Muratahan Aykol
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Muratahan Aykol has a total of 146 co-authors. Here are the top co-authors by number of shared publications:
- Joseph H. Montoya
Toyota Research Institute
7 shared publications - Anubhav Jain
Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, USA
4 shared publications - Chris Wolverton
International Institute for Nanotechnology Northwestern University Evanston Illinois USA
4 shared publications Jens S. Hummelshøj
Toyota Research Institute, Los Altos , California 94022, USA
4 shared publicationsJens Strabo Hummelshøj
4 shared publicationsJoseph H. Montoya
4 shared publications
Latest Publications
- Open Access
- 18 December 2024
- Published by Springer Nature in Nature Computational Science
- Vol. 5 (2) , 105-111
Predicting emergence of crystals from amorphous precursors with deep learning potentials
Crystallization of amorphous precursors into metastable crystals plays a fundamental role in the formation of new matter, from geological to biological processes in nature to the synthesis and development of new materials in the laboratory. Reliably predicting the outcome of such a process would enable new research directions in these areas, but has remained beyond the reach of molecular modeling or ab initio methods. Here we show that candidates for the crystallization products of amorphous precursors can be predicted in many inorganic systems by sampling the local structural motifs at the atomistic level using universal deep learning interatomic potentials. We show that this approach identifies, with high accuracy, the most likely crystal structures of the polymorphs that initially nucleate from amorphous precursors, across a diverse set of material systems, including polymorphic oxides, nitrides, carbides, fluorides, chlorides, chalcogenides and metal alloys.
- 18 December 2024
- Published by Springer Nature in Nature Computational Science
- Vol. 5 (2) , 105-111
Open AccessPredicting emergence of crystals from amorphous precursors with deep learning potentials
Crystallization of amorphous precursors into metastable crystals plays a fundamental role in the formation of new matter, from geological to biological processes in nature to the synthesis and development of new materials in the laboratory. Reliably predicting the outcome of such a process would enable new research directions in these areas, but has remained beyond the reach of molecular modeling or ab initio methods. Here we show that candidates for the crystallization products of amorphous precursors can be predicted in many inorganic systems by sampling the local structural motifs at the atomistic level using universal deep learning interatomic potentials. We show that this approach identifies, with high accuracy, the most likely crystal structures of the polymorphs that initially nucleate from amorphous precursors, across a diverse set of material systems, including polymorphic oxides, nitrides, carbides, fluorides, chlorides, chalcogenides and metal alloys.
- 7 October 2024
- Published by American Chemical Society (ACS) in Chemistry of Materials
- Vol. 36 (20) , 10205-10215
Lithium Transport in Crystalline and Amorphous Cathode Coatings for Li-Ion Batteries
- 7 October 2024
- Published by American Chemical Society (ACS) in Chemistry of Materials
- Vol. 36 (20) , 10205-10215
Lithium Transport in Crystalline and Amorphous Cathode Coatings for Li-Ion Batteries
- Top Cited PapersOpen Access
- 29 November 2023
- Published by Springer Nature in Nature
- Vol. 624 (7990) , 80-85
Scaling deep learning for materials discovery
Novel functional materials enable fundamental breakthroughs across technological applications from clean energy to information processing1–11. From microchips to batteries and photovoltaics, discovery of inorganic crystals has been bottlenecked by expensive trial-and-error approaches. Concurrently, deep-learning models for language, vision and biology have showcased emergent predictive capabilities with increasing data and computation12–14. Here we show that graph networks trained at scale can reach unprecedented levels of generalization, improving the efficiency of materials discovery by an order of magnitude. Building on 48,000 stable crystals identified in continuing studies15–17, improved efficiency enables the discovery of 2.2 million structures below the current convex hull, many of which escaped previous human chemical intuition. Our work represents an order-of-magnitude expansion in stable materials known to humanity. Stable discoveries that are on the final convex hull will be made available to screen for technological applications, as we demonstrate for layered materials and solid-electrolyte candidates. Of the stable structures, 736 have already been independently experimentally realized. The scale and diversity of hundreds of millions of first-principles calculations also unlock modelling capabilities for downstream applications, leading in particular to highly accurate and robust learned interatomic potentials that can be used in condensed-phase molecular-dynamics simulations and high-fidelity zero-shot prediction of ionic conductivity.
- 29 November 2023
- Published by Springer Nature in Nature
- Vol. 624 (7990) , 80-85
Top Cited PapersOpen AccessScaling deep learning for materials discovery
Novel functional materials enable fundamental breakthroughs across technological applications from clean energy to information processing1–11. From microchips to batteries and photovoltaics, discovery of inorganic crystals has been bottlenecked by expensive trial-and-error approaches. Concurrently, deep-learning models for language, vision and biology have showcased emergent predictive capabilities with increasing data and computation12–14. Here we show that graph networks trained at scale can reach unprecedented levels of generalization, improving the efficiency of materials discovery by an order of magnitude. Building on 48,000 stable crystals identified in continuing studies15–17, improved efficiency enables the discovery of 2.2 million structures below the current convex hull, many of which escaped previous human chemical intuition. Our work represents an order-of-magnitude expansion in stable materials known to humanity. Stable discoveries that are on the final convex hull will be made available to screen for technological applications, as we demonstrate for layered materials and solid-electrolyte candidates. Of the stable structures, 736 have already been independently experimentally realized. The scale and diversity of hundreds of millions of first-principles calculations also unlock modelling capabilities for downstream applications, leading in particular to highly accurate and robust learned interatomic potentials that can be used in condensed-phase molecular-dynamics simulations and high-fidelity zero-shot prediction of ionic conductivity.
