3.3.2. XPS

Detailed XPS spectra close to the Co 2p binding energies were collected to identify the chemical state of cobalt in the films (Figure 4). Previous reports of ALD-grown cobalt oxide using β-diketonates and ozone indicated a mixed 2+/3<sup>+</sup> valence. The presence of Co2+ could indicate detrimental inclusions of Co3O4 in the grown films. Co2+ can be identified by an intense shake-up satellite feature at around 786 eV, whereas the Co3+ satellite is shifted towards 790 eV. Currently, we only observed Co3+ satellite features, indicating that the films are dominated by GdCoO3. Based on the data and the complexity of Co XPS, however, we cannot rule out that some Co2+ is present in the films. Survey spectra and detailed scans of O 1 s and C 1 s can be found in the Supplementary Materials (Figure S3).

**Figure 4.** Co 2p XPS of 30 nm GdCoO3 thin films on LaAlO3 (100), with no observation of Co2+. The black line is the recorded data, the green line is the background, and the light blue line is the total fit.

#### 3.3.3. Deposition on a High-Aspect-Ratio Substrate

The ability to deposit catalytically active complex oxides on high aspect ratios is of high importance. This can, e.g., enable the coating of mesoporous γ-alumina, and thereby provide catalysts with a significantly enhanced surface area compared to nanoparticles (10–100 nm) obtained from wet chemical synthesis and/or ball milling. The conformality of the two gadolinium cobaltite-based films was investigated by applying the presented deposition process onto high-aspect-ratio substrates. Figure 5 shows cross section images of a GdCoO3 film deposited on a high-aspect-ratio trench Si wafer. The film is conformally deposited on all surfaces of the substrate. As shown in Figure 5b, the bottom of the trench is characterized by the presence of agglomerates, possibly resulting from turbulence during growth and/or the preparation of cross section SEM samples.

## **4. Conclusions**

We have developed an ALD process for crystalline and homogeneous Gd-Ca-Co-O thin films on three different substrates, which provides a route towards coatings with potential application within catalysis. The good crystallinity of the obtained films gives insight into the crystal structure of the product and orientation of crystallization, which is essential since catalytic performance depends on key parameters connected with the structure, chemistry, and electronic states of exposed surfaces. These features are shown to be tuned by appropriate choices of ALD precursors, substrates, and deposition/annealing conditions. The proof of concept of depositing conformal Gd-Ca-Co-O films on high-aspect-ratio substrates is important, since practical applications in catalysis would require high surface areas.

The gadolinium cobaltite-based catalysts represent a particularly interesting system, not only with respect to catalysis, but also to physical properties. For instance, an expanded lattice triggered by the substrate may possibly stabilize a high-spin Co(III) state. Such scenarios sugges<sup>t</sup> that ALD can be used to tune resulting properties by means of appropriate lattice-matching substrates.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1996-1944/13/1/24/s1: Figure S1: XRD patterns of 30 nm Gd0.9Ca0.1CoO3 films grown on (a) YAP(100) and (b) YAP(001), post-annealed for 30 minutes at 650 ◦C. Figure S2: XRD pattern of the GdCoO3 (040) reflection of as deposited (black) and annealed (green) 30 nm films grown on LAO (100)pc, used for Scherrer analysis of crystallite size. Figure S3: (a) XPS of C 1s, (b) XPS of O 1s and (c) Survey spectra showing identification of Gd, Co, O and carbon species.

**Author Contributions:** Investigation, M.D. and H.H.S.; formal analysis, M.D. and H.H.S.; methodology, H.H.S. and O.N.; project administration, O.N., A.O.S. and H.F.; writing—original draft preparation, M.D.; writing—review and editing, M.D., H.H.S., O.N., A.O.S. and H.F.; supervision, A.O.S. and H.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This project received financial support from the Research Council of Norway via the ASCAT project (contract no. 247753).

**Acknowledgments:** The authors acknowledge the Department of Geology at the University of Oslo for access to XRF instrumentation. Henrik H. Sønsteby acknowledges the Research Council of Norway for funding via the RIDSEM project (contract no. 272253). In addition, they thank Jon Einar Bratvold for his assistance with the ALD reactor, Kristian Weibye for recording XPS spectra, and Martin Jensen for recording SEM images.

**Conflicts of Interest:** The authors declare no conflict of interest.
