Sensitive endoscopic diamond magnetometer for non-contact sensing in confined environments

This paper presents a miniaturized 6 mm endoscopic diamond magnetometer that overcomes the trade-off between sensor size and magnetic sensitivity by utilizing a multi-core fiber bundle and FPGA-based control to achieve high-resolution, real-time magnetic field imaging of lithium-ion batteries in unshielded, confined environments.

The Gist

Imagine trying to listen to a whisper in a crowded, noisy room, but you have to do it through a tiny, 6-millimeter-wide straw that you can't take apart. That is the challenge this paper tackles: building a super-sensitive magnetic sensor that is small enough to fit into tight spaces (like inside a battery or a machine) but still powerful enough to detect incredibly faint magnetic signals.

Here is a simple breakdown of what the researchers did and why it matters, using everyday analogies.

The Problem: The "Flashlight vs. Camera" Dilemma

Usually, to see something clearly with a quantum sensor (which uses tiny defects in diamonds called "NV centers"), you need two things:

1. A bright flashlight to excite the diamond.
2. A big camera lens to catch the faint light (fluorescence) bouncing back.

In the past, scientists had to choose:

• Option A: A big, bulky system with a great camera lens. It works perfectly but is too heavy to fit into tight spots.
• Option B: A tiny, endoscopic probe (like a medical camera). It fits anywhere, but because the "lens" (the fiber optic cable) is so small, it misses most of the light coming back from the diamond. It's like trying to catch rain with a thimble instead of a bucket. The signal is too weak to be useful.

The Solution: The "Split-Path" Straw

The researchers solved this by redesigning the straw. Instead of using one single fiber optic cable for both sending light in and catching light out, they used a bundle of fused fibers (like a bundle of straws glued together).

• The Center Straw: A very thin, single straw in the middle sends the laser light in. This keeps the "flashlight" beam tight and focused on a tiny spot.
• The Surrounding Straws: Four larger straws surround the center one. These act as the "bucket," catching the light bouncing back from the diamond.

The Analogy: Imagine a group of people in a dark room. One person (the center straw) shines a laser pointer at a specific spot on the wall. Instead of one person trying to catch the reflection, four other people (the outer straws) stand around with wide nets to catch the light. This allows them to use a tiny probe head (6mm wide) while still catching enough light to make a clear measurement.

The "Brain" of the Operation

The sensor head is just the tip of the iceberg. It is connected by a long cable to a "brain" (a computer chip called an FPGA) sitting far away.

• Real-Time Tracking: Usually, to measure a magnetic field, you have to scan through a whole range of frequencies, which is slow. This system acts like a smart autopilot. It constantly adjusts the frequency to stay locked onto the exact magnetic signal it's looking for. It doesn't need to scan the whole range; it just "follows the needle." This makes the measurement fast and robust, even if the environment is noisy or the object being measured is moving.

The Real-World Test: The Battery Detective

To prove this works, the team used their sensor to look inside a commercial lithium-ion battery (the kind found in phones and laptops) while it was charging and discharging.

• The Challenge: Batteries are metal and crowded with parts. They create their own magnetic noise. Also, you can't put a giant lab machine inside a battery pack.
• The Result: The sensor was placed just 2 millimeters away from the battery surface. It successfully mapped the invisible flow of electricity inside the battery without touching it.
• The Outcome: They created a "heat map" of the electric current. They could see exactly where the electricity was flowing in and out, and how it changed when the battery was charging versus discharging.

Why This Matters

This paper demonstrates that you don't have to sacrifice sensitivity for size anymore.

• Before: You had big, sensitive sensors OR small, weak sensors.
• Now: You have a small sensor (6mm wide) that is sensitive enough to detect magnetic fields as weak as 91 picotesla (that's a trillionth of a Tesla) in a noisy, unshielded room.

In short: They built a "magnetic stethoscope" that is small enough to peek into tight, crowded places but sensitive enough to hear the faintest whispers of electricity flowing through a battery, all without needing a giant, expensive lab setup.

