MATERIALS DAY SPEAKERS
 Welcome & Introduction Carl V. Thompson Professor and Director Materials Research Laboratory |
 TBA Brian Storey Director, Accelerated Materials Design & Discovery TOYOTA Research Institute Abstract and Bio |
 Text and Data Mining for Material Synthesis Elsa Olivetti Professor Materials Science & Engineering,MIT Abstract & Bio
|
 Learning matter: Materials Design Through Atomistic Simulations and Machine Learning Rafael Gomez-Bombarelli Professor Materials Science & Engineering, MIT Abstract & Bio
|
 Advanced Chemical Development Through Process Intensification, Automation, and Machine Learning Klavs F. Jensen Professor Chemical Engineering, MIT Abstract & Bio |
 Elastic Strain Engineering for Unprecedented Properties Ju Li Professor Nuclear Science & Engineering, MIT Abstract & Bio
|
 Machine Learning in Optics: From Spectrum Reconstruction to Metasurface Design Juejun Hu Professor Materials Science & Engineering, MIT Abstract & Bio |
 TBA Asu Ozdaglar Professor & Department Head Electrical Engineering & Computer Science, MIT Abstract & Bio
|
Dr. Brian Storey
Director, Accelerated Materials Design & Discovery
Toyota Research Institute
Keynote:
TBA
Elsa Olivetti
Professor
Department of Materials Science & Engineering, MIT
Text and Data Mining for Material Synthesis
Abstract
Predictive materials modeling can provide properties of real and virtual compounds and will be available on demand, thereby enabling rapid iteration time in materials design. However, the allure (and necessity) of accelerated discovery that motivates computational materials design is diminished by the prevalent heuristic approaches to materials synthesis and optimization. This delay in moving from promising materials concept to validation, optimization, and scale‐up is a significant burden to commercialization. I will describe our work to extract information from peer reviewed academic literature across a range of inorganic solid state materials synthesis approaches. We have demonstrated not only the potential of the natural language processing (NLP) approach to assemble materials data from the literature, but we have also shown that one can develop hypotheses for what synthesis conditions drive a particular target material outcome using learning approaches.
Biography
Elsa Olivetti is the Atlantic Richfield Associate Professor of Energy Studies in the Department of Materials Science and Engineering. Her research focuses on improving the environmental and economic sustainability of materials using methods informed by materials economics, machine learning, and techno-economic analysis. She has received the NSF Career award for her experimental research focused on beneficial use of industrial waste materials.
Rafael Gomez-Bombarelli
Professor
Department of Materials Science & Engineering, MIT
Learning matter: Materials Design Through Atomistic Simulations and Machine Learning
Rafael Gomez-Bombarelli
Abstract
Machine learning tools combined with theoretical simulations can effectively accelerate the design of novel materials. Data-driven approaches can access the information embedded in years of experiments, perform rapid optimization of high-dimensional experimental conditions and design parameters, increase the accuracy and speed of physics-based simulations, or design new molecules and crystals automatically. By deploying automated atomistic simulations (molecular dynamics, electronic structure) to create interpretable, bottom-up representations of materials, and by using those as inputs to machine learning models, we can build effective and accurate predictors. Here, we will describe recent results and ongoing work in using machine learning as the connector between multiple scales of simulation and experiment in materials design. These include (i) high-throughput screening of molecular materials such as organic light emitting-diodes, small molecule battery electrolytes, or photovoltaics using electronic structure simulations; (ii) inverse design tools based on deep generative models for automatic chemical discovery; (iii) automated learning of all-atom and coarse-grained potentials for discovery of soft materials like ion-conducting polymers; (iv) graph-based representations that accurately predict and rationalize polymorphism in nanoporous zeolite materials.
Biography
Rafael (Rafa) Gomez-Bombarelli received his BSc, MSc (2006) and PhD (2011) in Chemistry from Universidad de Salamanca (Spain). After postdoctoral work at Heriot-Watt University (Edinburg, UK 2012-2014) and Harvard’s Department of Chemistry and Chemical Biology (2014-2016), Rafa co-founded Calculario, a materials discovery startup and joined Kyulux North America Inc., an OLED startup focused in thermally-assisted delayed fluorescence materials for display. Since January 2018 Rafa is an Assistant professor at MIT DMSE. There, he leads a 10-people group working at the interface between deep learning and atomistic simulations for materials design, with a strong focus on molecular and nanoporous materials, inverse design and dimensionality reduction, and supramolecular recognition. He was recently awarded the Google Faculty Research award (2019).
