Anubhav Jain, profile picture
Uploaded byAnubhav Jain
PDF, PPTX540 views

High-throughput computation and machine learning methods applied to materials design

High-throughput computation and machine learning methods can be applied to materials design problems at scale. Density functional theory (DFT) allows modeling of materials at the quantum mechanical level but large computational resources are required. "High-throughput DFT" uses automation, parallelization across supercomputers, and data mining approaches to rapidly screen millions of potential new materials in silico before experimental validation. This helps address the challenge of discovering new materials for applications like energy technologies by searching the vast space of possible compositions and structures more efficiently than traditional experimentation alone.

Embed presentation

Download as PDF, PPTX
High-throughput computation and machine learning methods applied to materials design Anubhav Jain Energy Technologies Area Lawrence Berkeley National Laboratory Berkeley, CA Citrine Informatics Talk July 3, 2018 Slides (already) posted to hackingmaterials.lbl.gov
New materials discovery for devices is difficult •  Novel materials with enhanced performance characteristics could make a big dent in sustainability, scalability, and cost •  In practice, we tend to re-use the same fundamental materials for decades –  solar power w/Si since 1950s –  graphite/LiCoO2 (basis of today’s Li battery electrodes) since 1990 –  Bi2Te3 and PbTe thermoelectrics first studied ~1910 •  Although there are lots of improvements to manufacturing, microstructure, etc., there not many new basic compositions •  Why is discovering better materials such a challenge? 2
3 A material is defined at multiple length scales – stick to the fundamental scale for now
4 A material is defined at multiple length scales – stick to the fundamental scale for now
5 Atoms in a box – the materials universe is huge! •  Bag of 30 atoms •  Each atom is one of 50 elements •  Arrange on 10x10x10 lattice •  Over 10108 possibilities! –  more than grains of sand on all beaches (1021) –  more than number of atoms in universe (1080)
6 Finding the right material is like “finding a needle in a haystack”
What constrains traditional experimentation? 7 “[The Chevrel] discovery resulted from a lot of unsuccessful experiments of Mg ions insertion into well-known hosts for Li+ ions insertion, as well as from the thorough literature analysis concerning the possibility of divalent ions intercalation into inorganic materials.” -Aurbach group, on discovery of Chevrel cathode for multivalent (e.g., Mg2+) batteries Levi, Levi, Chasid, Aurbach J. Electroceramics (2009)
Outline 8 ①  From quantum mechanics to density functional theory (DFT) ②  “High-throughput” DFT ③  Data mining approaches to materials design
The basis of density functional theory is quantum mechanics 9 −!2 2m ∇2 Ψ(r)+V (r)Ψ(r) = EΨ(r) Schrödinger equation describes all the properties of a system through the wavefunction: Time-independent, non-relativistic Schrödinger equation
•  There aren’t too many real situations where we can get a closed solution to the Schrödinger equation •  Let’s pretend we want to approach things numerically for 1000 electrons –  There are ~500,000 electron-electron interactions to worry about. –  Even storing the wavefunction would take ~101000 GB! •  Discretize the x,y,z, position of each electron into a 1000- element grid = 1 billion positions per electron •  Need the wavefunction output (real + complex part) for each combination of all electron positions, i.e. 1E9 ^ (1000) * 2, or 2E9000 values •  even at 1 byte per wavefunction value (low resolution), you have about 2E1000 GB needed needed to store the wavefunction! 10 The wave function is formidable
Dirac summarized it best … 11 “The underlying physical laws necessary for the mathematical theory of a large part of physics and the whole of chemistry are thus completely known, and the difficulty is only that the exact application of these laws leads to equations much too complicated to be soluble.” “It therefore becomes desirable that approximate practical methods of applying quantum mechanics should be developed, which can lead to an explanation of the main features of complex atomic systems without too much computation.”
What is density functional theory (DFT)? 12 DFT theory: •  replaces many-body interactions with a mean field interaction that reproduces the same charge density as the original formulation •  proves that, given the correct charge density, it in principle possible to compute all ground state properties of quantum mechanics exactly So, (for the ground state properties) we went from many-body wavefunctions to mean field charge density! This was worthy of the 1998 Nobel Prize. DFT practice: •  accuracy depends on the choice of (some) parameters, the type of material, the property to be studied, and whether the simulated system (crystal) is a good approximation of reality. e– e– e– e– e– e–
How does one use DFT to design new materials? 13 A. Jain, Y. Shin, and K. A. Persson, Nat. Rev. Mater. 1, 15004 (2016).
How accurate is DFT in practice? 14 Shown are typical DFT results for (i) Li battery voltages, (ii) electronic band gaps, and (iii) bulk modulus (i) (ii) (iii) (i) V. L. Chevrier, S. P. Ong, R. Armiento, M. K. Y. Chan, and G. Ceder, Phys. Rev. B 82, 075122 (2010). (ii) M. Chan and G. Ceder, Phys. Rev. Lett. 105, 196403 (2010). (iii) M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, M. Sluiter, C. K. Ande, S. Van Der Zwaag, J. J. Plata, C. Toher, S. Curtarolo, G. Ceder, K.A. Persson, and M. Asta, Sci. Data 2, 150009 (2015). battery voltages band gaps bulk modulus
•  System size is essentially limited to ~1000 atoms –  many important materials phenomena simply do not occur at this length scale •  Certain materials, such as those with strong electron correlation, remain difficult to model accurately •  Certain properties, including excited state properties such as band gap, remain difficult to model accurately •  These are all active areas of research and improvement to the theory, and the situation is improving on all fronts 15 Limitations of density functional theory
Outline 16 ①  From quantum mechanics to density functional theory (DFT) ②  “High-throughput” DFT ③  Data mining approaches to materials design
17 From a needle in a haystack to …
18 … hiring an army to search through the haystack
High-throughput DFT: a key idea 19 Automate the DFT procedure Supercomputing Power FireWorks Software for programming general computational workflows that can be scaled across large supercomputers. NERSC Supercomputing center, processor count is ~100,000 desktop machines. Other centers are also viable. High-throughput materials screening G. Ceder & K.A. Persson, Scientific American (2015)
•  The answer is “it really varies a lot” –  how big / complicated are the materials you are modeling? –  how complex / expensive are the properties you are modeling? •  Ballpark numbers: –  Low range: optimize structure of ~3-atom compounds •  time to do a million materials ~ 10 million CPU-hours –  Medium range: bulk modulus of ~50 atom compounds •  time to do a million materials ~ 2 billion CPU hours •  The largest CPU allocations from the DOE are typically in the order of ~100 million CPU-hours 20 How much computer time is needed for high-throughput DFT?
Examples of (early) high-throughput studies 21 Application Researcher Search space Candidates Hit rate Scintillators Klintenberg et al. 22,000 136 1/160 Curtarolo et al. 11,893 ? ? Topological insulators Klintenberg et al. 60,000 17 1/3500 Curtarolo et al. 15,000 28 1/535 High TC superconductors Klintenberg et al. 60,000 139 1/430 Thermoelectrics – ICSD - Half Heusler systems - Half Heusler best ZT Curtarolo et al. 2,500 80,000 80,000 20 75 18 1/125 1/1055 1/4400 1-photon water splitting Jacobsen et al. 19,000 20 1/950 2-photon water splitting Jacobsen et al. 19,000 12 1/1585 Transparent shields Jacobsen et al. 19,000 8 1/2375 Hg adsorbers Bligaard et al. 5,581 14 1/400 HER catalysts Greeley et al. 756 1 1/756* Li ion battery cathodes Ceder et al. 20,000 4 1/5000* Entries marked with * have experimentally verified the candidates. See also: Curtarolo et al., Nature Materials 12 (2013) 191–201.
Computations predict, experiments confirm 22 Sidorenkite-based Li-ion battery cathodes LED phosphors YCuTe2 thermoelectrics Wang, Z., Ha, J., Kim, Y. H., Im, W. Bin, McKittrick, J. & Ong, S. P. Mining Unexplored Chemistries for Phosphors for High-Color-Quality White-Light-Emitting Diodes. Joule 2, 914–926 (2018). Chen, H.; Hao, Q.; Zivkovic, O.; Hautier, G.; Du, L.-S.; Tang, Y.; Hu, Y.-Y.; Ma, X.; Grey, C. P.; Ceder, G. Sidorenkite (Na3MnPO4CO3): A New Intercalation Cathode Material for Na-Ion Batteries, Chem. Mater., 2013 Aydemir, U; Pohls, J-H; Zhu, H; Hautier, G; Bajaj, S; Gibbs, ZM; Chen, W; Li, G; Broberg, D; White, MA; Asta, M; Persson, K; Ceder, G; Jain, A; Snyder, GJ. Thermoelectric Properties of Intrinsically Doped YCuTe2 with CuTe4-based Layered Structure. J. Mat. Chem C, 2016 More examples here: A. Jain, Y. Shin, and K. A. Persson, Nat. Rev. Mater. 1, 15004 (2016).
With HT-DFT, we can generate data rapidly – what to do next? 23 M. de Jong, W. Chen, H. Geerlings, M. Asta, and K. A. Persson, Sci. Data, 2015, 2, 150053.! M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, M. Sluiter, C. K. Ande, S. Van Der Zwaag, J. J. Plata, C. Toher, S. Curtarolo, G. Ceder, K. a Persson, and M. Asta, Sci. Data, 2015, 2, 150009.! >4500 elastic tensors >900 piezoelectric tensors >48000 Seebeck coefficients + cRTA transport Ricci, Chen, Aydemir, Snyder, Rignanese, Jain, & Hautier (in submission)!
