Roadmap on UV-C photodetectors: materials, applications and industry perspectives

This paper is a comprehensive roadmap that provides an overview of the current landscape of UV-C photodetection, covering established and emerging material platforms and their applications.

The abstract states: "UV-C photodetectors are poised to play an increasingly important role in future photonic technologies, driven by the rapid emergence of UV-C light sources and new wide bandgap semiconductors. These advances are enabling new levels of spectral selectivity, radiation hardness, sensitivity, and device integration, while opening opportunities across a broad range of applications. This roadmap provides a comprehensive overview of the current landscape of UV-C photodetection, spanning established and emerging material platforms (Ga2O3, AlGaN, BN, diamond, MgZnO, 2-dimensional materials, metal halide perovskites, micro-electromechanical systems), and their applications in metrology, astronomy, communications, environmental monitoring, fire detection, missile warning, gas sensing, and medical diagnostics. By identifying opportunities, bottlenecks, and future directions, this roadmap aims to support both newcomers and established researchers, with the aim of accelerating the translation of UV-C photodetectors into impactful technologies."

The introduction explains the historical context and motivation: "Photodetectors have a long history, with early observations of light-to-electricity phenomena dating back to 1839... The UV region of the electromagnetic spectrum spans wavelengths from 400 to 100 nm and is commonly divided into UV-A (400–320 nm), UV-B (320–280 nm), and UV-C (280–100 nm). Of relevance to this roadmap is the UV-C range, which has attracted growing attention in recent years owing to its germicidal properties and the parallel development of UV-C light-emitting diodes (LEDs). It also notes the growing industrial demand: In 2020, the global UV sensor market (encompassing UV-A, UV-B and UV-C devices) was estimated at approximately USD 2.69 billion. By 2025, this figure had risen to around USD 8.54 billion, with continued growth projected into the 2030s at a compound annual growth rate of approximately 25%."

Section 2 on photodetector performance figures of merit states: The appropriateness and relevance of photodetector Performance Figures of Merit (FOM) will always depend on the prescribed application, be it imaging, sensing, communications, or otherwise. It details the essential FOMs: "To fully characterise a photodetector it is necessary to report the NSD, R(λ), D*(λ), a Bode Plot of R(f, λoptimum) from which f3dB and roll-off is evaluated, a photodetector response curve with a square wave modulated radiation field to evaluate τr, and D*(f, λoptimum). Importantly, all these must be measured under the conditions that create gain, if gain is an important feature of the device."

Section 3 on incumbent technology notes: "Current approaches to measure UV-C radiation employ photodiodes and vacuum discharge tubes. The photodiodes may consist of SiC, AlGaN, UV enhanced Si, or Si detector chips with conversion filters... Si and SiC-based photodetectors are the dominant technologies in terms of market size. It highlights a key challenge: A challenge is detecting a flame or an arc in the presence of sunlight... A suitable photodiode with a sufficient solar rejection rate (eight orders of magnitude) would enable new applications in the area of fire detection. Such a photodiode is currently not available."

Section 4 covers emerging materials. For Ga2O3: "Ga2O3 is inherently suited for UV-C detection due to its bandgap energy of approximately 4.3-5.3 eV, depending on the polymorph... This wide bandgap enables solar-blind spectral response with natural rejection of wavelengths above 280 nm, eliminating the need for additional filtering. Key challenges include Defects understanding," p-doping, Cost, Suitability for practical applications, and Low technology readiness level (TRL). The paper notes: At present, Ga2O3 UV-C photodetectors are best suited to applications where temporal response is not a primary requirement, for example in exoplanet spectroscopy, environmental monitoring, or gas sensing.

For AlGaN: "The development of AlGaN UV-C photodetectors began in the late 1990s, driven by the need for solar-blind UV detection in aerospace, environmental monitoring, and industrial applications. Interest in AlGaN for UV-C photodetection lies in its tunable direct bandgap ranging from 3.4 eV for GaN to over 6 eV for AlN, allowing excellent rejection of visible and near-UV light without the need for filters. Challenges include The high-AlN contents required for UV-C detection lead to significant lattice mismatch with commonly-used sapphire substrates, and The growth and processing of high-quality, high-AlN layers remain costly and technically demanding."

