Login / Signup

Synergistic effect of alloying on thermoelectric properties of two-dimensional PdPQ (Q = S, Se).

S ShahabfarYi XiaM H MorshedsoloukM MohammadiS Shahab Naghavi
Published in: Physical chemistry chemical physics : PCCP (2023)
Hosts of 2D materials exist, yet few allow compositional and structural tailoring as the MQ 2 (M = Mo, W; Q = S, Se) family does, for which various structural superlattices have been synthesized. Using thorough first-principles calculations, we show how bonding hierarchy contributes to the structural resilience of 2D PdPQ and allows for full-range alloying of sulfur and selenium. Within the structural unit of Pd 2 P 2 Q 2 , the covalently-bonded [P 2 Q 2 ] 4- polyanions hold the structure together with their molecular-like P-P bonds while ionically bonded Pd-Qs allow the S/Se substitution. Here, the bonding hierarchy imparts superior electronic and structural features to the PdPQ monolayers. As such, the flat-and-dispersive valence band and the eight degenerate valleys of the conduction band benefit the p-type and n-type thermoelectricity of pristine PdPQ, which can be further enhanced by alloying. The high-entropy alloying synergistically suppresses the lattice heat transport from 75 to 30 W m -1 K -1 and increases the band degeneracy of PdPQ monolayers, resulting in an overall improvement in zT . Combining these features, in a naïve approach, results in a large zT approaching two for both p-type and n-type doping. However, accurate fully-fledged electron-phonon calculations rebut this promise, showing that at high temperatures, the increased electron scattering results in a stagnant power factor in the flat-and-dispersive valence band. Using a realistic first-principles scattering, we finally calculate the thermoelectric efficiency of PdPQ (Q = S, Se) and highlight the importance of an accurate estimation of electron relaxation time for thermoelectric predictions.
Keyphrases
  • ionic liquid
  • density functional theory
  • gas chromatography mass spectrometry
  • solid phase extraction
  • machine learning
  • single molecule
  • mass spectrometry
  • solar cells
  • big data
  • artificial intelligence
  • deep learning