Structure-property relationships and mobility optimization in sputtered La-doped $\mathrm{BaSn}{\mathrm{O}}_3$ films: Toward $100\mathrm{c}{\mathrm{m}}^2{\mathrm{V}}^{-1}{\mathrm{s}}^{-1}$ mobility

The wide band gap semiconducting perovskite $\mathrm{BaSn}{\mathrm{O}}_3$ is of high current interest due to outstanding room temperature mobility at high electron density, fueled by potential applications in oxide, transparent, and power electronics. Due in part to a lack of lattice-matched substrates, $\mathrm{BaSn}{\mathrm{O}}_3$ thin films suffer from high defect densities, however, limiting electron mobility. Additionally, the vast majority of $\mathrm{BaSn}{\mathrm{O}}_3$ thin film research has focused on pulsed laser deposition or molecular beam epitaxy. Here, we present an exhaustive optimization of the mobility of ${\mathrm{Ba}}_{0.98}{\mathrm{La}}_{0.02}\mathrm{Sn}{\mathrm{O}}_3$ films grown by a scalable, high-throughput method: high-pressure-oxygen sputter deposition. Considering target synthesis conditions, substrate selection, buffer layer structure, deposition temperature, deposition rate, thickness, and postdeposition annealing conditions, and by combining high-resolution x-ray diffraction, reciprocal space mapping, rocking curve analysis, scanning transmission electron microscopy, atomic force microscopy, and temperature-dependent electronic transport measurements, detailed understanding of synthesis-structure-property relationships is attained. Optimized room temperature mobility of $96\mathrm{c}{\mathrm{m}}^2{\mathrm{V}}^{\text{–}1}{\mathrm{s}}^{\text{–}1}$ is achieved in vacuum-annealed $\mathrm{GdSc}{\mathrm{O}}_3$(110)/$\mathrm{BaSn}{\mathrm{O}}_3$(120 nm)/${\mathrm{Ba}}_{0.98}{\mathrm{La}}_{0.02}\mathrm{Sn}{\mathrm{O}}_3$(200 nm) heterostructures, as well as $92\mathrm{c}{\mathrm{m}}^2{\mathrm{V}}^{\text{–}1}{\mathrm{s}}^{\text{–}1}$ on unbuffered substrates and $87\mathrm{c}{\mathrm{m}}^2{\mathrm{V}}^{\text{–}1}{\mathrm{s}}^{\text{–}1}$ without postdeposition annealing. These results, including important trends in defect densities and a surprising dependence of mobility on lattice mismatch, substantially expand the understanding of the interplay between deposition conditions, microstructure, and transport in doped $\mathrm{BaSn}{\mathrm{O}}_3$ films, establishing competitive mobilities in films fabricated via a scalable, high-throughput, industry-standard technique.

Figure 1

Representative powder x-ray diffraction pattern (blue) from a bulk polycrystalline ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) target sintered at 1450 °C. The red vertical lines depict the reference pattern [3]. (b) and (c) Sintering temperature ($T_{\mathrm{sinter}}$) dependence of the sputtering target lattice parameter ($a_{\mathrm{target}}$) and energy dispersive spectroscopy (EDS) Sn:Ba atomic ratio. The horizontal dotted lines are bulk literature values for polycrystalline ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ [2]. Blue shaded regions indicate the deduced optimal $T_{\mathrm{sinter}}\approx 1450{}^{\circ }\mathrm{C}$. Error bars in (b) represent standard errors on the intercepts of linear regressions of $a_{\mathrm{target}}$ vs $\mathrm{co}{\mathrm{s}}^2\left({\theta }_{hkl}\right)/\sin \left({\theta }_{hkl}\right)$. Error bars in (c) represent standard deviations of Sn and Ba atomic percentages from EDS.

