Saurabh Bajaj

Computational and experimental investigation of TmAgTe2 and XYZ2 compounds, a new group of thermoelectric materials identified by first- principles high-throughput screening†

A new group of thermoelectric materials, trigonal and tetragonal XYZ2 (X, Y: rare earth or transition metals, Z: group VI elements), the prototype of which is TmAgTe2, is identified by means of high-throughput computational screening and experiment. Based on density functional theory calculations, this group of materials is predicted to attain high zT (i.e. B1.8 for p-type trigonal TmAgTe2 at 600 K). Among approximately 500 chemical variants of XYZ2 explored, many candidates with good stability and favorable electronic band structures (with high band degeneracy leading to high power factor) are presented. Trigonal TmAgTe2 has been synthesized and exhibits an extremely low measured thermal conductivity of 0.2–0.3 W m � 1 K � 1 for T 4 600 K. The zT value achieved thus far for p-type trigonal TmAgTe2 is approximately 0.35, and is limited by a low hole concentration (B10 17 cm � 3 at room temperature). Defect calculations indicate that TmAg antisite defects are very likely to form and act as hole killers. Further defect engineering to reduce such XY antisites is deemed important to optimize the zT value of the p-type TmAgTe2.

Understanding thermoelectric properties from high-throughput calculations: trends, insights, and comparisons with experiment

We present an overview and preliminary analysis of computed thermoelectric properties for more than 48 000 inorganic compounds from the Materials Project (MP). We compare our calculations with available experimental data to evaluate the accuracy of different approximations in predicting thermoelectric properties. We observe fair agreement between experiment and computation for the maximum Seebeck coefficient determined with MP band structures and the BoltzTraP code under a constant relaxation time approximation (R2 = 0.79). We additionally find that scissoring the band gap to the experimental value improves the agreement. We find that power factors calculated with a constant and universal relaxation time approximation show much poorer agreement with experiment (R2 = 0.33). We test two minimum thermal conductivity models (Clarke and Cahill–Pohl), finding that both these models reproduce measured values fairly accurately (R2 = 0.82) using parameters obtained from computation. Additionally, we analyze this data set to gain broad insights into the effects of chemistry, crystal structure, and electronic structure on thermoelectric properties. For example, our computations indicate that oxide band structures tend to produce lower power factors than those of sulfides, selenides, and tellurides, even under the same doping and relaxation time constraints. We also list families of compounds identified to possess high valley degeneracies. Finally, we present a clustering analysis of our results. We expect that these studies should help guide and assess future high-throughput computational screening studies of thermoelectric materials.

YCuTe2: a member of a new class of thermoelectric materials with CuTe4-based layered structure

Intrinsically doped samples of YCuTe2 were prepared by solid state reaction of the elements. Based on the differential scanning calorimetry and the high temperature X-ray diffraction analyses, YCuTe2 exhibits a first order phase transition at ∼440 K from a low-temperature-phase crystallizing in the space group Pm1 to a high-temperature-phase in P. Above the phase transition temperature, partially ordered Cu atoms become completely disordered in the crystal structure. Small increases to the Cu content are observed to favour the formation of the high temperature phase. We find no indication of superionic Cu ions as for binary copper chalcogenides (e.g., Cu2Se or Cu2Te). All investigated samples exhibit very low thermal conductivities (as low as ∼0.5 W m−1 K−1 at 800 K) due to highly disordered Cu atoms. Electronic structure calculations are employed to better understand the high thermoelectric efficiency for YCuTe2. The maximum thermoelectric figure of merit, zT, is measured to be ∼0.75 at 780 K for Y0.96Cu1.08Te2, which is promising for mid-temperature thermoelectric applications.

Calculation of dopant solubilities and phase diagrams of X–Pb–Se (X = Br, Na) limited to defects with localized charge

The control of defects, particularly impurities, to tune the concentrations of electrons and holes is of utmost importance in the use of semiconductor materials. To estimate the amount of dopant that can be added to a semiconductor without precipitating secondary phases, a detailed phase diagram is needed. The ability of ab initio computational methods to predict defect stability can greatly accelerate the discovery of new semiconductors by calculating phase diagrams when time-consuming experimental ones are not available. DFT defect energy calculations are particularly successful in identifying doping strategies by determining the energy of multiple defect charge states in large band gap semiconductors and insulators. In metals, detailed phase diagrams can be determined from such calculations but only one, uncharged defect is needed. In this work, we have calculated dopant solubilities of Br and Na in the thermoelectric material PbSe by mapping its solvus boundaries in different regions of the respective ternary phase diagrams using DFT defect energy calculations. The narrow gap PbSe provides an example where defects with nominal charge state (based on valence counting) have properly-localized charge states. However, defects with unexpected charge states produce delocalized electrons, which are then, in effect, defects with the expected charge state. Simply applying the methods for calculating multiple defect charge states in PbSe and treating them as separate defects fails to predict properties measured by experiments. Performing thermodynamic calculations using only the expected charge states, excluding others, enables accurate prediction of experimentally measured doping efficiencies and phase diagrams. Identifying which defect charge states to include in thermodynamic calculations will expedite the use of such calculations for other semiconductors in understanding phase diagrams and devising effective doping strategies.