Guan Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Guan Zhang has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
Jin Li
Technology Innovation Center for Land Engineering and Human Settlements, School of Human Settlement and Civil Engineering Xi'an Jiaotong University Xi'an Shaanxi China
4 shared publications- Zongkuan Liu
Technology Innovation Center for Land Engineering and Human Settlements, Shaanxi Land Engineering Construction Group Co., Ltd., School of Human Settlements and Civil Engineering, Xi’an Jiaotong University, Xi’an 710049, China
4 shared publications - Quanhua Sun
State Key Laboratory of High Temperature Gas Dynamics, Institute of Mechanics, Chinese Academy of Sciences 1 , 100190 Beijing
3 shared publications - Yonghao Zhang
Center for Interdisciplinary Research in Fluids, Institute of Mechanics, Chinese Academy of Sciences 3 , Beijing 100190
3 shared publications - Yuan Hu
State Key Laboratory of High Temperature Gas Dynamics, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, China
3 shared publications - Shiwei Hu
State Key Laboratory of High Temperature Gas Dynamics, Institute of Mechanics, Chinese Academy of Sciences , Beijing 100190
2 shared publications
Latest Publications
- 2 November 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 165 (17)
Erratum: “Deep potential modeling of oxygen adsorption and surface reconstruction on silica” [J. Chem. Phys. 165, 114702 (2026)]
- 2 November 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 165 (17)
Erratum: “Deep potential modeling of oxygen adsorption and surface reconstruction on silica” [J. Chem. Phys. 165, 114702 (2026)]
- 15 September 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 165 (11)
Deep potential modeling of oxygen adsorption and surface reconstruction on silica
The binding energy of atomic oxygen on silica has remained uncertain for decades. This quantity directly governs catalytic heating during atmospheric re-entry. Classical potentials overestimate it by an order of magnitude (ReaxFFSiOGSI: ∼31 eV vs experiment: ∼3 eV), while first-principles methods cannot reach the scales required for adsorption equilibration. Although neural network potentials bridge this accuracy–efficiency gap, conventional sampling fails to provide training data spanning rare reactive events within collective phase transitions. Here, we train a neural-network potential (HyGSI) using a hybrid enhanced sampling-machine learning framework coupling on-the-fly probability enhanced sampling with deep potential molecular dynamics. HyGSI covers SiO2 melting, pressure-dependent surface reconstruction, and oxygen adsorption–recombination. The predicted adsorption energy is 2.96 ± 0.15 eV, consistent with the 1991 experimental estimate (3.00 ± 0.31 eV) and an order of magnitude lower than ReaxFFSiOGSI predictions. Furthermore, surface reconstruction reshapes available catalytic sites, coupling structural evolution to reaction chemistry. These findings provide corrected energetics for hypersonic re-entry heat-flux models and validated force fields for semiconductor plasma etching and fusion reactor wall modeling.
- 15 September 2026
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 165 (11)
Deep potential modeling of oxygen adsorption and surface reconstruction on silica
The binding energy of atomic oxygen on silica has remained uncertain for decades. This quantity directly governs catalytic heating during atmospheric re-entry. Classical potentials overestimate it by an order of magnitude (ReaxFFSiOGSI: ∼31 eV vs experiment: ∼3 eV), while first-principles methods cannot reach the scales required for adsorption equilibration. Although neural network potentials bridge this accuracy–efficiency gap, conventional sampling fails to provide training data spanning rare reactive events within collective phase transitions. Here, we train a neural-network potential (HyGSI) using a hybrid enhanced sampling-machine learning framework coupling on-the-fly probability enhanced sampling with deep potential molecular dynamics. HyGSI covers SiO2 melting, pressure-dependent surface reconstruction, and oxygen adsorption–recombination. The predicted adsorption energy is 2.96 ± 0.15 eV, consistent with the 1991 experimental estimate (3.00 ± 0.31 eV) and an order of magnitude lower than ReaxFFSiOGSI predictions. Furthermore, surface reconstruction reshapes available catalytic sites, coupling structural evolution to reaction chemistry. These findings provide corrected energetics for hypersonic re-entry heat-flux models and validated force fields for semiconductor plasma etching and fusion reactor wall modeling.