Figure 2

(a), (c), and (e) High-resolution wide-angle specular x-ray diffraction around the 002 film and substrate reflections and (b), (d), and (f) corresponding 002 film rocking curves, from ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films sputtered at varied deposition temperature ($T_{\mathrm{dep}}$). Data are offset for clarity and are shown on MgO, $\mathrm{SrTi}{\mathrm{O}}_3$ (STO), and $\mathrm{LaAl}{\mathrm{O}}_3$ (LAO) substrates. Oxygen pressure ($P_{\mathrm{O}2}$), deposition rate (R), and thickness ($t$) were held constant at 1.9 Torr, $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$, and ∼80 nm, respectively. The vertical black dotted line in (a), (c), and (e) corresponds to the bulk lattice parameter. Small peaks marked with an asterisk are hybrid reflections [84]. $T_{\mathrm{dep}}$ dependence of (g) the out-of-plane lattice parameter ($a_{\mathrm{op}}$), (h) the Scherrer length-to-thickness ratio ($\mathrm{\Lambda }/t$), and (i) the 002 film rocking curve full width at half maximum (RC FWHM) on MgO, STO, LAO, and (${\mathrm{LaAlO}}_3{)}_{0.3}$(${\mathrm{Sr}}_2{\mathrm{TaAlO}}_6{)}_{0.7}$ LSAT substrates. The horizontal black dotted lines in (g) and (h) mark the bulk lattice parameter and $\mathrm{\Lambda }=t$, respectively. The colored dashed lines in (g), (h), and (i) are guides to the eye.

Figure 3

Deposition temperature ($T_{\mathrm{dep}}$) dependence of the 300 K (a) resistivity (ρ), (b) Hall electron density ($n$), and (c) Hall mobility (μ) for ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films on MgO, $\mathrm{LaAl}{\mathrm{O}}_3$ (LAO), $\mathrm{SrTi}{\mathrm{O}}_3$ (STO), and ${\left(\mathrm{LaAl}{\mathrm{O}}_3\right)}_{0.3}{\left({\mathrm{Sr}}_2\mathrm{TaAl}{\mathrm{O}}_6\right)}_{0.7}$ (LSAT) substrates. Oxygen pressure, deposition rate, and thickness were held constant at 1.9 Torr, $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$, and ∼80 nm, respectively. Temperature dependence of (d) ρ, (e) $n$, and (f) μ for the films on MgO in (a), (b), and (c), respectively. The horizontal black dotted line in (b) marks the nominal $n$ for full dopant activation. Colored dashed lines are guides to the eye.

Figure 4

(a), (c), and (e) High-resolution wide-angle specular x-ray diffraction around the 002 film and 002/220 substrate reflections and (b), (d), and (f) corresponding 002 film rocking curves, from ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films with various thickness ($t$). Data are offset for clarity and are shown on MgO, $\mathrm{GdSc}{\mathrm{O}}_3$ (GSO), and $\mathrm{SrTi}{\mathrm{O}}_3$ (STO) substrates. Deposition temperature ($T_{\mathrm{dep}}$), oxygen pressure ($P_{\mathrm{O}2}$), and deposition rate (R) were held constant at $900{}^{\circ }\mathrm{C}$, 1.9 Torr, and $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$, respectively. The vertical black dotted line in (a), (c), and (e) corresponds to the bulk lattice parameter. $t$ dependence of (g) the out-of-plane lattice parameter ($a_{\mathrm{op}}$), (h) the Scherrer length (Λ), and (i) the 002 film rocking curve full width at half maximum (RC FWHM) on MgO, STO, GSO, ${\mathrm{PrScO}}_3$ (PSO), and ${\mathrm{LaAlO}}_3$ (LAO) substrates (note the break on the RC FWHM axis). The black dotted lines in (g) and (h) mark the bulk lattice parameter and $\mathrm{\Lambda }=t$, respectively. The colored dashed lines in (g), (h), and (i) are guides to the eye.

