Correlated Metals Transparent Conductors with High UV to Visible Transparency on Amorphous Substrates

Correlated metals with high carrier density and strongly correlated electron effects provide an alternative route to achieve transparent conducting materials, different from the conventional degenerately doped wide‐bandgap transparent conducting oxides (TCO). The extremely low electrical resistivity and high optical transparency in the ultraviolet‐visible spectral range shown in 4d correlated metals present an advantage over conventional TCOs. However, most of the 4d correlated metals are grown epitaxially on single crystal substrates. Here, it has been shown that Ca2Nb3O10 nanosheets with different buffer layers promote the growth of high‐quality 4d2 SrMoO3 films on fused silica substrates, overcoming the use of expensive and size‐limited single‐crystal substrates. The room temperature electrical resistivity of SrMoO3 is as low as 61 µΩ cm, the lowest reported value on amorphous transparent substrates to date, without compromising its high optical transmittance. 4d1 correlated metal SrNbO3 on Ca2Nb3O10 nanosheets also exhibits similarly high optical transmittance but a higher room temperature resistivity of 174 µΩ cm. These findings facilitate the use of highly conducting and transparent 4d correlated metals not only as TCOs on technologically relevant substrates for the applications in the ultraviolet‐visible spectral range but also as electrodes for other oxide‐based thin film technologies.

Correlated metals with high carrier density and strongly correlated electron effects provide an alternative route to achieve transparent conducting materials, different from the conventional degenerately doped wide-bandgap transparent conducting oxides (TCO). The extremely low electrical resistivity and high optical transparency in the ultraviolet-visible spectral range shown in 4d correlated metals present an advantage over conventional TCOs. However, most of the 4d correlated metals are grown epitaxially on single crystal substrates. Here, it has been shown that Ca 2 Nb 3 O 10 nanosheets with different buffer layers promote the growth of high-quality 4d 2 SrMoO 3 films on fused silica substrates, overcoming the use of expensive and size-limited singlecrystal substrates. The room temperature electrical resistivity of SrMoO 3 is as low as 61 µΩ cm, the lowest reported value on amorphous transparent substrates to date, without compromising its high optical transmittance. 4d 1 correlated metal SrNbO 3 on Ca 2 Nb 3 O 10 nanosheets also exhibits similarly high optical transmittance but a higher room temperature resistivity of 174 µΩ cm. These findings facilitate the use of highly conducting and transparent 4d correlated metals not only as TCOs on technologically relevant substrates for the applications in the ultraviolet-visible spectral range but also as electrodes for other oxide-based thin film technologies.

Introduction
The continuously increasing demand of high-performance transparent conductors with broadband transparency and with cost-effective materials resulted from the fast-growing energy and electronic markets of photovoltaics and display technologies. [1,2] Regarding oxide materials, the widely used design optical resistivity values of 260 and 274 µΩ cm, respectively, and the high optical transmittance of about 75% at 550 nm (2.25 eV). [27] We also observed similar resistivity values of SVO on CNO nanosheets [27] due to the low growth temperature and/or the influence of the substrate materials (see SrVO 3 note in Supporting Information). Furthermore, it was revealed that 4d correlated metals SrMoO 3 (SMO) and SrNbO 3 (SNO) have higher UV-vis transmittance compared to 3d SVO due to interband transitions occurring at higher photon energies of 4.3-4.8 eV, [9,14,17,19] while these three correlated metals on single crystal substrates were reported to have similar resistivity values of around 30 µΩ cm. [10,11,28] It is worth noting that the use of SrTiO 3 (STO) and KTaO 3 single crystal substrates at the reduced growth conditions of these correlated metals may affect the measurements of their intrinsic resistivity because the substrates may become conducting. [29] In this study, we showed that high-quality SMO films on any substrates were obtained using CNO nanosheets with different buffer layers, affecting the lattice mismatch and SMO's wettability. These SMO films with the RT electrical resistivity of 64 ± 4 µΩ cm are to our knowledge the lowest values reported for conducting transition metal oxide films on amorphous transparent substrates so far, almost on par with those on single crystal substrates, [9][10][11][13][14][15][16][17][18][19] without compromising the high optical transmittance. Similarly, highly conducting and transparent SNO films were also realized on amorphous transparent fused silica substrates using only CNO nanosheets. The observed performance of both compounds on industrially relevant substrates, normally only limited to expensive and size-limited single crystal substrates, is expected to bring these correlated metal transparent electrodes a step closer to practical applications.

