Quantum tunneling, most prominently hydrogen tunneling, allows particles to cross energy barriers without climbing over them, in contrast to the predictions of classical transition-state theory. Traditionally, tunneling has been associated with cryogenic conditions, where over-the-barrier pathways are suppressed. Yet many important transformations in organic and organometallic catalysis occur under ambient or elevated temperatures, where thermally activated tunneling can play a decisive role. We combine quantum chemical calculations, tunneling-corrected kinetic models, and mechanistic simulations to investigate:

• C–H, O–H, and Si–H bond activations where tunneling accelerates rate-determining state.
• High-valent metal–oxo complexes that exploit tunneling to control oxidation pathways.
• Catalytic cycles under normal laboratory conditions, where tunneling can govern rate and selectivity.

We aim to establish thermally activated tunneling as a design principle for next-generation catalysts that deliver faster, greener, and more selective chemistry.

We investigate the chemistry of bioluminescent proteins, particularly luciferases. These enzymes emit visible light through electronically excited intermediates, providing a window into how biological systems control color and efficiency at the molecular level. A central challenge is understanding how subtle changes in the protein environment tune emission wavelength, brightness and quantum yield across organisms. To address these questions, we combine advanced multiscale simulation methods with machine-learning approaches to:

• Identify the reactive intermediates responsible for photon release.
• Rationalize how protein microenvironments modulate emission color.
• Map excited-state pathways and decay processes that govern brightness and efficiency.

Beyond mechanistic insight, our work seeks to translate nature’s light-emitting strategies into the rational design of brighter, color-tunable bioluminescent probes for bioimaging and biosensing.

We study the design and mechanisms of covalent inhibitors, with emphasis on cysteine and serine proteases such as SARS-CoV-2 Main Protease and cathepsin K. Covalent inhibitors offer unique therapeutic advantages by forming strong, often long-lasting bonds with their targets. A key difficulty in their development lies in precisely tuning binding behavior, as some inhibitors bind irreversibly, while others form reversible covalent adducts depending primarily on the nature of the warhead (the reactive functional group) and the active-site environment. To address these challenges, we employ density functional theory (DFT) and advanced QM/MM free-energy simulation methods to:

• Distinguish reversible from irreversible covalent binding inhibitors.
• Rationalize potency of the drugs and selectivity by quantifying reactivity.
• Provide transition-state level insights that are more predictive than static bound structures.

Our overarching goal is to guide the rational design of safer and more effective covalent drugs, particularly against challenging enzymatic targets relevant to human disease.

Plastics such as polyethylene terephthalate (PET) pose a major environmental challenge due to their persistence and resistance to degradation. The discovery of PETase and engineered FAST-PETase has opened new possibilities for sustainable recycling, but these enzymes remain limited by poor stability under industrial conditions and an incomplete mechanistic understanding of their activity.

We employ Empirical Valence Bond (EVB) simulations to probe how electrostatics, protein dynamics, and active-site chemistry govern efficiency. By building transition-state level mechanistic insights, We aim to define strategies for engineering more robust and efficient variants.

Our long-term goal is to enable sustainable enzymatic recycling of plastics to accelerate the development of industrially viable biocatalysts.