[Paper Review] Electronic structure and thermoelectric properties of CuRh(1-x)MgxO2

A. Maignan, Volker Eyert+4 more|arXiv (Cornell University)|Aug 13, 2009

Copper-based nanomaterials and applications1 references4 citations

TL;DR

This study investigates Mg-doped CuRhO₂ as a p-type thermoelectric oxide, demonstrating that Mg²⁺ substitution for Rh³⁺ up to 12% induces hole doping in the narrow Rh 4d bands, leading to a nearly temperature-independent power factor of ~7×10⁻⁴ W K⁻² m⁻¹ up to 1000 K. Electronic structure calculations and transport measurements confirm that the thermopower arises primarily from Rh 4d t₂g orbitals, with a temperature-independent contribution explaining the weak T-dependence of the power factor beyond 300 K.

ABSTRACT

Electronic structure calculations using the augmented spherical wave method have been performed for CuRhO2. For this semiconductor crystallizing in the delafossite structure, it is found that the valence band maximum is mainly due to the 4d t2g orbitals of Rh^{3+}. The structural characterizations of CuRh(1-x)MgxO2 show a broad range of Mg^{2+} substitution for Rh^{3+} in this series, up to about 12%. Measurements of the resistivity and thermopower of the doped systems show a Fermi liquid-like behavior for temperatures up to about 1000K, resulting in a large weakly temperature dependent power factor. The thermopower is discussed both within the Boltzmann equation approach as based on the electronic structure calculations and the temperature independent correlation functions ratio approximation as based on the Kubo formalism.

Motivation & Objective

• To understand the electronic and thermoelectric transport mechanisms in Mg-doped CuRhO₂, a p-type delafossite oxide.
• To determine the solubility limit of Mg²⁺ in CuRhO₂ and its impact on electrical transport.
• To clarify whether charge transport occurs primarily in RhO₂ or Cu layers, resolving conflicting interpretations in the literature.
• To quantify the role of electronic structure and Fermi surface topology in the observed weak temperature dependence of the thermopower and power factor.
• To validate theoretical models (GGA, TICR) against experimental transport data for predictive insight into thermoelectric performance.

Proposed method

• Electronic structure calculations using the augmented spherical wave (ASW) method with GGA functional to determine the density of states and orbital contributions to the valence band maximum.
• Structural characterization via X-ray diffraction to confirm solid solution formation and determine the solubility limit of Mg in CuRhO₂.
• Measurement of resistivity and thermopower from 300 K to 1000 K to assess temperature dependence and power factor.
• Application of the Boltzmann transport equation to model thermopower based on the calculated electronic structure.
• Extension of the temperature-independent correlation functions ratio (TICR) model by adding a temperature-independent thermopower contribution S₀ to fit experimental data.
• Use of the rigid band model to relate changes in carrier concentration to the observed transport behavior across different Mg doping levels.

Experimental results

Research questions

• RQ1 What is the maximum solubility of Mg²⁺ in CuRhO₂, and how does it affect the crystal structure?
• RQ2 Which electronic bands—Rh 4d or Cu 3d—dominate charge transport in Mg-doped CuRhO₂?
• RQ3 Why does the power factor remain nearly constant with temperature up to 1000 K, despite the expected T-dependence of thermopower and resistivity?
• RQ4 To what extent can the experimental thermopower be quantitatively explained by the GGA-based Boltzmann transport model?
• RQ5 What is the origin of the temperature-independent contribution to the thermopower, and how does it relate to the Fermi surface topology?

Key findings

• Mg²⁺ can substitute for Rh³⁺ in CuRhO₂ up to a solubility limit of approximately 12%, confirmed by structural analysis of samples synthesized above 1000 °C.
• The valence band maximum in CuRhO₂ is primarily composed of Rh 4d t₂g orbitals, confirming that transport is dominated by the RhO₂ layers.
• The resistivity exhibits a T² dependence, indicating Fermi liquid-like behavior up to 1000 K, consistent with hole doping in a narrow band.
• The thermopower shows weak temperature dependence, with values ranging from 70 to 130 μV K⁻¹ at 300 K, and the power factor remains nearly constant at ~7×10⁻⁴ W K⁻² m⁻¹ for x = 0.10.
• Theoretical modeling using GGA and Boltzmann transport equations shows good qualitative agreement with experiment, but a temperature-independent thermopower contribution S₀ (40–100 μV K⁻¹) is required for quantitative fit, suggesting localized states at H and A points on the Fermi surface.
• The same density of states from GGA calculations successfully describes the transport behavior across all doping levels, supporting the validity of the rigid band model for this system.