Advanced Photonics Research最新文献

Type-I Heterojunction Photodetectors

In article number 2400190, Yingjian Wang, Liang Li, Dongfeng Shi, and co-workers present type I van der Waals heterojunction photodetectors that demonstrate exceptional optoelectronic performance, characterized by a broadband spectral response spanning from ultraviolet to near-infrared regions, ultrafast response kinetics, and superior photoresponsivity. These advanced photodetectors, when synergistically combined with state-of-the-art Hadamard single-pixel imaging technology, enable high-fidelity image reconstruction across extensive spectral ranges and under challenging environmental conditions.

The breakthrough in van der Waals heterojunction diodes composed of 2D and 3D materials for optoelectronic devices has paved the way for advancements in broadband optical imaging. However, fabricating traditional array-based imaging detectors with these materials remains challenging. Cadmium sulfide (CdS), a historically significant semiconductor material, has been extensively used in optoelectronic devices due to its remarkable photoelectric properties and chemical stability. Notably, a unique type-I heterojunction can be formed by combining 2D CdS, prepared through chemical vapor deposition, with the first-generation semiconductor germanium (Ge). His heterojunction photodetector exhibits outstanding photoelectric performance, achieving a responsivity of 54 mA W⁻¹ and a detectivity of 1.4 × 10⁹ Jones under zero bias, with a spectral response range spanning from 265 to 1550 nm. Herein, the CdS/Ge heterojunction photodetector with the emerging single-pixel Hadamard algorithm, addressing challenges in nonvisible imaging that conventional imaging systems traditionally encounter, is integrated. This approach facilitates low-sampling-rate image reconstruction across a broad spectral range and under scattering conditions. It is anticipated that this work will significantly contribute to future advancements in broadband imaging applications.

Computer-Generated Holography

In article number 2400164, Zilong Zhang, Hongzhi Yang, and co-workers propose a new optical method based on self-interference to completely decompose the eigenmode spectrum of complex structured light. A diffractive optical method is designed and validated to extract the complete information of complex structured light fields composed with eigenmode superposition states, including the order of eigenmode, amplitude weight coefficient, and relative phase delay.

Light-Stimulated Artificial Synapses

The cover image is an attractive combination of advanced semiconductor technology and biology. The photos of biological synapses are placed on top of the chip that is displayed at the center and lower part. This implies the intersection of computing and neuroscience, in line with the research on developing photonic synapses using GaN materials. More information can be found in article number 2400146 by Xiaoqin Yang, Bingcheng Luo, Hong Gu, and co-workers.

Multimodal optical microscopy can be used to visualize multiple biological targets at the same location and at similar spatial resolutions, enabling more comprehensive studies of complex biological samples. However, current multimodal imaging systems typically use multiple or specially designed excitation lasers. Furthermore, previous studies have reported only three or four types of imaging modalities, limiting the number of imaging targets. This study develops an imaging system based on a single commercialized 1550 nm fiber laser and integrates five imaging modalities: second harmonic generation, third harmonic generation, two-photon fluorescence, three-photon fluorescence, and reflected confocal microscopy. Utilizing the characteristics of different imaging modalities, various biological components in biological tissues can be revealed simultaneously and cell movement in a complex biological environment can be traced noninvasively. The capacity of deep imaging has also been demonstrated in the mouse brain. The results show that the proposed multimodal microscopy system has great application potential in biomedical fields, such as tumor diagnosis and immunotherapy.

The conventional von Neumann architecture is increasingly losing the capacity to satisfy the urgent demand for high-speed parallel computing, energy efficiency, and ultralow power consumption owing to the rapid growth of information. Brain-inspired neuromorphic computing presents an opportunity to overcome the inherent limitations of conventional computers. In recent years, photoelectric neuromorphic devices have garnered significant attention for their potential applications in brain–machine interfaces, intelligent sensing, and neuromorphic computing. Herein, a simple two-terminal light-stimulated synaptic device is fabricated using GaN thin films through metal-organic chemical vapor deposition. The device demonstrates the ability to mimic various biological synaptic functions, including learning-experience behavior, the transition from short-term to long-term memory, paired-pulse facilitation, and visual recognition and memory. In this research, an effective strategy for developing photonic synapses using GaN-based materials in neuromorphic computing and bio-realistic artificial intelligence systems is presented.

Optical Switches

In article number 2400118, Yuexin Yin and co-workers fabricate a 1 × N generalized Mach-Zehnder interferometer (GMZI) based switch on a polymer platform. This scalable structure is realized for 1 × 2, 1 × 3, and 1 × 4 switches. With phase compensations introduced, for the 1 × 4 switch, the bandwidth is 13 times larger for 10 dB crosstalk. Moreover, the all-logic optical switch is demonstrated based on the switch proposed.

Holographic Optical Elements

In article number 2400152, Jisoo Hong and co-workers present a holographic optical element (HOE) printer, which produces consistent shapes of hogels across various focal conditions, and aberration-corrected results by a printed HOE. The hogels, recorded by the interference of two beams, are printed sequentially with optimized gratings.

A single micrometer-size spherical colloid has been set in rotation by transfer of light orbital angular momentum. This particle is floating at an air–water interface. Steady-state rotational frequencies of the order of one hertz have been observed, depending on the topological charge of the beam and on its power, in agreement with expected values. The detection is performed using the rotational Doppler shift of the diffused light. Two time constants have been evidenced in the rotational velocity dynamics. The first one is related to the friction of the colloid with the fluid (air and water), whereas the other one is principally associated with the wall friction of the air–liquid interface with the container. This measurement technique makes it possible to identify dynamic parameters of the rotational movement of any spherical object, which is usually impossible to detect.

Optical Convolutional Neural Networks

In article number 2400108, Weiwei Tan, Jiale He, Guanhai Li, Xiaoshuang Chen, and co-workers propose an optical neural network that leverages GST-based microring waveguides for on-chip computing, offering 64 levels of transmission contrast with 6-bit resolution. It achieves high accuracy in image edge detection and recognition with potential for large-scale photonic neural networks.