Traditional image sensors primarily capture intensity and color, but neglect the polarization aspect, which can provide richer contextual data about surfaces, materials, and textures. The ability to integrate polarization sensitivity directly into neuromorphic vision systems opens vast new frontiers, from enhanced robotic perception and autonomous navigation to medical diagnostics and advanced environmental monitoring. Neuromorphic systems, inspired by the human brain’s processing architecture, mimic biological visual processing, offering low power consumption and real-time responsiveness. The challenge lies in discovering materials and device architectures capable of seamlessly converting subtle polarization cues into meaningful electrical signals with high fidelity.

At the forefront of this innovation are Zhang, Zhu, Tian, and their collaborators, who have successfully harnessed the intrinsic anisotropic electronic and optical properties of pristine black arsenic-phosphorus to construct a polarization-sensitive neuromorphic vision sensor. Black arsenic-phosphorus, a layered two-dimensional material, exhibits remarkable in-plane anisotropy, making its electrical conductivity and photoresponse strongly dependent on the polarization direction of incident light. This material’s unique crystalline structure enables an intrinsic response to polarized photons without requiring complex external optical elements or filters.

The researchers designed their sensor device to exploit the natural anisotropy of b-AsP by fabricating an array of phototransistors sensitive to differing polarization orientations. This design translates the polarization state of incident light directly into variations in electrical signals, elegantly encoding polarization information at the sensor level. This capability dramatically enhances data richness and processing efficiency compared to traditional setups that capture and decode polarization externally. Such integration reduces hardware complexity, cost, and energy consumption, positioning the technology for widespread adoption in practical systems.

A crucial aspect of the device’s remarkable performance lies in the purity and crystalline quality of the black arsenic-phosphorus material employed. The team developed advanced synthesis and fabrication protocols to obtain pristine b-AsP flakes with minimal defects and superior layer uniformity. These attributes ensure consistent anisotropic behavior and stable long-term operation, overcoming typical challenges faced by two-dimensional materials such as environmental degradation or performance variability. Their meticulous material engineering efforts underscore the importance of controlled production techniques in realizing neuromorphic devices with practical viability.

In testing, the polarization-sensitive neuromorphic sensor demonstrated highly distinguishable photoresponses under linearly polarized light at various angles, with a clear modulation of photocurrent corresponding to polarization direction. The anisotropic phototransistor arrays effectively mimicked neuro-inspired recognition patterns, enabling the extraction of both visual intensity and polarization features from complex scenes. This dual-information acquisition enriches visual data sets for downstream machine learning algorithms, facilitating enhanced object detection, edge recognition, and texture discrimination — capabilities critical for autonomous systems operating in dynamic and visually cluttered environments.

The practical implications extend beyond robotics and computer vision into biomedical fields, where polarization imaging can reveal subtle changes in tissue properties associated with diseases or structural abnormalities. The integration of polarization-sensitive phototransistors into flexible, wearable devices could empower new diagnostic tools providing real-time, non-invasive monitoring with improved contrast and specificity. Furthermore, environmental sensing applications could benefit from enhanced polarization contrast to detect pollutants or assess water quality, enabling smarter ecological management strategies.

Neuromorphic computing architectures capitalize on reduced power consumption by mimicking human neural networks’ event-driven processing paradigm. By embedding polarization sensitivity at the sensor level, this technology takes a significant leap towards developing compact, efficient visual systems that capture richer input modalities akin to biological vision. This advancement paves the way for next-generation artificial intelligence systems that interpret the visual world with greater nuance and energy efficiency, overcoming bottlenecks imposed by conventional sensors and bulky optical components.

The study also delves into the device physics underpinning the polarization-sensitive behavior, revealing that the anisotropic response arises from directional-dependent carrier mobility and photogenerated charge separation within the b-AsP layers. The careful alignment of crystal axes with electrode configurations optimizes photodetection performance, highlighting the interplay between material properties and device architecture. These insights provide a valuable foundation for engineering bespoke two-dimensional materials tailored to specific neuromorphic sensing tasks.

Moreover, the research identifies avenues for scaling up the sensor arrays while maintaining uniformity in polarization response across larger areas. Such scalability is essential for practical deployment in complex imaging systems requiring high spatial resolution and consistent performance. The integration of these polarization-sensitive units with complementary metal-oxide-semiconductor (CMOS) technology also represents a promising direction for developing compact, commercially viable devices compatible with existing electronics manufacturing processes.

