⚛️ quantum physics

Quantum electrometry in a silicon carbide power device

This paper demonstrates that silicon vacancies in silicon carbide power devices function as unique quantum sensors capable of mapping arbitrary high electric fields with high spatial resolution, enabling the detection of fields up to ~2.3 MV/cm to facilitate early failure diagnosis and data-driven reliability improvements.

Original authors: Yuichi Yamazaki, Akira Kiyoi, Naoyuki Kawabata, Yuki Watanabe, Ryosuke Akashi, Shunsuke Daimon, Nobumasa Miyawaki, Yu-ichiro Matsushita, Makoto Kohda, Takeshi Ohshima

Published 2026-03-17

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Yuichi Yamazaki, Akira Kiyoi, Naoyuki Kawabata, Yuki Watanabe, Ryosuke Akashi, Shunsuke Daimon, Nobumasa Miyawaki, Yu-ichiro Matsushita, Makoto Kohda, Takeshi Ohshima

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 fix a high-performance electric car, but the engine is made of a super-strong material called Silicon Carbide (SiC). This material is amazing because it can handle huge amounts of electricity and heat, making it perfect for electric vehicles and AI data centers.

However, there's a problem: when these devices get stressed, tiny "hot spots" or weak points can form inside them. If you don't find these weak spots early, the whole engine could explode (or in technical terms, "break down").

The problem is that these weak spots are microscopic, and the electric fields inside are so intense that traditional tools can't see them without breaking the device or averaging out the details. It's like trying to find a single crack in a brick wall by looking at the whole wall from a mile away.

The Solution: A Quantum "Flashlight"

This paper introduces a brilliant new tool: a Quantum Electrometer using a specific defect in the silicon carbide called a Silicon Vacancy (VSi).

Here is how it works, using some simple analogies:

1. The "Silicon Vacancy" is a Tiny, Sensitive Antenna

Think of the Silicon Carbide crystal as a perfectly organized dance floor where every dancer (atom) is holding hands in a specific pattern. A "Silicon Vacancy" is like a dancer who is missing from the floor. This empty spot creates a unique "dance move" (a quantum spin state) that is incredibly sensitive to the music (the electric field).

When an electric field pushes on this empty spot, the dancer changes their rhythm. By shining a laser on this spot and listening to the rhythm (using a technique called ODMR), scientists can tell exactly how hard the electric field is pushing.

2. The "All-Direction" Superpower

Previous tools (like the "Divacancy") were like flashlights that only shine in one direction. If the electric field was pushing from the side, the flashlight couldn't see it well. They were also picky about temperature, needing to be frozen in ice to work.

The new Silicon Vacancy (VSi) tool is different. It's like a 360-degree omnidirectional radar.

Isotropic Sensitivity: It reacts equally whether the electric field is pushing from the top, bottom, or side. This is crucial because in complex power devices, electricity flows in all sorts of weird directions.
Heat Resistant: Unlike its cousins, this sensor works perfectly at room temperature and even gets hotter (up to 300°C) without losing its ability to "hear" the electric field.

3. The "Surgical" Approach

How do you put this sensor inside a working device without breaking it?
The researchers used a Particle Beam Writer (PBW). Imagine a super-precise paintbrush that shoots tiny helium ions. Instead of painting the whole wall, they use this brush to "paint" tiny dots of these sensors only where they need to look.

They created a grid of these sensors at different depths inside the chip.
Because they only made tiny dots, they didn't damage the rest of the device. It's like placing a few listening devices in a room without tearing down the walls.

4. The Record-Breaking Test

The team tested this on a device that was being pushed to its absolute limit.

The Stress Test: They applied a voltage so high that it was 90% of the point where the device would normally explode (2.3 million Volts per centimeter).
The Result: The sensors didn't break, and they didn't stop working. They successfully mapped the electric field right up to the edge of disaster.

Why Does This Matter?

Think of this technology as giving engineers X-ray vision for electric power chips.

Before: Engineers had to guess where the weak spots were or wait for the device to fail.
Now: They can see exactly where the electric field is too strong, find the "hot spots" before they cause a fire, and design better, safer, and more efficient devices for our electric cars and power grids.

In a Nutshell:
The researchers found a tiny, naturally occurring "hole" in a silicon crystal that acts like a super-sensitive, 360-degree electric field detector. They learned how to place these detectors inside working power chips without breaking them, allowing them to map dangerous electric fields with incredible precision, even when the chips are running at their breaking point. This is a giant leap toward making our future electronics safer and more reliable.

1. Problem Statement

The reliability of high-bias Silicon Carbide (SiC) power devices (e.g., SiC MOSFETs, super-junction devices) is critical for modern energy systems. Failures often originate from localized distortions in the internal electric field distribution, such as hotspots, premature breakdown, or trap-induced leakage.

Limitations of Current Methods: Conventional techniques like Kelvin probe force microscopy (KPFA), differential phase contrast STEM, and nonlinear optical approaches suffer from macroscopic spatial averaging, invasive contact requirements, or insufficient sensitivity to nanoscale field gradients.
Limitations of Existing Quantum Sensors: While quantum sensors based on point defects (spin defects) offer high sensitivity and resolution, previous candidates like the divacancy ( $V_{Si}V_C$ ) in SiC have intrinsic anisotropy. They respond strongly to electric fields parallel to their defect axis but weakly to perpendicular components. Furthermore, they often require cryogenic temperatures or high-temperature annealing, making them unsuitable for in-situ diagnostics of operating power devices.
The Gap: There is a lack of a method capable of mapping arbitrary electric field vectors (both parallel and perpendicular to the crystal $c$ -axis) with high spatial resolution under the extreme electric fields ( $\sim$ 2 MV/cm) typical of operating SiC power devices.

