Mode control and nonlinear effect suppression are technical problems to obtain 10 kW-level fiber laser with high beam quality. For the purpose of high beam quality and output power, a counter-tandem-pumped fiber laser employing a double clad fiber with a fiber core diameter of 30 μm and a clad diameter of 250 μm is demonstrated. A laser output with an output power of 10.03 kW, a beam quality factor M2 of 1.92, and a Raman inhibition ratio of 38 dB is achieved. A beam quality factor M2 better than 2 is realized, which confirms the possibility of generation and amplification of 10 kW-level fiber laser of high beam quality with conventional double clad fibers.

In order to investigate the influence of the photodarkening effect of ytterbium-doped fiber (YDF) on high-power and high-brightness fiber lasers, a fiber laser oscillator is established by using 25/400 μm large-mode-area YDF and 915 nm laser diodes. The maximum output power is up to 5 kW, and the output laser is near single mode. When the fiber laser oscillator is operated at full power, the photodarkening effect appears. The output power of the oscillator drops sharply, and strong transverse mode instability (TMI) occurs in the time domain of the output laser. After many tests, it is found that the pump power and output power thresholds corresponding to the TMI effect of the oscillator under the backward pump has a certain decline (about 14%), which is consistent with the traditional theory. However, the pump power and output power thresholds corresponding to the TMI effect of the oscillator under the forward pump shows a certain increase (about 15%), which is inconsistent with the traditional theory. Therefore, further investigation is required. Finally, the output power of the fiber laser oscillator fails to be maintained at 5 kW due to the photodarkening effect.

Based on the femtosecond laser inscribing technology and side-pump coupling technology, a kilowatt-level monolithic all-fiber laser oscillator without fusion splice in the main optical path is obtained. In addition, a pair of fiber Bragg gratings are written in a large-mode-area and double-cladding ytterbium-doped fiber by combining femtosecond laser and phase mask, so as to form a resonant cavity. Meanwhile, two side-pump couplers are fabricated by tapered-fused technology in the same ytterbium-doped fiber. With a semiconductor laser of 976 nm as the pump source, the maximum output power of 1052 W is obtained at a center wavelength of 1070 nm, and the optical-optical conversion efficiency and the beam qualityfactor M2 are about 73% and 1.8, respectively. A compact and stable monolithic fiber laser oscillator system is presented, and its potential in realizing laser output with high power and high beam quality is demonstrated, which is of great value to the development and application of high-power fiber lasers.

The demodulation accuracy of the traditional microwave photonic filtering interrogation technique based on the frequency-time transform is limited by the radio frequency (RF) response measurement bandwidth when it is applied to multi-points or quasi-distributed sensing systems. By the idea of the zoom fast Fourier transform (Zoom-FFT) of the spectrum refinement algorithm, the expression of time-domain refinement is derived, which effectively solves the problem of mutual restriction between RF response measurement bandwidth, demodulation rate, and demodulation accuracy. Compared with the direct and zero-padding inverse discrete Fourier transforms, the proposed algorithm can greatly reduce the requirements for the RF response measurement bandwidth and the computation load of frequency-time transform under the same time-domain resolution, and the demodulation rate is effectively improved. In the verification experiment, a multi-point sensing system containing five fiber Bragg gratings (FBGs) is constructed. The test results reveal that under the same time-domain resolution, the calculated sampling points and time consumption of the proposed algorithm are reduced to 1/200 and 10/145 of the results of the zero-padding algorithm, respectively; for a given 5 GHz sweep-frequency bandwidth, the time-domain resolution in sub-picosecond level can be achieved when the number of sampling points is greater than 1000, which corresponds to the picometer-level wavelength demodulation precision of FBGs.

This paper proposes a multi-frequency instantaneous signal detection scheme based on a non-uniform optical frequency comb. Furthermore, the paper takes the comb's power ratio with non-uniform power drop as a reference, utilizes the beat frequency power ratio of the signal to be measured and the comb to determine the frequency range of the signal, and then calculates the exact frequency of the signal based on the demodulated frequency information. Through the simulation, the instantaneous measurement of multi-frequency signals in the range of 0-20 GHz is realized, with an error ranging from 0 to 23 MHz. In addition, the paper analyzes the effects of factors on the measurement results, such as the laser's linewidth, the center frequency of fiber Bragg grating (FBG), the bias voltage of the electro-absorption modulator, and the phase difference of the optical hybrid coupler. Results show that the system is insensitive to changes in the laser's linewidth and the phase difference of the coupler, but the center frequency of FBG and the bias voltage of the modulator can affect the power ratio of the comb signal. When the power ratio is different to a large extent, the system is more stable, and the error gets smaller.

