A machine learning approach to analyze the structural formation of soft matter via image recognition

ABSTRACT A novel method is developed to analyze the structural formation of colloidal particles based on image recognition via a convolutional neural network (CNN). This makes it possible to analyze various complex structures that are difficult to study using a traditional bond-order parameter analysis. Molecular dynamics simulations on soft colloidal systems are performed in quasi two-dimensional and three-dimensional systems, and the efficiency of the proposed method is demonstrated.

[1]  Hai Su,et al.  Deep Learning in Microscopy Image Analysis: A Survey , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[2]  David R. Nelson,et al.  Theory of Two-Dimensional Melting , 1978 .

[3]  H. Stanley,et al.  Unusual phase behavior of one-component systems with two-scale isotropic interactions , 2009, Journal of physics. Condensed matter : an Institute of Physics journal.

[4]  Athanassios Z. Panagiotopoulos,et al.  Multi-atom pattern analysis for binary superlattices. , 2017, Soft matter.

[5]  Athanassios Z Panagiotopoulos,et al.  Automated crystal characterization with a fast neighborhood graph analysis method. , 2018, Soft matter.

[6]  David R. Nelson,et al.  Dislocation-mediated melting in two dimensions , 1979 .

[7]  Gianpietro Malescio,et al.  Stripe phases from isotropic repulsive interactions , 2003, Nature materials.

[8]  P. Steinhardt,et al.  Bond-orientational order in liquids and glasses , 1983 .

[9]  F. Saija,et al.  Hexatic phase in the two-dimensional Gaussian-core model. , 2011, Physical review letters.

[10]  F. Saija,et al.  Hexatic phase and water-like anomalies in a two-dimensional fluid of particles with a weakly softened core. , 2012, The Journal of chemical physics.

[11]  Christoph Dellago,et al.  Neural networks for local structure detection in polymorphic systems. , 2013, The Journal of chemical physics.

[12]  Peter Sollich,et al.  Demixing cascades in cluster crystals. , 2014, The Journal of chemical physics.

[13]  M. Dijkstra,et al.  On the stability of a quasicrystal and its crystalline approximant in a system of hard disks with a soft corona. , 2015, The Journal of chemical physics.

[14]  Philip J Camp,et al.  Structure and phase behavior of a two-dimensional system with core-softened and long-range repulsive interactions. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[15]  G. Grest,et al.  Dynamics of entangled linear polymer melts: A molecular‐dynamics simulation , 1990 .

[16]  G. Malescio,et al.  Anomalous melting behavior under extreme conditions: hard matter turning "soft". , 2008, The Journal of chemical physics.

[17]  Werner Krauth,et al.  Two-step melting in two dimensions: first-order liquid-hexatic transition. , 2011, Physical review letters.

[18]  Michael A. Webb,et al.  Recent advances in machine learning towards multiscale soft materials design , 2019, Current Opinion in Chemical Engineering.

[19]  Sharon C Glotzer,et al.  Machine learning for crystal identification and discovery , 2017, 1710.09861.

[20]  Kremer,et al.  Molecular dynamics simulation for polymers in the presence of a heat bath. , 1986, Physical review. A, General physics.

[21]  T. Ramakrishnan.  Density-wave theory of first-order freezing in two dimensions , 1982 .

[22]  M. Dijkstra,et al.  Phase behaviour of quasicrystal forming systems of core-corona particles. , 2017, The Journal of chemical physics.

[23]  M. J. Ruiz-Montero,et al.  Numerical evidence for bcc ordering at the surface of a critical fcc nucleus. , 1995, Physical review letters.

[24]  J. Q. Broughton,et al.  Molecular-dynamics study of melting in two dimensions. Inverse-twelfth-power interaction , 1982 .

[25]  Andrea J. Liu,et al.  Relationship between local structure and relaxation in out-of-equilibrium glassy systems , 2016, Proceedings of the National Academy of Sciences.

[26]  P. Ziherl,et al.  Mosaic two-lengthscale quasicrystals , 2014, Nature.

[27]  F. Stillinger.  Phase transitions in the Gaussian core system , 1976 .

