Projekt

Zurück zur Übersicht

Development of neural network interatomic potentials for phase change materials

Gesuchsteller/in Parrinello Michele
Nummer 137631
Förderungsinstrument Projektförderung (Abt. I-III)
Forschungseinrichtung Computational Science Dept. of Chemistry and Applied Biosciences ETH Zürich
Hochschule ETH Zürich - ETHZ
Hauptdisziplin Physik der kondensierten Materie
Beginn/Ende 01.10.2011 - 31.10.2013
Bewilligter Betrag 228'324.00
Alle Daten anzeigen

Alle Disziplinen (2)

Disziplin
Physik der kondensierten Materie
Materialwissenschaften

Keywords (3)

phase-change materials; molecular dynamic simulations; neural network potentials

Lay Summary (Englisch)

Lead
Lay summary

Non-volatile memories are nowadays employed in a wide series of applications ranging from memory cards used in several electronic devices (cameras, mobile phones ...), mass storage devices (USB sticks, MP3 players, SSDs ...), and to embedded systems or even automotive applications. Traditional non-volatile memory implementations are based on the flash technology. The 0/1 bit information is stored via trapping/removal of electrons in an electrically isolated part of the device. Their presence or absence opens or closes a conducting channel between two metal contacts in a transistor-like setup. Readout operation is performed via conductivity measure between such metal contacts. Due to the high demand for larger and larger capacity, NAND-type flash memory is the most aggressively scaled technology among electronic devices.  However, the feature size of flash memory cells is close to reach its intrinsic minimum limit for scaling to higher capacity.

Phase-change materials based on chalcogenide alloys are promising candidates for flash memory replacement as next-generation non-volatile electronic memories. A phase-change memory is essentially a resistor formed by a thin chalcogenide film with a low field resistance which changes by several orders of magnitude, depending on the state of the chalcogenide, metallic in the crystalline form and insulating in the amorphous phase. The large difference in conductivity between the two states is the feature which enables information storage. Together with better scalability/higher memory density, phase-change memories offer a wide set of advantages over flash memories: higher write bandwidth, reduced power consumption, smaller latency and better endurance.

However, in spite of the great technological importance of this class of materials, the microscopic origin of several of their properties is still matter of debate. On the other hand, the performance and reliability of the device can be optimized by an accurate selection of the active chalcogenide compound. For instance, a change of several tens of degrees of the crystallization temperature can be achieved by tuning the composition of the ternary GeSbTe alloys. This dramatic change of physical properties offers wide opportunities to tailor the memory performance to specific applications and/or to improve its scalability. However, this optimization procedure cannot be left to mere trial-and-error, but requires a detailed physical understanding of the material properties. In this respect, atomistic simulations can provide crucial insights to aid the experimental activity, as for instance elucidating the correlation between functional and structural properties of these materials at the atomistic level.

Although fully ab-initio atomistic simulations have provided crucial insights on the properties of phase change materials in the lat five years,  several key issues such as the thermal conductivity at the nanoscale, the crystallization dynamics, and  the properties of the crystalline/amorphous interface, just to name a few, are presently beyond the reach of  fully ab-initio simulations.

The development of reliable classical interatomic potentials is a possible route to overcome the limitations in system size and time scale of ab-initio molecular dynamics.  Traditional approaches based on the fitting of simple functional forms for the interatomic potentials turned out to be unfeasible due to the complexity and variability of the chemical bonding in the crystal and amorphous phases revealed by the ab-initio simulations.

A possible solution is the development of empirical interatomic potentials with close to ab-initio accuracy by fitting large ab-initio databases within a neural network (NN) scheme to allow simulating thousands of atoms for tens of ns. This method has been applied successfully to study elemental sodium, carbon and silicon in the last few years. By means of this approach, we plan to devise NN potentials for phase change materials which will allow us to address the study of the crystallization dynamics, thermal transport and the amorphous/crystalline interface and their dependence on composition. We are confident that the insights provided by the atomistic simulation will aid the experimental search for better performing materials in this class for non volatile memories applications.  

Direktlink auf Lay Summary Letzte Aktualisierung: 21.02.2013

Verantw. Gesuchsteller/in und weitere Gesuchstellende

Mitarbeitende

Publikationen

Publikation
Density functional simulations of hexagonal Ge2Sb2Te5 at high pressure
Caravati S. Sosso G.C. Bernasconi M. Parrinello M. (2013), Density functional simulations of hexagonal Ge2Sb2Te5 at high pressure, in Physical Rev. B, 87(9), 094117-094128.
Density functional simulations of hexagonal GST at high pressure
Caravati Sebastiano Sosso Gabriele C. Bernasconi Marco Parrinello Michele (2013), Density functional simulations of hexagonal GST at high pressure, in Phys. Rev. B, 87(9), 094117-1-094117-12.
Density functional simulations of sb-rich GeSbTe phase change alloys
Gabardi S., Caravati Sebastiano, Bernasconi Marco, Parrinello Michele (2012), Density functional simulations of sb-rich GeSbTe phase change alloys, in Journal of Physics: Condensed Matter , 24(38), 385803-1-385803-11.
Neural-network interatomic potential for the phase change material GeTe
Sosso Gabriele C., Miceli Giacomo, Caravati Sebastiano, Behler Jörg, Bernasconi Marco (2012), Neural-network interatomic potential for the phase change material GeTe, in Physical Review B, 85(17), 174103-1-174103-13.
Thermal transport in phase-change materials from atomistic simulations
Sosso Gabriele C., Donadio Davide, Caravati Sebastiano, Behler Jörg, Bernasconi Marco (2012), Thermal transport in phase-change materials from atomistic simulations, in Physical Review B, 86(10), 104301-1-104301-5.
Large scale atomistic simulations of phase change materials
Sosso Gabriele C., Miceli Giacomo, Caravati Sebastiano, Behler Jörg, Bernasconi Marco, Large scale atomistic simulations of phase change materials, Proceedings E\PCOS 2012, Proceedings E\PCOS 2012.