In this paper, a spurious interference model of Rayleigh scattering for short-range transmission link is constructed, and the relationship between Rayleigh backscattering noise of the system and parameters such as interrogation pulse width, fiber length of transmission link, and Rayleigh scattering rate is summarized by using the parameter characteristics of interference visibility. Both theoretical and experimental results show that the effect of Rayleigh scattering under the short-range transmission does not increase linearly as the fiber length of the transmission link or Rayleigh scattering rate increases. Instead, it tends to be discontinuous. The Rayleigh scattering under the short-range transmission is also closely related to the modulation and demodulation parameters of the interrogation system and changes dynamically as these parameters vary. The research results quantitatively reveal the degradation of fiber-optic hydrophone array with Rayleigh scattering under short-range transmission links and provide significant support for the optimal design of the system under short-range transmission.

This paper demonstrates the coherent optical frequency transfer over a 60 km fiber link between Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences and Huazhong University of Science and Technology. A 1560 nm ultra-stable laser is used as the transfer laser, and an acousto-optic modulator is employed to actively compensate for the phase noise introduced by optical fibers to achieve noise suppression of 67 dB at 1 Hz Fourier frequency. The end-to-end transfer of the all-fiber structure is adopted by the system, and tracking oscillators are used to suppress the amplitude fluctuation of the beating signals caused by fiber polarization change. In this way, the system can operate without interruption all day. The optical frequency offset introduced by the link is at the order of magnitude of 10-20 in the four-day measurement, and the frequency instability is measured to be 2.4×10-17 in 1 s, 6.6×10-20 in 1000 s, and 6.5×10-21 in 65000 s. This result demonstrates the system is applicable for remote optical clock comparison.

In this paper, a new local zoom imaging system is proposed, which contains a liquid crystal lens and a glass lens, so as to solve the problems of large mechanical structure, difficult design of nonlinear motion, and discrete zoom number in the traditional zoom imaging system. Specifically, the proposed system is firstly described, and then the relationship among the parameters of the system is derived. After that, the experimental setup is introduced, and the liquid crystal lens and the dual polarization structures are emphasized. The proposed system is then experimentally validated by changing the voltage applied to the liquid crystal lens group to adjust the imaging zoom ratio in the aperture. Furthermore, the experiment demonstrates that local zoom imaging can be achieved by adjusting the voltage applied to the liquid crystal lens group when system components do not move mechanically, and the zoom ratio is continuously adjustable. This system not only simplifies the mechanical structure of the traditional zoom imaging system, but also makes it possible to lighten and miniaturize the zoom system. In addition, it provides a new method for acquiring and resolving image detail information.

This paper proposes a rigorous nonlinear measurement method of polarization aberrations for the large numerical aperture (NA=0.55) variable-magnification extreme ultraviolet lithography (EUVL) projection objective. First, on the basis of the rigorous variable-magnification extreme ultraviolet (EUV) vector imaging model, the nonlinear overdetermined equations are established with the nonlinear relationships between polarization aberrations and spatial image spectra. Then, a synchronous rotation measurement method is proposed, which constructs and trains a deep neural network algorithm to solve the rigorous nonlinear overdetermined equations and thus achieves the Jones pupil measurement of polarization aberrations for the EUV projection objective with high precision and efficiency. The simulation results show that this method can realize the measurement precision of 10-4λ (λ denotes the wavelength), which will support the online quality monitoring of EUVL at 3-7 nm technical nodes.

Camera calibration accuracy determines the precision of vision-based measurement. To address the issues of limited inclination angle detection and low calibration accuracy, this paper proposes a binocular camera calibration method for target images with large inclination angles. By clustering the geometric feature data of target marked points, the paper designs a marked point extraction algorithm without prior threshold parameters to enhance the capability of detecting target images with large inclination angles. Meanwhile, the paper uses local deformation matching of marked points to replace direct detection according to the matching correlation between the ideal target plane images without inclination angles and target images with inclination angles. In addition, in order to improve the detection accuracy of the real circle center, the projection deviation is estimated by calculating the optimal local deformation parameter. Simulation and experimental results demonstrate that the proposed calibration method is more sensitive in detecting inclination angles than the traditional method. The calibration accuracy for the simulation images is improved by up to 82%, and that for experimental calibration images is enhanced by up to 60%.

