Diamond-on-chip magnetic field camera for mobile imaging

At a Glance

Metadata	Details
Publication Date	2025-03-12
Journal	Physical Review Applied
Authors	Julian M. Bopp, Hauke Conradi, Felipe Perona Martínez, Anil Palaci, Jonas Wollenberg
Institutions	Leibniz Institute of Surface Engineering, Humboldt-Universität zu Berlin
Citations	1
Analysis	Full AI Review Included

Executive Summary

This research introduces an integrated, fiber-packaged magnetic field camera utilizing Nitrogen-Vacancy (N-V) centers in diamond, designed for robust, mobile imaging applications.

Core Innovation: The camera employs a 3x3 pixel array defined by perpendicularly intersecting green pump (532 nm) and infrared (1042 nm) laser beams within a diamond substrate, enabling spatially resolved magnetic field sensing without bulky free-space optics or moving components.
Sensing Mechanism: Infrared Absorption Optically Detected Magnetic Resonance (IRA ODMR) is used, exploiting the N-V^- singlet transition (¹A₁ ↔ ¹E).
Performance (Multipixel): The best pixel achieved a magnetic-sensitive Amplitude Spectral Density (ASD) of 10.5 µT Hz^-1/2, with a pixel spacing of 500 µm.
Imaging Capability: The camera successfully reconstructed the position of a current-driven solenoid coil with high precision, yielding a position uncertainty of (30, 20, 10) µm (δx, δy, δz).
Benchmark Sensitivity: An optimized single-pixel reference setup demonstrated a significantly improved sensitivity of 44.3 nT Hz^-1/2, confirming the potential for detecting biological signals (e.g., neural activity).
Bandwidth Potential: The single-pixel setup resolved a 1 kHz ramp signal, indicating the possibility of achieving measurement bandwidths required for millisecond-regime neural sensing.
Future Outlook: Sensitivity can be enhanced by optimizing pump/IR beam overlap, increasing laser power (using nonabsorbing adhesives), and incorporating optical cavities or magnetic flux concentrators.

Technical Specifications

Parameter	Value	Unit	Context
Sensor Architecture	3 x 3	Pixels	Integrated array (one pixel excluded from analysis)
Pixel Spacing	500	µm	Determined by fiber diameter and integration
Pixel Diameter (D_px)	80	µm	Pump and infrared beam waists inside diamond
Diamond Substrate	(111)	-	HPHT CD1411 (Sumitomo Electric)
N-V Density (Multipixel)	3.0 x 10²³	m^-3	Substrate A
N-V Density (Single-Pixel)	1.1 x 10²³	m^-3	Substrate B
Ground State Splitting (D)	2.87	GHz	Zero-field splitting of N-V triplet ³A₂
Best Pixel Sensitivity (ASD)	10.5	µT Hz^-1/2	Pixel (3,3) of multipixel camera
Single-Pixel Sensitivity (ASD)	44.3	nT Hz^-1/2	Optimized free-space setup (0.8 Hz to cutoff)
Spin Dephasing Time (T₂*)	520(110)	ns	Derived from single-pixel sensitivity
Inhomogeneous Linewidth	6.4(3)	MHz	Pixel-averaged standard deviation
ODMR Contrast (Pixel Avg.)	23(16) x 10^-7	-	Multipixel sensor
Pump Wavelength	532	nm	Green excitation
Infrared Wavelength	1042	nm	Absorption detection
Microwave Power	16	W	Applied to AlN PCB inductor lines
Solenoid Position Uncertainty	(30, 20, 10)	µm	(δx, δy, δz) reconstruction accuracy
Time-Varying Field Resolution	1	kHz	Max frequency resolved in single-pixel setup

Key Methodologies

The magnetic field camera relies on a highly integrated, fiber-coupled platform and frequency modulation ODMR techniques.

