Dr. Shaama Mallikarjun Sharada's Talk

Nov 28, 2016 - 16:00

Nov 28, 2016 - 17:00

Rm. No. 240 Chemical Engg. Dept.

Seminars

Dr. Shaama Mallikarjun Sharada Stanford University.

Abstract: The discovery and design of catalysts necessitate an understanding of reaction chemistry at the molecular level. In recent years in silico chemistry has made significant strides in the large-scale screening of catalyst materials. In particular density functional theory (DFT) owing to its simplicity and low cost has been instrumental in making quantum chemistry a viable field of study. The development of DFT involves approximations to the exchange-correlation (XC) potential between interacting electrons. These parameters have been determined based on either satisfaction of certain exact constraints or by fitting the functional to benchmark databases. I will discuss the development and performance of the Bayesian Error Estimation Functional (BEEF) which employs principles of machine learning to systematically determine the optimal compromise between XC model complexity and accuracy. We are also examining the space of available dispersion-correction schemes within the BEEF framework in order to accurately predict adsorption energies and barrier heights on transition metal surfaces. In addition to accurate quantum chemical methods there is also a need for computationally efficient and automated means to determine transition states (TS) of reactions. TS search requires information about the reaction coordinate which typically involves calculation of expensive second derivatives of the electronic energy (hessian). We have developed a computationally efficient gradient-based finite differences Davidson method as an alternative to full hessian calculation. This approach can significantly lower costs associated with the search and characterization of TS’s especially when systems are large and hessians intractable. We have applied this approach to examine the sensitivity of reaction rates of light alkanes to the size and shape of the nanopores containing active sites in zeolite catalysts.Bio: Shaama Mallikarjun Sharada will be joining the Department of Chemical Engineering and Materials Science at the University of Southern California Los Angeles as an Assistant Professor in Fall 2017. Her research will focus on the development of computational tools for in silico electrochemistry and corrosion prevention. She completed her Bachelor’s and Master’s degree in Chemical Engineering at the Indian Institute of Technology Bombay where she also received the Institute Gold Medal for academic excellence. She completed her PhD from the University of California at Berkeley where her graduate research focused on the development of automated reaction path search methods for complex catalytic reactions. She is currently a postdoctoral researcher at Stanford University where she is developing quantum chemistry methods using principles from machine learning for applications in surface chemistry.