新闻中心
NEWS
2024-10-19 00:00:00
Click:
Cite this article: Yang Yongying, Ling Tong, Cao Pin, Jiang Jiabin. Wavefront sensing technology and applications based on Quadriwave Lateral Shearing Interferometry (Invited) [J]. Infrared and Laser Engineering, 2024, 53(9): 20240331. DOI: 10.3788/IRLA20240331shu
Wavefront Sensing Technology and Applications Based on Quadriwave Lateral Shearing Interferometry (Invited)
Yongying Yang 1,4*,TongLing 2,3,4*,PinCao 4,JiabinJiang 4
(1. College of Optical Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China;
2. School of Chemistry, Chemical Engineering and Biotechnology, Nanyang Technological University, Singapore 639798, Singapore;
3. School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798, Singapore; 4. Hangzhou Zernike Optoelectronic Technology Co., Ltd., Hangzhou, Zhejiang 310027, China)
Fund Projects:
National Natural Science Foundation of China (11275172, 6187517); National Academic Monograph Publishing Fund Project; NTU Start-up Grant (SUG); NRF Fellowship Grant (NRF-NRFF14-2022-0005)
Corresponding Authors:
Yongying Yang, female, Professor, Ph.D., primarily engaged in research on precision interferometric metrology, particularly common-path multiwave lateral shearing interferometry.
Tong Ling, male, Assistant Professor, Ph.D., primarily engaged in research on precision interferometric imaging techniques for biomedical imaging.
The demand for label-free imaging techniques continues to grow in modern optics and biomedical fields. These techniques don't rely on traditional dyes or fluorescent markers, allowing for in situ observation and analysis without interfering with the sample's natural state. Among these, Quantitative Phase Microscopy (QPM) has garnered significant attention in research and clinical diagnostics due to its ability to provide additional information about transparent biological samples, such as their volume and dry mass [1–4]. QPM's greatest advantage lies in its ability to significantly enhance the contrast of microscopic images without any dyes or fluorescent labels, making it particularly suitable for long-term tracking of dynamic processes in live cells.

QPM techniques can be categorized into different branches based on their implementation, including Fourier Ptychographic Microscopy, the Transport of Intensity Equation, Diffraction Phase Microscopy, and Digital Holographic Microscopy (see Table 1). Fourier Ptychographic Microscopy (FPM) reconstructs images by analyzing the phase and amplitude information of a sample's scattered wavefield [5–8]. This technique uses a ptychographic iterative engine that processes the sample's diffraction patterns with complex algorithms, generating high-contrast, artifact-free images. However, a major drawback of this method is its low temporal efficiency; it requires acquiring a large number of ptychographic diffraction patterns and lengthy post-processing, which limits its use in real-time applications.
The Transport of Intensity Equation (TIE) technique recovers a sample's wavefront phase changes by analyzing image differences at various focal planes and can be directly applied to conventional bright-field microscopes [1, 9]. By introducing partially coherent light sources [10], such as light-emitting diodes, and using electrically tunable lenses [11], the image acquisition rate and signal-to-noise ratio of TIE technology have significantly improved, leading to more accurate and faster reconstruction of phase images.

