《Chip》发表云南大学史衍丽团队综述论文：近红外单光子探测器发展
FUTURE | 远见 闵青云 文
近日，云南大学史衍丽团队的综述论文「Advances in near-infrared avalanche diode single-photon detectors」发表于全球新锐综合性研究期刊Chip，第一作者为博士生刘辰。该综述论文介绍了近红外单光子探测器的发展。
近年来，单光子探测技术在量子保密通信、量子计算、人工智能、军事探测等多个领域展现了广阔的应用前景，使得单光子探测器成为当下的一个研究热点。
Photons are absorbed to produce hole-electron pairs, and holes enter the multiplication region unde the electric field to trigger avalanche multiplication.
半导体雪崩二极管探测器由于内部具备雪崩增益，响应快、体积小、成本低、易于集成，通过半导体制冷即可实现约零下40℃的工作温度，成为单光子探测器的一个比较好的选择。其中InP/InGaAs近红外单光子探测器是目前发展最为成熟的近红外雪崩二极管单光子探测器，其单管和阵列都实现了商业化产品，主要的性能指标包括：在零下40℃单光子探测效率超过30%，暗计数率低于10kHz，后脉冲概率低于5%，时间抖动低于100ps，最大计数率超过100MHz。焦平面阵列中心距50um、阵列规格256×128，可采用光子计数和光子计时两种成像方式进行三维成像。目前国内外正在研发更小中心距（25微米）、更大规格的单光子焦平面阵列（320×256规格或以上）。InP/InGaAs近红外单光子探测器产品的出现和新进展为这类探测器的规模应用提供了可能。
单光子探测器的发展方向
目前，单光子探测器研发主要沿着两条路径快速推进：一条路径是不断优化现有InP/InGaAs SPAD的性能，向着更小的暗计数、更低的后脉冲、更高的计数率和更高的工作温度等方向发展。存在的挑战是倍增层里的材料缺陷导致后脉冲效应大、死时间长（不发生雪崩的时间）、计数率降低。目前的技术手段存在一定的相互制约困境：为了降低后脉冲，通过降低雪崩电荷量，导致了探测效率低的问题；通过延长死时间，导致了计数率低的问题；通过提高温度，减小俘获载流子的寿命，导致了暗计数率高的问题。现有的解决方法是借助集成淬灭电路的发展，最大限度降低寄生电容和后脉冲, 与此同时, 进一步提高倍增层材料的生长质量。
单光子探测器的性能优化和应用离不开淬灭电路的支撑。淬灭电路一方面需要及时停止雪崩，另外一方面还需要抑制来自器件和电路等的噪声，把雪崩信号提前出来。目前通过器件自身集成淬灭电阻或势垒，通过外部淬灭电路的单片集成化，通过正弦门控和不同淬灭电路的组合，显著地降低了器件的暗计数、后脉冲，提高了器件的计数率。
Working principle of SPAD-based single-photon detectors.
另一条路径是寻找更有发展前景的新材料、新机理器件。在巨量应用需求的刺激下，基于低k因子的低噪声材料开发、基于低维材料的扩展波长、基于离化倍增工程降低噪声、基于低维材料弹道输运实现高增益倍增等新材料、新机理器件应运而生，这些新技术的发展为进一步降低单光子探测器的噪声、提高信噪比、拓展波长和改善器件工作条件开辟了新的方向。
近红外单光子探测器的重要应用领域
近红外单光子探测器具有光子探测的极限灵敏度，其工作波段是传统光纤使用的主要波段，同时又是人眼安全波段，因此，近红外单光子探测器在量子保密通信、激光雷达、无人驾驶、三维成像、微弱信号检测等领域具有重要的应用。
激光雷达采用1550nm的激光作为发射光源时，由于对人眼的安全性，可以使用更大功率的激光，从而能探测更远距离，使目前的探测距离从100m到200m，另外波长1550nm的光在自然光环境下具有更干净的背景噪声，这对于无人驾驶或车载激光雷达的应用都具有重要的意义，可以预期1550nm单光子探测器的发展将显著推动无人驾驶以及激光雷达技术的快速发展。
Compared to Si-LiDAR, LiDAR at 1550 nm can see much further.
(a) L3Harris' Geiger-mode LiDAR data; (b) Aurora’s Frequency Modulated Continuous Wave lidar data; (c) BB84 implementation using a single SPAD.
From https://www.l3harris.com/all-capabilities/geiger-mode-lidar;
https://aurora.tech/blog/fmcw-lidar-the-self-driving-game-changer;
Martelli, P., Brunero, M., Fasiello, A., Rossi, F., Tosi, A., & Martinelli, M. (2019). Single-SPAD implementation of quantum key distribution. In 21st International Conference on Transparent Optical Networks, ICTON 2019 (Vol. 2019, pp. 1-3). IEEE Computer Society.
该文对基于InP/InGaAs的单光子探测器技术和相关的新技术发展进行介绍，旨在帮助增加对近红外单光子探测器的了解和认识，扩展其应用，为进一步的性能提升和发展提供参考。
Advances in Near-Infrared Avalanche Diode Single-Photon Detectors
In recent years, single-photon detection technology has shown broad application prospects in communication, military, computing, artificial intelligence, detection and other fields, making single-photon detectors (SPDs) a current research hotspot.
