scispace - formally typeset
Search or ask a question
Author

Maxim Ziatdinov

Bio: Maxim Ziatdinov is an academic researcher from Oak Ridge National Laboratory. The author has contributed to research in topics: Physics & Computer science. The author has an hindex of 21, co-authored 156 publications receiving 1482 citations. Previous affiliations of Maxim Ziatdinov include University of Maryland, College Park & GlaxoSmithKline.

Papers published on a yearly basis

Papers
More filters
Journal ArticleDOI
14 Dec 2017-ACS Nano
TL;DR: This work develops a "weakly supervised" approach that uses information on the coordinates of all atomic species in the image, extracted via a deep neural network, to identify a rich variety of defects that are not part of an initial training set.
Abstract: Recent advances in scanning transmission electron and scanning probe microscopies have opened exciting opportunities in probing the materials structural parameters and various functional properties in real space with angstrom-level precision. This progress has been accompanied by an exponential increase in the size and quality of data sets produced by microscopic and spectroscopic experimental techniques. These developments necessitate adequate methods for extracting relevant physical and chemical information from the large data sets, for which a priori information on the structures of various atomic configurations and lattice defects is limited or absent. Here we demonstrate an application of deep neural networks to extract information from atomically resolved images including location of the atomic species and type of defects. We develop a “weakly supervised” approach that uses information on the coordinates of all atomic species in the image, extracted via a deep neural network, to identify a rich vari...

299 citations

Journal ArticleDOI
01 Feb 2019
TL;DR: In this paper, a deep learning framework was developed for dynamic transmission electron microscopy (STEM) imaging that is trained to find the lattice defects and apply it for mapping solid state reactions and transformations in layered WS2.
Abstract: Recent advances in scanning transmission electron microscopy (STEM) allow the real-time visualization of solid-state transformations in materials, including those induced by an electron beam and temperature, with atomic resolution. However, despite the ever-expanding capabilities for high-resolution data acquisition, the inferred information about kinetics and thermodynamics of the process, and single defect dynamics and interactions is minimal. This is due to the inherent limitations of manual ex situ analysis of the collected volumes of data. To circumvent this problem, we developed a deep-learning framework for dynamic STEM imaging that is trained to find the lattice defects and apply it for mapping solid state reactions and transformations in layered WS2. The trained deep-learning model allows extracting thousands of lattice defects from raw STEM data in a matter of seconds, which are then classified into different categories using unsupervised clustering methods. We further expanded our framework to extract parameters of diffusion for sulfur vacancies and analyzed transition probabilities associated with switching between different configurations of defect complexes consisting of Mo dopant and sulfur vacancy, providing insight into point-defect dynamics and reactions. This approach is universal and its application to beam-induced reactions allows mapping chemical transformation pathways in solids at the atomic level.

118 citations

Journal ArticleDOI
TL;DR: Recent work focusing on generation and application of libraries from both experiment and theoretical tools, across length scales is reviewed, showing how modeling, macroscopic experiments and atomic-scale imaging can be combined to dramatically accelerate understanding and development of new material systems via a statistical physics framework.
Abstract: The use of statistical/machine learning (ML) approaches to materials science is experiencing explosive growth. Here, we review recent work focusing on the generation and application of libraries from both experiment and theoretical tools. The library data enables classical correlative ML and also opens the pathway for exploration of underlying causative physical behaviors. We highlight key advances facilitated by this approach and illustrate how modeling, macroscopic experiments, and imaging can be combined to accelerate the understanding and development of new materials systems. These developments point toward a data-driven future wherein knowledge can be aggregated and synthesized, accelerating the advancement of materials science.

