Berhane Temelso's research efforts are devoted to many topics linked together by the prominent role hydrogen bonding plays in each. Along with Provost George C. Shields, he investigates the structure and property of molecular clusters using computational methods.
Figure 1. We apply computational chemistry tools to problems in physical and atmospheric chemistry.
Initial Stages of Sulfate Aerosol Formation
We apply computational chemistry methods to model the formation of atmospheric aerosols at a molecular level. The end goal is to explain the growth of nanoscale small gas phase clusters to large aerosols and cloud droplets in the micrometer range, as illustrated in Figure 2. While current understanding of aerosol formation mechanisms has improved a lot in the last ten years, many questions, particularly those regarding sub-critical size regime of 3nm diameter or less, remain unanswered. Understanding aerosol formation pathways in the presence of different component vapors, temperature and pressure conditions will provide valuable information about the size and distribution of aerosols in the atmosphere. That will minimize the large uncertainty associated with the role of aerosols on the global climate and refine models used to understand the severity of global warming and aerosols’ possible role in mitigating it.
Figure 2. We study the effect of bases on the initial stages of sulfate aerosol formation.
Structures and Properties of Water Clusters and Other Hydrogen Bonded Systems
We they study the structure and dynamics of small water clusters in collaboration with experimental colleagues. Water is the most fundamental molecule for life and it plays a key role in many processes. However, developing a water model that can describe all its unusual and crucial properties has proven difficult. We predict the most stable water clusters under different conditions and compare these predictions with experimental findings. For example, the next figures show three water hexamers (prism, cage and book) that were predicted to be stable in silico (top) and their experimentally observed rotational spectrum (bottom).
Figure 3. Computations predicted the Prism, Cage, and Book (H2O)6 isomers to be equally stable at 0 Kelvin.
Figure 4. Rotational spectroscopy by our experimental collaborators was able to definitively identify the co-existence of the Prism, Cage, and Book (H2O)6 isomers around 0 Kelvin.
Developing Tools to Study Hydrogen Bonded Systems
Because these molecular clusters are held together by weak and dynamic hydrogen bonds, the kinds of structures they can form and their relative stability is very hard to determine. We develop and apply different tools to 1) sample the large number of configurations these clusters can adapt efficiently and 2) determine which ones are important. For example, we have applied the protocol below to systems ranging from water clusters and sulfate aerosols to small peptides.
Figure 5. A protocol to efficiently search a large number of configurations and determine the most stable molecular clusters.
One recently developed resource is ArbAlign which is a web and command line tool to align any two isomers for accurately assessing their similarity. Another work under development is a Molecular Clusters Repository to compile and share published structures and energies of molecular clusters in a convenient way.
These research projects are largely funded by National Science Foundation (NSF). The calculations require significant computational resources and take advantage of large computer clusters such as a local MERCURY cluster as well as others managed by NSF (XSEDE) and Department of Energy (NERSC).