2. Methodology

The authors utilized the Silicon Vacancy ( $V_{Si}$ ) defect in 4H-SiC as a quantum sensor.

Sensor Selection: Unlike the anisotropic divacancy, $V_{Si}$ is a monovacancy with $C_{3v}$ symmetry. Theoretical analysis suggests it possesses an isotropic electrostatic environment, potentially offering equivalent sensitivity to electric fields parallel ( $E_{\parallel}$ ) and perpendicular ( $E_{\perp}$ ) to the $c$ -axis.
Device Fabrication:
- A pn diode with edge termination was fabricated on a 4°-off n-type SiC substrate.
- Selective Formation: An ensemble of $V_{Si}$ defects was created at specific locations using Particle Beam Writing (PBW) with a focused Helium ion beam. This allowed for the creation of "dots" of $V_{Si}$ at precise depths (2.1 to 8.1 $\mu$ m) without degrading the overall device performance.
- Verification: Current-Voltage (I-V) characteristics were measured before and after PBW to confirm no device degradation occurred.
Measurement Setup:
- The device was biased under reverse voltage (up to 1500 V) to generate high internal electric fields.
- The setup was immersed in Fluorinert (an insulating liquid) to prevent high-voltage arcing while maintaining optical transparency.
- ODMR (Optically Detected Magnetic Resonance): A confocal microscope with a 785 nm laser and a 900 nm long-pass filter was used. RF fields were applied via an enameled wire to manipulate the spin states.
- Simulation: Technology Computer-Aided Design (TCAD) simulations were used to model the expected electric field distributions, including corrections for the presence of the RF wire and the dielectric properties of the immersion fluid.

3. Key Contributions

Demonstration of Isotropic Sensitivity: The paper experimentally proves that $V_{Si}$ exhibits comparable electric dipole moments for both parallel ( $d_{\parallel}$ ) and perpendicular ( $d_{\perp}$ ) field components. This is a unique feature distinguishing it from other spin defects (like $V_{Si}V_C$ or NV centers), enabling the detection of arbitrary electric field orientations.
Operation Near Breakdown Limits: The sensor successfully operated in electric fields up to $\sim$ 2.3 MV/cm, which is approximately 90% of the breakdown electric field of the specific 4H-SiC sample. This demonstrates the sensor's robustness under extreme operating conditions.
High-Resolution 3D Mapping: By varying the ion energy during PBW, the authors mapped electric field distributions at different depths within the device, achieving nanometer-scale spatial resolution.
Temperature Stability: The study highlights that the zero-field splitting parameter ( $D$ ) of $V_{Si}$ remains stable from room temperature up to $\sim$ 590 K, allowing for accurate electric field measurements without complex temperature correction, even in areas with potential local heating.

4. Key Results

Dipole Moment Determination:
- Parallel Component ( $d_{\parallel}$ ): Measured at a point with minimal perpendicular field. The value was determined to be $-15.0 \pm 0.5$ MHz/(MV/cm).
- Perpendicular Component ( $d_{\perp}$ ): Measured at a point with a significant perpendicular field ratio ( $E_{\perp}/E_{\parallel} \approx 0.19$ ). The ratio $|d_{\perp}/d_{\parallel}|$ was found to be $1.1 \pm 0.06$ .
- Significance: The near-unity ratio confirms that $V_{Si}$ responds almost equally to fields in all directions, enabling full 3D vector reconstruction.
High-Field Stability: ODMR spectra remained linear and stable even at 1500 V bias ( $\sim$ 2.3 MV/cm). The experimental data matched theoretical predictions, confirming that the dipole moments do not change under high stress.
Field Distribution Mapping:
- Experimental field maps matched TCAD simulations well in regions far from external conductors.
- Discrepancies near the surface were successfully attributed to the perturbation caused by the enameled RF wire and the dielectric constant of the Fluorinert. Simulations including these factors perfectly reproduced the experimental data, validating the measurement technique's ability to detect field distortions.
Comparison with Theory: First-principles calculations (DFT) predicted $|d_{\perp}/d_{\parallel}| \approx 1.0$ , which aligns closely with the experimental value of 1.1.

5. Significance and Impact

Device Diagnostics: This work provides the first practical method for non-invasive, high-resolution, 3D mapping of electric fields inside operating high-power SiC devices. This is crucial for identifying failure mechanisms like hotspots and leakage paths that are invisible to conventional tools.
Next-Gen Power Electronics: The ability to measure fields up to the material's breakdown limit allows for the optimization of cutting-edge device architectures (e.g., trench gate MOSFETs and super-junction devices) where lateral field control is critical.
Data-Driven R&D: The technique enables a shift toward data-driven development, allowing engineers to validate TCAD models against real-world in-situ data and accelerate the design of more reliable power electronics.
Fundamental Physics: It establishes $V_{Si}$ as a superior quantum sensor for electrometry due to its unique isotropic response and high-temperature stability, overcoming the limitations of previous spin-defect sensors.

In conclusion, the paper demonstrates that selectively formed $V_{Si}$ ensembles serve as robust, isotropic quantum sensors capable of mapping electric fields in SiC power devices with nanometer precision, even under extreme bias conditions approaching material breakdown.