🔬 applied physics

Phonon Trapping Lateral Field Excited Suspended Bulk Acoustic Wave Resonators (XBARs)

This paper proposes lateral field excited suspended bulk acoustic wave resonators (XBARs) as a superior platform for microwave-to-optical transducers by overcoming the modal confinement limitations of traditional FBARs, and demonstrates a 4-fold improvement in quality factor and modal confinement through the fabrication of a phonon-trapping microresonator with a spherical piezoelectric lens.

Original authors: Elnaz Shokati, Robert Thomas, Krishna C. Balram

Published 2026-07-08

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Elnaz Shokati, Robert Thomas, Krishna C. Balram

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 you are trying to build a super-efficient translator that can convert a whisper from a microwave signal into a beam of light. This is the goal of the researchers in this paper: creating a device that acts as a bridge between the microwave world (used in our phones and quantum computers) and the optical world (used in fiber optic cables).

To do this, they need a "middleman" made of sound waves (phonons). The challenge is that sound waves usually spread out like butter on toast, making it hard to concentrate them enough to talk to light effectively.

Here is how the paper explains their solution, using simple analogies:

The Problem: The "Flat Mirror" Trap

Traditionally, these sound-based devices (called FBARs) work like a flat mirror. You have a flat piece of material with metal electrodes on top. When you send electricity through it, sound waves bounce up and down between the metal and the bottom.

The Issue: While the sound bounces up and down, it also leaks out sideways, like water spilling over the edge of a flat tray. Because the sound spreads out, it's hard to squeeze it into a tiny, powerful spot where it can efficiently talk to light.
The Consequence: To build a good translator, you need the sound to be trapped in a tiny, dense box. If it leaks, the translation is weak and inefficient.

The Solution: The "Acoustic Lens"

The researchers decided to stop using flat trays and start using bowl-shaped bowls.

The Analogy: Think of a flat mirror versus a curved mirror in a bathroom. A flat mirror reflects light straight back, but if you tilt it, the light goes everywhere. A curved mirror (like a magnifying glass) catches light from different angles and focuses it into a single, bright spot.
The Innovation: They took their sound-generating material (a thin film of a special ceramic) and shaped it into a tiny, spherical lens (like a microscopic dome).
How it works: When the sound waves try to leak out sideways, the curved shape of the lens acts like a funnel. Instead of escaping, the sound waves are gently pushed back toward the center. It's like a "phonon trap" that keeps the sound energy locked in a small, high-quality box.

The Specifics: "XBAR" and the "Overtone"

The paper introduces a specific type of device called an XBAR (Lateral Field Excited Bulk Acoustic Wave Resonator).

The Setup: Instead of pushing the sound straight up and down, they push it sideways between two metal fingers.
The "Overtone": Imagine a guitar string. You can pluck it to get a low note (the fundamental), or you can force it to vibrate in a way that creates a much higher, sharper note (the overtone). The researchers are using these high-pitched "overtones" of sound.
Why? They need the sound to vibrate at a very specific, high speed (around 7.25 billion times a second) to match the speed of modern quantum computers. By using this specific "overtone" mode, they can hit that target speed while keeping the sound trapped in the lens.

The Results: A 4x Improvement

The team built these lens-shaped devices and tested them.

The Comparison: They compared a "flat" device (the old way) with a "lensed" device (their new way).
The Outcome: The lensed device was much better at keeping the sound trapped. They measured the "Quality Factor" (a score for how long the sound rings out before fading). The lensed version scored about 4 times higher than the flat version.
What this means: The sound stays in the box longer and is more concentrated. This makes the device much more ready to become an efficient translator between microwaves and light.

The Bottom Line

The paper doesn't claim to have built the final, perfect translator yet. Instead, they have successfully demonstrated the first crucial step: making a sound trap that actually works.

They showed that by shaping the material into a tiny lens, they can stop sound from leaking out and concentrate it into a small, powerful space. This is the essential foundation needed to eventually build the high-speed bridges between quantum computers and the internet of the future.

Technical Summary: Phonon Trapping Lateral Field Excited Suspended Overtone Bulk Acoustic Wave Resonators (XBARs) for Microwave-to-Optical Transduction

Problem Statement
Film bulk acoustic wave resonators (FBARs) are the standard for compact RF filters in modern wireless communication but typically operate with quasi-plane acoustic waves where the transverse field extent is much larger than the acoustic wavelength. While this suits filtering, it is suboptimal for microwave-to-optical signal transduction (MW-OT), which requires strong modal confinement to maximize the acousto-optic (AO) interaction strength ( $g$ ) and achieve high photon conversion efficiency.

