Improving the spatial and spectral resolution of 2D X-ray near-edge absorption structure(XANES)has been a decade-long pursuit to probe local chemical reactions at the nanoscale.However,the poor signal-to-noise ratio i...Improving the spatial and spectral resolution of 2D X-ray near-edge absorption structure(XANES)has been a decade-long pursuit to probe local chemical reactions at the nanoscale.However,the poor signal-to-noise ratio in the measured images poses significant challenges in quantitative analysis,especially when the element of interest is at a low concentration.In this work,we developed a postimaging processing method using deep neural network to reliably improve the signal-to-noise ratio in the XANES images.展开更多
Domain wall structures form spontaneously due to epitaxial misfit during thin film growth.Imaging the dynamics of domains and domain walls at ultrafast timescales can provide fundamental clues to features that impact ...Domain wall structures form spontaneously due to epitaxial misfit during thin film growth.Imaging the dynamics of domains and domain walls at ultrafast timescales can provide fundamental clues to features that impact electrical transport in electronic devices.Recently,deep learning based methods showed promising phase retrieval(PR)performance,allowing intensity-only measurements to be transformed into snapshot real space images.While the Fourier imaging model involves complex-valued quantities,most existing deep learning based methods solve the PR problem with real-valued based models,where the connection between amplitude and phase is ignored.To this end,we involve complex numbers operation in the neural network to preserve the amplitude and phase connection.Therefore,we employ the complex-valued neural network for solving the PR problem and evaluate it on Bragg coherent diffraction data streams collected from an epitaxial La_(2-x)Sr_(x)CuO_(4)(LSCO)thin film using an X-ray Free Electron Laser(XFEL).Our proposed complex-valued neural network based approach outperforms the traditional real-valued neural network methods in both supervised and unsupervised learning manner.Phase domains are also observed from the LSCO thin film at an ultrafast timescale using the complex-valued neural network.展开更多
As a critical component of coherent X-ray diffraction imaging(CDI),phase retrieval has been extensively applied in X-ray structural science to recover the 3D morphological information inside measured particles.Despite...As a critical component of coherent X-ray diffraction imaging(CDI),phase retrieval has been extensively applied in X-ray structural science to recover the 3D morphological information inside measured particles.Despite meeting all the oversampling requirements of Sayre and Shannon,current phase retrieval approaches still have trouble achieving a unique inversion of experimental data in the presence of noise.Here,we propose to overcome this limitation by incorporating a 3D Machine Learning(ML)model combining(optional)supervised learning with transfer learning.The trained ML model can rapidly provide an immediate result with high accuracy which could benefit real-time experiments,and the predicted result can be further refined with transfer learning.More significantly,the proposed ML model can be used without any prior training to learn the missing phases of an image based on minimization of an appropriate‘loss function’alone.We demonstrate significantly improved performance with experimental Bragg CDI data over traditional iterative phase retrieval algorithms.展开更多
Multimodal hard X-ray scanning probe microscopy has been extensively used to study functional materials providing multiple contrast mechanisms.For instance,combining ptychography with X-ray fluorescence(XRF)microscopy...Multimodal hard X-ray scanning probe microscopy has been extensively used to study functional materials providing multiple contrast mechanisms.For instance,combining ptychography with X-ray fluorescence(XRF)microscopy reveals structural and chemical properties simultaneously.While ptychography can achieve diffraction-limited spatial resolution,the resolution of XRF is limited by the X-ray probe size.Here,we develop a machine learning(ML)model to overcome this problem by decoupling the impact of the X-ray probe from the XRF signal.The enhanced spatial resolution was observed for both simulated and experimental XRF data,showing superior performance over the state-of-the-art scanning XRF method with different nano-sized X-ray probes.Enhanced spatial resolutions were also observed for the accompanying XRF tomography reconstructions.Using this probe profile deconvolution with the proposed ML solution to enhance the spatial resolution of XRF microscopy will be broadly applicable across both functional materials and biological imaging with XRF and other related application areas.展开更多
基金supported by the LDRD project 24255 received from Brookhaven National Laboratory.Z.LM.T.are supported by the Department of Energy,Basic Energy Sciences under project DE-SC0019111。
文摘Improving the spatial and spectral resolution of 2D X-ray near-edge absorption structure(XANES)has been a decade-long pursuit to probe local chemical reactions at the nanoscale.However,the poor signal-to-noise ratio in the measured images poses significant challenges in quantitative analysis,especially when the element of interest is at a low concentration.In this work,we developed a postimaging processing method using deep neural network to reliably improve the signal-to-noise ratio in the XANES images.