[28]  Jean-Pierre Hansen,et al.  A Monte Carlo study of the classical two-dimensional one-component plasma , 1982 .

[29]  Yilong Han,et al.  Two-dimensional freezing criteria for crystallizing colloidal monolayers. , 2010, The Journal of chemical physics.

[30]  Andrew L. Ferguson,et al.  Machine learning for autonomous crystal structure identification. , 2017, Soft matter.

[31]  Matteo Salvalaglio,et al.  DeepIce: A Deep Neural Network Approach To Identify Ice and Water Molecules , 2019, J. Chem. Inf. Model..

[32]  Gerd E. Schröder-Turk,et al.  Shortcomings of the bond orientational order parameters for the analysis of disordered particulate matter. , 2012, The Journal of chemical physics.

[33]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[34]  Christoph Dellago,et al.  Accurate determination of crystal structures based on averaged local bond order parameters. , 2008, The Journal of chemical physics.

[35]  Melissa C. Smith,et al.  A generalized deep learning approach for local structure identification in molecular simulations , 2019, Chemical science.

[36]  Andrew L. Ferguson,et al.  Machine learning and data science in soft materials engineering , 2018, Journal of physics. Condensed matter : an Institute of Physics journal.

[37]  G. Malescio,et al.  Anomalous phase behavior of a soft-repulsive potential with a strictly monotonic force. , 2009, Physical review. E, Statistical, nonlinear, and soft matter physics.

[38]  C. Santangelo,et al.  Soft spheres make more mesophases , 2006, cond-mat/0609570.

[39]  Ranganathan,et al.  Freezing transition of two-dimensional Lennard-Jones fluids. , 1992, Physical review. A, Atomic, molecular, and optical physics.

[40]  Katherine J. Strandburg,et al.  Two-dimensional melting , 1988 .

[41]  John Stewart,et al.  An accurate perception method for low contrast bright field microscopy in heterogeneous microenvironments , 2017 .

[42]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[43]  H. Löwen,et al.  Low-Temperature Crystal Structures of the Hard Core Square Shoulder Model , 2017, Materials.

[44]  B. A. Lindquist,et al.  Unsupervised machine learning for detection of phase transitions in off-lattice systems. I. Foundations. , 2018, The Journal of chemical physics.

[45]  T. Terao.  Monte Carlo Simulation of Polymer–Nanoparticle Composites , 2012 .

[46]  Jennifer M. Rieser,et al.  Structure-property relationships from universal signatures of plasticity in disordered solids , 2017, Science.

[47]  H. Emmerich,et al.  Molecular dynamics study of colloidal quasicrystals. , 2016, Soft matter.

[48]  A. P. Young,et al.  Melting and the vector Coulomb gas in two dimensions , 1979 .

[49]  F. Müller-Plathe,et al.  Local bond order parameters for accurate determination of crystal structures in two and three dimensions. , 2018, Physical chemistry chemical physics : PCCP.

[50]  T. Terao,et al.  Generalised local bond order parameter analysis: application to colloidal particles with dendritic polymer brushes , 2019, Molecular Simulation.

[51]  Bianca M. Mladek,et al.  Clustering in the absence of attractions: density functional theory and computer simulations. , 2007, The journal of physical chemistry. B.

[52]  Carolyn L. Phillips,et al.  Discovering crystals using shape matching and machine learning , 2013 .

[53]  E. A. Jagla.  PHASE BEHAVIOR OF A SYSTEM OF PARTICLES WITH CORE COLLAPSE , 1998 .

[54]  A. Rucklidge,et al.  Soft-core particles freezing to form a quasicrystal and a crystal-liquid phase. , 2015, Physical review. E, Statistical, nonlinear, and soft matter physics.

[55]  Steve Plimpton,et al.  Fast parallel algorithms for short-range molecular dynamics , 1993 .

[56]  D. Thouless,et al.  Ordering, metastability and phase transitions in two-dimensional systems , 1973 .

[57]  G. Grest,et al.  Effective interactions between grafted nanoparticles in a polymer matrix , 2012 .

[58]  Andrea J. Liu,et al.  A structural approach to relaxation in glassy liquids , 2015, Nature Physics.