Klavs F. Jensen
Professor
Department of Chemical Engineering, MIT
Advanced Chemical Development Through Process Intensification, Automation, and Machine Learning
Abstract
Process intensification via continuous operation combined with automated optimization and screening techniques offer opportunities for faster development and more efficient manufacture of diverse chemical products. This presentation starts with advances in process intensification and green chemistry achieved through micro-reaction technology and continuous multistep synthesis (flow chemistry). Case studies include individual intensified reaction units as well as on-demand synthesis of common pharmaceuticals in a plug-and-play, manually reconfigurable, refrigerator-sized manufacturing platform. Next, advances in automated screening and optimization of chemical reactions are highlighted as methods to accelerate translation of laboratory discoveries to manufacturing. Examples include optimization of thermal- and photo-chemical reactions in 15 microliter droplets with respect to continuous process conditions (temperature, time, concentrations…) and discrete process choices (catalysts, solvents, bases …). Finally, machine learning of chemical information is applied to computer aided chemical synthesis - the planning of reaction paths to a given molecular target from purchasable starting materials. With expert user input, the synthesis plans are converted into recipes that are executed by an automated modular continuous flow platform configured by a robotic arm, which sets up the required unit operations. Examples of automatic continuous syntheses of pharmaceutical compounds and libraries illustrate the promise of this combined approach of machine learning, reaction engineering, and robotics.
Biography
Klavs F. Jensen is Warren K. Lewis Professor in Chemical Engineering and Materials Science and Engineering at the Massachusetts Institute of Technology. From 2007- July 2015 he was the Head of the Department of Chemical Engineering. His research interests include on-demand multistep synthesis, methods for automated synthesis, and machine learning techniques for chemical synthesis and interpreting large chemical data sets. He is a co-director of MIT's Pharma AI consortium that aims to bring machine learning technology into pharmaceutical discovery and development. Catalysis, chemical kinetics and transport phenomena are also topics of interest along with development of methods for predicting performance of reactive chemical systems. He chairs the Editorial Board for the new Royal Society of Chemistry Journal Reaction Chemistry and Engineering. He serves on advisory boards to universities, companies, professional societies, and governments. He is the recipient of several awards, including the Allan P. Colburn, Charles C.M. Stine, R.H. Wilhelm, W.H. Walker, and Founders Awards of the American Institute of Chemical Engineers, and the inaugural IUPAC-ThalesNano Prize in Flow Chemistry. Professor Jensen is a member of the US National Academy of Sciences, the US National Academy of Engineering as well as the American Academy of Arts and Science. He is a Fellow of the American Association for the Advancement of Science (AAAS), and the American Institute of Chemical Engineers, and the Royal Society of Chemistry.
Ju Li
Professor
Department of Nuclear Science & Engineering, MIT
Elastic Strain Engineering for Unprecedented Properties
Abstract
Strain Engineering uses strain to guide the interactions of material structures with electrons, photons, etc.1and control energy, mass and information flows. The success of Strained Silicon technology today harbingers what Strain Engineering may do for human civilization in the future, with potential breakthroughs in electronics, photonics, ferroics, superconductivity, catalysis, sensing, etc. In this talk I will give examples of exploiting the strain design space with machine learning. Homogenous and inhomogeneous elastic strain, bending, interlayer twist5and slip6lead to tunable, low-energy artificial atoms, artificial superlattices and pseudoheterostructures that can regulate quasiparticle motion. Strain also governs ferroelastic and band topology transitions in these materials. Lastly, we demonstrate production of kilogram-scale nanowires under large tensile elastic strain, that leads to improved superconductivity. By controlling the strain tensor and strain gradient statically or dynamically, one opens up a much larger parameter space -on par with alloying -for optimizing the functional properties of materials, which imparts a new meaning to Feynman’s statement “There's Plenty of Room at the Bottom”.