Figure 5

Annular dark-field (ADF) scanning transmission electron microscopy (STEM) images of 200 nm thick ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films grown at 1.9 Torr oxygen pressure, 900 °C deposition temperature and $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$ deposition rate on $\mathrm{LaAl}{\mathrm{O}}_3$ (LAO), $\mathrm{SrTi}{\mathrm{O}}_3$ (STO), $\mathrm{GdSc}{\mathrm{O}}_3$ (GSO), and $\mathrm{PrSc}{\mathrm{O}}_3$ (PSO) substrates. (a) Cross-sectional high-angle ADF (HAADF; top) and low-angle ADF (LAADF) STEM images (bottom); scale bars are 100 nm. (b) Plan-view HAADF (left) and LAADF-STEM (right) images of the same films grown on LAO and PSO substrates; scale bars are 50 nm. In the LAADF images, yellow dots mark threading dislocations (TDs) of edge character, yellow arrows mark bright, high-strained regions (screw dislocations or grain boundary regions), and green arrows mark the partial voids discussed in the text. The insets show magnified regions around pairs of dissociated TDs (of edge character) connected via a short stacking fault.

Figure 6

Atomic-resolution cross-sectional high-angle annular dark-field (HAADF) scanning transmission electron microscopy (STEM) images of the substrate-film interfaces for the films shown in Fig. 5. These are 200 nm thick ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films grown at 1.9 Torr oxygen pressure, 900 °C deposition temperature, and $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$ deposition rate on $\mathrm{LaAl}{\mathrm{O}}_3$ (LAO), $\mathrm{SrTi}{\mathrm{O}}_3$ (STO), $\mathrm{GdSc}{\mathrm{O}}_3$ (GSO), and $\mathrm{PrSc}{\mathrm{O}}_3$ (PSO) substrates. Scale bars are 2 nm. Misfit dislocations (MDs) with in-plane and out-of-plane Burgers vector are marked yellow and red, respectively.

Figure 7

Thickness ($t$) dependence of the 300 K (a) resistivity (ρ), (b) Hall electron density ($n$), and (c) Hall mobility (μ) for ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films on MgO, $\mathrm{LaAl}{\mathrm{O}}_3$ (LAO), $\mathrm{SrTi}{\mathrm{O}}_3$ (STO), $\mathrm{GdSc}{\mathrm{O}}_3$ (GSO), and $\mathrm{PrSc}{\mathrm{O}}_3$ (PSO) substrates. Oxygen pressure, deposition temperature, and rate were held constant at 1.9 Torr, $900{}^{\circ }\mathrm{C}$, and $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$, respectively. The horizontal black dotted line in (b) marks the nominal $n$ for full dopant activation. Other black dashed lines are guides to the eye. (d) Lattice mismatch dependence of the 300 K mobility under the same growth conditions for 200 nm thick films. The relevant substrates (all perovskites) are marked.

Figure 8

Temperature (T) dependence of (a) resistivity (ρ), (b) Hall electron density ($n$), and (c) Hall mobility (μ), for ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films of various thickness ($t$) on representative $\mathrm{SrTi}{\mathrm{O}}_3$ (STO) substrates. Oxygen pressure, deposition temperature, and rate were held constant at 1.9 Torr, $900{}^{\circ }\mathrm{C}$, and $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$, respectively. Colored dashed lines are guides to the eye.

Figure 9

Deposition rate (R) dependence of (a) the out-of-plane lattice parameter ($a_{\mathrm{op}}$), (b) the Scherrer length-to-thickness ratio ($\mathrm{\Lambda }/t$), (c) the 002 rocking curve full width at half maximum (RC FWHM), (d) the 300 K resistivity (ρ), (e) the 300 K Hall electron density ($n$), and (f) the 300 K Hall mobility (μ) for ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) films on $\mathrm{LaAl}{\mathrm{O}}_3$ (LAO), $\mathrm{SrTi}{\mathrm{O}}_3$ (STO), $\mathrm{GdSc}{\mathrm{O}}_3$ (GSO), and $\mathrm{PrSc}{\mathrm{O}}_3$ (PSO) substrates. The deposition temperature ($T_{\mathrm{dep}}$), oxygen pressure ($P_{\mathrm{O}2}$), and thickness ($t$) were held constant at $900{}^{\circ }\mathrm{C}$, 1.9 Torr, and ∼35 nm, respectively. The horizontal black dashed lines in (a), (b), and (e) mark the bulk lattice parameter $\mathrm{\Lambda }=t$ and the nominal $n$ for full dopant activation, respectively. Colored dashed lines are guides to the eye. The arrows in the right panels indicate that lower R values resulted in much larger ρ, lower $n$, and lower μ, off scale in (d) and (e).

