RUI: Correlated Methods for Calculation of the Electronic Coupling Element

Harvey Mudd College, Claremont CA

Investigators

Abstract

Robert Cave of Harvey Mudd College is supported by an award from the Chemical Theory, Models and Computational Methods program in the Chemistry division to develop computational methods for studying electron transfer reactions. Electron transfer reactions are the simplest chemical reactions but have central roles in biological systems and in human-engineered devices that seek to harness solar energy as alternative sources. The initial stages of plant photosynthesis are exquisitely engineered to transfer electrons across large distances, wasting as little energy as possible, in order to make high-energy molecules for later use. The mitochondria in the cells of animals use an analogous "reverse" electron transfer pathway to synthesize molecules that allow muscle movement from high-energy precursors. As we seek to mimic the effectiveness of plants in harnessing solar energy it is critical to understand electron transfer processes at a detailed level. The rate of electron transfer depends on many quantities, but the factor that controls the distance- and orientation-dependence of the rate is called the "electronic coupling element", a quantity that has the potential to yield unprecedented control of rates if we can understand it at a fundamental level. Cave and his coworkers develop new methods for the calculation of the electronic coupling that are more accurate than existing approaches. The new methods and results are used to better design new synthetic solar energy conversion devices and understand mechanisms of biological electron transfer. The work is carried out Harvey Mudd College, an undergraduate institution. The work provides an excellent educational opportunity for students likely to pursue graduate work in chemistry and also supports the annual presentation of a theoretical chemistry workshop to the Society of Women Engineers? On-Campus Day for high school women, where 100-150 high school women are introduced to opportunities in science, engineering and mathematics. Great strides have been made over the past two decades in developing diabatization techniques for extracting the electronic coupling from standard quantum chemical approaches. However, because of the poor scaling of correlated quantum chemistry methods with system size one is often faced with the choice between accuracy and tractability, especially for large electron transfer systems. Given the success of these diabatization methods, it is imperative to turn attention to developing new electronic structure theory approaches, tailored to the electron transfer problem, which are accurate and able to treat considerably larger systems. This work addresses this challenge by developing a series of approximate methods that include correlation in a balanced fashion for all of the zeroth-order states relevant to the electron transfer process. These approaches greatly extend the size of electron transfer systems for which the coupling can be obtained accurately and thus provide useful guidance about the suitability of more approximate methods. In particular a family of correlated methods is developed to calculate the electronic coupling element based on approximations to the Equation of Motion Coupled Cluster approach. The methods scale no worse than MP2/MBPT2, giving access to dramatically larger systems than available to CI or CCSD. Use of PT-based coefficients in place of CCSD coefficients in the similarity transformed Hamiltonian, coupled with truncated excitation spaces tailored to electron transfer systems, lead to the increased scope and speed. Tests of these new methods include a series of model systems where high-accuracy calculations can also be performed, application to calculate the electronic coupling in near-degenerate donor/acceptor and bridge systems, and the study of through-solvent electron transfer. Both of the latter two computational targets have been studied experimentally.