Periodic Reporting for period 4 - TIMP (Ultrahigh-speed nanometer-scale microscopy)
Período documentado: 2023-09-01 hasta 2025-02-28
Ultrahigh-speed microscopy at THz frame rate is essential for exploring ultrafast non-repetitive events, for which the pump-probe technique is inapplicable. The spatial resolutions of such microscopes is to date limited to the micrometer scale. In this project, we develop an ultrahigh speed microscope with nanometric scale resolution, for exploring laser-induced ultrafast phase transitions (e.g. femto-magnetism).
The new microscope is based on a new approach for ultrahigh-speed imaging that we recently proposed: time-resolved imaging by multiplexed ptychography (TIMP). In TIMP, multiple frames of the object are recovered algorithmically from data measured in a single CCD exposure of a single-shot ptychographic microscope (we recently demonstrated experimentally reconstruction of 36 complex-valued frames). A burst of ultrashort pulses (finite pulse train) illuminates the object. For each pulse, the imaging system (single-shot ptychograpic microscope) produces a diffraction pattern on the CCD. The diffraction pattern is highly redundant: it contains more information than needed to determine uniquely the object at high resolution. This redundancy is critical because the CCD is slow, and so integrates over all the overlapping diffraction patterns and records the sum of them – the multiplexed diffraction pattern. Thanks to the redundancy, the recorded multiplexed pattern can determine all the frames, without decreasing the spatial resolution of the single-shot ptychographic system (the cost for multi-framing can be in other quantities, e.g. field of view or setup complexity). In order to arrange the frames in the correct order, each pulse in the burst should be different from all the other pulses (e.g. in their spectrum, polarization or spatial mode). The number of recovered frames corresponds to the number of pulses.
The new microscope is based on a new approach for ultrahigh-speed imaging that we recently proposed: time-resolved imaging by multiplexed ptychography (TIMP). In TIMP, multiple frames of the object are recovered algorithmically from data measured in a single CCD exposure of a single-shot ptychographic microscope (we recently demonstrated experimentally reconstruction of 36 complex-valued frames). A burst of ultrashort pulses (finite pulse train) illuminates the object. For each pulse, the imaging system (single-shot ptychograpic microscope) produces a diffraction pattern on the CCD. The diffraction pattern is highly redundant: it contains more information than needed to determine uniquely the object at high resolution. This redundancy is critical because the CCD is slow, and so integrates over all the overlapping diffraction patterns and records the sum of them – the multiplexed diffraction pattern. Thanks to the redundancy, the recorded multiplexed pattern can determine all the frames, without decreasing the spatial resolution of the single-shot ptychographic system (the cost for multi-framing can be in other quantities, e.g. field of view or setup complexity). In order to arrange the frames in the correct order, each pulse in the burst should be different from all the other pulses (e.g. in their spectrum, polarization or spatial mode). The number of recovered frames corresponds to the number of pulses.
We have been progressing in several fronts in the TIMP microscope.
1. Experimental development of TIMP microscope in the visible spectral range, and its applications.
1.1. In Wengrowicz et al., Opt. Exp. 2019 we demonstrated experimentally the concept of TIMP. Specifically, we reconstructed 36 complex-valued frames from data recorded in a single exposure of the CCD.
1.2. In Veler et al., Optics Letters 2024, we transferred TIMP to the ultrafast domain, demonstrating TIMP movies of four frames at nanosecond frame time. Our system is capable of capturing dynamics on the femtosecond scale. We utilized the microscope to explore the dynamics of soot layer following laser pulse illumination.
1.3. In Veler et al., IEEE Photonics Journal 2024, we utilized the TIMP microscope for spatial characterization of the modes of a train of ultrafast (femtosecond scale duration) pulses of a laser in which each pulse has a different spatial profile.
1.4. In a still ongoing experiment, we utilize our TIMP microscope for exploring femto-magnetism.
2. Developing TIMP microscope in short-wavelength spectral region.
In Levitan et al., Optics Letters (2025), we demonstrated a method for single-shot ptychography in the x-ray spectral region. Single-shot ptychography is the first step for TIMP. We continue to develop TIMP in the x-ray spectral region. In parallel, we develop TIMP microscope in the extreme ultraviolet spectral region.
