Chaozheng He

While the electrocatalytic CO ₂ reduction reaction (CO ₂ RR) represents a cornerstone for sustainable energy conversion, the development of dual-efficient catalysts—balancing high activity with selectivity—presents a major ongoing challenge.

Utilizing the advanced crystal structure prediction algorithm CBD-GM, this study reports the first prediction of a novel 2D MoS ₃ configuration.

Isotope substitution affects lattice thermal conductivity by altering both the average atomic mass and the mass disorder, which influence phonon frequency and Tamura scattering, respectively. Previously, these two effects were combined into a single measure of thermal conductivity change and never separated. This study examines GaAs and GaN to cover a range of isotope effects from weak to strong. A comprehensive set of g-parameter isotope configurations was designed. By using the g=0 configuration to remove the mass scaling effect, the contribution from disorder scattering was isolated. Results were validated using the NEP-kappa iterative Boltzmann Transport Equation method and HNEMD molecular dynamics simulations. The results show that at room temperature, the total isotope effect in GaAs is 4.12%, with mass scaling contributing about 1.08% and disorder scattering about 3.58%, the latter being dominant. At 100 K, the total effect rises to 12.37%. For GaN, the total isotope effect at room temperature is 30.87%, nearly an order of magnitude higher than GaAs. The mass scaling effect in GaN strongly correlates with temperature, linked to its phonon bandgap structure. This study successfully separates the two mechanisms in both weak- and strong-effect systems, validating the universal "configuration design—mechanism separation—multi-method cross-validation" approach and providing quantitative support for isotope engineering in thermal management.

In this study, Zn-Co and Cu-Co bimetallic Co-doped MoS2 systems were constructed using density functional theory (DFT) to systematically investigate the activation mechanisms of the B-H and N-H bonds in AB molecules. The results show that undoped MoS2 can only cleave the N-H bond, with a B-H bond activation capability of merely 1.97%; Zn-Co@MoS2, despite forming redox dual centers, can only cleave the B-H bond and exhibits an N-H bond activation capability of 2.15%; whereas Cu-Co@MoS2 demonstrates optimal bifunctional performance, absorbing 1.369eV of energy to simultaneously cleave both B-H and N-H bonds, at an energy cost significantly lower than that of undoped MoS2. Charge analysis reveals that the Cu region acts as an oxidation center, accepting electrons released from the B-H bond, while the Co region functions as a reduction center, capturing and stabilizing the protons generated from N-H bond cleavage; their synergy enables the simultaneous cleavage of both bonds. Density of states analysis further confirms that the Cu/Co 3d orbitals hybridize with the B/N 2p and H s orbitals near the Fermi level, effectively weakening the bonding interactions within the B-H and N-H bonds. By constructing efficient dual active sites through Cu-Co co-doping, this work overcomes the limited activation capability of a single substrate and provides a theoretical basis for the design of bifunctional catalysts for hydrogen generation from ammonia borane dehydrogenation.

SIn this paper, the crystal structure prediction software (CBD-GM) based on Bayesian optimization algorithm and deep learning has achieved remarkable results in the field of materials science, especially in the exploration of copper-oxygen compound thin film structures. Through this innovative approach, not only a new two-dimensional copper oxide thin-film structure, H-type-CuO2, is discovered, but also three other never-before-seen novel structures, M1-Cu4O3, M2-Cu4O3, and N-Cu3O2, are successfully predicted. We have calculated and investigated the above four structures using first principles. First, we verified the thermodynamic and kinetic stability of all candidate structures. Second, we calculated the elastic constants using the “stress-strain” method, and the results show that the Young's modulus and Poisson's ratio of the materials exhibit excellent ductile mechanical properties. In addition, we accurately calculate the energy band structure and density of states using the hybridized generalized function (SHE) method to reveal their electronic properties. Combined with negative formation energies and good kinetic, thermodynamic and mechanical stability, the findings in this paper provide a clear direction for the experimental synthesis of copper oxides. At the same time, these research results also provide a solid theoretical foundation for the potential application of these materials in industrial production.

Hollow cerium vanadate nanospheres as a multifunctional host were synthesized for promoting sulfur utilization and long-term cycling stability in Li–S batteries.

Leveraging the CBD-GM platform developed by our group, this study presents the first theoretical prediction of a novel two-dimensional layered MoS ₃ structure.