Significance Polarization imaging technology has the advantages of non-invasive detection, rich information, sensitivity to the microstructure of the sample, and compatibility with traditional optical imaging technology, which makes it suitable for combining with microscopy techonology based on staining methods to distinguish the characteristics of different microstructures of pathological tissues. By adding the polarization state analyzer (PSA) and polarization state generator (PSG) modules to the commercial transmission and colinear reflection optical microscopes, the Mueller matrix microscopic imaging can be performed. The Mueller matrix can realize the complete characterization of the polarization properties of the sample.

Progress We have established the upright transmission Mueller matrix microscope ( Fig. 1, DRR-UTMMM) and collinear reflection Mueller matrix microscope ( Fig. 2, DRR-CRMMM) based on dual rotating retarders in which PSA and PSG modules consist of a fixed linear polarizer and a rotatable quarter-wave plate, respectively. The working principles are based on Fourier coefficient analysis. During each measurement process, two quarter-wave plates rotate with a fixed step angle ratio, and 30 images with different polarization states are collected to reconstruct the Mueller matrix of the sample. In order to eliminate the possible systematic errors and improve the measurement accuracy, the calibration of the Mueller matrix measurement system is necessary. For the transmission system, the analytic calibration method (ACM) is adopted. The systematic errors can be accurately solved by establishing the error model between each systematic error and the measurement signal. For the collinear reflection system, due to the complex relationship between the measurement signal and errors, it is difficult to derive the expression directly, so the numerical calibration method (NCM) is used to calculate the systematic errors and rebuild the system instrument matrix, which has a higher applicable scope and flexibility. In order to perform the fast Mueller matrix imaging, linear polarization CCD based on division of focal plane (DoFP) is used, which is capable of measuring the linear polarization states of light by equipping micro-polarizer array in front of the ordinary imaging sensor. We designed and implemented the upright transmission Mueller matrix microscope based on dual DoFP linear polarization CCDs ( Fig. 4, DoFPs-UTMMM). Two linear polarization CCDs are fixed in the transmission and reflection ends of a non-polarized beam splitter, respectively, one of which is equipped with a fixed angle quarter-wave plate. The real-time polarization state analyzer is realized by combining the multi-channel polarization data from two linear polarization CCDs. In order to eliminate the parasitic polarization artifacts introduced by the beam splitter, the Mueller matrix of the beam splitter is considered in PSA instrument matrix after calibration. Two different schemes of PSG are also discussed. The performance of DoFPs-UTMMM is validated by conducting Mueller matrix imaging on standard polarization samples with different azimuths. The results show that DoFPs-UTMMM has a higher measurement accuracy and faster measurement speed compared to DRR-UTMMM, which make it suitable for monitoring dynamic processes or living tissues.

In this article, we also introduce some biomedical applications of the Mueller matrix microscope. By perfoming Mueller matrix imaging of cancerous liver tissues in different stages and calculating polarization parameters, it can realize the characterization of the degree of liver fibrosis in the cancerous liver tissue (Figs. 5--7). By combining with data technologies including machine learning, new polarization parameters used to quantitatively characterize the microstructure of biological tissues can be derived, which can accurately distinguish specific pathological structures (Fig. 8). Through the fast and accurate polarization measurement of blood cells (Fig. 9), it can be tested that the system has the potential for real time and accurate polarization monitoring of the dynamic process of living cells in the future.

Conclusions and Prospects In this article, we summarize several modular polarization microscopes implemented in our previous studies, including the transmission Mueller matrix microscope and the collinear reflection Mueller matrix microscope based on dual rotating retarders, as well as the transmission Mueller matrix microscope based on dual linear polarization CCDs. Then we introduce some applications of modular polarization microscope in the biomedicine field. With the combination of microscopy and polarization imaging technology, Mueller matrix microscope can be directly upgraded from ordinary optical microscopy methods. It has the following advantages: suitable for the studies of biological living systems, capable of obtaining cross-scale image information, and easy to be combined with data science technologies. Besides biomedicine, the Mueller matrix microscope can be also applied to many fields including material science, defect detection, etc. There also exsit some aspects that need to be improved: relatively small imaging area, and high requirement for system stability and residual polarization artifacts of the optics inside the system. This article puts forward specific suggestions for the above problems. In the future, the development tread of Mueller matrix microscope is faster measurement speed and higher measurement accuracy. With the improvements of polarization modulation and measurement technology, the Mueller matrix microscope is expected to perform real-time and accurate full-polarization measurement of living cells and in-vivo tissues, and becomes an important tool to promote biomedical applications.

Significance Photoacoustic tomography (PACT) is a novel medical-imaging modality. During the imaging process, biological tissues are irradiated by nanosecond ultrashort light pulse. Fast energy deposition in tissues causes thermoelastic expansion and generates ultrasound emission. Such emissions can be detected by ultrasonic detectors. The penetration depth of PACT (>5 cm) is much higher than that of most optical-imaging modalities, approaching that of ultrasonic imaging. Therefore, PACT has wide potential applications in areas such as blood pressure monitoring, cancer detection, and small-animal studies.

PACT reconstruction requires knowledge of distribution of speed-of-sound (SoS). In practice, acoustic properties of biological tissues are inhomogeneous and unknown, resulting in image degradation in PACT. In its simplest implementation, PACT reconstruction assumes single SoS for both biological tissues and surrounding acoustic-coupling medium (i.e., water in most cases). However, considering that the SoS inside soft tissues varies from 1350 m/s (fat) to 1700 m/s (skin), such an assumption can sometimes be too idealized, resulting in splitting, blurring, and distortion of structural features. Furthermore, the SoS of a range of tumors is much higher than that of normal tissues. Therefore, imaging degradation is much severe in tumor imaging.

To solve this problem, a more rigorous acoustic model is required during the reconstruction process. Some studies have employed ultrasound devices for directly imaging the distribution of SoS. However, the most effective method is the joint reconstruction of initial sound pressure (IP) and SoS. Presently, there are several methods for joint reconstruction. Each method has advantages and disadvantages and is effective in different scenarios. Herein, we introduce several methods developed by our teams and summarize their advantages and disadvantages, supporting the biomedical application of PACT.

Progress The ring-array PACT system (Fig. 1) is widely used in IP-SoS joint reconstruction. Most joint-reconstruction methods are based on delay-and-sum (DAS) or back-projection (BP) reconstruction.

Several conventional joint-reconstruction methods have been developed, including dual-SoS, passive element, and model-based methods. Each of the methods has disadvantages and cannot meet practical application demand.

We have developed four joint-reconstruction methods, including feature coupling (FC), multisegmented feature coupling (MSFC), adaptive PACT (APACT), and signal compensating (SC) methods. The FC method, separates ring arrays into two halves. Two images are reconstructed using signals detected by the two halves separately. The correlation coefficient between the two images is then calculated to measure their similarities. The aim of optimization is to maximize the correlation coefficient. In vivo experiments have shown promising results (Fig. 2). The MSFC method further separates the ring array into eight subarrays. In numerical studies, the MSFC method has shown better reconstruction results for both IP and SoS distributions, with lower computational complexity (Fig. 3). The APACT method, inspired by adaptive optics methods widely used in optical imaging, estimates the inhomogeneity-induced wavefront distortions of photoacoustic signals in the frequency domain. The SoS distribution is calculated based on the estimates of the wavefront distortion. In vivo experiments have also shown promising results (Fig. 5). Numerical studies have further shown the advantages and disadvantages of the APACT method compared with the FC method. In summary, when using the FC method, prior knowledge of the distribution of SoS is vital. However, currently, the APACT method works well in the absence of prior knowledge (Fig. 6). The SC method utilizes the characteristics of electronic impulse response (EIR). In photoacoustic imaging, the EIR of ultrasonic detectors has positive and negative peaks. For the reconstructed IP images, when the positive peak detected by one detector is superposed by the negative peak detected by the opposite detector, the intensity of the reconstructed image is the lowest. Therefore, by minimizing the intensity of the reconstructed IP image, the inhomogeneity-induced wavefront distortion can be estimated. Fourier transform is applied to the reconstructed image to further analyze the relationship between delay time and image intensity in different directions. Numerical studies have shown promising results (Fig. 7).

Conclusion and Prospect Although PACT is a promising imaging modality for various biomedical applications, a better solution to the joint-reconstruction problem is needed. In summary, there are several methods for joint reconstruction, and each is effective in different scenarios. However, all these methods are designed for ring- or hemisphere-array systems, which limits their applications. Therefore, further research on joint reconstruction in linear array and multimodal systems is required to promote the development of this imaging modality.

Significance With the growing demand for health care and the development of medical examinations, more accurate and minimally invasive diagnosis and treatment technologies have received extensive global attention. Currently, a rapid and effective intraoperative diagnostic method is still lacking for major diseases in clinical practice. Traditional medical imaging modalities, such as magnetic resonance imaging, computed tomography, ultrasound imaging, positron emission tomography, and single-photon emission computed tomography, are commonly used to present the global anatomical structure or functional information of human tissue, but their resolution is too low to show fine structures. Histopathological examination is the gold standard for malignant tumors and other diseases, which detect pathological changes in cells on a microscopic scale with the best accuracy. However, the process is complex and time-consuming and depends on the distribution of biopsy samples; therefore, it cannot cover a wide range of tissues. In addition, diagnosis and treatment procedures are relatively independent, leading to the mismatch of preoperative and intraoperative information, and surgical operations largely depend on the personal experience of surgeons.

Represented by emerging optical imaging and spectroscopic methods, biomedical images at mesoscopic and macroscopic levels provide a good foundation for multimodal rapid and precise diagnosis, such as optical coherence tomography, two-photon microscopy, photoacoustic imaging, Raman spectroscopy and microscopic imaging, and fluorescence spectroscopy and imaging. Because of their excellent real-time performance, high accuracy, and resolution in intraoperative use, many of these methods are known as “optical biopsy”. In terms of treatment, optical methods with high spatio-temporal selectivity, such as laser ablation, photodynamic therapy, and photothermal therapy have gradually entered clinical practice. At the same time, the development of computer vision, precision instruments, automation, and other research fields has promoted more intelligent, accurate, and personalized diagnosis and treatment technology, including artificial intelligence-assisted medical image processing, minimally invasive surgical robots, intelligent treatment planning, and navigation. On this basis, by combining optical imaging and treatment, we can build an intelligent theranostic system, which can break down the barrier between traditional diagnosis and treatment, improving the current surgical process. The accurate intraoperative diagnosis results are directly used for treatment planning and control, which can achieve intelligent, quantitative, and accurate lesion clearance. These emerging technologies are of great significance for the diagnosis and treatment of tumors and other major diseases in clinical practice. Therefore, summarizing the existing studies regarding emerging optical theranostics technologies is necessary to guide the future development and clinical transformation of this field.

Progress In this paper, the research progress of intelligent precise optical diagnosis and therapy technology, specifically for malignant tumor theranostics, is reviewed based on three aspects: 1) optical imaging and intelligent diagnosis methods (Fig. 2); 2) precise optical treatment methods (Fig. 4); 3) optical diagnosis and therapy instruments and theranostic methods (Fig. 6). Through intelligent optical diagnosis, the location of lesions could be automatically determined through computer-aided image processing, and we can plan and control precision optical treatment using theranostic algorithms and hardware systems. There are various optical imaging methods used in clinical or preclinical experiments, and some clinical optical diagnosis standards are established preliminarily. However, most doctors have not been trained to read optical images (or spectra); thus, computer-aided automated or quantitative diagnosis is currently the most appropriate method (example given in Fig. 3), which involves quantitative parameter extraction, machine learning, deep learning, and other methods. We focus on several conventional mainstream optical diagnostic modalities with intelligent diagnostic algorithms, including fluorescence and imaging spectroscopy, Raman spectroscopy and microscopy, optical coherence tomography, and photoacoustic imaging. Then, we describe several emerging optical treatment methods, including laser ablation, photodynamic, photothermal, and other light-activated therapies. Precision theranostic devices and methods are divided into four categories and reviewed: imaging field enlargement, improving image quality, multimodal imaging, and the integration of diagnosis and treatment.

Conclusions and Prospects Optical diagnosis and treatment of major diseases, especially integrated diagnosis and treatment technology, can considerably improve the clinical processes and treatment prognosis. We expect that intelligent, quantitative, and accurate optical diagnosis and treatment technology will play a more significant role in human life and health, promoting the development and progress of clinical diagnosis and treatment of malignant tumors.

Significance Recently, neuroscience has attracted great attention around the world. To prompt the study of neuroscience, a lot of countries have launched brain projects, in which the development of advanced neural techniques is regarded as the driving force. Optical techniques own the advantages of being less-invasive and high spatial resolution, etc, promising for neural activity recording and manipulation. Compared to traditional electrode stimulation methods, optical stimulation based on optogenetics could selectively excite or inhibit specific neural ensembles, benefiting from the introduction of gene engineering. So far, a variety of opsins have been developed for the activation or inhibition of neural activity. On the other hand, to achieve selective manipulation at single-neuron resolution, the techniques for two-photon optogenetics are emerging. Here, we review various strategies for illumination in two-photon optogenetics. We summarize their technical principles, and discuss their advantages and disadvantages.

Progress Illumination strategies in optogenetics can be classified as conventional illumination strategies based on single-photon absorption and high-spatial-resolution illumination strategies based on two-photon absorption.

In the early days of optogenetics, the wide-field illumination strategy based on single-photon absorption was used to manipulate neurons with opsins. Due to the scattering and absorption of biological tissues, the power of illumination light decreases significantly with the increase of penetration depth in wide-field illumination strategy. For neuron manipulation in deep tissues, fiber-coupled illumination is performed, in which the excitation light is guided through the fiber to the targeted depth. However, specific manipulation of neural activity at single-neuron resolutions is not achieved in neither wide-field illumination nor fiber-coupled illumination, due to the fact that all neurons with opsins in the illumination region would be excited.

For specific manipulation of neural activity at single-neuron resolution or sub-neuron resolution (such as a dendrite or a dendritic spine), the two-photon illumination strategy has to be adopted, which ensures high spatial resolution in three-dimensional (3D) space. Besides, the longer wavelength in two-photon excitation is more robust to scattering and leads to a deeper penetration depth than that of the conventional illumination strategies.