To solve the core problem of how to search and converge to the feature points faced by the automatic knife-edge instrument, this paper proposes an aspheric surface testing technology based on the axial movement of the knife-edge instrument to realize the extraction of ring-belt errors during the search process. By this technology, the wavefront characteristics are dynamically obtained during the search process, and the convergence can be accelerated to find the feature points. The mirror ring-belt errors of different parameters are tested, and the relative error between the tested ring-belt error position and the test result of the interferometer is less than 3.3%, which verifies the effectiveness of the method. The proposed method provides a convenient solution for the automatic measurement of the knife-edge instrument and provides technical support for further improving the efficiency of aspheric surface processing.

This paper proposes a nonmetric correction method for lens distortion based on the collinear vanishing point constraint. Regarding the error in distortion center positioning, the fundamental matrix of the distortion model and the least-squares method are used to achieve the high-precision positioning of the distortion center. Furthermore, the distortion measure function for the joint measure by the vanishing points and the straight-lines is defined according to the collinear vanishing point constraint, and the Nelder-Mead algorithm is employed for nonlinear optimization. Accurate distortion model coefficients are thereby calculated iteratively. The experimental results show that the proposed method can effectively and accurately correct lens distortion. Moreover, the method is simple and easy to operate and offers high correction accuracy.

To suppress the vibration-induced measurement error in the Φ600 mm wavelength-tuning interferometer, this paper presents a wavelength-tuning vibration suppression algorithm. In this algorithm, a theoretical vibration model based on wavelength-tuning phase shifting is constructed, and the relationship curve between the light intensity and the initial phase is established. The actual phase is retrieved after the initial phase is corrected by the harmonic coefficients extracted from the spectrum. The error suppression coefficient is introduced as an evaluation parameter. Furthermore, the algorithm is applied to a Φ600 mm wavelength-tuning phase-shifting plane interferometer, and the vibration-induced double-frequency "ripple" error is suppressed well by the algorithm. Specifically, the peak-to-valley (PV) value of the wavefront is decreased from 0.1180λ to 0.0951λ (λ is wavelength), and the root-mean-square (RMS) value is reduced from 0.0164λ to 0.0135λ. The above results verify the effectiveness of the proposed vibration suppression algorithm on wavelength-tuning phase-shifting interferometers.

High-quality photonic materials are important basis for the development of integrated photonics. In recent years, integrated silicon-based chalcogenide photonic devices have been widely researched for optical information processing chips and systematic application. This paper reviews the influencing relationships among chalcogenide glass (ChG) materials, integrated photonic devices, and systematic application. Then, we outlines the technical route of preparing integrated chalcogenide photonic devices with ultra-low loss and the latest research progress of such devices in optical information processing. Integrated chalcogenide photonic devices show great advantages, such as multi-spectrum, low threshold, and multi-functional integration, in their application in optical information processing as ChGs are characterized by ultra-broadband transmission window, high Kerr nonlinearity, large photoelastic coefficient, and readiness for on-chip hybrid integration. Finally, on the basis of the characteristics of chalcogenide materials, this paper presents the opportunities and challenges for integrated chalcogenide photonic devices in future multifunctional integrated photonic devices and their application in high-speed optical information processing.

A standing-wave cavity regenerative amplifier without Faraday rotator is demonstrated. In addition, according to the residual reflectivity of the thin film polarizer, the beams of injected seed laser and output regenerative amplified laser are separated. Finally, under a repetition frequency of 10 kHz, an amplified picosecond pulse laser output with an average power of 4.87 W is obtained, and the pulse width is 53 ps. Furthermore, the laser beam quality factor M2 is smaller than 1.19.

The chamber environment of a laser inertial confinement fusion (ICF) driver is usually complex. Specifically, signals will be affected by the coupling currents induced by ionizing radiation and electromagnetic radiation during the laser shots. In this paper, radiation responses of two kinds of shielded cables are calculated by using the self-written code and CST software, respectively, and the irradiation experiment is carried out based on SG-Ⅲ laser driver, with the results compared and analyzed. It is found that the electromagnetic radiation response of the RG142 cable is slight, but its ionizing radiation is strong, which is the opposite of the CERN SPA6 cable. Finally, according to the coupling law, a composite shielding structure of cable is proposed.