The Growth of 4d Correlated Metal Films on Ca 2 Nb 3 O 10 Nanosheets
To epitaxially grow high-quality thin films, one would choose single crystal substrates with similar crystal structures and small lattice mismatch with the grown thin films. In this study, we used CNO nanosheets deposited on amorphous fused silica substrates, unless otherwise stated, where individual CNO nanosheets can be considered as micro-sized single crystal perovskite substrates with the square 2D lattice constants of 3.85 Å. [30] The schematic illustration of SMO film/10 unit-cells of STO buffer layer/CNO nanosheets/fused silica is presented in Figure 1. SMO has the bulk lattice constants of 3.975 Å, [31] giving rise to a lattice mismatch with CNO nanosheets of 3.2%. We observed that the direct growth of SMO on fused silica or on CNO nanosheets/fused silica resulted in poorly crystallized, polycrystalline thin films (see Figure S1, Supporting Information).
On the other hand, it is worth mentioning here that the direct growth of SNO, which has the bulk lattice constants of 4.024 Å, [32] on CNO nanosheets formed highly crystalline single out-of-plane oriented-(001) thin films (see Figure S2, Supporting Information) even though the lattice mismatch of SNO with CNO nanosheets is 4.4%, larger than that of SMO. This is similar to reports of (La x Ba 1-x )SnO 3 grown on lattice mismatch STO. [33] Apparently, when the lattice mismatch is such that the film material grows relaxed, crystallinity is not necessarily hampered.
To resolve the poor wettability of SMO on CNO nanosheets (see Figure S1, Supporting Information) and its favorable wettability on STO reported earlier, [11,34,35] we introduced a 10 unitcells, or 4 nm, STO buffer layer grown on the CNO nanosheets. With this, we achieved the growth of highly crystallized single out-of-plane oriented-(001) SMO thin films (Figure 1b), confirmed by the out-of-plane lattice constant of 3.97 Å (close to the bulk value of SMO) calculated from reflections of SMO in Figure 1b. In the inset of Figure 1b, a spotty reflection highenergy electron diffraction (RHEED) pattern, which was collected in situ during pulsed laser deposition (PLD), indicates the 3D island growth of highly crystallized SMO, which was consistent with its observed surface morphology (see Figure S1, Supporting Information). Electron backscatter diffraction (EBSD) data in Figure 1c-e shows that SMO was epitaxially grown on individual CNO nanosheets, but there was no long-range coherence in the in-plane orientation among SMO on different CNO nanosheets due to the random in-plane distribution of CNO nanosheets during the Langmuir-Blodgett deposition of the nanosheets [25] on fused silica substrates.
As mentioned above, highly crystalline single out-of-plane oriented-(001) SNO films were achieved on CNO nanosheets without the STO buffer layer thanks to the excellent wettability of SNO on CNO nanosheets (see Figure S2, Supporting Information). To confirm the role of the nanosheets, Figures S1 and S2, Supporting Information showed that direct growth of SMO and SNO on fused silica substrates only formed poorly crystallized polycrystalline thin films. The introduction of CNO nanosheets with and without buffer layers on amorphous fused silica substrates, denoted as CNO-based substrates, is therefore needed for the growth of SMO and SNO perovskite films. This indicates that the growth is mainly driven by the initial crystallization of the SMO and SNO at the interface with the CNO nanosheets, having a small lattice-mismatch. Introducing STO as a buffer layer for the SMO growth results in high-quality SMO films due to an improved lattice-match and improved wettability as explained above.

