🔬 materials science

Moiré-resonant surface state in ultrathin RuO $_2$

This study reveals that ultrathin RuO $_2$ (110) films on Ru(0001) exhibit a nonmagnetic charge order driven by moiré-induced flat-band scattering and a reversible metastable surface reconstruction, while spin-polarized measurements confirm the absence of magnetic order on the surface.

Original authors: Philipp Keßler, Andreas Feuerpfeil, Armando Consiglio, Hendrik Hohmann, Ronny Thomale, Jonas Erhardt, Bing Liu, Vedran Jovic, Ralph Claessen, Patrick Härtl, Matteo Dürrnagel, Simon Moser

Published 2026-07-21

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Philipp Keßler, Andreas Feuerpfeil, Armando Consiglio, Hendrik Hohmann, Ronny Thomale, Jonas Erhardt, Bing Liu, Vedran Jovic, Ralph Claessen, Patrick Härtl, Matteo Dürrnagel, Simon Moser

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the world of materials science as a giant, bustling city where atoms are the citizens. Sometimes, these citizens arrange themselves in perfect, rigid grids, like soldiers on parade. Other times, they get a little wild, forming waves of charge or even spinning like tiny magnets. For a long time, scientists have been hunting for a very specific type of "magnetic citizen" in a material called Ruthenium Dioxide ( $\text{RuO}_2$ ). They were looking for something called "altermagnetism," a fancy new kind of magnetic order that acts like a mix between a ferromagnet (where everyone spins the same way) and an antiferromagnet (where neighbors spin in opposite directions). It's a bit like a dance where the steps alternate, but the rhythm creates a unique, hidden energy. Why does this matter? Because if we can find and control these magnetic dances, we might build faster, smarter computers and better ways to split water for clean energy. But to find the dance, you need a stage that's perfectly flat and clean, which is incredibly hard to build in the lab.

Enter the story of a team of scientists who decided to build the ultimate stage: an ultrathin, atomically perfect sheet of $\text{RuO}_2$ grown on a bed of pure Ruthenium. They wanted to see if this thin film would finally show off its magnetic moves. Instead, they found something even more fascinating: a complex, non-magnetic tango between the surface electrons and the pattern of the floor beneath them.

The researchers started by growing a super-thin film of $\text{RuO}_2$ (about 3 nanometers thick) on a crystal of Ruthenium. Think of the Ruthenium floor as a tiled floor with a specific pattern, and the $\text{RuO}_2$ film as a second layer of tiles laid on top. Because the two layers don't match up perfectly—like trying to lay a grid of square tiles over a grid of hexagonal ones—they created a giant, rippling pattern called a "moiré" effect. You've probably seen this when you hold two fine mesh screens over each other and see a swirling, wavy pattern emerge.

On this surface, the electrons behave like they are stuck in a very flat, one-dimensional highway. The scientists found that these electrons form a "flat-band surface state," which is a bit like a traffic jam where all the cars are moving at the same speed in a straight line. Because the highway is so flat and crowded, the electrons are very sensitive to any bumps in the road. And the moiré pattern from the floor underneath provided the perfect bump.

Here is the magic part: the scientists discovered that the spacing of the moiré ripples on the floor matched almost perfectly with the natural "wavelength" of the electron traffic jam. It's like if the bumps in the road were spaced exactly the right distance to make the cars bounce in rhythm. This created a "resonance," a special condition where the electrons' charge density (how crowded they are) started to wiggle in sync with the moiré pattern. The team used a super-powerful microscope called a Scanning Tunneling Microscope (STM) to watch this happen. They could see the electrons getting denser and sparser in a repeating pattern, a "charge order," but crucially, this was purely about where the electrons were, not how they were spinning.

While investigating, they also stumbled upon a secret trick. By poking the surface with the tip of their microscope, they could flip the surface atoms into a different, temporary arrangement called a "c(2 × 2) reconstruction." Imagine a puzzle where you can push a few pieces to make the whole picture look slightly different, and then push them back to restore the original. This new shape was caused by the atoms relaxing into a more comfortable position, and the team could switch it back and forth like a light switch.

The big question, however, was: where is the magnetism? The paper explicitly rules out the presence of any magnetic order on this surface. The scientists used a special magnetic microscope tip (coated with Gadolinium) to look for tiny magnetic spins, hoping to find the "altermagnetic" dance they were expecting. They looked very closely, but the result was a flat line: no magnetic contrast, no magnetic domains, and no evidence of the spins aligning or anti-aligning. Their computer simulations, which modeled the electrons interacting with each other, also confirmed that the ground state of this surface is non-magnetic. The electrons are busy dancing a charge tango, but they are not spinning in a magnetic way.

So, what did they find? They found that in these ultrathin films, the electrons are dominated by the geometry of the surface and the moiré pattern, creating a beautiful, non-magnetic charge order. They proved that while the bulk material (the thick block) might have been debated for its magnetic properties, this specific, atomically thin surface is a playground for electronic correlations and structural tricks, but not for magnetism. It turns out that $\text{RuO}_2$ on a Ruthenium floor is a fantastic platform for studying how electrons organize themselves when they are squeezed into a thin layer and shaken by a moiré pattern, even if they aren't the magnetic stars some thought they would be.

