Emergent interface vibrational structure of oxide superlattices

Abstract

As the length scales of materials decrease, the heterogeneities associated with interfaces become almost as important as the surrounding materials. This has led to extensive studies of emergent electronic and magnetic interface properties in superlattices^1-9. However, the interfacial vibrations that affect the phonon-mediated properties, such as thermal conductivity^10,11, are measured using macroscopic techniques that lack spatial resolution. Although it is accepted that intrinsic phonons change near boundaries^12,13, the physical mechanisms and length scales through which interfacial effects influence materials remain unclear. Here we demonstrate the localized vibrational response of interfaces in strontium titanate-calcium titanate superlattices by combining advanced scanning transmission electron microscopy imaging and spectroscopy, density functional theory calculations and ultrafast optical spectroscopy. Structurally diffuse interfaces that bridge the bounding materials are observed and this local structure creates phonon modes that determine the global response of the superlattice once the spacing of the interfaces approaches the phonon spatial extent. Our results provide direct visualization of the progression of the local atomic structure and interface vibrations as they come to determine the vibrational response of an entire superlattice. Direct observation of such local atomic and vibrational phenomena demonstrates that their spatial extent needs to be quantified to understand macroscopic behaviour. Tailoring interfaces, and knowing their local vibrational response, provides a means of pursuing designer solids with emergent infrared and thermal responses.

Fig. 1. Period-dependent changes in the symmetry of STO–CTO superlattices.

a, Superlattice structures calculated from DFT with coloured-bar schematics denoting the chemically (left) and structurally (right) defined interfaces. Here green, blue and cyan rectangles correspond to STO, CTO and interface layers, respectively; the same colours are used in e, f, h, i, k, l. Green, blue, grey and red circles in a, e, f, h, i, k correspond to Sr, Ca, Ti and O atoms, respectively. b–d, The [100] zone-axis SADP for SL27 (b), SL4 (c) and SL2 (d) grown on NGO. The coloured arrows correspond to ordered reflections from the three possible domains. The solid arrows indicate ordered reflections that exist and the hollow arrows indicate absences. Insets: ball-and-stick models of the orientations present with border colours matching the arrows. The red and blue arrows and insets are viewed along an out-of-phase tilt axis and the yellow are viewed along an in-phase tilt axis. In c, d, superlattice reflections are seen in the 001 direction. In b, closely spaced superlattice reflections appear as streaking of the fundamental reflections. e–m, ADF images (e, h, k), iDPC images (f, i, l) and octahedral tilt angles (g, j, m) of SL27 (e–g), SL4 (h–j) and SL2 (k–m). The legend in g illustrates the in-plane (green) and out-of-plane (black) tilt angles (θ), which are defined as half of the projected O–Ti–O bond angle . The tilt angles for a one unit-cell column are overlayed in each iDPC image to demonstrate the changing in-plane (green triangles) and out-of-plane (grey triangles) tilt angles. In g, j, m, solid and dashed curves are from experimental measurements and calculations, respectively. The error bars represent one standard deviation. Chemically abrupt interfaces are illustrated to the left of the ADF images (e, h, k) and model structures (a), illustrating the abrupt change between STO (green) and CTO (blue) layers. Chemically diffuse interfaces are illustrated to the right of the iDPC images (f, i, l) and model structures (a), illustrating the non-abrupt symmetry changes that are occurring as a result of octahedral coupling.

Fig. 2. Second-harmonic intensity indicates short-period superlattices lack interfaces.

Second-harmonic intensity of STO–CTO superlattices with varying periodicity, demonstrating various regimes of structural transitions and their role in the electronic/optical properties of heterostructures. The error bars are calculated from the mean square deviation of a parabolic fit to the measured second-harmonic intensity versus incident electric field. Ball-and-stick models are included to pictorially show the connection between octahedral tilt and the presence structurally diffuse interfaces, or lack thereof.

Fig. 3. Localized vibrational response of superlattices indicates the emergent role of the interfacial symmetry.

a, DFT-calculated PDOS projected on the octahedron O and Ti atoms of the SL27, SL4 and SL2 models. The arrows indicate the dominant phonon peaks. b, Cascade of DFT-calculated PDOS projected on STO (green), CTO (purple) and interface (orange) layers and the total DOS (black) for each superlattice model. c–k, Monochromated STEM-EELS line profile analyses of the three SL structures SL27 (c–e), SL4 (f–h) and SL2 (i–k) with the ADF intensity (I) profile (c, f, i), EELS profile (d, g, j) and integrated spectra from each layer (e, h, k) (as indicated by coloured regions in the ADF profile). Energy-loss spectra are normalized by multiplying intensity by the energy squared (IE²). The colour bars in d, g, j share the same labels and scales as e, h, k.

Fig. 4. FTIR and TDBS response of the STO–CTO superlattices.

a, Raw (solid), fitted (dashed) and residual (dot-dashed) data for FTIR from the superlattices on an NGO substrate. The 200-nm STO and CTO thin films on NGO substrates used to fit the superlattice spectra are shown in Supplementary Figs. 13b, 15e, f. The difference curves are scaled by a factor of two for clarity. b, Phonon lifetimes (black squares) as measured via TDBS compared with the thermal conductivity (red triangles, from ref. ) of STO–CTO superlattices with varying periodicities. The strong correlation between the two techniques conclusively demonstrates a transition in phonon scattering rates across the structural transitions elucidated with STEM/EELS. The error bars represent the standard deviation.