Transport Properties
It is well-known that the physical properties of thin films are strongly affected by the film crystal quality. Figure 2 shows the electrical transport properties of SMO thin films grown on different buffer layers on CNO nanosheets. In Figure 2a, the 40 nm SMO film is grown on 10 unit-cells STO on CNO nanosheets exhibited the temperature-dependent resistivity of a typical correlated metal with the critical temperature T* = 128 K consistent with earlier reported values. [36] Below this T*, the electron-electron scattering is the dominant mechanism shown by the T 2 dependence. The RT resistivity of 63 µΩ cm for transition metal oxide thin films on amorphous substrates is only a factor of two higher than the best SMO reported value of 30 µΩ cm on single crystal substrates www.advmatinterfaces.de with a similar film thickness (≈40 nm). [11,29] However, as the film thickness decreased, the RT resistivity of SMO films on 10 unit-cells STO on CNO nanosheets exponentially increased (Figure 2b), i.e., to 716 µΩ cm for the 12 nm-thick film. This was a direct consequence of the significant decrease of carrier mobility, µ, measured from 3.9 cm 2 V −1 s −1 to 0.3 cm 2 V −1 s −1 with the average measured carrier density, N, of 2.5 × 10 22 cm −3 (Figure 2c,d).
To investigate the role of the lattice mismatch on the conductivity of SMO, we introduced two thick buffer layers, 100 nm DyScO 3 (DSO) with the bulk pseudocubic lattice constant of 3.94 Å, and 100 nm STO with a bulk lattice constant of 3.905 Å, on CNO nanosheets. The RT resistivity values of 12 nm SMO films decreased to 386 and 127 µΩ cm, respectively. Apparently, the effect of the electron surface scattering contributed to the increase of the resistivity as the films got thinner on three different buffer layers. However, thin films grown on larger lattice mismatch substrates generally are expected to have a higher defect density, such as point defects and dislocations, which reduces the carrier mobility. We observed that the resistivity of SMO was lower on DSO buffer layers than on 10 unit-cells STO buffer layers. It has been shown that SMO has a better wettability on STO compared to Sc-based substrates, resulting in a different microstructure and thus higher carrier mobility for the grown films on STO. [11,34,35] We grew a sample with an additional 10 unit-cells STO on 100 nm DSO buffer layer. The 12 nm SMO film had a resistivity of 237 µΩ cm, lower than that on only 100 nm DSO buffer layer because of the higher carrier mobility. Apparently, both the lattice mismatch and the wettability of SMO on the buffer layers play important roles in determining the electrical properties of SMO. However, we observed that in the thickness range of less than 20 nm, SMO films had e) crystal structure identification of 40 nm SMO film on 10 unit-cells of STO buffer layer/CNO nanosheets/fused silica substrate using diffraction techniques: b) the symmetrical X-ray diffraction (XRD) scan shows the (00l) reflections of the SMO film. The inset is the spotty RHEED pattern of the SMO film, indicating the crystalline thin film formed by a 3D island growth mode. In EBSD, the band contrast image c) exhibits the boundaries of individual CNO nanosheets on which the SMO film was grown, while the inverse pole figure maps of the out-of-plane d) and in-plane e) directions indicate the epitaxial growth of SMO on individual CNO nanosheets. Although the SMO film formed the single out-of-plane (001)-orientation, its in-plane orientation is random among different CNO nanosheets.