Technical Summary: Moiré-Resonant Surface State in Ultrathin RuO₂(110)

Problem Statement
Ruthenium dioxide (RuO₂) has emerged as a prototypical candidate for altermagnetism, a magnetic phase characterized by alternating magnetic moments within the unit cell that enable spin-split magnon excitations. While bulk RuO₂ has been the subject of intense debate regarding the existence of intrinsic magnetic order—with recent sensitive measurements (muon spin rotation, neutron scattering) largely excluding long-range magnetic order or sizable local moments—theoretical interest has shifted toward ultrathin films and surfaces. It has been proposed that dimensionality reduction and enhanced electron correlations in ultrathin films could drive surface-specific magnetic instabilities. However, experimental verification has been hindered by the lack of high-quality, atomically ordered RuO₂(110) surfaces, which are difficult to achieve via conventional epitaxial methods like sputtering or molecular beam epitaxy.

Methodology
The authors investigated atomically ordered, ultrathin RuO₂(110) films grown on a Ru(0001) substrate via controlled oxidation and reactive segregation. This growth method yields stoichiometric films with minimal defect density, suitable for high-resolution surface characterization. The study employed a multi-faceted approach:

Experimental: Scanning Tunneling Microscopy (STM) and Spectroscopy (STS) were used for atomic-resolution topography and local density of states (LDOS) mapping. Spin-polarized STM (SP-STM) with a Gd-coated tip was utilized to probe for magnetic order. Angle-resolved photoemission spectroscopy (ARPES) was performed at the Advanced Light Source to map the Fermi surface.
Theoretical: Density Functional Theory (DFT) was used to model structural relaxations and surface reconstructions. Density Matrix Renormalization Group (DMRG) methods were applied to a minimal Hamiltonian model to investigate strongly correlated electronic states in quasi-one-dimensional chains, incorporating the moiré potential and electron-electron interactions.

Key Contributions and Results

Characterization of the Flat-Band Surface State (FBSS):
The study confirms the existence of a quasi-one-dimensional flat-band surface state (FBSS) on the RuO₂(110) surface, arising from the hybridization of Ru 4d and O 2p orbitals in [001]-oriented chains. ARPES measurements reveal a canonical Fermi surface dominated by linear features of this FBSS, which exhibits strong Fermi surface nesting with a vector $q_N \approx 0.24(4) \, \text{\AA}^{-1}$ .
Moiré-Induced Charge Modulation:
STM topography reveals a distinct, quasi-periodic "beating" in the apparent height of the Ru–O chains, attributed to an incommensurate moiré pattern resulting from the lattice mismatch between the RuO₂ film and the Ru(0001) substrate. The moiré wavelength is determined to be $\lambda_M \approx 22.9 \, \text{\AA}$ .
- Thickness Dependence: The coupling to the substrate is thickness-dependent. Thinner regions show a stronger moiré potential and shifted LDOS features, while thicker regions exhibit a substrate-decoupled electronic structure.
- Resonant Scattering: A critical finding is the near-coincidence between the moiré wave vector ( $q_M \approx 0.27 \, \text{\AA}^{-1}$ ) and the Fermi surface nesting vector ( $q_N$ ). STS measurements show a pronounced resonance in the moiré scattering intensity at a bias of $-5$ meV. DMRG calculations confirm that this resonance arises from a constructive interference between the moiré potential and electronic fluctuations (Fermi surface scattering) within the FBSS. The charge modulation is primarily driven by the moiré potential but is significantly enhanced by electron correlations at the Fermi surface.
Metastable Surface Reconstruction:
The authors identify a previously unreported, metastable $c(2 \times 2)$ surface reconstruction. This reconstruction, which doubles the surface unit cell, can be reversibly toggled between two symmetry-equivalent states using STM tip manipulation. DFT calculations attribute this to a structural relaxation driven by surface phonon softening, which is energetically more stable than the unreconstructed surface by $\approx 25$ meV.
Absence of Magnetic Order:
Despite the theoretical motivation for surface magnetism, the study finds no evidence of magnetic order.
- Spin-Polarized STM: Measurements with a Gd-coated tip show no magnetic contrast, spin-density waves, or magnetic domains, even at the sub-atomic resolution limits of the technique.
- Theoretical Ground States: DMRG calculations, even with on-site Coulomb repulsions exceeding the FBSS bandwidth, yield spin-singlet ground states with no local magnetization. The observed charge modulations are purely non-magnetic.

Significance
The paper establishes ultrathin RuO₂(110) as a versatile platform for exploring moiré-assisted electronic orders in correlated electron systems. The primary significance lies in the demonstration of a nonmagnetic charge order that is imprinted by the substrate's moiré potential and resonantly enhanced by Fermi surface nesting within a flat-band surface state.

The findings argue against the propensity of stoichiometric, ultrathin RuO₂(110) to exhibit magnetic instabilities or altermagnetism under these specific growth conditions. Instead, the system is characterized by complex structural and electronic phenomena—specifically the interplay between moiré potentials and Fermi surface scattering—that result in tunable, non-magnetic ordering. This work provides a crucial experimental benchmark, suggesting that substrate-induced effects and the specific nature of the surface termination play decisive roles in the electronic and magnetic properties of RuO₂ thin films, potentially reconciling conflicting reports on altermagnetism in different substrate environments (e.g., TiO₂ vs. Ru).

Technical Summary: Moiré-Resonant Surface State in Ultrathin RuO₂(110)

More like this