Competitive Growth of Scrutinyite (α-PbO2) and Rutile Polymorphs of SnO2 on All Orientations of Columbite CoNb2O6 Substrates
Julia Wittkamper, Zhongnan Xu, Boopathy Kombaiah, Farangis Ram, Marc De Graef, John R. Kitchin, Gregory S. Rohrer and Paul A. Salvador, “Competitive Growth of Scrutinyite (α-PbO2) and Rutile Polymorphs of SnO2 on All Orientations of Columbite CoNb2O6 Substrates”, Crystal Growth & Design 17(7), 3929-3939 (2017). https://doi.org/10.1021/acs.cgd.7b00569
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
Scrutinyite (the α-PbO2 structure) dioxides are interesting targets for epitaxial stabilization: scrutinyite IrO2 had been predicted to be an excellent oxygen evolution catalyst, but it and scrutinyite RuO2 have never been made, and isostructural single-crystal substrates are not commercially available. SnO2 is a convenient test case because scrutinyite SnO2 is known and lies only about 6 kJ/mol above stable rutile SnO2. Rutile and scrutinyite share nearly the same hexagonally stacked oxygen sublattice and differ mainly in how cations fill the octahedral sites, so anion packing cannot discriminate between them. The paper asks how rutile and scrutinyite SnO2 grow epitaxially on columbite CoNb2O6 (a cation-ordered derivative of scrutinyite), and whether rutile is destabilized enough for the metastable phase to form, across all substrate orientations (Wittkamper et al. 2017).
Approach
The authors used combinatorial substrate epitaxy (CSE), in which a locally epitaxial film is deposited on a polished, large-grained polycrystal and each film grain is related to the substrate grain beneath it by electron backscatter diffraction (EBSD). Dense CoNb2O6 pellets were made by solid-state reaction and sintered (final anneal 40 h at 1240 °C) to grain sizes of 5 to 70 μm. SnO2 films were grown by pulsed laser deposition at 700 °C in about 50 mTorr O2, at a low growth rate (0.03 Å/pulse, about 20 nm film) and a high growth rate (0.6 Å/pulse, about 100 nm film). Because the two polymorphs give similar Kikuchi patterns, phase assignments from commercial Hough-based indexing were checked with an in-house dictionary method that compares experimental patterns to simulated patterns of all orientations. Selected grains were examined by focused ion beam lift-out and TEM. Orientation relationships (ORs) were rationalized qualitatively using tabulated misfit strains and the matching of cation packing patterns across the interface (Wittkamper et al. 2017).
Key findings
Growth was grain-over-grain: each substrate grain carried a single film grain of one phase and one orientation, and the film was a mixture of scrutinyite and rutile SnO2. Whole-pattern dictionary indexing agreed with the commercial software and showed the two phases can be distinguished by EBSD alone. Of 135 film/substrate grain pairs, 66 were scrutinyite and 69 rutile. Sixty-five of the 66 scrutinyite grains adopted a unit-cell over unit-cell OR aligning the primary eutactic planes, (100)[001] of film parallel to (100)[001] of the substrate (Pcnb setting), with mean misorientations of 5.6° out of plane and 6.2° in plane. Sixty-four of 69 rutile grains adopted a secondary eutaxial OR, {101}<010> rutile parallel to (010)[001] columbite, with mean misorientations of 7.5°; the authors note this matches the relationship seen geologically for scrutinyite-type TiO2 lamellae at rutile {101} twins, whose planes resemble scrutinyite (010).
Phase selection depended on substrate orientation. Substrate grains between the (001) and (100) poles supported scrutinyite, while grains near (010) supported rutile, with a demarcation line roughly joining (012) and (110); scrutinyite covered slightly more than half of orientation space. The high growth rate gave fewer scrutinyite grains, more of them (100)-oriented, suggesting (100) CoNb2O6 is the most robust surface for the metastable phase. TEM of a (100)-oriented grain in the 100 nm film showed columnar epitaxial growth with good atomic registry and a single set of diffraction spots, and the film lacked the threefold superstructure along [100] seen in the substrate, consistent with scrutinyite SnO2.
The observed ORs were the ones that both minimize misfit strain and give identical cation packing across the interface, while other low-strain alignments (for example scrutinyite {010} on columbite {012}) were not seen. Where strain and interface terms are similar for both phases, as on columbite {010}, rutile wins by its bulk stability; elsewhere, large strains and cation packing mismatch penalize rutile enough for scrutinyite to grow. The authors interpret orientation dependence through a terrace-ledge-kink picture in which any surface is built from low-index facets. The numerical strain values of Table 1 are not legible in the extracted text, so they are not reported here (Wittkamper et al. 2017).
Significance
This work shows that a polycrystalline, cation-ordered isostructural substrate can stabilize a metastable scrutinyite dioxide and maps, in one experiment, which substrate orientations do so. It supplies experimental counterparts to the bulk, strain and interface energy framework developed computationally by the same group, and identifies cation packing penalties as the key quantity for choosing substrates for scrutinyite targets such as IrO2 (Wittkamper et al. 2017).