Biography
Professor of Materials Science and Engineering at MIT. His group investigates the mechanical, electrochemical and transport behaviors of materials as well as novel means of energy storage and conversion. Ju is a recipient of the 2005 Presidential Early Career Award for Scientists and Engineers, the 2006 Materials Research Society Outstanding Young Investigator Award, and the TR35 award from Technological Review. Ju was elected Fellow of the American Physical Society in 2014 and a Fellow of the Materials Research Society in 2017. In 2014/2018Thomson Reuters/Clarivate included him in the Highly Cited Researchers list in Materials Science category. In 2016 Ju Li co-founded one of the MIT Energy Initiative (MITEI) Low-Carbon Energy Centers, the Center for Materials in Energy and Extreme Environments
Juejun Hu
Associate Professor
Materials Science & Engineering, MIT
Machine Learning in Optics: From Spectrum Reconstruction to Metasurface Design
Abstract
Machine learning in optics: from spectrum reconstruction to metasurface design
In this talk, we discuss two examples in our work where machine learning is applied to solving problems in optics. In the first example, we developed a machine learning regularization method to reconstruct optical spectra based on data measured from a chip-scale optical spectrometer. We show that the algorithm can significantly boost signal-to-noise ratio in optical spectrum acquisition, enabling two-fold enhanced spectral resolution compared to the classical Rayleigh criterion. In the second example, we introduce a generic deep neural network approach to vastly improve the throughput and accuracy of free-form optical metasurface design. Examples of neural networks capable of producing on-demand designs for meta-atoms, meta-optical filters, and reconfigurable metasurfaces are demonstrated.
Biography
Juejun (JJ) Hu received the B.S. degree from Tsinghua University, China, in 2004, and the Ph.D. degree from Massachusetts Institute of Technology, Cambridge, MA, USA, in 2009, both in materials science and engineering. He is currently
the Merton C. Flemings Career Development Associate Professor at MITs Department of Materials Science and Engineering. His primary research interest is enhanced photon-matter interactions in nanophotonic structures. Prior to joining MIT, he was an Assistant Professor at the University of Delaware from 2010 to 2014. Hu has authored and coauthored more than 70 refereed journal publications since 2006 and has been awarded six U.S. patents. He has been recognized with the National Science Foundation Faculty Early Career Development award, the Robert L. Coble Award from the American Ceramic Society, the Gerard J. Mangone Young Scholars Award, the University of Delaware College of Engineering Outstanding Junior Faculty Member, the University of Delaware Excellence in Teaching Award, among others.
Asu Ozdaglar
Professor & Department Head
Department of Electrical Engineering & Computer Science, MIT
TBA
Biography
Since completing her graduate studies at MIT in 2003, Dr. Ozdaglar has been a faculty member in the Electrical Engineering and Computer Science Department at MIT. She is affiliated with the Laboratory for Information & Decision Systems (LIDS) and the Operations Research Center. Her research focuses on problems that arise in the analysis and optimization of large-scale dynamic multi-agent networked systems including communication networks, transportation networks, and social and economic networks.
Carl V. Thompson
Director
Materials Research Laboratory
Stavros Salapatas Professor of Materials Science & Engineering, MIT
Biography
Professor Thompson received an S.B. in Materials Science and Engineering from MIT and an S.M. and Ph.D. in Applied Physics from Harvard University. He was an IBM postdoctoral fellow in the Research Laboratory of Electronics at MIT in 1982 and joined the faculty of the Department of Materials Science and Engineering in 1983. Prof. Thompson was an SERC Fellow at Cambridge University from 1990 - 1991, a Humboldt Senior Scientist awardee at the Max Planck Institute for Metallurgy in Stuttgart Germany from 1997 - 1998, and a visiting scientist at the Institute for Applied Materials at the Karlsruhe Institute of Technology in 2012. He served as the president of the Materials Research Society in 1996 and served on the MRS council from 1991 - 1997. He has also been active in MIT’s programs in Singapore since 1998, including serving for 12 years as the co-Chair of the Program for Advanced Materials for Micro- and Nano-Systems of the Singapore-MIT Alliance. He was the Director of MIT’s Materials Processing Center from 2008 - 2017 and is currently the Director of MIT’s Materials Research Laboratory and Co-Director of the Skoltech Center for Electrochemical Energy Storage.
Professor Thompson worked briefly for U.S. Steel and General Electric and has been a consultant for over 30 companies including DEC, IBM and Intel and other microelectronics companies as well as smaller enterprises based on microelectromechanical devices and systems. He has also worked with a number of legal firms. His group has collaborated in research with IBM, Intel, AMD, TI, Motorola, and Sematech.
Professor Thompson’s research interests include structure evolution during processing of thin films and nanostructures, and incorporation of thin films and nanostructures into electronic, microelectromechanical and electrochemical devices and systems. Specific research topics include control of structure in single crystal and polycrystalline films and nanostructures; stress development, evolution and control in polycrystalline films and nanostructures; morphological evolution and stability of films and nanostructures, including templating of morphological instabilities as method for pattern formation; fabrication of thin film batteries using CMOS compatible materials and processes; and characterization and modeling of the reliability of IC interconnects and GaN-based HEMTs and LEDs.
Save
Save
Save
Save
Save