www.advmatinterfaces.de the lowest resistivity on 100 nm STO buffer layers compared with those on the other two buffer layers. Especially, for the film thickness of 12 nm, it had the resistivity lower than the one on additional 10 unit-cells STO on 100 nm DSO buffer layer even though it was expected that the 10 unit-cell STO on 100 nm DSO buffer layer had smaller lattice mismatch with SMO than that of 100 nm STO buffer layer because of the epitaxial strain to 100 nm DSO buffer layer, and SMO had the same wettability on both STO surfaces. To investigate any role of the conductivity of the buffer layer itself, a control sample of 100 nm STO layer was grown on CNO nanosheets under the same conditions as 100 nm STO buffer layers for SMO thin films and was kept under the reduced growth conditions of 40 nm SMO for 1 h without an actual SMO deposition. The 100 nm STO layer was extremely insulating with the RT resistance value out of the measurement range of our physical property measurement system even though STO single crystal substrates were reported to become conducting in the similar reduced conditions. [29] Yet, the contribution of a conducting STO layer with a thickness higher than 10 unit-cells might be possible, because of the risk of Mo doping through inter diffusion, instead of oxygen vacancies, at the interface of 100 nm STO layer. However, when the SMO film thickness was 40 nm, the resistivity values of SMO on three buffer layers were 64 ± 4 µΩ cm, suggesting the lattice mismatch and wettability of SMO on buffer layers and the possible conducting STO layer had minor effects on the resistivity of thick SMO films.
Regarding SNO films on CNO nanosheets, the 50 nm SNO film had a typically T 2 -dependent resistivity of a correlated metal with the critical temperature T* = 183 K (see Figure S3, Supporting Information), consistent with the reported value in literature. [28] As the film thickness decreased in the range of 12-50 nm, the RT resistivity increased from 174 to 270 µΩ cm ( Figure S3, Supporting Information) due to the larger electron surface scattering contribution to the film resistivity for thinner films. In contrast with SMO, the exponential increase of resistivity was not observed as SNO got thinner even on only CNO nanosheets with the large lattice mismatch of 4.4% probably thanks to the excellent wettability of SNO.
We note that the reported resistivity values of SMO, which were grown on various single crystal substrates at different growth conditions, vary greatly from 18 to 127 µΩ cm in the thickness range of 40-50 nm. [9,14,19,28] Here, we grew SMO on a single crystal DSO substrate at the same growth conditions as the one on 100 nm DSO/10unit-cells STO/CNO nanosheets/fused silica substrate with the same thickness. The former SMO film had the resistivity value of 53 µΩ cm, lower than that of the latter, 61 µΩ.cm, while both had the similar residual resistivity ratio ρ 300K/2K of 1.9 (see Figure S4, Supporting Information), comparable to the reported values on single crystal substrates. [14,29] The carrier mobility and carrier density values of the former SMO were 2.7 cm 2 V −1 s −1 and 4.3 × 10 22 cm −3 , whereas those of the latter were 3.2 cm 2 V −1 s −1 and 3.1 × 10 22 cm −3 , respectively. Similar trends in the resistivity, carrier mobility, and carrier density

www.advmatinterfaces.de
were observed for SNO on single crystal DSO substrate and 100 nm DSO/10 unit-cells STO/CNO nanosheets/fused silica substrate (see Figure S4 in SI). It has been suggested that the strain imposed by single crystal substrates on oxide thin films favors the formation of oxygen vacancies. [15,37] From XRD data (see Figure S5, Supporting Information), the out-of-plane lattice constant values of SMO and SNO on single crystal DSO substrates were 4.00 and 4.10 Å, larger than their bulk values of 3.975 and 4.024 Å, respectively, due to the in-plane compressive strain caused by the lattice mismatch with DSO. On 100 nm DSO buffer layers, the out-of-plane lattice constant value of SNO was 4.02 Å, while that of SMO shifted toward the bulk value (see Figure S5, Supporting Information). Apparently, the degree of strain imposed on SMO and SNO was higher on DSO single crystal substrate than on 100 nm DSO on CNO-based substrates. We hypothesize that the strain could provide additional carriers from oxygen vacancies for SMO and SNO on single crystal DSO substrates, resulting in higher free carrier densities (more pronounced effect on SMO) for the films deposited on the single crystal DSO ( Figure S4, Supporting Information). In addition, one would expect that the randomness of the in-plane orientation of SMO and SNO films and their boundaries between CNO nanosheets (Figure 1c-e) would lead to the lower carrier mobility compared to the ones on single crystal substrates. But the resulting mobility values for the films on 100 nm DSO/10 unit-cells STO/CNO and on single crystal DSO do not show significant difference ( Figure S4, Supporting Information). This suggests that the higher carrier density of SMO and SNO lead to the lower carrier mobility, which is opposite to the trend observed in SVO. [15] A further study on the film transport properties of SMO and SNO on individual CNO nanosheets could elucidate the effects of the random in-plane orientations and boundaries between CNO nanosheets on the overall mobility of the films.