We proposed and numerically studied a method for chaotic synchronization of three-section distributed Bragg reflection (DBR) semiconductor lasers driven by a common noise. The analysis of cross-correlation and correlation dimension shows that a pair of parameter-matched DBR lasers can achieve chaos synchronization when the injection strength of noise light is within 0.34-0.83. More importantly, chaos synchronization is very sensitive to the mismatch of parameters in grating region and phase region of DBR laser, and DBR laser has a large space for hardware parameters. By modulating the current in the grating region, the chaos synchronization on-off keying can be realized, and the synchronization recovery time is only about 4 ns. Using chaos synchronization of DBR laser, we are expected to achieve high-speed and secure chaotic laser key distribution.

Iris textures are easily hidden or even forged by textured contact lenses, which further threatens the security of the iris recognition system. Considering the tiny differences in the optical properties and texture features of authentic irises and irises forged by textured contact lenses, this paper proposes an anti-spoofing detection method for contact lens irises based on recurrent attention, namely recurrent attention iris net (RAINet). Specifically, the recurrent attention mechanism is employed to locate the key regions that can be used to distinguish authentic irises from forged ones in an unsupervised manner, and multi-level feature fusion is applied to improve the anti-spoofing detection accuracy. An end-to-end anti-spoofing detection network is built for the direct detection of authentic and forged features without image pre-processing. MobileNetV2 is used as the feature classification network to reduce the number of parameters and amount of computation of the network in addition to maintaining the detection accuracy. Experimental verification is performed on two public databases (IIITD CLI and ND series) containing both authentic iris samples and contact lens iris samples. The results show that the proposed RAINet outperforms other anti-spoofing detection networks in detection accuracy. Its average correct classification rates under intra-sensor, inter-sensor, and inter-database experimental conditions reach 99.93%, 97.31%, and 97.86%, respectively.

In order to meet the integrated application requirements of full-spectrum hyperspectral imaging and detection, this paper designs a visible-to-long-wave infrared (LWIR) co-aperture imaging spectrometer. The front image square telecentric objective is realized by an off-axis three-mirror structure, and the spectrum is divided into four segments by the field of view separation and beam splitter. The post spectrometer adopts an Offner structure with asymmetric convex grating, and the magnifications of the middle-wave infrared (MWIR) and LWIR spectrometers are set to 0.90 and 0.61, so as to meet the requirements of design specification and imaging quality. The design and analysis results show that the proposed system achieves a ground coverage width of 30 km at an orbital height of 600 km. The F number of visible spectral band to short-wave infrared (SWIR) spectral band is 2.6, and resolutions of visible/near infrared (VNIR) and SWIR are better than 5 nm and 10 nm, respectively, with a pixel resolution of 30 m. The F number of the MWIR and LWIR spectral bands is 2.3 and 1.5, respectively, and their spectral resolutions are better than 50 nm and 100 nm, with a pixel resolution of 50 m and 74 m. The modulation transfer function (MTF) of each spectral band is close to the diffraction limit at the Nyquist frequency, and the spectral line curvature and color distortion are less than 1/10 of the detector's pixel size.

Due to the high sensitivity and real-time detection, surface plasmon resonance microscopy (SPRM) has been widely used in nano-detection, biomedicine, and environmental monitoring. As the evanescence wave propagates along the interface, a special point spread function of SPRM is formed, from which rich information of the analyte can be retrieved. However, the defocus can affect the imaging pattern, which hinders the acquisition of accurate analyte information. As a result, the quantitative study on the effect of defocus on SPRM is crucial. This study quantitatively analyzes the effect of defocus on SPRM both theoretically and experimentally, and realizes the SPRM imaging of single polystyrene nanoparticles. The proposed method can be used to rapidly judge the defocus status of SPRM, retrieve the accurate defocus displacement to achieve fast re-focusing, and improve the SPRM performance in long-term observation.

We use the Magnus expansion method to study the problem of coherent population transfer from the ground state to the indirect exciton state in a three-level asymmetric double quantum-dot system driven by a single laser pulse. By the first-order Magnus expansion of the time-evolution unitary operator of the quantum system, we analytically obtain the pulse-area conditions for achieving complete population transfer of the asymmetric double quantum-dot system without rotation wave approximation. Then, we solve the time-dependent Schrödinger equation of the system for numerical verification of the pulse-area conditions. Furthermore, we compare the performance of the Gaussian pulse, the multi-period cosine pulse, and the single-period cosine pulse in population transfer. The results show that all three pulses can achieve complete population transfers as long as the pulse-area conditions are satisfied, which reveals the physical significance of optical field amplitude in population transfer. The robustness analysis of the three pulse schemes indicates that the single-period cosine pulse is superior to the other two in the resistance against unstable laser pulse parameters and the decoherence effect. This work offers a vital reference for precise optical field control over the quantum state of asymmetric quantum-dot systems, which is of significant value in quantum optics and quantum information science fields.

