MHManami Hayashi

Manami Hayashi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Manami Hayashi has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Analytic Hessian for mixed-reference spin-flip time-dependent density functional theory

    by Yugo Hirano, Manami Hayashi, Naoto Inai, Takeshi Yanai

    Mixed-reference spin-flip time-dependent density functional theory (MRSF-TDDFT) describes diradicals and nearly degenerate states largely free of the spin contamination of spin-flip (SF) TDDFT, but its Hessian has been available only numerically. Within the Tamm–Dancoff approximation, we derive and implement the analytic Hessian of MRSF-TDDFT with restricted open-shell references, including SF-TDDFT as a special case, and the analytic polarizability derivative for excited-state Raman intensities. For stilbene, the analytic Hessian reproduces finite-difference frequencies to within 0.07 cm⁻¹ at one-nineteenth of the estimated three-point finite-difference cost. The Hessians show that the twisted S₁ intermediate (the phantom state) is an MRSF minimum and characterize the diradical S₀ saddle point. Computed ¹³C isotope shifts are consistent with the assignment of the S₁ resonance Raman band of the central C=C bond; at the phantom state, this band lies about 170 cm⁻¹ lower than at the planar S₁ minimum, a candidate marker of the twist. At the saddle point, SF-TDDFT gives energies and a reaction-coordinate curvature close to those of MRSF-TDDFT, but in a direction orthogonal to the reaction coordinate its Hessian has an unphysical mode at 6005 cm⁻¹ and, at a nearby point on the path, an imaginary frequency of 3358i cm⁻¹. S₀ and the M S = 0 component of T₁ are roots of the same SF response matrix; spin contamination couples them in the Hessian through a term inversely proportional to their gap. MRSF-TDDFT, with separate singlet and triplet response matrices, has no such term. Spin contamination can thus affect curvatures much more than energies.

  • Open Access

    Analytical nuclear gradients for the CC2 method using pair-natural orbitals based on localized virtual molecular orbitals for excited states

    by Manami Hayashi, Shunya Nagae, Masaya Hagai, Takeshi Yanai

    Analytical nuclear energy gradients are indispensable for optimizing excited-state structures and exploring excited-state potential energy surfaces. Local correlation methods based on pair-natural orbitals (PNOs) have extended correlated wavefunction calculations to large molecules, and their analytical gradients have been developed for ground states but, to our knowledge, not for excited states. In our previous work [M. Hayashi, S. Nagae, M. Saitow, and T. Yanai, J. Chem. Phys. 165, 164118 (2026)], we established the analytical gradient theory of the PNO-based second-order approximate coupled cluster (CC2) method using localized virtual molecular orbitals (LVMOs), which was limited to the ground state. In this work, we report the development of analytical nuclear gradients for singlet and triplet excited states of the LVMObased PNO-CC2 method. Using the ground-state PNOs for all excited states, we inherit the ground-state Lagrangian, including the PNO-construction constraints, and add only excited-state constraints. To reduce the excitation-energy errors arising from the ground-state PNOs, we introduce a PNO truncation correction and develop the gradients including this correction. The formulation is extended to a projected-atomic-orbital-based variant and to the spin-component-scaled scheme. To verify the implementation, we extended our automated symbolic derivation program to spin-orbital algebra and to the excitation vectors of the excited states, and the analytical gradients were validated against selection-fixed finite differences. Using the present method, we accomplished analytical excited-state gradient calculations for real-life molecules with 80–91 atoms.

