⚡ electrical engineering

Spurious-Free Lithium Niobate Bulk Acoustic Wave Resonator with Grounded-Ring Electrode

This paper presents a mechanically robust, 20 µm thick single-crystal lithium niobate bimorph PMUT with a 6.4% electromechanical coupling coefficient and 65 nm/V transmit efficiency, demonstrating stable operation up to 600°C and survival up to 900°C.

Original authors: Vakhtang Chulukhadze, Kristi Nguyen, Eric Stolt, Kilian Shambaugh, Weston Braun, Tzu-Hsuan Hsu, Osama Jameel, Juan Rivas-Davila, Ruochen Lu

Published 2026-04-09

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Vakhtang Chulukhadze, Kristi Nguyen, Eric Stolt, Kilian Shambaugh, Weston Braun, Tzu-Hsuan Hsu, Osama Jameel, Juan Rivas-Davila, Ruochen Lu

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

The Big Picture: Swapping Heavy Magnets for Tiny Crystals

Imagine you are trying to build a super-efficient, tiny power converter for your phone or a drone. Usually, these devices need big, heavy magnetic coils (inductors) to store energy and smooth out electricity. But as we try to make these devices faster and smaller, those magnetic coils become too bulky and inefficient.

Scientists have been trying to replace these heavy magnets with piezoelectric crystals (like a specific type of rock called Lithium Niobate). These crystals vibrate when you apply electricity, acting like a tiny, super-fast spring that stores energy. They are perfect for this job because they are small, light, and efficient.

The Problem:
Think of a piezoelectric crystal like a drum. When you hit it, it vibrates. You want it to vibrate in one specific, clean way (like a perfect drumbeat) to store energy efficiently. However, in real life, hitting the drum also causes it to wobble, rattle, and make weird, unwanted noises (called spurious modes).

In the world of power converters, these "weird noises" are disastrous. They act like potholes on a highway, causing energy loss and making the device inefficient. For years, scientists had to choose between a crystal that vibrated strongly (good for power) or one that vibrated cleanly (good for efficiency), but rarely both.

The Solution: The "Grounded-Ring" Trick

This paper introduces a clever new design that solves this problem. The researchers built a special electrode (a metal contact) that looks like a bullseye:

The Center: The active area where the magic happens.
The Gap: A tiny, empty ring of space surrounding the center.
The Ring: A metal ring on the outside that is connected to the ground (like a safety net).

The Analogy: The "Piston" vs. The "Wobbly Drum"

To understand why this works, imagine two scenarios:

1. The Old Way (The Wobbly Drum):
Imagine a large, flat drum skin. When you hit the center, the whole skin vibrates up and down. But because the edges are free to move, the vibration bounces back and forth, creating messy "standing waves" (ripples) that travel across the surface. These ripples are the spurious modes. They mess up the energy storage.

2. The New Way (The Piston):
Now, imagine the drum skin is surrounded by a rigid, heavy ring that is glued down tight. When you hit the center, the vibration cannot ripple out to the edges. Instead, the entire center section moves up and down together, like a solid piston in an engine.

This "piston-like" motion is perfect. It's smooth, predictable, and has no messy ripples. The "Grounded-Ring" design forces the crystal to behave like this piston.

How They Proved It

The researchers didn't just guess; they used two main tools to prove their theory:

Electrical Testing: They measured the electricity flowing in and out. The old designs showed jagged, messy lines (the "potholes"). The new design showed a smooth, clean line, proving the "potholes" were gone.
Laser Vibrometry (The "Slow-Motion Camera"): They used a super-precise laser to watch the crystal vibrate in slow motion.
- Old Design: The laser showed the crystal rippling and wobbling all over the place.
- New Design: The laser showed the center moving up and down in perfect unison, while the outer ring stayed still. It was a perfect "piston."

The Results: A Record-Breaking Performance

Because they eliminated the "wobbles," the new device is a champion:

Speed: It operates at 10.14 MHz (very fast).
Efficiency: It has a "Quality Factor" (Q) of 5,230. Think of this as how long a bell rings after you hit it. A higher number means it rings longer and loses less energy.
Power: It has a high "Coupling Coefficient" (29.6%), meaning it converts electricity to vibration (and back) very efficiently.

Why This Matters

This paper is a breakthrough because it finally allows us to use these tiny, high-speed crystals for power conversion (charging batteries, running motors) without the energy losses that used to make them impractical.

In summary: The researchers figured out how to put a "guard rail" (the grounded ring) around a vibrating crystal. This guard rail stops the messy vibrations from spreading, forcing the crystal to move like a smooth, powerful piston. This makes it possible to build smaller, faster, and more efficient power electronics for the future.

