skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Extracting an Empirical Intermetallic Hydride Design Principle from Limited Data via Interpretable Machine Learning

Abstract

An open question in the metal hydride community is whether there are simple, physics-based design rules that dictate the thermodynamic properties of these materials across the variety of structures and chemistry they can exhibit. While black box machine learning-based algorithms can predict these properties with some success, they do not directly provide the basis on which these predictions are made, therefore complicating the a priori design of novel materials exhibiting a desired property value. In this work we demonstrate how feature importance, as identified by a gradient boosting tree regressor, uncovers the strong dependence of the metal hydride equilibrium H2 pressure on a volume-based descriptor that can be computed from just the elemental composition of the intermetallic alloy. Elucidation of this simple structure–property relationship is valid across a range of compositions, metal substitutions, and structural classes exhibited by intermetallic hydrides. Finally, this permits rational targeting of novel intermetallics for high-pressure hydrogen storage (low-stability hydrides) by their descriptor values, and we predict a known intermetallic to form a low-stability hydride (as confirmed by density functional theory calculations) that has not yet been experimentally investigated.

Authors:
ORCiD logo [1]; ORCiD logo [2];  [2];  [2];  [1]; ORCiD logo [1]; ORCiD logo [1]
  1. Sandia National Lab. (SNL-CA), Livermore, CA (United States)
  2. Univ. of Nottingham, University Park, Nottingham (United Kingdom)
Publication Date:
Research Org.:
Sandia National Lab. (SNL-CA), Livermore, CA (United States)
Sponsoring Org.:
USDOE Office of Energy Efficiency and Renewable Energy (EERE), Transportation Office. Fuel Cell Technologies Office
OSTI Identifier:
1595021
Report Number(s):
SAND-2019-14626J
Journal ID: ISSN 1948-7185; 682475; TRN: US2100658
Grant/Contract Number:  
AC04-94AL85000; NA-0003525
Resource Type:
Accepted Manuscript
Journal Name:
Journal of Physical Chemistry Letters
Additional Journal Information:
Journal Volume: 11; Journal Issue: 1; Journal ID: ISSN 1948-7185
Publisher:
American Chemical Society
Country of Publication:
United States
Language:
English
Subject:
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; machine learning; metal hydrides

Citation Formats

Witman, Matthew, Ling, Sanliang, Grant, David M., Walker, Gavin S., Agarwal, Sapan, Stavila, Vitalie, and Allendorf, Mark D. Extracting an Empirical Intermetallic Hydride Design Principle from Limited Data via Interpretable Machine Learning. United States: N. p., 2019. Web. https://doi.org/10.1021/acs.jpclett.9b02971.
Witman, Matthew, Ling, Sanliang, Grant, David M., Walker, Gavin S., Agarwal, Sapan, Stavila, Vitalie, & Allendorf, Mark D. Extracting an Empirical Intermetallic Hydride Design Principle from Limited Data via Interpretable Machine Learning. United States. https://doi.org/10.1021/acs.jpclett.9b02971
Witman, Matthew, Ling, Sanliang, Grant, David M., Walker, Gavin S., Agarwal, Sapan, Stavila, Vitalie, and Allendorf, Mark D. Sat . "Extracting an Empirical Intermetallic Hydride Design Principle from Limited Data via Interpretable Machine Learning". United States. https://doi.org/10.1021/acs.jpclett.9b02971. https://www.osti.gov/servlets/purl/1595021.
@article{osti_1595021,
title = {Extracting an Empirical Intermetallic Hydride Design Principle from Limited Data via Interpretable Machine Learning},
author = {Witman, Matthew and Ling, Sanliang and Grant, David M. and Walker, Gavin S. and Agarwal, Sapan and Stavila, Vitalie and Allendorf, Mark D.},
abstractNote = {An open question in the metal hydride community is whether there are simple, physics-based design rules that dictate the thermodynamic properties of these materials across the variety of structures and chemistry they can exhibit. While black box machine learning-based algorithms can predict these properties with some success, they do not directly provide the basis on which these predictions are made, therefore complicating the a priori design of novel materials exhibiting a desired property value. In this work we demonstrate how feature importance, as identified by a gradient boosting tree regressor, uncovers the strong dependence of the metal hydride equilibrium H2 pressure on a volume-based descriptor that can be computed from just the elemental composition of the intermetallic alloy. Elucidation of this simple structure–property relationship is valid across a range of compositions, metal substitutions, and structural classes exhibited by intermetallic hydrides. Finally, this permits rational targeting of novel intermetallics for high-pressure hydrogen storage (low-stability hydrides) by their descriptor values, and we predict a known intermetallic to form a low-stability hydride (as confirmed by density functional theory calculations) that has not yet been experimentally investigated.},
doi = {10.1021/acs.jpclett.9b02971},
journal = {Journal of Physical Chemistry Letters},
number = 1,
volume = 11,
place = {United States},
year = {2019},
month = {12}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Save / Share:

Works referenced in this record:

Hydrogen-storage materials for mobile applications
journal, November 2001

  • Schlapbach, Louis; Züttel, Andreas
  • Nature, Vol. 414, Issue 6861
  • DOI: 10.1038/35104634

High capacity hydrogenstorage materials: attributes for automotive applications and techniques for materials discovery
journal, January 2010

  • Yang, Jun; Sudik, Andrea; Wolverton, Christopher
  • Chem. Soc. Rev., Vol. 39, Issue 2
  • DOI: 10.1039/B802882F

Thermal decomposition of complex metal hydrides
journal, June 1972


Structure of the Complex Metal Hydride BaReH9
journal, September 1994

  • Stetson, N. T.; Yvon, K.; Fischer, P.
  • Inorganic Chemistry, Vol. 33, Issue 20
  • DOI: 10.1021/ic00098a032

Formation of metal hydrides by mechanical alloying
journal, February 1995


Nanocrystalline metal hydrides
journal, May 1997


The renaissance of hydrides as energy materials
journal, December 2016


Exploring the High-Pressure Materials Genome
journal, November 2018


Nanostructured Metal Hydrides for Hydrogen Storage
journal, October 2018


Metal hydride hydrogen compressors: A review
journal, April 2014


Metal hydrides used as negative electrode materials for Li-ion batteries
journal, February 2016


Application of hydrides in hydrogen storage and compression: Achievements, outlook and perspectives
journal, March 2019

  • Bellosta von Colbe, Jose; Ares, Jose-Ramón; Barale, Jussara
  • International Journal of Hydrogen Energy, Vol. 44, Issue 15
  • DOI: 10.1016/j.ijhydene.2019.01.104

A correlation between the interstitial hole sizes in intermetallic compounds and the thermodynamic properties of the hydrides formed from those compounds
journal, November 1977


Relationships between intermetallic compound structure and hydride formation
journal, March 1981


An investigation of R6Fe23Hx thermodynamics
journal, October 1983


A microcalorimetric investigation of the thermodynamics and kinetics of hydriding-dehydriding reactions
journal, September 1996

  • Zhang, Wenlin; Sridhar Kumar, M. P.; Visintin, Arnaldo
  • Journal of Alloys and Compounds, Vol. 242, Issue 1-2
  • DOI: 10.1016/0925-8388(96)02282-7

Intermetallic compounds as negative electrodes of Ni/MH batteries
journal, February 2001

  • Cuevas, F.; Joubert, J. -M.; Latroche, M.
  • Applied Physics A Materials Science & Processing, Vol. 72, Issue 2
  • DOI: 10.1007/s003390100775

The electronegativity parameter for transition metals: Heat of formation and charge transfer in alloys
journal, July 1973


Hydrogen absorption in LaNi5 and related compounds: Experimental observations and their explanation
journal, March 1974


Hydrogen absorption in intermetallic compounds of thorium
journal, September 1975


Hydrides of intermetallic compounds: A review of stabilities, stoichiometries and preferred hydrogen sites
journal, May 1983


Calculation of the enthalpy of metal hydride formation
journal, November 1987


Identification of Destabilized Metal Hydrides for Hydrogen Storage Using First Principles Calculations
journal, May 2006

  • Alapati, Sudhakar V.; Johnson, J. Karl; Sholl, David S.
  • The Journal of Physical Chemistry B, Vol. 110, Issue 17
  • DOI: 10.1021/jp060482m

Using first principles calculations to identify new destabilized metal hydride reactions for reversible hydrogen storage
journal, January 2007

  • Alapati, Sudhakar V.; Karl Johnson, J.; Sholl, David S.
  • Physical Chemistry Chemical Physics, Vol. 9, Issue 12
  • DOI: 10.1039/b617927d

Discovery of novel hydrogen storage materials: an atomic scale computational approach
journal, January 2008


Theoretical Limits of Hydrogen Storage in Metal–Organic Frameworks: Opportunities and Trade-Offs
journal, August 2013

  • Goldsmith, Jacob; Wong-Foy, Antek G.; Cafarella, Michael J.
  • Chemistry of Materials, Vol. 25, Issue 16
  • DOI: 10.1021/cm401978e

What Are the Best Materials To Separate a Xenon/Krypton Mixture?
journal, June 2015


Machine learning for molecular and materials science
journal, July 2018


Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules
journal, January 2018

  • Gómez-Bombarelli, Rafael; Wei, Jennifer N.; Duvenaud, David
  • ACS Central Science, Vol. 4, Issue 2
  • DOI: 10.1021/acscentsci.7b00572

Metallic Metal–Organic Frameworks Predicted by the Combination of Machine Learning Methods and Ab Initio Calculations
journal, July 2018

  • He, Yuping; Cubuk, Ekin D.; Allendorf, Mark D.
  • The Journal of Physical Chemistry Letters, Vol. 9, Issue 16
  • DOI: 10.1021/acs.jpclett.8b01707

New tolerance factor to predict the stability of perovskite oxides and halides
journal, February 2019