Beyond the demonstrated phototransistor arrays, the principles established by this work lay the groundwork for exploring other anisotropic layered materials and heterostructures to further customize spectral range, sensitivity, and polarization selectivity. By expanding the material palette and combining different two-dimensional crystals, researchers could build multifunctional neuromorphic sensors capable of simultaneously detecting polarization, intensity, wavelength, and even phase, thereby offering holistic visual perception akin to natural biological systems.

This pioneering research not only advances the scientific understanding of two-dimensional material optoelectronics but also concretely pushes forward the technological frontier of neuromorphic vision sensing. Bridging the gap between material innovation and practical device design, it provides a tangible pathway to embedding sophisticated sensory functions into compact, low-power systems. As autonomous devices and artificial intelligence increasingly permeate everyday life, such enhancements in visual perception will be crucial to unlocking their full potential safely and effectively.

In conclusion, the introduction of polarization-sensitive neuromorphic vision sensing based on pristine black arsenic-phosphorus marks a seminal achievement in the quest for advanced, biologically inspired artificial vision systems. The compelling combination of material anisotropy, device ingenuity, and neuromorphic design principles culminates in a sensor capable of capturing richer visual cues while operating under practical constraints. This breakthrough sets the stage for a host of transformative applications across robotics, healthcare, environmental monitoring, and beyond, heralding a future where machines see the world through eyes as refined and sensitive as those of living beings.

As research progresses, continued refinement of material quality, integration techniques, and system architectures will further enhance performance and durability. Interdisciplinary collaborations spanning physics, engineering, computer science, and materials chemistry will be critical to translating these advances into real-world products. The exciting developments reported underscore the vibrant potential of two-dimensional materials to reshape not only fundamental science but also the practical capabilities of next-generation technologies that emulate and extend natural sensory processes.

Subject of Research: Polarization-sensitive neuromorphic vision sensing enabled by pristine black arsenic-phosphorus

Article Title: Polarization-sensitive neuromorphic vision sensing enabled by pristine black arsenic-phosphorus

Article References:
Zhang, S., Zhu, S., Tian, S. et al. Polarization-sensitive neuromorphic vision sensing enabled by pristine black arsenic-phosphorus. Light Sci Appl 15, 100 (2026). https://doi.org/10.1038/s41377-025-02125-0

Image Credits: AI Generated

DOI: 02 February 2026

Single-photon LiDAR systems operate by emitting low-intensity laser pulses and detecting the faintest reflections from distant objects. The technique hinges on counting individual photons to construct detailed three-dimensional maps of environments. However, conventional detectors often struggle with noise, inefficiencies, and limited resolution, hindering their practical usage in scenarios demanding ultra-high sensitivity and accuracy. The recent advance by Xu et al. introduces a quantum-enhanced detection scheme that curtails these limitations substantially.

At the heart of this innovation lies photon-number-resolving detection — a technology capable of discerning not only whether photons have arrived but precisely how many are received within a given temporal window. Unlike traditional single-photon avalanche diodes (SPADs), which register only the presence or absence of photons, this new detection approach leverages superconducting nanowire detectors and advanced signal processing to quantify photon numbers with ultra-high fidelity. The result is a LiDAR system that more closely adheres to the fundamental quantum measurement limits prescribed by the standard quantum limit, thus maximizing information extraction from minimal light levels.

Approaching the standard quantum limit is no trivial feat. It demands exquisite control over noise minimization, timing resolution, and system efficiency. The research team meticulously engineered the detection apparatus to suppress dark counts and mitigate afterpulsing effects, factors traditionally degrading performance. Their integration of cryogenically cooled detectors with optimized readout electronics allowed them to achieve a negligible noise floor, enabling the identification of single photons amidst ambient background noise — a challenge intensified in real-world operating conditions.

Furthermore, the research addresses the temporal resolution necessary for precise ranging. Typical timing uncertainties in photon counting blur the accuracy of depth measurements inherent to LiDAR technology. By enhancing timing jitter to the sub-nanosecond regime, the system could differentiate photons arriving from closely separated distances, thus refining the three-dimensional imaging capacity to an unprecedented level of detail. This temporal precision is crucial for applications like autonomous vehicles, where split-second decisions based on accurate spatial data are essential for safety.