  • Analytical nuclear gradients for the CC2 method using pair-natural orbitals based on localized virtual molecular orbitals for ground states

    by Manami Hayashi, Shunya Nagae, Masaaki Saitow, Takeshi Yanai

    Analytical nuclear energy gradients are indispensable for geometry optimizations and potential-energy-surface explorations, yet their development for pair-natural-orbital (PNO)-based methods remains limited. In this work, we present the analytical nuclear gradient theory for the ground-state PNO-based second-order approximate coupled cluster (CC2) model using localized virtual molecular orbitals (LVMOs). The gradient theory is formulated within the Lagrangian framework as an extension of our previously developed LVMO-based PNO-MP2 analytical gradient theory. Although the use of LVMOs requires additional localization constraints in the Lagrangian, orthonormal LVMOs simplify the coupled-perturbed equations and energy gradient expressions, and these advantages are exploited in the PNO-CC2 analytical gradient formulation. The implementation is extended to the spin-component-scaled variant. To facilitate the development and implementation, we extended our previously developed automated symbolic derivation framework, which systematically generates the derivative expressions for gradient calculations. Benchmark calculations on phenylalkane chains demonstrate favorable scaling of the analytical gradient evaluation. Gradients vary smoothly along a bond-dissociation coordinate of a ketene derivative, indicating that PNO orbital truncation does not introduce noticeable artifacts. Geometry optimizations of difluorodiene reproduce the characteristic elongation of electron-rich bonds observed for the CC2 theory relative to the MP2 theory. Applications to intermolecular complexes show that the PNO-CC2 framework provides optimized structures and binding energies comparable with those from the PNO-MP2 framework. Single-point gradient calculations for realistic molecular systems of around one hundred atoms demonstrate applicability of the implementation. This work establishes the analytical gradient formulation for ground-state PNO-CC2 theory and provides a foundation for future extensions to excited states.

  • Open Access

    Analytical nuclear gradients for the CC2 method using pair-natural orbitals based on localized virtual molecular orbitals for ground states

    by Manami Hayashi, Shunya Nagae, Masaaki Saitow, Takeshi Yanai

    Analytical nuclear energy gradients are indispensable for geometry optimizations and potential-energy-surface explorations, yet their development for pair-natural-orbital (PNO)-based methods remains limited. In this work, we present the analytical nuclear gradient theory for the ground-state PNO-based second-order approximate coupled cluster (CC2) model using localized virtual molecular orbitals (LVMOs). The gradient theory is formulated within the Lagrangian framework as an extension of our previously developed LVMO-based PNO-MP2 analytical gradient theory. Although the use of LVMOs requires additional localization constraints in the Lagrangian, orthonormal LVMOs simplify the coupled-perturbed equations and energy gradient expressions, and these advantages are exploited in the PNO-CC2 analytical gradient formulation. The implementation is extended to the spin-component-scaled (SCS) variant. To facilitate the development and implementation, we extended our previously developed automated symbolic derivation framework, which systematically generates the derivative expressions for gradient calculations. Benchmark calculations on phenylalkane chains demonstrate favorable scaling of the analytical gradient evaluation. Gradients vary smoothly along a bond-dissociation coordinate of a ketene derivative, indicating that PNO orbital truncation does not introduce noticeable artifacts. Geometry optimizations of difluorodiene reproduce the characteristic elongation of electron-rich bonds observed for the CC2 theory relative to the MP2 theory. Applications to intermolecular complexes show that the PNO-CC2 framework provides optimized structures and binding energies comparable to those from the PNO-MP2 framework. Single-point gradient calculations for realistic molecular systems of around one hundred atoms demonstrate applicability of the implementation. This work establishes the analytical gradient formulation for ground-state PNO-CC2 theory and provides a foundation for future extensions to excited states.

  • Open Access

    Analytical first-order derivatives for second-order Møller–Plesset perturbation theory with pair-natural orbitals using localized virtual molecular orbitals: electronic response properties

    by Manami Hayashi, Masaaki Saitow, Takeshi Yanai

    Accelerating analytical energy derivative calculations for large systems remains challenging within wave function-based quantum chemistry frameworks. The pair-natural orbital (PNO) method provides a powerful approach to reduce computational costs while maintaining accuracy. The authors recently reported a fully analytical nuclear gradient theory for the second-order Møller–Plesset perturbation theory using orthogonal localized virtual molecular orbital-based PNOs. This paper provides an overview of the theoretical framework and describes its extension to electronic response property calculations.


