It is now possible to calculate many properties including the energetics (total bond dissociation energies or heats of formation) of molecules containing light elements to high accuracy by using correlation-consistent basis sets, coupled cluster theory and including additive corrections for core-valence and relativistic effects and careful treatment of the zero point energy. We propose to develop software for ab initio electronic structure calculations based on molecular orbital theory and density functional theory with the proper treatment of relativistic effects to study complexes of heavy elements in order to assist in understanding and predicting the chemistry of the actinides, lanthanides, and heavy transition metals, molecules critical to DOE missions including environmental management. The proposed work will focus on the development of these electronic structure methods and their implementation in software on advanced massively parallel processor (MPP) computer architectures capable of multi-tens of teraflops to petaflops. The core of the software will be developed within the NWChem and Columbus software suites. We propose to make the software broadly available so that other scientists can use these tools to address the complex environmental problems facing the Department of EnergyÃÂÃÂÃÂâÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂs nuclear production sites as well as other waste sites in the Nation. Our implementation of relativistic quantum chemical methods for massively parallel computers will enable us to simulate the behavior of heavy-element compounds at the same type of level currently available for light-element compounds. In addition, this work will enable us to provide better methods for benchmarks of the additive energetic schemes currently available for light atom compounds. The theoretical and computational methodology so developed will be an invaluable supplement to current, very expensive experimental studies of the actinides, lanthanides, and radioactive heavy transition metal elements, allowing limited experimental data to be extrapolated to many other regimes of interest. The program objectives will be attained through a multi-site collaboration from PNNL, Ohio State University, University of Memphis and Eloret that includes leading researchers in the areas of high-performance computational chemistry and relativistic theoretical chemistry. The new tools can be used to study, for example, the interaction of actinides with organic complexing agents present in tank wastes and with natural aqueous systems (carbonates) in order to better understand their fate and transport in the environment, as well as interactions with new materials such as phosphates and amides for the design of innovative in situ remediation technologies and separation materials. In addition, the proposed work will allow scientists to tackle the complexity of excited states in heavy element compounds especially those comprised of actinide, lanthanide, and heavy transition metal atoms.
