Microwave-acoustic-based isolated gate driver for power electronics
Pith reviewed 2026-05-17 20:54 UTC · model grok-4.3
The pith
A lithium niobate SAW device delivers galvanic isolation of 2.75 kV with 0.032 pF capacitance for gate drivers.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
We demonstrate a mechanically-isolated gate driver based on microwave-frequency surface acoustic wave (SAW) device on lithium niobate that achieves galvanic isolation of 2.75 kV with ultralow isolation capacitance (0.032 pF) over 1.25 mm mechanical propagation length, delivering 13.4 V open-circuit voltage and 44.4 mA short-circuit current. We demonstrate isolated gate driving for a gallium nitride (GaN) high-electron-mobility transistor, achieving a turn-on time of 108.8 ns comparable to commercial drivers and validate its operation in a buck converter. In addition, our SAW device operates over an ultrawide temperature range from 0.5 K to 544 K.
What carries the argument
The microwave-frequency surface acoustic wave (SAW) device on lithium niobate, which propagates mechanical waves across a 1.25 mm gap to achieve galvanic isolation while keeping capacitance low.
If this is right
- Isolated gate driving of GaN transistors reaches 108.8 ns turn-on time matching commercial drivers.
- The driver functions inside a buck converter topology.
- Performance holds across an ultrawide temperature window from 0.5 K to 544 K.
- Acoustic transmission supplies inherent immunity to electromagnetic interference.
- The structure supports heterogeneous integration onto multiple semiconductor platforms for compact designs.
Where Pith is reading between the lines
- The low capacitance could lower switching losses when the driver is paired with very high-frequency converters.
- Cryogenic operation opens possible use in superconducting or quantum control circuits.
- Mechanical isolation may improve long-term reliability in high-vibration industrial settings provided the substrate survives.
- The same acoustic channel might be adapted for bidirectional power and signal exchange in other isolated converter topologies.
Load-bearing premise
The 1.25 mm mechanical gap and lithium niobate substrate will maintain reliable galvanic isolation and mechanical integrity under repeated high-voltage stress, thermal cycling, and vibration in real power-electronic environments.
What would settle it
Subject the fabricated device to repeated 2.75 kV bias while cycling temperature between 0.5 K and 544 K and measure whether breakdown occurs or isolation capacitance rises above 0.032 pF.
read the original abstract
Electrical isolation is critical to ensure safety and minimize electromagnetic interference (EMI), yet existing methods struggle to simultaneously transmit power and signals through a unified channel. Here we demonstrate a mechanically-isolated gate driver based on microwave-frequency surface acoustic wave (SAW) device on lithium niobate that achieves galvanic isolation of 2.75 kV with ultralow isolation capacitance (0.032 pF) over 1.25 mm mechanical propagation length, delivering 13.4 V open-circuit voltage and 44.4 mA short-circuit current. We demonstrate isolated gate driving for a gallium nitride (GaN) high-electron-mobility transistor, achieving a turn-on time of 108.8 ns comparable to commercial drivers and validate its operation in a buck converter. In addition, our SAW device operates over an ultrawide temperature range from 0.5 K (-272.6 {\deg}C) to 544 K (271 {\deg}C). The microwave-frequency SAW devices offer inherent EMI immunity and potential for heterogeneous integration on multiple semiconductor platforms, enabling compact, high-performance isolated power and signal transmission in advanced power electronics.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript demonstrates a mechanically-isolated gate driver based on microwave-frequency surface acoustic wave (SAW) devices fabricated on lithium niobate. It reports galvanic isolation of 2.75 kV with 0.032 pF isolation capacitance across a 1.25 mm mechanical gap, delivering 13.4 V open-circuit voltage and 44.4 mA short-circuit current. The device is validated by driving a GaN HEMT with 108.8 ns turn-on time in a buck converter and is shown to operate from 0.5 K to 544 K while providing inherent EMI immunity.
Significance. If the short-term performance metrics prove reproducible and the mechanical isolation remains reliable under sustained high-voltage bias, this approach could provide a compact, low-capacitance alternative to conventional isolation techniques with added benefits of wide-temperature operation and EMI immunity. The functional buck-converter demonstration and cryogenic-to-high-temperature data are concrete strengths that support practical relevance in power electronics.
major comments (2)
- [Abstract and results sections] Abstract and results sections: the reported isolation voltage (2.75 kV), capacitance (0.032 pF), and current/voltage delivery values are presented without error bars, repeatability statistics, or explicit measurement protocols, which limits independent verification of the quantitative performance claims.