The sheet resistance values in the range of 15-125 Ω/square of SMO and SNO on CNO-based substrates for film thicknesses of more than 20 nm (Figure 2e) are comparable to those of other high-performance transparent conductors in the visible spectrum [2,3] and are about two orders of magnitude lower than those of UV transparent conductors. [5,38] In contrast, poorly crystallized SMO, SNO without buffered CNO nanosheets and even SMO with buffered CNO nanosheets on fused silica substrates had extremely high RT sheet resistance values.

Optical Properties
Besides the requirement of low sheet resistance (Figure 2e), 4d correlated metals are required to have high optical transmittance in the UV-visible range for transparent electrode applications. Figure 3a,b,c shows the transmittance of SMO on CNO-based fused silica substrates in the photon energy range of 0.5-4.5 eV. As shown in Figure 3a, SMO films on 10 unit cells STO on CNO nanosheets on fused silica substrates are highly transparent in the visible range with the transmittance of 82%-62% at 2.25 eV (550 nm). This is due to the weak t 2gt 2g intraband and t 2g -e g interband transitions in the region 2.5-3.5 eV. [39] Moreover, the films are moderately transparent  Panels (a,b,c) show the optical transmittance of SMO films on 10 unit-cells STO, 100 nm DSO/10 unit-cells STO, and 100 nm STO buffer layers on CNO nanosheets, respectively, in the range of 0.5-4.5 eV. The performance in the visible (d) and UV (e) spectra of SMO and SNO films were compared with SNO, SVO, and ITO films in this work and the literature. [4,10,19,42] Note that the transmittance values of ITO were calculated using data from. [4,42] www.advmatinterfaces.de in the UV range with the transmittance of 45%-30% at 4.5 eV (275 nm) mainly due to the absorption of 10 unit cells STO at the photon energies of 3.25 and 3.75 eV [40] (see Figure S6 intra band) and the interband transition from O 2p to Mo 4d t 2g at the photon energy of 4.3 eV. [14] Likewise, the transmittance values of SNO films were similar to those of the SMO films in the visible range but were slightly higher in the UV range simply because no STO buffer layer was used and the strong interband transition from O 2p to Nb 4d e g occurs at the photon energy of 4.8 eV [9] (see Figure S7, Supporting Information). The plasma frequency of SMO was determined to be 1.97-2.10 eV from the point of the inflection in the reflectance spectra (see Figure S8, Supporting Information), which is consistent with the reported value. [11] Note that the insertion of thick buffer layers, i.e., 100 nm STO or 100 nm DSO, modified the transmittance spectra of SMO film in the whole photon energy range due to the multiple reflection and interference effects. In addition, the inserted buffer layers lead to higher absorptance because of their band gaps (see Figure S6, Supporting Information). Moreover, it is worth noting that the averaged transmittance of SMO and SNO on CNO-based fused silica substrates is 5%-10% higher than those on single crystal substrates [19] with the similar film thickness mainly due to the higher transmittance of bare fused silica substrates.