1. Problem Statement

Piezoelectric resonators are promising candidates for replacing bulky magnetic inductors in high-frequency DC-DC power converters due to their compact size and low loss. However, their practical application is severely limited by a critical trade-off:

High Electromechanical Coupling ( $k^2$ ): Essential for wide bandwidth and high power density.
Spurious Modes: High- $k^2$ materials (like Lithium Niobate, LN) often excite unwanted lateral acoustic modes (spurious modes) within the inductive operating band (between resonance $f_s$ and anti-resonance $f_p$ ).
Consequence: These spurious modes create sharp impedance spikes, introducing resistive losses, degrading converter efficiency, and limiting the usable voltage conversion ratio. Existing solutions often compromise material performance or fail to provide broadband suppression.

2. Methodology

The authors propose a novel Grounded-Ring Electrode architecture to suppress spurious modes while maintaining high performance.

Material & Mode Selection:
- Material: Single-crystal 36°Y-cut Lithium Niobate (LN), chosen for its high piezoelectric coupling coefficient ( $k^2_{M33}$ ) and low parasitic coupling.
- Mode: Thickness-Extensional (TE) mode, which offers better dispersion characteristics for spurious-free operation compared to Thickness-Shear (TS) modes.
Device Architecture:
- The resonator consists of a central active area (signal top electrode, ground bottom electrode).
- Surrounding the active area is a non-metalized sub-wavelength gap.
- Encircling the gap is a metalized ring connected to ground on both top and bottom surfaces.
Analysis Techniques:
- Simulation: 3D Finite Element Analysis (FEA) using COMSOL to model impedance, resistance, and mode shapes.
- Fabrication: Devices were fabricated on 300 µm thick 36°Y-cut LN wafers with 300 nm Gold electrodes.
- Characterization:
  - Electrical: Vector Network Analyzer (VNA) measurements to extract $f_s$ , $f_p$ , $k^2$ , and Bode Quality Factor ( $Q_{Bode}$ ).
  - Mechanical: Laser Doppler Vibrometry (LDV) to visualize surface velocity ( $v_z$ ) and displacement profiles across the device surface.
- Theoretical Modeling: Analysis of Lamb wave dispersion curves (S1 and A1 modes) to explain the suppression mechanism.

3. Key Contributions

Novel Grounded-Ring Design: Introduction of a grounded ring separated by a narrow gap to alter effective boundary conditions, forcing the device into a piston-like modal response rather than a standard TE mode with lateral standing waves.
Physical Mechanism Elucidation:
- The design utilizes the dispersion characteristics of LN. The gap width is engineered to be smaller than the decay length of evanescent A1 waves, while the ring width supports propagating S1 and A1 overtones.
- The interaction between the S1 (even symmetry) and A1 (odd symmetry) modes at the mid-plane creates a cancellation of lateral particle displacement ( $u_x$ ), effectively enforcing a "piston" boundary condition ( $u_x \approx 0$ ) that suppresses lateral spurious modes.
Geometric Design Rules: The paper establishes scaling rules linking the gap width and ring width to the Lamb wave dispersion characteristics (specifically decay lengths and wavelengths) to ensure spurious suppression.
Experimental Validation: Comprehensive validation using LDV confirms the shift from complex lateral standing waves (in conventional designs) to a uniform piston-like velocity profile in the proposed design.

4. Key Results

The fabricated device operates at 10.14 MHz and demonstrates the following performance metrics:

Spurious-Free Operation: The impedance spectrum shows a smooth resistance curve across the entire inductive band, with no sharp spikes from lateral overtones.
Electromechanical Coupling ( $k^2$ ): Achieved 29.6%, indicating that the suppression mechanism does not degrade the material's transduction efficiency.
Quality Factor ( $Q$ ):
- Maximum in-band Bode Quality Factor ( $Q_{Bode}$ ): 5230.
- (Note: While a rectangular reference showed a higher peak $Q$ of 7924, it occurred at a spurious resonance and was unusable for power conversion).
Figure of Merit (FoM): $Q \cdot k^2 = \mathbf{1548}$ .
Comparison: The device outperforms prior state-of-the-art LN and PZT resonators in terms of combining high frequency, high $k^2$ , and spurious-free operation.

5. Significance

Enabling Piezoelectric Power Conversion: This work removes the primary bottleneck (spurious modes) preventing the use of high-performance acoustic resonators in DC-DC converters. It allows for the realization of compact, high-power-density converters operating at MHz frequencies.
Design Framework: The paper provides a generalizable design framework based on Lamb wave dispersion. The grounded-ring concept can be scaled to different frequencies (by adjusting dimensions relative to plate thickness) and applied to other piezoelectric materials (e.g., Lithium Tantalate, AlN).
Broader Applications: Beyond power conversion, this spurious-free design approach is highly relevant for RF filters and sensors where clean impedance responses and high $Q$ are critical.

In conclusion, this paper successfully demonstrates that a grounded-ring electrode architecture can transform a standard high- $k^2$ LN resonator into a spurious-free, high-performance energy storage element, paving the way for the next generation of piezoelectric power electronics.