Diamond Preparation:
- (111)-oriented HPHT diamond substrates (1.4 x 1.4 x 0.2 mm³) were irradiated with an electron beam and subsequently annealed to achieve high N-V ensemble densities (up to 3.0 x 10²³ m^-3).
- The four side facets of the diamond were polished to facilitate optical coupling.
Chip Integration and Optical Coupling:
- The diamond substrate and single-mode fibers (SMFs) were integrated into trenches etched into a polymer substrate, which covers a silicon submount for stability.
- Fibers were terminated with Graded-Index (GRIN) lenses to collimate the pump (532 nm) and infrared (1042 nm) beams, defining 80 µm beam waists inside the diamond.
- Pixels were defined by the intersection points of the pump (row) and infrared (column) beams, spaced 500 µm apart.
Microwave and Magnetic Field Application:
- Microwave inductor lines (three connected rings) were fabricated onto an AlN printed-circuit board and placed parallel to the diamond surface, centered over the active pixels.
- A 16 W microwave signal was frequency-modulated (f_mod = 20 kHz, f_dev = 10 MHz) to drive the N-V spin transitions.
- An offset magnetic field was applied along the [011] direction to split the N-V orientations into two ensembles, enabling magnetic field sensitivity via the Zeeman effect.
IRA ODMR Measurement and Readout:
- The IRA ODMR technique measures the change in infrared (1042 nm) absorption corresponding to spin flips between the m_s = 0 and m_s = ±1 ground states.
- A balanced photodetector (BPD) was used for common mode rejection (CMR) of laser intensity fluctuations, splitting 10% of the IR light to a reference path.
- The BPD difference signal was demodulated using a lock-in amplifier (LIA) with a 100 ms time constant to extract the ODMR signal U(f_c).
Imaging and Reconstruction:
- Magnetic field imaging was achieved by sequentially addressing pixels: turning on the pump beam for row i and the infrared beam for column j.
- The magnetic field B_i,j was calculated from the measured frequency splitting (Δf_i,j) of the N-V ensemble 2 resonances using the N-V gyromagnetic ratio (γ = 28 GHz/T).
- The camera’s position relative to an external solenoid was reconstructed by minimizing the difference between measured and simulated magnetic fields across the array.

Commercial Applications

This integrated, robust magnetic field camera technology is suitable for miniaturized, hand-held devices and integrated lab-on-a-chip platforms, targeting fields where high spatial resolution and mobility are critical.

Neuroscience and Medical Diagnostics:
- Detection of single-neuron action potentials and neural signals (requiring millisecond temporal resolution).
- Millimeter-scale magnetocardiography (MCG) and magnetoencephalography (MEG).
- Imaging of magnetic nanoparticles for biomedical applications (e.g., malarial hemozoin detection).
Energy and Industrial Inspection:
- Lifetime and quality assessment of lithium-ion batteries via non-invasive current density imaging.
- Imaging damage in steel and other materials using diamond magnetometers.
Microfluidics and Chemical Sensing:
- Two-dimensional Nuclear Magnetic Resonance (NMR) spectroscopy in microfluidic devices.
- Lab-on-a-chip platforms for drug discovery and chemical analysis.
Quantum Technology and Sensing:
- Miniaturized, adjustment-free magnetic field imagers for mobile quantum sensing.
- Spaceborne quantum magnetometry (due to robust, integrated design).
- Multipixel temperature sensing (by adapting the measurement procedure).

View Original Abstract

Integrated and fiber-packaged magnetic field sensors with a sensitivity sufficient to sense electric pulses propagating along nerves and a spatial resolution fine enough to resolve their propagation directions will trigger tremendous steps ahead in medical diagnostics and research. Nitrogen-vacancy centers in diamond are best suitable for such sensing tasks under ambient conditions. Current research on uniting a good sensitivity and high spatial resolution is facilitated by scanning or imaging techniques. However, these techniques employ moving parts or bulky microscopes. Both approaches cannot be miniaturized to build robust, adjustment-free, hand-held devices. In this work, we introduce concepts for spatially resolved magnetic field sensing and two-dimensional gradiometry with an integrated magnetic field camera. The camera utilizes infrared absorption optically detected magnetic resonance (IRA ODMR) mediated by perpendicularly intersecting infrared and pump laser beams forming a pixel matrix. We demonstrate our scalable <a:math xmlns:a=“http://www.w3.org/1998/Math/MathML” display=“inline” overflow=“scroll”><a:mn>3</a:mn><a:mo>×</a:mo><a:mn>3</a:mn></a:math> pixel sensor’s capability to reconstruct the position of an electromagnet. In a reference measurement, we show an IRA ODMR sensitivity of <d:math xmlns:d=“http://www.w3.org/1998/Math/MathML” display=“inline” overflow=“scroll”><d:mn>44</d:mn><d:mspace width=“0.1em”/><d:mi>nT</d:mi><d:mspace width=“0.1em”/><d:msup><d:mi>Hz</d:mi><d:mrow><d:mo>−</d:mo><d:mn>1</d:mn><d:mo>/</d:mo><d:mn>2</d:mn></d:mrow></d:msup></d:math>.

Tech Support

Original Source

DOI: https://doi.org/10.1103/physrevapplied.23.034024