Density Functional Theory Study of Edge-Induced Atomic-Scale Structural Phase Transitions of MoS₂ Nanocrystals: Implications for a High-Performance Catalyst

Abstract

Molybdenum disulfide (MoS2) has attracted much attention as a material to replace the noble-metal-based hydrogen evolution reaction catalyst. Polymorphism is an important factor in improving the catalytic performance of transition-metal dichalcogenides (TMDs) including MoS2. Several methods have been proposed to synthesize the 1T/1T′ phase with high catalytic efficiency, and a gas–solid reaction has recently been proposed as one of the alternative methods. However, the atomic-scale reaction mechanism between gas molecules and MoS2 has not been clarified. Here, we report a detailed atomic-scale mechanism of structural phase transition of MoS2 nanocrystals under reaction with CO gas molecules using density functional theory calculations. We confirm that the evaporation of S atoms at the edge is much faster than the evaporation at the basal plane of MoS2 nanocrystals. It is found that the S evaporation at the edge induces the structural change from 2H to 1T/1T′ in the basal plane of nanocrystals. The structural change is also attributed to the chain reaction due to the sequential migration of S atoms to the octahedral sites, which is energetically favorable. The present results provide a guideline for the gas–solid reaction-based phase control of TMDs including MoS2 to synthesize a high-performance catalyst.

Phase Engineering of Transition Metal Dichalcogenides via a Thermodynamically Designed Gas–Solid Reaction

“…To understand the atomistic reaction behavior in anion extraction-based polymorph conversion, DFT calculations for the reaction barriers of CO-mediated V S formation and S diffusion for polymorph conversion of WS 2 were conducted in the same process as the mechanism that was presented in the previous work. 43 The calculation results showed that the reaction barrier for the CO-mediated polymorph conversion of WS 2 was higher than that of MoS 2 , and the difference in reaction barriers was consistent with the experimental results. We modeled WS 2 nanocrystals with a triangular shape, which was observed in several previous reports.…”

“…Through the repeated diffusion of S atoms at the trigonal prismatic site to the octahedral site, the whole nanocrystal could be changed to the T phase, as suggested in a previous study. 43 Figure 5a shows a higher reaction barrier for the ratelimiting step (1.31 eV) than that of MoS 2 (0.49 eV). 43 Therefore, S diffusion to the octahedral site in WS 2 needs to overcome a higher reaction barrier than that in MoS 2 .…”

“…43 Figure 5a shows a higher reaction barrier for the ratelimiting step (1.31 eV) than that of MoS 2 (0.49 eV). 43 Therefore, S diffusion to the octahedral site in WS 2 needs to overcome a higher reaction barrier than that in MoS 2 . The energy of the final state (state 9 in Figure 5) was lowered by 1.74 eV compared to the initial state (state 1 in Figure 5).…”

“…Furthermore, this structure exhibited a S dimer structure on the edge similar to the initial structure of MoS 2 used in the previous work. 43 The reaction barriers were calculated for CO-mediated V S formation on the initial structure of WS 2 . In the reaction barrier calculation process, we set the distance between the C atom of CO and S atom of WS 2 as a constraint and calculated the DFT energy of optimized structure by changing the distance.…”

“…Among the seven V S , the sixth V S showed the highest value, which was higher than that of MoS 2 (0.64 eV). 43 This meant that CO-mediated V S formation on the edge of WS 2 was more difficult than that on the edge of MoS 2 . This result is well-matched to the experimental result that the T/T′ phase of WS 2 was obtained at a higher temperature than MoS 2 .…”