🔬 applied physics

Single-Crystal AlN Wafer-Based Bulk Acoustic Resonators for Piezoelectric Power Conversion

This paper presents the first single-crystal aluminum nitride (AlN) wafer-based bulk acoustic resonator for piezoelectric power conversion, demonstrating high quality factors and coupling coefficients alongside effective spurious mode suppression, thereby establishing AlN as a promising, thermally robust platform for next-generation power electronic systems.

Original authors: Ziqian Yao, Clarissa Daniel, Kaicheng Pan, Tzu-Hsuan Hsu, Heather Chang, Mark S Goorsky, Juan Rivas-Davila, Ruochen Lu

Published 2026-03-23

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Ziqian Yao, Clarissa Daniel, Kaicheng Pan, Tzu-Hsuan Hsu, Heather Chang, Mark S Goorsky, 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

Imagine you have a very old, bulky, and heavy transformer in your power supply. It's like a giant, clunky brick that takes up a lot of space and gets hot. Engineers have been trying to replace this brick with something tiny, efficient, and cool-running. They found a way to use sound waves (specifically, vibrations) instead of magnetic fields to move electricity around. This is called Piezoelectric Power Conversion.

Think of it like a musical instrument. When you pluck a guitar string, it vibrates at a specific note. If you can make that vibration move electricity efficiently, you can build a power converter that is the size of a coin instead of a brick.

However, there's a catch. Most materials used to make these "electric guitars" have a problem: they overheat.

The Problem: The "Hot Rock" vs. The "Cool Metal"

In the past, scientists used materials like PZT (a ceramic) or Lithium Niobate (a crystal) to make these resonators.

The Old Materials: Imagine trying to run a marathon while wearing a heavy winter coat. These materials generate heat when they vibrate, but they are terrible at letting that heat escape. It's like trying to cool down a hot rock by blowing on it; the heat gets trapped inside, causing the device to drift out of tune or even break.
The New Material (AlN): This paper introduces a new champion: Single-Crystal Aluminum Nitride (AlN). Imagine AlN as a high-tech heat sink or a super-conductive copper pipe. It doesn't just vibrate; it's incredibly good at spreading heat away instantly.

The Innovation: A "Silent Ring"

The researchers built a new device using a wafer of this super-cool AlN crystal. But they faced a new challenge: Spurious Modes.

The Analogy: Imagine shouting in a canyon. You want to hear one clear echo (the main note). But if the canyon walls are uneven, you might hear weird, jumbled echoes bouncing off the side rocks. These are "spurious modes"—unwanted noise that messes up the power conversion.
The Solution: The team added a grounded ring around the edge of the device. Think of this like putting a sound-dampening fence around the canyon. It traps the unwanted side echoes and forces the sound to stay focused on the main note. This ensures the device sings a pure, strong tone without getting confused by its own noise.

The Results: A High-Performance Engine

The team tested their new "electric guitar string" and found some impressive stats:

It's Efficient: It has a high "Quality Factor" (Q). In our analogy, this means the string vibrates for a long time with very little energy loss. It's like a bell that rings clearly for minutes instead of fading instantly.
It's Tunable: It converts electricity to vibration (and back) very effectively.
It Stays Cool: Because AlN is so good at conducting heat, this device can handle much higher power levels without melting or going out of tune. It's the difference between a car engine that needs a massive radiator and one that runs cool on a simple breeze.

Why This Matters

This isn't just about making a better speaker. This is about the future of electronics.

Smaller Gadgets: Because these resonators are tiny and efficient, we could eventually see power converters in your phone or laptop that are 10 times smaller than what we have today.
Better Efficiency: Less energy is wasted as heat, meaning your devices could last longer on a battery.
Robustness: Because it handles heat so well, it could be used in extreme environments, like inside a jet engine or a deep-space probe, where other materials would fail.

In a nutshell: The researchers took a material that is naturally excellent at staying cool (AlN), shaped it into a precise vibrating crystal, and added a "noise-canceling fence" (the grounded ring) to keep it singing a perfect note. The result is a tiny, powerful, and heat-resistant engine for the next generation of electronics.