112 citations

Journal ArticleDOI
10 Aug 2017
TL;DR: A ‘machine vision’ approach, adapting techniques already used in cancer detection and satellite imaging, is developed to analyze up to 1000 of individual molecular configurations from microscopic images of so-called ‘buckybowl’ molecules on gold surfaces, which revealed details of how buckybowls interact with their neighbors to build complex arrays, which could help in the design of molecular memory devices.
Abstract: Recent advances in high resolution scanning transmission electron and scanning probe microscopies have allowed researchers to perform measurements of materials structural parameters and functional properties in real space with a picometre precision. In many technologically relevant atomic and/or molecular systems, however, the information of interest is distributed spatially in a non-uniform manner and may have a complex multi-dimensional nature. One of the critical issues, therefore, lies in being able to accurately identify (‘read out’) all the individual building blocks in different atomic/molecular architectures, as well as more complex patterns that these blocks may form, on a scale of hundreds and thousands of individual atomic/molecular units. Here we employ machine vision to read and recognize complex molecular assemblies on surfaces. Specifically, we combine Markov random field model and convolutional neural networks to classify structural and rotational states of all individual building blocks in molecular assembly on the metallic surface visualized in high-resolution scanning tunneling microscopy measurements. We show how the obtained full decoding of the system allows us to directly construct a pair density function—a centerpiece in analysis of disorder-property relationship paradigm—as well as to analyze spatial correlations between multiple order parameters at the nanoscale, and elucidate reaction pathway involving molecular conformation changes. The method represents a significant shift in our way of analyzing atomic and/or molecular resolved microscopic images and can be applied to variety of other microscopic measurements of structural, electronic, and magnetic orders in different condensed matter systems. Complex patterns formed by thousands of molecules on a surface can now be automatically recognized and classified by a computer. Sergei Kalinin, Maxim Ziatdinov and Artem Maksov at Oak Ridge National Lab have developed a ‘machine vision’ approach, adapting techniques already used in cancer detection and satellite imaging, to analyze up to 1000 s of individual molecular configurations from microscopic images of so-called ‘buckybowl’ molecules on gold surfaces. These bowl-shaped molecules may rest face up or down, and can adopt different rotational orientations, resulting in a rich tapestry of molecular patterns which are too complicated to sort manually or with existing computational methods. However, this machine vision approach revealed details of how buckybowls interact with their neighbors to build complex arrays, which could help in the design of molecular memory devices.

86 citations

Journal ArticleDOI
TL;DR: A comprehensive, atomically resolved real-space study by scanning transmission electron and scanning tunnelling microscopies on a novel layered material displaying Kitaev physics, α-RuCl3, revealing considerable variations in the geometry of the ligand sublattice in thin films of α- RuCl3 that opens a way to realization of a spatially inhomogeneous magnetic ground state at the nanometre length scale.
Abstract: A pseudospin-1/2 Mott phase on a honeycomb lattice is proposed to host the celebrated two-dimensional Kitaev model which has an elusive quantum spin liquid ground state, and fascinating physics relevant to the development of future templates towards topological quantum bits. Here we report a comprehensive, atomically resolved real-space study by scanning transmission electron and scanning tunnelling microscopies on a novel layered material displaying Kitaev physics, α-RuCl3. Our local crystallography analysis reveals considerable variations in the geometry of the ligand sublattice in thin films of α-RuCl3 that opens a way to realization of a spatially inhomogeneous magnetic ground state at the nanometre length scale. Using scanning tunnelling techniques, we observe the electronic energy gap of ≈0.25 eV and intra-unit cell symmetry breaking of charge distribution in individual α-RuCl3 surface layer. The corresponding charge-ordered pattern has a fine structure associated with two different types of charge disproportionation at Cl-terminated surface. The two-dimensional Kitaev model is a quantum spin liquid state that theory predicts should appear in some materials with a honeycomb lattice. Here, the authors use atom-resolution scanning transmission electron and scanning tunnelling microscopies to characterize one such candidate material, α-RuCl3.

82 citations


Cited by
More filters
01 May 1993
TL;DR: Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems.
Abstract: Three parallel algorithms for classical molecular dynamics are presented. The first assigns each processor a fixed subset of atoms; the second assigns each a fixed subset of inter-atomic forces to compute; the third assigns each a fixed spatial region. The algorithms are suitable for molecular dynamics models which can be difficult to parallelize efficiently—those with short-range forces where the neighbors of each atom change rapidly. They can be implemented on any distributed-memory parallel machine which allows for message-passing of data between independently executing processors. The algorithms are tested on a standard Lennard-Jones benchmark problem for system sizes ranging from 500 to 100,000,000 atoms on several parallel supercomputers--the nCUBE 2, Intel iPSC/860 and Paragon, and Cray T3D. Comparing the results to the fastest reported vectorized Cray Y-MP and C90 algorithm shows that the current generation of parallel machines is competitive with conventional vector supercomputers even for small problems. For large problems, the spatial algorithm achieves parallel efficiencies of 90% and a 1840-node Intel Paragon performs up to 165 faster than a single Cray C9O processor. Trade-offs between the three algorithms and guidelines for adapting them to more complex molecular dynamics simulations are also discussed.