Technical Summary

Technical Summary: Sensitive Endoscopic Diamond Magnetometer for Non-Contact Sensing in Confined Environments

Problem Statement
The transition of quantum magnetometry from laboratory settings to real-world applications is hindered by a persistent trade-off between sensor miniaturization and magnetic sensitivity. While bulky systems achieve high sensitivity, endoscopic probes typically suffer from inefficient fluorescence collection, leading to reduced performance. Existing miniaturized Nitrogen-Vacancy (NV) magnetometer designs face a physical bottleneck: separating the remote optical/electronic backend from the sensing probe via optical fibers limits the collection of NV fluorescence due to the mismatch between the large optical etendue of the diamond and the limited etendue of the fiber. Single-fiber solutions struggle to increase collection efficiency without expanding the excitation volume, while integrating detectors into the probe head increases size and complexity, negating the benefits of an endoscopic form factor.

Methodology
The authors present a miniaturized diamond quantum magnetometer designed to resolve this trade-off through a novel optical architecture and a compact digital backend.

• Sensor Head Architecture: The core innovation is a 6 mm diameter sensor head utilizing a fused multi-core fiber bundle. Instead of a single fiber, the design separates excitation and collection into distinct cores. A small central fiber (14 µm core, 0.12 NA) delivers 532 nm excitation light to a tightly confined volume within the diamond. Four surrounding larger fibers (200 µm core, 0.22 NA) collect the resulting NV fluorescence. A custom high-numerical-aperture micro-objective, composed of a 0.25-pitch gradient-index (GRIN) lens and a 1 mm spherical lens, couples the fiber bundle to the diamond. This geometry maximizes fluorescence collection without requiring detectors or bulk optics at the probe tip, preserving the probe's deployability in confined spaces.
• Remote Backend and Control: The optical and electronic components (laser, detector, microwave synthesis) are housed in a remote backend connected to the sensor via the fiber bundle and electrical wiring. Microwave control is generated by a Red Pitaya FPGA platform.
• Real-Time Tracking: To enable fast magnetic imaging, the system employs a digital closed-loop frequency tracking scheme. The FPGA generates a frequency-modulated microwave signal and performs digital lock-in detection. A discrete integral controller continuously updates the microwave carrier frequency to track the Optically Detected Magnetic Resonance (ODMR) zero-crossing in real time. This eliminates the need for repeated full-spectrum ODMR acquisitions at each scan point, allowing for continuous magnetic field measurement even when the ODMR contrast or linewidth varies (e.g., near ferromagnetic materials).
• Experimental Setup: The sensor was tested in a magnetically unshielded environment. For characterization, an external bias field was aligned along the diamond [100] direction to address all four NV orientations simultaneously, enhancing optical contrast.

Key Results

• Sensitivity: The sensor achieved an open-loop magnetic-field sensitivity of 91 pT/√Hz over a 2 kHz measurement bandwidth in an unshielded environment. This performance was measured with the bias field aligned to the [100] direction. A single-axis [111] alignment yielded 169 pT/√Hz.
• Performance Comparison: The device occupies a distinct region in the sensitivity-diameter landscape, combining sub-100 pT/√Hz sensitivity with a 6 mm endoscopic sensor head. It represents the lowest reported noise floor for an endoscopic, fiber-coupled NV sensor head below 10 mm in diameter.
• Battery Imaging Application: The sensor was used to image the magnetic field of a commercial lithium-ion pouch cell during charge and discharge cycles without electrical contact. The system successfully mapped the stray fields, distinguishing between the ferromagnetic nickel anode tab and the non-magnetic aluminum cathode tab.
• Current Density Reconstruction: Using the measured magnetic field maps, the authors reconstructed depth-integrated sheet-current density maps ($K$) inside the battery. The reconstruction utilized a real-space regularized inversion method, revealing large-scale current flow patterns consistent with the battery geometry, including current spreading from the cathode tab and converging toward the anode tab during charging.

Significance and Claims
The paper claims to demonstrate the first endoscopic NV magnetometer that successfully addresses the fluorescence collection bottleneck by separating excitation and collection paths within a fused multi-core fiber bundle. This architecture enables high sensitivity without compromising the probe's small form factor.

The authors assert that their platform combines high sensitivity, a probe geometry suitable for confined and crowded environments, and real-time tracking capabilities. This combination opens new avenues for non-destructive testing and in-operando diagnostics, specifically demonstrated here through the quantitative magnetic imaging of current flow in packaged lithium-ion batteries in unshielded settings. The work suggests that such sensors can operate effectively where traditional quantum magnetometers (requiring cryogenics or shielding) and other compact sensors (lacking sensitivity) cannot, particularly in applications involving dynamic magnetic environments or spatially restricted access.