For BN: "Boron nitride (BN) possesses several allotropes... all of which exhibit ultra-wide bandgap (typically 6.0–6.4 eV) and other distinctive properties including high band-edge absorption coefficient... exceptional dielectric strength, chemical inertness, thermal stability and mechanical robustness. Challenges include the efficient and low-cost production of wafer-scale, uniform, low-defect, and high-crystallized h-BN films with appropriate thickness remains the primary bottleneck, and the performance of BN-based detectors is still lower than expected in terms of dark current, responsivity, and response speed."

For diamond: "Diamond's exceptional intrinsic properties, such as a wide direct bandgap energy (5.5 eV), outstanding radiation hardness, and remarkable chemical stability, render it a promising material platform for UV-C photodetection. Challenges include The growth of large SCD wafers with high crystal quality, low dislocation density, and controllable impurity doping remains a great challenge, and Realizing reliable n-type diamond is another major hurdle."

For MgZnO: The bandgap of MgZnO can be tuned from 3.37 to 7.8 eV, covering full UV spectral range. Challenges include "One of the challenges lies in monocrystal MgZnO synthesis. Due to the crystal lattice difference between ZnO (hexagonal) and MgO (cubic), there is a limited solid solubility of MgO in ZnO (4 mol%) in equilibrium conditions, and Native point defects ubiquitously exist in oxide semiconductors... One major challenge is that the slow photo-generated carrier recombination process through these defects will lead to persistent photocurrent."

For 2D materials: "2D ultrawide bandgap semiconductors have gained increasing interest in UV-C photodetectors due to their unique mechanical (flexibility) and physical properties (high optical transparency, tunable electrical conductivity, high carrier mobility, ultra-high gate dielectrics). Challenges include Low cost, large scale, and uniform synthesis," Defect control at surfaces and interfaces, and Integration and intelligent applications.

For metal halide perovskites: Among MHPs, chloride-based compositions (e.g., CsPbCl3) are particularly suited for UV-C photodetection because their wide bandgaps enable solar-blind operation and selective UV-C absorption. Challenges include UV-C-induced photochemical degradation, Ion migration, interfacial deterioration, and detection sensitivity limits, Environmental stress coupling and toxic material concerns, and Solubility limitations and challenges in large-area uniform deposition.

For MEMS: "Microelectromechanical systems (MEMS) technology has commonly been used in development of free space optical elements... Recently there has been increasing development aiming at using MEMS in photodetectors to improve device performances. The fundamental mechanism is to apply mechanical strain/stress to the photodetector devices to realise performance tuneability or enhancement. Three directions are identified: piezoelectric microstructures are used together with heterojunctions to create built-in potentials," MEMS devices fabricated with wide bandgap materials are constructed to achieve multi-functional or dual-modality photodetectors, and MEMS technology has been used to overcome the limitation imposed by this balance between photoresponsivity and photo-to-dark current ratio.

Section 5 covers applications. For metrology: "Within optical metrology, the UV-C spectral range is less developed than the visible and near IR spectral ranges... The limitations of available sensing devices in the UV-C range is a key challenge since standard Si-based detectors exhibit poor quantum efficiency for wavelengths below 240 nm."

For astronomy: "UV-C astronomy explores one of the most information-rich yet technically challenging regions of the electromagnetic spectrum. Because Earth's atmosphere is opaque to UV-C radiation, all observations must be conducted from space. The paper notes: The Hubble Space Telescope is now operating well beyond its original intended lifetime, and so, continuity of UV-C observations is increasingly at risk."

