Microstructure analysis of IrO2 thin films
Introduction
Recent interest in Ir oxides has been stoked by the prediction of topological states appearing in pyrochlore-type iridate films strained along the [111] crystal direction. [1], [2] Among the iridate pyrochlores, Pr2Ir2O7 has been shown by angle-resolved photoemission spectroscopy (ARPES) to possess a single-point Fermi node at the Γ-point [3], making it a good candidate for epitaxial strain experiments if stoichiometric and atomically flat surfaces can be prepared. However, a common problem for the growth of iridate thin films by physical vapor deposition techniques such as pulsed laser deposition (PLD), is the volatility of nonstoichiometric iridium oxides. This can lead to severe stoichiometry deviations in thin films and it is the main reason why pyrochlore iridate thin films have so far been fabricated mostly by solid-phase epitaxy, where the film precursor material is deposited at room temperature and crytallization occurs during post-annealing. [4] While this process produces thin films with excellent structural properties, it unavoidably leads to a relaxed lattice with a compositionally degraded surface, making the films unsuitable for ARPES studies. We therefore explore in this work the thin film growth process of the simplest iridate, IrO2, to determine the PLD process parameter window where the Ir4+ state can be stabilized. Besides the process optimization, we recognize that IrO2 is an interesting material for a number of reasons, and we analyze the microstructure, transport behavior, and the strain relaxation mechanism in IrO2(110) films grown on TiO2(110) substrates.
Iridium dioxide belongs to a small family of oxides that are good metals, together with RuO2, ReO2, OsO2, MoO2, and WO2. [5] The Fermi level in IrO2 is located close to the top of the Ir band, leading to nonmagnetic metallic behavior. [6], [7], [8] IrO2 thin films have been used as a spin-to-current detector in spin valves [9] and may be useful in devices with magnetically switchable majority carrier type. [10] Other well-known applications of bulk or polycrystalline IrO2 are biocompatible electrodes, [11] oxygen barriers in semiconductor devices, [12] electrodes in ferroelectric devices, [13] and electrocatalytic oxygen evolution electrodes for water splitting. [14].
The main difficulty in growing IrO2 films is the difficulty of oxidizing Ir metal and the volatility of nonstoichiometric Ir oxides. [15], [16] IrO2 films have been grown by various techniques, including molecular beam epitaxy (MBE), [10] sputtering, [17] and PLD. [18], [19] Although it is common to use a metallic Ir target for film deposition and allow the ambient oxygen to oxidize the deposited metal at the film surface, it is known that in PLD growth, oxygen is also transported to the film from an oxide target. [20], [21] In an attempt to obtain stoichiometric films over a wide range of process parameters and for the results to be relevant for oxide ablation, we chose to use an IrO2 target instead of Ir metal to grow IrO2 films by PLD.
Unlike bulk iridate synthesis, which is often done in an enclosed quartz tube, thin film growth is a non-equilibrium process where the evaporation of volatile species and resputtering from the film surface by the ablation plume cannot be prevented. A particularly severe problem for iridate growth is the formation of a metastable IrO3 phase, which is volatile even at low temperatures and therefore severely restricts the useful parameter space where a stoichiometric IrO2 phase can form. Although the Ir oxide volatility is not unique to IrO2 and applies to pyrochlore iridate thin film growth as well, the loss of Ir from the film surface appears to be dependent on the particular crystal structure and the surface layer composition. For example, the growth of perovskite iridates such as SrIrO3 and related Ruddlesden-Popper phases appears to be less affected by Ir loss.
We show here that it is possible to grow IrO2 films that are stoichiometric and possess an atomically well-defined surface morphology. The grain structure of the films is associated with strain relaxation, which occurs by the nucleation of misfit dislocation boundaries. Some of the misfit dislocations lead to extended defects that can be observed on the film surface as distinct grain boundaries and affect the low-temperature transport characteristics of the films.
Section snippets
Experiments
The IrO2 films were grown by pulsed laser deposition [22] on TiO2(110) single crystal substrates that were annealed in air for one hour at 900 °C before film growth to form a regular step-and-terrace surface. [23] The ablation target was a nominally stoichiometric IrO2 pellet, which was ablatied with a KrF (Thin Film Star, ) laser operating at 5 Hz. Although an oxide target was used instead of Ir metal, the deposition rate was relatively low. The film growth experiments were therefore done
Results and discussions
The oxidation rate of an IrOx film depends on the growth temperature and the ambient oxygen pressure. The structure and composition of films was therefore mapped over a range of growth temperatures, from 500 °C to 900 °C and oxygen pressures, from 4 m Torr to 500 m Torr. Fig. 1 shows a comparison of several characteristic Cu XRD patterns of IrOx films deposited at different growth conditions. Clear IrO2 diffraction peaks were obtained at 500 °C and 100 m Torr. All film peaks were sharp and
Conclusion
We have determined the growth window for the formation of IrO2 thin films by PLD on TiO2 substrates. Optimal films were obtained at 500 °C at an oxygen pressure of 100 m Torr. The resistivity behavior of the films is dominated by the presence of stacking-fault grain boundaries in the film. The formation of the grain structure is driven by strain relaxation. This strain relaxation leads to the formation of columnar grains with relatively high grain boundary resistance. However, despite the grain
Acknowledgments
The authors thank Shintaro Kobayashi of Rigaku Co. for grazing-incidence in-plane XRD measurements. This work was supported by a Grant-in-Aid for Scientific Research (Grant Nos. 25706022, and 26105002) from the Japan Society for the Promotion of Science. X. H. was supported by the Japan Society for the Promotion of Science Fellowship and the Program for Leading Graduate Schools (MERIT).
References (27)
- et al.
J. Electroanal. Chem.
(2013) - et al.
J. Less-Common Met.
(1963) - et al.
Appl. Surf. Sci.
(2006) - et al.
Phys. Rev. Lett.
(2014) - et al.
Phys. Rev. B
(2010) - et al.
Nat. Commun.
(2015) - et al.
Sci. Rep.
(2015) - et al.
Inorg. Chem.
(1969) - et al.
J. Phys. Chem. C
(2015) - et al.
Phys. Rev. B
(2006)
Nat. Commun.
Phys. Rev. B
Cited by (14)
Influence of chromium and lanthanum incorporation on the optical properties, catalytic activity, and stability of IrO<inf>x</inf> nanostructured films for hydrogen generation
2023, International Journal of Hydrogen EnergyHigh-temperature X-ray scattering studies of atomic layer deposited IrO<inf>2</inf>
2020, Journal of Applied CrystallographyEngineering metal oxidation using epitaxial strain
2023, Nature Nanotechnology