  • Analytical nuclear gradients for second-order Møller–Plesset perturbation theory using pair-natural orbitals based on localized virtual molecular orbitals

    by Manami Hayashi, Masaaki Saitow, Takeshi Yanai

    Accelerating analytical nuclear energy gradient calculations within wavefunction-based quantum chemical frameworks and extending their applicability to large systems remain significantly challenging. Recently, Pinski and Neese developed fully analytical nuclear energy gradients for second-order Møller–Plesset perturbation theory (MP2) using pair natural orbitals (PNOs) based on projected atomic orbitals (PAOs) [P. Pinski and F. Neese, J. Chem. Phys. 148, 031101 (2018); P. Pinski and F. Neese, J. Chem. Phys. 150, 164102 (2019)]. Here, we present an analytical nuclear gradient method based on the PNO-MP2 theory using orthonormal and non-redundant localized virtual molecular orbitals (LVMOs) instead of conventional PAOs. Compared to PAO-based approaches, our method exhibits two main differences. First, constraints from the construction of LVMOs are incorporated into the Lagrangian, requiring the solution of an additional Z-vector equation—the coupled-perturbed virtual localization (CP-VL) equation. Second, owing to the orthonormality of the LVMOs, the derivatives of the PNO coefficients with respect to both the MO coefficients and the nuclear coordinates vanish. The latter property significantly simplifies the formulation of analytical energy gradients. In addition, to mitigate the technical complexity of deriving gradient formulas with the PNO treatment, we used an extended scheme implemented in our automatic derivation program. Benchmark calculations on phenylalkane chains, docetaxel, and a host–guest complex demonstrated high computational scalability of our nuclear gradient implementations, additionally confirming that the computational effort required to solve the CP-VL equation is relatively minor. We illustrated that the energy gradients as functions of nuclear coordinates are smooth and that the errors in the predicted structures are small.


  • Open Access

    Data-driven approach to elucidate the correlation between photocatalytic activity and rate constants from excited states

    by Ryuga Kunisada, Manami Hayashi, Tabea Rohlfs, Taiki Nagano, Koki Sano, Naoto Inai, Naoki Noto, Takuya Ogaki, Yasunori Matsui, Hiroshi Ikeda, et al.

    Even though excited-state properties play a crucial role in photocatalysis, directly correlating these with photocatalytic activity remains challenging. Herein, we propose a method to elucidate the correlations between the catalytic activity of organic photosensitizers and the rate constants of various excited-state processes through integrating machine learning (ML), quantum chemical calculations, and chemical experiments. This approach was applied to interpolative predictions of the yield of the nickel/photocatalytic formation of C–O bonds and radical additions to alkenes using various organic photosensitizers with satisfactory accuracy (R2 = 0.83 and 0.77 on the test set, respectively). The calculated rate constants obtained through quantum chemical calculations proved to be comparable or even superior to the experimentally measured excited-state lifetimes as descriptors. SHAP-based visual analysis revealed that the rate constants corresponding to transitions from the T1 state provide significant contributions to the interpolative prediction of photocatalytic activity. Additionally, the non-radiative decay process between the S1 and S0 states helps describe the low catalytic activity of poorly emissive photosensitizers. These findings highlight the potential of the proposed method to provide insights into photocatalytic properties that are difficult to obtain using conventional approaches.


  • Open Access

    Data-driven approach to visualize the correlation between photocatalytic activity and rate constants from excited states

    by Ryuga Kunisada, Manami Hayashi, Tabea Rohlfs, Taiki Nagano, Koki Sano, Naoto Inai, Naoki Noto, Takuya Ogaki, Yasunori Matusi, Hiroshi Ikeda, et al.

