Dunyou Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Dunyou Wang has a total of 174 co-authors. Here are the top co-authors by number of shared publications:
Delu Gao
College of Physics and Electronics
12 shared publicationsPeng Liu
Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China
7 shared publicationsChen Li
College of Physics and Electronics
6 shared publicationsRuoqi Zhang
Shandong Normal University
6 shared publicationsYuping Wang
College of Physics and Electronics, Shandong Normal University, Jinan, People’s Republic of China
6 shared publicationsJingxue Zhang
School of Physics, Nankai University, Tianjin, China
5 shared publications
Latest Publications
- 30 July 2026
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 17 (32) , 9408-9415
Beyond d -Band Models: Unoccupied Valence Band Ratio Governs CO Oxidation on BiCuSeO-Supported Single-Atom Catalysts
Developing reliable descriptors that correlate catalyst electronic structure with performance is crucial for designing efficient single-atom catalysts (SACs). Here, we report a systematic first-principles study of CO oxidation on single transition-metal atoms (Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt) anchored on 2D BiCuSeO. These catalysts primarily follow the Eley–Rideal mechanism, exhibiting remarkably low rate-limiting barriers ranging from 0.19 to 0.39 eV. We propose a comprehensive, orbital-resolved descriptor, the unoccupied valence band ratio, which integrates both s and d state contributions and serves as a superior descriptor for catalytic activity (R2 = 0.90), outperforming conventional d-band center and unoccupied d-band ratio models. Electronic structure analyses clarify that metal s orbitals play a critical, often-overlooked role in synergistic electron donation/back-donation during adsorbate activation. These findings position BiCuSeO as a promising SAC support for CO oxidation, underscoring the importance of leveraging non-d valence states for rational SAC design.
- 30 July 2026
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 17 (32) , 9408-9415
Beyond d -Band Models: Unoccupied Valence Band Ratio Governs CO Oxidation on BiCuSeO-Supported Single-Atom Catalysts
Developing reliable descriptors that correlate catalyst electronic structure with performance is crucial for designing efficient single-atom catalysts (SACs). Here, we report a systematic first-principles study of CO oxidation on single transition-metal atoms (Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt) anchored on 2D BiCuSeO. These catalysts primarily follow the Eley–Rideal mechanism, exhibiting remarkably low rate-limiting barriers ranging from 0.19 to 0.39 eV. We propose a comprehensive, orbital-resolved descriptor, the unoccupied valence band ratio, which integrates both s and d state contributions and serves as a superior descriptor for catalytic activity (R2 = 0.90), outperforming conventional d-band center and unoccupied d-band ratio models. Electronic structure analyses clarify that metal s orbitals play a critical, often-overlooked role in synergistic electron donation/back-donation during adsorbate activation. These findings position BiCuSeO as a promising SAC support for CO oxidation, underscoring the importance of leveraging non-d valence states for rational SAC design.
- 20 November 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 129 (48) , 11127-11135
Mechanistic Insights and Solvent Effects for the Ambident SCN – + CH 3 Cl Reaction in Aqueous Solution: A Multilevel QM/MM Study
The ambident SN2 reaction of SCN- + CH3Cl can proceed via either S-side or N-side attack, leading to distinct product formations. While previous studies have explored this reaction in the gas phase, the mechanistic details and solvent effects in aqueous solution remain unresolved. We employ a multilevel quantum mechanics/molecular mechanics approach with explicit water modeling to investigate the atomic-level ambident mechanisms, potentials of mean force, and solvent contributions for both S-side and N-side attack pathways. Our calculations reveal distinct structural and charge transfer dynamics between the S-side attack and N-side attack pathways, with the S-side transition state resembling the reactant complex more closely than the N-side transition state. Charge analysis indicates more efficient charge transfer in the S-side mechanism, correlating with its lower activation barrier. The CCSD(T) potentials of mean force yield free energy barriers of 21.4 kcal/mol for the S-side and 26.3 kcal/mol for the N-side attack, confirming the S-side preference persists in solution. Solvent effects raise the barriers by ∼31% for the S-side and ∼34% for the N-side compared to gas-phase values, with polarization and solvent energy contributions destabilizing the N-side transition state more significantly. These findings provide atomic-level insights into ambident reactivity in solution and highlight the critical role of solvent in modulating SN2 reaction energetics and mechanisms.
