Professor Hiromi Nakai has pioneered cutting-edge computational chemistry methods by integrating quantum chemistry with molecular simulations, enabling the detailed molecular-level elucidation of a wide range of dynamic phenomena in large-scale complex systems, including chemical reactions, energy transport, and photoexcited processes. In addition, through the development of machine-learning-based density functional theory, the discovery of universal chemical principles governing excited states, and the establishment of data-driven links with experimental chemistry, he has significantly expanded the theoretical framework of modern chemistry and opened new research directions across diverse chemical disciplines. His principal achievements are summarized below.

1. Development of Quantum Mechanical Molecular Dynamics Methods and Their Application to Large-Scale Complex Systems
To enable simulations of chemical reactions in large-scale complex systems, Professor Nakai developed the divide-and-conquer density-functional tight-binding (DC-DFTB) method and pursued its large-scale and high-performance implementation. In particular, by designing original massively parallel algorithms optimized for supercomputers such as K and Fugaku, he achieved the world's first quantum chemical calculations for systems containing on the order of one hundred million atoms, dramatically extending the practical limits of quantum chemistry. Building on this foundation, he established the DC-DFTB molecular dynamics (DC-DFTB-MD) method, which allows for the explicit treatment of bond formation and cleavage. Using this approach, he quantitatively analyzed proton diffusion in water and ice, clarifying the respective contributions of the vehicle and Grotthuss mechanisms and their dependence on crystalline phase. He further elucidated the origin of anomalously high proton conductivity in nanotubes as arising from collective Grotthuss diffusion supported by ice-like structured water. In studies of CO₂ chemical absorption processes, he demonstrated that fundamentally different microscopic reaction pathways govern absorption and desorption, with hydrated-proton Grotthuss transport dominating absorption and ion-pair mechanisms prevailing during desorption. He also identified the pathways and timing of multistep proton transfer in bacteriorhodopsin, providing a molecular interpretation of time-resolved X-ray free electron laser (XFEL) structural data. In addition, his simulations of lithium and post-lithium secondary batteries revealed that carrier-ion diffusion in highly concentrated electrolytes is controlled by ligand-exchange reactions, underscoring the crucial role of theoretical chemistry in electrolyte design. Through these extensive studies, he refined and released the DCDFTBMD program, which is now widely used by researchers and computational centers worldwide.

2. Development of Nonadiabatic Dynamics Methods and Their Application to Large-Scale Complex Systems
Professor Nakai extended the divide-and-conquer approach to time-dependent DFTB (TD-DFTB) and incorporated nonadiabatic transitions into molecular dynamics simulations, thereby enabling the quantitative tracking of excited-state electron-nuclear dynamics in large-scale systems previously inaccessible to conventional methods. Applying this framework to the photoisomerization of azobenzene, he extended earlier gas-phase studies to solution environments and demonstrated that, while N=N bond rotation dominates the rapid deactivation pathway in the gas phase, solvent-induced constraints enhance the contribution of inversion pathways in solution. Furthermore, he successfully reproduced, for the first time by simulation, ultrafast charge-separation processes occurring immediately after photoexcitation in perovskite and donor-acceptor organic solar cells, where multiple nonadiabatic transitions are involved. These results provide a new theoretical foundation for understanding and designing high-efficiency photoenergy conversion materials.

3. Development of Machine-Learning-Based Density Functional Theory
Professor Nakai introduced machine learning models using the electron density and its gradients as descriptors to construct kinetic energy functionals that significantly surpass the accuracy of all previously proposed analytical functionals, thereby realizing orbital-free density functional theory (OF-DFT). He further extended this approach to correlation energies, establishing a machine-learning-based electronic correlation model that achieves accuracy comparable to that of CCSD(T) at the complete basis set limit, but at a dramatically reduced computational cost. These pioneering contributions represent a major step toward the realization of true density functional theory and constitute a paradigm shift in computational chemistry.

4. Machine Learning Approaches to Experimental Chemistry
Professor Nakai developed systematic machine-learning frameworks that combine electronic-structure-based descriptors with predictive models for reaction outcomes, solvent selection, and reaction-condition optimization, substantially rationalizing experimental chemistry beyond traditional trial-and-error approaches. Moreover, he pioneered a system that automatically collects multimodal sensor data and records them in electronic laboratory notebooks, greatly enhancing reproducibility and enabling efficient data utilization. These developments provide a robust foundation for data-driven experimental chemistry.

5. Discovery of Chemical Principles Governing Excited States
Through systematic theoretical studies of excited states in metal complexes and highly symmetric molecules, Professor Nakai discovered a symmetry rule for degenerate excitations, demonstrating that among multiple states sharing the same dominant excitation configuration, the highest-energy state becomes optically allowed. He established the generality of this rule across a wide range of molecules. In addition, he elucidated that changes in internal rotation barriers observed in the excited states of toluene derivatives originate from a unique π*-σ* hyperconjugation manifested in the lowest unoccupied molecular orbital. Furthermore, he demonstrated that the formation of conical intersections governing nonradiative S₀/S₁ transitions is dictated by exchange interactions between the highest occupied and lowest unoccupied molecular orbitals, thereby formulating a novel chemical bonding theory fundamentally distinct from conventional concepts of stability in ground or excited states. These principles significantly deepen the theoretical understanding of photochemical reactions and provide essential guidelines for the design of photofunctional materials.

 In summary, Professor Nakai has elucidated the fundamental nature of chemical reactions, transport phenomena, and photoexcited processes through the development of cutting-edge computational chemistry methods and their sophisticated application to large-scale complex systems. Through the creation of machine-learning-based theories and the discovery of new chemical principles, he has substantially expanded the horizons of theoretical chemistry. Accordingly, these achievements are recognized as being of exceptional academic and practical significance and are deemed fully worthy of the Chemical Society of Japan Award.