Quantum Sensor Arrays Overview

Updated 13 December 2025

Quantum sensor arrays are engineered networks of spatially distributed quantum systems that use coherence, entanglement, and collective effects to surpass classical detection limits.
They employ a variety of platforms—such as NV centers, semiconductor quantum dots, optical tweezers, and SPAD arrays—with advanced readout, multiplexing, and machine-learning integration.
Applications include quantum imaging, biosensing, nanoscale magnetometry, and fundamental physics searches, demonstrating high sensitivity and scalable performance.

Quantum sensor arrays are engineered networks of spatially distributed, individually addressable quantum systems whose measurement correlations, sensitivity, and spatial or temporal multiplexing capabilities enable detection and imaging tasks beyond single-sensor architectures. By leveraging quantum coherence, entanglement, or collective operation, such arrays transcend classical sensor paradigms in fields ranging from quantum imaging and biosensing to condensed-matter metrology and fundamental physics searches.

1. Architectures and Physical Platforms

Quantum sensor arrays are realized across diverse physical systems, each optimizing for spatial resolution, operational bandwidth, quantum state controllability, and application-driven constraints.

Solid-State Defect Arrays: Nitrogen-vacancy (NV) centers in diamond and boron-vacancy (V $_\mathrm{B}^-$ ) centers in hBN are patterned into two-dimensional arrays with spacings from sub-micron (hBN: 100 nm pitch) to hundreds of microns (diamond: 225 μm), targeting biosensing, magnetic imaging, and scalable multiplexing (Chi-Durán et al., 15 Aug 2025, Sasaki et al., 2023, Weng et al., 14 Nov 2025). Arrays are formed via lithographic stamping, helium-ion microscope writing, or integration onto photonic platforms, with antifouling surface chemistry and patterning dictating specificity and density.

Quantum-Dot and Semiconductor Architectures: Silicon quantum-dot arrays implemented in fully-depleted silicon-on-insulator (FD-SOI) nanowire FETs leverage split-gate schemes to define 2 $\times$ n or bilinear layouts (Duan et al., 2020). Capacitive coupling, floating gates, and on-chip electronics enable charge or spin-state readout as well as remote sensing and gate operation.

Atom-Based Tweezer and Lattice Arrays: Neutral atoms are trapped in two- or three-dimensional site-ordered arrays via microlens-generated optical tweezers, achieving 7 μm site separation, high parallelism, and single-particle addressability for vector field mapping and quantum-enhanced metrology (Schäffner et al., 2023).

Photonic and Integrated Sensor Arrays: Arrays of single-photon avalanche diodes (SPADs) enable massively parallel quantum-correlation detection, photon number resolution, and wide-field quantum imaging with sub-nanosecond time resolution (Lubin et al., 2019, Wang et al., 2023). Foundry-fabricated silicon nitride photonic-integrated circuits (PICs) with deterministic NV-nanodiamond deposition establish scalable, low-loss, and low-crosstalk sensor grids suitable for simultaneous multi-point quantum magnetometry (Weng et al., 14 Nov 2025, Gurses et al., 13 Jun 2024).

Platform	Max. Density / Pitch	Example Application
NV-in-diamond microarray	49 sites / 225 μm	Multiplexed biosensing
hBN V $_\mathrm{B}^-$	100 nm pitch	Nanoscale magnetometry
Optical tweezers	7 μm pitch, 270 sites	Field mapping, atom arrays
SPAD camera + NV	64×32 pixels, 150 μm	Wide-field quantum imaging

2. Measurement Principles and Quantum Enhancement

Quantum sensor arrays detect relevant physical observables by monitoring transitions, coherence, or correlations among array elements subject to external fields or particle interactions.

Spin-Based Quantum Sensing: Defects such as NV centers or V $_\mathrm{B}^-$ are addressed via optically detected magnetic resonance (ODMR) to extract local magnetic fields, temperature, or strain. The contrast between eigenstate populations in response to external perturbations provides the measurement signal, while $T_1$ or $T_2^*$ relaxation or dephasing times set sensitivity and integration timescales (Chi-Durán et al., 15 Aug 2025, Sasaki et al., 2023, Weng et al., 14 Nov 2025).

Photon Counting and Correlation: Arrays of SPADs or other single-photon detectors perform time-correlated single-photon counting (TCSPC), enabling measurement of photon statistics (e.g., $g^{(2)}$ , $g^{(3)}$ ) and super-resolution imaging beyond diffraction limits (quantum image scanning microscopy, Q-ISM) (Lubin et al., 2019).

Continuous-Variable and Entangled Probes: Distributed quantum sensing leverages multipartite entanglement or squeezed states distributed across an array to surpass the standard quantum limit (SQL) for parameter estimation, achieving Heisenberg-limited scaling of sensitivity ( $\propto 1/M$ for $M$ sensors) in lossless scenarios (Zhang et al., 2020). Photonic sensor arrays utilize joint homodyne detection to extract quadrature-level signals with shot-noise clearance and programmable spatial filtering (Gurses et al., 13 Jun 2024).

3. Readout, Multiplexing, and Machine-Learning Integration

Readout schemes in quantum sensor arrays are engineered for high throughput, minimal crosstalk, and robust error mitigation.

Parallelized Readout: Optical, electrical, or photonic circuits route signals from each sensor site to dedicated detectors with timing resolution down to 10 ns (SPAD arrays) or spatial mapping across 2048 or more channels (wide-field NV + SPAD implementation) (Lubin et al., 2019, Wang et al., 2023).