- 20 November 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 129 (48) , 11127-11135
Mechanistic Insights and Solvent Effects for the Ambident SCN – + CH 3 Cl Reaction in Aqueous Solution: A Multilevel QM/MM Study
The ambident SN2 reaction of SCN- + CH3Cl can proceed via either S-side or N-side attack, leading to distinct product formations. While previous studies have explored this reaction in the gas phase, the mechanistic details and solvent effects in aqueous solution remain unresolved. We employ a multilevel quantum mechanics/molecular mechanics approach with explicit water modeling to investigate the atomic-level ambident mechanisms, potentials of mean force, and solvent contributions for both S-side and N-side attack pathways. Our calculations reveal distinct structural and charge transfer dynamics between the S-side attack and N-side attack pathways, with the S-side transition state resembling the reactant complex more closely than the N-side transition state. Charge analysis indicates more efficient charge transfer in the S-side mechanism, correlating with its lower activation barrier. The CCSD(T) potentials of mean force yield free energy barriers of 21.4 kcal/mol for the S-side and 26.3 kcal/mol for the N-side attack, confirming the S-side preference persists in solution. Solvent effects raise the barriers by ∼31% for the S-side and ∼34% for the N-side compared to gas-phase values, with polarization and solvent energy contributions destabilizing the N-side transition state more significantly. These findings provide atomic-level insights into ambident reactivity in solution and highlight the critical role of solvent in modulating SN2 reaction energetics and mechanisms.
- 13 October 2025
- Published by Taylor & Francis in Molecular Physics
- Vol. 124 (11)
Reduced-dimensional quantum dynamic study of the F – +CH3I→FCH3+I – system
- 13 October 2025
- Published by Taylor & Francis in Molecular Physics
- Vol. 124 (11)
Reduced-dimensional quantum dynamic study of the F – +CH3I→FCH3+I – system
- 29 July 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 16 (31) , 7937-7943
Ultralow-Barrier CO Oxidation on Bi2O2S-Supported Single-Atom Catalysts: Mechanistic Insights and Electronic Descriptors
Efficient CO oxidation is critical for environmental catalysis. Here, we investigate CO oxidation on single transition metal atoms (TM1 = Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt) supported by Bi2O2S using density functional theory and ab initio molecular dynamics. The reaction proceeds exclusively via the Eley-Rideal mechanism, except for Pd, with activation barriers ranging from 3.6 to 43.7 kJ/mol. Group 9 (Co, Rh, and Ir) and 10 (Ni and Pt) metals demonstrate superior performance, exhibiting ultralow barriers below 10 kJ/mol. We identify two electronic descriptors: a static descriptor quantifying pre-transition-state O-TM1 bond stability, which is inversely correlated with activation barriers, and a dynamic descriptor tracking the shift of the occupied orbital center from the pre-transition-state intermediates to transition states, directly linked to barrier heights. Bi2O2S-supported single-atom catalysts enable ultralow-barrier CO oxidation, indicating great potential for environmental applications and providing design principles for efficient single-atom catalysts for CO oxidation.
- 29 July 2025
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 16 (31) , 7937-7943
Ultralow-Barrier CO Oxidation on Bi2O2S-Supported Single-Atom Catalysts: Mechanistic Insights and Electronic Descriptors
Efficient CO oxidation is critical for environmental catalysis. Here, we investigate CO oxidation on single transition metal atoms (TM1 = Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt) supported by Bi2O2S using density functional theory and ab initio molecular dynamics. The reaction proceeds exclusively via the Eley-Rideal mechanism, except for Pd, with activation barriers ranging from 3.6 to 43.7 kJ/mol. Group 9 (Co, Rh, and Ir) and 10 (Ni and Pt) metals demonstrate superior performance, exhibiting ultralow barriers below 10 kJ/mol. We identify two electronic descriptors: a static descriptor quantifying pre-transition-state O-TM1 bond stability, which is inversely correlated with activation barriers, and a dynamic descriptor tracking the shift of the occupied orbital center from the pre-transition-state intermediates to transition states, directly linked to barrier heights. Bi2O2S-supported single-atom catalysts enable ultralow-barrier CO oxidation, indicating great potential for environmental applications and providing design principles for efficient single-atom catalysts for CO oxidation.