- 1 August 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (8)
Chemical energy accommodation of atomic oxygen recombination on silica surfaces under high-temperature nonequilibrium flow environments: Insights from atomistic-scale simulations
The energy transfer and accommodation for the catalytic recombination of atomic oxygen (O) on silica surfaces, a key process to be understood for the accurate prediction of aerothermal heating of hypersonic vehicles, are studied using a combination of the density functional theory (DFT) and reactive molecular dynamics calculations. The key elementary reactions are determined by the DFT calculations, along with the barriers and changes of free energy for each reaction. The energy carried by the recombined oxygen molecules (O2) and its partition into different internal modes for various reactions are obtained by averaging the corresponding molecular information from a significant number of molecular dynamics trajectories. The chemical energy accommodation (CEA) coefficients, for various reactions, internal modes, and surface structures, are then computed based on the free energy changes and energies carried by the recombined O2. Moreover, the detailed energy distributions of O2 are also provided. It is found that CEA for the recombination of O on the silica surfaces depends greatly on the reaction type and internal energy mode but is less profoundly influenced by the surface structures. The results of the present study offer better insights into the mechanisms of chemical energy transfer and accommodation for the catalytic recombination of O on silica surfaces and can help improve the modeling of the relevant gas–surface interactions for more reliable aerothermal heating prediction.
- 1 August 2025
- Published by AIP Publishing in Physics of Fluids
- Vol. 37 (8)
Chemical energy accommodation of atomic oxygen recombination on silica surfaces under high-temperature nonequilibrium flow environments: Insights from atomistic-scale simulations
The energy transfer and accommodation for the catalytic recombination of atomic oxygen (O) on silica surfaces, a key process to be understood for the accurate prediction of aerothermal heating of hypersonic vehicles, are studied using a combination of the density functional theory (DFT) and reactive molecular dynamics calculations. The key elementary reactions are determined by the DFT calculations, along with the barriers and changes of free energy for each reaction. The energy carried by the recombined oxygen molecules (O2) and its partition into different internal modes for various reactions are obtained by averaging the corresponding molecular information from a significant number of molecular dynamics trajectories. The chemical energy accommodation (CEA) coefficients, for various reactions, internal modes, and surface structures, are then computed based on the free energy changes and energies carried by the recombined O2. Moreover, the detailed energy distributions of O2 are also provided. It is found that CEA for the recombination of O on the silica surfaces depends greatly on the reaction type and internal energy mode but is less profoundly influenced by the surface structures. The results of the present study offer better insights into the mechanisms of chemical energy transfer and accommodation for the catalytic recombination of O on silica surfaces and can help improve the modeling of the relevant gas–surface interactions for more reliable aerothermal heating prediction.
- 14 August 2023
- Published by Wiley in Journal of Physical Organic Chemistry
- Vol. 36 (11)
Research on the reaction mechanism of 2,4,6‐trinitrotoluene resource utilization I: Formation of 2,4,6‐trinitrobenzoic acid in acetic acid with N,N′,N″‐trihydroxyisocyanuric acid catalyst
As an organic molecule catalyst, N,N′,N″‐trihydroxyisocyanuric acid can selectively catalyze the oxidation of the methyl group of waste 2,4,6‐trinitrotoluene to generate 2,4,6‐trinitrobenzoic acid. This reaction can avoid environmental pollution by inorganic heavy metal catalysts. In this study, four reaction stages of this catalytic reaction were designed and validated computationally at the M06‐2X‐D3ZERO/6‐311G(d,p) level using the acetic acid solvent model. These validations include transition state searches, intrinsic reaction coordinate calculations, reactant and product optimizations, and frequency calculations. The final reaction network of 23 transition states shows that after N,N′,N″‐trihydroxyisocyanuric acid activation and common reaction, the network bifurcates into two stages: alcohol to carboxylic acid and aldehyde to carboxylic acid. Although the former stage releases about 155 kcal/mol of Gibbs free energy, less than the 177 kcal/mol from the latter stage, the overall reaction equation shows that the pathway including former stage does not consume the catalytically active substance IM_T2, which saves the energy required for reactivation and is thus more favorable. Furthermore, the key transition states in the reaction network include bimolecular substitution reactions and proton‐hopping transfer reactions. Analyses of their interaction region indicators and intrinsic reaction coordinate results demonstrate strong selectivity. Additionally, the energy barriers and heat releases of the latter are twice and 1.3 times greater than those of the former, respectively. In summary, this study elucidated two competitive reaction pathways and identified the more energetically favorable and selective pathway, and it provides useful insights for further optimization of industrial utilization of 2,4,6‐trinitrotoluene.