基金supported by the U.S.Department of Energy,Office of Science,Office of Basic Energy Sciences,under Contract No.DE-SC0012704supported by EPSRC.Work at Argonne National Laboratory was supported by the U.S.Department of Energy,Office of Science,Office of Basic Energy Sciences,Materials Science and Engineering Division+2 种基金X.H.was supported by the Gordon and Betty Moore Foundation’s EPiQS Initiative through Grant No.GBMF9074S.D.M.and P.G.E.gratefully acknowledge support from the U.S.DOE Office of Science under grant no.DE-FG02-04ER46147from the US NSF through the University of Wisconsin Materials Research Science and Engineering Center(DMR-2309000 and DMR-1720415).
文摘Domain wall structures form spontaneously due to epitaxial misfit during thin film growth.Imaging the dynamics of domains and domain walls at ultrafast timescales can provide fundamental clues to features that impact electrical transport in electronic devices.Recently,deep learning based methods showed promising phase retrieval(PR)performance,allowing intensity-only measurements to be transformed into snapshot real space images.While the Fourier imaging model involves complex-valued quantities,most existing deep learning based methods solve the PR problem with real-valued based models,where the connection between amplitude and phase is ignored.To this end,we involve complex numbers operation in the neural network to preserve the amplitude and phase connection.Therefore,we employ the complex-valued neural network for solving the PR problem and evaluate it on Bragg coherent diffraction data streams collected from an epitaxial La_(2-x)Sr_(x)CuO_(4)(LSCO)thin film using an X-ray Free Electron Laser(XFEL).Our proposed complex-valued neural network based approach outperforms the traditional real-valued neural network methods in both supervised and unsupervised learning manner.Phase domains are also observed from the LSCO thin film at an ultrafast timescale using the complex-valued neural network.
基金Work at Brookhaven National Laboratory was supported by the U.S.Department of Energy,Office of Science,Office of Basic Energy Sciences,under Contract No.DE-SC0012704J.D.received funding from the China Scholarship Council(CSC).Work at UCL was funded by EPSRC.Measurements were carried out at the Advanced Photon Source(APS)beamline 34-ID-C,which was supported by the U.S.Department of Energy,Office of Science,Office of Basic Energy Sciences,under Contract No.DE-AC02-06CH11357The beamline 34-ID-C was built with U.S.National Science Foundation grant DMR-9724294.
文摘As a critical component of coherent X-ray diffraction imaging(CDI),phase retrieval has been extensively applied in X-ray structural science to recover the 3D morphological information inside measured particles.Despite meeting all the oversampling requirements of Sayre and Shannon,current phase retrieval approaches still have trouble achieving a unique inversion of experimental data in the presence of noise.Here,we propose to overcome this limitation by incorporating a 3D Machine Learning(ML)model combining(optional)supervised learning with transfer learning.The trained ML model can rapidly provide an immediate result with high accuracy which could benefit real-time experiments,and the predicted result can be further refined with transfer learning.More significantly,the proposed ML model can be used without any prior training to learn the missing phases of an image based on minimization of an appropriate‘loss function’alone.We demonstrate significantly improved performance with experimental Bragg CDI data over traditional iterative phase retrieval algorithms.
基金This work uses the 3-ID Hard X-ray Nanoprobe(HXN)beamline of the National Synchrotron Light Source II(NSLS-II),which was supported by the U.S.Department of Energy(DOE).NSLS-II is an Office of Science user facility operated by Brookhaven National Laboratory under Contract No.DE-SC0012704.The work at UCL was supported by EPSRCThis work was partially carried out at the MERF facility at Argonne National Laboratory,which is supported within the core funding of the Applied Battery Research for Transportation Program.Argonne,a U.S.DOE,Office of Science laboratory,is operated under Contract No.DE-AC02-06CH11357.We acknowledge the support of the U.S.DOE,Office of Energy Efficiency and Renewable Energy,Vehicle Technologies Office,and in particular the support of Peter Faguy and Dave Howell.
文摘Multimodal hard X-ray scanning probe microscopy has been extensively used to study functional materials providing multiple contrast mechanisms.For instance,combining ptychography with X-ray fluorescence(XRF)microscopy reveals structural and chemical properties simultaneously.While ptychography can achieve diffraction-limited spatial resolution,the resolution of XRF is limited by the X-ray probe size.Here,we develop a machine learning(ML)model to overcome this problem by decoupling the impact of the X-ray probe from the XRF signal.The enhanced spatial resolution was observed for both simulated and experimental XRF data,showing superior performance over the state-of-the-art scanning XRF method with different nano-sized X-ray probes.Enhanced spatial resolutions were also observed for the accompanying XRF tomography reconstructions.Using this probe profile deconvolution with the proposed ML solution to enhance the spatial resolution of XRF microscopy will be broadly applicable across both functional materials and biological imaging with XRF and other related application areas.