- 21 July 2025
- Published by American Chemical Society (ACS) in Langmuir
CO Oxidation on Single Transition Metal Atoms Supported on 2D ZnO: Role of Antibonding-State Center Shifts in Catalytic Efficiency
Catalytic oxidation is widely regarded as an effective method of eliminating CO pollutants. Given that single transition metal atoms supported on 2D ZnO monolayers have been successfully synthesized experimentally, we herein conduct a systematic study of CO oxidation on single transition metal atoms supported on 2D ZnO (TM1/gZnO, TM = Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt) using density functional theory. Our findings show that within the same period, a larger empty d-band ratio in a single transition metal atom enhances its adsorption of gas molecules on TM1/gZnO. The common mechanisms for CO oxidation on TM1/gZnO are the Langmuir-Hinshelwood and Eley-Rideal pathways, with Pt1/gZnO emerging as the optimal single-atom catalyst due to its lowest rate-limiting adiabatic barriers in both mechanisms. We discover that the activation barrier height correlates closely with the upward shift of the antibonding-state center in the transition state relative to its preceding adsorption precursor; specifically, a greater upshift corresponds to a lower barrier. These insights deepen the understanding of CO oxidation mechanisms on supported single-atom catalysts and provide a predictive framework for designing efficient catalysts for CO oxidation.
- 21 July 2025
- Published by American Chemical Society (ACS) in Langmuir
CO Oxidation on Single Transition Metal Atoms Supported on 2D ZnO: Role of Antibonding-State Center Shifts in Catalytic Efficiency
Catalytic oxidation is widely regarded as an effective method of eliminating CO pollutants. Given that single transition metal atoms supported on 2D ZnO monolayers have been successfully synthesized experimentally, we herein conduct a systematic study of CO oxidation on single transition metal atoms supported on 2D ZnO (TM1/gZnO, TM = Fe, Co, Ni, Ru, Rh, Pd, Os, Ir, and Pt) using density functional theory. Our findings show that within the same period, a larger empty d-band ratio in a single transition metal atom enhances its adsorption of gas molecules on TM1/gZnO. The common mechanisms for CO oxidation on TM1/gZnO are the Langmuir-Hinshelwood and Eley-Rideal pathways, with Pt1/gZnO emerging as the optimal single-atom catalyst due to its lowest rate-limiting adiabatic barriers in both mechanisms. We discover that the activation barrier height correlates closely with the upward shift of the antibonding-state center in the transition state relative to its preceding adsorption precursor; specifically, a greater upshift corresponds to a lower barrier. These insights deepen the understanding of CO oxidation mechanisms on supported single-atom catalysts and provide a predictive framework for designing efficient catalysts for CO oxidation.
- 10 June 2025
- Published by Elsevier in Surface Science
- Vol. 761
H2 dissociation barrier governed by antibonding-state center in defective graphene-supported Cu19 cluster
- 10 June 2025
- Published by Elsevier in Surface Science
- Vol. 761
H2 dissociation barrier governed by antibonding-state center in defective graphene-supported Cu19 cluster
- 9 December 2024
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 128 (50) , 10766-10774
Potentials of Mean Force and Solvent Effects of the CN– + CH3X (X = F, Cl, Br, and I) Reactions by the N-Side Attack in Aqueous Solution
A combined multilevel quantum mechanics and molecular mechanics approach is performed to investigate the nucleophilic substitution reactions of CN- + CH3X (X = F, Cl, Br, and I) by the N-side attack in aqueous solution. The water molecules are treated microscopically using an explicit SPC/E model, and the potentials of mean force are characterized by both the DFT and CCSD(T) levels of theory for the solute. Calculations demonstrate that the shielding effect of the solvent reduces the nucleophile-substrate and substrate-leaving group interactions in solution, leading to stationary point structures that are quite different from those in the gas phase. The structure and charge evolution along the reaction paths reveal that the reaction is not only a synchronous bonding and bond-breaking Walden-inversion mechanism but also a synchronous charge transfer process. The activation barriers calculated at the CCSD(T) level of theory are 27.5 (F), 22.6 (Cl), 21.7 (Br), and 21.2 (I) kcal/mol, respectively, which are larger than the corresponding experimental values for the C-side attack. The polarization effect of water molecules causing solute polarization contributes to the activation barrier in the order of F > Cl > Br > I. The solvent energy contribution to the activation barrier is in the order of F < Cl < Br < I because the F leaving group has the most compact transition state structure and the I leaving group has the loosest transition state structure. As a result, the total contributions of the solvent effects to the activation barriers are 7.9 (F), 10.7 (Cl), 15.3 (Br), and 15.7 (I) kcal/mol. Our results show that the solvent effects have a significant influence on both the structure and the energetics of the N-side attack reactions in aqueous solution.
