Novel doping alternatives for single-layer transition metal dichalcogenides

Abstract

Successful doping of single-layer transition metal dichalcogenides (TMDs) remains a formidable barrier to their incorporation into a range of technologies. We use density functional theory to study doping of molybdenum and tungsten dichalcogenides with a large fraction of the periodic table. An automated analysis of the energetics, atomic and electronic structure of thousands of calculations results in insightful trends across the periodic table and points out promising dopants to be pursued experimentally. Beyond previously studied cases, our predictions suggest promising substitutional dopants that result in p-type transport and reveal interesting physics behind the substitution of the metal site. Doping with early transition metals (TMs) leads to tensile strain and a significant reduction in the bandgap. The bandgap increases and strain is reduced as the d-states are filled into the mid TMs; these trends reverse as we move into the late TMs. Additionally, the Fermi energy increases monotonously as the d-shell is filled from the early to mid TMs and we observe few to no gap states, indicating the possibility of both p- (early TMs) and n- (mid TMs) type doping. Quite surprisingly, the simulations indicate the possibility of interstitial doping of TMDs; the energetics reveal that a significant number of dopants, increasing in number from molybdenum disulfide to diselenide and to ditelluride, favor the interstitial sites over adsorbed ones. Furthermore, calculations of the activation energy associated with capturing the dopants into the interstitial site indicate that the process is kinetically possible. This suggests that interstitial impurities in TMDs are more common than thought to date and we propose a series of potential interstitial dopants for TMDs relevant for application in nanoelectronics based on a detailed analysis of the predicted electronic structures.