A ground-state cooling strategy for the mechanical oscillator is proposed to suppress the thermal noise excited by the external thermal environment. For this purpose, the quantitative relationship between the rotation angular velocity and the amplitude of the output light field signal is determined. Then, the influence of the radiation pressure-induced fluctuation spectrum on the number of phonons in the mechanical oscillator is discussed. Finally, the cooling rate and the number of steady-state phonons are investigated to optimize the system parameters and thereby cool the mechanical oscillator to its ground state, that is, to the extent that the number of steady-state phonons is less than 1. Theoretical analysis and simulation results show that the peak value of the radiation pressure-induced fluctuation spectrum can be leveraged to strengthen the cooling process, and its valley value can be utilized to suppress the heating process. The proposed dual-cavity quantum gyroscope model can cool the mechanical oscillator by reducing the number of steady-state phonons to 0.19 and ultimately reduce the thermal noise in the system.

In order to realize the real-time, secure, and high-speed post-processing for quantum random number generation, this work experimentally takes four independent high-frequency sideband modes from quantum vacuum noise as entropy sources using balanced homodyne measurement. In addition, the paper performs four-channel parallel extraction under a sampling rate of 240 MSa/s and 16-bit analog-to-digital conversion in each channel and achieves multiplex real-time, and high-speed Toeplitz-Hash post-processing in a field programmable gate array (FPGA). The large-scale Toeplitz matrix is decomposed, and multi-cycle distributed processing is performed to ensure the stable operation of hardware. Furthermore, the paper investigates the hardware resource occupancy rate of secure post-processing with different matrix sizes and channel numbers, and finally realizes four-channel Toeplitz-Hash post-processing with an FPGA logical resource occupancy rate of 62% and quantum random number generation with a real-time rate of 10.44 Gbit/s. The cross correlation and mutual information of quantum random numbers between every two channels are below 10-3 and 10-6, respectively, and the accumulatively generated quantum random numbers pass the NIST, Diehard, and TestU01 tests. Therefore, this work provides important support for the practical applications of quantum random numbers in high-speed and secure communication.

The four-hole amplitude-modulated wavefront sensor (FHAM-WS) introduces amplitude modulation in each sub-aperture of Shack-Hartmann sensor (SHS) to measure the slope and curvature of the wavefront in the sub-aperture. The precise alignment of FHAM-WS is essential to achieve high precision wavefront sensing. In this paper, the scalar diffraction theory is used to analyze the slope and curvature measurement errors of the wavefront introduced by the alignment error of the microlens array in FHAM-WS in each sub-aperture. Taking the error as input, the aberration introduced by the microlens array alignment error in the whole wave surface measurement is obtained by using the slope and curvature mixed wavefront reconstruction technique. The sensitivities of various aberrations introduced by the focal plane offset error and tilt error of the microlens array in FHAM-WS are simulated and analyzed, and the alignment technique scheme of FHAM-WS is established. The validity of this method is verified by the zero-test experiment of FHAM-WS. The experimental results show that the absolute measurement accuracy of FHAM-WS can reach 0.005λ (root mean square, wavelength of λ=635 nm) after the calibration by this method.

A temperature sensor cascading a reflective Fabry-Perot (FP) etalon with a microring resonator is proposed. This temperature sensor is composed of a fiber mirror, an FP etalon, and a microring resonator. The intensity interrogation method is employed, and both the simulation calculation and experimental verification results show that the proposed temperature sensor offers high sensitivity and a low detection limit. A combination of a broadband light source and a filter is used as the input light source for the experiment. The reflective FP etalon plays the role of a filter, and the temperature sensing element is the microring resonator. The sensitivity of the proposed temperature sensor yielded by the simulation calculation is 2.1406 dB/℃. The experimental results show that the sensitivity of the temperature sensor is 1.9434 dB/℃, and its detection limit is 0.01 ℃.

The polarization effect of ferroelectric films and the hot electrons induced by the nonradioactive damping of surface plasmon in metal micro-nano structures can be used to improve the photoelectric conversion efficiency of traditional ferroelectric films, which has broad application in photovoltaics, photocatalysis, and photoelectric detection. In this paper, uniform BiFeO3 films are fabricated by the sol-gel method, and the electron beam thermal evaporation is employed to make the Au nanoparticles deposited on the upper and lower surfaces, so as to obtain Au/BiFeO3 composite films. The results show that compared with the pure BiFeO3 films, the light absorption of the composite films with Au nanoparticles in the visible region is significantly enhanced, and the photocurrent density is also increased. In addition, the interface barrier of ferroelectric thin films is controlled by the polarization effect of ferroelectric thin films, so as to control the transfer of hot electrons in Au nanoparticles and photogenerated carriers in BiFeO3 thin films, realizing the manipulation of the photocurrent polarity in composite films.

