3
the density of states of a similar p-type half-Heusler alloy ZrCo(Sn0.3Sb0.7)1-xAlx, yielding
“resonant states” near the band edge. The present findings could pave the way for refocusing
traditional doping techniques to yield state-of-the-art thermoelectric performance in common
thermoelectric materials.
Experimental Procedures
Elements Hf, Zr, Co, Sn, Sb, and Al were weighted according to the nominal composition
Hf0.3Zr0.7Co(Sn0.3Sb0.7)1-xAlx (x=0, 0.005, 0.01, 0.015, 0.02). Following prior work20, a small
amount (~5%) of excess Sb was added to compensate for the evaporation of Sb during arc melting.
Since the preferential doping site of Al in Hf0.3Zr0.7Co(Sn0.3Sb0.7) is not known a priori, additional
alloys were synthesized to investigate the dopability of the (Hf,Zr) sublattice and Co sublattice.
The elements were loaded into an arc furnace in preparation for melting under the Argon
atmosphere. Because of the low concentration of Al as a dopant element, Al was pre-melted with
small amounts of Hf and Zr to ensure homogeneous dissolution of Al in the final product. The
ingot was re-melted twice to improve the overall homogeneity. Afterward, the ingots were
pulverized into fine 10-30 µm size powders, followed by consolidation using Spark Plasma
Sintering (Thermal Technologies® SPS 10-4) under an axial pressure of 50 MPa at 1073 K for 10
minutes and then at 1423 K for 5 minutes in vacuum. Details of the sample preparation were
reported in previous publications with appropriate modifications.21, 22
The crystal structure was investigated via X-ray diffraction (XRD) technique using Cu-Kα X-rays
(1468.7 eV) on a PANalytical Empyrean Diffractometer at a 3°/min scan rate. Microstructures and
elemental composition profiles of the samples were analyzed via scanning electron microscopy
(SEM) using backscattered electron (BSE) imaging on an FEI Quanta 650 operating at an
accelerating voltage of 15 keV, a spot size of 4 nm, and a working distance of approximately 10
mm. The microstructures of the half-Heusler alloys were characterized by electron backscatter
diffraction (EBSD) using a Helios UC G4 Dual Beam FIB-SEM. The samples were mounted in
nonconductive epoxy. Before EBSD, the surfaces of the samples were mechanically polished with
SiC abrasive papers of grit sizes 600, 1200, 2500, and 4000. The surface treatment was continued
by polishing with diamond polishing suspensions of 0.25 micrometer, followed by 0.05-
micrometer colloidal silica suspension. The (EBSD) images were evaluated to plot the grain size
distribution curves using the Crystal Imaging software attached to the instrument.
The temperature dependence of electrical resistivity and Seebeck coefficient were measured with
a ZEM 3 (ULVAC Riko, Japan) electrical properties measurement system. Thermal diffusivity
(D) was measured via the Laser Flash technique (LFA 457 MicroFlash System). Heat capacity,
Cp, was measured via Differential Scanning Calorimetry (Netzsch DSC 404C), and material
density (ρ) was measured using the Archimedes method. Thermal conductivity was calculated
using the relation κ = DρCν. The lattice component to the thermal conductivity was estimated by
applying the Weidemann-Franz law to the electrical part of the total thermal conductivity equation,
κL = κ - κe. The Wiedemann-Franz law represents a standard model of heat conduction through
current carriers that gives κe = LσT, where L is the Lorenz number, and σ is the electrical
conductivity. Room-temperature mobility (µ) and carrier concentration (n) was obtained via the
Hall Effect using the electrical transport option (ETO) of Versalab. Samples used for the
measurement were polished to a thickness of approximately 0.3 mm. A magnetic field (B) was
applied perpendicular to the supplied electrical current (I). In this configuration, the resistance
developed across the transverse leads (R⊥) is measured while the magnetic field (B) was varied
from 0.1 T to -0.1 T at constant I. The slope obtained from the R⊥ vs. B plot was used to estimate