    Even though excited-state properties play a crucial role in photocatalysis, directly correlating these with photocatalytic activity remains challenging. Herein, we propose a method to elucidate the correlations between the catalytic activity of organic photosensitizers and their rate constants of various excited-state processes through integrating machine learning (ML), quantum chemical calculations, and chemical experiments. This approach was applied to interpolative predictions of the yield of the nickel/photocatalytic formation of C–O bonds and radical additions to alkenes using various organic photosensitizers with satisfactory accuracy (R2 = 0.83 and 0.77 on the test set, respectively). The calculated rate constants obtained through quantum chemical calculations proved to be comparable or even superior to the experimentally measured excited-state lifetimes as descriptors. SHAP-based visual analysis revealed that the rate constants corresponding to transitions from the T1 state provide significant contributions to the interpolative prediction of photocatalytic activity. Additionally, the non-radiative decay process between the S1 and S0 states helps describe the low catalytic activity of poorly emissive photosensitizers. These findings highlight the potential of the proposed method to provide insights into photocatalytic properties that are difficult to obtain using conventional approaches.

  • Open Access

    Analytical nuclear gradients for second-order Møller-Plesset perturbation theory using pair-natural orbitals based on localized virtual molecular orbitals

    by Manami Hayashi, Masaaki Saitow, Takeshi Yanai

    Accelerating analytical nuclear energy gradient calculations within wavefunction-based quantum chemical frameworks and extending their applicability to large systems remain significant challenges. Recently, Pinski and Neese developed fully analytical nuclear energy gradients for second-order Møller-Plesset perturbation theory (MP2) using pair natural orbitals (PNOs) based on projected atomic orbitals (PAOs) [J. Chem. Phys. 148, 031101 (2018); J. Chem. Phys. 150, 164102 (2019)]. Here, we present an analytical nuclear gradient method based on the PNO-MP2 theory using orthonormal and non-redundant localized virtual molecular orbitals (LVMOs) instead of conventional PAOs. Compared with PAO-based approaches, our method exhibits two main differences. First, constraints from the construction of LVMOs are incorporated into the Lagrangian, requiring the solution of an additional Z-vector equation---the coupled-perturbed virtual localization (CP-VL) equation. Second, owing to the orthonormality of the LVMOs, the derivatives of the PNO coefficients with respect to both the MO coefficients and the nuclear coordinates vanish. The latter property significantly simplifies the formulation of analytical energy gradients. Additionally, to mitigate the technical complexity of deriving gradient formulas with the PNO treatment, we used an extended scheme implemented in our automatic derivation program. Benchmark calculations on phenylalkane chains, docetaxel, and a host–guest complex demonstrated high computational scalability of our nuclear gradient implementations, additionally confirming that the computational effort required to solve the CP-VL equation is relatively minor. We illustrated that the energy gradients as functions of nuclear coordinates are smooth and that the errors in the predicted structures are small.

  • Open Access

    Transfer learning across different photocatalytic organic reactions

    by Naoki Noto, Ryuga Kunisada, Tabea Rohlfs, Manami Hayashi, Ryosuke Kojima, Olga García Mancheño, Takeshi Yanai, Susumu Saito

    While seasoned organic chemists can often predict suitable catalysts for new reactions based on their past experiences in other catalytic reactions, developing this ability is costly, laborious and time-consuming. Therefore, replicating this remarkable expertize of human researchers through machine learning (ML) is compelling, albeit that it remains highly challenging. Herein, we apply a domain-adaptation-based transfer-learning (TL) approach to photocatalysis. Despite being different reaction types, the knowledge of the catalytic behavior of organic photosensitizers (OPSs) from photocatalytic cross-coupling reactions is successfully transferred to ML for a [2+2] cycloaddition reaction, improving the prediction of the photocatalytic activity compared with conventional ML approaches. Furthermore, a satisfactory predictive performance is achieved by using only ten training data points. This experimentally readily accessible small dataset can also be used to identify effective OPSs for alkene photoisomerization, thereby showcasing the potential benefits of TL in catalyst exploration.