29,323 citations

Journal ArticleDOI
26 Jul 2018-Nature
TL;DR: A future in which the design, synthesis, characterization and application of molecules and materials is accelerated by artificial intelligence is envisaged.
Abstract: Here we summarize recent progress in machine learning for the chemical sciences. We outline machine-learning techniques that are suitable for addressing research questions in this domain, as well as future directions for the field. We envisage a future in which the design, synthesis, characterization and application of molecules and materials is accelerated by artificial intelligence.

2,295 citations

25 Apr 2017
TL;DR: This presentation is a case study taken from the travel and holiday industry and describes the effectiveness of various techniques as well as the performance of Python-based libraries such as Python Data Analysis Library (Pandas), and Scikit-learn (built on NumPy, SciPy and matplotlib).
Abstract: This presentation is a case study taken from the travel and holiday industry. Paxport/Multicom, based in UK and Sweden, have recently adopted a recommendation system for holiday accommodation bookings. Machine learning techniques such as Collaborative Filtering have been applied using Python (3.5.1), with Jupyter (4.0.6) as the main framework. Data scale and sparsity present significant challenges in the case study, and so the effectiveness of various techniques are described as well as the performance of Python-based libraries such as Python Data Analysis Library (Pandas), and Scikit-learn (built on NumPy, SciPy and matplotlib). The presentation is suitable for all levels of programmers.

1,338 citations

Journal ArticleDOI
08 Aug 2019
TL;DR: A comprehensive overview and analysis of the most recent research in machine learning principles, algorithms, descriptors, and databases in materials science, and proposes solutions and future research paths for various challenges in computational materials science.
Abstract: One of the most exciting tools that have entered the material science toolbox in recent years is machine learning. This collection of statistical methods has already proved to be capable of considerably speeding up both fundamental and applied research. At present, we are witnessing an explosion of works that develop and apply machine learning to solid-state systems. We provide a comprehensive overview and analysis of the most recent research in this topic. As a starting point, we introduce machine learning principles, algorithms, descriptors, and databases in materials science. We continue with the description of different machine learning approaches for the discovery of stable materials and the prediction of their crystal structure. Then we discuss research in numerous quantitative structure–property relationships and various approaches for the replacement of first-principle methods by machine learning. We review how active learning and surrogate-based optimization can be applied to improve the rational design process and related examples of applications. Two major questions are always the interpretability of and the physical understanding gained from machine learning models. We consider therefore the different facets of interpretability and their importance in materials science. Finally, we propose solutions and future research paths for various challenges in computational materials science.

1,301 citations

Journal Article
TL;DR: In this paper, the authors used electron beams instead of photons to detect plasmons as resonance peaks in the energy-loss spectra of sub-nanometre electron beams rastered on nanoparticles of well-defined geometrical parameters.
Abstract: Understanding how light interacts with matter at the nanometre scale is a fundamental issue in optoelectronics and nanophotonics. In particular, many applications (such as bio-sensing, cancer therapy and all-optical signal processing) rely on surface-bound optical excitations in metallic nanoparticles. However, so far no experimental technique has been capable of imaging localized optical excitations with sufficient resolution to reveal their dramatic spatial variation over one single nanoparticle. Here, we present a novel method applied on silver nanotriangles, achieving such resolution by recording maps of plasmons in the near-infrared/visible/ultraviolet domain using electron beams instead of photons. This method relies on the detection of plasmons as resonance peaks in the energy-loss spectra of subnanometre electron beams rastered on nanoparticles of well-defined geometrical parameters. This represents a significant improvement in the spatial resolution with which plasmonic modes can be imaged, and provides a powerful tool in the development of nanometre-level optics.

803 citations