吸附
从头算
单层
化学物理
解吸
电荷密度
分子动力学
价(化学)
材料科学
密度泛函理论
分子
化学
计算化学
物理化学
纳米技术
物理
有机化学
量子力学
标识
DOI:10.1021/acs.jpcc.0c03786
摘要
The potential energy surface for NO2 physisorbed on a MoS2 monolayer, acting as a chemical sensor, is complex with several configurations having similar adsorption energies (ΔEads) and charge-transfer characteristics. Hence, this can be considered to be a difficult system to model. A careful exploration of the energy surface is necessary in order to identify any sites at which strong adsorption and/or enhanced charge transfer can occur, which would affect sensor operation. Beyond this, a general computational approach is needed that can efficiently identify the lowest-energy configuration for a molecule weakly adsorbed on MoS2 since this pertains to many potential sensor applications. In the present study, the computational methods are first carefully tested. Then ab initio molecular dynamics simulations are employed for an unbiased sampling of configuration space, which makes use of the approximate correlation between increasing ΔEads and decreasing NO2-MoS2 separation. A simulation temperature of ∼225 K, which promotes surface diffusion of NO2 but not rapid desorption, appears to be nearly ideal. A series of simulations identified several transient configurations with relatively small separations, and each of these was evaluated and compared with previously reported adsorption models. A well-defined structure with the highest ΔEads is thus identified and characterized, and further insight into adsorption and surface structure is obtained by computing the charge density at the valence band maximum and the effects of NO2 on charge density. These results were extended to a preliminary investigation of the intercalation of NO2 between MoS2 layers. This is found to be energetically unfavorable and to destabilize a 2H MoS2 bilayer at 300 K. However, when relaxed at 0 K, intercalated NO2 is seen to affect the layer stacking and to increase the charge transfer relative to adsorption on a monolayer. The implications of these results for sensor applications are briefly discussed.
科研通智能强力驱动
Strongly Powered by AbleSci AI