Traditional MW-OT architectures based on sub-micron optomechanical crystals face challenges in efficient piezoelectric excitation and high optical power handling due to large surface-to-volume ratios. Conversely, traditional bulk acoustic wave approaches, such as high-overtone bulk acoustic wave resonators (HBARs), suffer from two main limitations: (1) the acoustic field is confined directly under metal electrodes, hindering the engineering of small-mode-volume, high-quality-factor optical cavities; and (2) the longitudinal wave velocities often push interaction frequencies beyond the preferred 5–8 GHz range of superconducting qubits. Consequently, existing bulk MW-OTs exhibit very weak transduction efficiency ( $\approx 10^{-8}$ ).

Methodology and Design
The authors propose an alternative platform using lateral field excited suspended overtone bulk acoustic wave resonators (XBARs) to overcome these geometric and physical constraints. The design strategy involves three key adaptations:

Overtone Shear Mode Selection: Instead of the standard fundamental $A_1$ Lamb wave, the device utilizes a higher-order $A_{21}$ overtone resonance. By employing a thin-film lithium niobate (LN) or scandium-doped aluminum nitride (ScAlN) layer with a silica underlayer, the device excites a shear wave resonance. This configuration is critical for achieving a Brillouin scattering frequency ( $f_{BS}$ ) compatible with superconducting qubits (approx. 7.25 GHz) while maintaining a small mode volume.
Phonon Trapping via Acoustic Lensing: To address lateral mode leakage inherent in Lamb wave modes, the piezoelectric layer is shaped into a spherical lens. This curvature acts as a spherical mirror for the acoustic field, refocusing off-axis propagation and trapping the phonons within a small mode volume ( $A_{eff}L$ ). This approach aims to increase the mechanical quality factor ( $Q_m$ ) and the AO interaction strength ( $g$ ).
Material and Fabrication Trade-offs: While z-cut LN offers superior piezoelectric coupling, the authors initially demonstrate the concept using ScAlN on a silicon-on-insulator (SOI) substrate. This choice was made to circumvent fabrication difficulties associated with reliable grayscale lithography and etching in LN, which often lead to parasitic modes and insufficient lateral leakage suppression. The ScAlN device features a spherical lens embedded in the piezoelectric film, fabricated via deep reactive ion etching (DRIE) and laser lithography.

Key Results
The paper presents the fabrication and characterization of these phonon-trapping XBAR microresonators:

Modal Confinement: Finite Element Method (FEM) simulations and k-space analysis confirm that shaping the ScAlN layer into a spherical lens significantly suppresses lateral leakage. In the k-space representation, the leaky modes (non-zero $k_x$ components) present in flat-flat cavities are nearly absent in the lensed devices, indicating tight 3D confinement.
Quality Factor Improvement: Experimental measurements of the RF reflection spectrum ( $S_{11}$ ) reveal that the lensed devices exhibit a distinct doublet structure. One mode corresponds to the untrapped background resonance, while the other corresponds to the trapped mode within the lens. The trapped modes demonstrate a significant improvement in the mechanical quality factor.
Performance Metrics: The authors report an overall improvement in the $f \times Q$ product of approximately 4 $\times$ for the lensed devices compared to reference flat-flat devices. The measured $Q_m$ for reference devices ranged from 120 to 365, while the lensed devices achieved higher values, validating the phonon trapping mechanism.
Coupling Efficiency: Although the $A_{21}$ overtone mode has a piezoelectric coupling coefficient ( $k^2_{eff}$ ) roughly 75 times lower than the fundamental $A_1$ mode, the authors argue it remains sufficiently high for efficient microwave-to-acoustic transduction, provided the $Q_m$ is enhanced as demonstrated.

Significance and Claims
The paper positions this work as a "first step towards viability" for building efficient MW-OTs using bulk acoustic wave platforms. The primary significance claimed is the demonstration of a geometry that successfully combines:

Electrical Actuation: The lateral field excitation allows for electrical driving without interfering with the optical path.
Optical Compatibility: By moving the acoustic field away from the metal electrodes (via the suspended, lensed geometry), the platform becomes suitable for integrating small-mode-volume, high- $Q_o$ optical cavities, a prerequisite for high transduction efficiency.
Scalability: The approach leverages established piezoelectric resonator designs, offering a potential pathway to utilize foundry infrastructure, unlike complex 1D photonic crystal approaches.

The authors modestly frame these results as a foundational acoustic demonstration. They acknowledge that while the acoustics aspects (trapping and $Q_m$ enhancement) are proven, the full realization of an efficient MW-OT requires further optimization of the optical interface and careful balancing of the reduced piezoelectric coupling against the gains in modal confinement. The work establishes that phonon trapping in suspended XBARs is a viable route to achieving the necessary high $Q_m$ and small mode volumes required for quantum state transduction.