In general, the two-photon illumination strategies can be classified as serial scanning illumination and parallel illumination based on phase modulation. In the former, a single focus is steered to perform spiral scanning on a neuron to open enough ion channels for neuron excitation before being switched to another neuron (Fig. 1). Or, serial scanning with soma-patterned illumination can be employed (Fig. 2). However, the temporal resolution in these methods is low, which makes it only compatible with opsins of slow turn-off time. Besides, parallel illumination can be achieved based on phase modulation (Fig. 3), in which the phase can be calculated by the generalized phase contrast (GPC) method and the computer generated holography (CGH) method. The phase modulation plane of GPC is the conjugate plane of the focal plane, while that of CGH is the conjugate plane of the Fourier plane. The GPC method has good uniformity of excitation patterns but is of a poor energy efficiency. Based on the theory of CGH, two illumination schemes have been developed, i.e. the multi-foci generation combined with the spiral scanning strategy, and the multiple extended-pattern generation strategy. The former has a high energy efficiency and can realize the excitation of 3D distributed neurons, but it is of poor temporal resolution and only works with opsins of slow turn-off response time. The latter can directly generate a multiple expanded-pattern and excites multiple neurons simultaneously. Combined with the temporal focusing technology, this method has high temporal resolution and high axial resolution, however, its energy efficiency is low. At the same time, we summarize the commonly used CGH algorithms for parallel illumination based on phase modulation, which contain superposition algorithm, Gerchberg-Saxton (GS) algorithm, non-convex optimization (NOVO) algorithm, and DeepCGH algorithm. The diagrams of different algorithms are presented (Fig. 4). The basic ideas, advantages, and disadvantages of these algorithms are briefly pointed out.

Conclusion and Prospect Conventional neural manipulation in vivo relies on single-photon illumination, which is not good for specific excitation of neural ensembles at high spatial resolution. To this end, several techniques of two-photon optogenetics have been proposed, and have achieved in vivo neural manipulation at high-spatial-resolution. We summarize the development of two-photon illumination strategies, and compare their advantages and disadvantages. Then we discuss the potential issues in the practical employment of two-photon optogenetics, such as excitation precision and field-of-view. We expect that, the all-optical physiology, in which two-photon imaging and two-photon optogenetics are combined, is promising in neuroscience, benefiting from the simultaneous monitoring and manipulating of neural circuits in vivo.

Objective Optical coherence tomography (OCT) has become the de facto gold standard of diagnosis in ophthalmology. In recent years, with the rapid improvement in imaging speed and the wide adoption of OCT angiography (OCTA), a large amount of retinal OCT B-Scan can be generated in one clinical scan. Automatic and effective retinal tissue segmentation is required to realize this trend. Conventional segmentation algorithms based on path searching are time consuming and error prone when dealing with morbid retinas. In such cases, neural-network (NN)-based methods such as U-Net and ReLayNet are promising approaches. These NN-based methods differ in complexity and performance. For the desktop computer in the current mainstream OCT equipment, an NN with moderate parameter sets and high performance is highly desirable. In this study, we demonstrate a novel end-to-end segmentation method for retina images, named multiple-scale inpainting convolutional NN (MsiNet) for retinal layer segmentation. MsiNet is based on human visual characteristics and can be implemented on a desktop computer with high performance. The framework was validated on two retina image datasets with comparisons against U-Net and ReLayNet, which are well established in retinal OCT image segmentation. MsiNet showed better performance than the other two methods in terms of both retinal layer segmentation accuracy and morbid tissue segmentation, with a moderate parameter set size suitable for desktop computers.

Methods MsiNet is based on semantic segmentation with encoder-decoder architecture and convolutional NNs (CNNs). We regarded retinal layers as different categories and predicted a pixel’s probability to different categories. The human visual system usually detects objects in two steps: first, it tends to obtain semantic (outline, location, etc.) information; second, it is used to focusing on details. Inspired by this fact, we employed a small-scale network as a decoder to refine semantic information and then used inpainting networks to extract spatial structures from high-resolution feature maps for inpainting low-resolution results. Thus, semantic and detailed information in different stages could be refined directly with less redundancy. To fit MsiNet into the limited computation resource, we reduced the number of parameters and floating-point operations using new structures: interlaced residual unit (IRU) and biased fusion unit (BFU). We also adopted a single-stage decoder instead of a traditional decoder and improved the segmentation results stage by stage. Further, we designed a joint weighting method for some special pixels to intensify punishment. Multiple losses were provided in different resolution stages for obtaining different resolution results. Compared with two well-established NN-based methods, the segmentation accuracy on edges was significantly improved.

Results and Discussions MsiNet was tested on two retinal OCT datasets: the SD-OCT dataset and a dataset provided by a third-party company (TOWARD π Medical Technology). We compared MsiNet with two well-established NN-based methods: U-Net and RaLayNet. First, using the GSE(generalized semantic boundary) weighting method, the accuracy of edge and disease tissue prediction of MsiNet was better than those of U-Net and RaLayNet (Table 1). Second, by comparing the outputs of different stages in Fig. 6, we confirmed that a high-level decoder significantly improves the accuracy of low-level outputs. Third, MsiNet outperformed U-Net and RaLayNet in terms of both retinal layer and morbid tissue segmentation accuracy, with a moderate parameter set size suitable for desktop computers. The results of the three methods are demonstrated in Fig.7 and Table 1.

Conclusions Based on human visual characteristics, we propose MsiNet for retinal tissue segmentation. MsiNet replaces the traditional decoder that merges different-resolution feature maps with a single-stage decoder in low resolution, and an inpainting network is designed to rectify segmentation errors and add structural information to phased results. An extended GSE mask is applied to the loss function to adjust the weights of edge pixels. Because of the clear semantic information, the parameter set size is significantly reduced. Experiments show that MsiNet outperforms U-Net and ReLayNet in terms of both layer segmentation and morbid tissue segmentation, mainly due to the improvement in edge point classification.

Significance Optical coherence tomography (OCT) is a depth-resolved biomedical in vivo imaging technique providing cross-sectional and three-dimensional (3D) images of tissue microstructure with a micrometer-scale resolution. In the past few decades, OCT has been employed in various applications, including clinical and material research areas. In this paper, some studies on improving the performance of OCT are introduced. Among them, imaging resolution, imaging speed, and multifunctional integration are the basic core indicators of the imaging system. Recently, research on OCT conducted at the Department of Physics of Tsinghua University has achieved a series of important advances in improving imaging system performance.

Progress For high-resolution imaging, we achieved ultrahigh-resolution OCT with a supercontinuum by coupling femtosecond pulses generated from a commercial Ti∶sapphire laser into an air-silica microstructure fiber. The visible supercontinuum from 450 to 700 nm centered at 540 nm was generated. A free-space axial OCT resolution of 0.64 μm was achieved. The sensitivity of the OCT system was 108 dB with an incident light power of 3 mW at a sample, only 7 dB below the theoretical limit.

For subcellular and long-time in vivo imaging, we developed a novel system of full-field OCT (FF-OCT) for label-free 3D subcellular in vivo imaging of preimplantation mouse embryos. As the sample received much less optical dose than in conventional confocal imaging, the preimplantation mouse embryos were alive even after several days’ live imaging. Various typical preimplantation stages, including zygote, two-cell, four-cell, and blastocyst (at embryonic day 3.5, E3.5 for short), were investigated with a spatial resolution of 0.7 μm and imaging rate of 24 frame/s. These are the first in vivo studies with mammalian embryos at the beginning of their embryonic lives for understanding early patterning and polarity.

For high-speed imaging, a high-speed swept source is necessary for system setup. Real-time 3D high-definition OCT imaging requires 1000 x-scan×1000 y-scan×30 times/s refresh rate, i.e., 1000×1000×30=30 MHz, implying that the laser should have a sweeping speed of at least 30 MHz. Therefore, we devised an all-optical swept source with an A-scan rate of 40 MHz, the fastest one thus far. The inertia-free swept source, having an output power of 41.2 mW, tuning range of 40 nm, and high scan linearity in wavenumber with a Pearson correlation coefficient r of 0.9996, comprised a supercontinuum laser, an optical band-pass filter, a linearly chirped fiber Bragg grating, an erbium-doped fiber amplifier, and two buffer stages. With a sensitivity of 87 dB and a 6-dB fall-off depth of 0.42 mm, ultrahigh-speed sweeping OCT based on sweep laser (SS-OCT) imaging of biological tissue in vivo was demonstrated.

Ultrahigh-speed real-time optical imaging requires the real-time information processing of big image data. Real-time 3D high-definition OCT has a 1000 pixel (X)×1000 pixel (Y)×1000 pixel (Z)×30 times/s refresh rate, indicating that 30-Gbit/s data flow needs to be processed in real time. To address this challenge, we proposed a novel all-optical computing technique to process the signal in the spectral domain with a fiber-optics system other than compute interpolation based on fast Fourier transform algorithm (FFT) using an electronic computer, resulting in a significant reduction of processing time and enahancement of imagin speed. In the so-called optical computing OCT, the Fourier transform of the A-scan signal was optically processed in real time before the light was detected by a photoelectric detector. Low-coherence continuous wave (CW) light centered at 1550 nm with an average power of 22 mW and a bandwidth of 40 nm was generated by a superluminescent diode. The CW light was modulated by a 10-GHz-bandwidth intensity modulator, known as the Mach-Zehnder modulator (MZM). The MZM was biased by a power supply and driven by a cos(at²) waveform signal generated by an arbitrary waveform generator (AWG). With this optical computing system, a processing rate of 10⁷ A-scan per second was experimentally achieved, which is the highest speed for OCT imaging to the best of our knowledge.

Because tiny endoscopic probes work as its sample arms, the applied range of the proposed OCT technique can be expanded to various internal organs such as arteries and esophagus. To further enhance the feasibility of the proposed OCT technique, we also proposed and fabricated a prototype focus-adjustable endoscopic probe with an outer diameter of 2.5 mm and a rigid length of 32 mm based on a two-way shape-memory-alloy (SMA) spring and an in-house hollow-core ultrasonic motor. This novel probe has adjustable focus and hence a larger scanning range, with high resolution and no sensitivity loss. The focus-adjustable range was more than 1.5 mm, with a 100.3-dB sensitivity and the best lateral resolution of ~4 μm. With the use of a hollow-core motor, the probe can provide an unobstructed 360° field of view. To the best of our knowledge, this is the first demonstration of a focus-adjustable probe for C-mode scanning in endoscopic OCT. We believe that this novel probe will be useful in future biomedical applications.

Conclusion and Prospect This paper presents some remarkable research advances at Tsinghua University and is dedicated to celebrating the 110th anniversary of Tsinghua University.

Significance As targets become increasingly complicated, image detection with high sensitivity, high resolution, and high precision has become essential. However, owing to their high cost and large size, monostatic radars fail to meet the abovementioned requirements. To overcome this problem, we propose a distributed coherent-aperture-imaging radar (DCAIR). DCAIR is a novel radar system that uses multiple spatially dispersed small-aperture unit radars for cooperative detection and imaging. By employing the signal-level coherent fusion of the unit radars, the DCAIR obtains target images with high signal-to-noise ratios (SNRs). Thus, it is an important means to deal with long-range, low radar cross sections (RCS), or small threats. Additionally, DCAIR has many advantages such as flexibility, high survivability, low cost, and strong maintainability, making DCAIR an important development in the direction of imaging radars. Further improving the detecting resolution requires DCAIR to generate and process high-frequency broadband signals. However, traditional DCAIR realized using purely electronic technologies suffers severely from the “electronic bottlenecks” , making it difficult to generate and process high-frequency broadband-radar signals directly. Moreover, conventional time and frequency synchronization technologies fail to strike a balance among the transmission distance, stability performance, and synchronization precision. Microwave photonics has been considered a promising solution to these bottlenecks. Because of the broad bandwidth, flat response, low loss transmission, and multidimensional multiplexing of photonics devices, microwave photonic technologies have merits in high-frequency broadband signal generation, transmission, and processing. Combined with microwave photonic technologies, DCAIR exhibits better performance in terms of range resolution, velocity resolution, angular resolution, and SNR gain. With the funding of major programs of the National Natural Science Foundation of China, many achievements have been made. This paper highlights the achievements of DCAIR based on microwave photonic technologies proposed by researchers at Tsinghua University.

Progress In this paper, the international developing status of the three key modules is briefly reviewed, including generation of a dynamic reconfigurable waveform, optical fractional Fourier domain receiver front-end, and high-precision fiber-optic time-frequency synchronization network (OTFSN), and the the achievements funded by the major program of the National Natural Science Foundation of China are highlighted. From these achievements, the first experimental X-band distributed coherent broadband imaging radar system using microwave photonics was built and the staged experimental results were obtained. To generate multichannel orthogonal waveforms and achieve dynamic switching to coherent waveforms, a generation method for dynamic reconfigurable radar waveform using photonics-based broadband is proposed. We use the phase-coded linear frequency modulated waveform (PCLFMW) as the orthogonal waveform and the linear frequency modulated waveform (LFMW) as the coherent waveform. Here, two PCLFMWs in X-band with a bandwidth of 3.5 GHz are generated and the orthogonality between the waveforms reaches about 29 dB ( Fig.3). The proposed scheme achieves arbitrary generation and dynamic reconfiguration of the waveform. Furthermore, an optical FrFD receiver front-end is proposed to eliminate ghost targets produced by multiple echoes that are overlapped in both the time and frequency domains. The received broadband LFMW echo signals are projected on the optimal fractional Fourier domain formed using the photonic rotation of the time-frequency plane. By controlling the fractional Fourier transform spectrum, the proposed receiver front-end cancels ghost targets in multitarget circumstances. Experimental results show that the proposed receiver front-end can adapt to multiple noncooperative target environments and is immune to ghost targets under optimal working conditions ( Fig.6). An all-optical stable quadruple frequency dissemination scheme using photonic microwave phase conjugation is presented over a fiber-optic loop link. The relative frequency stability of 10 ^-16 at 1000-s averaging time can be obtained at every remote site located at a 20-km fiber loop link ( Fig.8). Moreover, a fiber-optic two-way time transfer method based on the time-frequency domain transforms (TFDT) is proposed. The TFDT directly obtains the two-way transmission delay by chirp frequency mixing and time-frequency analysis. With the proposed method, the time offset fluctuation and TDEV can reach 5.6 and 0.36 ps at 10000 s, respectively ( Fig.10). Further, using the stable frequency transfer technology, a three-node time-frequency synchronization network over a 20-km fiber loop link is presented. At 1000-s averaging time, frequency stability levels reach 10 ^-16 and time deviations reach 0.5 and 0.8 ps for two sites, respectively ( Fig.12). Two X-band two-unit microwave photonic DCAIRs based on the static and dynamic OTFSN are demonstrated. From the static OTFSN, the SNR gain ratio relative to the coherence-on-transmit mode is ~5.98 dB and that for the full coherence mode is ~8.6 dB. The range and cross-range resolution of 3.4 cm and 4.3 cm, respectively, are achieved in the experiment for rotating-target imaging ( Fig.14). From the dynamic OTFSN, the fully coherent SNR ratio gain can be increased by 8.1 and 7.9 dB, respectively, for the two-unit radars ( Fig.16). Thus, weak targets can be imaged and probed using the mutually coherent operation, while they are undetectable using the single radar.

