Table of Contents
Fetching ...

Strain Engineering of Altermagnetic Symmetry in Epitaxial RuO$_2$ Films

Johnathas D. S. Forte, Seung Gyo Jeong, Anand Santhosh, Seungjun Lee, Bharat Jalan, Tony Low

Abstract

The magnetic ground state of RuO$_2$ has been under intense debate. Using first-principles calculations, we show that compressive strain along [001] direction stabilizes an altermagnetic phase in RuO$_2$ thin films grown on (100) and (110) TiO$_2$ substrates. We further identify that compressive strain enhances the density of states near the Fermi level, resulting in a Fermi surface instability and the emergence of altermagnetism. The magnitude of strain and the associated increase in the density of states can be tuned by varying the film thickness, as systematically confirmed by x-ray diffraction and photoemission spectroscopy measurements. Symmetry analysis further reveals that (100) RuO$_2$ hosts an ideal altermagnetic order, whereas broken symmetry in (110) films leads to an uncompensated ferrimagnetic state. Finally, we discuss the effects of Hubbard $U$ parameters and evaluate the realistic tunneling magnetoresistance of (100) RuO$_2$.

Strain Engineering of Altermagnetic Symmetry in Epitaxial RuO$_2$ Films

Abstract

The magnetic ground state of RuO has been under intense debate. Using first-principles calculations, we show that compressive strain along [001] direction stabilizes an altermagnetic phase in RuO thin films grown on (100) and (110) TiO substrates. We further identify that compressive strain enhances the density of states near the Fermi level, resulting in a Fermi surface instability and the emergence of altermagnetism. The magnitude of strain and the associated increase in the density of states can be tuned by varying the film thickness, as systematically confirmed by x-ray diffraction and photoemission spectroscopy measurements. Symmetry analysis further reveals that (100) RuO hosts an ideal altermagnetic order, whereas broken symmetry in (110) films leads to an uncompensated ferrimagnetic state. Finally, we discuss the effects of Hubbard parameters and evaluate the realistic tunneling magnetoresistance of (100) RuO.

Paper Structure

This paper contains 7 sections, 1 equation, 4 figures.

Figures (4)

  • Figure 1: (a-c) Schematic illustrations of epitaxial strain in RuO2 films grown on TiO2 substrates with three orientations: (001), (100), and (110). Orange and pink arrows represent tensile and compressive strain components, respectively. (d-f) XRD RSMs of 12 nm thick RuO2 films grown on (001), (100), and (110) TiO2 substrates. (g) Summary of lattice constants (d) and calculated strain factors (s) for each crystallographic orientation of RuO2/TiO2 heterostructures.
  • Figure 2: Strain-magnetization Neel vector phase diagrams for (a) (001), (b) (100) and (c) (110) RuO$_2$. The red stars and blue diamonds in each panel indicate lattice constants of fully strained and partially strained RuO$_2$ on TiO$_2$, respectively. (d) Strain-net magnetic moments for (110) RuO$_2$.
  • Figure 3: (a) Hubbard $U$ and (b) hole doping-strain phase diagrams for (100) RuO$_2$ showing the Neel vector. Density of states for (c) charge neutral (100) RuO$_2$ from pristine to fully strained, (d) unstrained (100) RuO$_2$ as a function of hole-doping and (e) unstrained, charge charge neutral (100) RuO$_2$ as a function of the Hubbard $U$ strength. The insets show the increase of the density of states at the Fermi level, which dictates the itinerant electron magnetism in (c) and (d), while the decreasing DOS($E_F$) in (e) reveals the localized character of this type of magnetism. (f) XPS valence band spectra of RuO$_2$ films on TiO$_2$ (100) substrates with different thickness $t$. The vertical arrows in the right panel indicate the estimated peak positions of the Ru $4d$ states, obtained by fitting with Voigt functions. (g) Summary of the Ru $4d$ peak positions as a function of $t$.
  • Figure 4: Band structures for (a) unstrained (100) RuO$_2$ with $U = 0$ eV and fully strained (100) RuO$_2$ with (b) $U = 0$ eV and (c) $U = 2$ eV. Strain alone can drive the spin-splitting of the bands, which is further enhanced by the Hubbard $U$ interaction. In (d), we evaluate the TMR as a function of the Fermi level for different $U$, showing that it increases for largeer spin-splittings, as expected.