For communications: "The attractive characteristics of the UV-C region specifically are, firstly, the UV-C region is solar-blind with little to no natural or artificial background interference. Secondly, Rayleigh scattering of light is inversely proportional to the fourth power of wavelength, thus strongly scattered UV-C lends itself to non-line-of-sight (NLOS) communications. A key challenge is that the rise and fall times of solar-blind photoreceivers are at the moment typically in the range of milliseconds to seconds, in contrast with nanoseconds from Si-based visible photoreceivers."

For environmental monitoring: UV-C photodetectors, operating within the 200-280 nm wavelength range, have become increasingly relevant in environmental sensing, offering rapid, reagent-free analysis of water quality and organic matter dynamics. Challenges include spectral complexity, Matrix variability, Standardisation across platforms, and Scalability.

For fire detection: "UV-C signals can play a role in remote optical fire detection/localisation in forests, large-scale facilities, cultural assets or transport vehicles... state-of-the-art remote optical fire sensors already combine an IR sensor with a high sensitivity UV-C PMT. The latter eliminates false positives from the IR sensors but requires high-voltage driving, has a limited lifetime and is fragile. For these reasons, there is a drive to replace PMTs with solid-state UV-C photodetectors."

For missile warning: "The primary advantage of UV-C photodetection lies in its intrinsic insensitivity to solar background radiation at ground and near-ground levels, owing to strong atmospheric absorption below ~280 nm. This 'solar-blind' characteristic enables high signal-to-noise ratios during daylight operation. Challenges include Ultra-low false alarm rates are essential," rapid response times, often in the microsecond-to-millisecond range, and Long-term stability.

For gas detection: "Many atmospheric pollutant gases such as ozone (O3), nitrogen oxides (NOx), sulphur dioxide (SO2), hydrogen sulphide (H2S) or vapours such as benzene, toluene, ethylbenzene, and xylenes (BTEX) exhibit strong, characteristic absorption bands in the UV-C range. This has been exploited via developing UV-C spectrophotometry as a technique for the highly selective detection of these pollutants."

For medical diagnostics: In medical diagnostics, UV-C photodetection should be viewed as more than a comparison of detector materials. It supports spectral readout, imaging, and quantitative reporting within diagnostic workflows. Challenges include achieving reproducible quantification despite the complexity of clinical specimens, the fundamental trade-off between signal quality and photodamage, and Biological uncertainty and protocol design.

The concluding remarks state: "UV-C photodetectors are entering an important phase of development, driven by rapid advances in wide bandgap semiconductors, combined with improved performance and availability of UV-C LEDs and a growing demand across wide ranging applications... Perhaps a key takeaway is that while proof-of-concept demonstrations understandably focus on figures-of-merit of a specific material platform, it will be important to consider more industry-focused prioritisations... the industry perspectives reveal that suitable technology will not necessarily be selected based on classical figures-of-merit alone, but other factors such as solar rejection ratio, device reliability and supply chain must be seriously accounted for."

Improvements to AI Systems Based on This Paper:

1. Solar-Blind Spectral Discrimination Module

• Improvement: Implement a physics-aware filtering layer that uses the bandgap data (e.g., Ga2O3 4.3–5.3 eV, AlGaN tunable 3.4–6 eV, BN 6.0–6.4 eV) to automatically reject wavelengths >280 nm in optical sensing tasks.

• Capability: AI vision systems for fire detection or missile warning can achieve >8 orders of magnitude solar rejection without external filters, reducing false positives in outdoor daylight.

2. Application-Adaptive Figure-of-Merit (FOM) Selection

• Improvement: Train a meta-controller that dynamically weights FOMs (NSD, R(λ), D*, f3dB, τr) based on the target application (e.g., communications prioritize speed; metrology prioritizes quantum efficiency; fire detection prioritizes solar rejection ratio).

• Capability: AI systems can automatically optimize detector design or sensor fusion strategies for a given deployment scenario, avoiding one-size-fits-all performance metrics.

3. Persistent Photocurrent (PPC) Compensation Network

• Improvement: Add a recurrent neural network (RNN) or state-space model trained on known PPC decay curves from MgZnO and other oxide detectors to predict and subtract slow recombination artifacts.