Conclusions and Prospect This paper introduces the achievements of high-frequency broadband DCAIRs using microwave photonics technologies proposed by researchers of Tsinghua University. Combined with microwave photonic technologies, DCAIR realizes high-resolution and high-precision imaging. The abovementioned achievements will promote the development of DCAIR. The microwave photonics-based DCAIR has a wide application potential in both civil and military fields.

Objective Imaging is the most intuitive way to perceive the world. The resolution of traditional imaging method is limited by the diffraction limit caused by the limited aperture of lens system. Ptychography is a non-lens imaging method based on coherent light, which avoids the diffraction limit problem caused by lens system. At present, the mechanical structure is often used to translate the optical probe in Ptychography, which brings errors of the position of the optical probe, resulting in the degradation of the imaging quality. Thus in this paper, we proposed a beam steering chip to avoid the errors of the position of the optical probe. The chip can replace the traditional mechanical optical probe to be used in ptychography. The 100 nanometer processing technology ensures that the position of the optical probe will not have errors, which greatly improves the quality of imaging and the stability of the system. On the other hand, because the size of the chip is only millimeter level, the size of the imaging system is also greatly reduced.

Methods The silicon-based integrated beam steering chip is processed on a 220 nm silicon on insulator (SOI) platform. The main structure of the chip is cascaded optical transmitting antennas with a filter system. The CMOS process with a linewidth of 180 nm is used to ensure that the position error of the optical probe caused by the actual processing process will not exceed one ten thousandth of the distance between adjacent optical transmitting antennas. Therefore, the position error of optical probe can almost be ignored. Light with different wavelengths will be sent to its corresponding optical transmitting antenna. The transverse spacing between adjacent optical transmitting antennas is 120 μm while the longitudinal spacing is 150 μm. After collimated by the lens, the light emitted from each light transmitting antenna can act as a light probe. Because the corresponding wavelength of each optical probe is different, we can adjust the wavelength to decide which optical probe to scan. At the same time, because the position of the optical transmitting antenna on the chip is fixed, there is almost no error in the position of the optical probe in the imaging process. This avoids the influence of optical probe position error on imaging quality.

Results and Discussions When the beam steering chip on the integrated platform is used to replace the mechanical structure for Ptychography, the whole image-forming system can be shown in Fig. 5. Fig.6 shows the diffraction patterns obtained from the CCD, the amplitude and phase information of the sample recovered by the PIE algorithm, and the curve of the error function during the operation of the algorithm. It can be seen that according to the diffraction pattern of CCD obtained in our experiment, the amplitude and phase information of the sample are successfully recovered. Because all 16 diffraction patterns are recorded in one exposure, there is actually overlap between the adjacent diffraction patterns on CCD, and the overlapped part is the high frequency component of the sample, which leads to the crosstalk between the adjacent diffraction patterns. Second, compared with the high-frequency component in the diffraction pattern, the power of the low-frequency component is much stronger. It will lead the center frequency to be overexposed, which also affects the final image quality. According to the introduction, the problem of center frequency overexposure can be solved by using baffle to block the zero order diffraction light. To sum up, we have successfully proved that the integrated beam steering chip can be used to complete space stack diffraction imaging and solve the problem of optical probe position error.

Conclusions In this paper, Ptychography based on integrated beam steering chip is proposed. After collimating, the light emitted from the optical transmitting antenna on the chip becomes the optical probe. Because the filter system is cascaded in the front end of the optical probe, the switch of the optical probe can be controlled by controlling the wavelength of the input light, which increases the diversity of imaging methods. At the same time, due to the fixed position of the optical transmitting antenna on the chip, there is almost no error in the position of the optical probe, which solves the huge impact of the position error of the optical probe on the imaging quality. Compared with the previous research on the error of the position of optical probe, our method solves the problem by device rather than by correction on the algorithm. At the same time, our method not only does not increase the complexity of the imaging system, but also greatly reduces the size of the whole system. The stability and robustness introduced by integrated platform processing will play an important role in the field of laser imaging. Using a small chip instead of a large volume device to realize imaging not only reduces the cost of the system, but also avoids the impact of environmental vibration and noise on the optical path of precision imaging. So using the power of integrated photonics to achieve more complex imaging system will be the focus of our future work.

Significance Recently, super multi-view light-field display devices have made great progress, showing the characteristics of increasing views, higher resolution, and larger viewing angles. The virtual scene to be displayed is becoming increasingly complex.

Herein, we review and summarize the existing virtual stereo content generation technology for super multiview light-field display devices. Their applicable scenarios, merits, and defects are illustrated. The main challenges of content generation of exisistng super multiview light-field display devices are highlighted. Real-time light-field content generation tehcnology for large-scale virtual scenes of super multiview is not fully mature. A new rendering system needs to be explored to break through the bottleneck of existing algorithms. With the technical breakthrough, the popularization and application of light-field in key fields such as military affairs, agriculture, and fire protection could be solved.

Progress Recently, the super multiview virtual content generation technology has made consistent progress in the following three aspects, which are summarized in table 2.

Graphic processing unit (GPU) instancing is used to improve the view-by-view light-field rendering technology (Fig. 18). When there are many repetitive objects in the virtual scene that needs to be drawn, the GPU instancing technology only requires to set the original data buffer once and then draw calls. Compared with the original view-by-view rendering algorithm based on geometric shaders, as the number of views increases, GPU instancing requires no more memory. When there is no object repeatedly drawn in the virtual scene, the super multiview light-field image of a large-scale scene can be rendered simultaneously.

The delay shading technique is introduced to improve the super multiview light-field rendering technique based on depth rendering (Fig. 23). Depth-based calculations can produce false illumination information, which is considerably different from the illumination characteristic in real space. The three-dimensional technology team of the Beijing University of Posts and Telecommunications proposed an image generation method of super multiview light-field based on delay coloring in 2020. Its reference image comprises one or more pairs of color, depth, highlight, and environment images. Although the lighting problem is solved using this method, it still leads to the degradation of multiview image quality and depth-based rendering.

An ultrafast experimental light-field rendering pipeline is constructed using geometric correlation (Fig. 24). In 2020, a super multiview rendering pipeline for polygon rasterization based on the principle of geometric correlation was developed. In the case of a 2×10⁸ surface slice with a total resolution of 7680×4320, the rendering rate reached 60 frame/s. Other factors have not been considered, such as lighting model and texture mapping, so it cannot be compatible with the traditional hardware rendering pipeline.

Conclusions and Prospect Many rendering algorithms can be used to generate super multiview light-field content. The ray tracing algorithm has the highest rendering quality, which can be used for nonreal-time applications. The algorithm with the fastest speed is the rendering algorithm based on geometric correlation; however, it is incompatible with the existing hardware rendering pipeline. Therefore, that lighting and occlusion need to be treated specially in the rendering process. Currently, the two most practical real-time rendering technologies are the rendering algorithm based on GPU instancing and the deferred rendering method. The former could degrade the image quality, whereas the latter is not fully competent for high-quality rendering of large-scale scenes. In general, the real-time light-field content generation technology for large-scale virtual scenes of super multiview is not yet fully developed. Therefore, it is still necessary to explore a new rendering system to break through the bottlenecks of existing algorithms.

Objective Automatic registration of optical and synthetic aperture radar (SAR) images is challenging owing to the significant geometric and radiometric differences between optical and SAR images. In this study, the phase congruency algorithm was used to calculate the phase because of its radiation invariance, construct feature direction information and feature intensity information, and establish a local feature descriptor, i.e., maximum phase index map (MPIM). Consequently, the corresponding points are obtained from the input image using the correlation measure of MPIM. Moreover, the projection transformation is used to achieve registration. Experimental results show that the proposed method shows strong adaptability to the radiation difference between optical and SAR images and exhibits a high registration accuracy.

Methods In this study, a multimodal image matching algorithm based on phase congruency was proposed. First, the definition of phase congruency was presented based on the frequency domain transformation. Then, the energy concept was introduced into the phase congruency calculation and the energy function was calculated using the odd and even filters of the Log-Gabor wavelet filter. The phase congruency transformation was obtained and found to be consistent with human vision. Further, the maximum moment M_ψ was calculated based on the phase congruency transformation and the corner points of the image were extracted using the M_ψ graph. Thereafter, a multimodal image matching algorithm based on MPIM of the Log-Gabor filter response was obtained.

Results and Discussions As shown in Fig.2, the image texture is preserved and noise is suppressed after the phase congruency transformation. By comparing the Harris corner points extracted from the original image and M_ψ based on the phase congruency transformation, the corner distribution extracted from M_ψ is more reasonable and conducive for subsequent image matching. Then, to test the performance of the MPIM operator, the information extraction ability of the operator is assessed by changing the template size and observing the number of matching points. The matching results of MPIM, histogram of oriented gradients, and multi-innovation algorithms are observed by changing the template size from 20 to 100. The results show that the proposed MPIM operator-based method exhibits a strong ability to extract the image texture. To verify the matching ability of the proposed algorithm for multimodal images, four sets of experiments were designed to perform optical, SAR, LiDAR, and electronic map registration to evaluate the algorithm in terms of neutrosophic c-means and root mean square error. The resolution and time span of the experimental images are large, and the radiation difference is obvious; hence, image matching is difficult. Table 2 shows the registration results. The proposed MPIM algorithm can adapt to radiation changes and stably extract the image texture for registration, thereby achieving a good matching effect.

Conclusions The registration of optical and SAR images is difficult owing to the large nonlinear radiation difference between optical and SAR images. To solve this problem, a registration method based on the MPIM descriptor was constructed based on the phase congruency to extract the image texture and resist noise. A descriptive operator with radiation invariance and texture characterization was obtained. Using the intermediate results of phase congruency calculation and combining them with the Harris operator, MPIM, normalized cross correlation measure, and random sample consensus algorithm, the problem of large local similarity difference between optical and SAR images owing to radiation differences is solved. In the experiment, the Harris operator was used to compare the feature points extracted from the M_ψ graph and the original image. Moreover, it is verified that the combination of M_ψ graph and Harris operator can extract more convincing feature points. Using four sets of multisource image experiments with large differences, it is verified that the proposed method can register optical and SAR images with robustness and obtain high matching accuracy. This method is not only applicable to optical and SAR image matching but also to the other images with large radiation differences. However, the proposed algorithm shows weak resistance to image scale and rotation and cannot complete matching in the case of large rotation and scale differences between images. Furthermore, the phase congruency requires the image texture to be very rich and mismatching can easily occur when the texture information is not rich. Future researches should focus on this problem to improve the ability of operators to adapt to more complex environments.

Significance Optical frequency combs (OFCs) with ultrahigh stability have revolutionized metrology science and facilitated a series of new research disciplines and directions, such as optical clock, precision spectroscopy, absolute distance measurement, and astrophysics. Recently, dual-comb spectroscopy (DCS), as an emerging broadband spectroscopic technique, has attracted substantial attention owing to its outstanding advantages of ultrahigh-frequency resolution, accuracy, and sampling rate. DCS has significant applications in various areas, such as gas absorption spectroscopy, greenhouse gas monitoring, and nonlinear spectral imaging. To achieve DCS, two coherent OFCs with a small offset of comb spacing are demanded. Various approaches could be used to generate these two OFCs, including phase-locking two independent mode-locked lasers, modulating a continuous-wave (CW) laser using two electro-optic modulators, or by directionally pumping a microresonator using a CW laser ( Fig. 3). However, those dual-comb systems generally require high-cost and bulky electronic configurations, a limited number of comb lines, or sophisticated fabrication processes, which pose substantial challenges for their practical applications. Dual-comb lasing in a single fiber laser (known as single-cavity dual-comb mode-locked fiber laser) is a viable alternative for releasing DCS, owing to mode-locking of direction/wavelength/polarization/cavity multiplexing. Because both combs are created from a single cavity, a compact and low-complexity DCS is possible owing to intrinsic high phase coherence and the absence of complex servo locking systems. This article reviews the recent research progress of single-cavity dual combs, ranging from their generation approaches, soliton dynamics to applications.

Progress Four types of multiplexing methods to generate dual combs from a single cavity-directions ( Fig. 4), wavelengths ( Fig. 5), polarization ( Fig. 6), and cavity spaces are first summarized. Generally, the spectra of wavelength multiplexing dual combs do not overlapp, leading to the necessary implementations of amplification and spectral broadening for practical applications, which increases the complexity of the entire system. Polarization multiplexing and bidirectional mode-locking create dual combs with overlapped spectra, but their tunable range is a limitation. Although noncommon optical-path structures could flexibly tune the offset repetition rate in a larger range, most spatially optical delay lines cause the system to be complicated. Therefore, novel single-cavity dual-comb lasers with all fiber and flexible tunability of the offset repetition rate are vigorously explored to overcome those pitfalls.

Mode locking is a complicated process that locks a huge number of longitudinal modes with the same phase difference to form ultrashort pulses. Single-cavity dual-comb mode-locked lasers could produce two stable pulses with slightly varying repetition rates. These novel asynchronous have a range of spectral and temporal features as well as differing intracavity evolution tendencies. This distinctive behavior distributes different net gains on both pulses leading to their distinguishable buildup dynamics. The entire buildup processes of asynchronous vector solitons in a polarization-multiplexed dual-comb fiber laser include relaxation oscillation, quasi-mode locking, spectral beating dynamics, and stable mode-locking (Fig.7). Polarization hole burning in the erbium-doped fiber might facilitate the different buildup periods of both vector solitons. Besides, counter-propagating pulses in a bidirectional ultrafast fiber laser undergo splitting after beating dynamics caused by modulation instability and finally annihilate to a stable bidirectional pulse. Buildup dynamics of counter-propagating ultrashort pulses might also undergo a long-time Q-switched mode-locking, accompanied by the formation of multisoliton structures.

