Gate-Tunable and Thickness-dependent Electronic and Thermoelectric Transport in few-layer MoS2 (1505.05891v2)
Abstract: Over the past few years, there has been a growing interest in layered transition metal dichalcogenides (TMD) such as molybdenum disulfide (MoS2). Most studies so far have focused on the electronic and optoelectronic properties of single-layer MoS2, whose band structure features a direct bandgap, in sharp contrast to the indirect bandgap of thicker MoS2. In this paper, we present a systematic study of the thickness-dependent electrical and thermoelectric properties of few-layer MoS2. We observe that the electrical conductivity () increases as we reduce the thickness of MoS2 and peaks at about two layers, with six-time larger conductivity than our thickest sample (23-layer MoS2). Using a back-gate voltage, we modulate the Fermi energy () of the sample where an increase in the Seebeck coefficient () is observed with decreasing gate voltage () towards the subthreshold (OFF state) of the device, reaching as large as in a four-layer MoS2. While previous reports have focused on a single-layer MoS2 and measured Seebeck coefficient in the OFF state, which has vanishing electrical conductivity and thermoelectric power factor (), we show that MoS2-based devices in their ON state can have as large as in the two-layer sample. The increases with decreasing thickness then drops abruptly from double-layer to single-layer MoS2, a feature we suggest as due to a change in the energy dependence of the electron mean-free-path according to our theoretical calculation. Moreover, we show that care must be taken in thermoelectric measurements in the OFF state to avoid obtaining erroneously large Seebeck coefficients when the channel resistance is very high. Our study paves the way towards a more comprehensive examination of the thermoelectric performance of two-dimensional (2D) semiconductors.
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.