Antimony sulfide (Sb2S3) thin films possess n-type and p-type conductivity. Design and defect analysis on promising light-harvesting material Sb2S3 homojunction solar cells with various electron transport layers and hole transport layers are performed by using wxAMPS. The device structure consisting of glass/FTO/ZnS/(n)Sb2S3/(p)Sb2S3/Spiro-OMeTAD/Au is proposed. In Sb2S3 homojunction solar cells, a built-in electric field is formed that increase the bending of the energy band and therefore leads to the increase of open-circuit voltage. Bulk defects in the (p)Sb2S3 have stronger impact on the device performance than that in the (n)Sb2S3, but defects at ZnS/(n)Sb2S3 interface and (p)Sb2S3/Spiro-OMeTAD interface have the same effects on device performance. When the bulk defect density in (n)Sb2S3 and (p)Sb2S3 is 1015 cm-3, and the interface defect density at ZnS/(n)Sb2S3 interface and (p)Sb2S3/Spiro-OMeTAD interface is 109 cm-2, then the optimized conversion efficiency of the solar cells can reach 23.96%. Simulation results show that device design with Sb2S3 based homojunction is an effective structure to achieve highly efficient solar cells.

Full-field transmission X-ray microscope (TXM) has been widely applied in many research fields owing to its various strengths, such as in-suit non-destructive three-dimensional (3D) imaging with a nanoscale spatial resolution. 3D nano image beamline, a part of the Shanghai Synchrotron Radiation Facility (SSRF) phase-II project, focuses on cutting-edge scientific problems and national strategic needs. The main experimental methods are TXM and nano-computed tomography (nano-CT), and the energy range is 5-14 keV. The design goal for spatial resolution is 20 nm. Based on a bending magnet source, the beamline is built with a cylindrical collimating mirror, a double crystal monochromator, and a toroidal condenser. The experimental endstation adopts a self-designed and integrated full-field TXM system, and the mono-capillary condenser, the TXM mechanical system, and the nano-CT control and data acquisition software of the experimental endstation are all developed independently. The commissioning and performance tests of the SSRF 3D nano image beamline (BL18B) have been completed in 2021, with a resolution of 20 nm achieved for TXM imaging. This beamline is the first TXM beamline based on a bending magnet source in the world with a 20 nm imaging resolution. All the test results have reached the design goals for this beamline, and it will open to users in 2022.

An experimental platform based on a microfocus X-ray grating interferometer is established to support the wavefront control of high-performance light sources and the development of advanced experimental technologies, and facilitate laboratory-level surface shape measurement at a working wavelength. X-ray grating interferometry is a highly sensitive wavefront sensing technique and can be used to quantitatively measure the X-ray wavefront distortion. Furthermore, the phase of the fringes and the wavefront radius of curvature distribution are reconstructed by phase stepping and Fourier analysis, so as to calculate the wavefront angle and mirror slope error distribution. The measurement results obtained by Fourier analysis are in good agreement with the long trace profiler, with the root mean square of their difference less than 200 nrad. The proposed technique can be used for online wavefront feedback and control in X-ray active optics, error detection of reflection, refraction and diffraction devices, and quality evaluation of X-ray beams of large scientific devices.

Light sources such as high-repetition-rate free-electron lasers and low-emittance synchrotron radiation diffraction-limited storage rings place higher demands on the thermal deformations of the mirrors. Given the high average thermal power and the wavefront preservation demand of the Shanghai high repetition rate XFEL and extreme light facility (SHINE), the thermal power distribution on the first mirror named M1 for the first beamline at different energy points is calculated. A finite-element analysis model with thermal and structural coupling is built to calculate the thermal deformation of M1 and conduct wavefront propagation simulation. Finally, the mirror cooling design is optimized by multi-segment cooling and compound utilization. The results show that when the incident light has an energy of 7.0 keV and an grazing angle of 4 mrad, the rated thermal power is improved from 0.48 W to 3.06 W, and the working repetition rate is enhanced by 6.4 times accordingly. According to the optical simulation results, the rated thermal power is also increased by 2.0 to 8.3 times at other energy points.