Comparative Raman study of Ruddlesden-Popper nickelates and the monolayer-trilayer polymorph

TL;DR

This paper addresses phase identification and the interplay between lattice dynamics and electronic excitations in Ruddlesden-Popper nickelates, with a focus on the monolayer-trilayer (ML-TL) polymorph. Using polarized Raman spectroscopy on high-quality, oxygen-optimized crystals and supporting DFPT phonon calculations, the authors establish spectral fingerprints for ML, BL, TL, and ML–TL. They find that ML–TL mimics TL at room temperature in its phonon spectrum but hosts distinct low-temperature electronic Raman features, including a $910\ \mathrm{cm^{-1}}$ hump tied to density-wave fluctuations and a unique $680\ \mathrm{cm^{-1}}$ peak, implying a distinct correlated electronic state arising from ML insertion. Overall, the study shows that the ML block actively modulates both lattice dynamics and electronic excitations, offering a framework for phase identification and insights into layer-dependent phenomena that may be relevant to superconductivity in RP nickelates.

Abstract

Ruddlesden-Popper (RP) nickelates have attracted intense interest following the discovery of superconductivity in several members of the series, including bilayer (BL) La$_3$Ni$_2$O$_7$, trilayer (TL) La$_4$Ni$_3$O$_{10}$, and structural polymorphs composed of monolayer-bilayer or monolayer-trilayer (ML-TL) units. However, an inherent propensity of the RP series to form intergrown phases during single-crystal synthesis, together with spatial variations in oxygen stoichiometry, has complicated the determination of their intrinsic material properties. As a consequence, conflicting reports have emerged on both their electronic phase transitions and lattice dynamics. In this work, we perform a comparative study of the phononic and electronic Raman responses of high-quality ML-TL single crystals and contrast them with those of other RP nickelates, using samples with optimized oxygen content. We establish several Raman spectral features that enable unambiguous phase identification across the series. Moreover, we uncover characteristics in the phononic and electronic Raman response of ML-TL that are not reflected in the pure ML and TL compounds. We attribute these differences to a distinctive electronic structure arising from self-doping and confinement effects induced by the ML unit within the ML-TL lattice architecture.

Figures (5)

• Figure 1: (a-d) Schematics of the crystal structures of (a) ML La$_2$NiO$_4$, (b) BL La$_3$Ni$_2$O$_7$, (c) TL La$_4$Ni$_3$O$_{10}$, and (d) ML-TL La$_3$Ni$_2$O$_7$, with $Bmab$, $Cmcm$, $P2_1/a$, and $Fmmm$ unit cells, respectively. (e) STEM-HAADF images of a ML-TL single crystal. The low-resolution image (left) demonstrates absence of secondary phases on large length scales. The contrast between bright and dark vertical stripes corresponds to minor topographic differences due to curtaining effects caused by ion milling during specimen preparation. The atomic-resolution image (right) resolves the stacking sequence of ML (red) and TL (green) units. The [001] and [010] directions of the $Fmmm$ unit cell are indicated.
• Figure 2: (a-d) Raman spectra of the (a) ML, (b) BL, (c) TL, and (d) ML-TL phases acquired at room temperature in $\bar{z}(xu)z$ scattering configuration (without analyzer). The inset in (a) shows the top-view of a Ni-O plaquette in the ab-plane of layered nickelates, with octahedral tilts omitted for simplicity. The $x$ and $y$ directions of the laboratory frame of the Raman experiment point along the Ni-O-Ni bond directions. The $x'$ direction is also indicated. Vertical dashed lines in (c) and (d) indicate the position of characteristic phonon modes of the TL phase that are absent in ML-TL. Green arrows mark a broad electronic Raman feature present in both TL and ML-TL.
• Figure 3: Polarization-resolved Raman spectra of the phonon sector of ML-TL at (a) 315 K and (b) 60 K, acquired in $xx$, $x'x'$, $xy$, and $x'y'$ scattering configuration, respectively. For clarity, the $xy$ and $x'x'$ spectra are offset in vertical direction by 0.5 a.u. and the $xx$ spectra by 2 a.u. (c) Temperature dependence of selected low-energy phonon modes. Spectra are offset in vertical direction by 1 a.u. Solid yellow lines are fits of the data using Voigt profiles. Individual profiles of the 34 K fit are shown as black lines at the bottom of the panel. The phonon mode around 140 cm$^{-1}$ is highlighted by the triangle symbol and its fitted peak profile is shaded in purple. (d) Temperature dependence of higher-energy phonon modes, with modes around 400 and 425 cm$^{-1}$ highlighted by green and orange triangle symbols, respectively, and the corresponding peak profiles shaded in the same colors. (e) Temperature dependence of the Raman shift, integrated intensity, and linewidth of the modes evolving around 140, 400, and 425 cm$^{-1}$ with the same color coding as in panels (c) and (d).
• Figure 4: (a-f) Atomic displacement patterns of selected phonon modes of ML-TL for the tetragonal $P4/mmm$ unit cell. The $E_g$, $B_{1g}$, and $A_{1g}$ mode symmetries are indicated along with the computed phonon frequencies. Green arrows display the directions and amplitudes of the atomic vibrations. For clarity, only displacements with large amplitudes are shown for each mode. The $b$ and $c$ directions of the $P4/mmm$ unit cell are also indicated.
• Figure 5: Electronic Raman scattering across a wide energy range of (a-d) ML-TL, (e-h) TL, and (i-l) ML phases. Spectra were acquired at different temperatures (light to dark shades) and in $xx$, $x'x'$, $xy$, and $x'y'$ scattering configurations (top to bottom panel), respectively. Green arrows mark a feature around 900 cm$^{-1}$ that occurs most prominently in the $xx$ and $x'x'$ channels of the ML-TL and TL phases. Purple arrows highlight a feature occurring most pronounced in the $xy$ and $x'y'$ channels of ML-TL. The TL spectra in panel (e-h) are adopted from Ref. SutharVS2025.