- 14 August 2023
- Published by Wiley in Journal of Physical Organic Chemistry
- Vol. 36 (11)
Research on the reaction mechanism of 2,4,6‐trinitrotoluene resource utilization I: Formation of 2,4,6‐trinitrobenzoic acid in acetic acid with N,N′,N″‐trihydroxyisocyanuric acid catalyst
As an organic molecule catalyst, N,N′,N″‐trihydroxyisocyanuric acid can selectively catalyze the oxidation of the methyl group of waste 2,4,6‐trinitrotoluene to generate 2,4,6‐trinitrobenzoic acid. This reaction can avoid environmental pollution by inorganic heavy metal catalysts. In this study, four reaction stages of this catalytic reaction were designed and validated computationally at the M06‐2X‐D3ZERO/6‐311G(d,p) level using the acetic acid solvent model. These validations include transition state searches, intrinsic reaction coordinate calculations, reactant and product optimizations, and frequency calculations. The final reaction network of 23 transition states shows that after N,N′,N″‐trihydroxyisocyanuric acid activation and common reaction, the network bifurcates into two stages: alcohol to carboxylic acid and aldehyde to carboxylic acid. Although the former stage releases about 155 kcal/mol of Gibbs free energy, less than the 177 kcal/mol from the latter stage, the overall reaction equation shows that the pathway including former stage does not consume the catalytically active substance IM_T2, which saves the energy required for reactivation and is thus more favorable. Furthermore, the key transition states in the reaction network include bimolecular substitution reactions and proton‐hopping transfer reactions. Analyses of their interaction region indicators and intrinsic reaction coordinate results demonstrate strong selectivity. Additionally, the energy barriers and heat releases of the latter are twice and 1.3 times greater than those of the former, respectively. In summary, this study elucidated two competitive reaction pathways and identified the more energetically favorable and selective pathway, and it provides useful insights for further optimization of industrial utilization of 2,4,6‐trinitrotoluene.
- 13 February 2023
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 127 (7) , 1609-1618
Heuristic-Based Alkaline Hydrolysis Mechanism of Nitrate Ester (Nitrocellulose Monomer) and Nitroamine (Hexogen) Compounds: Electrostatic Attraction Effect of the Nitro Group
The alkaline hydrolysis reaction of energetic materials is important and complex. With improved performance, AMK_Mountain was used to systematically study the alkaline hydrolysis of the nitrocellulose monomer and hexogen. The reaction pathways showed that the nitrocellulose monomer produces the nitrate anion and nitrite anion differently, while hexogen only produces the nitrite anion. Electronic structure results at the M06-2X/6-311G(d,p)/PCM(Pauling) level showed that the nitrocellulose monomer and hexogen have a similar pathway in their main energy-releasing process (nitrite anion production): with electrostatic attraction effects after proton transfer, the nitrite anion dissociates from the original structure with a low barrier. Moreover, during the alkaline hydrolysis of the nitrocellulose monomer, the metastable intermediates after proton transfer may be directly generated following transition states that, structurally, tend to produce nitrite anions "proximal" to the proton transfer site and produce nitrate anions "distal" to the proton transfer site. Electronic structure analysis showed that representative metastable intermediates revealed that the charge transfer caused by electrostatic attraction may be the direct cause of these reactions.
