Home

We are a computational materials science research group based in the Department of Materials Science and Engineering at MIT. Our work is focused on elucidating the mechanisms of microstructural evolution for systems of relevance in materials science broadly construed. We employ a combination of computational, theoretical, and data-driven techniques to perform physics-based modeling of materials.

Access the article here.
Access the article here.
Access here. See also: MIT News, DMSE News, ACCESS News, and Kudos
Access here. See also: MIT News, DMSE News, ACCESS News, and Kudos
Review article on the application of Machine Learning potentials to crystal defects (open access ).
Review article on the application of Machine Learning potentials to crystal defects (open access here ).
Initial stages of silicon growth. Liquid atoms are colored according to a Machine-Learning defined structural parameter. Atoms in the crystalline phase are colored in gray.
Initial stages of silicon growth. Liquid atoms are colored according to a Machine-Learning defined structural parameter. Atoms in the crystalline phase are colored in gray.
Surface step configuration from a Molecular Dynamics simulation. Using information collected from atomistic simulations we formulated a quantitative thermodynamic theory of step properties on faceted crystalline surfaces. Our approach was substantiated by further developing a statistical-mechanical framework for the configurational contributions to step free energies.
Surface step configuration from a Molecular Dynamics simulation. Using information collected from atomistic simulations we formulated a quantitative thermodynamic theory of step properties on faceted crystalline surfaces. Our approach was substantiated by further developing a statistical-mechanical framework for the configurational contributions to step free energies.
Crystal growth of silicon using Molecular Dynamics. The last frame shows the dislocation network formed during the growth process. Atomistic simulations of this magnitude have been called "cross-scale" for their ability to represent mesoscale elements of the microstructure while accounting for each constituent atom individually.
Crystal growth of silicon using Molecular Dynamics. The last frame shows the dislocation network formed during the growth process. Atomistic simulations of this magnitude have been called "cross-scale" for their ability to represent mesoscale elements of the microstructure while accounting for each constituent atom individually.
Machine Learning assisted characterization of crystal growth of copper. Atoms are colored according a Machine Learning constructed parameter that encodes all atomic-scale mechanisms of crystallization.
Machine Learning assisted characterization of crystal growth of copper. Atoms are colored according a Machine Learning constructed parameter that encodes all atomic-scale mechanisms of crystallization.
Illustration of how interface-induced ordering of the liquid alters the local structure around crystallizing atoms.
Illustration of how interface-induced ordering of the liquid alters the local structure around crystallizing atoms.
Capillary wave fluctuations of a rough surface step.
Capillary wave fluctuations of a rough surface step.
Machine Learning based crystal-structure classification of a Molecular Dynamics simulation of copper crystallization.
Machine Learning based crystal-structure classification of a Molecular Dynamics simulation of copper crystallization.
Atomistic simulation of a screw dislocation where quantum effects on the mechanical properties metals were investigated.
Atomistic simulation of a screw dislocation where quantum effects on the mechanical properties metals were investigated.
Dynamical heterogeneity in supercooled liquids detected through Machine Learning methods.
Dynamical heterogeneity in supercooled liquids detected through Machine Learning methods.
Phase transitions between two different phases of the Σ5(310)[001] grain boundary in fcc crystals.
Phase transitions between two different phases of the Σ5(310)[001] grain boundary in fcc crystals.
Island nucleation and growth mechanism in silicon during solidification from the melt.
Island nucleation and growth mechanism in silicon during solidification from the melt.
Capillary wave fluctuation of a rough surface step.
Capillary wave fluctuation of a rough surface step.
Atomic resolution of surface step directions.
Atomic resolution of surface step directions.
Distribution of interfaces to which crystallizing atoms attach, showing a strong preference for (111) vicinals. Step–step separation distances for steps on (111) surfaces.
Distribution of interfaces to which crystallizing atoms attach, showing a strong preference for (111) vicinals. Step–step separation distances for steps on (111) surfaces.