Analysis of simple scattering models on the thermoelectric performance of analytical electron dispersions
Cameron Rudderham, Jesse Maassen

TL;DR
This study compares simple scattering models for thermoelectric predictions across various analytical electron dispersions, revealing significant differences and highlighting the importance of accurate scattering models for material design.
Contribution
It systematically evaluates the impact of different scattering models on thermoelectric properties using analytical band structures, emphasizing the limitations of common approximations.
Findings
DOS scattering model can predict different thermoelectric performance than constant relaxation-time or mean-free-path models.
1D parabolic and 2D quartic bands show highest power factors with DOS model.
Constant relaxation-time model tends to be overly optimistic compared to mean-free-path model.
Abstract
Recent first-principles electron-phonon scattering calculations of heavily-doped semiconductors suggest that a simple DOS scattering model, wherein the electronic scattering rates are assumed to be proportional to the density-of-states, better approximates the rigorous scattering characteristics compared to the commonly used constant relaxation-time and constant mean-free-path approximations. This work investigates how the thermoelectric properties predicted with the DOS model compare to the other two scattering models, using three analytical electron dispersions (parabolic band in 3D/2D/1D, Kane band in 3D/2D/1D, and ring-shaped quartic band in 2D). Our findings show that the scattering models can lead to significant differences, and can disagree about whether certain band structures can provide benefits. A constant relaxation-time is found to always be optimistic compared to a…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
