Skip to main content
QUICK REVIEW

[Paper Review] Surface acoustic wave stimulated Brillouin scattering in thin-film lithium niobate waveguides

Kaixuan Ye, Hanke Feng|arXiv (Cornell University)|Nov 9, 2023
Photorefractive and Nonlinear Optics7 citations
TL;DR

This paper reports the first experimental observation of backward stimulated Brillouin scattering (SBS) in thin-film lithium niobate (TFLN) waveguides using surface acoustic waves (SAW). A peak Brillouin gain coefficient of 84.9 m⁻¹W⁻¹ was achieved in z-cut TFLN waveguides at a 20° crystal rotation angle, demonstrating strong optomechanical interaction and high potential for integrated photonic applications.

ABSTRACT

We report the first-ever experimental observation of backward stimulated Brillouin scattering (SBS) in thin-film lithium niobate (TFLN) waveguides. The peak Brillouin gain coefficient of the z-cut LN waveguide with a crystal rotation angle of 20$^{\circ}$ is as high as 84.9m$^{-1}$W$^{-1}$, facilitated by surface acoustic waves (SAW) at 8.06GHz.

Motivation & Objective

  • To experimentally demonstrate stimulated Brillouin scattering (SBS) in thin-film lithium niobate (TFLN) waveguides, a platform with high potential for integrated nonlinear photonics.
  • To investigate the role of surface acoustic waves (SAW) in enhancing SBS gain in TFLN waveguides by comparing air-clad and silica-clad configurations.
  • To explore the crystal orientation dependence of SBS gain in z-cut TFLN waveguides by varying the rotation angle (θ) from 0° to 40°.
  • To validate the SBS signal using multiple measurement techniques, including lock-in amplifier and vector network analyzer (VNA), ensuring signal reliability.
  • To establish TFLN as a scalable, low-loss, and high-performance platform for integrated Brillouin-based devices such as lasers and filters.

Proposed method

  • Fabricated 1-cm-long z-cut and x-cut TFLN waveguides on a 400-nm-thick lithium niobate wafer using reactive ion etching with 200-nm and 250-nm etch depths, respectively.
  • Employed a pump-probe modulated lock-in amplifier setup to measure the SBS gain profile, comparing SBS peaks from the waveguide to those from a 5-m fiber reference.
  • Used COMSOL Multiphysics to simulate the optical and acoustic field overlap in z-cut TFLN waveguides at θ = 20°, identifying surface acoustic wave excitation at 7.95 GHz.
  • Characterized SBS using vector network analyzer (VNA) to measure linewidth and signal fidelity, particularly for Stokes and anti-Stokes components.
  • Systematically varied the crystal rotation angle (θ = 0°, 20°, 40°) in z-cut waveguides to map the dependence of SBS gain on crystal orientation.
  • Compared air-clad and silica-clad x-cut TFLN waveguides to isolate the effect of acoustic wave confinement on SBS enhancement.
Fig. 1 : (a) Artistic representation of the air-cladded z-cut TFLN waveguides with crystal orientation angle $\theta$ of 0 $\degree$ , 20 $\degree$ , and 40 $\degree$ . (b) Simulated electric field and (c) the corresponding acoustic response of the air-cladded z-cut TFLN waveguides at the Brillouin
Fig. 1 : (a) Artistic representation of the air-cladded z-cut TFLN waveguides with crystal orientation angle $\theta$ of 0 $\degree$ , 20 $\degree$ , and 40 $\degree$ . (b) Simulated electric field and (c) the corresponding acoustic response of the air-cladded z-cut TFLN waveguides at the Brillouin

Experimental results

Research questions

  • RQ1Can backward stimulated Brillouin scattering (SBS) be experimentally observed in thin-film lithium niobate (TFLN) waveguides using surface acoustic waves (SAW) as the mechanical excitation source?
  • RQ2How does the crystal orientation (θ) of z-cut TFLN waveguides affect the Brillouin gain coefficient, and is this dependence consistent with theoretical predictions?
  • RQ3To what extent does acoustic wave confinement—via air cladding versus silica cladding—affect the SBS gain in x-cut TFLN waveguides?
  • RQ4What is the maximum achievable Brillouin gain coefficient in TFLN waveguides under SAW excitation, and how does it compare to other integrated platforms?
  • RQ5Can the SBS signal be reliably measured and validated using both lock-in amplifier and vector network analyzer (VNA) techniques in a chip-scale photonic platform?

Key findings

  • The first experimental observation of backward stimulated Brillouin scattering (SBS) in thin-film lithium niobate (TFLN) waveguides was achieved using surface acoustic waves (SAW) at 8.06 GHz.
  • A peak Brillouin gain coefficient of 84.9 m⁻¹W⁻¹ was measured in z-cut TFLN waveguides with a 20° crystal rotation angle, significantly exceeding values in silicon nitride and approaching those in chalcogenide and suspended-silicon platforms.
  • The SBS gain coefficient in z-cut TFLN waveguides shows strong dependence on crystal orientation, with 9.45 m⁻¹W⁻¹ at 0°, 84.9 m⁻¹W⁻¹ at 20°, and 25.6 m⁻¹W⁻¹ at 40°, matching simulated trends.
  • The SBS signal in x-cut TFLN waveguides exhibited a narrow linewidth of 11.9 MHz for the Stokes signal, indicating high spectral purity and potential for high-resolution filtering.
  • Silica cladding reduced the SBS gain coefficient by nearly fourfold compared to air-clad x-cut waveguides, confirming that acoustic wave confinement is critical for high SBS efficiency.
  • The measured SBS gain coefficients were approximately 50% higher than simulated values, likely due to an underestimation of the acoustic quality factor (Q_ac = 1000) in the simulation model.
Fig. 2 : (a)Lock-in amplifier measurement results of the x-cut TFLN waveguides with and without silica cladding, displayed together with SBS peaks from 5 m of fibers in the setup. Measured (b) Stokes and (c) anti-Stokes signal from the x-cut TFLN waveguides without silica cladding.
Fig. 2 : (a)Lock-in amplifier measurement results of the x-cut TFLN waveguides with and without silica cladding, displayed together with SBS peaks from 5 m of fibers in the setup. Measured (b) Stokes and (c) anti-Stokes signal from the x-cut TFLN waveguides without silica cladding.

Better researchstarts right now

From reading papers to final review, dramatically reduce your research time.

No credit card · Free plan available

This review was created by AI and reviewed by human editors.