Enhanced Low-Temperature Thermoelectric Performance in the Two-Dimensional A$\mathrm{Mn}$X Family

Thermoelectric (TE) coolers show considerable promise for replacing aging refrigeration devices, which suffer from the poor low-temperature performance of most TE materials due to the small entropy of carriers and weak lattice vibrational anharmonicity. Here we propose that two-dimensional (2D) semiconductors with multiple bonding layers can be potential TE cooling materials, and the 2D A$\mathrm{Mn}$X (A = $\mathrm{Sr},\mathrm{Ba}$; X = $\mathrm{Sn},\mathrm{Pb}$) family is identified as an example. Owing to the d-orbital inversion of the $\mathrm{Mn}$ element, both the valence and conduction bands possess continuously square-shaped band edges with slight energy divergence, which trigger high mobility and sharply enhanced density of states near the Fermi level, thereby producing large power factors for both p- and n- type materials. Moreover, the coexistence of soft covalent bonds and ionic layers lowers phonon group velocity and strengthens phonon scattering, resulting in extremely low lattice thermal conductivity. At optimal carrier concentrations, both p- and n-type $\mathrm{SrMnPb}$ quintuple layers can reach ultrahigh room-temperature ZT above 2.0, with excellent average ZT beyond 1.8 from 150 to 350 K. This work not only proposes a tellurium-free alternative for TE refrigeration, but also provides general indicators for guiding the search for high-performance TE cooling materials.

Figure 1

(a) Crystal structure of bulk A$\mathrm{Mn}$X (b) The side view, (c) top view, and (d) 2D BZ of A$\mathrm{Mn}$X QL.

Figure 2

(a) Calculated density of states (DOS) and crystal orbital Hamilton population (COHP) projected on $\mathrm{Mn}$—$\mathrm{Pb}$, $\mathrm{Pb}$—$\mathrm{Sr}$, and $\mathrm{Mn}$—$\mathrm{Sr}$ bonds. (b) Electronic localization function (ELF) of $\mathrm{SrMnPb}$. The ELF value of the isosurface is 0.6. (c) The 2D display of electrostatic potential along the (010) plane of $\mathrm{SrMnPb}$.

Figure 3

(a) Orbital decomposed band structures of $\mathrm{SrMnPb}$. The black, green, and orange balls represent the d orbital of the $\mathrm{Sr}$ atom, the d orbital of the $\mathrm{Mn}$ atom, and the p orbital of the $\mathrm{Pb}$ atom, respectively. The 3D morphology of the (b) valence band and (c) conduction band in momentum space.

Figure 4

(a) Seebeck coefficient, (b) conductivity, and (c) power factor as a function of carrier concentration for p-type $\mathrm{SrMnPb}$ QL. (d) Seebeck coefficient, (e) conductivity, and (f) power factor as a function of carrier concentration for n-type $\mathrm{SrMnPb}$ QL. The insets in (b),(e) are the p- and n-type electronic thermal conductivity, respectively.

Figure 5

(a) Phonon dispersions and corresponding PDOS of the $\mathrm{SrMnPb}$ QL. (b) Phonon scattering rate of acoustic (ZA, TA, LA) and optical phonon and (c) phonon group velocity as a function of frequency. (d) Temperature-dependent lattice thermal conductivity of $\mathrm{SrMnPb}$ QL. The lattice thermal conductivity has been renormalized to compare with those of bulk $\mathrm{SrMnPb}$. The inset shows the cumulative lattice thermal conductivity as a function of phonon frequency.

Figure 6

Figure of merit ZT as a function of carrier concentration for (a) p- and (b) n-type $\mathrm{SrMnPb}$ at 150–350 K.