- [Device characterization and application sections] Device characterization and application sections: no accelerated-life, partial-discharge, or post-stress re-characterization data are supplied after combined 2.75 kV bias, thermal cycling, and mechanical vibration, leaving the long-term integrity of the 1.25 mm lithium niobate gap untested despite its centrality to the galvanic-isolation claim.
minor comments (2)
- [Abstract] The temperature range statement in the abstract mixes kelvin and Celsius with inconsistent formatting; a uniform presentation would improve clarity.
- [Introduction] Additional references to prior acoustic or SAW-based isolation work would better situate the novelty relative to existing literature.
Simulated Author's Rebuttal
We thank the referee for their constructive comments, which help clarify the presentation of our results. We address each major point below and indicate revisions made to the manuscript.
read point-by-point responses
-
Referee: [Abstract and results sections] Abstract and results sections: the reported isolation voltage (2.75 kV), capacitance (0.032 pF), and current/voltage delivery values are presented without error bars, repeatability statistics, or explicit measurement protocols, which limits independent verification of the quantitative performance claims.
Authors: We agree that error bars, repeatability statistics, and explicit protocols would strengthen verifiability. In the revised manuscript we have added a dedicated Methods subsection describing the measurement setup, equipment, and protocols for the isolation voltage, capacitance, and drive current/voltage characterizations. Where repeated measurements on the same device were performed, we now include error bars or standard deviations. The reported values remain representative single-device results from the proof-of-concept fabrication run; a multi-device statistical study lies outside the scope of this initial demonstration but is noted as future work. revision: partial
-
Referee: [Device characterization and application sections] Device characterization and application sections: no accelerated-life, partial-discharge, or post-stress re-characterization data are supplied after combined 2.75 kV bias, thermal cycling, and mechanical vibration, leaving the long-term integrity of the 1.25 mm lithium niobate gap untested despite its centrality to the galvanic-isolation claim.
Authors: We acknowledge that long-term reliability data under combined electrical, thermal, and mechanical stress would be valuable for validating the mechanical gap integrity. The present work focuses on initial high-voltage performance and operation from 0.5 K to 544 K; no accelerated-life, partial-discharge, or post-stress re-characterization results are available. We have added a short limitations paragraph in the Discussion section that explicitly states this gap and identifies combined-stress reliability testing as an important direction for subsequent studies. revision: no
- No accelerated-life, partial-discharge, or post-stress re-characterization data after combined 2.75 kV bias, thermal cycling, and mechanical vibration
Circularity Check
No significant circularity: experimental demonstration rests on direct measurements
full rationale
This is an experimental device paper reporting measured performance of a SAW-based isolated gate driver on lithium niobate. The central claims (2.75 kV isolation, 0.032 pF capacitance over 1.25 mm gap, 13.4 V / 44.4 mA outputs, 108.8 ns turn-on, buck-converter validation, and 0.5–544 K operation) are presented as direct experimental results with no equations, fitted parameters, or derivations shown in the provided text. No self-definitional loops, fitted-input predictions, load-bearing self-citations, uniqueness theorems, or ansatz smuggling appear. The derivation chain is absent; results are self-contained against external benchmarks via physical testing.
Axiom & Free-Parameter Ledger
axioms (1)
- standard math Surface acoustic waves propagate on lithium niobate with known velocity and coupling coefficients under standard piezoelectric theory.
Reference graph
Works this paper leans on
-
[1]
AI load dynamics–a power electronics perspective,
The 223-MHz input RF signals are generated by a vector signal generator (Rohde & Sch warz SMW200A). The RF signals are then modulated by square waves from an arbitrary waveform generator (Rigol DG 852 Pro) . The modulated signals are amplified by 53 dB using a high power RF amplifier (Mini-circuits ZHL-10M2G0020+) before delivering to the input port of ou...
-
[2]
Why Small, Cold and Quiet DC-DC Conversion is Impossible
1484-1489 (IEEE). 14 Rothmund, D., Bortis, D. & Kolar, J.W. Highly compact isolated gate driver with ultrafast overcurrent protection for 10 kV SiC MOSFETs. CPSS Transactions on Power Electronics and Applications 3, 278-291 (2018). 15 Huang, Z., Wong, S.C. & Tse, C.K. Control Design for Optimizing Efficiency in Inductive Power Transf er Systems. IEEE Tran...
work page internal anchor Pith review Pith/arXiv arXiv 2018
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.