  • Bartel, Christopher J.; Sutton, Christopher; Goldsmith, Bryan R.
  • Science Advances, Vol. 5, Issue 2
  • DOI: 10.1126/sciadv.aav0693

Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces
journal, April 2007


Predicting density functional theory total energies and enthalpies of formation of metal-nonmetal compounds by linear regression
journal, February 2016


Insightful classification of crystal structures using deep learning
journal, July 2018


Unsupervised word embeddings capture latent knowledge from materials science literature
journal, July 2019


Data-Driven Discovery of Photoactive Quaternary Oxides Using First-Principles Machine Learning
journal, August 2019


End-to-End Differentiable Learning of Protein Structure
journal, April 2019


Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
journal, April 2018


Linking synthesis and structure descriptors from a large collection of synthetic records of zeolite materials
journal, October 2019


SISSO: A compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates
journal, August 2018


Machine learning based prediction of metal hydrides for hydrogen storage, part I: Prediction of hydrogen weight percent
journal, March 2019

  • Rahnama, Alireza; Zepon, Guilherme; Sridhar, Seetharaman
  • International Journal of Hydrogen Energy, Vol. 44, Issue 14
  • DOI: 10.1016/j.ijhydene.2019.01.261

Machine learning based prediction of metal hydrides for hydrogen storage, part II: Prediction of material class
journal, March 2019

  • Rahnama, Alireza; Zepon, Guilherme; Sridhar, Seetharaman
  • International Journal of Hydrogen Energy, Vol. 44, Issue 14
  • DOI: 10.1016/j.ijhydene.2019.01.264

A simple constrained machine learning model for predicting high-pressure-hydrogen-compressor materials
journal, January 2018

  • Hattrick-Simpers, Jason R.; Choudhary, Kamal; Corgnale, Claudio
  • Molecular Systems Design & Engineering, Vol. 3, Issue 3
  • DOI: 10.1039/C8ME00005K

Commentary: The Materials Project: A materials genome approach to accelerating materials innovation
journal, July 2013

  • Jain, Anubhav; Ong, Shyue Ping; Hautier, Geoffroy
  • APL Materials, Vol. 1, Issue 1
  • DOI: 10.1063/1.4812323

Materials Design and Discovery with High-Throughput Density Functional Theory: The Open Quantum Materials Database (OQMD)
journal, September 2013


AFLOWLIB.ORG: A distributed materials properties repository from high-throughput ab initio calculations
journal, June 2012


Destabilization of the Mg-H System through Elastic Constraints
journal, June 2009


Metal-hydrogen systems with an exceptionally large and tunable thermodynamic destabilization
journal, November 2017


Optimizing nanoporous materials for gas storage
journal, January 2014

  • Simon, Cory M.; Kim, Jihan; Lin, Li-Chiang
  • Physical Chemistry Chemical Physics, Vol. 16, Issue 12
  • DOI: 10.1039/c3cp55039g

Role of Associative Charging in the Entropy–Energy Balance of Polyelectrolyte Complexes
journal, October 2018

  • Rathee, Vikramjit S.; Sidky, Hythem; Sikora, Benjamin J.
  • Journal of the American Chemical Society, Vol. 140, Issue 45
  • DOI: 10.1021/jacs.8b08649

Correlation between thermodynamical stabilities of metal borohydrides and cation electronegativites: First-principles calculations and experiments
journal, July 2006


Complex hydrides for energy storage
journal, March 2019


A general-purpose machine learning framework for predicting properties of inorganic materials
journal, August 2016


Discovery of Intermetallic Compounds from Traditional to Machine-Learning Approaches
journal, December 2017


On the heat of formation of solid alloys
journal, July 1975


On the heat of formation of solid alloys. II
journal, April 1976


Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set
journal, October 1996


Generalized Gradient Approximation Made Simple
journal, October 1996

  • Perdew, John P.; Burke, Kieron; Ernzerhof, Matthias
  • Physical Review Letters, Vol. 77, Issue 18, p. 3865-3868
  • DOI: 10.1103/PhysRevLett.77.3865

Projector augmented-wave method
journal, December 1994


Fully unconstrained noncollinear magnetism within the projector augmented-wave method
journal, November 2000


High-precision sampling for Brillouin-zone integration in metals
journal, August 1989


The hydrogen activation of LaNi5
journal, July 1992


A new study of the structure of LaNi5D6.7 using a modified Rietveld method for the refinement of neutron powder diffraction data
journal, February 1987


Compounds of uranium with the transition metals of the first long period
journal, January 1950


Synthesis of novel deuterides in several Laves phases by using gaseous deuterium under high pressure
journal, October 2002

  • Filipek, S. M.; Paul-Boncour, V.; Gu gan, A. Percheron
  • Journal of Physics: Condensed Matter, Vol. 14, Issue 44
  • DOI: 10.1088/0953-8984/14/44/464

Investigation on high-pressure metal hydride hydrogen compressors
journal, November 2007


Materials Databases: The Need for Open, Interoperable Databases with Standardized Data and Rich Metadata
journal, September 2019