A Crossgrating Lateral Shearing Interferometer (CGLSI) uses a conventional two-dimensional amplitude grating to generate different diffraction orders [19]. The grating consists of a periodic array of square apertures, typically with pitches ranging from a few to tens of micrometers. When an input wavefront illuminates the grating, the grating diffracts light into multiple orders, including the ±1 orders in both x and y directions. For unwanted diffraction orders, such as the zero order and higher-order components, a CGLSI requires an additional order-selection window placed at the Fourier focal plane of the lens [20]. This window only allows the ±1 orders to pass while blocking other orders. However, the need for an order-selection window increases system complexity and limits the flexibility of grating parameters and the magnification of the optical system under test.
The modified Hartmann mask technique combines a two-dimensional amplitude grating with a phase chessboard to reduce unwanted diffraction orders [22–23]. The phase chessboard consists of periodically arranged square cells with a π phase shift, with a period twice that of the amplitude grating. This chessboard-like phase modulation can eliminate zero-order diffraction. Furthermore, by optimizing the duty cycle of the amplitude grating to 2/3 (i.e., the aperture size is 2/3 of the grating pitch), the modified Hartmann mask technique can also suppress ±3 order diffraction components. By eliminating the zero order and adjacent ±3 order components, the modified Hartmann mask technique significantly improves the contrast of the interferogram and removes the need for an order-selection window in the system, making the system more compact and flexible.
Since an ideal sinusoidal transmittance distribution would only produce the desired ±1 diffraction orders, the randomly encoded hybrid grating further improves the grating design by approximating an ideal sinusoidal transmittance distribution [17]. A randomly encoded hybrid grating is formed by superimposing a two-dimensional binary amplitude grating and a phase chessboard similar to the modified Hartmann mask technique. However, instead of a periodic amplitude grating, the randomly encoded hybrid grating employs a pseudo-random encoding scheme to approximate the ideal sinusoidal transmittance distribution. The amplitude grating consists of numerous tiny pixels, typically 1–2 μm in size, with each pixel randomly set to 0 (blocking) or 1 (passing) according to an ideal sinusoidal transmittance distribution [18]. This design breaks the periodic structure of the amplitude grating, minimizing artifacts and unwanted diffraction orders caused by periodic structures. Consequently, it can produce high-contrast Quadriwave Lateral Shearing Interferometry interferograms, enhancing the accuracy and robustness of the phase retrieval process, and enabling high-quality wavefront reconstruction with the same grating in optical systems with different parameters.After obtaining the Quadriwave Lateral Shearing Interferometry interferogram, the original wavefront distribution under test needs to be reconstructed computationally. Its phase retrieval process primarily involves two steps. As shown in Figure 2, first, by performing a Fourier transform on the interferogram generated by the interference of the four tilted wavefronts, the +1 orders in orthogonal directions (x and y) are extracted. Based on this, an inverse Fourier transform is performed, and algorithms such as Goldstein, quality map, or differential unwrapping [16] are used for phase unwrapping to obtain the sheared wavefronts in both x and y directions. To recover the original wavefront from the unwrapped sheared wavefronts, a differential Zernike polynomial fitting method [24] or Fourier transform-based algorithms [25–26] can be used. Differential Zernike polynomial fitting can effectively suppress random noise but filters out high-frequency information, while Fourier transform methods can preserve high-frequency information. For small shear amounts, the original wavefront can be reconstructed in the Fourier frequency domain by approximating the sheared wavefront as partial derivatives; for large shear amounts, the original wavefront can be reconstructed using a least-squares based method.

The original wavefront function W(x,y) can be expressed as the following N-term Zernike polynomial:

In polar coordinates, represented by the radial coordinate ρ and the azimuthal angle θ, the Zernike polynomials can be written in the following form:




Experimental results show that the aberration measurements obtained using the Quadriwave Lateral Shearing Interferometry system are very close to those from the ZYGO GPI interferometer. This validates the accuracy and reliability of Quadriwave Lateral Shearing Interferometry based on randomly encoded hybrid gratings for optical component testing.Given that the Four-wave Interferometric Sensor (FIS4), hereafter referred to as the FIS4 interferometric sensor, utilizes the latest randomly encoded digital grating technology to form a common-path interferometric system through its inherent Quadriwave Lateral Shearing Interferometry, it holds broad application prospects. To facilitate its wider domestic application in defense and national economic sectors and to aid in understanding its principle, the most intuitive understanding of this interferometric sensor can be as the Four-wave Interferometric Sensor (FIS4).Because the system performs self-interference, it requires no reference mirror, boasts a compact structure, and exhibits strong resistance to external environmental interference. It can achieve high-precision and stable detection of the wavefront under test without demanding experimental conditions such as vibration isolation or constant temperature. By leveraging shear differential interference, the system offers a large dynamic range and broader applications. The FIS4 interferometric sensor is equipped with a 5-megapixel camera and features digital image processing software for high-precision FIS4 wavefront reconstruction. Depending on the application, it can output rich data images of the measured wavefront or precision component surface topography, including PV (peak-to-valley) values, RMS (root mean square) values, contour maps, and 3D plots.

