Chiral-phonon generation of orbital currents in light transition metals

Abstract

Orbital angular momentum offers a new channel for information transport in a vast set of materials. Its coherent generation and detection remain, however, largely unexplored. Here, we demonstrate that chiral surface acoustic waves (SAWs) generate sizable orbital currents in light-metal/ferromagnet bilayers through both the acoustic orbital Hall effect and acoustic orbital pumping. Using symmetry analysis of SAW-driven voltages, we disentangle vorticity-sensitive orbital currents arising from lattice rotation in the non-magnetic layer from angular-momentum pumping from the ferromagnet. Strong signals are observed only in nickel/chromium and nickel/titanium, while nickel/aluminum and all cobalt-based bilayers show negligible responses, revealing the critical roles of orbital Hall conductivity, phonon-orbital coupling, and interfacial orbital transparency. Comparison with spin-torque ferromagnetic resonance and second-harmonic measurements -- where electrically driven orbital angular momentum are weaker -- demonstrates that phonon excitation generates orbital currents more efficiently. These results establish chiral SAWs as an effective route for orbitronic functionality and open pathways toward phonon-controlled orbital magnetism.

Figures (11)

• Figure 1: (A) Schematic diagram of the experimental setup. The device consists of a LiNbO$_3$ substrate with two opposing IDTs that can generate SAWs propagating with opposite wavevectors. A FM/LM bilayer (FM = Ni, Co, and LM = Cr, Al) patterned in a bar geometry is grown perpendicular to the SAW propagation direction and in the acoustic path. Gold contacts are grown on each end of the bar to detect acoustically induced voltages. An in-plane magnetic field is applied with an electromagnet, and the sample can be rotated within the field plane. Transmission measurements ($S{_{21}}$) are performed with a Vector Network Analyzer (VNA) at the resonant IDT frequencies as a function of magnetic field, while voltage measurements are carried out by exciting the IDT with low-frequency amplitude modulated microwave pulses and detected the synchronized response with a lock-in amplifier. (B) Schematics of acoustic pumping (AP) (left-side), where a spin or orbital current is pumped into an adjacent non-magnetic layer and transformed into a charge current through ISHE or IOHE. Schematics of acoustic orbital hall effect (AOHE) (right-side) rising from the SAW lattice rotations coupled to the orbital degree of freedom in the LM, generating an orbital current polarized in the $\mathbf{y}$ direction. This current can interact with the magnetization of the FM, with a modulation that depends on their relative orientation. (C) Acoustic-FMR transmission measurements ($S_{21}$) at 1.3 GHz of Ni (left) and Co (right) displaying the expected 4-fold symmetry and resonance peaks.
• Figure 2: of transverse voltage contributions in Ni/Cr. (A) Representative voltage signals measured as a function of the magnetic field ($\mu_0 H$) at an angle $\varphi_H=45^\circ$, showing the total (top), even (middle), and odd (bottom) contributions. The magnetic field sweeps from a large magnetic field to zero. (B) Two-dimensional maps of the total (top), even (middle), and odd (bottom) voltage components as a function of in-plane ($\mu_0H_x$ and $\mu_0H_y$) magnetic fields. The left column shows the results for a SAW traveling in the positive x-direction, and the right column shows the results for a SAW traveling in the negative x-direction. (C) Angular dependence of the even voltage component at $\mu_0H = 35$ mT for both $+k$ (blue) and $-k$ (red) SAWs. (d) Angular dependence of the odd voltage components at $\mu_0H = 3.5$ mT for both $+k$ (blue) and $-k$ (red) SAWs.
• Figure 3: (A) Normalized AOHE signal for each sample as a function of the angle between the SAWs and the magnetic field at a fixed magnetic field of $\mu_0H = 35$ mT. (B) Normalized AP signal for each sample as a function of the angle between the SAWs and the magnetic field at a fixed magnetic field of $\mu_0H = 3.5$ mT for Ni-based samples and $\mu_0H = 1.3$ mT for Co-based samples. (C) Summary of the results, where the blue bars represent the AOHE signal and the gold bars represent the AP signal. For comparison, we include Ni$|$Ti measured in stripped bars; however, this sample was not part of the same batch and has a different geometry.
• Figure 4: (A) Plot of the voltage $V_{xy}$ for each contribution: pumping (gold) and AOHE (blue) for a single propagating SAW in the +k direction. (B) Quotient $V^{\rm AP}/V^{\rm AOHE}$ as a function of frequency at a fixed power $P= 34$ mW for a single propagating SAW in the +k direction. (C) Voltage signals with both IDTs emitting SAWs at different power ratios, but keeping the total power constant. The top panel shows the AOHE voltage $V_{xy}^{\rm AOHE}$ for different power ratios of IDTs; the color code represents each power ratio, where blue is maximum power in IDT2 and no power on IDT1. At the same time, red is maximum power in IDT1 and no power in IDT2. In green dots with black circles, we show the result of having both IDTs at the same power, which should have a dominant standing wave with some propagating. The bottom panel shows the pumping signal for the different power ratios.
• Figure S1: (A) Schematics of the magnetoacoustic device with the dimensions. The IDTs are separated by 2 mm, and in the acoustic path, we have two transversal rectangles of the material under study. The transverse bars are 10 $\mu$m in width, and the FM$|$NM are 10nm/10nm. (B) SAW transmission $S_{21}$ from 0 to 2.5 GHz. The peaks show the resonant behavior of the IDTs at which SAWs are efficiently sent. The fundamental frequency is 111 MHz. (C) Transmission measurements of Ni-based sample (left) and Co-based sample (right) as a function of the strength and direction of the magnetic field.