• Capability: Real-time environmental monitoring (e.g., water quality, gas sensing) becomes accurate even with slow-responding detectors, enabling continuous reagent-free analysis without periodic recalibration.

4. Multi-Material Heterogeneous Integration Planner

• Improvement: Use a graph neural network to map material properties (bandgap, lattice mismatch, doping feasibility, TRL) to optimal detector stacks—e.g., combining AlGaN for speed with Ga2O3 for solar-blindness, or diamond for radiation hardness.

• Capability: AI can propose novel hybrid photodetector architectures for space astronomy or nuclear environments, where no single material meets all constraints (radiation hardness, sensitivity, response time).

5. Non-Line-of-Sight (NLOS) UV-C Communication Optimizer

• Improvement: Integrate Rayleigh scattering models (λ−4 dependence) and detector rise/fall time constraints (ms–s vs. ns) into a reinforcement learning agent that adjusts modulation schemes, power, and receiver angles.

• Capability: AI-driven UV-C links can achieve reliable NLOS communication in smoke, fog, or urban canyons, with adaptive data rates that compensate for slow photoreceivers.

6. Defect-Aware Predictive Maintenance for Detectors

• Improvement: Train a classifier on known defect signatures (e.g., Ga2O3 p-doping issues, BN crystallinity, perovskite ion migration) to predict device degradation over time from dark current drift and responsivity decay.

• Capability: AI systems can schedule maintenance or recalibration for missile warning or medical diagnostics, ensuring long-term stability and ultra-low false alarm rates.

7. Spectral Unmixing for Multi-Gas Sensing

• Improvement: Implement a blind source separation algorithm using UV-C absorption cross-sections of O3, NOx, SO2, H2S, and BTEX to decompose overlapping spectral signatures from a single detector readout.

• Capability: AI-based environmental monitors can simultaneously quantify multiple pollutants with high selectivity, even in complex mixtures, without needing separate spectrometers.

8. Photodamage-Aware Medical Diagnostic AI

• Improvement: Add a constraint layer that limits UV-C exposure based on tissue photodamage thresholds, using reinforcement learning to optimize signal-to-noise ratio while minimizing biological harm.

• Capability: AI-driven diagnostic tools (e.g., skin lesion analysis, pathogen detection) can operate safely on living tissue, balancing detection sensitivity with patient safety.

9. Supply-Chain-Aware Detector Selection Advisor

• Improvement: Incorporate industry data (market size, TRL, cost, material availability) into a recommendation engine that scores candidate detector technologies not just by FOMs but by manufacturability and supply chain robustness.

• Capability: AI can guide R&D investment or procurement decisions, prioritizing Ga2O3 or AlGaN over exotic 2D materials when time-to-market and cost are critical.

10. Space-Based UV-C Observation Continuity Planner

• Improvement: Use a transformer model trained on Hubble and upcoming mission data to predict observation gaps and suggest alternative detector configurations (e.g., diamond-based) that maintain spectral coverage despite aging infrastructure.

• Capability: AI can autonomously schedule space telescope observations and flag when detector degradation requires switching to backup material platforms, ensuring uninterrupted UV-C astronomy.

UV-C photodetectors are poised to play an increasingly important role in future photonic technologies, driven by the rapid emergence of UV-C light sources and new wide bandgap semiconductors. These advances are enabling new levels of spectral selectivity, radiation hardness, sensitivity, and device integration, while opening opportunities across a broad range of applications. This roadmap provides a comprehensive overview of the current landscape of UV-C photodetection, spanning established and emerging material platforms (Ga2O3, AlGaN, BN, diamond, MgZnO, 2-dimensional materials, metal halide perovskites, micro-electromechanical systems), and their applications in metrology, astronomy, communications, environmental monitoring, fire detection, missile warning, gas sensing, and medical diagnostics. By identifying opportunities, bottlenecks, and future directions, this roadmap aims to support both newcomers and established researchers, with the aim of accelerating the translation of UV-C photodetectors into impactful technologies.