Fast,precise structure determination of unknown compounds has been the foundation but with a persistent challenge in the field of chemical research.Among various chemical characterization techniques,single-crystal X-r...Fast,precise structure determination of unknown compounds has been the foundation but with a persistent challenge in the field of chemical research.Among various chemical characterization techniques,single-crystal X-ray diffraction(SCXRD)stands out as the most straightforward and accurate method in modern structural chemistry.By precisely determining the three-dimensional arrangement of atoms within a crystal,it provides direct atomic-level evidence for understanding the relationship between material structure and properties[1].展开更多
X-ray Diffraction analysis is crucial for understanding material structures but is hindered by complex patterns and the need for expert interpretation.Deep learning offers automation in phase identification but faces ...X-ray Diffraction analysis is crucial for understanding material structures but is hindered by complex patterns and the need for expert interpretation.Deep learning offers automation in phase identification but faces challenges such as data scarcity,overconfidence in predictions and lack of interpretability.This study addresses these by employing Template Element Replacement to generate a perovskite chemical space containing physically unstable virtual structures,enhancing model understanding of XRD-crystal structure relationships and improving classification accuracy by~5%.A Bayesian-VGGNet model was developed,achieving 84%accuracy on simulated spectra and 75%on external experimental data,while simultaneously estimating prediction uncertainty.Evaluation using Bayesian methods revealed low entropy values,indicating high model confidence.Quantifying the importance of input features to crystal symmetry,aligning significant features of seven crystal systems with physical principles.These approaches enhance the model’s robustness and reliability,making it suitable for practical applications.展开更多
The discovery of the quantum Hall effect in the presence of a relatively strong magnetic field has profoundly inspired the study of topological phase of matter[1],[2],[3],which not only deepens our understanding of co...The discovery of the quantum Hall effect in the presence of a relatively strong magnetic field has profoundly inspired the study of topological phase of matter[1],[2],[3],which not only deepens our understanding of condensed materials beyond the scope of symmetry breaking but also holds significant promise in device application with low or even vanishing energy dissipation.In principle,since the role of magnetic field can be completely replaced by magnetic ordering,quantum Hall effect and its anomalous counterpart,termed quantum anomalous Hall effect(QAHE),typically appear as complementary pair.展开更多
基金the National Natural Science Foundation of China(Nos.22471014 and 22271013)the Beijing Natural Science Foundation(No.2232024)for financial support.
文摘Fast,precise structure determination of unknown compounds has been the foundation but with a persistent challenge in the field of chemical research.Among various chemical characterization techniques,single-crystal X-ray diffraction(SCXRD)stands out as the most straightforward and accurate method in modern structural chemistry.By precisely determining the three-dimensional arrangement of atoms within a crystal,it provides direct atomic-level evidence for understanding the relationship between material structure and properties[1].
基金support from the National Key Research and Development Program of China(grant nos.2021YFA1200904,2020YFA0710700)the National Natural Science Foundation of China(grant nos.12375326)+1 种基金the Innovation Program for IHEP(grant nos.E35457U210)the Postdoctoral Fellowship Program and China Postdoctoral Science Foundation(grant nos.BX20240205),and the directional institutionalized scientific research platform relying on Beijing Synchrotron Radiation Facility of Chinese Academy of Sciences.
文摘X-ray Diffraction analysis is crucial for understanding material structures but is hindered by complex patterns and the need for expert interpretation.Deep learning offers automation in phase identification but faces challenges such as data scarcity,overconfidence in predictions and lack of interpretability.This study addresses these by employing Template Element Replacement to generate a perovskite chemical space containing physically unstable virtual structures,enhancing model understanding of XRD-crystal structure relationships and improving classification accuracy by~5%.A Bayesian-VGGNet model was developed,achieving 84%accuracy on simulated spectra and 75%on external experimental data,while simultaneously estimating prediction uncertainty.Evaluation using Bayesian methods revealed low entropy values,indicating high model confidence.Quantifying the importance of input features to crystal symmetry,aligning significant features of seven crystal systems with physical principles.These approaches enhance the model’s robustness and reliability,making it suitable for practical applications.
基金supported by the National Key R&D Program of China(2022YFA1403700 and 2024YFA1409003)the National Natural Science Foundation of China(12204044,12350401,and 12404056)the Shanghai Science and Technology Innovation Action Plan(24LZ1400800).
文摘The discovery of the quantum Hall effect in the presence of a relatively strong magnetic field has profoundly inspired the study of topological phase of matter[1],[2],[3],which not only deepens our understanding of condensed materials beyond the scope of symmetry breaking but also holds significant promise in device application with low or even vanishing energy dissipation.In principle,since the role of magnetic field can be completely replaced by magnetic ordering,quantum Hall effect and its anomalous counterpart,termed quantum anomalous Hall effect(QAHE),typically appear as complementary pair.