To have a proper evaluation of the performance of correlated metals, the figure of merit (FOM) proposed by Haacke [41] FOM = T 10 /R s is often used to examine both sheet resistance R s and averaged optical transmittance T over a certain photon energy range. It has been shown that in the visible range and/ or UV range, single crystal SVO, SNO, and SMO films outperform conventional ITO with higher FOM values in the so-called freestanding forms, whose transmittance is normalized to the transmittance of substrates [11] or is calculated from ellipsometry data to remove the contributions from the substrates and film thickness-dependent effects. [9,10] In this study, we normalized the averaged optical transmittance to calculate the FOM values of the correlated metal oxide films. We noted that the averaged optical transmittance of SMO films in the visible spectrum -normalized to the averaged transmittance of the buffer layers/ CNO nanosheets/fused silica substrates-had similar values on the different buffer layers (see Table S1, Supporting Information). Meanwhile, the sheet resistance of SMO was lowest on 100 nm STO buffer layers (Figure 2e), so 100 nm STO buffer layers had negligible influence on the optical transmittance of SMO films in the visible spectrum. Figure 3d shows the FOM values of the so-called freestanding SMO and SNO grown on CNO nanosheets/fused silica substrates in comparison with other reported data from the literature in the visible range of 1.59-3.26 eV (380 nm-780 nm). We also deposited SVO films on CNO nanosheets on fused silica to have a complete evaluation of the performance of correlated metals as TCOs (see SrVO 3 note in Supporting Information). SMO films on 100 nm STO buffer layers/ CNO nanosheets/ fused silica had FOM values comparable to those of SVO films grown on LSAT single crystal substrates using molecular beam epitaxy (MBE) and outperform ITO films. Meanwhile, SMO films on other buffer layers and SNO films on CNO nanosheets had lower FOM values in the range of 0.4 × 10 −3 -2 × 10 −3 . The high FOM of our SMO/100 nm STO/CNO nanosheets/ fused silica transparent electrode demonstrates that the use of expensive single crystal substrates for the growth of SMO and SNO can be overcome by using CNO nanosheets on transparent amorphous fused silica. We noted that the FOM values of SVO films grown on CNO nanosheets were lower than those of SVO films grown on single crystals by MBE. That is not consistent with the reported data in reference [27] even though SVO films in this study had lower resistivity than the reported data [27] and similar transmittance values (see SrVO 3 note in SI). Furthermore, Figure 3e shows FOM values of SMO and SNO in comparison with other films reported in the literature in the UV spectrum of 3.85-4.5 eV. SMO films on CNO-based substrates with the high FOM values outperform SNO, SVO, and ITO films thanks to their high averaged optical transmittance in the UV spectrum and low sheet resistance (see Table S1, Supporting Information). SNO films have FOM values as high as ITO films. It is worth noting the thickness-dependent effects on the optical transmittance of the correlated metals (see Figure S9, Supporting Information). These interference effects can be tuned with the oxide buffer or correlated metal thickness to match specific application requirements.

Conclusion
We have shown the potential of 4d correlated metals SMO and SNO as TCOs in the UV-visible spectra on technologically relevant transparent substrates. In addition to the essential role of CNO nanosheets, the intrinsic structural and chemical properties of 4d correlated metals, such as wettability and lattice match, determine whether additional buffer layers are needed for the growth of high-quality thin films. The extremely low resistivity and high optical transmittance of SMO and SNO on amorphous transparent substrates offer the high performance in the UV-visible spectra in terms of FOM, comparable to the ones on single crystal substrates and outperforming the ITO films. It is worth noting that the high-temperature growth of 4d metallic correlated oxides in this study is suitable for back electrode applications where the electrodes are directly contacted with the supporting substrates. Using CNO nanosheets and different buffer layers, 4d correlated metal SMO and SNO films on Si substrates had the same properties in terms of crystal structure and resistivity as the ones on fused silica substrates. Our findings also facilitate the use of highly conducting 4d correlated metals as electrodes on application-relevant substrates for other oxide-based thin film technologies.