Machine-Learning-Assisted Reconstruction: For spatially distributed sensing tasks (e.g., magnetic localization of microscale objects), feed-forward convolutional neural networks trained on multichannel sensor data achieve localization errors subwavelength relative to the object size, exploiting channel redundancy and optimizing for robustness under measurement noise (Weng et al., 14 Nov 2025).

Shot-Noise and Crosstalk Suppression: Integration of high-efficiency optical or electronic coupling (e.g., inverse-tapered Si $_3$ N $_4$ waveguides, balanced photodiode detection) delivers channel isolation $<0.1\%$ and shot-noise clearance $>30$ dB for quantum-enhanced imaging and communications (Gurses et al., 13 Jun 2024).

4. Performance Metrics and Scaling Laws

Key performance metrics in quantum sensor arrays include detection efficiency, spatial resolution, signal-to-noise ratio (SNR), multiplexing capability, and limits set by quantum or thermal noise.

Photon Detection Efficiency: CMOS SPADs reach $55\%$ at $520$ nm, with timing jitter $\sim 120$ ps and crosstalk probability of $0.14\%$ per nearest-neighbor pixel (Lubin et al., 2019).
Sensitivity: NV-based ensemble sensors demonstrate field sensitivities down to $\sim 25\,\mu\mathrm{T}/\sqrt{\mathrm{Hz}}$ per pixel, with demonstrated spatial resolution set by array pitch (e.g., $150\,\mu$ m, single-cell footprint) (Chi-Durán et al., 15 Aug 2025) or $100$ nm for hBN arrays (Sasaki et al., 2023).
Frame Rate: Wide-field quantum sensor arrays leveraging SPAD cameras attain $\sim100$ kHz frame rates (10 $\mu$ s readout) over $64\times32$ pixels (Wang et al., 2023).
Scaling with Array Size: For distributed quantum sensing, deploying entangled resources transitions SQL scaling $(1/\sqrt{M})$ to the Heisenberg limit $(1/M)$ , contingent on loss and decoherence (Zhang et al., 2020). In mechanical sensor arrays aimed at dark-matter searches, SNR and event rates scale as $\sqrt{N}$ with the number of sensors $N$ (Amaral et al., 10 Dec 2025).

5. Applications across Domains

Quantum Imaging and Super-Resolution: SPAD arrays combined with quantum emitters enable Q-ISM super-resolution, photon number-resolving (PNR) detection, and $g^{(n)}$ statistics for single-photon sources (Lubin et al., 2019).

Multiplexed Biosensing: Diamond NV arrays functionalized with antifouling subnanometer PEG layers and DNA microarrays provide parallel detection of up to 49 unique biomolecular targets with binary $T_1$ -based quantum readout, and can be extended to $>10^3$ spots/cm $^2$ for high-throughput diagnostics (Chi-Durán et al., 15 Aug 2025).

Nanoscale Magnetometry: hBN V $_\mathrm{B}^-$ arrays with 100 nm pitch perform vector-resolved, sub-diffraction magnetic field imaging of current-carrying nanowires, with practical sensitivities of $73.6\,\mu\mathrm{T}/\sqrt{\mathrm{Hz}}$ (Sasaki et al., 2023).

Quantum Field Mapping and Distributed Sensing: Tweezer-based individual-atom arrays map DC and gradient magnetic fields with $7\,\mu$ m grid resolution and sub- $0.1\,\mu$ m addressability, and are extensible to 3D lattices for volumetric metrology (Schäffner et al., 2023). Photonic-integrated NV arrays achieve real-time tracking of magnetic micro-objects and demonstrate low-latency, multi-point localization (Weng et al., 14 Nov 2025).

Fundamental Physics Searches: Large-scale quantum accelerometer arrays are proposed for direct detection of composite, ultraheavy dark matter, exploiting SNR scaling laws and array geometry to probe gravitational and Yukawa-coupled signals with sensitivity determined by the interplay of DM size, sensor pitch, and noise suppression techniques (Amaral et al., 10 Dec 2025).

6. Future Directions and Open Challenges

Critical future directions include achieving higher spatial densities, real-time fluidic and optical integration, multi-modal field imaging, and error-corrected distributed sensing networks.

Scaling and Integration: On-chip photonics, multilayer routing, and V-groove fiber arrays pave the way for 2D and 3D sensor matrix architectures with $>100$ operational channels (Weng et al., 14 Nov 2025, Gurses et al., 13 Jun 2024).
Quantum Enhancement under Decoherence: Maximizing the advantage from entanglement and squeezing in large arrays remains limited by photon loss, local dephasing, and crosstalk. Engineering high-Q guided modes, programmable beamforming, and optimal resource allocation are active research frontiers (Zhang et al., 2020, Gurses et al., 13 Jun 2024).
Machine-learning-optimized Sensing: Hybrid quantum-classical networks may tune entanglement structure to task-specific loss landscapes or classification boundaries (Zhang et al., 2020).
Application-Driven Advancements: Targeted fields include scalable molecular diagnostics, navigation and control for magnetic microrobots, quantum-enhanced imaging for semiconductor diagnostics, and fundamental studies in gravitational and dark-matter physics (Weng et al., 14 Nov 2025, Amaral et al., 10 Dec 2025).

Quantum sensor arrays synthesize precision quantum control, scalable classical engineering, and algorithmic advances, providing a framework for the next generation of multiplexed, high-sensitivity detection and measurement platforms in science and technology.