Beyond the mere improvement of hardware components, the study innovates in data interpretation and signal reconstruction methods. The team implemented sophisticated statistical inference algorithms grounded in quantum estimation theory to maximize information retrieval from the detected photon counts. By modeling photon arrival distributions and leveraging Bayesian frameworks, the researchers could reconstruct depth profiles with minimal uncertainty, surpassing classical LiDAR limits.

One exciting implication of this advancement is its potential to reduce laser power requirements drastically. Operating closer to the fundamental quantum limit means that illuminating scenes with fewer photons still yields highly accurate distance measurements. This reduction mitigates eye safety concerns and lowers the system’s optical footprint, a vital consideration for consumer electronics and mobile platforms. It also opens the door for stealth applications requiring minimal detectable emissions.

Moreover, the photon-number-resolving detection approach enhances discrimination capabilities in challenging environments. In scenarios with dense fog, rain, or urban clutter where photon returns are weak and scattered, the system’s improved sensitivity and resolution maintain reliable operation. This robustness positions it uniquely for deployment in autonomous navigation systems operating in adverse weather — a significant step forward from current LiDAR technologies that often degrade under such conditions.

The technique’s compatibility with integrated photonics further strengthens its prospects for widespread adoption. Advances in nanofabrication enable the construction of compact superconducting nanowire arrays directly on chip-scale platforms. This integration promises to scale the technology with cost-effectiveness and reduced power consumption, facilitating its incorporation into handheld devices, drones, and even satellite payloads where weight and size constraints are paramount.

Xu and colleagues also explored noise reduction strategies leveraging quantum correlations and squeezing states, aiming to push beyond the standard quantum limit toward the Heisenberg limit. Although their current setup achieves proximity to the standard quantum limit, ongoing research in quantum light sources and entangled photon pairs suggests the possibility of even greater sensitivity enhancements in the near future.

Parallel improvements in the LiDAR transmitter’s coherence and pulse shaping ensure optimal coupling with the photon-number-resolving detector architecture. By matching pulse properties to the temporal response of detectors, the system minimizes losses and detector saturation effects, cementing its efficiency gains. These holistic system-level optimizations underscore the importance of concerted advances across both emitter and receiver components.

Beyond the realm of terrestrial applications, this pioneering work portends benefits for spaceborne LiDAR missions tasked with planetary mapping and orbital debris tracking. The ability to operate with low photon flux and superb ranging accuracy could yield richer datasets with smaller instruments, revolutionizing remote sensing and astronomical observation capabilities.

Importantly, by pushing single-photon LiDAR systems toward fundamental quantum limits, this technology aligns with broader quantum metrology initiatives aiming to redefine measurement science. The methods and insights derived here may extend into other photonic sensing regimes, such as quantum communication channels and biological imaging, where photon budget constraints and noise are critical.

As quantum technologies mature, this breakthrough exemplifies how marrying quantum detection schemes with classical sensing modalities can unlock performance regimes previously considered unattainable. The path forward includes refining detector arrays for larger-scale imaging, integrating machine learning for real-time data interpretation, and further minimizing environmental noise impacts.

In summary, the demonstration of photon-number-resolving detection in single-photon LiDAR systems marks a seminal advance in the quest for ultra-precise, quantum-limited remote sensing. By conquering key noise and resolution challenges, this approach promises to accelerate the deployment of next-generation LiDAR platforms with transformative implications spanning autonomy, environmental monitoring, and fundamental physics research. The vision of LiDAR systems operating at the edge of quantum measurement limits is no longer a distant dream but an emerging reality.

—

Subject of Research: Photon-number-resolving detection in single-photon LiDAR systems achieving performance near the standard quantum limit.

Article Title: Photon-number-resolving detection enables single-photon LiDAR approaching the standard quantum limit

Article References:

Xu, F. Photon-number-resolving detection enables single-photon LiDAR approaching the standard quantum limit.
Light Sci Appl 14, 206 (2025). https://doi.org/10.1038/s41377-025-01880-4

Image Credits: AI Generated