3. Exploring theoretically and experimentally new approaches for single-shot ptychography, TIMP, and related apperatuses.
3.1 In Haham et al., J. Opt., 22, 075608 (2020), we proposed and demonstrated a new approach for single shot ptychography and TIMP. It is based on upgrading a conventional single-shot microscope into a single-shot ptychographic microscope, without impairing its optical performance. It splits the microscope’s intermediate image plane into multiple replicas, and detecting a set of their coded Fourier transform magnitudes, using a different sensor for each replica. To code each beam, it is transmitted through a distinct coding mask.
3.2 In Haham et al., Journal of Physics: Photonics (2021), we proposed and demonstrated a device for characterizing vectorial laser pulses in a single shot that is based on concepts from ptychography.
3.3 In Haham et al., Optics letters (2025), we proposed a single-shot diffractive imaging scheme, relying on illuminating an object with several mutually in-coherent modulating probes. The resulting multiplexed diffraction pattern is recorded and used to reconstruct the object. Compared with standard CDI, our method is robust to dynamic range limitations and noise, and the reconstruction process does not stagnate. Compared to other coherent modulating schemes, it enables better single-shot reconstruction and requires less stringent assumptions. We demonstrate the method experimentally in the visible spectral range.
3.4 In Neufeld et al., Optics Express (2022), we proposed an all-optical method for detecting and characterizing multiple chirality centers in chiral molecules. The method utilize algorithmic methods inspired by our works on TIMP.
4. Reconstruction algorithims. We have put major effort to develop more effcient reconstruction algorithims in TIMP and single-shot ptychography.
4.1 In Wengrowicz et al., Optics express (2020), we developed and explored a deep learning based single-shot ptychography reconstruction method. We show that a deep neural network, trained using only experimental data and without any model of the system, leads to reconstructions of natural real-valued images with higher spatial resolution and better resistance to systematic noise than common iterative algorithms.
4.2 In Wengrowicz et al., Optics express (2024), we developed and explored an unsupervised, physics-informed, deep learning-based reconstruction technique for TIMP. In this method, the untrained deep learning model replaces the iterative algorithm’s update step, yielding superior reconstructions of multiple dynamic object frames compared to conventional methodologies. More precisely, we demonstrated improvements in image quality and resolution, while reducing sensitivity to the number of recorded frames, the mutual orthogonality of different probe modes, overlap between neighboring probe beams and the cutoff frequency of the ptychographic microscope – properties that are generally of paramount importance for ptychographic reconstruction algorithms.
1. Experimental development of TIMP microscope in the visible spectral range, and its applications.
1.1. In Wengrowicz et al., Opt. Exp. 2019 we demonstrated experimentally the concept of TIMP. Specifically, we reconstructed 36 complex-valued frames from data recorded in a single exposure of the CCD.
1.2. In Veler et al., Optics Letters 2024, we transferred TIMP to the ultrafast domain, demonstrating TIMP movies of four frames at nanosecond frame time. Our system is capable of capturing dynamics on the femtosecond scale. We utilized the microscope to explore the dynamics of soot layer following laser pulse illumination.
1.3. In Veler et al., IEEE Photonics Journal 2024, we utilized the TIMP microscope for spatial characterization of the modes of a train of ultrafast (femtosecond scale duration) pulses of a laser in which each pulse has a different spatial profile.
1.4. In a still ongoing experiment, we utilize our TIMP microscope for exploring femto-magnetism.
2. Developing TIMP microscope in short-wavelength spectral region.
In Levitan et al., Optics Letters (2025), we demonstrated a method for single-shot ptychography in the x-ray spectral region. Single-shot ptychography is the first step for TIMP. We continue to develop TIMP in the x-ray spectral region. In parallel, we develop TIMP microscope in the extreme ultraviolet spectral region.
3. Exploring theoretically and experimentally new approaches for single-shot ptychography, TIMP, and related apperatuses.
