Structural phase transition in epitaxial perovskite films

Three different film systems have been systematically investigated to understand the effects of strain and substrate constraint on the phase transitions of perovskite films. In $\mathrm{Sr}\mathrm{Ti}{\mathrm{O}}_3$ films, the phase transition temperature $T_c$ was determined by monitoring the superlattice peaks associated with rotations of $\mathrm{Ti}{\mathrm{O}}_6$ octahedra. It is found that $T_c$ depends on both $\mathrm{Sr}\mathrm{Ti}{\mathrm{O}}_3$ film thickness and $\mathrm{Sr}\mathrm{Ru}{\mathrm{O}}_3$ buffer layer thickness. However, lattice parameter measurements showed no sign of the phase transitions, indicating that the tetragonality of the $\mathrm{Sr}\mathrm{Ti}{\mathrm{O}}_3$ unit cells was no longer a good order parameter. This signals a change in the nature of this phase transition, the internal degree of freedom is decoupled from the external degree of freedom. The phase transitions occur even without lattice relaxation through domain formation. In $\mathrm{Nd}\mathrm{Ni}{\mathrm{O}}_3$ thin films, it is found that the in-plane lattice parameters were clamped by the substrate, while the out-of-plane lattice constant varied to accommodate the volume change across the phase transition. This shows that substrate constraint is an important parameter for epitaxial film systems, and is responsible for the suppression of external structural change in $\mathrm{Sr}\mathrm{Ti}{\mathrm{O}}_3$ and $\mathrm{Nd}\mathrm{Ni}{\mathrm{O}}_3$ films. However, in $\mathrm{Sr}\mathrm{Ru}{\mathrm{O}}_3$ films we observed domain formation at elevated temperature through x-ray reciprocal space mapping. This indicated that internal strain energy within films also played an important role, and may dominate in some film systems. The final strain states within epitaxial films were the result of competition between multiple mechanisms and may not be described by a single parameter.

Figure 1

Raw XRD scan for (0 0 2) peaks on $200\text{−}\mathrm{nm}$ $\mathrm{STO}∕350\text{−}\mathrm{nm}$ SRO/LAO sample, showing well resolved peaks. Photon energy is $10\mathrm{keV}$.

Figure 2

Temperature dependence of ($1∕2$, $1∕2$, $7∕2$) STO superlattice peak for $1\text{−}\mu \mathrm{m}$ $\mathrm{STO}∕350\text{−}\mathrm{nm}$ STO/LAO sample. The other peak is from the SRO layer.

Figure 3

Temperature dependence of superlattice peak intensity on STO film thickness. Bulk value from Ref. 19.

Figure 4

Temperature dependence of superlattice peak intensity on samples with different SRO thickness. Bulk value from Ref. 19.

Figure 5

Temperature dependence of superlattice peak intensity, in-plane, and out-of-plane lattice parameters of STO films. Dashed lines indicate $T_c$. LAO and STO bulk data (Ref. 20) are shown for comparison.

Figure 6

Out-of-plane and in-plane lattice parameters of $\mathrm{Nd}\mathrm{Ni}{\mathrm{O}}_3$ films and LAO substrates during heating up, shown in pseudocubic notation.

Figure 7

Temperature dependence of $\mathrm{Nd}\mathrm{Ni}{\mathrm{O}}_3$ unit-cell volume. Bulk data from Ref. 21.

Figure 8

Resistivity of $\mathrm{Nd}\mathrm{Ni}{\mathrm{O}}_3$ films.

Figure 9

Reciprocal space mapping shows SRO forming domain structure. The lower peaks are SRO ${\left(206\right)}_o$ and ${\left(026\right)}_o$. The upper weak peaks are from STO ${\left(113\right)}_c$. The unit is in reciprocal lattice unit of SRO.

Figure 10

Temperature dependence of SRO lattice parameters.

Figure 11

Order parameters for phase transitions in STO bulk and films, (a) superlattice peak intensity in bulk, (b) tetragonality in bulk, (c) superlattice peak intensity, and (d) tetragonality in $200\text{−}\mathrm{nm}$ $\mathrm{STO}∕50\text{−}\mathrm{nm}$ SRO/LAO sample. Dashed lines indicate $T_c$. Bulk data from Refs. 19, 27.

Figure 12

Lattice parameters for STO films of different thickness on $350\text{−}\mathrm{nm}$ SRO layers. Both experiment data (a) and calculated value (b) are shown. From Ref. 16.

Figure 13

Lattice parameters for $200\text{−}\mathrm{nm}$ STO films on SRO buffer layers of varying thickness. (a) Experiment results; inset is the detail of thin SRO region. (b) Calculated STO lattice parameters. From Ref. 16.

Figure 14

Dependence of $T_c$ on (a) in-plane strain, (b) tetragonality.