Various group-velocities of both pulses cause the inevitable soliton collisions in dual-comb fiber lasers. Strong soliton-soliton interactions during collisions lead to their fascinating transient spectral evolutions and rich nonlinear phenomena. We summarize the collision processes of asynchronous scalar and vector solitons in dual-wavelength and polarization-multiplexed mode-locked fiber lasers, respectively. Scalar solitons experience the collision-induced self-reshaping process, such as the central wavelength shifts, dynamic spectral fringes, and rebuilding process of the Kelly sidebands. Vector-soliton collisions, however, could yield substantial four-wave mixing sidebands owing to cross-polarization coupling and the formation of another subordinate pulse on each polarization component (Fig. 8).

Intrinsic mutual phase coherence of single-cavity dual combs ensures the simple configuration for practical applications. Both frequency combs’ heterodyne beats map the molecular fingerprint from optical domain to radio frequency domain, which could be reconstructed using the Fourier transform of the time-domain interferogram obtained by a photodetector. Thus, DCSs by free-running dual-comb mode-locked fiber lasers have been leveraged to measure the gas absorption spectrum of hydrogen cyanide, water vapor, and methane, the coherent anti-stokes Raman scattering spectroscopy, and especially the optical sensing with fiber Bragg gratings (Fig. 9), as well as the application domains in absolute distance measurement and fiber optic gyroscopes.

Conclusions and Prospects Single-cavity dual-comb mode-locked fiber lasers exhibit remarkable merits of simple configuration and low cost, presenting broadband applications in enormous disciplines. However, small repetition rates, weak tunability, difficult self-starting, limited operating wavebands, are some of its limitations. Thus, some works ought to be examined in the future.

Significance The laser has had a revolutionary impact on scientific research and technological applications since it was invented more than 60 years ago. Extensive theoretical and application research has been conducted, and many important developments have been reported. For example, the size of lasers has both increased and decreased. The linear dimensions of lasers have increased by more than 10 orders of magnitude, and ultra-small semiconductor lasers that are uniquely important for many applications have been developed.

Driven by Moore's law, continuous progress of microelectronics technology has resulted in unprecedented challenges and requirements. Developments in microelectronics technology have presented significant possibilities related to the transition from electronics to photonics for information transmission and processing. However, the field of integrated nanophotonics, particular in relation to lasers, still faces obstacles, including dimensions, energy consumption, and integration with silicon photonic devices. Currently, semiconductor lasers are generally more than tens of microns. To be more compatible with electronic devices, the size must be reduced by two to three orders of magnitude. According to system level analysis, the energy consumption of on-chip optical interconnection needs to be less than 10 fJ/bit, and the data transfer rate must be greater than 10 Gbit/s. Studies have shown that the power-to-bandwidth ratio decreases as the device size decreases. Silicon is an indirect bandgap semiconductor that emits light inefficiently; thus, it is necessary to integrate lasers based on other materials with silicon-based electronic chips. Nanoscale lasers have the potential to overcome the influences of mechanical strain caused by lattice mismatch and to be integrated with silicon. Therefore, the development of nanolasers is significant no matter which aspect is considered.

In addition to the on-chip interconnection required by future information technology, detection, sensing and high-definition display based on nanolasers are also important application areas. Currently, continuous miniaturization and stable operation under electrical pumping are being pursued. The emergence of novel cavity designs and gain materials have created new opportunities for nanolaser research.

Progress The emergence of semiconductor nanolasers followed naturally from the development of semiconductor lasers. Since first developed in 1962, semiconductor lasers have undergone several breakthroughs in cavity designs, and each breakthrough has led to improved performance, lower thresholds, reduced size, and the appearance of new application scenarios. The early semiconductor laser cavity based on the Fabry- Pérot etalon was naturally formed by the crystal cleavage plane. In the 1970s and 1980s, the distributed feedback laser and distributed Bragg reflection (DBR) laser with distributed feedback mechanisms were developed. These developments had a decisive influence on reducing the laser threshold, improving the monochromaticity, and increasing the modulation speed, and such developments played a critical role in the use of semiconductor lasers in the field of optical communications. The vertical cavity surface emitting laser based on the DBR structure appeared in the 1980s, followed by various microcavity concepts in the 1990s, and subsequently photonic crystal lasers. In the 21st century, the development of ever smaller lasers has led to many novel nanoscale laser designs. The typical feature of these lasers is that, in at least one dimension, the size is on the order of submicron or much shorter, representing the dawn of the nanolaser age. Nanolasers are primarily divided into two categories. One category is represented by lasers based on various nanomaterials and nanostructures, such as nanowires, nanobelts, and nanofilms (Figs. 1 and Figs. 2). In 2001, Yang's research group realized an ultraviolet laser based on ZnO nanowires at room temperature for the first time. The other category is lasers based on a plasmonic mode at the metal-dielectric interface (Figs. 4 and Figs. 5). Plasmonic devices use free electron oscillations on the metal surface to enhance light confinement, which allows the size of laser to break the diffraction limit. In 2009, three teams independently demonstrated the first plasmonic nanolasers, or spasers, with different structures based on a surface-plasmon polariton mode or a localized surface-plasmon mode.

Conclusions and Prospect Lasers with ever decreasing sizes, i.e., down to nanoscales, or nanolasers, have evolved rapidly in recent years. We briefly describe the historical background of nanolasers including various types, their basic features, possible applications, current status, existing problems, and future trends. The types of nano-cavities include nanowire cavities, whispery-gallery mode cavities, Fabry-Pérot cavities, as well as surface-plasmon polariton cavities. The types of gain media include conventional compound semiconductors as well as newly emerging materials, such as perovskites and transition-metal dichalcogenides.

Significance High-energy rep-rated nanosecond diode-pumped solid-state lasers (DPSSLs), mainly referring to nanosecond lasers with pulse energies greater than 10 J and repetition rates greater than 10 Hz, are crucial in major fundamental and applied research domains and are emerging as one of the hot topics at the frontier of scientific research. This study first analyzes the preferred technical paths of high-energy rep-rated nanosecond DPSSLs in terms of the gain medium and amplifier geometry and then reviews the representative achievements and research progress of high-energy rep-rated nanosecond DPSSLs in detail. Furthermore, the prospects of future development of DPSSLs are discussed herein.

Progress Favored for its moderate saturation fluence and high thermal conductivity, Yb∶YAG at the cryogenic temperature and Nd∶LuAG at room temperature have been proven to be the most promising gain media in achieving rep-rated nanosecond DPSSLs with even higher energy. Conversely, gain medium with high saturation fluence at room temperature, such as Yb∶YAG, and that with low saturation fluence, such as Nd∶YAG, is not suitable for high energy lasers primarily owing to the defect of high pump threshold from the quasi-three-level structure and high passive loss from too many stages. In addition, the three preferred amplifier geometries are the multislab, active mirror, and zigzag slab ( Fig. 1), as categorized by the representative achievements of high energy rep-rated nanosecond DPSSLs summarized in Table 2.

For the multislab geometry, the Mercury system developed by Lawrence Livermore National Laboratory (LLNL) produced a nanosecond output with the pulse energy of 61 J at the repetition rate of 10 Hz based on Yb∶S-FAP multislabs, with ultralow wavefront aberration using a new high-speed gas cooling technology at room temperature (Fig. 2), a classical approach that was then widely used. Using similar gas cooling technology but operating at cryogenic temperature, the DiPOLE system based on Yb∶YAG ceramic achieved 105 J, 10 Hz, and 10 ns in 2017, which was the world’s first demonstration of a kW-level high energy DPSSL (Fig. 4). Researchers from STFC Rutherford Appleton Laboratory and HiLASE solved scientific and engineering problems in efficiency optimization, thermal effect management, depumping suppression, and other aspects. In the same year, LLNL reported the output level at 97 J, 3.3 Hz of a nanosecond Nd∶glass multislab laser for pumping the petawatt-level HAPLS system, using high-power intelligent laser diode system (HILADS), the highest peak power and brightest pulsed diode light delivery system in the world (Fig. 8).

For the active mirror mode, the LUCIA system reached 13.9 J at 2 Hz in 2013, using the Yb∶YAG laser head at room temperature at the pump intensity of 11 kW/cm² by focusing on the mirror (Fig. 12). To improve the energy above the 30 J level, the researchers invented a static helium gas cooling technology and plan to use a cryogenically cooled cosintered Yb∶ YAG ceramic, which may suffer from much stronger thermal lensing and higher depolarization losses than the crystal counterpart. Total-reflection active-mirror (TRAM) and multi-TRAM structures have been proposed by researchers at Osaka University, which achieved 1 J, 100 Hz laser amplification in 2015, despite unstable operation. Later this year, they released a new configuration of the conductive-cooled active-mirror amplifier (CcAMA) and reported the 9.3 J, 33.3 Hz laser scaling, suppressing the wavefront distortion by an elaborate heat sink design (Fig. 16). In 2016, our group at Tsinghua University demonstrated excellent compatibility of the Nd∶YAG seeder and Nd∶LuAG booster (Fig. 17), and then proposed a new concept called distributed active mirror amplifier chain (DAMAC) to disperse the gain and thermal deposition among several gain modules, thus achieving in 2019 a room temperature 10.3 J, 10 Hz, 10 ns laser from a large-aperture Nd∶YAG-Nd∶LuAG active mirror hybrid chain (Fig. 18), its output was recently raised to 100 J, 10 Hz at room temperature.

For the zigzag slab design, Hamamatsu developed the HALNA system, which demonstrated an output of 21.3 J, 10 Hz, 8.9 ns in 2008, with an optical-optical efficiency of 11.7%. The beam quality was controlled as 1.8 times diffraction limit, combining a thermally edge-controlled zigzag slab (TECS) design (Fig. 22) and a stimulated Brillouin scattering (SBS) mirror. In addition, the Chinese Academy of Sciences built an Nd∶YAG system in 2017 (Fig. 23), which generated pulse energy of 5 J at 1064 nm with a pulse duration of 6.6 ns and a repetition rate of 200 Hz, while the output energy stability was 4.9% peak-to-valley over 6000 shots. It was verified that the beam quality could be improved to 1.7 times the diffraction limit by an SBS mirror or by a deformable mirror.

Conclusions and Prospects Over the past two decades, extensive efforts have been made into achieving the first milestone, that is, the output target of 100 J, 10 Hz, and 10 ns, which has been achieved in the development of high energy rep-rated nanosecond DPSSLs with breakthroughs in both cryogenic and room temperature. In the next two decades, as new geometry, new gain medium, and new technical approach will inevitably emerge, the main trend expected will be the continuous upgrade in beamlet pulse energy (beyond kJ level), repetition rate (hundreds to kilohertz), and plug efficiency (over 20%), whereas potential directions of development may include system miniaturization, open and flexible access to other operating mechanisms, such as chirped pulse amplification, and programmable control over temporal, spatial, and frequency tuning.

Significance Ytterbium-doped fiber lasers have gained rapid development in the past two decades. Since the breakthrough of hundred-watt-level power at the end of the last century, nowadays the output power of the fiber lasers can exceed ten-kilowatt level. Thanks to the waveguide nature of optical fibers and the laser diode pumping techniques, high-power ytterbium-doped fiber laser owns several advantageous features, including high conversion efficiency, high beam brightness, compact and flexible architecture, easy thermal management, and stable operation. High-power ytterbium-doped fiber lasers have become preferred laser sources for many applications in various fields, such as industrial manufacturing, biomedical treatment, scientific researches, and defense.

Technologies for improving the output characteristics of high-power ytterbium-doped fiber lasers, especially for scaling output power, realizing more control of the output spectrum and expanding the range of output wavelengths, have received considerable interests from the fields. The characteristics are influenced by pumping schemes of the lasers, of which there are two major kinds as direct pumping and tandem-pumped. Direct pumping generally means that fiber lasers are pumped by laser diodes, which usually emit at around 915 nm and 976 nm for ytterbium-doped fibers (YDFs). In contrast, tandem-pumped means that fiber lasers are pumped by other fiber lasers, of which the wavelength generally ranges from 1000 nm to 1030 nm. Both pumping schemes have demonstrated high capability for realizing high-performance ytterbium-doped fiber lasers, and have realized outstanding advances in increasing power and brightness, controlling output spectrum, and expanding the available range of output wavelength. Particularly, tandem-pumped scheme has achieved increasing performance from recent studies. A few 20 kW fiber lasers have been proposed in the last several years.

Owing to the rapid development of light sources such as laser diodes, direct pumping has been the majority of the adopted pumping schemes. However, for higher power levels, namely 10 kW level, direct pumping may be a bottleneck to ensuring safe operation. Direct pumping offers usually low pump brightness [typically <0.2 W/(μm²·sr)]. For gaining a higher absorption, the available pump wavelength will be very limited, as it needs to be at around the absorption peak of the gain media (for YDFs, ~976 nm). This causes a large quantum defect in the pump-to-laser conversion process and thus severe heat management problem; for higher-power fiber lasers, the problem can be too hard to controll. In contrast, in tandem-pumped scheme where pump brightness can be enhanced by 3 orders to even >1000 W/(μm²·sr), it is possible to use double cladding YDFs of much smaller cladding diameters while having a good pump-gain overlap. In this way, it is possible to use longer pump wavelengths while having good pump absorption, and the heat problem can be mitigated. In recent years, an increasing number of high-power fiber lasers at from 1000 nm to 1030 nm has been proposed and demonstrated. There is an important landmark, as in 2009 when the IPG company for the first time announced their 10 kW fiber laser that was pumped by combined 47 fiber lasers at 1018 nm. The news has pushed the studies of fiber lasers for tandem-pumped (including 1018 nm fiber lasers) to the fast lane.

Progress This paper reviews the latest research progress about tandem-pumped high-power ytterbium-doped fiber lasers. We discuss the key technologies in realizing the tandem-pumped high-power fiber lasers with leading performances, and look forward to possible directions and challenges in future studies. In the second chapter, we review recent development of high-power 1018 nm fiber lasers. The features and performance of the lasers are introduced (Table 1, Fig. 1—3); the main problem, amplified spontaneous emission (ASE), that hinders further power scaling and several promising methods to mitigate it are discussed. In the third chapter, the recent development of tandem-pumped high-power ytterbium-doped fiber lasers working at traditional wavelengths, which have scaled the output powers, are introduced (Fig. 4—7). The fourth chapter discusses recent results of tandem-pumped high-power random fiber lasers, which can offer more stable spectral characteristics (Table 2, Fig. 8—9). The fifth chapter discusses tandem-pumped high-power Raman fiber lasers, which expands the range of output wavelength of ytterbium-doped fiber lasers to around 1.2 μm (Table 3, Fig. 10—11).