3.1 In Haham et al., J. Opt., 22, 075608 (2020), we proposed and demonstrated a new approach for single shot ptychography and TIMP. It is based on upgrading a conventional single-shot microscope into a single-shot ptychographic microscope, without impairing its optical performance. It splits the microscope’s intermediate image plane into multiple replicas, and detecting a set of their coded Fourier transform magnitudes, using a different sensor for each replica. To code each beam, it is transmitted through a distinct coding mask.
3.2 In Haham et al., Journal of Physics: Photonics (2021), we proposed and demonstrated a device for characterizing vectorial laser pulses in a single shot that is based on concepts from ptychography.
3.3 In Haham et al., Optics letters (2025), we proposed a single-shot diffractive imaging scheme, relying on illuminating an object with several mutually in-coherent modulating probes. The resulting multiplexed diffraction pattern is recorded and used to reconstruct the object. Compared with standard CDI, our method is robust to dynamic range limitations and noise, and the reconstruction process does not stagnate. Compared to other coherent modulating schemes, it enables better single-shot reconstruction and requires less stringent assumptions. We demonstrate the method experimentally in the visible spectral range.
3.4 In Neufeld et al., Optics Express (2022), we proposed an all-optical method for detecting and characterizing multiple chirality centers in chiral molecules. The method utilize algorithmic methods inspired by our works on TIMP.
4. Reconstruction algorithims. We have put major effort to develop more effcient reconstruction algorithims in TIMP and single-shot ptychography.
4.1 In Wengrowicz et al., Optics express (2020), we developed and explored a deep learning based single-shot ptychography reconstruction method. We show that a deep neural network, trained using only experimental data and without any model of the system, leads to reconstructions of natural real-valued images with higher spatial resolution and better resistance to systematic noise than common iterative algorithms.
4.2 In Wengrowicz et al., Optics express (2024), we developed and explored an unsupervised, physics-informed, deep learning-based reconstruction technique for TIMP. In this method, the untrained deep learning model replaces the iterative algorithm’s update step, yielding superior reconstructions of multiple dynamic object frames compared to conventional methodologies. More precisely, we demonstrated improvements in image quality and resolution, while reducing sensitivity to the number of recorded frames, the mutual orthogonality of different probe modes, overlap between neighboring probe beams and the cutoff frequency of the ptychographic microscope – properties that are generally of paramount importance for ptychographic reconstruction algorithms.
Within the project period, we did not progress beyond the expected results in the core topics of the project. We did progressed beyond the expected results and state of the art in one topic - developing squeezed bright short-wavelength (UV and extreme UV) radiation that has potential to increase the resolution of TIMP as well as other microscopes. Relevant papers within this topic:
1. Matan Even Tzur, Michael Birk, Alexey Gorlach, Michael Krüger, Ido Kaminer and Oren Cohen, Generation of squeezed high-order harmonics, Physical Review Research, 6, 033079 (2024)
2. Matan Even Tzur; Michael Birk; Alexey Gorlach; Michael Krüger; Ido Kaminer; Oren Cohen, Photon-statistics force in ultrafast electron dynamics, Nature Photonics 2025
3. Matan Even Tzur and Oren Cohen, Motion of charged particles in bright squeezed vacuum, Light: Science & Applications, 13, 41 (2024)
4. Alexey Gorlach, Matan Even Tzur, Michael Birk, Michael Krüger, Nicholas Rivera, Oren Cohen, Ido Kaminer, High harmonic generation driven by quantum light, Nature Physics, 19, 1689 (2023)
1. Matan Even Tzur, Michael Birk, Alexey Gorlach, Michael Krüger, Ido Kaminer and Oren Cohen, Generation of squeezed high-order harmonics, Physical Review Research, 6, 033079 (2024)
2. Matan Even Tzur; Michael Birk; Alexey Gorlach; Michael Krüger; Ido Kaminer; Oren Cohen, Photon-statistics force in ultrafast electron dynamics, Nature Photonics 2025
3. Matan Even Tzur and Oren Cohen, Motion of charged particles in bright squeezed vacuum, Light: Science & Applications, 13, 41 (2024)
4. Alexey Gorlach, Matan Even Tzur, Michael Birk, Michael Krüger, Nicholas Rivera, Oren Cohen, Ido Kaminer, High harmonic generation driven by quantum light, Nature Physics, 19, 1689 (2023)