Research Areas

Our research group aims to revolutionize the materials development cycle by establishing a robust data-driven framework. By transitioning from the traditional Edisonian trial-and-error approach to a predictive paradigm, we significantly accelerate the identification and optimization of next-generation functional materials. We integrate high-throughput computational workflows with advanced data analytics to navigate vast chemical and structural spaces, transforming raw data into actionable scientific insights.

• Materials Ontology for Efficient Materials Design Decision Making

  To ensure data interoperability and enable autonomous knowledge discovery, we focus on the construction of a formal Materials Ontology. This structured framework standardizes the complex relationships between synthesis parameters, crystal structures, and multi-scale properties. By establishing a unified language for materials data, we create a searchable and reusable knowledge graph that provides the logical infrastructure required for AI-driven autonomous research systems.

• Materials Informatics for Accelerated Materials Discovery

  We employ Materials Informatics to extract hidden patterns and governing laws from high-dimensional materials datasets. Beyond simple regressions, our research focuses on advanced feature engineering and descriptor development to quantify complex crystal symmetries and electronic environments. By utilizing deep learning architectures and statistical learning, we perform multi-objective optimization to discover novel candidates for energy applications, ensuring a fundamental understanding of the structure-property-performance relationship.

• AI-driven Predictive Modeling for Materials Properties

  We develop robust AI-based predictive models to explore the vast and complex landscape of material properties. By integrating advanced statistical learning with physical descriptors, we build high-fidelity surrogate models that enable rapid and accurate property estimation across diverse chemical spaces. Our approach focuses on enhancing model generalizability and data efficiency, allowing for the reliable discovery of novel materials even when experimental or computational data is limited.

We explore the frontier of low-dimensional systems and nanostructures to engineer novel functionalities at the atomic scale. By leveraging quantum mechanical simulations, we design and optimize materials with tailored electronic, optical, and magnetic properties, providing the fundamental blueprints for next-generation energy devices.

• Atomic Scale Materials Engineering for Futuristic Optoelectronics

  We engineer electronic band structures and excitonic properties to maximize the quantum efficiency of semiconductor materials for advanced optoelectronic applications. By precisely controlling charge carrier dynamics and optical transitions in emerging functional materials, including perovskites and heterojunctions, we identify optimal candidates for next-generation light-harvesting and emitting devices.

• Energy Storage and Production System

  We conduct multi-scale modeling of electrochemical interfaces to enhance the performance of batteries and catalysts. By investigating ion transport kinetics and reaction pathways at the atomic level, we design stable and high-capacity electrodes for lithium-ion batteries and highly active, cost-effective catalysts for water splitting and fuel cells, addressing the core challenges of sustainable energy systems.

• Spin-controlled Materials for Energy Applications

  We investigate the spin-dependent properties of materials to develop energy-efficient spintronic and magnetic devices. Our research leverages the interplay between lattice, spin, and orbital degrees of freedom to design materials with programmable magnetic properties, enabling the development of high-performance spintronic devices and next-generation magnetic materials.

We utilize high-fidelity simulations to provide a fundamental understanding of materials at the atomic and electronic levels. By modeling the response of matter to external stimuli, we bridge the gap between theoretical predictions and experimental observables, facilitating the precise identification of local structures and active sites in complex energy and functional materials.

• Advanced Spectroscopy & Microscopy Simulation

  We perform ab initio simulations of advanced characterization techniques, including X-ray Absorption Spectroscopy (XAS), Electron Energy Loss Spectroscopy (EELS), and Scanning Tunneling Microscopy (STM). Our approach provides a direct mapping between electronic structures and experimental signals, enabling the unambiguous determination of coordination environments and surface electronic configurations at the atomic scale.

• Theory-Experiment Complementary Materials Characterization

  We implement a collaborative feedback loop between computational modeling and experimental characterization. By synthesizing theoretical "fingerprints" of candidate structures, we enable the unambiguous identification of metastable phases and defect configurations observed in operando experiments. This synergistic approach ensures the structural integrity of our material models and provides a rigorous physical basis for interpreting experimental data in real-time.