Conclusions and Prospects The advances in high-power fiber lasers continues to be driven by novel concepts and innovative techniques. Studies on high-power 1018 nm fiber lasers and the various tandem-pumped high-power fiber lasers have contributed a lot of useful solutions for the future development of many fields. In scaling output power, the application of the high-power 1018 nm fiber lasers in tandem-pumped high-power fiber lasers has been making excellent results. However, development of the high-power 1018 nm fiber lasers, as well as the others for tandem-pumped, is the key for further increasing the final performance. In controlling output spectral characteristics, tandem-pumped high-power random fiber lasers have shown promising effects in stable time-domain characteristics and can be used as seed in main oscillator power amplifier (MOPA) configuration for controlling the broadening of laser bandwidth. In expanding the range of output wavelength, tandem-pumped high-power Raman fiber lasers have exhibited outstanding performance in realizing high-power output at up to 1.2 μm. Meanwhile, it is further possible to combine the merits of erbium-ytterbium-co-doped fibers to reach the output of >1.5 μm. The future of tandem-pumped high-power lasers and the fiber lasers for tandem-pumped is bright and awaiting further studies.

Significance Optical fibers are appealing objects that are both fragile and powerful. However, they are micro-meter-class waveguides that are easily fractured by most external forces. Alternatively, another fact that is surprisingly unnoticed is that they are the core make-up of our society. For example, tons of information at every moment are loaded in the laser blood transmitting in the blood vessels of optical fibers and are used freely and timely worldwide. Many objects are processed, welded, and machined by lasers, scaled up to bulk materials in buildings and vehicles and down to micro- and nano-electronics. Optical fibers and fiber lasers are essential in various industrial areas, including industrial manufacturing, biomedical sensing, smart wearables, or even quantum-encrypted communications. Currently, the smooth running of day-to-day activities follows the safe, stable, and reliable operation of optical fiber systems. Therefore, potential threats to the operation of optical fiber systems cannot be treated with insufficient care.

A fiber fuse is a chain damage effect that propagates in optical fibers transmitting light. It was first reported in 1987 by Raman Kashyap, who at that time worked at a laboratory of British Telecom. Since then, fiber fuse damage effects have been observed in almost all types of optical fibers made from a variety of materials, including silica and organic polymers. It resembles a burning fuse emitting bright light from a moving spot; it happens spontaneously provided suitable conditions and causes irreversible damage to online fiber components it passes through in the inverse direction of the laser light. Therefore, it imposes a serious threat to many important technologies and applications nowadays, such as fiber communication networks and high-power fiber lasers. Studying mechanisms and characteristics of fiber fuse are not only important for controlling the hazards but also beneficial for realizing novel and effective methods for modifying fiber waveguide structures with more intimate aid from the laser inside.

Progress Studies discussing fiber fuse, ranging from former to the recent ones, are reviewed herein. Following authors' experiences with respect to the fiber fuse over the past decade, this study provides an introductory and to-date knowledge on physical mechanisms of fiber fuse, prevention of fiber fuse, monitoring or mitigating its propagation, and applications demonstrated using fiber fuse itself. Thus, directions for future research and major existing problems are also discussed. Regarding the characteristics of fiber fuse, the black-body assumption for obtaining the temperature of fiber fuse can be less rigorous because of the significant omission of other important mechanisms of radiation. Also, oxygen is formed during the fiber fuse and has been left inside the in-fiber bubbles. This implies that physical models that describe fiber fuse should focus more on chemical changes in the materials and their effects on other characteristics of the ongoing phenomenon. The critical temperature and the critical laser power conditions correlated with the initiation of fiber fuse, wherein the mathematical derivation directly suggests that the initiation of fiber fuse is dominated by a chemical process of formation energy around 1 eV, depending on the respective type of optical fiber, which may be attributed to the oxygen diffusion in the silica substances of the fibers. Furthermore, applications of fiber fuse have been developed to an extent wherein highly sensitive fiber sensors of various parameters are made using optical fibers damaged by fiber fuses as raw materials. Moreover, a possible method that uses controlled initiation of fiber fuse without propagating fiber fuse as an effective noninvasive one-step method to fabricate in-fiber microcavities, which are both highly cost-effective and hundreds of times faster than conventional methods, is proposed herein.

Conclusions and Prospects Studies discussing the fiber fuses in the past three decades have yielded many important and useful findings. With such considerable knowledge on the external characteristics of the damaging effect, however, there is room for deeper and further studies. Regarding the propagation characteristics of fiber fuse, the propagation velocity of fiber fuse increases with increasing laser power in fibers but the acceleration rate decreases. Nevertheless, in kilowatt-level high-power fiber lasers, the propagation velocity can be tens of meters per second, which hinders safe operation. Regarding physical mechanisms, oxygen is formed during fiber fuses; the initiation of fiber fuse is dominated by the diffusion of oxygen, which caused critical temperature and power conditions for the initiation; the fiber materials during fiber fuse can be plasma state. A more inclusive and comprehensive physical model for revealing more details and hidden characteristics of fiber fuse is necessary. Future studies on fiber fuses could be extremely beneficial. Groundbreaking changes can root from studies of physical mechanisms. If more physics of fiber fuse is revealed with rigorous theoretical and experimental proof, a wider connection among parameters of design can be built for fiber systems and the stochastic spontaneous initiation of fiber fuse can be avoided with assurance, which will profoundly secure many fields. Moreover, this will benefit the direct application of fiber fuse as a material modification tool that can bring potentially various new in-fiber microstructures into reality.

Significance Since the invention of the fiber laser is more than half a century ago, the performance index of fiber beam output has continuously improved, which is mainly reflected in the improvement of output pulse power, shortening of output pulse width, improvement of beam quality, and shaping of the laser spectrum. Femtosecond fiber laser, as an important branch of the fiber laser field, has been widely used in the industry, mainly because of its advantages over solid-state femtosecond lasers in the following three aspects.1) All-optical slimming and high optical efficiency. Because of the pump light in the optical fiber waveguide transmission, the effective action distance is longer, which improves the conversion efficiency. Additionally, the fiber has the advantages of soft and easy-to-coil integration, so it can be connected through the welding process.2) Close to the diffraction limit of the beam quality and good heat dissipation. As the signal light is confined to the optical fiber waveguide in the single-mode form, the output from the photonic crystal fiber with a large mode field diameter and chirality coupling optical fiber can help maintain a good speckled pattern. Additionally, the large cooling area of the heating effect of the solid gain medium can help overcome the inherent effect. It is an important way to realize high-power laser.3) The ion emission spectrum in the fiber is wider, so that the fiber laser can achieve the output pulse with a narrower pulse width and a wider spectral width. Thanks to the ultra-short pulse width, ultra-wide spectrum, and ultra-high peak power of the fiber laser, it has wide application prospects in industrial processing, communication detection, biological medicine, high-energy physics, material preparation, and chemistry, and other fields.

High energy, narrow pulse width, high stability of ultra-short pulses are also being researched throughout the optical fiber mode-locked oscillator development. Therefore, new principles, new technologies, and a variety of new laser materials have been revealed at historic moments, from active to passive mode-locking, from small energy width with a broad pulse width to high energy with a narrow pulse width, from a space structure to an all-fiber structure, from theoretical research to practical applications. In the past several decades, the fiber mode-locked oscillator has undergone rapid and comprehensive development, which has greatly promoted the development of new research fields and industrial markets. Now, the fiber mode-locked oscillator has become one of the core members of the laser family.

Progress Development of a fiber mode-locked oscillator revolves mainly around two themes. One theme involves a narrow pulse width and large energy in two main directions. One direction is based on theoretical study and is given priority to explore a new type of mode-locking mechanism. In this direction, it has experienced the traditional soliton, dispersion-management soliton, self-similar soliton, dissipative soliton resonance, and Mamyshev mode-locked soliton. The other direction is based on the use of new laser materials, including a new optical fiber, pump source, saturable absorption materials, and so on. To realize the all-optical fiber integration of the oscillator, adopting a special fiber structure is generally not suitable and realizing self-starting using the Mamyshev mode-locking is difficult; therefore, the nonlinear polarization evolution (NPE)-based all normal dispersion mode-locked oscillator is still the mainstream trend. NPE mode-locking can achieve a high single-pulse energy and a narrow pulse width. However, a polarization controller is necessary to adjust the cavity polarization state. Introducing a mechanical structure makes this type of oscillator sensitive to the environment, and ensuring the long-term stable mode-locking state is difficult. The mode-locked laser based on nonlinear amplified loop mirror (NALM) is the best choice for all-fiber integration. Such an oscillator can be used for all polarization-maintaining devices, which effectively avoid mode-locked dependence on the mechanical structure and have high environmental stability and reliability. Additionally, its output is linearly polarized—used as a CPA system of the seed source—and it can achieve high compression efficiency using a grating pair. However, when the femtosecond pulse is directly output by the oscillator, the dispersion management in the cavity becomes critical, especially in the 1-μm band. Because of a lack of negative dispersion devices, construction of such lasers becomes more difficult. Recently, the microfiber has injected new vitality in femtosecond fiber lasers. Therefore, we will review the development of these two types of mode-locked fiber oscillators herein.

Conclusion and Prospect Mode-locked fiber lasers are gradually becoming a powerful tool in many fields. High-energy and highly stable oscillators based on all-fiber structures are more attractive; among these, NALM and microfiber laser have undergone rapid development recently. However, the stability, reliability, and startup performance of these lasers need more detailed research to ensure their wider commercial use.

Significance With the increasing demand for the miniaturization and multifunctioning of micro/nanoelectronics and optical-electromechanical devices or systems, high-performance integration of materials at small scales is urgently needed. Nanojoining technology, which involves interconnection among low-dimension materials and even cross-dimensional nano-micro-macro materials, has shown great potential in manufacturing and developing micro-nano devices/systems.

Conventional welding/joining technologies such as ultrasonic welding, cold-pressure welding, and laser welding can be employed for the integration of diverse nanomaterials. However, to achieve high-performance and low-damage nanojoints, high-precision manipulation of input energy sources (e.g., thermal energy or laser beam) or mechanical tools (e.g., ultrasonic head or indent) is crucial in the nanojoining process, which poses great technical difficulties, thereby limiting the mass production of nanojoints and adaptability for multimaterial nanojoining. Ultrafast laser (UFL), with an extremely high pulse energy and short pulse duration, has been widely used in the precise manufacturing of a broad range of materials. Notably, a plasmonic effect, arising at the metal-dielectric interface, is a new approach to redistribute the incident optical (e.g., laser) energy with a “self-limited” effect, which is used for low-damage nanojoining. Therefore, nanojoining with UFL is promising for high-precision nanojoint formation, not only in homogeneous metal-metal structures but also in heterogeneous metal-oxide-semiconductor combinations.

Progress Local energy input within nanostructures can be redistributed because of incident laser-induced plasmonic effects (Figs.1--4). By constructing a metal-dielectric structure and simultaneously adjusting incident-laser parameters, spatial energy can be confined at local positions without the need to accurately control the spot size and spatial location of the incident laser beams. The self-limited effect of input energy contributes to the highly efficient formation of nanojoints with low damage, which permits the fabrication of complex micro-nano structures.

The interface structure of a nanojoint determines the electronic band diagram and thereby affects the electron transportation at the junction (Figs. 5 and 6). As indicated, a highly defective region with a large number of oxygen vacancies will be formed at a metal-oxide heterojunction after UFL nanojoining, which can reduce the Schottky barrier, thereby facilitating the origin and conduction of electrons at the junction. Based on this, high-performance heterojunctions with a variety of material combinations have been successfully achieved using UFL nanojoining technology. By regulating the interface structures, our team has developed several electronic nanodevices, which shows that UFL is promising for high-precise nanodevice fabrication and performance modification.

In addition, pulsed laser deposition (PLD) based on UFL developed by our group provides an alternative approach for low-temperature nanojoining (Figs.7--11). Nanoparticles fabricated via PLD possess high surface energy and fast diffusivity, which will be combined to form a connection layer composed of submicron particles. Therefore, the melting point will be close to that of bulk silver, which could be used for “low-temperature nanojoining and high-temperature service”. These UFL-deposited metal/alloy nanoparticle films have been successfully used for large-scale packaging of SiC wafers, which is expected to be widely used in the packaging of third-generation semiconductor devices that require high-temperature service and high reliability.

Accordingly, functional micro-nano devices can be fabricated based on the UFL nanojoining technology (Figs.12--19). We have developed several micro-nano devices by nanojoining nanomaterials at diverse scales, which will be used in devices such as optical waveguides, multilevel memristors, rectifier units, field-effect transistors, p-n junction units, supercapacitors, and flexible pressure sensors. This implies that the UFL nanojoining is promising for constructing basic micro-nano device units and is conducive to realizing multifunction device miniaturization.

Conclusion and Prospect UFL nanojoining is a vital way for micro-nano device fabrication. The local energy distribution within nanostructures can be well controlled, and the energy band diagram of the heterogeneous interface can be modified by interface metallurgy. In addition, low-temperature nanojoining can be realized using PLD based on UFL. The UFL nanojoining technology can be used to fabricate many micro-nano devices and has been proven to achieve excellent performance. Further research in the aspects of accurate positioning of three-dimensional structures, uniform fabrication of micro-nano patterns, common mechanism and quality control of heterogeneous interface formation, mass production, reliable performance evaluation, special equipment for nanojoining, and engineering applications will enrich and perfect nanojoining theory and technology, thereby permitting a broad range of applications in advanced micro-nano fabrications in various fields.

Significance Numerous multi-scale surfaces are available in nature with special micro-nano structures, such as lotus leaves, rice leaves, rose petals, gecko toes, shark skin, butterfly wings, and insect compound eyes. Scientists confirm that, among these multi-scale structures, the micro-scale structures function to strengthen the mechanical stability for protecting the nanostructures, while the nanoscale structures exhibit 13 magical functions, including superhydrophobicity, superhydrophilicity, directional wetting, self-cleaning, drag reduction, reversible adhesion, directional adhesion, anti-reflection, structural color, high sensitivity, selective filtration, cyto-biocompatibility, and the regulation of cell behaviors. How to artificially fabricate these bionic multi-scale structures to achieve the goal of imitating and surpassing nature is a major topic in the fields of materials and manufacturing.

