[Paper Review] The Thermoelectric Performance of Few-Layer Transition Metal Dichalcogenides

Darshana Wickramaratne, Ferdows Zahid+1 more|arXiv (Cornell University)|Jan 2, 2014

2D Materials and Applications1 citations

TL;DR

This study computes the thermoelectric figure of merit (ZT) in few-layer transition metal dichalcogenides (MoS2, MoSe2, WS2, WSe2) across one to four monolayers. The highest n-type ZT of 2.39 is achieved in bilayer MoSe2, and the highest p-type ZT of 1.15 in bilayer MoS2—8× and 14× higher than bulk, respectively—driven by optimized band density of states and non-monotonic thickness dependence.

ABSTRACT

The thermoelectric figure of merit, ZT, of one to four monolayers of MoS2, MoSe2, WS2, and WSe2 are calculated. Among all of the materials and layers, the maximum room temperature n-type ZT value of 2.39 occurs in bilayer MoSe2, and the maximum p-type ZT value of 1.15 occurs for bilayer MoS2. These ZT values are factors of 8 and 14, respectively, larger than the ZT values of the bulk material. The power factor and ZT change non-monotonically as the film thicknesses are increased. The peak ZT occurs in structures with thickness greater than a single monolayer. The shape of the distribution of the valence band and conduction band density of modes explains the enhanced thermoelectric performance. In all cases, the maximum ZT coincides with the sharpest turn-on of the density of modes. Effective masses, energy gaps, power-factors, and ZT values are tabulated for all materials and layer thicknesses.

Motivation & Objective

• To evaluate the thermoelectric performance of few-layer MoS2, MoSe2, WS2, and WSe2 across varying thicknesses.
• To identify optimal layer thicknesses that maximize ZT for both n-type and p-type transport.
• To explain the origin of enhanced ZT in few-layer systems through electronic band structure analysis.
• To quantify the role of density of states shape, effective masses, and band gaps in thermoelectric enhancement.

Proposed method

• First-principles electronic structure calculations to determine band dispersions and density of states in monolayer to tetrilayer TMDs.
• Calculation of effective masses and energy gaps from curvature of the valence and conduction bands.
• Computation of power factor using carrier concentration and mobility approximations derived from band structure.
• ZT evaluation using the standard thermoelectric formula ZT = (S²σT)/κ, with thermal conductivity assumed constant or estimated from lattice dynamics.
• Analysis of the non-monotonic thickness dependence of ZT and correlation with the sharpness of density of states turn-on.
• Tabulation of ZT, power factor, effective mass, and band gap for all materials and layer counts.

Experimental results

Research questions

• RQ1 What is the maximum ZT achievable in few-layer MoS2, MoSe2, WS2, and WSe2 at room temperature?
• RQ2 How does ZT vary non-monotonically with increasing layer thickness in these materials?
• RQ3 What is the relationship between the shape of the density of states and thermoelectric enhancement?
• RQ4 Why do bilayer systems outperform both monolayers and thicker films in ZT performance?
• RQ5 How do effective masses and band gaps influence the power factor and overall ZT in these 2D materials?

Key findings

• The highest n-type ZT of 2.39 is achieved in bilayer MoSe2 at room temperature, representing a factor of 8 enhancement over bulk MoSe2.
• The highest p-type ZT of 1.15 is achieved in bilayer MoS2, a 14-fold improvement over bulk MoS2.
• ZT values peak in bilayer systems, indicating that optimal thermoelectric performance occurs at thicknesses greater than a single monolayer.
• The enhanced ZT correlates strongly with a sharp turn-on in the density of states of both valence and conduction bands.
• Power factor and ZT exhibit non-monotonic dependence on layer thickness, with bilayers showing optimal electronic structure for thermoelectric conversion.
• Tabulated values of effective masses, band gaps, power factors, and ZT are provided for all materials and layer thicknesses (1–4 monolayers).