With HT-DFT, we can generate data rapidly – what to do next? 24 M. de Jong, W. Chen, H. Geerlings, M. Asta, and K. A. Persson, Sci. Data, 2015, 2, 150053.! M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, M. Sluiter, C. K. Ande, S. Van Der Zwaag, J. J. Plata, C. Toher, S. Curtarolo, G. Ceder, K. a Persson, and M. Asta, Sci. Data, 2015, 2, 150009.! >4500 elastic tensors >900 piezoelectric tensors >48000 Seebeck coefficients + cRTA transport Ricci, Chen, Aydemir, Snyder, Rignanese, Jain, & Hautier (in submission)! Goal: make it easy to generate comparable data sets on your own
A “black-box” view of performing a calculation 25 “something”! Results!! researcher! What is the GGA-PBE elastic tensor of GaAs?
Unfortunately, the inside of the “black box” is usually tedious and “low-level” 26 lots of tedious, low-level work…! Results!! researcher! What is the GGA-PBE elastic tensor of GaAs? Input file flags SLURM format how to fix ZPOTRF? q  set up the structure coordinates q  write input files, double-check all the flags q  copy to supercomputer q  submit job to queue q  deal with supercomputer headaches q  monitor job q  fix error jobs, resubmit to queue, wait again q  repeat process for subsequent calculations in workflow q  parse output files to obtain results q  copy and organize results, e.g., into Excel
What would be a better way? 27 “something”! Results!! researcher! What is the GGA-PBE elastic tensor of GaAs?
What would be a better way? 28 Results!! researcher! What is the GGA-PBE elastic tensor of GaAs? Workflows to run! q  band structure! q  surface energies! ü  elastic tensor! q  Raman spectrum! q  QH thermal expansion!
Ideally the method should scale to millions of calculations 29 Results!! researcher! Start with all binary oxides, replace O->S, run several different properties Workflows to run! ü  band structure! ü  surface energies! ü  elastic tensor! q  Raman spectrum! q  QH thermal expansion! q  spin-orbit coupling!
Atomate tries make it easy, automatic, and flexible to generate data with existing simulation packages 30 Results!! researcher! Run many different properties of many different materials!
Each simulation procedure translates high-level instructions into a series of low-level tasks 31 quickly and automatically translate PI-style (minimal) specifications into well-defined FireWorks workflows What is the GGA-PBE elastic tensor of GaAs? M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, et al., Charting the complete elastic properties of inorganic crystalline compounds, Sci. Data. 2 (2015).
Atomate contains a library of simulation procedures 32 VASP-based •  band structure •  spin-orbit coupling •  hybrid functional calcs •  elastic tensor •  piezoelectric tensor •  Raman spectra •  NEB •  GIBBS method •  QH thermal expansion •  AIMD •  ferroelectric •  surface adsorption •  work functions Other •  BoltzTraP •  FEFF method •  LAMMPS MD Mathew, K. et al Atomate: A high-level interface to generate, execute, and analyze computational materials science workflows, Comput. Mater. Sci. 139 (2017) 140–152.
33 Full operation diagram job 1 job 2 job 3 job 4 structure! workflow! database of all workflows! automatically submit + execute!output files + database!
Atomate thus encodes and standardizes knowledge about running various kinds of simulations from domain experts 34 K. Mathew J. Montoya S. Dwaraknath A. Faghaninia All past and present knowledge, from everyone in the group, everyone previously in the group, and our collaborators, about how to run calculations M. Aykol S.P. Ong B. Bocklund T. Smidt H. Tang I.H. Chu M. Horton J. Dagdalen B. Wood Z.K. Liu J. Neaton K. Persson A. Jain +
Outline 35 ①  From quantum mechanics to density functional theory (DFT) ②  “High-throughput” DFT ③  Data mining approaches to materials design
36 From a needle in a haystack to …
37 … hiring an army to search through the haystack to …
38 Armies with metal detectors
Can we build a general optimizer? 39 Generalizable forward solver Supercomputing Power Statistical optimization FireWorks NERSC Various optimization libraries (Figure: J. Mueller)
Rocketsled: Automatic materials screening that selects materials to compute AND submits them to supercomputer 40 screening space of ~20,000 potential ABX3 perovskite combinations as water splitting materials – precomputed in DFT by different group if a machine learning algorithm was in charge of picking the next compound based on past data, how efficient would it be?
Machine learning: the big problem in my view is connecting data to ML algorithms through features 41 Lots of data on complex objects that you want to interrelate Clustering, Regression, Feature extraction, Model-building, etc. Well developed data-mining routines that work only on numbers (ideally ones with high relevance to your problem) Need to transform materials science objects into a set of physically relevant numerical data (“features” or “descriptors”)
Goal of matminer: connect materials data with data mining algorithms and data visualization libraries 42 Ward, L. et al. Matminer: An open source toolkit for materials data mining. Comput. Mater. Sci. 152, 60–69 (2018).
>40 featurizer classes can generate thousands of potential descriptors 43 Matminer contains a library of descriptors for various materials science entities feat = EwaldEnergy([options]) y = feat.featurize([input_data]) •  compatible with scikit-learn pipelining •  automatically deploy multiprocessing to parallelize over data •  include citations to methodology papers
44 Interactive Jupyter notebooks demonstrate use cases https://github.com/hackingmaterials/matminer_examples! Many examples available: •  Retrieving data from various databases •  Predicting bulk / shear modulus •  Predicting formation energies: •  from composition alone •  with Voronoi-based structure features included •  with Coulomb matrix and Orbital Field matrix descriptors (reproducing previous studies in the literature) •  Making interactive visualizations •  Creating an ML pipeline
Defining local order parameters for various environments 45 Use a given local order parameter with a threshold for motif recognition: If qtet > qthresh, then motif is tetrahedron. Else not (too much) a tetrahedron. Tetrahedral order parameter, qtet, [1]: [1] Zimmermann et al., J. Am. Chem. Soc., 2017, 10.1021/jacs.5b08098
We have now developed mathematical order parameters for various types of local environments 46
How well do these work? 47 1. Order parameters clearly distinguish different environments even after thermal distortion 2. Work well in applications (defect site finding, diffusion characterization) [1] Zimmermann et al., Frontiers of Materials, 2017, doi: 10.3389/fmats.2017.00034
48 Can cluster crystal structures by “local environment similarity”
Results on MP web site, e.g. for BCC-like structures 49 https://www.materialsproject.org/materials/mp-91/! Target: W similar structures (distance near 0) Cs3Sb! TiGaFeCo! CeMg2Cu!
50 Text mining: learning from scientific abstracts Matstract corpus Unlabeled data Data labels Feature engineering Text cleaning Tokenization POS tag labels Word embeddings (word2vec) Text processing Hand crafted features Supervised learning Neural network (LSTM) Logistic regression Train/test sets Named Entities Named Entities “Learning” what a scientific study is about from >2 million materials science abstracts
51 Application: a revised materials search engine Auto-generated summaries of materials based on text mining
52 Application: materials compositions of interest … A search for “unconventional” (i.e., non-perovskite) ferroelectrics
53 •  The Materials Project today mostly compiles fundamental simulation output data •  Many users don’t really know what to “do” with this data –  i.e., they would be interested in lattice thermal conductivity, but don’t know they can get there using MP data + other models •  How can we decorate MP database with additional properties and clearly show how we got there? How to connect different materials properties together? “Constitutive relations”
54 A materials map / network / atlas •  What we need is a “connected property network” •  This is a graph in which nodes are materials properties and edges are relationships between those properties •  Given a set of known properties, e.g. simulation data, one can easily figure out what are all the derived engineering properties one can get Starting with the three properties in blue, one can derive many additional properties using one (orange), two (purple), or three (green) physical models.! ! The value of computations are not only in the direct simulation outputs!!
55 Materials Atlas
56 We are now feeding the “atlas” into the Materials Project to derive new data
•  High-throughput density functional theory, materials databases, and machine learning are a new set of tools for doing materials science •  We are developing many methods and software implementations to try to advance the field •  If you are interested, give the software a try! 57 Conclusions Quantum mechanics Density functional theory High-throughput DFT e– e– e– e– e– e– Materials databases Machine learning
•  Atomate –  K Matthew (project lead) & team •  Structure order parameters –  N. Zimmermann (project lead) & team •  Rocketsled –  A. Dunn, J. Brenneck •  Matminer –  L. Ward (project lead, U. Chicago) & team •  Text mining –  V. Tshitoyan, J. Dagdelen, L. Weston •  Propnet –  M. K. Horton, D. Mrdjenovich •  All that provided feedback & contributed code to open-source software efforts! •  Funding: DOE-BES (Early Career + Materials Project Center) •  Computing: NERSC 58 Thank you!