An ultrafast laser is a pulsed laser with a pulse duration ranging from tens of femtoseconds to 10 picoseconds. Owing to its extremely short pulse duration, ultrafast lasers have high instantaneous energy density and high pulse repetition frequency. This enables its photon energy to interact directly with the internal lattice and electronic structures of materials in a considerably short time (the magnitude of pulse width) through various phase transformation mechanisms, such as phase and Coulomb explosions. Therefore, ultrafast lasers can process materials rapidly and accurately. These lasers can significantly reduce the heat-affected zone in the ablated area and then achieve extremely high accuracy and resolution. Thus, such a laser is a reliable tool for fabricating various micro-nano structures with high flexibility. However, owing to the diffraction limit, its fabrication capability for nanostructures is far more restricted, along with its lower fabrication efficiency.

Progress For the past decade, the Laser Materials Processing Research Center, Tsinghua University, has been equipped with a new generation of high-power ultrafast lasers (a pulse duration of 400 fs--10 ps, a repetition frequency of 100 kHz--2 MHz, and an average power of 40--100 W), supported by various projects, such as the National Key Research and Development (R & D) Program of China, the 973 project, the National Natural Science Foundation of China, the major international cooperation projects, and the Tsinghua University Initiative Scientific Research Program. Based on the suitable equipment and projects, a research team led by Prof. Zhong Minlin has performed a systematic investigation to expand the ultrafast laser fabrication ability for micro and nano structures and explore the functionalization of these fabricated bionic micro-nano structures. This team developed a series of novel approaches for micro-nano structure fabrication and dual-scale precise modulation through ultrafast lasers. Moreover, the team discovered several innovative applications of micro-nano structured surfaces in relation to the aspects of superhydrophobicity, high anti-reflection, high sensitivity, and biomedical detection.

Specifically, they achieved an efficient fabrication of large-area superhydrophobic metal surfaces with a higher contact angle and lower rolling angle than those of lotus leaves (Fig. 2). The team fabricated novel surfaces comprising periodical micro-pillar arrays that were covered by dense nanograsses and dispersed microflowers, exhibiting the highest available Cassie state stability and lowest ice adhesion strength compared to the state-of-the-art superhydrophobic surfaces (Fig. 4). Inspired by cacti, beetles, and redbud leaves, they developed a patterned superhydrophobic/superhydrophilic surface using an ultra-contrasting wettability venation network with hierarchical micro-nano structures as the skeleton for the massive water collection with high efficiency (Fig. 6). The team proposed a unique oil-triggered surface (OTS) by combining the “lotus-leaf-like” superhydrophobicity, the “nepenthes-like” slippery liquid-infused surface hydrophobicity, and the “re-entrant structure-induced” superamphiphobicity to achieve a high throughput manipulation of droplets, avoiding pinning, droplet loss, and cross-contamination (Fig. 8). They fabricated two kinds of surfaces: micropillar arrays accompanying nanowires and microcolumns covered by nanoparticles to obtain ultra-low infrared reflectivity and visible reflectivity, respectively (Figs. 10 and 12). Further, the research team explored a new field of micro/nano-textured electrodes for efficient hydrogen and oxygen production through water splitting (Figs. 15--17). They produced a novel patterned surface-enhanced Raman scattering (SERS) platform consisting of a superhydrophilic central area surrounded by superhydrophobic structures with an extremely high Cassie-Baxter state stability. Based on the evaporation enrichment of water droplets on such a platform, they achieved atto-molar SERS detection (Fig. 20). Furthermore, the team addressed challenging issues in SERS research and applications, such as stability, uniformity, and manufacturability, thus to expand SERS applications to cancer diagnosis and food safety evaluation. Based on the above works, the research team has published more than 100 papers, granted over 20 patents, and developed more than 50 know-hows. More importantly, the technologies achieved significant applications in various critical fields.

Conclusions and Prospects Fabricating various micro-nano structures by ultrafast laser and realizing biomimetic functionalization is an attractive research area. However, challenges still remain, such as the fabrication of typical nanostructures with a size of 1--100 nm via the breaking of the diffraction limit, the design of novel diverse functional micro-nano structures and their free fabrication, and the efficient fabrication of large-area micro-nano structures. Facing these challenges, the authors summarize their research results in the past decade and select four representative research areas: the controllable fabrication of micro-nano structures for special wettings, the two-stage tuning of micro-nano structures for high anti-reflection, the nanostructures for electrocatalytic water splitting, and the laser-induced patterned surfaces for SERS. This paper is written for the special issue of the Journal of Chinese Lasers—Celebration of the 110th Anniversary of Tsinghua University—to summarize the past and address the future as well as to exchange, discuss, and promote R & D in the field. All comments are welcome.

Significance Graphene is a two-dimensional carbon crystal with a single atomic layer, which was first discovered using the “Scotch tape method” in 2004. Compared with other carbon materials, graphene has shown unique physical and chemical properties, such as high carrier mobility (150000 cm ²·V ^-1·s ^-1), high thermal conductivity, high Young's modulus, optical transmittance, flexibility, and biocompatibility. Because of these properties, graphene has become a common material in electronic, energy storage, optical, micromachining, and biological devices.

The unique physical and chemical properties of graphene have stimulated the rapid development of graphene preparation technology. Researchers have developed a mature system of graphene preparation methods, such as mechanical/chemical exfoliation of graphite, silicon carbide epitaxial growth method, and chemical vapor deposition (CVD). According to the different applications of graphene, there are various advantages and disadvantages of different preparation methods. The mechanically exfoliated graphene has a better crystal structure, but lower preparation efficiency; the CVD method is effective to prepare high-quality graphene, but transferring graphene from metal catalytic substrates to target substrates is necessary for electronic devices; reduced graphene oxide (RGO) can be prepared in batch, but there exist several defects. The different properties of graphene are determined by the different preparation methods so that the application fields are not the same. No matter what device it is applied to, the processing of graphene is the primary technical problem to be solved. Thus, the device-based processing of graphene is a critical challenge for its practical applications.

Laser processing technology has numerous advantages such as fast processing speed, controllable processing procedure, unaffected by harsh reaction conditions, high precision and flexibility, friendly environment, and large-scale preparation. Therefore, it has become a simple and effective method for the preparation, modification, and device processing of graphene materials. Using laser processing technology, we can not only control the heteroatom species and concentration of graphene but can also induce its carbonization on polymeric substrates. In addition, the ability to pattern and structure without a mask also facilitates the development of graphene devices. Laser processing of graphene has been commonly used in optoelectronic devices, energy storage components, sensors, microelectronics, and other fields.

Recently, significant progress has been made in the development of laser-processed graphene-based sensors and actuators. As a two-dimensional material, each carbon atom of graphene is a surface atom, making it a unique sensitive material. Laser processing technology can not only achieve the patterning of graphene but can also effectively control its surface structure and heteroatom content, which is an effective means for preparing graphene-based sensors. In terms of actuators, graphene is sensitive to the stimulation of optical, electrical, and chemical fields because of its unique optical, electrical, and mechanical properties. Therefore, the actuators based on laser-processed graphene have been successfully developed and can be driven and controlled by a variety of physical and chemical fields.

In this paper, we reviewed the recent progress of laser-processed graphene and graphene derivatives, for example, graphene oxide (GO), for developing sensors and actuators. The topic is significant because laser processing of graphene and its derivatives hold great potential for developing carbon-based micro-electromechanical system (MEMS) sensors and actuators that are important for future smart devices.

Progress Recently, laser processing technologies have promoted the rapid progress of graphene-based devices, especially sensors and actuators. First, laser processing enables the preparation, modification, patterning, and structuring of graphene, revealing the great potential for manufacturing graphene-based electronic devices. Subsequently, typical laser processing strategies for graphene preparation, including laser reduction of GOs, laser-induced graphene (LIG), and laser-processed CVD graphene have been reviewed.

Second, using laser-processed graphene, sensors capable of detecting stress, pressure, chemicals, and biomolecules have been successfully developed. When developing graphene-based sensors, laser-processed graphene can act as electrode, semiconductor, dielectric, and host material. Laser processing facilitates device fabrication because it permits flexible and mask-free patterning, hierarchical structuring, and heteroatom doping. As a typical work, Ren's research group developed an epidermal electronic skin based on laser-scribed graphene that can be used as a strain sensor. The patterned epidermal electronic skin can be used to detect the finger bending motion and pulse. In addition, human breath can also be detected when the sensor is attached to masks and throats (Fig.6).

Third, graphene and its derivatives are sensitive to external stimuli, and thus can be employed to develop actuators. By full utilizing the moisture responsiveness of GO, the photothermal property, and conductivity of graphene, stimuli-responsive actuators that can dynamically deform in response to external stimuli have been reported. Ma et al. reported a laser-reduced RGO/GO bilayer humidity actuator that exhibits a fast response to humidity (Fig. 7). Wang et al. developed a light-driven actuator based on the Marangoni effect using patterned LIG tape. The actuator can achieve translation and rotation motion under photothermal actuation, providing a broad prospect for the development of light-driven microrobots (Fig. 8).

Conclusions and Prospects Graphene is promising for developing novel electronic devices; however, the lack of graphene-processing technologies compatible with the device manufacturing techniques may limit its progress. Laser processing of graphene has become a solution to address this limitation. Within the past decade, rapid progress has been made in the laser preparation of graphene, and the resultant sensors and actuators have revealed great potential for developing carbon-based MEMS devices. However, challenges in the preparation of high-quality graphene (or GO) and low laser processing efficiency still exist. The situation will soon improve because graphene preparation methods and laser processing systems have developed rapidly in recent years. Laser-processed graphene is expected to find practical applications in the near future, especially for developing smart carbon-based devices and flexible e-skins.

Significance Due to the increasing energy crisis and environmental pollution problems, new energy technologies have been a research hotspot among scientists. Although several energy devices, such as supercapacitors, lithium-ion batteries, solar cells, new energy storage, and power generation devices, having been developed, low energy density, complex preparation process, monotonous structure, and poor mechanical properties have severely limited their application in practical scenarios. Traditional processing methods suffer from complex processes, difficulty in processing microdevices, and the inability to precisely regulate material properties, making it difficult to process high-performance energy devices. With high peak energy density, small heat-affected zone, wide material applicability, high spatial resolution, and customizability, laser processing is often used to increase the energy density of energy devices and integrate microdevices. Thus, it has great research value and application potential in the field of precision processing of advanced materials and devices.

Progress The ultra-high peak power density of the laser can produce strong interactions with the material in a localized area of action, which can finely modulate the microstructure of the electrode surface and significantly increase the energy density of the double-layer capacitor ( Fig. 1). For pseudocapacitor supercapacitors, laser treatment can significantly improve the energy storage capacity by doping active materials or heteroatoms into the electrode material or enhance the kinetics of redox reactions and improve cycling stability through the surface morphology modulation ( Fig. 2).

Lithium-ion batteries play a significant role in life and production. However, the slow reaction kinetic process limits the output power of lithium-ion batteries. The expansion of electrodes during charging and discharging affects the life and safety of lithium batteries. The pulsed laser deposition technology can finely regulate the microstructure of electrodes and develop composite materials, which is conducive to improving the working area of electrodes, reducing the ion transport distance, and increasing the electrical conductivity of electrons and ions; thereby, significantly increasing the energy density and output power of lithium-ion batteries (Fig. 3). Additionally, pulsed laser deposition technology can prepare ultra-thin, dense, and uniform films, which can significantly reduce interface resistance to ensure high ion conductivity and high output power (Fig. 4). Besides, the use of laser etching technology can conveniently control the microstructure of electrode materials or selectively construct composite materials. It can also alleviate the problems of poor cycle stability and low capacity retention caused by volume changes during charging and discharging (Fig. 5).

Laser processing also has important applications in constructing light-absorbing and interfacial layer materials for solar cells. Nanoscale-oriented structures can be prepared on the electrode surface using pulsed laser deposition techniques, which can prevent electron-hole complexation, reduce interfacial resistance, and improve the energy conversion efficiency (Fig. 6). Functional light-absorbing for solar cells can also be prepared using pulsed laser deposition technology. Compared with traditional methods, the grain size of the photoabsorption layer prepared using pulsed laser deposition technology is more uniform and dense. It can easily regulate the structure and band gap of the film, and the microstructure can easily be adjusted to achieve higher light-absorption efficiency (Fig. 7).

Due to the local interaction between the laser and graphene oxide, the direct laser writing technique is often used to prepare water-enabled electric generators. Direct laser writing technology can reduce irradiated graphene oxide regions to conductive reduced graphene oxide to serve as electrodes and interconnecting circuits. Thus, enabling easy in situ construction and integration of water-enabled electric generators on graphene oxide assemblies is realized. However, the thermal effect of laser processing can partially reduce the graphene oxide material. Thus, three-dimensional water-enabled electric generators can be prepared in a controlled manner to obtain higher output power (Fig. 8).

Laser processing has the characteristics of high spatial resolution and strong customization. It can be used to prepare electrode materials with high light transmittance and good flexibility. Simultaneously, the device can be designed into a specific structure to obtain ductility; thus, it has important applications in the field of flexible electronics (Fig. 9).

Conclusions and Prospects Laser processing has obvious advantages and potential in the field of high-performance energy devices. Currently, pulsed laser deposition technology, direct laser writing technology, laser induction technology, and laser etching technology have achieved many successful applications in material modification, device functionalization and miniaturization, electrode preparation, and structural modulation. Supercapacitors, lithium-ion batteries, new energy devices, and flexible electronics prepared using laser processing have significant advantages in terms of energy density, stability, and energy conversion efficiency miniaturization and integration. Although laser processing has made considerable progress in the field of energy devices, there are still many issues to be addressed. For example, the knowledge and understanding of the mechanism of laser-material interaction are still imperfect. It is also difficult to control the spatial thickness or form of structural defects using laser processing. Additionally, the current research on laser processing mainly focused on regulating precursor materials; its application in more complex scenarios needs to be further investigated. With a deeper understanding of laser action mechanisms and further development of laser precision control, it is expected that the applications of laser processing in the field of new energy devices will make breakthroughs.

