Brief introduction of the results
The computer design of high-efficiency electrocatalysts for oxygen reduction and oxygen evolution reactions (ORR OER) is critical for the development of the hydrogen economy. However, practical design principles remain relatively lacking due to the difficulty of assessing the catalytic properties of materials under electrochemical conditions, such as activity and stability.
Here, based on a 2DMappedia 2D materials database containing more than 6,300 materials,Huang Shiping of Beijing University of Chemical Technology, Zhang Shengli of Nanjing University of Science and TechnologyChen Zhongfang, University of Puerto Rico, et alA data-driven framework has been developed for the discovery of potential 2D materials that are electrochemically stable to oxygen electrocatalysis.
Calculation method:
Based on the Perdew Burke Ernzerhof (PBE) functional in the Generalized Gradient Approximation (GGA), the authors utilize VASP for spin polarization first-principles calculations, and the valence electrons are extended to a plane wave with a kinetic energy cut-off of 400 eV. The convergence threshold for energy is set to 10 10 5, the convergence threshold of the atomic force is 20×10−2evå-1。The author uses 004*2 1 monkhorst pack k-point grid samples the Brillouin zone. In order to eliminate periodic interactions between adjacent conformations, the authors set up a vacuum layer of 15 in the vertical direction of the 2D plane. The authors used the DFT-D3 method of Grimme et al. to describe van der Waals interactions.
Results & Discussion
Fig. 1 (a) Screening the layering stage in the material database. (b-c) Classification of layered materials according to the number of elements (b) and crystal system (c).
As shown in Figure 1A, the authors identified possible prototype compounds from 2DMATPEDIA's open compute database. The authors focus on the study of monary, binary, ternary, and quaternary systems, and exclude unit cells with pure gas phases (e.g., O, I, and Br), atomic numbers over 40, and materials containing radioactive elements (e.g., TC, PM, and PO-LR).
Thus, the authors identified 2854 layered structures, including 22 unmemberate, 728 binary, 1558 ternary, and 546 quaternary compounds (Figure 1b). When further classified into different crystal systems based on space groups (Figure 1C), most 2D materials have monoclinic and orthorhombic crystal systems instead of 2h (hexagonal) and 1t (triangular) configurations.
Figure 2 (a) Decomposition distribution of experimentally available layer compounds and unrealized hypothetical structures. (b) Peeling energy of 20 representative 2D materials. (c) Changes in decomposition and peeling energies of 2854 layered materials. (d) Band gap distribution of 1411 experimentally feasible 2D materials.
In order to reasonably assess the phase stability of 2D materials, the authors established a general benchmark for studying experimentally available metastable phases and hypothetical, unobserved metastable compounds, (Figure 2A). In addition to phase stability, the peel energy is an indicator to evaluate the feasibility of mechanical peeling of 2D materials, therefore, the authors then studied the peel energies of 20 monolayers to reveal the limits of the mechanical peeling method (Figure 2B).
Based on the benchmark values (80 mEV Atom for decomposition energy and 200 MeV Atom for peeling energy), the authors divided each material in the library into four distinct categories (see Figure 2C), which are represented by blue, pink, green, and purple. Among the 2854 2D materials, only 1411 materials in the blue area had good phase stability and mechanical peel properties. The authors studied the band structures of 1411 experimental 2D materials at the GGA-PBE theoretical level to evaluate their conductivity (Fig. 2D). Considering that GGA tends to underestimate the band gap, the authors chose 05EV bandgap threshold, after which only 339 compounds could pass conductivity screening.
Fig.3 (a) Reaction pathway of oxygen electrocatalysis. (b h) Total (b) and partial (c h) distribution of oxygen adsorption energy at 896 possible active sites.
The ORR OER can be performed by an association mechanism or dissociation pathway at the catalyst surface, and the authors describe an acidic electrolyte-based reaction pathway (see Figure 3A), as the intermediates involved in the ORR OER are identical in both acidic and alkaline environments. In the association mechanism, the adsorption of O2 H2O is followed by a hydrogenation and dehydrogenation step, resulting in a series of reaction intermediates that bind to the catalyst surface.
Depending on the catalyst, both the two-electron and four-electron pathways are possible. Hydrogen peroxide (H2O2) is used as a product in the two-electron pathway, and only one reaction intermediate (OOH* for ORR and OH* for OER) is involved. The four-electron reaction involves O*, Oh*, and OOH* reaction intermediates and produces H2O and O2 for ORR and OER, respectively. According to δ