- Open Access
Predicting emergence of crystals from amorphous matter with deep learning
Crystallization of the amorphous phases into metastable crystals plays a fundamental role in the formation of new matter, from geological to biological processes in nature to synthesis and development of new materials in the laboratory. Predicting the outcome of such phase transitions reliably would enable new research directions in these areas, but has remained beyond reach with molecular modeling or ab-initio methods. Here, we show that crystallization products of amorphous phases can be predicted in any inorganic chemistry by sampling the crystallization pathways of their local structural motifs at the atomistic level using universal deep learning potentials. We show that this approach identifies the crystal structures of polymorphs that initially nucleate from amorphous precursors with high accuracy across a diverse set of material systems, including polymorphic oxides, nitrides, carbides, fluorides, chlorides, chalcogenides, and metal alloys. Our results demonstrate that Ostwald's rule of stages can be exploited mechanistically at the molecular level to predictably access new metastable crystals from the amorphous phase in material synthesis.
Open AccessPredicting emergence of crystals from amorphous matter with deep learning
Crystallization of the amorphous phases into metastable crystals plays a fundamental role in the formation of new matter, from geological to biological processes in nature to synthesis and development of new materials in the laboratory. Predicting the outcome of such phase transitions reliably would enable new research directions in these areas, but has remained beyond reach with molecular modeling or ab-initio methods. Here, we show that crystallization products of amorphous phases can be predicted in any inorganic chemistry by sampling the crystallization pathways of their local structural motifs at the atomistic level using universal deep learning potentials. We show that this approach identifies the crystal structures of polymorphs that initially nucleate from amorphous precursors with high accuracy across a diverse set of material systems, including polymorphic oxides, nitrides, carbides, fluorides, chlorides, chalcogenides, and metal alloys. Our results demonstrate that Ostwald's rule of stages can be exploited mechanistically at the molecular level to predictably access new metastable crystals from the amorphous phase in material synthesis.
- 3 August 2023
- Published by American Chemical Society (ACS) in Chemistry of Materials
- Vol. 35 (16) , 6346-6356
Nature of the Amorphous–Amorphous Interfaces in Solid-State Batteries Revealed Using Machine-Learned Interatomic Potentials
- 3 August 2023
- Published by American Chemical Society (ACS) in Chemistry of Materials
- Vol. 35 (16) , 6346-6356
Nature of the Amorphous–Amorphous Interfaces in Solid-State Batteries Revealed Using Machine-Learned Interatomic Potentials
- Open Access
- 14 June 2022
- Published by Springer Nature in Scientific Data
- Vol. 9 (1) , 1-9
Novel inorganic crystal structures predicted using autonomous simulation agents
We report a dataset of 96640 crystal structures discovered and computed using our previously published autonomous, density functional theory (DFT) based, active-learning workflow named CAMD (Computational Autonomy for Materials Discovery). Of these, 894 are within 1 meV/atom of the convex hull and 26826 are within 200 meV/atom of the convex hull. The dataset contains DFT-optimized pymatgen crystal structure objects, DFT-computed formation energies and phase stability calculations from the convex hull. It contains a variety of spacegroups and symmetries derived from crystal prototypes derived from known experimental compounds, and was generated from active learning campaigns of various chemical systems. This dataset can be used to benchmark future active-learning or generative efforts for structure prediction, to seed new efforts of experimental crystal structure discovery, or to construct new models of structure-property relationships.
- 14 June 2022
- Published by Springer Nature in Scientific Data
- Vol. 9 (1) , 1-9
Open AccessNovel inorganic crystal structures predicted using autonomous simulation agents
We report a dataset of 96640 crystal structures discovered and computed using our previously published autonomous, density functional theory (DFT) based, active-learning workflow named CAMD (Computational Autonomy for Materials Discovery). Of these, 894 are within 1 meV/atom of the convex hull and 26826 are within 200 meV/atom of the convex hull. The dataset contains DFT-optimized pymatgen crystal structure objects, DFT-computed formation energies and phase stability calculations from the convex hull. It contains a variety of spacegroups and symmetries derived from crystal prototypes derived from known experimental compounds, and was generated from active learning campaigns of various chemical systems. This dataset can be used to benchmark future active-learning or generative efforts for structure prediction, to seed new efforts of experimental crystal structure discovery, or to construct new models of structure-property relationships.
