Scientific Research

Below summarized are some of my current and past research projects. For more information, feel free to drop me a line.

AI-powered Materials Discovery

One major challenge in energy storage devices like rechargeable batteries and super-capacitors, is the need for materials with simultaneously high energy and high power storage density. MXenes have shown tremendous potential in this regard. Since their initial discovery, many compositional and structural variations of the two-dimensional MXenes (Mn+1XnTx) have been proposed and synthesized. Each has distinct structural, electronic, and electrochemical properties.

My work combines density functional theory (DFT) and deep learning techniques to select MXene nanomaterials for specific applications, showcasing the synergy between theoretical and AI-driven approaches in nanomaterials design.

High throughput computational materials discovery is a transformative approach that helps rapidly screen and identify new materials with desirable properties. This method significantly accelerates the discovery process compared to traditional experimental methods, enabling the efficient exploration of vast material spaces. Specific to MXenes, instead of varying the “M” metal atom site as has mostly commonly been done, we vary the “X” site atoms, by substituting carbons for combinations of boron, carbon, and nitrogen.

Additive Manufacturing & in situ monitoring and repair

Additive manufacturing (AM), commonly known as 3D printing, is revolutionizing the manufacturing industry by enabling the production of complex and customized parts with unprecedented precision and efficiency. However, to fully realize the potential of AM, the implementation of in situ monitoring and repair techniques is critical. These methods significantly enhance the quality, reliability, and cost-effectiveness of AM processes.

My work focuses on the innovative application of deep learning techniques to analyze sensor data, with the goal of uncovering complex correlations between defects and sensor readings. By leveraging innovative AI algorithms, I aim to enhance the precision and reliability of defect detection and prediction, thereby providing solutions for improving overall manufacturing quality and efficiency.

Force-field development

Force fields are the functional relationships between potential energy and atomic positions which enable inexpensive evaluation of multidimensional potential energy surface (PES). Classical force fields have been used for more than two decades but they limitations in handling complex PES of high dimensional systems due to the ‘simple’ functional form, human expertise, prior knowledge of functional forms, and ad hoc fitting constants. Neural Network Potentials (NNP) have highly nonlinear functional form, making it general and flexible allowing a very accurate representation of reference data from electronic structure calculations. All types of atomic interactions can be described without bias at the same level of accuracy with lesser human expertise and higher opportunities of automation.

My work involves the development of classical reactive and non-reactive force fields for chemical reaction modeling, phonon transport simulations, etc. as well as state-of-the-art NNP's such as E(3)-equivariant interatomic potentials for complex electrolyte-electrode systems.

Computational Spectroscopy

Spectroscopic techniques that non-invasively probe atomic/molecular systems to investigate their structure, properties, and dynamics are steadily growing. However, the accompanied growth in sophistication of these methods makes it challenging to interpret spectroscopic results without the help of computational chemistry. Computational spectroscopy, a derivative of quantum chemistry has the potential to provide predictions of spectroscopy, as well as in developing generalized benchmarked models and simulations for researchers with no access to expensive and sophisticated experimental tools.

My research is focused on computational characterization of nanosized systems which essentially are simulations of Wide angle X-ray Scattering (WAXS), Infrared (IR) Spectroscopy, X-ray absorption near edge structure (XANES), and scanning transmission electron microscopy (STEM) coupled with electron energy loss spectroscopy (EELS). I collaborate with experimentalists to explore deeper theoretical insights to their datasets.