More Related Content

PDF
Methods, tools, and examples (Part II): High-throughput computation and machi...
byAnubhav Jain
 
PDF
Lecture: Interatomic Potentials Enabled by Machine Learning
byDanielSchwalbeKoda
 
PPT
Organic- Inorganic Perovskite Solar Cell
byRajan K. Singh
 
PDF
A DFT & TDDFT Study of Hybrid Halide Perovskite Quantum Dots
byAthanasiosKoliogiorg
 
PDF
Software tools for calculating materials properties in high-throughput (pymat...
byAnubhav Jain
 
PDF
Wien2k getting started
byABDERRAHMANE REGGAD
 
PPTX
Perovskite solar cells
bykidist w.
 
PDF
Plasma physics by Dr. imran aziz
byDr.imran aziz
 
Methods, tools, and examples (Part II): High-throughput computation and machi...
byAnubhav Jain
 
Lecture: Interatomic Potentials Enabled by Machine Learning
byDanielSchwalbeKoda
 
Organic- Inorganic Perovskite Solar Cell
byRajan K. Singh
 
A DFT & TDDFT Study of Hybrid Halide Perovskite Quantum Dots
byAthanasiosKoliogiorg
 
Software tools for calculating materials properties in high-throughput (pymat...
byAnubhav Jain
 
Wien2k getting started
byABDERRAHMANE REGGAD
 
Perovskite solar cells
bykidist w.
 
