in computational chemistry, molecular mechanics and ab initio. I will not attempt to describe either in any but the most general terms here. Molecular mechanics models use algebraic and geometric equations to determine molecular energy (Equation 1) based upon bond distance, bond angle, bond bending, electrostatic interactions, steric interactions, and other interactions. These equations all contain parameters whose value is optimized to establish the best possible correlation with experimental observables. For example, one common method for determining the energy associated with a particular bond distance is to calculate the difference between the distance A-B and the ideal distance between atoms A and B (Equation 2). The presence of these parameters is both the greatest advantage of molecular mechanics - it makes it "easy" to create a model that compares well with experiment - and its greatest liability - it makes it impossible to create a model to describe a chemical system for which there is a lack of experimental data.

Ab initio models calculate the molecular energy using the kinetic energy of the electrons and nuclei, the attractions between electrons and nuclei, the repulsion between electrons, and the repulsion between nuclei of the molecule (Equation 3). Exact solutions to this equation are not possible for any but the simplest species (e.g., H*), so the development of ab initio models is interested primarily in determining methods to approximate the mathematically difficult portions of the equation. Examples include assuming that nuclei move much more slowly than electrons (Born-Oppenheimer Approximation) and that the electrons move independently of one another (Hartree-Fock Approximation).

Finally, there are two points to clarify. First, there are many methods that lie between the molecular mechanics and ab initio models described above (e.g., semi-empirical and density functional theory). For further reading I recommend Computational Chemistry: A Practical Guide for Applying Techniques to Real World Problems by David Young for math-phobes and Essentials of Computational Chemistry: Theories and Models by Christopher J. Cramer for those interested in more of the mathematical details of computational chemistry. The second aspect of computational chemistry that needs to be clarified is that the models described above calculate the properties of an isolated molecule in the gas phase, not one interacting with other molecules of itself or, more importantly, solvent molecules. The ability to include solvent molecules in computational chemistry models of chemical systems is a relatively recent development described below.

Solvent Efects

Interactions between a molecule of interest (solute) and solvent molecules can dramatically affect structure and reactivity. This is increasingly true for nonpolar, polar aprotic, and polar protic solvents. The presence of interactions between solute(s) and solvent indicates a stabilization of the solute(s). This stabilization is often more significant for some species than others and can result in significantly different structure and reactivity than would be observed in the gas phase. For example, it is known that SN2 reactions proceed more slowly in polar protic solvents because they hydrogen bond to the nucleophile, stabilizing it, and thereby increasing the activation energy of the reaction. Polar aprotic solvents cannot be hydrogen bond donors so they don't stabilize the nucleophile as well and SN2 reactions therefore proceed more rapidly.

Solvent effects can be included in computational models through a variety of methods. The most obvious way is to include explicit solvent molecules in the calculation. The problem is that it is necessary to include 100's-1000's of solvent molecules that dramatically increases the amount of time required to complete the calculation and requires using molecular mechanics methods. Another obvious way to include the effects of solvent is to just consider the affect of its dielectric constant upon the electrostatic interactions within the solute and disregard actual interactions between the solute and solvent. This wouldn't significantly increase computational time, but is also not sufficient for most applications. The two most common methods for including solvation effects in ab initio and density functional theory calculations, polarizable continuum model (PCM), and solvation model (SMx), use the dielectric effect and include solute-solvent interactions through some rather convoluted manipulations to which I will refer you to Computational Chemistry: A Practical Guide for Applying Techniques to Real World Problems or Essentials of Computational Chemistry: Theories and Models for explanation.

Writing a Research Proposal

There are several committees on campus that provide funding for student research over the summer. Everyone should plan to apply to the Faculty Research Fund Board, but the Computational Science Committee will also fund student research in the area of Computational Science. Finally, the Chemistry Department is typically able to supplement the FRFB stipends and fully support several students each summer. The FRFB deadline for Student Summer Research Grants is March 1, 2008. The COSC deadline is always a little later and the deadline for the Chemistry department is flexible. The best way to acquire summer funding is to meet with your prospective faculty advisor early in the Spring semester to discuss the project and have a rough draft of the proposal ready by late January/ early February. Other things to keep in mind:

The audience, particularly for the FRFB proposal, is general. If your non-science major roommate, parent, sibling, etc. cannot understand your proposal it is too technical.

The audience wants to know that you understand the topic, know what you will be doing, and know how you will be doing it.

Explain why the research is important.

Provide sufficient background for a general audience to understand, at some level, what your research is about.

Provide a detailed description of how the research will proceed and how results will be analyzed. Be clear about anticipated challenges and how they will be overcome.