Significance The sustained demand for the joining techniques with high quality, high efficiency, high flexibility, and environmental protection, as well as the emerging demand for joining various novel materials, jointly promote the research and applications of laser-welding techniques. Benefited from the rapid development of laser components and supporting equipment, as well as the understanding of the interaction mechanism between the laser and materials, laser welding has achieved the transformation from laboratory development to large-scale industrial applications in the past 30 years. The laser energy can be precisely controlled in time and space. High energy density and low heat input enable laser welding to control the weld shape accurately and restrain the adverse effect of the welding thermal cycle on the base metal. On the one hand, high energy density causes metal evaporation, thereby forming keyhole, which is a common defect in laser welding. On the other hand, low heat input leads to a fast cooling rate, affecting the microstructure of the weld and heat-affected zone. The control of penetration, porosity, and microstructure is closely related to the laser energy. By analyzing four typical cases, welding of steel, magnesium alloy, titanium alloy, and dissimilar material, this study introduces the influence of laser energy on the welding penetration, porosity, and microstructure and its control methods. Moreover, it reflects the characteristics and advantages of laser welding and provides a useful reference for solving the welding problems of new materials and structures using laser-welding techniques.

Progress Laser welding can reduce the heat input, obtain a high aspect ratio, and improve the welding efficiency and accuracy. It can also accurately control the temperature change of the base metal and obtain the desired melting shape and temperature distribution. The semi penetration laser welding of stainless-steel car body can avoid the burning traces of the outer plate and ensure joint strength ( Fig. 1). Using double beams, the hump of thin-walled stainless-steel welded by laser can be avoided, and the welding speed can be greatly improved ( Fig. 2). By adjusting the laser incidence position and heat input, the fusion ratio or heating temperature of dissimilar materials can be controlled, and the bad microstructure, holes, and cracks can be avoided ( Fig. 3).

The metallurgical pores of laser welding are mainly affected by the gas source. The impurity on the surface of a workpiece is removed using the method of laser surface cleaning before welding. This suppresses the pores of laser welding, improves welding efficiency and flexibility, and reduces environmental pollution. For materials with high gas content in the base metal, such as die-casting magnesium alloy, with high energy density and appropriate heat input, the precipitation of solid solution gas can be avoided, reducing the porosity (Fig. 5). The porosity process of laser welding is affected by the stability of the keyhole. Appropriate laser energy is essential to improve the stability of the keyhole and reduce the tendency of porosity. Due to the influence of gravity, the influence of the laser energy on the pore tendency under different welding positions differs; thus, it is necessary to select appropriate laser energy according to specific welding positions (Fig. 6).

Laser welded joints undergo heating, cooling, melting, and solidification processes, which lead to changes in structures, deteriorating performance. Laser welding is the ideal-welding method for ultrafine grain steel due to the low heat input, which has little influence on the entire welded joint. The rapid cooling rate of laser-welding results in a great difference between the microstructure of the welded joint and base metal. The laser surface heat treatment after welding can control the microstructure of the weld and heat-affected zone properly, providing more possibility for controlling the microstructure of the joints by adjusting the laser energy (Fig. 9).

Conclusions and Prospect The essence of laser welding is the interaction between laser energy and material. Penetration, porosity, and microstructure are key factors affecting the properties of laser welded joints. In this study, typical examples of laser welding of steel, magnesium alloy, titanium alloy, and dissimilar materials are selected. The influence of laser energy on the welding penetration, porosity, and microstructure and their control methods are summarized. For future development, the laser energy modulated by wavelength, pulse, and compound heat source, and other laser-processing methods combined with the pretreatment or posttreatment of welding provide more possibilities for controlling the melting shape, porosity, and microstructure of laser welded joints, improving the quality and efficiency of laser welding and expanding the applications of laser welding. From the application perspective, laser-welding techniques may have great development and breakthrough in microjoining, nanojoining, and joining materials with special properties or functions. Moreover, due to the rapid development of artificial intelligence, more techniques, such as big data and machine learning, will be used in laser welding in the future.

Objective Tungsten (W) has been widely applied in electronic and biological fields owing to its high melting point, excellent strength, and oxidation resistance. It has also been selected as the plasma-facing material for diverters in future fusion reactors due to its excellent radiation-shielding properties against heat and plasma fluxes. Recently, tungsten components with complex and fine structures have been required to face the increasing demand of material customization, which is difficult to manufacture through powder metallurgy because of the high hardness and intrinsic brittleness of tungsten. Laser power bed fusion (LPBF), as an important method of additive manufacturing technique, is a rapid forming technology based on 3D models. However, cracks would initiate and propagate due to steep temperature gradient and excessive thermal stress during rapid cooling of the LPBF process. Possible mitigation methods have achieved less crack density via preheating substrates, introducing secondary phases, and optimizing process parameters, with limited success in these cases. This study introduces tantalum carbide (TaC) into the tungsten matrix to explore a new method of reducing the crack density by forming submicron scale substructure strenthening phases. As a high melting point ceramic phase, TaC is well-matched with the lattice constant of tungsten. We reported optimized formability as well as achieving reduced cracks in LPBF tungsten while considering in-situ alloying. The tungsten caibide (W₂C) phase formation through in-situ reaction during LPBF process should be focused.

Methods The 5--25 μm tungsten powders are prepared by the electric explosion method of metal wires. The alloying powders of tungsten-5%TaC are prepared by spherical tungsten powders above and 5% spongy TaC powders with 5 μm median diameter and are mechanically mixed using a low-energy blender mixer. Pure tungsten and tungsten-5%TaC parts are fabricated using an SLM280 2.0 machine with appropriate process parameters. Samples were built on stainless substrates. The scanning mode is rotated by 67° between adjacent layers with a “Zig-Zag” scan strategy. To limit oxidation during the LPBF process, the oxygen content inside the chamber is kept at less than 5×10 ^-4(volume fraction). In the next step, the as-built samples are mechanically grounded and polished, followed by electropolishing in a 1.5% NaOH solution at a voltage of 9.8 V for seconds to reveal the morphology of true cracks. The microstructure of top and side view cross-sections and phase analysis are characterized using a scanning electron microscope and X-ray diffractometer (XRD) with Cu-Kα radiation. The microhardness of tungsten and tungsten-5%TaC parts is also measured.

Results and Discussions Tungsten-5%TaC parts with good surface morphology and formability are obtained with a laser power of 350--400 W and scanning speed of 300 mm/s [Fig. 2(a)]. Warping and interlayer cracking will occur with improper laser energy input [Fig. 2(b)]. With suitable laser energy input, the interaction of powders and laser can achieve complete melt spreading of tungsten-5%TaC. Compared with pure tungsten, the defect density of pores and mesoscopic cracks in the microstructure has been considerably reduced in tungsten-5%TaC system (Fig. 4). After adding TaC particles, the grains are refined and many small-angle interfaces are formed. Further, transgranular cracks and hot cracks are initiated (Fig. 5). According to XRD analysis, the tungsten carbide (W₂C) phase appears in the tungsten-5%TaC parts (Fig.6). Submicron substructures with various morphologies are observed in the tungsten-5%TaC microstructure (Fig. 7). During the LPBF ultra-high-speed solidification process, the substructure is formed when the solidification is near the absolute stability limit. The tantalum element exists in the matrix and cell walls, and the W₂C phase exists on most of the cell walls. The morphology of the W₂C phase may be closely related to the aggregation of tantalum and solidification conditions. These submicron scale substructure combined with good thermal conductivity and high strength of W₂C phase segregation at the interface can effectively improve the strength of the material. The meandering crack path at the cell boundary increases the resistance of cracking propagation to some extent. The solid solution of tantalum in the matrix can strengthen the matrix and improve the material's intragranular and grain boundary strength. After adding TaC, the microhardness of the microstructure is increased from 400 HV of pure tungsten to 666 HV (Fig. 9). The refinement of grains can increase crack propagation resistance and alleviate the stress concentration during the LPBF process.

Conclusions This study proposes a new method of strengthening the tungsten matrix and reducing cracks via in-situ reaction by forming submicron scale substructure strengthening phases. By adjusting the appropriate laser process parameters, tungsten-5%TaC samples fabricated by the LPBF technique have achieved good formability and pores in the microstructure have considerably reduced. Neither lower energy input (pores and spheroidization) nor higher energy input (overheated phenomenon) can produce samples with good formability. In the range of appropriate laser parameters, the tungsten-5%TaC system cracks are considerably reduced compared to pure tungsten cracks, which are mainly achieved via alloy strengthening, substructure strengthening, and grain refinement. The microhardness of tungsten-5%TaC samples is increased by 50% compared with pure tungsten, and the intrinsic strength of the material is also improved. Meanwhile, various morphologies of substructures are observed in the microstructure of tungsten-5%TaC, and there exists W₂C segregation in the cell walls. The evolution of substructures is mainly affected by melt flow, temperature gradient, and thermal history during LPBF solidification and cooling.

Significance There is no science without measurement. More accurate measurement of physical quantities is highly desired in modern science and technologies. Laser interferometric precision measurements have outstanding advantages, including traceability, nanometer or even picometer resolution, and ultralong measuring range up to several meters, kilometers, or even thousands of kilometers. It is widely used in advanced technologies and frontier research, such as IC devices, CNC machines, ultraprecision micromanufacturing, and gravitational wave detection.

However, laser interferometric precision measurements have many key problems that demand urgent solutions. The most crucial one is that the laser source requires to be independent, whereas the current domestic market cannot produce satisfactory dual-frequency lasers for heterodyne interferometry. Traditional laser sources have frequency differences lower than 3 MHz, which limits the maximum measuring speed. This severely restricts the processing efficiency of IC chips or machine tools. Furthermore, the output power of the widely used lasers is only 0.5 mW, which is low for further multidimensional measurements. More importantly, dual-frequency lasers exhibit a nonlinear error of several nanometers, thus affecting the precision of the interferometers. For example, the Agilent dual-frequency interferometer has a nonlinear error of 3 nm [Fig. 10(a)]. When it is used for collimation at the precise location of machine tools, this error is considered in the final precision evaluation. Another troublesome problem is that, to generate an interference signal with a high signal-to-noise ratio, traditional laser heterodyne interferometry requires a target mirror or highly reflective surface of the test object to reflect a sufficiently strong beam. However, in many cutting-edge science and ultraprecision applications, a target mirror cannot be placed on the test object and the measured surface is not highly reflective, such as flexible film deformable mirrors for laser fusion, thermal and gravitational deformation of space camera primary and secondary mirrors in low-temperature vacuum environments, and large-travel Abbe error calibration of machine tools and the like. Laser interferometry requires not only nanometer precision measurements but also to match the optical path, which has become a bottleneck in the field. Thus, it restricts the technological innovation and development of precise measurements.

Many scientific frontier studies, such as gravitational wave detection, lithography machine positioning, and interstellar exploration, require ultrahigh precision measurement technology. Since the advent of laser interference technology, it has been crucial in precision measurements, and the demand for accurate measurements will increase from micro-nano level to picometers, or even femtometers, in the future. Therefore, independently developing novel interferometers with better performance is required for domestic laser interferometry precision measurements. Furthermore, summarizing the characteristics and limitations of existing interferometers is crucial to guide future development in this field more rationally.

Progress Owing to the above application requirements and technical problems, we have been devoted to investigating laser and laser-feedback interference in the past several decades. We have recorded great breakthroughs in dual-frequency innovative lasers with large frequency difference and high-power maintenance and in feedback interference for nanometer measurements without target mirrors.

On one hand, a new laser source based on the principle of the Zeeman-birefringence dual-frequency has been developed and produced independently with a higher dual-frequency difference and output power (Fig. 13). The core technical parameters are comparable with or surpass similar advanced lasers worldwide. With this laser source, various interferometers for displacement, angle, linearity, and flatness measurement have been developed. Zeeman-birefringence dual-frequency laser interferometers in China are at the mass manufacturing and production level for the first time. On the other hand, to address the measurements without target mirrors or measuring a low-reflectivity target, we consider the basic principles of self-mixing interferometry and have successfully developed a laser feedback interferometer (Fig. 23). Due to the high sensitivity of the self-mixing modulation, the produced interferometer can achieve an ultrahigh gain amplification of the detected signals. Therefore, the feedback interferometer has a wide application range, including displacement and velocity measurements, vibration recovery, refractive index sensing, and biological imaging. The developed interferometers solve the key measuring problems in laser interferometry precision measurements. They have demonstrated notable performance in nanometer precision measurements; thus, they are employed in many frontier research and industrial applications.

Conclusions and Prospects Herein, we presented the latest achievements and research progress in the research team on dual-frequency laser measurement and feedback interferometry technology in the past decade. We also presented the prospect of laser interferometry in precision measurements. Based on the results obtained herein, we shall focus on innovation, seek breakthroughs in new measurement principles and methods, and continuously improve the performance of the developed interferometers, bringing breakthroughs to laser interference ultraprecision measurements and their applications.

Significance Distance measurement is a common basic technology in the field of geometric measurement and has broad applications in scientific research and industry. Currently, high-precision distance measurement is normally achieved using the interferometric method, and the distance results can be directly traced to the optical wavelength. However, the phase ambiguity hinders the application of the traditional interferometric method in long-distance absolute positioning, such as space missions, including tight formation-flying satellites, antenna measurement, spacecraft rendezvous and docking, as well as precision manufacturing and assembly, including aircraft manufacturing, satellite equipment manufacturing, and synthetic aperture optical system assembly. Fortunately, invention of the optical frequency comb (OFC) provides great opportunities for geometric measurements.

In recent years, several OFC-based methods have been proposed for the measurement of distances, e.g., the intermode beat, dispersive interferometry, pulse alignment, and dual-comb methods. Compared with conventional methods, OFC-based methods are capable of resolving the problem of phase ambiguity and measuring the absolute distance. Among them, the dual-comb ranging method makes full use of the characteristics of OFC in the time and frequency domains and exhibits reasonable dynamics, precision, and unambiguity range. The dual-comb method opens up a new direction for distance measurement and is expected to bring great benefits to optical metrology. Since 2009, many advances have been achieved in dual-comb ranging techniques. However, there are still several challenges involving principle research and industrial applications. Hence, it is necessary to summarize progress of the dual-comb ranging technique to guide future development in this field more rationally.