Plasma physics by Dr. imran aziz
byDr.imran aziz
 

What's hot

PDF
Computational materials design with high-throughput and machine learning methods
byAnubhav Jain
 
PDF
Graphs, Environments, and Machine Learning for Materials Science
byaimsnist
 
PPTX
Perovskite Solar Cell
byHimanshu Dixit
 
PDF
HiPIMS: technology, physics and thin film applications.
byInstituto Nacional de Engenharia de Superfícies
 
PDF
Quantum Computation for Predicting Electron and Phonon Properties of Solids
byKAMAL CHOUDHARY
 
PDF
Electronic structure of strongly correlated materials
byABDERRAHMANE REGGAD
 
PPTX
Presentation on Perovskite Solar Cell.
byFazlul K. Palash
 
PDF
Exp zncoo
byyahya291985
 
ODP
CERN-LHC presentation
byAlfonso de la Fuente Ruiz
 
PDF
Solar Cells Lecture 5: Organic Photovoltaics
byTuong Do
 
PPT
ION IMPLANTATION
byAJAL A J
 
PPTX
The Large Hadron Collider
bySree Bhargava Kothapalli
 
PPTX
Methods available in WIEN2k for the treatment of exchange and correlation ef...
byABDERRAHMANE REGGAD
 
PDF
UCSD NANO106 - 12 - X-ray diffraction
byUniversity of California, San Diego
 
PPTX
Perovskite solar cells
bysudhinsukumaran
 
PPT
Physics M5 Nature of radioactive decay
byeLearningJa
 
PPTX
Tandem solar cell slide share
bykamatlab
 
PPTX
Perovskite Solar Cells - an Introduction
byDawn John Mullassery
 
PPTX
Organic photovoltaic cells : OPV
byMalak Talbi
 
PPTX
Synthesis of graphene
byAPRATIM KHANDELWAL
 
Computational materials design with high-throughput and machine learning methods
byAnubhav Jain
 
Graphs, Environments, and Machine Learning for Materials Science
byaimsnist
 
Perovskite Solar Cell
byHimanshu Dixit
 
HiPIMS: technology, physics and thin film applications.
byInstituto Nacional de Engenharia de Superfícies
 
Quantum Computation for Predicting Electron and Phonon Properties of Solids
byKAMAL CHOUDHARY
 
Electronic structure of strongly correlated materials
byABDERRAHMANE REGGAD
 
Presentation on Perovskite Solar Cell.
byFazlul K. Palash
 
Exp zncoo
byyahya291985
 
CERN-LHC presentation
byAlfonso de la Fuente Ruiz
 
Solar Cells Lecture 5: Organic Photovoltaics
byTuong Do
 
ION IMPLANTATION
byAJAL A J
 
The Large Hadron Collider
bySree Bhargava Kothapalli
 
Methods available in WIEN2k for the treatment of exchange and correlation ef...
byABDERRAHMANE REGGAD
 
UCSD NANO106 - 12 - X-ray diffraction
byUniversity of California, San Diego
 
Perovskite solar cells
bysudhinsukumaran
 
Physics M5 Nature of radioactive decay
byeLearningJa
 
Tandem solar cell slide share
bykamatlab
 
Perovskite Solar Cells - an Introduction
byDawn John Mullassery
 
Organic photovoltaic cells : OPV
byMalak Talbi
 
Synthesis of graphene
byAPRATIM KHANDELWAL
 

Similar to High-throughput computation and machine learning methods applied to materials design

PDF
Combining density functional theory calculations, supercomputing, and data-dr...
byAnubhav Jain
 
PDF
A*STAR Webinar on The AI Revolution in Materials Science
byUniversity of California, San Diego
 
PDF
Density functional theory calculations and data mining for new thermoelectric...
byAnubhav Jain
 
PDF
The Materials Project: An Electronic Structure Database for Community-Based M...
byAnubhav Jain
 
PDF
Materials Modelling: From theory to solar cells (Lecture 1)
bycdtpv
 
PDF
Software tools, crystal descriptors, and machine learning applied to material...
byAnubhav Jain
 
PDF
Combining density functional theory calculations, supercomputing, and data-dr...
byAnubhav Jain
 
PDF
Efficient methods for accurately calculating thermoelectric properties – elec...
byAnubhav Jain
 
PDF
Materials discovery through theory, computation, and machine learning
byAnubhav Jain
 
PDF
Introduction (Part I): High-throughput computation and machine learning appli...
byAnubhav Jain
 
PDF
Discovering advanced materials for energy applications (with high-throughput ...
byAnubhav Jain
 
PDF
Combining density functional theory calculations, supercomputing, and data-dr...
byAnubhav Jain
 
PPTX
Density Functional Theory imtroduction.pptx
bysaduddeen99
 
PDF
The Materials Project and computational materials discovery
byAnubhav Jain
 
PDF
ICME Workshop Jul 2014 - The Materials Project
byUniversity of California, San Diego
 
PDF
Smart Metrics for High Performance Material Design
byKAMAL CHOUDHARY
 
PDF
Prediction and Experimental Validation of New Bulk Thermoelectrics Compositio...
byAnubhav Jain
 
PDF
Overview of accelerated materials design efforts in the Hacking Materials res...
byAnubhav Jain
 
PDF
Smart Metrics for High Performance Material Design
byaimsnist
 
PDF
Software tools for data-driven research and their application to thermoelectr...
byAnubhav Jain
 
Combining density functional theory calculations, supercomputing, and data-dr...
byAnubhav Jain
 
A*STAR Webinar on The AI Revolution in Materials Science
byUniversity of California, San Diego
 
Density functional theory calculations and data mining for new thermoelectric...
byAnubhav Jain
 
The Materials Project: An Electronic Structure Database for Community-Based M...
byAnubhav Jain
 
Materials Modelling: From theory to solar cells (Lecture 1)
bycdtpv
 
Software tools, crystal descriptors, and machine learning applied to material...
byAnubhav Jain
 
Combining density functional theory calculations, supercomputing, and data-dr...
byAnubhav Jain
 
Efficient methods for accurately calculating thermoelectric properties – elec...
byAnubhav Jain
 
Materials discovery through theory, computation, and machine learning
byAnubhav Jain
 
Introduction (Part I): High-throughput computation and machine learning appli...
byAnubhav Jain
 
Discovering advanced materials for energy applications (with high-throughput ...
byAnubhav Jain
 
Combining density functional theory calculations, supercomputing, and data-dr...
byAnubhav Jain
 
Density Functional Theory imtroduction.pptx
bysaduddeen99
 
The Materials Project and computational materials discovery
byAnubhav Jain
 
ICME Workshop Jul 2014 - The Materials Project
byUniversity of California, San Diego
 
Smart Metrics for High Performance Material Design
byKAMAL CHOUDHARY
 
Prediction and Experimental Validation of New Bulk Thermoelectrics Compositio...
byAnubhav Jain
 
Overview of accelerated materials design efforts in the Hacking Materials res...
byAnubhav Jain
 
