Optical neural networks (ONNs), enabling low latency and high parallel data processing without electromagnetic interference, have become a viable player for fast and energy-efficient processing and calculation to meet the increasing demand for hash rate. Photonic memories employing nonvolatile phase-change materials could achieve zero static power consumption, low thermal cross talk, large-scale, and high-energy-efficient photonic neural networks. Nevertheless, the switching speed and dynamic energy consumption of phase-change material-based photonic memories make them inapplicable for in situ training. Here, by integrating a patch of phase change thin film with a PIN-diode-embedded microring resonator, a bifunctional photonic memory enabling both 5-bit storage and nanoseconds volatile modulation was demonstrated. For the first time, a concept is presented for electrically programmable phase-change material-driven photonic memory integrated with nanosecond modulation to allow fast in situ training and zero static power consumption data processing in ONNs. ONNs with an optical convolution kernel constructed by our photonic memory theoretically achieved an accuracy of predictions higher than 95% when tested by the MNIST handwritten digit database. This provides a feasible solution to constructing large-scale nonvolatile ONNs with high-speed in situ training capability.

信息技术的发展使得通讯波段集成光子技术在过去的几十年中被广泛重视并取得了突出进展，目前已走向商业化，这一技术的发展也激发了人们对于中红外波段（2~20 μm）片上光子集成的兴趣。中红外波段在空间光通信、热成像、物质探测分析等关乎国家发展、**安全、民生改善等技术领域具有重要的应用前景。利用半导体工艺实现中红外光电子系统芯片小型化在尺寸、功耗以及大规模量产部署具有重大优势。因此，发展中红外片上集成光电子技术具有重大意义。文中主要针对于中红外波段片上集成的一些关键基础器件（如：调制器、探测器）的突破性进展及代表性工作进行了回顾；对各器件的种类、性能、参数及加工手段分别进行了较为全面的调研与比较；同时，也对器件的发展进程、亟待解决的问题以及对未来的展望进行了总结。

The rapid development of information technology has fueled an ever-increasing demand for ultrafast and ultralow-energy-consumption computing. Existing computing instruments are pre-dominantly electronic processors, which use electrons as information carriers and possess von Neumann architecture featured by physical separation of storage and processing. The scaling of computing speed is limited not only by data transfer between memory and processing units, but also by RC delay associated with integrated circuits. Moreover, excessive heating due to Ohmic losses is becoming a severe bottleneck for both speed and power consumption scaling. Using photons as information carriers is a promising alternative. Owing to the weak third-order optical nonlinearity of conventional materials, building integrated photonic computing chips under traditional von Neumann architecture has been a challenge. Here, we report a new all-optical computing framework to realize ultrafast and ultralow-energy-consumption all-optical computing based on convolutional neural networks. The device is constructed from cascaded silicon Y-shaped waveguides with side-coupled silicon waveguide segments which we termed “weight modulators” to enable complete phase and amplitude control in each waveguide branch. The generic device concept can be used for equation solving, multifunctional logic operations as well as many other mathematical operations. Multiple computing functions including transcendental equation solvers, multifarious logic gate operators, and half-adders were experimentally demonstrated to validate the all-optical computing performances. The time-of-flight of light through the network structure corresponds to an ultrafast computing time of the order of several picoseconds with an ultralow energy consumption of dozens of femtojoules per bit. Our approach can be further expanded to fulfill other complex computing tasks based on non-von Neumann architectures and thus paves a new way for on-chip all-optical computing.

As a promising spectral window for optical communication and sensing, it is of great significance to realize on-chip devices at the 2 µm waveband. The development of the 2 µm silicon photonic platform mainly depends on the performance of passive devices. In this work, the passive devices were fabricated in the silicon photonic multi-project wafer process. The designed micro-ring resonator with a 0.6 µm wide silicon ridge waveguide based on a 220 nm silicon-on-insulator platform achieves a high intrinsic quality factor of 3.0×105. The propagation loss is calculated as 1.62 dB/cm. In addition, the waveguide crossing, multimode interferometer, and Mach–Zehnder interferometer were demonstrated at 2 µm with good performances.

Optical filters are essential parts of advanced optical communication and sensing systems. Among them, the ones with an ultrawide free spectral range (FSR) are especially critical. They are promising to provide access to numerous wavelength channels highly desired for large-capacity optical transmission and multipoint multiparameter sensing. Present schemes for wide-FSR filters either suffer from limited cavity length or poor fabrication tolerance or impose an additional active-tuning control requirement. We theoretically and experimentally demonstrate a filter that features FSR-free operation capability, subnanometer optical bandwidth, and acceptable fabrication tolerance. Only one single deep dip within a record-large waveband (S+C+L band) is observed by appropriately designing a side-coupled Bragg-grating-assisted Fabry–Perot filter, which has been applied as the basic sensing unit for both the refractive index and temperature measurement. Five such basic units are also cascaded in series to demonstrate a multichannel filter. This work provides a new insight to design FSR-free filters and opens up a possibility of flexible large-capacity integration using more wavelength channels, which will greatly advance integrated photonics in optical communication and sensing.

A new optical microscopy technique, termed high spatial and temporal resolution synthetic aperture phase microscopy (HISTR-SAPM), is proposed to improve the lateral resolution of wide-field coherent imaging. Under plane wave illumination, the resolution is increased by twofold to around 260 nm, while achieving millisecond-level temporal resolution. In HISTR-SAPM, digital micromirror devices are used to actively change the sample illumination beam angle at high speed with high stability. An off-axis interferometer is used to measure the sample scattered complex fields, which are then processed to reconstruct high-resolution phase images. Using HISTR-SAPM, we are able to map the height profiles of subwavelength photonic structures and resolve the period structures that have 198 nm linewidth and 132 nm gap (i.e., a full pitch of 330 nm). As the reconstruction averages out laser speckle noise while maintaining high temporal resolution, HISTR-SAPM further enables imaging and quantification of nanoscale dynamics of live cells, such as red blood cell membrane fluctuations and subcellular structure dynamics within nucleated cells. We envision that HISTR-SAPM will broadly benefit research in material science and biology.