- 13 February 2023
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 127 (7) , 1609-1618
Heuristic-Based Alkaline Hydrolysis Mechanism of Nitrate Ester (Nitrocellulose Monomer) and Nitroamine (Hexogen) Compounds: Electrostatic Attraction Effect of the Nitro Group
The alkaline hydrolysis reaction of energetic materials is important and complex. With improved performance, AMK_Mountain was used to systematically study the alkaline hydrolysis of the nitrocellulose monomer and hexogen. The reaction pathways showed that the nitrocellulose monomer produces the nitrate anion and nitrite anion differently, while hexogen only produces the nitrite anion. Electronic structure results at the M06-2X/6-311G(d,p)/PCM(Pauling) level showed that the nitrocellulose monomer and hexogen have a similar pathway in their main energy-releasing process (nitrite anion production): with electrostatic attraction effects after proton transfer, the nitrite anion dissociates from the original structure with a low barrier. Moreover, during the alkaline hydrolysis of the nitrocellulose monomer, the metastable intermediates after proton transfer may be directly generated following transition states that, structurally, tend to produce nitrite anions "proximal" to the proton transfer site and produce nitrate anions "distal" to the proton transfer site. Electronic structure analysis showed that representative metastable intermediates revealed that the charge transfer caused by electrostatic attraction may be the direct cause of these reactions.
- 14 May 2022
- Published by Wiley in Journal of Computational Chemistry
- Vol. 43 (22) , 1513-1523
Automated reaction mechanisms and kinetics with the nudged elastic band method‐based AMK_Mountain and its description of the preliminary alkaline hydrolysis of nitrocellulose monomer
To improve the transition state (TS) search capability in complex chemical environments, AMK_Mountain is constructed based on the automated reaction mechanisms and kinetics (AutoMeKin) package. AMK_Mountain does not distinguish the reaction type of the TSs, which is beneficial to obtaining a more comprehensive reaction mechanism. In this study, the first step of the alkaline hydrolysis process of nitrocellulose monomer was adopted as the research object, and 730 possible initial configurations are constructed and 22 TSs pass high-level calculations. Energy difference and interaction region indicator reveal that the first step of alkaline hydrolysis is mainly the combination of nitrogen-containing functional groups at the positions alpha and beta with hydroxide anions, followed by the formation of nitric acid and the further loss of protons to form nitrate. Overall, in combination with GFN(2)-xTB and ORCA, the AMK_Mountain technique provides a reliable method for the location of the TSs in complex environments.
- 14 May 2022
- Published by Wiley in Journal of Computational Chemistry
- Vol. 43 (22) , 1513-1523
Automated reaction mechanisms and kinetics with the nudged elastic band method‐based AMK_Mountain and its description of the preliminary alkaline hydrolysis of nitrocellulose monomer
To improve the transition state (TS) search capability in complex chemical environments, AMK_Mountain is constructed based on the automated reaction mechanisms and kinetics (AutoMeKin) package. AMK_Mountain does not distinguish the reaction type of the TSs, which is beneficial to obtaining a more comprehensive reaction mechanism. In this study, the first step of the alkaline hydrolysis process of nitrocellulose monomer was adopted as the research object, and 730 possible initial configurations are constructed and 22 TSs pass high-level calculations. Energy difference and interaction region indicator reveal that the first step of alkaline hydrolysis is mainly the combination of nitrogen-containing functional groups at the positions alpha and beta with hydroxide anions, followed by the formation of nitric acid and the further loss of protons to form nitrate. Overall, in combination with GFN(2)-xTB and ORCA, the AMK_Mountain technique provides a reliable method for the location of the TSs in complex environments.