Experimental Section
Preparation of CNO Nanosheets: Anhydrous potassium carbonate K 2 CO 3 (Fluka), calcium carbonate CaCO 3 (Sigma-Aldrich), and niobium(V) oxide Nb 2 O 5 (Sigma-Aldrich) had a purity of 99.0% or higher and were used as received. Nitric acid HNO 3 (65%, ACROS Organics) and tetra-n-butylammonium hydroxide (TBAOH) (40 wt. % H2O, Alfa Aesar) were used as received. Demineralized water was used throughout the experiments. A stoichiometric mixture of K 2 CO 3 , CaCO 3 , and Nb 2 O 5 was used in the solid-state synthesis. The following temperature program was used: heating to 1150 °C at 3 °C min −1 , holding for 24 h, and then cooling down to room temperature at 5 °C min −1 .

www.advmatinterfaces.de
The obtained KCa 2 Nb 3 O 10 (KCNO) powders were protonated in 5 M of HNO 3 solution (250 mL) while stirring at RT. The acid solution was replaced daily for three days. Subsequently, the powders were filtered and washed with 1 L of water and then air-dried overnight at RT to obtain HCa 2 Nb 3 O 10 (HCNO) powders. The exfoliation of HCNO powders in TBAOH with a molar ratio HCNO/TBAOH of 1:2 was carried out. HCNO powders (0.1 g) was stirred in a TBAOH solution (10 mL) for 3 h to obtain a CNO nanosheet solution. Using the obtained CNO nanosheet solution, CNO nanosheets were deposited on fused silica and Si substrates with almost fully coverage using the LB method. [25] Pulsed Laser Deposition of SrMoO 3 and SrNbO 3 : PLD was performed in a vacuum system with a base pressure of 2 × 10 −8 mbar, equipped with a (krypton fluoride) KrF excimer laser of 248 nm (COMPexPro from Coherent Inc.) and in situ RHEED. Stoichiometric SrMoO 4 (homemade) and Sr 2 Nb 2 O 7 (from Praxair Inc.) targets were used. The growth temperature of 700 °C, substrate surface-target distance of 50 mm, spot size of 1.8 mm 2 , fluence of 1.3 J/cm 2 , and laser-frequency of 1 Hz were used for 10 unit-cells STO, SMO, and SNO. The same parameters were used for 100 nm DSO and 100 nm STO but the laser frequency was 3 Hz. STO and DSO were deposited in O 2 pressure of 10 −2 mbar and 10 −3 mbar, respectively. SMO and SNO were deposited in vacuum with the base pressure 2 × 10 −6 mbar at 700 °C without background gas. PLD growth of SVO films was described in SrVO 3 note in Supporting Information.
Characterization and Analysis: Thin film X-ray diffraction (XRD) data were measured using PANalytical X'Pert Pro MRD Inc. beam Monochr. 4xGe220 Cu asym. LF monochromator to select Cu Kα 1 radiation. CNO nanosheets, SMO, and SNO on supporting substrates were investigated by atomic force microscopy (AFM), Bruker Dimension ICON, operating in tapping mode. The average coverage area of CNO nanosheets on supporting substrates was 95%-97%, which was calculated from AFM data at three different locations. The film thickness of SMO, SNO, and SVO films was determined using cross-section high-resolution scanning electron microscopy (HR-SEM) ( Figure S10, Supporting Information) and EBSD was carried out by Merlin field emission microscopy (Zeiss 1550) equipped with an angle-selective backscatter detector. The transport measurements were performed in the 4-probe van der Pauw configuration using a Quantum Design Physical Properties Measurement System Dynacool. Normal incidence unpolarized transmission and reflectance measurements were performed using a Perkin-Elmer Lambda 950 UV-Vis-NIR spectrophotometer in the wavelength range from 250 to 2500 nm with the step of 5 nm. The detected transmitted and reflected intensities were normalized against the detected intensity without a sample in the optical path.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.