- 9 December 2024
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 128 (50) , 10766-10774
Potentials of Mean Force and Solvent Effects of the CN– + CH3X (X = F, Cl, Br, and I) Reactions by the N-Side Attack in Aqueous Solution
A combined multilevel quantum mechanics and molecular mechanics approach is performed to investigate the nucleophilic substitution reactions of CN- + CH3X (X = F, Cl, Br, and I) by the N-side attack in aqueous solution. The water molecules are treated microscopically using an explicit SPC/E model, and the potentials of mean force are characterized by both the DFT and CCSD(T) levels of theory for the solute. Calculations demonstrate that the shielding effect of the solvent reduces the nucleophile-substrate and substrate-leaving group interactions in solution, leading to stationary point structures that are quite different from those in the gas phase. The structure and charge evolution along the reaction paths reveal that the reaction is not only a synchronous bonding and bond-breaking Walden-inversion mechanism but also a synchronous charge transfer process. The activation barriers calculated at the CCSD(T) level of theory are 27.5 (F), 22.6 (Cl), 21.7 (Br), and 21.2 (I) kcal/mol, respectively, which are larger than the corresponding experimental values for the C-side attack. The polarization effect of water molecules causing solute polarization contributes to the activation barrier in the order of F > Cl > Br > I. The solvent energy contribution to the activation barrier is in the order of F < Cl < Br < I because the F leaving group has the most compact transition state structure and the I leaving group has the loosest transition state structure. As a result, the total contributions of the solvent effects to the activation barriers are 7.9 (F), 10.7 (Cl), 15.3 (Br), and 15.7 (I) kcal/mol. Our results show that the solvent effects have a significant influence on both the structure and the energetics of the N-side attack reactions in aqueous solution.
- 2 October 2024
- Published by Elsevier in Molecular Catalysis
- Vol. 569
CO oxidation on single Pt atom supported by two-dimensional ZnO monolayer: Reaction mechanism and precursor activation
- 2 October 2024
- Published by Elsevier in Molecular Catalysis
- Vol. 569
CO oxidation on single Pt atom supported by two-dimensional ZnO monolayer: Reaction mechanism and precursor activation
- 13 February 2024
- Published by Wiley in Chemphyschem
- Vol. 25 (6) , e202300861
Theoretical study of NH3, H2S, and HCN adsorption enhancement on defective graphene‐supported Cu19 clusters
Recent studies have shown that graphene‐supported metal clusters can enhance catalytic reactivity compared with corresponding metal clusters. In this study, the adsorptions of NH3, H2S, and HCN on Cu19 and defective graphene‐supported Cu19 clusters are investigated using plane‐wave density functional theory. The results reveal the three gas molecules can be adsorbed on three types of top sites of Cu atoms, respectively. The adsorption energies of the corresponding adsorption sites on the defective graphene‐supported Cu19 clusters are all increased compared with those on the Cu19 clusters. The orbital‐resolved, crystal orbital Hamilton population analysis demonstrates that the larger the integrated crystal orbital Hamilton population, the stronger the adsorption between the gas molecule and the bonded Cu atom. The center of antibonding states on the defective graphene‐supported Cu19 is shifted upward relative to Fermi level compared to the corresponding one on pure Cu19, which explains the enhanced adsorption energy on defective graphene‐supported Cu19. In addition, the closer d‐band center to the Fermi level on the defective graphene‐supported Cu19 indicates a stronger adsorption capacity than on pure Cu19.