Smart Metrics for High Performance Material Design
byaimsnist
 
Software tools for data-driven research and their application to thermoelectr...
byAnubhav Jain
 

More from Anubhav Jain

PDF
Machine Learning for Catalyst Design
byAnubhav Jain
 
PDF
Discovering new functional materials for clean energy and beyond using high-t...
byAnubhav Jain
 
PDF
Available methods for predicting materials synthesizability using computation...
byAnubhav Jain
 
PDF
Predicting the Synthesizability of Inorganic Materials: Convex Hulls, Literat...
byAnubhav Jain
 
PDF
Applications of Large Language Models in Materials Discovery and Design
byAnubhav Jain
 
PDF
Natural Language Processing for Data Extraction and Synthesizability Predicti...
byAnubhav Jain
 
PDF
Discovering advanced materials for energy applications: theory, high-throughp...
byAnubhav Jain
 
PDF
Discovering and Exploring New Materials through the Materials Project
byAnubhav Jain
 
PDF
Research opportunities in materials design using AI/ML
byAnubhav Jain
 
PDF
An AI-driven closed-loop facility for materials synthesis
byAnubhav Jain
 
PDF
Perspectives on chemical composition and crystal structure representations fr...
byAnubhav Jain
 
PDF
A Career at a U.S. National Lab: Perspective from a Mid-Career Scientist
byAnubhav Jain
 
PDF
DuraMat CO1 Central Data Resource: How it started, how it’s going …
byAnubhav Jain
 
PDF
Accelerating materials discovery with big data and machine learning
byAnubhav Jain
 
PDF
The Materials Project
byAnubhav Jain
 
PDF
Natural language processing for extracting synthesis recipes and applications...
byAnubhav Jain
 
PDF
Evaluating Chemical Composition and Crystal Structure Representations using t...
byAnubhav Jain
 
PDF
Accelerating New Materials Design with Supercomputing and Machine Learning
byAnubhav Jain
 
PDF
Best practices for DuraMat software dissemination
byAnubhav Jain
 
PDF
Best practices for DuraMat software dissemination
byAnubhav Jain
 
Machine Learning for Catalyst Design
byAnubhav Jain
 
Discovering new functional materials for clean energy and beyond using high-t...
byAnubhav Jain
 
Available methods for predicting materials synthesizability using computation...
byAnubhav Jain
 
Predicting the Synthesizability of Inorganic Materials: Convex Hulls, Literat...
byAnubhav Jain
 
Applications of Large Language Models in Materials Discovery and Design
byAnubhav Jain
 
Natural Language Processing for Data Extraction and Synthesizability Predicti...
byAnubhav Jain
 
Discovering advanced materials for energy applications: theory, high-throughp...
byAnubhav Jain
 
Discovering and Exploring New Materials through the Materials Project
byAnubhav Jain
 
Research opportunities in materials design using AI/ML
byAnubhav Jain
 
An AI-driven closed-loop facility for materials synthesis
byAnubhav Jain
 
Perspectives on chemical composition and crystal structure representations fr...
byAnubhav Jain
 
A Career at a U.S. National Lab: Perspective from a Mid-Career Scientist
byAnubhav Jain
 
DuraMat CO1 Central Data Resource: How it started, how it’s going …
byAnubhav Jain
 
Accelerating materials discovery with big data and machine learning
byAnubhav Jain
 
The Materials Project
byAnubhav Jain
 
Natural language processing for extracting synthesis recipes and applications...
byAnubhav Jain
 
Evaluating Chemical Composition and Crystal Structure Representations using t...
byAnubhav Jain
 
Accelerating New Materials Design with Supercomputing and Machine Learning
byAnubhav Jain
 
Best practices for DuraMat software dissemination
byAnubhav Jain
 
Best practices for DuraMat software dissemination
byAnubhav Jain
 

Recently uploaded

PDF
Deep-Behavioral Phenotyping in Systems Neuroscience for Functional Atlasing a...
byAna Luísa Pinho
 
PPTX
Polarity - Polar and non-polar, electronegativity and molecular geometry
byJovenAlvarado
 
PDF
Understanding the Metabolism of Proteolysis Targeting Chimeras (PROTACs): The...
byVirat Ojha
 
PDF
12th Bio NCERT Important Questions @irfanullah_ mehar.pdf
byWorld of Wisdom
 
PPTX
Photosynthesis (Light dependent reaction and Calvin cycle).pptx
byRashmiMG2
 
PPTX
Cohesin And Condesin In Chromsomal Segregation.pptx
bykeesarisuchith
 
PPTX
CABG.pptx Coronary..heart surgery... Bypass grafting
bypoulomipal936
 
PDF
Presentation.pdf for Biology students cell cycle
byfarahnasr15
 
PPTX
SCIENCE Quiz Bee 2025-School - Based.pptx
byMaryJean78
 
PPTX
Data Collection and data types Chapter 2.pptx
byModern College Shivajinagar, Pune-5
 
PDF
20251030 Jonh Kemp Energy system transformations and energy security
byddaher1
 
PDF
Sciences of Europe No 180 (2025)
bySciences of Europe
 
PPTX
Sci-Tech Filler - Quiz for Coeus 2023.pptx
bynvkeerthana2002
 
PPTX
Tissue Expansion , a technique used in plastic surgery
byrandil1
 
PDF
Star Constollations and Quasars.pdf Files
bycastilloronelparro22
 
PDF
HND Networking and Cloud Computing (NCC).pdf
byssuser8dcccc
 
PPTX
Sedimentary basins by Team Maverick.pptx
byolonabasit
 
PPTX
Ianalumab_ITP_Journal_Presentation_FULL.pptx
bysandiponroy342
 
PPTX
PRE NATAL GROWTH OF MAXILLA in orthodontics.pptx
byharanip03
 
PPTX
ALLOSTERIC INHIBITION : ITS CHARACTERISTICS, STRUCTURE AND MECHANISMS
byaubrey02judd
 
Deep-Behavioral Phenotyping in Systems Neuroscience for Functional Atlasing a...
byAna Luísa Pinho
 
Polarity - Polar and non-polar, electronegativity and molecular geometry
byJovenAlvarado
 
Understanding the Metabolism of Proteolysis Targeting Chimeras (PROTACs): The...
byVirat Ojha
 
12th Bio NCERT Important Questions @irfanullah_ mehar.pdf
byWorld of Wisdom
 
Photosynthesis (Light dependent reaction and Calvin cycle).pptx
byRashmiMG2
 
Cohesin And Condesin In Chromsomal Segregation.pptx
bykeesarisuchith
 
CABG.pptx Coronary..heart surgery... Bypass grafting
bypoulomipal936
 
Presentation.pdf for Biology students cell cycle
byfarahnasr15
 
SCIENCE Quiz Bee 2025-School - Based.pptx
byMaryJean78
 
Data Collection and data types Chapter 2.pptx
byModern College Shivajinagar, Pune-5
 
