Research

Quantum materials refer to the broad classes of compounds that exhibit exotic properties such as unconventional superconductors, topological materials, 2D materials, van der Waals heterostructures, and twisted layered materials. The versatility of materials platforms and emergence of exotic quantum phenomena make quantum materials attractive for exploration in quantum information science and technologies (QIST) such as quantum computation, communication, and sensing. These quantum technologies rely on quantum mechanical states originating from their electronic structures, defects, spin-orbit coupling, spin, and valley degrees of freedom. However, detailed investigations of stability, electronic structures, defects in quantum materials and their heterostructures are only handful for QIST applications.

We will perform systematic study of band engineering in quantum materials to tune the Fermi level and topological bands crossing (e.g., Dirac, Weyl points) using DFT calculations. The work will be performed in two steps. First, we will investigate the tunability of the electronic states as a function of external perturbations such as strain, pressure, and electrical field. In the second step, we will utilize this knowledge to design solid solutions (through alloy engineering), heterostructures, and superlattices to propose specific experiments to realize exotic topological states and associated transport phenomena. An interesting research avenue in this project would be to couple topological orders in materials with properties like ferroelectricity and thermoelectricity. Particularly, we will first focus on topological thermoelectric materials, where the topology of the band structure can enhance the thermoelectric effect. I will also design and screen various quantum materials (e.g., 2D topological materials) for qubit applications.

Recently, halide perovskites have emerged as next-generation solar cell materials showing impressive power conversion efficiency (PCE) of ~ 25%, reaching close to the performance of commercially used Si based solar cells. Due to the relatively easier synthetic and processing conditions of halide perovskites and excellent performance, these materials have a huge potential to reduce the overall cost of solar cell device fabrication and deployment. Although Pb based halide perovskites were shown to exhibit excellent PCE and stability, the use of Pb-containing materials possesses a challenge due to their high toxicity. However, Sn-based perovskites despite being similar to Pb-based compounds, suffer from stability issues and low PCE. There exists a lack of through understanding of the thermodynamics, kinetics, electronic, defect, and transport properties of Sn-based halide Perovskites, which is essential for improving the PCE’s of Sn-based compounds.

Although most of the studies on halide perovskites have focused on examining their potential for photovoltaics and optoelectronic applications, perovskites are multifunctional which exhibits tunability of their properties such as band gap, defect states, carrier concentration through compositional and dimensional modifications. Therefore, halide perovskites should potentially be explored for the other energy related applications such as thermoelectrics. The conversion efficiency of thermeoelctrics is defined by the figure-of-merit, ZT = S2𝜎/(𝜅e + 𝜅l), that has complex interdependence on multiple materials properties such as the Seebeck coefficient (S), electrical conductivity (𝜎), electronic (𝜅e) and lattice (𝜅l) thermal conductivities. Although thermoelectric investigations of the halide perovskites are quite rare in literature, it has been shown that halide perovskites possess ultralow 𝜅l and high S, which make these materials quite attractive for explorations. Currently, the known halide perovskites do not exhibit high electrical conductivity, which hinders further research and potential applications in thermoeletrics. Therefore, there is a huge opportunity to engineer the electrical conductivity of halide perovskites through band, defect, compositional, dimensional engineering. Recently, we have shown that dimensional reduction of all-inorganic halide perovskites lead to glasslike thermal conductivity in crystalline sample, which is one of the desired requirements for high-performance thermoelectrics.

Therefore, a clear knowledge gap exists in understanding the thermodynamics, kinetic, defect and transport properties of the Sn based halide perovskites that would be extremely beneficial for advancement of next-generation photovoltaics, optoelectronics, and thermoelectric devices.

Materials with extreme thermal conductivity are important for device applications. For example, materials with ultrahigh thermal conductivity are used in microelectronic devices to quickly dissipate heat, and materials with ultralow thermal conductivity materials are used in barrier coating, thermoelectrics and thermal data storage devices. Hence, investigation of the heat transfer mechanism is extremely important to under stand the complex correlation between crystal structure, lattice dynamics and phonon transport. This knowledge would help us in engineering the thermal transport properties of existing compounds as well as help us in designing and discovering hitherto unknown compounds. In our research, We recently showed that a class of quaternary AMM'Q3 chalcogenides exhibit very low thermal conductivity due to the presence of rattler cations. In another work,  we showed that higher-order phonon interactions are extremely important in Tl3VSe4.   

In this theme, we will search for semiconductors and insulators having extreme lattice thermal conducitvity using materials informatics approaches. We would investigate the microscopic heat transfer mechanism in ultralow thermal conductivity cry-stalline materials using first-principles parameter free method and also machine-learning based methods which is much faster than the direct approaches. Also, we will determine the thermal transport properties of complex materials such as those with defects, moire superlattices using simpler model which is also very expensive if we calculate it using the direct conventional methods. 

The experimental discovery of novel materials with target properties through trial and error-based approaches can be both significantly time and resource-expensive. However, with the advent of state-of-the-art predictive materials modelling techniques, artificial intelligence methods, high-performance computational resources, and open-source materials databases, new materials with tailored properties can be computationally designed in a much faster time scale. This can significantly narrow down the search space of the compounds to be explored and engineered experimentally. Data-driven studies can also shed light into complicated structure-property relationship of materials which can provide important design principles for new materials. Although very useful, conventional data-driven research tends to use brute-force approach towards novel materials design and hence lacks efficiency. Therefore, it is important to combine data-driven approaches such as machine-learning models or material screening workflow with physics-informed insights to enhance the fidelity and efficiency of the materials design and property prediction models.

However, all DFT-calculated materials databases contain properties of materials such as formation energies, electronic structures, bandgaps, phonon dispersions calculated at zero-temperature, it possesses a fundamental challenge in modelling thermodynamic and transport properties of materials at finite temperature. Although these 0K data are extremely useful in modelling various properties of materials, integration of data-driven methods to generate large scale thermodynamic properties and incorporating them in the design cycles of new materials in currently missing. Therefore, it is important to combine advanced theory and materials informatics approaches to generate data largescale finite-temperature data using e.g., machine learning interatomic potential, which can help in understanding of

thermodynamic and transport properties above 0K. A long-term ambition in this direction is to design closed-loop autonomous infrastructure that would be able design new materials computationally, perform experimental synthesis and characterization utilizing feedback mechanism and active learning.