1. Problem Statement

Piezoelectric power conversion is an emerging technology aimed at replacing bulky inductors in power converters with miniature acoustic resonators to improve efficiency and power density. However, existing state-of-the-art (SoA) piezoelectric materials face significant limitations for high-power applications:

Lead Zirconate Titanate (PZT): Offers high electromechanical coupling ( $k_t^2$ ) but suffers from nonlinearity and poor thermal robustness.
Lithium Niobate (LN): Provides high figures of merit (FoM) and power handling but has a large temperature coefficient of frequency (TCF $\approx$ -70 ppm/K). This causes self-heating-induced frequency drift, often requiring external cooling.
Lithium Tantalate (LT): Improved thermal stability over LN, but still suffers from low thermal conductivity, leading to heat accumulation at high power levels.
General Issue: Most incumbent materials (PZT, LN, LT, ScAlN) have low thermal conductivity (single-digit W·m⁻¹·K⁻¹), making them prone to localized hot spots and thermally driven frequency drift during sustained high-power operation.

2. Methodology

The authors propose utilizing single-crystal Aluminum Nitride (AlN) wafers to address thermal accumulation issues. The methodology involved three main stages:

Material Selection: The team selected single-crystal AlN due to its wurtzite crystal structure, strong bonding, and light constituent atoms, which theoretically provide intrinsically high thermal conductivity (hundreds of W·m⁻¹·K⁻¹) compared to other piezoelectrics.
Device Design & Simulation:
- A Thickness-Extensional (TE) mode bulk acoustic resonator was designed on a 0.4-mm-thick (0001)-oriented AlN wafer.
- A grounded ring structure was implemented around the central drive electrode to suppress in-band spurious modes. This design was adapted from previous Lithium Niobate (LN) resonators.
- Finite Element Analysis (FEA) using COMSOL was performed to simulate electrical admittance and mechanical displacement, predicting a fundamental $S_1$ mode at ~13.66 MHz with $k_t^2 \approx 7.52\%$ .
Fabrication & Integration:
- Devices were fabricated on 2-inch single-crystal AlN wafers (HexaTech) using standard cleanroom processes (photolithography, electron-beam evaporation of Cr/Au electrodes).
- The top electrode featured a 14.6-mm central drive region surrounded by a 0.9-mm grounded ring.
- Devices were mounted on FR-4 PCBs using UV-cure epoxy, with the active area suspended in air to minimize acoustic leakage and enable efficient TE-mode excitation.

3. Key Contributions

First Demonstration: This work presents the first single-crystal AlN wafer-based TE-mode bulk acoustic resonator specifically targeted for piezoelectric power conversion.
Spurious Mode Suppression: The successful implementation and experimental validation of a grounded ring electrode structure to effectively suppress in-band spurious modes in AlN resonators.
Thermal Advantage: Establishing single-crystal AlN as a superior platform for power conversion due to its orders-of-magnitude higher thermal conductivity compared to PZT, LN, and LT, promising better heat dissipation and frequency stability.

4. Experimental Results

Resonance Performance: The fabricated device exhibited a clear resonance at 13.52 MHz.
Quality Factor ( $Q$ ): The device achieved a high series resonance 3-dB $Q$ of 1677. While lower than the simulated $Q$ of 2000 (due to electrode resistance, boundary imperfections, and packaging losses), it remains competitive for low-MHz TE-mode operation.
Coupling Coefficient ( $k_t^2$ ): The measured electromechanical coupling coefficient was 6.1%.
Spurious Suppression: The frequency-domain impedance response showed a spurious-free spectrum above the resonance, confirming the effectiveness of the grounded ring design in mitigating parasitic coupling.
Figure of Merit (FoM): Despite a lower $k_t^2$ compared to PZT/LN, the combination of $k_t^2$ and $Q$ yields a competitive FoM and $f \cdot Q$ product.
Thermal Properties: While not fully tested in this specific paper, the material properties of single-crystal AlN (high thermal conductivity) are highlighted as the primary differentiator for future high-power stability.

5. Significance and Future Outlook

Thermal Robustness: The high thermal conductivity of single-crystal AlN is expected to significantly reduce self-heating and frequency drift during sustained high-power operation, addressing a critical bottleneck in current piezoelectric power converters.
Scalability: The use of wafer-scale single-crystal AlN offers a path toward compact, robust, and thermally stable power electronic systems.
Future Work: The authors plan to:
- Optimize wafer surface quality (e.g., double-side polishing) to further reduce scattering-induced parasitics.
- Conduct rigorous thermal cycling and large-signal power-handling tests to quantify frequency drift, $Q$ stability, and breakdown limits.
- Integrate the resonators into full power converter circuits to demonstrate real-world efficiency and power density improvements.

In conclusion, this paper validates single-crystal AlN as a promising, thermally superior alternative to traditional piezoelectric materials for next-generation high-power acoustic resonators.