20251030 Jonh Kemp Energy system transformations and energy security
byddaher1
 
Sciences of Europe No 180 (2025)
bySciences of Europe
 
Sci-Tech Filler - Quiz for Coeus 2023.pptx
bynvkeerthana2002
 
Tissue Expansion , a technique used in plastic surgery
byrandil1
 
Star Constollations and Quasars.pdf Files
bycastilloronelparro22
 
HND Networking and Cloud Computing (NCC).pdf
byssuser8dcccc
 
Sedimentary basins by Team Maverick.pptx
byolonabasit
 
Ianalumab_ITP_Journal_Presentation_FULL.pptx
bysandiponroy342
 
PRE NATAL GROWTH OF MAXILLA in orthodontics.pptx
byharanip03
 
ALLOSTERIC INHIBITION : ITS CHARACTERISTICS, STRUCTURE AND MECHANISMS
byaubrey02judd
 

High-throughput computation and machine learning methods applied to materials design

  • 1.
    High-throughput computation andmachine learning methods applied to materials design Anubhav Jain Energy Technologies Area Lawrence Berkeley National Laboratory Berkeley, CA Citrine Informatics Talk July 3, 2018 Slides (already) posted to hackingmaterials.lbl.gov
  • 2.
    New materials discoveryfor devices is difficult •  Novel materials with enhanced performance characteristics could make a big dent in sustainability, scalability, and cost •  In practice, we tend to re-use the same fundamental materials for decades –  solar power w/Si since 1950s –  graphite/LiCoO2 (basis of today’s Li battery electrodes) since 1990 –  Bi2Te3 and PbTe thermoelectrics first studied ~1910 •  Although there are lots of improvements to manufacturing, microstructure, etc., there not many new basic compositions •  Why is discovering better materials such a challenge? 2
  • 3.
    3 A material isdefined at multiple length scales – stick to the fundamental scale for now
  • 4.
    4 A material isdefined at multiple length scales – stick to the fundamental scale for now
  • 5.
    5 Atoms in abox – the materials universe is huge! •  Bag of 30 atoms •  Each atom is one of 50 elements •  Arrange on 10x10x10 lattice •  Over 10108 possibilities! –  more than grains of sand on all beaches (1021) –  more than number of atoms in universe (1080)
  • 6.
    6 Finding the rightmaterial is like “finding a needle in a haystack”
  • 7.
    What constrains traditionalexperimentation? 7 “[The Chevrel] discovery resulted from a lot of unsuccessful experiments of Mg ions insertion into well-known hosts for Li+ ions insertion, as well as from the thorough literature analysis concerning the possibility of divalent ions intercalation into inorganic materials.” -Aurbach group, on discovery of Chevrel cathode for multivalent (e.g., Mg2+) batteries Levi, Levi, Chasid, Aurbach J. Electroceramics (2009)
  • 8.
    Outline 8 ①  From quantummechanics to density functional theory (DFT) ②  “High-throughput” DFT ③  Data mining approaches to materials design
  • 9.
    The basis ofdensity functional theory is quantum mechanics 9 −!2 2m ∇2 Ψ(r)+V (r)Ψ(r) = EΨ(r) Schrödinger equation describes all the properties of a system through the wavefunction: Time-independent, non-relativistic Schrödinger equation
  • 10.
    •  There aren’ttoo many real situations where we can get a closed solution to the Schrödinger equation •  Let’s pretend we want to approach things numerically for 1000 electrons –  There are ~500,000 electron-electron interactions to worry about. –  Even storing the wavefunction would take ~101000 GB! •  Discretize the x,y,z, position of each electron into a 1000- element grid = 1 billion positions per electron •  Need the wavefunction output (real + complex part) for each combination of all electron positions, i.e. 1E9 ^ (1000) * 2, or 2E9000 values •  even at 1 byte per wavefunction value (low resolution), you have about 2E1000 GB needed needed to store the wavefunction! 10 The wave function is formidable
  • 11.
    Dirac summarized itbest … 11 “The underlying physical laws necessary for the mathematical theory of a large part of physics and the whole of chemistry are thus completely known, and the difficulty is only that the exact application of these laws leads to equations much too complicated to be soluble.” “It therefore becomes desirable that approximate practical methods of applying quantum mechanics should be developed, which can lead to an explanation of the main features of complex atomic systems without too much computation.”
  • 12.
    What is densityfunctional theory (DFT)? 12 DFT theory: •  replaces many-body interactions with a mean field interaction that reproduces the same charge density as the original formulation •  proves that, given the correct charge density, it in principle possible to compute all ground state properties of quantum mechanics exactly So, (for the ground state properties) we went from many-body wavefunctions to mean field charge density! This was worthy of the 1998 Nobel Prize. DFT practice: •  accuracy depends on the choice of (some) parameters, the type of material, the property to be studied, and whether the simulated system (crystal) is a good approximation of reality. e– e– e– e– e– e–
  • 13.
    How does oneuse DFT to design new materials? 13 A. Jain, Y. Shin, and K. A. Persson, Nat. Rev. Mater. 1, 15004 (2016).
  • 14.
    How accurate isDFT in practice? 14 Shown are typical DFT results for (i) Li battery voltages, (ii) electronic band gaps, and (iii) bulk modulus (i) (ii) (iii) (i) V. L. Chevrier, S. P. Ong, R. Armiento, M. K. Y. Chan, and G. Ceder, Phys. Rev. B 82, 075122 (2010). (ii) M. Chan and G. Ceder, Phys. Rev. Lett. 105, 196403 (2010). (iii) M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, M. Sluiter, C. K. Ande, S. Van Der Zwaag, J. J. Plata, C. Toher, S. Curtarolo, G. Ceder, K.A. Persson, and M. Asta, Sci. Data 2, 150009 (2015). battery voltages band gaps bulk modulus
  • 15.
    •  System sizeis essentially limited to ~1000 atoms –  many important materials phenomena simply do not occur at this length scale •  Certain materials, such as those with strong electron correlation, remain difficult to model accurately •  Certain properties, including excited state properties such as band gap, remain difficult to model accurately •  These are all active areas of research and improvement to the theory, and the situation is improving on all fronts 15 Limitations of density functional theory
  • 16.
    Outline 16 ①  From quantummechanics to density functional theory (DFT) ②  “High-throughput” DFT ③  Data mining approaches to materials design
  • 17.
    17 From a needlein a haystack to …
  • 18.
    18 … hiring anarmy to search through the haystack
  • 19.