Semiconductor avalanche diode detectors are a good choice for single-photon detectors due to their internal avalanche gain, fast response, small size, low cost, ease of integration, and the near room-temperature operating temperature that can be achieved by semiconductor thermo-electrical coolers. InP-based single-photon avalanche diodes (SPADs) including focal plane arrays are one of the most mature near-infrared SPADs, both the SPAD and arrays have commercialized products. The main performance parameters include: at a temperature of about 233K the single-photon detection efficiency ≥30%, the dark count rate is less than 10kHz, the after-pulsing probability is less than 5%, the time jitter is less than 100ps, and the maximum count rate exceeds 100MHz. The focal plane array achieves a pitch of 50um and an array size of 256×128. Two imaging methods, photon counting and photon timing, can be used for three-dimensional imaging. At present, single-photon focal plane arrays with smaller pitch (25 microns) and larger formats (320×256 or above) are being developed at home and abroad. The emergence of these products provides the possibility for the large-scale application of near-infrared SPADs.
Development directions of SPDs
At present, the research and development of single-photon detectors are advancing rapidly, mainly along two paths.
One research direction is to continuously optimize the performance of existing InP/InGaAs SPADs toward smaller dark counts, lower afterpulsing, higher count rates, and higher temperatures. The challenge is that material defects in the multiplication layer lead to large afterpulsing effects, long dead times (time without avalanches), and reduced count rates. Furthermore, there are certain mutual-restriction dilemmas for the current technical solution. In order to reduce the afterpulsing, by reducing the avalanche charge, the detection efficiency is low; by prolonging the dead time, the count rate is low; by increasing the temperature, the lifetime of trapped carriers is reduced, resulting in a high dark count rate. The present effective solution is to minimize parasitic capacitance and afterpulsing with the development of integrated quenching circuits, while at the same time by further improving the growth quality of the multiplication layer material.
The driving circuit plays a key role in the performance of SPAD-based SPDs. On the one hand, the quenching circuit needs to stop the avalanche in time, and on the other hand, it also acts to suppress the noise resulting from the device and circuit, etc. Through the integration of the quenching resistance or potential barrier of the device itself, through the monolithic integration of the external quenching circuit, and through the combination of sinusoidal gating and different quenching circuits, the dark count and post-pulse of the device are significantly reduced. Improvements in the quenching circuits have greatly contributed to SPD development.
Another research direction is to explore new materials and new mechanisms. Up to now, low-dimensional materials, low k factor, and thus low noise material were intensively studied. Novel mechanism devices based on ionization multiplication engineering, ballistic transport multiplication, etc. have opened up new directions for further reducing the noise of single-photon detectors, improving the signal-to-noise ratio, extending wavelengths, and optimizing device operation temperature conditions.
Important applications of near-infrared SPDs
The near-infrared single-photon detectors has the ultimate sensitivity of photon detection, at the same time its working band is the main band used by traditional optical fibers, and it is also the eye-safe band. Therefore, near-infrared single-photon detectors have important applications in the fields of quantum secure communication, LiDAR, unmanned driving, 3D imaging, and weak signal detection.
The near-infrared wavelengths >1400 nm are known as an eye-safe band for humans, which means that 1550 nm LiDAR is allowed to use higher-power laser sources. This advantage directly increases the detection distance of LiDAR in self-driving vehicles from ~100 m to over 200 m. It is expected that the development of near-infrared SPADs will also play a pivotal role in the development of the above fields.
This work analyzes the latest development and application of InGaAs/InP photodiodes, then briefly reviews other near-infrared single-photon detection technology based on new materials and new mechanisms. It aims to promote the understanding of near-infrared SPADs, expand their applications, and provide a reference for the further development and performance improvement of such devices.
文章预印版：https://www.sciencedirect.com/science/article/pii/S270947232200003X?v=s5
关于 Chip
Chip是全球唯一聚焦芯片类研究的综合性国际期刊，已入选由中国科协、教育部、科技部、中科院等单位联合实施的「中国科技期刊卓越行动计划高起点新刊项目」，为科技部鼓励发表「三类高质量论文」期刊之一。
Chip期刊由上海交通大学与Elsevier集团合作出版，并与多家国内外知名学术组织展开合作，为学术会议提供高质量交流平台。
Chip秉承创刊理念: All About Chip，旨在发表与芯片相关的各科研领域尖端突破，助力未来芯片科技发展。迄今为止，Chip已在其编委会汇集了来自12个国家的57名世界知名专家学者，其中包括多名中外院士及IEEE、ACM等知名国际协会终身会士（Fellow）。
Chip首刊于2022年3月在爱思唯尔Chip官网以完全开放获取形式（Open Access）发布，欢迎访问阅读文章。
爱思唯尔Chip官网：
https://www.journals.elsevier.com/chip
点击「阅读原文」直达文章预印版。
End