Figure 10

(a) High-resolution wide-angle specular x-ray diffraction around the 002 film and substrate reflections and (b) corresponding 002 film rocking curves from a ${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$ (BSO:La) film before and after vacuum annealing for 1 h at 750 °C. The vertical black dashed line in (a) corresponds to the bulk lattice parameter. All data in this figure are from films deposited on MgO at $900{}^{\circ }\mathrm{C}$, in 1.9 Torr of ${\mathrm{O}}_2$, at $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$, to ∼58 nm thickness. (c)–(f) Annealing temperature ($T_{\mathrm{anneal}}$) dependence of (c) the out-of-plane lattice parameter ($a_{\mathrm{op}}$), (d) the 002 film rocking curve full width at half maximum (RC FWHM), (e) the annealing-induced change in 300 K Hall electron density ($\mathrm{\Delta }n$), and (f) the annealing-induced change in 300 K Hall mobility (Δμ). Shown on the right axis in (f) is the annealing-induced change in Scherrer length-to-thickness ratio $\left(\mathrm{\Delta }\left[\mathrm{\Lambda }/t\right]\right)$. The horizontal black dashed lines in (c) and (d) represent average values for as-grown films; all other dashed lines are guides to the eye. Contact-mode atomic force microscopy (AFM) images of (g) an as-grown film, (h) a 750 °C annealed film, and (i) an 800 °C annealed film; scale bars are provided.

Figure 11

(a) High-resolution wide-angle specular x-ray diffraction around the 002 film and 220 substrate reflections and (b) 103 reciprocal space map from a representative $\mathrm{GdSc}{\mathrm{O}}_3/{\mathrm{La}}_{0.002}{\mathrm{Ba}}_{0.998}\mathrm{Sn}{\mathrm{O}}_3$ (270 nm)/${\mathrm{La}}_{0.02}{\mathrm{Ba}}_{0.98}\mathrm{Sn}{\mathrm{O}}_3$(200 nm) heterostructure after vacuum annealing at 650 °C for 15 min. The vertical black dashed line in (a) and the red X in (b) correspond to the bulk BSO:La lattice parameter. (c) and (d) 300 K Hall electron density ($n$) and mobility (μ) vs total film thickness ($t_{\mathrm{total}}=t_{\mathrm{buffer}}+t_{\mathrm{active}}$) for films on MgO and $\mathrm{GdSc}{\mathrm{O}}_3$ (GSO) substrates. Open points represent unbuffered films [as in Figs. 7 and 7] and solid points buffered films ($t_{\mathrm{buffer}}=100\text{–}270\mathrm{nm}$). (e) and (f) 300 K $n$ and μ vs $t_{\mathrm{buffer}}$ for constant $t_{\mathrm{active}}=200\mathrm{nm}$, again for films on MgO and GSO. Open and solid points represent as-grown and 650 °C, 15 min, vacuum-annealed films, respectively. All films in the figure were deposited at $900{}^{\circ }\mathrm{C}$ in 1.9 Torr of ${\mathrm{O}}_2$ at $\sim 1\mathrm{nm}\mathrm{mi}{\mathrm{n}}^{\text{–}1}$. The horizontal black dashed lines in (c) and (e) mark the nominal $n$ for full dopant activation; all other dashed and solid lines are guides to the eye.