Our Research

Simulating the quantum dynamics of processes in systems with many interacting particles is notoriously demanding. To make these problems tractable, one often turns to semiclassical and mixed quantum-classical approaches that reduce the computational cost while retaining essential quantum effects. Our group focuses on mixed quantum-classical methods, in which a subsystem of interest is treated quantum mechanically and its surrounding environment (or bath) is described in an effective classical-like fashion. The subsystem might be, for example, a chromophore or a key proton/electron in a charge-transfer reaction, while the environment may comprise a molecular scaffold, protein matrix, or solvent.

Over the past few decades, many mixed quantum-classical techniques have been proposed, differing mainly in how they handle subsystem-bath coupling and, consequently, in their regimes of validity. Persistent challenges include accurately describing decoherence, enforcing detailed balance, obtaining reliable long-time dynamics, and treating strong subsystem-bath interactions. One of the most rigorous formalisms, the mixed quantum-classical Liouville (MQCL) equation, is widely regarded as highly accurate but remains computationally demanding in practice. Our research aims to lower these barriers while preserving, and where possible improving, accuracy and robustness.

The transport of protons, electrons, and energy underpins a wide range of chemical and biological phenomena—including hydrogen bonding, enzyme catalysis, photochemistry, and photosynthesis—as well as energy-conversion technologies such as electrochemical and photovoltaic cells. A fundamental understanding of these processes requires theoretical studies of their underlying molecular dynamics and, in many cases, an explicitly quantum-mechanical treatment. At the same time, these processes typically occur in complex environments containing many atoms. To make simulations tractable, we employ mixed quantum–classical approaches in which the key degrees of freedom directly involved in transport (e.g., a proton, an electron, or an exciton) are treated quantum mechanically, while the surrounding environment is modeled in an effective classical-like fashion.

Our group develops and applies mixed quantum–classical methods to simulate charge and energy transport in chemically and technologically relevant systems, including proton transfer, photoinduced electron transfer, proton-coupled electron transfer, vibrational energy transfer, heat transport, and electronic/excitonic energy transport. We have implemented and benchmarked methods such as DECIDE (Deterministic Evolution of Coordinates with Initial Decoupled Equations) for nonequilibrium transport, demonstrating its accuracy for steady-state heat currents in the nonequilibrium spin–boson model and benchmarking it against numerically exact multilayer MCTDH and other approximate schemes. Our work has shown that classical versus quantum sampling of environmental (bath) degrees of freedom can significantly alter steady-state currents and the validity of steady-state fluctuation theorems, and that acoustic phonons can accelerate vibrational exciton transfer and tune sensitivity to reservoir coupling. These studies provide reliable, efficient alternatives to common approximations, clarify best practices for bath modeling, and yield design rules for phonon-assisted transport and thermal management.

Quantum batteries are nanoscale devices that use quantum resources such as coherence and entanglement to store and release energy with rates and control possibilities that can exceed those of classical batteries. They offer new paradigms for integrated energy storage in quantum technologies, photonic and excitonic circuits, and nanoscale electronics, where fast charging, minimal dissipation, and device miniaturization are crucial. In our program, we focus on excitonic quantum batteries that store energy in long-lived, symmetry-protected dark states—optically inactive exciton states that are largely immune to radiative loss. We developed practical laser charging protocols that populate these dark states via nearby bright states and showed that energy—rather than just excitation population—can be stored with minimal loss. Using para-benzene-like rings and stacked-ring motifs as proof-of-concept platforms, we demonstrated controlled discharge via symmetry-breaking perturbations and identified network geometries, operating conditions, and symmetry-breaking strategies that enable robust charging and on-demand energy release. These designs remain effective under realistic static disorder, thermal fluctuations, and a range of dephasing mechanisms, and we have further shown that tailored sink-battery coupling can substantially enhance exciton extraction rates. We also developed chemically explicit, electronic structure–parametrized exciton models that naturally host symmetry-protected dark states.

Together, this body of work provides concrete design principles—spanning network motifs, symmetry constraints, charging and discharging protocols, and environment coupling—that other theorists can adopt to model related open quantum devices, and that experimental groups can leverage to design and test prototype quantum batteries in molecular aggregates and nanoscale excitonic platforms. Our goal is to bridge fundamental quantum transport theory with practical device concepts, enabling realizable quantum batteries that can be integrated into future quantum, photonic, and energy-harvesting technologies.

Phosphonate metal-organic frameworks (MOFs) built from phosphonate linkers form a chemically and thermally robust class of porous materials with rich structural diversity, strong metal-oxygen bonding, and tunable electronic and transport properties. Their stability under harsh conditions makes them promising platforms for gas separations, energy storage, and optoelectronics. Our phosphonate MOF research program leverages density functional theory (DFT) and classical molecular dynamics (MD), in close collaboration with experimental synthesis and characterization, to design, screen, and understand phosphonate MOFs at the molecular level.

We have co-led experimental-computational studies on a series of phosphonate MOFs. For gas separation and stability, we showed that the mixed-linker MOF TUB41 exhibits exceptional pH stability, retains its structure after two years of adsorption cycling, and displays selective CO2/H2O uptake with sizable adsorption enthalpies and MD evidence of ordered CO2 arrangements—features relevant for separations under harsh conditions. For energy and electronic applications, combined DFT and experiment on TUB1, TUB40, TUB75, GTUB3, and GTUB4 revealed narrow and spin-dependent band gaps, directional conductivity (TUB75/TUB1), record single-crystal conductance in TUB40, and conductive/photoluminescent GTUB frameworks, pointing to their use as supercapacitor electrodes and in optoelectronic devices. Across these projects, we performed the DFT/MD studies to locate adsorption sites and energetics, map electronic and spin structures, and work hand-in-hand with experimental collaborators. Together, these efforts advance phosphonate MOFs toward robust gas separation and energy-storage applications.

Hydrogen-bonded organic frameworks (HOFs) are crystalline porous materials constructed from organic building blocks held together by directional hydrogen-bond networks rather than coordination or covalent bonds. Phosphonate-functionalized, porphyrin-based HOFs are particularly attractive because phosphonic acid groups provide strong, reversible hydrogen bonding, enhanced thermal and chemical stability, and potential proton-conducting pathways, while porphyrin units contribute tunable optoelectronic and redox properties. This combination positions phosphonate HOFs as promising candidates for sensors, membranes, photocatalysis, proton-conducting components, and low-cost electronic devices. Our computational HOF program uses density functional theory (DFT) and ab initio molecular dynamics (AIMD), in close collaboration with synthesis and characterization, to understand and design such frameworks at the molecular level.

We have co-led experimental-computational studies on phosphonate porphyrin-based HOFs that combine semiconductivity, microporosity, thermal robustness, and measurable proton conduction. Key results include the synthesis and DFT-validated characterization of GTUB-5, and the demonstration that metal insertion into the porphyrin core (Cu, Ni, Pd, Zn) systematically tunes band gaps, photoluminescence lifetimes, pore volumes, and HOMO/LUMO character—providing a simple design rule for optimizing optoelectronic and sorption properties. Collectively, this work establishes phosphonate porphyrin-based HOFs as a tunable materials platform and delivers metalation-based design strategies and protocols that can be readily adopted by other groups.