Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts...
-
Upload
silvia-mcdonald -
Category
Documents
-
view
219 -
download
6
Transcript of Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts...
![Page 1: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/1.jpg)
Ab Initio Crystal Structure Prediction:High-throughput and Data Mining
Dane Morgan
Massachusetts Institute of Technology
NSF Division of Materials Research
ITR Computational Workshop
June 17-19, 2004
![Page 2: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/2.jpg)
Why Does Structure Matter? Essential for Rational Materials Design
Structure key to understand properties and performance
Key input for property computational modeling
Performance
Properties
Structure and
Composition
Processing
![Page 3: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/3.jpg)
Why Do We Need Structure Predictions?Structural Information is Often Lacking
Binary alloys incomplete
Multi-component systems largely unknown
Massalski, Binary Alloy Phase Diagrams ‘90
![Page 4: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/4.jpg)
The Structure Prediction Problem
Given elements A, B, C, … predict the stable low-temperature phases
Crystalline phasesAb initio methods
Present focus
![Page 5: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/5.jpg)
Why is Structure Prediction Hard?
Ene
rgy
Atomic positions
Local minima
Global minimumTrue structure
Infinite structural space Rough energy surface – many local minima
Ab initio methods give accurate energies, but …
![Page 6: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/6.jpg)
Two New Tools
Data MiningCalculated/Experimental
Databases
High-Throughput Ab Initio
![Page 7: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/7.jpg)
High-Throughput Ab Initio
0
1
2
3
4
5
6
1965 1975 1985 1995 2005
Lo
g(#
ca
lcu
lati
on
s)
Cheilokowsky and Cohen, ‘74
Si
Metals Database
~14,000 EnergiesCurtarolo, et al., Submitted ‘04
Robust methods/codesAutomated tasksParallel computation
![Page 8: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/8.jpg)
Ab Initio Structure Prediction
Directly optimize ab initio Hamiltonian with Monte Carlo, genetic algorithms, etc. (too slow)
Simplified Hamiltonians – potentials, cluster expansion (fitting challenges, limited transferability/accuracy)
Intelligent guess at good candidates
Obtain a manageable list of likely candidate structures for high-throughput calculation
How good can this be?
![Page 9: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/9.jpg)
“Usual Suspects” Structure List
80 binary intermetallic alloys176 “usual suspects” structures(“usual suspects” = Most frequent in CRYSTMET, hcp, bcc, fcc superstructures)
Metals Database
~14,000 Energies
Calculate energiesConstruct convex hullsCompare to experiment
![Page 10: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/10.jpg)
High-Throughput Predictions
Metals Database
~14,000 Energies
95 predictions of new compounds
21 predictions for unidentified compounds
110 agreements
3 unambiguous errorsCurtarolo, et al., Submitted ‘04
But far too many structures + alloys to explore!! Need smart way to choose “sensible” structures!!
![Page 11: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/11.jpg)
Data Mining
New alloy system A,B,C,…
Predicted crystal structure
DatabaseData Miningto choose
“sensible” structures
![Page 12: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/12.jpg)
Data Mining with Correlations
Atomic positions
Ene
rgy
E( ) = E( )
E( ) = [E( ) + E( )]2
1
Linear correlations between energies
Faster to find low energies
All energies do not need to be calculated
Do linear correlations exist between structural energies across alloys?
![Page 13: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/13.jpg)
Structural Energy Correlations Exist!
Principal Component Analysis identifies correlations
0
0.05
0.1
0.15
0.2
0.25
0 20 40 60 80 100% Total Degrees of Freedom
RM
S E
rro
r (e
V/a
tom
)
N structural energies from N/3 independent variables
True data
Randomized data
![Page 14: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/14.jpg)
Using Correlations for Structure Prediction
Databasecalculated energies(AC, BC, etc.) + AB
New alloy system: AB
Predicted crystal structure
Correlations Predict likely stable structure i for alloy AB
Calculate Ei
Accurate convex hull?No
Yes
![Page 15: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/15.jpg)
Data Mining Example: AgCd
![Page 16: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/16.jpg)
Compound Forming Vs. Phase Separating
0
20
40
60
80
100
95 96 97 98 99 100Accuracy (%)
# C
alcu
lati
on
s
~2-8x speedup from Data Mining
No DM
With DM
![Page 17: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/17.jpg)
Ground State Prediction
0
20
40
60
80
100
120
50 60 70 80 90 100Accuracy (%)
# C
alcu
lati
on
s
No DM
With DM
~4x speedup from Data Mining
![Page 18: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/18.jpg)
Conclusions
Future workMore experimental/computed dataMore data mining tools Web interface
Practical tool to predict crystal structure
High-throughput ab initio approaches are a powerful tool for crystal structure prediction.
Data Mining of previous calculations can create significant speedup when studying new systems.
![Page 19: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/19.jpg)
Web Access to Database
Easy Interface Analysis: Convex hull, Ground States
Structural and Computational Data, Visualization
![Page 20: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/20.jpg)
Collaborators/Acknowledgements
Collaborators Mohan Akula (MIT) Stefano Curtarolo (Duke) Chris Fischer (MIT) Kristin Persson (MIT) John Rodgers (NRC Canada, Toth, Inc.) Kevin Tibbetts (MIT)
FundingNational Science Foundation Information Technology Research (NSF-ITR) Grant No. DMR-0312537
![Page 21: Ab Initio Crystal Structure Prediction: High-throughput and Data Mining Dane Morgan Massachusetts Institute of Technology NSF Division of Materials Research.](https://reader036.fdocuments.us/reader036/viewer/2022062805/5697c0061a28abf838cc58d7/html5/thumbnails/21.jpg)
END