图5为利用FIS4干涉传感器研发的3D活体细胞显微镜观察的红细胞动态。

Since the microscope offers label-free 3D imaging of live cells, it can monitor the morphology of live cells throughout the culture process at various stages. This enables research into cell morphology, size, and activity characteristics. Figure 6 shows the state changes of ESC live cells during autophagy at different times, imaged at 20x magnification. The dashed circles in the contour maps highlight the changes in cell morphology at 0, 4, and 8 hours.The cellular objects acquired by the microscope are label-free and do not damage internal cell structures, thus allowing for long-term 3D imaging of live cells. The 3D cell images deciphered from the system's interferograms possess the characteristic of quantitative phase calculation. These 3D images and contour maps can be used to study various parameters during live cell culture, such as cell count, size, volume, and dry mass.



A microscopy system can detect various microscopic features on optical component surfaces, such as the width and depth of scratch-like defects and surface roughness. Using a dual-camera imaging system, the 2D camera's wide field of view allows for rapid identification of surface microscopic feature information, while the digital wavefront interferometric sensor enables 3D depth detection of scratches.
Figure 9(a) presents a localized contour map and a 3D image of a protruding circular ring on a sample surface, as detected by the FIS4 white light profilometer. Figure 9(b) shows the detection of a standard etched line on a fused silica plate, fabricated using electron beam exposure and ion etching. Microscopic depth detection of this etched line provides contour and 3D images of the standard surface line. As seen in Figure 9(b), the peak-to-valley (PV) depth of the standard etched line on the fused silica plate is approximately 258 nm, which aligns with the depth calibrated by a step profiler at around 250 nm.
Figure 10(a) shows the contour map and surface roughness of an optical component surface detected by the FIS4 white light profilometer, revealing a surface roughness of approximately 4.7 nm. Figure 10(b) displays the contour map and microscopic morphology of a gauge block surface, with a detected peak-to-valley (PV) value of approximately 27 nm.This demonstrates that the reflective white light profilometer, built using the FIS4 interferometer sensor, can perform real-time, online measurements of the microscopic morphology and profiles of precision surfaces. Furthermore, due to the stability of its common-path interference, it's suitable for on-site and field inspections.







This document comprehensively introduces the principle, development history, wavefront reconstruction methods, and broad applications of Quadriwave Lateral Shearing Interferometry for phase imaging, particularly focusing on its FIS4 interferometer sensor. By comparing different grating designs, such as cross-grating lateral shearing interferometers, modified Hartmann mask technique, and random encoded hybrid gratings, the advantages of random encoded hybrid gratings in eliminating unwanted diffraction orders and improving interferogram quality are highlighted. Furthermore, the two crucial steps of wavefront reconstruction—sheared wavefront extraction and various algorithms for recovering the original wavefront from the sheared wavefront—are elaborated in detail. Finally, the extensive applications of Quadriwave Lateral Shearing Interferometry for phase imaging in fields such as optical component measurement, live cell imaging, super-resolution 3D localization, and thermal imaging are discussed, showcasing its potential as a cross-disciplinary, versatile tool.
The FIS4 interferometer sensor, utilizing Quadriwave Lateral Shearing Interferometry for phase imaging, demonstrates broad application prospects in numerous fields including biomedicine, optical metrology, and material characterization, owing to its unique advantages such as compactness, robustness, high temporal resolution, and compatibility with existing microscopy systems. With continuous improvements in design and optimization of wavefront reconstruction algorithms, Quadriwave Lateral Shearing Interferometry for phase imaging is expected to achieve higher measurement precision and imaging quality in the future, further expanding its application range. Its excellent anti-vibration performance and online detection capability also hold significant potential for wavefront detection in lithography objectives. The development of this technology not only provides new research tools for relevant fields but also opens up new possibilities for interdisciplinary innovation and discovery.