Progress The progress of the dual-comb ranging method is illustrated in Fig. 1. The concept of dual-comb was first proposed in 2002 by Schiller, and was first applied to absolute distance measurement in 2009 by Coddington et al from NIST. Recently, the proposed dual-comb ranging method primarily uses fiber combs as the light source and achieves absolute distance measurement based on two principles, namely, time-of-flight-based and phase-based ranging. Optimizing the parameter model and suppressing the frequency noise of a dual-comb system are two key technologies to achieve high-precision distance measurement. Here, the time-of-flight ranging method mainly focuses on the former, while achieving the phase-ranging method is based on the latter. Since 2013, domestic and foreign researchers have explored the performance improvement, function expansion, and applications of the dual-comb ranging method, including the dead-zone elimination of dual-comb ranging, nonambiguity range extension, improvement of ranging precision and response speed, dual-comb multi-degree-of-freedom sensing, and instrumentation and applications of the dual-comb ranging technique. Since 2018, research on the microresonator-based dual-comb ranging method has been reported. The microcomb has the characteristics of miniaturization and easy on-chip integration. Compared with the fiber comb, the microcomb offers a better application prospect in the field of ranging, and provides new opportunities for further developments of dual-comb ranging.

Conclusions and Prospects In summary, the dual-comb ranging technique provides an efficient tool for absolute distance measurement with a large unambiguity range, high precision, and high speed, and it has also become a hot spot in the field of ranging. Such an overall performance brings great benefits to various tasks in optical metrology. With the continuous in-depth and detailed explorations of dual-comb ranging, it is expected to become a portable instrument product widely used in scientific research and industry.

Significance The weak-measurement (WM) theory was proposed by Aharonov et al. in 1988. We can obtain the measurement value that is much larger than the eigenvalue using the WM theory by appropriately adjusting the pre-and post-selection states and maintaining a small interaction intensity between the system under test and the detector. The small interaction maintained here is what referred to as “weak” in the WM. In the process of weak interaction, an important parameter, named “weak value,” contained in the pointer state has always played an important role. Therefore, we call this process of significantly amplifying the actual parameters as the weak value amplification. However, because it is impossible to prove the existence of this “weak value” at the beginning, the WM theory has been questioned by the scientific community to the extent that some believed that the WM theory was an absurd idea. In 1989, Duck et al. re-explained the concept of WM, and then in 1991, Ritchie et al. verified the existence of weak values of key parameters in WM through experiments. Consequently, the WM theory is widely accepted.

Progress The WM theory provides a deeper explanation for quantum physics and shows potential for the precision measurement. However, in the next decade, WM development mainly revolved around the theoretical study, such as WM realization in some specific systems, related study on the pointer states in the WM systems, and the “weak value”. In this paper, we discussed important parameters, the WM techniques in classical and quantum physics, the prospect of applying WM techniques to the fields of high-precision measurement, and some other theoretical studies. After the first five years of the 21 ^st century, the WM techniques have gradually showed their own unique properties in the measurement field. In 2005 and 2007, Pryde and Jozsa, respectively, implemented the WM experiments in polarization detection and measured the complex weak value. They explained in detail the physical meaning of the real and imaginary parts of the weak value in the actual measurement. The WM has been improved to the stage of high-precision measurement. In 2008, Hosten et al. reported a related WM work that studies the spin Hall effect of observing light in Science, which refocused the spotlight on the WM, the promising techniques. Owing to the high measurement accuracy and potential, the amplification mechanism of the WM can be used to observe the physical phenomena and detect the physical parameters.

Furthermore, several related theories since 2010 have shown that the WM performance in the frequency domain has more obvious detection advantages than other fields. Particularly, in 2011, Li group from the University of Science and Technology, China, proposed that white light could be used to achieve high-sensitivity detection of optical phase in the WM system. Two years later, they verified the WM through experiments, which laid the foundation for the WM applications in frequency domain optics. In 2016, He Yonghong group used broadband high-brightness super-luminescent diode as a light source and realized an optical frequency domain WM system with a wide range of general values. Compared with the traditional optical interference detection, the WM system is 1--2 orders of magnitude higher in detection accuracy.

Currently, the WM techniques are widely implemented in four fields including the time domain, frequency domain, spatial domain, and polarization angle distribution based on the requirement of different applications. Relevant study results show that the WM technique has good applicability and high measurement accuracy in these four fields. The representative work in each field is as follows:

(1) Time-domain WM: single-photon tunneling time and observing the spin Hall effect of light.

(2) WM in frequency domain: subpulse width time delay, temperature measurement, and phase shift.

(3) WM in the spatial domain: ultrasensitive beam deflection measurement, Goos-H?nchen displacement, WM techniques to improve the SPR resolution.

(4) Weak-polarization angle measurement: polarization rotation, beam deflection angle measurement, light polarization measurement, chiral molecule measurement, and deviation of the beam angle in reflection or refraction.

Conclusions and Prospects In this paper, we introduce two types of measurement methods combining weak-value amplification based on practical applications of the weak-value amplification in high-precision measurement. These methods are used to measure changes in physical quantities by analyzing the lateral offset and frequency shift of the beam. Based on these systems, the weak-value amplification can be combined with many traditional measurement methods to improve the resolution of the system. Finally, we discuss the development trend of weak-value amplification. Combining the traditional detection methods and weak-value amplification techniques to achieve higher system resolution, indicators are a direct application of the WM in high-precision measurement. Because the weak-value amplification techniques can excellently suppress technical noise, the existing WM system has a resolution of ~1--2 orders of magnitude improvement compared with that of the traditional system. The WM techniques have potential applications in the fields of biology and chemistry.

Significance A series of new physical mechanisms and unique optoelectronic properties in nanostructures provide the possibility to explore and study new optoelectronic chips.

Progress Photons and phonons are important information carriers. Precise measurement of photons and phonons in micro-/nano- periodic structures allows the detection of multi-dimensional physical information and various physical quantities. Here, by manipulation of photons with metasurface units, a one-shot miniaturized ultraspectral camera with thousands of micro-spectrometers on a CMOS image sensor chip is proposed and applied to realize real-time on-chip spectral imaging. Micro-spectrometers with high center-wavelength accuracy of 0.1 nm and spectral resolution of 0.8 nm are realized and the ultraspectral imaging results for a plate of fruits just under the lighting of a fluorescent lamp are obtained. For phonons, opto-mechanical crystals are studied by dealing with the interaction between photons (light) and phonons (mechanical motion). A hetero opto-mechanical crystal is proposed and demonstrated by integrating two types of periodic structures into the system, and the optical and mechanical modes can be confined separately. This separate confinement gives rise to phonon lasing with a high mechanical frequency of 6.22 GHz. Moreover, radiation-pressure-antidamping enhanced opto-mechanical spring sensing based on a silicon nano-beam opto-mechanical structure is proposed and demonstrated, which allows for sensing resolution of ~10 ^-7.

Some interesting phenomena and novel devices are arising by free electrons interacting with various nano?structures. We demonstrate the first on?chip integrated free electron light source by greatly reducing electron energy to 0.25??1.4 keV for generating Cherenkov radiation (CR). In hexagonal boron nitride (hBN) with hyperbolic phonon polaritons, the theoretical and simulated results reveal that CR can be generated using free electrons with an extremely low kinetic energy of 1 eV, which is about two?orders of magnitude lower than that in multilayer plasmonic hyperbolic metamaterial. For generating Smith?Purcell radiation (SPR) in the deep UV region, we let an electron beam pass through a grating with 30 nm?wide slots and observe the SPR with the shortest wavelength of ~230 nm and the broadband SPR with wavelengths covering 230??1100 nm. We numerically investigate the SPASER excited by free electrons, and the tunable, deep?ultraviolet laser with output power density reaching about 30 W/μm2 and wavelength widely tuned by varying the electron energy. Our work opens up the possibility of exploring high performance on?chip integrated free electron light sources and optoelectronic devices, and provides a way for realizing an integrated free electron laser.

Introduced by Allen in 1992, orbital angular momentum (OAM) was characterized as a new freedom of lightwave. Since then, it has been attracted much research interest and shown the potential for various applications. Compared with the bulk optics, photonic integrated devices are much more compact and, the most importantly, compatible with the matured CMOS fabrication process. Since 2012, we has proposed and demonstrated integrated OAM emitters, plasmonic vortex arrays, angular momentum beam splitters and sorters, as well as the methods to identify the topological charges carried by OAM beams, etc. Here, three representative works are shown. Firstly, beams carrying OAM generated on chips are proposed for wireless optical interconnects and an integrated OAM emitter with a wide switching range of OAM modes is demonstrated. The independence of the micro-ring cavity and the gratings unit provides the flexibility to design the device and optimize the performances. Second, we propose an integrable method for generating vortex Smith-Purcell radiation by letting free electrons pass on holographic gratings and the numerical results indicate that the OAM wave with different topological charges can be obtained. Third, an angular momentum (AM) beam splitter has been demonstrated so that both spin and orbital components carried by lightwave can be distinguished simultaneously.

Photons are ideal flying qubits. Photonics provides an important way to develop quantum information technologies. A quantum information system based on photonics includes functional units for quantum state generation, manipulation, and detection. How to integrate these functions on a photonic chip is a crucial technology for future quantum information applications. We have taken part in the research of integrated quantum photonic circuits since 2010. In these ten years, we have developed comprehensive solutions on integrated quantum light sources for various quantum entangled state generation at telecom bands, based on spontaneous four-wave-mixing in silicon waveguides and micro-ring resonators. We have also developed technologies for quantum state manipulation and detection on a silicon photonic chip. Utilizing high-performance energy-time entangled photon pairs generated in silicon waveguides, we have proposed and demonstrated a scheme of temporal ghost imaging based on the frequency correlation in the photon pairs, and have developed a quantum secure ghost imaging scheme based on it. Recently, we have realized a fully connected quantum key distribution (QKD) network with 40 users and 780 QKD links based on a silicon photonic quantum light source. It is the entanglement-based QKD network with the largest user number to the knowledge of authors.

Conclusion and Prospect In summary, the research achievements of our research group in the field of nanostructured optoelectronic chips are reviewed. Various nanostructures have been successfully developed to control the mechanism of photons, electrons, phonons, surface plasmon polariton and their interactions, and a series of new functional optoelectronic chips have been realized, such as free electron radiation, real-time spectral imaging, phonon sensing, optical orbital angular momentum radiation, and quantum state generation and control. At present, some of the chips are being industrialized and expected to become practical in the near future.

Significance One of the most critical factors for the survival of living organisms is their gas exchange with the environment. Gas exchange is primarily to absorb oxygen and exhale carbon dioxide (CO₂), water vapor, and other gases. The main gases exhaled by the human body are nitrogen (78%), oxygen (15%--18%), water vapor (5%), carbon dioxide (4%--5%), and argon (1%). In addition, there are more than 3000 other gases and volatile organic compounds in the exhaled breath, and their emissions are much less than those of water vapor and CO₂ (with volume fractions of less than 1×10 ^-6). Many of these gaseous compounds have been identified as biomarkers for specific diseases or metabolic disorders. Breath diagnosis involves the detection of traces of gaseous components in exhaled breath that can be used as biomarkers. It can diagnose diseases without traumatic detection such as blood drawing.

The methods used for breath diagnosis can be divided into non-optical and optical methods. Gas chromatography, electrochemical sensors, and chemiluminescence are the common non-optical methods. Optical methods include non-dispersive infrared spectroscopy, cavity ring-down spectroscopy, Fourier-transform infrared spectroscopy, and tunable diode laser absorption spectroscopy (TDLAS). Compared with other breath diagnosis methods, TDLAS has many advantages: high spectral resolution, good gas selectivity, non-invasive measurement, high detection sensitivity (up to 10^-9 level), online measurement, and fast response. Thus, its devices are easy to obtain and low cost, and its detection system is easy to miniaturize and suitable for practical applications.

TDLAS has been widely used in breath diagnosis, and it is necessary to introduce and summarize the main TDLAS technologies, the types and latest developments of lasers and gas cells, as well as the application status of TDLAS technology in common breathing gases, to guide follow-up researches.

Progress Many gases that can be used for breath diagnosis, the common exhaled breath, and related diseases are summarized in Table 1. The laser wavelengths and corresponding detection limits for common exhaled gases estimated by the HITRAN database are shown in Table 2. At the same time, the normal concentration of the exhaled gases is obtained. TDLAS technology can meet the concentration measurement requirements of most exhaled gases, and the detection limit can be further reduced by increasing the absorption optical path of the gas cell. Direct absorption spectroscopy and wavelength modulation spectroscopy are the most widely used TDLAS technologies, a brief introduction of their principles is described in section 2.2 and section 2.3. The types of laser and gas cells commonly used in TDLAS technology are summarized in section 2.5 and section 2.6, respectively. Hollow waveguide is a new type of gas cell, which can achieve a long optical path in a small gas inflation volume.

In the current study on the measurement of ¹³CO₂ and ¹²CO₂ isotopes for human breath based on TDLAS, the wavelength of most lasers used is 4.3 μm, the Herriott cell is usually used, the accuracy of ratio measurement can reach 0.5‰, and the appearance of a hollow optical waveguide causes the inflation volume to decrease further and the measurement time to be significantly reduced. For CO breath detection, quantum cascade lasers with wavelengths at 4.6--5 μm are usually used, most of the systems can realize online measurement, the sensitivity of concentration detection can reach 10^-9 level, circular multi-pass cell proposed by Ghorbani and coworkers can reduce the gas volume to 38 mL. There are many types of laser wavelengths used to measure acetone based on TDLAS. Since many absorption peaks of acetone in the infrared are broadband, it is necessary to exclude the influence of other gases during the measurement.

Conclusion and Prospect To improve the accuracy of concentration detection, for existing breath detection based on TDLAS, mid-infrared quantum cascade laser is commonly used, in combination with the Herriott multi-path gas cell. The volume of the multi-pass cell used in the existing researches is relatively large. Therefore, strict pressure control is required to shorten the inflation time and reduce the volume of gas to be measured in the online breath diagnosis. In the future, smaller and longer path gas cells can be used in breath diagnosis to further improve the detection speed and accuracy. For some gases, such as isotopes of CO₂, CO, and NO, the laboratory research is relatively mature.The key research directions of the breath detection are to further improve the air intake device, collect respiratory data, give the quantitative corresponding relationship between the measured gas concentration and specific diseases, and integrate and commercialize the measurement system. When TDLAS is applied to the breath detection with broadband absorption, the absorption interference between different gases will affect the accuracy of the concentration measurement of the gas to be measured. Exploring new data processing methods, using wavelength locking technology, or other related technologies to achieve a single line gas measurement, and removing interference from other gases are important directions for future research. In addition, the existing TDLAS technology cannot achieve simultaneous measurement of a variety of exhaled gases. It is also important to improve the laser, gas cell, modulation technology, and data processing, and simultaneously measure more types of exhaled gas concentration through TDLAS.