- 13 January 2022
- Published by AIP Publishing in Applied Physics Reviews
- Vol. 9 (1) , 011405
Toward autonomous materials research: Recent progress and future challenges
The modus operandi in materials research and development is combining existing data with an understanding of the underlying physics to create and test new hypotheses via experiments or simulations. This process is traditionally driven by subject expertise and the creativity of individual researchers, who “close the loop” by updating their hypotheses and models in light of new data or knowledge acquired from the community. Since the early 2000s, there has been notable progress in the automation of each step of the scientific process. With recent advances in using machine learning for hypothesis generation and artificial intelligence for decision-making, the opportunity to automate the entire closed-loop process has emerged as an exciting research frontier. The future of fully autonomous research systems for materials science no longer feels far-fetched. Autonomous systems are poised to make the search for new materials, properties, or parameters more efficient under budget and time constraints, and in effect accelerate materials innovation. This paper provides a brief overview of closed-loop research systems of today, and our related work at the Toyota Research Institute applied across different materials challenges and identifies both limitations and future opportunities.
- 13 January 2022
- Published by AIP Publishing in Applied Physics Reviews
- Vol. 9 (1) , 011405
Toward autonomous materials research: Recent progress and future challenges
The modus operandi in materials research and development is combining existing data with an understanding of the underlying physics to create and test new hypotheses via experiments or simulations. This process is traditionally driven by subject expertise and the creativity of individual researchers, who “close the loop” by updating their hypotheses and models in light of new data or knowledge acquired from the community. Since the early 2000s, there has been notable progress in the automation of each step of the scientific process. With recent advances in using machine learning for hypothesis generation and artificial intelligence for decision-making, the opportunity to automate the entire closed-loop process has emerged as an exciting research frontier. The future of fully autonomous research systems for materials science no longer feels far-fetched. Autonomous systems are poised to make the search for new materials, properties, or parameters more efficient under budget and time constraints, and in effect accelerate materials innovation. This paper provides a brief overview of closed-loop research systems of today, and our related work at the Toyota Research Institute applied across different materials challenges and identifies both limitations and future opportunities.
- Open Access
- 24 December 2021
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 7 (52) , eabj5505
Machine learning–accelerated design and synthesis of polyelemental heterostructures
Machine learning accelerates materials discovery by suggesting targets, yielding exceptionally complex biphasic nanoparticles.
- 24 December 2021
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 7 (52) , eabj5505
Open AccessMachine learning–accelerated design and synthesis of polyelemental heterostructures
Machine learning accelerates materials discovery by suggesting targets, yielding exceptionally complex biphasic nanoparticles.
- Open Access
- 10 September 2021
- Published by Morressier GmbH
Scale bridging materials physics: Active learning workflows and integrable deep neural networks for free energy function representations in alloys
Scale bridging between the electronic structure of a solid and continuum descriptions of its non-equilibrium behavior can be realized with integrable deep neural networks (IDNN) that are trained to free energy derivative data generated by first-principles statistical mechanics simulations and then analytically integrating to recover a free energy density function. Here we combine
- 10 September 2021
- Published by Morressier GmbH
Open AccessScale bridging materials physics: Active learning workflows and integrable deep neural networks for free energy function representations in alloys
Scale bridging between the electronic structure of a solid and continuum descriptions of its non-equilibrium behavior can be realized with integrable deep neural networks (IDNN) that are trained to free energy derivative data generated by first-principles statistical mechanics simulations and then analytically integrating to recover a free energy density function. Here we combine
- Top Cited Papers
- 11 June 2021
- Published by American Chemical Society (ACS) in Journal of the American Chemical Society
- Vol. 143 (24) , 9244-9259
Rational Solid-State Synthesis Routes for Inorganic Materials
The rational solid-state synthesis of inorganic compounds is formulated as catalytic nucleation on crystalline reactants, where contributions of reaction and interfacial energies to the nucleation barriers are approximated from high-throughput thermochemical data and structural and interfacial features of crystals, respectively. Favorable synthesis reactions are then identified by a Pareto analysis of relative nucleation barriers and phase selectivities of reactions leading to the target. We demonstrate the application of this approach in reaction planning for the solid-state synthesis of a range of compounds, including the widely studied oxides LiCoO2, BaTiO3, and YBa2Cu3O7, as well as other metal oxide, oxyfluoride, phosphate, and nitride targets. Pathways for enabling the retrosynthesis of inorganics are also discussed.
- 11 June 2021
- Published by American Chemical Society (ACS) in Journal of the American Chemical Society
- Vol. 143 (24) , 9244-9259
Top Cited PapersRational Solid-State Synthesis Routes for Inorganic Materials
The rational solid-state synthesis of inorganic compounds is formulated as catalytic nucleation on crystalline reactants, where contributions of reaction and interfacial energies to the nucleation barriers are approximated from high-throughput thermochemical data and structural and interfacial features of crystals, respectively. Favorable synthesis reactions are then identified by a Pareto analysis of relative nucleation barriers and phase selectivities of reactions leading to the target. We demonstrate the application of this approach in reaction planning for the solid-state synthesis of a range of compounds, including the widely studied oxides LiCoO2, BaTiO3, and YBa2Cu3O7, as well as other metal oxide, oxyfluoride, phosphate, and nitride targets. Pathways for enabling the retrosynthesis of inorganics are also discussed.