    High-throughput DFT: akey idea 19 Automate the DFT procedure Supercomputing Power FireWorks Software for programming general computational workflows that can be scaled across large supercomputers. NERSC Supercomputing center, processor count is ~100,000 desktop machines. Other centers are also viable. High-throughput materials screening G. Ceder & K.A. Persson, Scientific American (2015)
  • 20.
    •  The answeris “it really varies a lot” –  how big / complicated are the materials you are modeling? –  how complex / expensive are the properties you are modeling? •  Ballpark numbers: –  Low range: optimize structure of ~3-atom compounds •  time to do a million materials ~ 10 million CPU-hours –  Medium range: bulk modulus of ~50 atom compounds •  time to do a million materials ~ 2 billion CPU hours •  The largest CPU allocations from the DOE are typically in the order of ~100 million CPU-hours 20 How much computer time is needed for high-throughput DFT?
  • 21.
    Examples of (early)high-throughput studies 21 Application Researcher Search space Candidates Hit rate Scintillators Klintenberg et al. 22,000 136 1/160 Curtarolo et al. 11,893 ? ? Topological insulators Klintenberg et al. 60,000 17 1/3500 Curtarolo et al. 15,000 28 1/535 High TC superconductors Klintenberg et al. 60,000 139 1/430 Thermoelectrics – ICSD - Half Heusler systems - Half Heusler best ZT Curtarolo et al. 2,500 80,000 80,000 20 75 18 1/125 1/1055 1/4400 1-photon water splitting Jacobsen et al. 19,000 20 1/950 2-photon water splitting Jacobsen et al. 19,000 12 1/1585 Transparent shields Jacobsen et al. 19,000 8 1/2375 Hg adsorbers Bligaard et al. 5,581 14 1/400 HER catalysts Greeley et al. 756 1 1/756* Li ion battery cathodes Ceder et al. 20,000 4 1/5000* Entries marked with * have experimentally verified the candidates. See also: Curtarolo et al., Nature Materials 12 (2013) 191–201.
  • 22.
    Computations predict, experimentsconfirm 22 Sidorenkite-based Li-ion battery cathodes LED phosphors YCuTe2 thermoelectrics Wang, Z., Ha, J., Kim, Y. H., Im, W. Bin, McKittrick, J. & Ong, S. P. Mining Unexplored Chemistries for Phosphors for High-Color-Quality White-Light-Emitting Diodes. Joule 2, 914–926 (2018). Chen, H.; Hao, Q.; Zivkovic, O.; Hautier, G.; Du, L.-S.; Tang, Y.; Hu, Y.-Y.; Ma, X.; Grey, C. P.; Ceder, G. Sidorenkite (Na3MnPO4CO3): A New Intercalation Cathode Material for Na-Ion Batteries, Chem. Mater., 2013 Aydemir, U; Pohls, J-H; Zhu, H; Hautier, G; Bajaj, S; Gibbs, ZM; Chen, W; Li, G; Broberg, D; White, MA; Asta, M; Persson, K; Ceder, G; Jain, A; Snyder, GJ. Thermoelectric Properties of Intrinsically Doped YCuTe2 with CuTe4-based Layered Structure. J. Mat. Chem C, 2016 More examples here: A. Jain, Y. Shin, and K. A. Persson, Nat. Rev. Mater. 1, 15004 (2016).
  • 23.
    With HT-DFT, wecan generate data rapidly – what to do next? 23 M. de Jong, W. Chen, H. Geerlings, M. Asta, and K. A. Persson, Sci. Data, 2015, 2, 150053.! M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, M. Sluiter, C. K. Ande, S. Van Der Zwaag, J. J. Plata, C. Toher, S. Curtarolo, G. Ceder, K. a Persson, and M. Asta, Sci. Data, 2015, 2, 150009.! >4500 elastic tensors >900 piezoelectric tensors >48000 Seebeck coefficients + cRTA transport Ricci, Chen, Aydemir, Snyder, Rignanese, Jain, & Hautier (in submission)!
  • 24.
    With HT-DFT, wecan generate data rapidly – what to do next? 24 M. de Jong, W. Chen, H. Geerlings, M. Asta, and K. A. Persson, Sci. Data, 2015, 2, 150053.! M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, M. Sluiter, C. K. Ande, S. Van Der Zwaag, J. J. Plata, C. Toher, S. Curtarolo, G. Ceder, K. a Persson, and M. Asta, Sci. Data, 2015, 2, 150009.! >4500 elastic tensors >900 piezoelectric tensors >48000 Seebeck coefficients + cRTA transport Ricci, Chen, Aydemir, Snyder, Rignanese, Jain, & Hautier (in submission)! Goal: make it easy to generate comparable data sets on your own
  • 25.
    A “black-box” viewof performing a calculation 25 “something”! Results!! researcher! What is the GGA-PBE elastic tensor of GaAs?
  • 26.
    Unfortunately, the insideof the “black box” is usually tedious and “low-level” 26 lots of tedious, low-level work…! Results!! researcher! What is the GGA-PBE elastic tensor of GaAs? Input file flags SLURM format how to fix ZPOTRF? q  set up the structure coordinates q  write input files, double-check all the flags q  copy to supercomputer q  submit job to queue q  deal with supercomputer headaches q  monitor job q  fix error jobs, resubmit to queue, wait again q  repeat process for subsequent calculations in workflow q  parse output files to obtain results q  copy and organize results, e.g., into Excel
  • 27.
    What would bea better way? 27 “something”! Results!! researcher! What is the GGA-PBE elastic tensor of GaAs?
  • 28.
    What would bea better way? 28 Results!! researcher! What is the GGA-PBE elastic tensor of GaAs? Workflows to run! q  band structure! q  surface energies! ü  elastic tensor! q  Raman spectrum! q  QH thermal expansion!
  • 29.
    Ideally the methodshould scale to millions of calculations 29 Results!! researcher! Start with all binary oxides, replace O->S, run several different properties Workflows to run! ü  band structure! ü  surface energies! ü  elastic tensor! q  Raman spectrum! q  QH thermal expansion! q  spin-orbit coupling!
  • 30.
    Atomate tries makeit easy, automatic, and flexible to generate data with existing simulation packages 30 Results!! researcher! Run many different properties of many different materials!
  • 31.
    Each simulation proceduretranslates high-level instructions into a series of low-level tasks 31 quickly and automatically translate PI-style (minimal) specifications into well-defined FireWorks workflows What is the GGA-PBE elastic tensor of GaAs? M. De Jong, W. Chen, T. Angsten, A. Jain, R. Notestine, A. Gamst, et al., Charting the complete elastic properties of inorganic crystalline compounds, Sci. Data. 2 (2015).
  • 32.
    Atomate contains alibrary of simulation procedures 32 VASP-based •  band structure •  spin-orbit coupling •  hybrid functional calcs •  elastic tensor •  piezoelectric tensor •  Raman spectra •  NEB •  GIBBS method •  QH thermal expansion •  AIMD •  ferroelectric •  surface adsorption •  work functions Other •  BoltzTraP •  FEFF method •  LAMMPS MD Mathew, K. et al Atomate: A high-level interface to generate, execute, and analyze computational materials science workflows, Comput. Mater. Sci. 139 (2017) 140–152.
  • 33.
    33 Full operation diagram job1 job 2 job 3 job 4 structure! workflow! database of all workflows! automatically submit + execute!output files + database!
  • 34.