- 13 February 2024
- Published by Wiley in Chemphyschem
- Vol. 25 (6) , e202300861
Theoretical study of NH3, H2S, and HCN adsorption enhancement on defective graphene‐supported Cu19 clusters
Recent studies have shown that graphene‐supported metal clusters can enhance catalytic reactivity compared with corresponding metal clusters. In this study, the adsorptions of NH3, H2S, and HCN on Cu19 and defective graphene‐supported Cu19 clusters are investigated using plane‐wave density functional theory. The results reveal the three gas molecules can be adsorbed on three types of top sites of Cu atoms, respectively. The adsorption energies of the corresponding adsorption sites on the defective graphene‐supported Cu19 clusters are all increased compared with those on the Cu19 clusters. The orbital‐resolved, crystal orbital Hamilton population analysis demonstrates that the larger the integrated crystal orbital Hamilton population, the stronger the adsorption between the gas molecule and the bonded Cu atom. The center of antibonding states on the defective graphene‐supported Cu19 is shifted upward relative to Fermi level compared to the corresponding one on pure Cu19, which explains the enhanced adsorption energy on defective graphene‐supported Cu19. In addition, the closer d‐band center to the Fermi level on the defective graphene‐supported Cu19 indicates a stronger adsorption capacity than on pure Cu19.
- 1 December 2023
- Published by AIP Publishing in Chinese Journal of Chemical Physics
- Vol. 36 (6) , 747-754
Ab initiomolecular dynamics study of dissociative adsorption of H2 on defective graphene-supported Cu19 cluster
The dissociative adsorption of H2 on Cu19 and defective graphene-supported Cu19 clusters (Cu19G) are investigated using ab initio molecular dynamics. The molecular-level trajectories show that, on Cu19, the preferred adsorption site is the bridge-hollow site, where the two H atoms are adsorbed at the bridge and hollow sites beside a Cu atom, with an adsorption energy of −0.74 eV. In contrast, on the defective graphene-supported Cu19 cluster, the favorite adsorption site is located where the two H atoms are adsorbed at hollow-hollow sites with an adsorption energy of −1.27 eV. In general, the average adsorption energy on the defective graphene-supported Cu19 cluster is −1.07 eV, which is about 84% larger than that of −0.58 eV on the Cu19 cluster. This indicates that the adsorption capacity is greatly enhanced for the dissociative adsorption of H2 on the defective graphene-supported Cu19 cluster. The d-band center shifts to the Fermi level, illustrating the enhanced adsorption capacity on the defective graphene-supported Cu19 cluster. The integrated crystal orbital Hamilton population analysis reveals that stronger bond interactions between hydrogen atoms with their bonded Cu atoms lead to much larger adsorption energies on the defective graphene-supported Cu19 cluster compared to the Cu19 cluster.
- 1 December 2023
- Published by AIP Publishing in Chinese Journal of Chemical Physics
- Vol. 36 (6) , 747-754
Ab initiomolecular dynamics study of dissociative adsorption of H2 on defective graphene-supported Cu19 cluster
The dissociative adsorption of H2 on Cu19 and defective graphene-supported Cu19 clusters (Cu19G) are investigated using ab initio molecular dynamics. The molecular-level trajectories show that, on Cu19, the preferred adsorption site is the bridge-hollow site, where the two H atoms are adsorbed at the bridge and hollow sites beside a Cu atom, with an adsorption energy of −0.74 eV. In contrast, on the defective graphene-supported Cu19 cluster, the favorite adsorption site is located where the two H atoms are adsorbed at hollow-hollow sites with an adsorption energy of −1.27 eV. In general, the average adsorption energy on the defective graphene-supported Cu19 cluster is −1.07 eV, which is about 84% larger than that of −0.58 eV on the Cu19 cluster. This indicates that the adsorption capacity is greatly enhanced for the dissociative adsorption of H2 on the defective graphene-supported Cu19 cluster. The d-band center shifts to the Fermi level, illustrating the enhanced adsorption capacity on the defective graphene-supported Cu19 cluster. The integrated crystal orbital Hamilton population analysis reveals that stronger bond interactions between hydrogen atoms with their bonded Cu atoms lead to much larger adsorption energies on the defective graphene-supported Cu19 cluster compared to the Cu19 cluster.