- 28 September 2021
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 23 (41) , 23673-23683
Automated reaction mechanisms and kinetics based transition state search process AMK-gau_xtb and its application to the substitution reaction of the nitroso group in 2,4,6-trinitrotoluene by hydroxide anion in the aqueous phase
The automated reaction mechanisms and kinetics (AutoMeKin) program evolved from a transition state search using chemical dynamics simulations (TSSCDS). It combines a series of empirical, semi-empirical and ab initio calculation methods to provide a two-step transition state search process from low-level calculation to high-level calculation. However, in this process, with the lack of solution keywords, low-level calculation has the problem of low accuracy or high computational cost. To address this problem, the gau_xtb interface that combines the high efficiency of xTB and the comprehensiveness of Gaussian09 was incorporated into the AutoMeKin2020 in this work and after adding some keywords, the AMK-gau_xtb software was obtained. Meanwhile, to adapt to the interface, the MD sampling results used Quadratic Synchronous Transit 3 (QST3) for the low-level transition state search. As an application, the reaction in which the nitroso group is replaced by hydroxide anion during the alkaline hydrolysis of 2,4,6-trinitrotoluene (TNT) in the water phase was studied with AMK-gau_xtb. The results of Intrinsic Reaction Coordinate (IRC) calculations revealed that the reactions on the front side and back side are different, with higher energy barriers obtained for the reactions on the front side. In addition, the hydrogen atom of the hydroxide anion has a slightly higher energy barrier for motion toward the inside of the benzene ring than for motion out of the benzene ring. Examination of the transition state structures of the low-level and high-level results showed that all reactions involve the stretching and restoration of the benzene ring. This process will lead to the incorrect identification of several transition states by the gau_xtb-based low-level calculation, while high-level calculation eliminates these incorrect results. The results of this research showed that AMK-gau_xtb has high efficiency and high fault tolerance, and has potential for use in large-scale system transition state searches.
- 28 September 2021
- Published by Royal Society of Chemistry (RSC) in Physical Chemistry Chemical Physics
- Vol. 23 (41) , 23673-23683
Automated reaction mechanisms and kinetics based transition state search process AMK-gau_xtb and its application to the substitution reaction of the nitroso group in 2,4,6-trinitrotoluene by hydroxide anion in the aqueous phase
The automated reaction mechanisms and kinetics (AutoMeKin) program evolved from a transition state search using chemical dynamics simulations (TSSCDS). It combines a series of empirical, semi-empirical and ab initio calculation methods to provide a two-step transition state search process from low-level calculation to high-level calculation. However, in this process, with the lack of solution keywords, low-level calculation has the problem of low accuracy or high computational cost. To address this problem, the gau_xtb interface that combines the high efficiency of xTB and the comprehensiveness of Gaussian09 was incorporated into the AutoMeKin2020 in this work and after adding some keywords, the AMK-gau_xtb software was obtained. Meanwhile, to adapt to the interface, the MD sampling results used Quadratic Synchronous Transit 3 (QST3) for the low-level transition state search. As an application, the reaction in which the nitroso group is replaced by hydroxide anion during the alkaline hydrolysis of 2,4,6-trinitrotoluene (TNT) in the water phase was studied with AMK-gau_xtb. The results of Intrinsic Reaction Coordinate (IRC) calculations revealed that the reactions on the front side and back side are different, with higher energy barriers obtained for the reactions on the front side. In addition, the hydrogen atom of the hydroxide anion has a slightly higher energy barrier for motion toward the inside of the benzene ring than for motion out of the benzene ring. Examination of the transition state structures of the low-level and high-level results showed that all reactions involve the stretching and restoration of the benzene ring. This process will lead to the incorrect identification of several transition states by the gau_xtb-based low-level calculation, while high-level calculation eliminates these incorrect results. The results of this research showed that AMK-gau_xtb has high efficiency and high fault tolerance, and has potential for use in large-scale system transition state searches.
- 23 July 2019
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry C
- Vol. 123 (32) , 19962-19969
Multiple Objective NSGA-II-Based Optimization Program and Its Application in Reactive Force Field for 2,4,6-Trinitrotoluene Diffusion in the Aqueous Phase
- 23 July 2019
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry C
- Vol. 123 (32) , 19962-19969
Multiple Objective NSGA-II-Based Optimization Program and Its Application in Reactive Force Field for 2,4,6-Trinitrotoluene Diffusion in the Aqueous Phase