Alex Ganose1 2 Winnie Leung1 Adam Jackson1 Robert Palgrave1 David Scanlon1 2

1, University College London, London, , United Kingdom
2, Diamond Light Source, Didcot, , United Kingdom

Thermoelectric materials, used to convert thermal into electrical energy, present a promising route for renewable energy generation. The range of applications for thermoelectrics is broad, with industries from manufacturing to the automotive likely to benefit from efficient recycling of waste thermal energy.1 The dimensionless figure of merit for thermoelectrics, ‘ZT’, depends on both electronic and thermal transport properties, with a material considered promising if its ZT exceeds ~1.5. Unfortunately, despite over 50 years’ development, the champion thermoelectric materials, such as Bi2Te3, show lack-lustre performance and are costly to produce due to their reliance on tellurium.2 Significant research effort has been spent attempting to produce oxide based thermoelectrics due to their earth-abundance, chemical stability and dramatically reduced costs. However, all attempts to produce high performance n-type oxide thermoelectrics have failed, often due to their high lattice conductivity which limits obtainable ZT.3

Standard packages now exist for calculating ZT from an electronic band structure, with the results being dependent on two major approximations: a fixed lattice thermal conductivity and electronic chemical potential (Fermi level). Typically, the Fermi level is assumed without knowledge of the true response of the material to defects and doping, which can lead to incorrect predictions of high ZT capability.

In this work, we have used rational chemical design to pinpoint a series of layered oxides that should exhibit degenerate n-type conductivity, whilst still possessing very low lattice thermal conductivity.4 We employ state of the art methods to calculate the lattice thermal conductivity, using many-body perturbation theory to capture phonon-phonon scattering processes. We also use rigorous defect chemistry analysis, performed using hybrid density functional theory, to explicitly consider the intrinsic and extrinsic defect behaviour and obtain a physical and realistic doping density and Fermi level. Combining these methods, we have predicted the largest ZT of any oxide thermoelectric material previously reported and provide guidance on the growth conditions to enhance thermoelectric power conversion.


1. L. E. Bell, Cooling, heating, generating power, and recovering waste heat with thermoelectric systems. Science 321, 1457–1461 (2008)
2. M. W. Gaultois, T. D. Sparks, C. K. H. Borg, R. Seshadri, W. D. Bonificio, and D. R. Clarke, Data-driven review of thermoelectric materials: performance and resource considerations, Chemistry of Materials 25, 2911–2920 (2013)
3. G. Tan, L-D. Zhao, and M. G. Kanatzidis, Rationally designing high-performance bulk thermoelectric materials, Chemical Reviews 116, 12123–12149 (2016)
4. Alex M. Ganose, W. W. Leung, Adam J. Jackson, R. G. Palgrave, and David O. Scanlon, Submitted (2017)