    Atomate thus encodesand standardizes knowledge about running various kinds of simulations from domain experts 34 K. Mathew J. Montoya S. Dwaraknath A. Faghaninia All past and present knowledge, from everyone in the group, everyone previously in the group, and our collaborators, about how to run calculations M. Aykol S.P. Ong B. Bocklund T. Smidt H. Tang I.H. Chu M. Horton J. Dagdalen B. Wood Z.K. Liu J. Neaton K. Persson A. Jain +
  • 35.
    Outline 35 ①  From quantummechanics to density functional theory (DFT) ②  “High-throughput” DFT ③  Data mining approaches to materials design
  • 36.
    36 From a needlein a haystack to …
  • 37.
    37 … hiring anarmy to search through the haystack to …
  • 38.
  • 39.
    Can we builda general optimizer? 39 Generalizable forward solver Supercomputing Power Statistical optimization FireWorks NERSC Various optimization libraries (Figure: J. Mueller)
  • 40.
    Rocketsled: Automatic materialsscreening that selects materials to compute AND submits them to supercomputer 40 screening space of ~20,000 potential ABX3 perovskite combinations as water splitting materials – precomputed in DFT by different group if a machine learning algorithm was in charge of picking the next compound based on past data, how efficient would it be?
  • 41.
    Machine learning: thebig problem in my view is connecting data to ML algorithms through features 41 Lots of data on complex objects that you want to interrelate Clustering, Regression, Feature extraction, Model-building, etc. Well developed data-mining routines that work only on numbers (ideally ones with high relevance to your problem) Need to transform materials science objects into a set of physically relevant numerical data (“features” or “descriptors”)
  • 42.
    Goal of matminer:connect materials data with data mining algorithms and data visualization libraries 42 Ward, L. et al. Matminer: An open source toolkit for materials data mining. Comput. Mater. Sci. 152, 60–69 (2018).
  • 43.
    >40 featurizer classescan generate thousands of potential descriptors 43 Matminer contains a library of descriptors for various materials science entities feat = EwaldEnergy([options]) y = feat.featurize([input_data]) •  compatible with scikit-learn pipelining •  automatically deploy multiprocessing to parallelize over data •  include citations to methodology papers
  • 44.
    44 Interactive Jupyter notebooksdemonstrate use cases https://github.com/hackingmaterials/matminer_examples! Many examples available: •  Retrieving data from various databases •  Predicting bulk / shear modulus •  Predicting formation energies: •  from composition alone •  with Voronoi-based structure features included •  with Coulomb matrix and Orbital Field matrix descriptors (reproducing previous studies in the literature) •  Making interactive visualizations •  Creating an ML pipeline
  • 45.
    Defining local orderparameters for various environments 45 Use a given local order parameter with a threshold for motif recognition: If qtet > qthresh, then motif is tetrahedron. Else not (too much) a tetrahedron. Tetrahedral order parameter, qtet, [1]: [1] Zimmermann et al., J. Am. Chem. Soc., 2017, 10.1021/jacs.5b08098
  • 46.
    We have nowdeveloped mathematical order parameters for various types of local environments 46
  • 47.
    How well dothese work? 47 1. Order parameters clearly distinguish different environments even after thermal distortion 2. Work well in applications (defect site finding, diffusion characterization) [1] Zimmermann et al., Frontiers of Materials, 2017, doi: 10.3389/fmats.2017.00034
  • 48.
    48 Can cluster crystal structures by“local environment similarity”
  • 49.
    Results on MPweb site, e.g. for BCC-like structures 49 https://www.materialsproject.org/materials/mp-91/! Target: W similar structures (distance near 0) Cs3Sb! TiGaFeCo! CeMg2Cu!
  • 50.
    50 Text mining: learningfrom scientific abstracts Matstract corpus Unlabeled data Data labels Feature engineering Text cleaning Tokenization POS tag labels Word embeddings (word2vec) Text processing Hand crafted features Supervised learning Neural network (LSTM) Logistic regression Train/test sets Named Entities Named Entities “Learning” what a scientific study is about from >2 million materials science abstracts
  • 51.
    51 Application: a revisedmaterials search engine Auto-generated summaries of materials based on text mining
  • 52.
    52 Application: materials compositionsof interest … A search for “unconventional” (i.e., non-perovskite) ferroelectrics
  • 53.
    53 •  The MaterialsProject today mostly compiles fundamental simulation output data •  Many users don’t really know what to “do” with this data –  i.e., they would be interested in lattice thermal conductivity, but don’t know they can get there using MP data + other models •  How can we decorate MP database with additional properties and clearly show how we got there? How to connect different materials properties together? “Constitutive relations”
  • 54.
    54 A materials map/ network / atlas •  What we need is a “connected property network” •  This is a graph in which nodes are materials properties and edges are relationships between those properties •  Given a set of known properties, e.g. simulation data, one can easily figure out what are all the derived engineering properties one can get Starting with the three properties in blue, one can derive many additional properties using one (orange), two (purple), or three (green) physical models.! ! The value of computations are not only in the direct simulation outputs!!
  • 55.
  • 56.
    56 We are nowfeeding the “atlas” into the Materials Project to derive new data
  • 57.
    •  High-throughput densityfunctional theory, materials databases, and machine learning are a new set of tools for doing materials science •  We are developing many methods and software implementations to try to advance the field •  If you are interested, give the software a try! 57 Conclusions Quantum mechanics Density functional theory High-throughput DFT e– e– e– e– e– e– Materials databases Machine learning
  • 58.
    •  Atomate –  KMatthew (project lead) & team •  Structure order parameters –  N. Zimmermann (project lead) & team •  Rocketsled –  A. Dunn, J. Brenneck •  Matminer –  L. Ward (project lead, U. Chicago) & team •  Text mining –  V. Tshitoyan, J. Dagdelen, L. Weston •  Propnet –  M. K. Horton, D. Mrdjenovich •  All that provided feedback & contributed code to open-source software efforts! •  Funding: DOE-BES (Early Career + Materials Project Center) •  Computing: NERSC 58 Thank you!