Zusammenarbeit

Gruppe / Person Land
Formen der Zusammenarbeit
Marco Bernasconi Italien (Europa)
- vertiefter/weiterführender Austausch von Ansätzen, Methoden oder Resultaten
- Publikation

Wissenschaftliche Veranstaltungen

Aktiver Beitrag

Titel Art des Beitrags Titel des Artikels oder Beitrages Datum Ort Beteiligte Personen
E/PCOS 2013 Vortrag im Rahmen einer Tagung Large scale atomistic simulations of the crystallization of GeTe 09.09.2013 Berlin, Deutschland Sosso Gabriele Cesare;
FisMat 2013 Poster Large Scale Molecular Dynamics Simulations of Supercooled Liquid GeTe 09.09.2013 Milano, Italien Sosso Gabriele Cesare;
CPMD - Leipzig 2013, Matter, life, light from ab inition molecular dynamics simulations Poster Homogeneus nucleation and growth of the phase change material GeTe via LARGE SCALE Molecular Dynamics simulation 02.09.2013 Leipzig, Deutschland Parrinello Michele; Sosso Gabriele Cesare;
CECAM Workshop - From cooperativity in supercooled liquids to plasticity of amorphous solids Poster Dynamical properties of supercooled liquid GeTe via large scale simulations: unravelling the key of the fast crystallization of phase change materials 26.06.2013 Zürich, Schweiz Sosso Gabriele Cesare;
International Symposium Fundamentals of Laser Assisted Micro- and Nanotechnologies Vortrag im Rahmen einer Tagung Large scale atomistic simulations of phase change materials for optical data storage (M. Bernasconi) 24.06.2013 St. Petersburg, Russland Sosso Gabriele Cesare;
Material Research Society Spring Meeting Vortrag im Rahmen einer Tagung Large scale molecular dynamics simulation of the crystallization dynamics of GeTe (Marco Bernasconi) 01.04.2013 San Francisco, Vereinigte Staaten von Amerika Sosso Gabriele Cesare;
CUSO Winterschool - Simulating activated processes in physics and chemistry: theoretical foundations Poster Dynamical properties of supercooled liquid GeTe via large scale simulations: unravelling the key of the fast crystallization of phase change materialsPoster 03.02.2013 Villars, Schweiz Sosso Gabriele Cesare;
16th Workshop on Computational Physics and Materials Science: Total Energy and Force Methods Vortrag im Rahmen einer Tagung Atomistic simulations of phaxe change materials for data storage (Marco Bernasconi) 10.01.2013 Triest, Italien Sosso Gabriele Cesare;
EPCOS 2012 Poster Large scale atomistic simulations of phase change materials 08.09.2012 Tampere, Finland, Finnland Caravati Sebastiano;
Nature Conference on "Frontiers in Electronic Materials: Correlation Effects and Memrisitive Phenomena" Poster Large Scale Molecular Dynamics Simulations of Phase Change Materials 17.06.2012 Aachen, Deutschland Caravati Sebastiano; Sosso Gabriele Cesare;
Challenges in the Atomic Scale Modelling of Glasses Vortrag im Rahmen einer Tagung Large scale simulations of amorphous phase change materials for data storage (Marco Bernasconi) 04.06.2012 Strasbourg, Frankreich Caravati Sebastiano;
European Material Research Society Meeting Vortrag im Rahmen einer Tagung Large scale molecular dynamics simulations of phase change materials (Marco Bernasconi) 14.05.2012 Strasbourg, Frankreich Caravati Sebastiano;
Material Research Society Spring Meeting Vortrag im Rahmen einer Tagung Invited talk 09.04.2012 San Francisco, USA, Vereinigte Staaten von Amerika Caravati Sebastiano;


Auszeichnungen

Titel Jahr
Enrico Fermi Prize, Italian Physical Society, Italy For the discovery of a Molecular Dynamics method known the world over as the Car-Parrinello method. This method has been a breakthrough in the field of numerical simulations, with great impact in many interdisciplinary contexts both theoretical and experimental, ranging from Material Science to Chemistry and Biology 2012
Grande Ufficiale della Repubblica Italiana 2012
Hirschfelder Prize, University of Wisconsin-Madison, USA 2012
Honorary Doctorate in Science, University of St. Andrews, Scotland 2012

Verbundene Projekte

Nummer Titel Start Förderungsinstrument
119882 Phase Change Materials from First Principles 01.04.2008 Projektförderung (Abt. I-III)

Abstract

Lead: Phase-change materials based on chalcogenide alloys are promising candidates for next-generation non-volatile electronic memories. In spite of the great technological importance of this class of materials, the microscopic origin of their properties is still largely unknown. Computer simulations provide a great help in elucidating the correlation between functional and structural properties of these materials at the atomistic level.Background: Phase-change materials are attracting an increasing interest for applications in non volatile memories. A phase-change memory is essentially a resistor formed by a thin chalcogenide film with a low field resistance which changes by several orders of magnitude, depending on the state of the chalcogenide, metallic in the crystalline form and insulating in the amorphous phase. Performance and reliability can be optimized by an accurate selection of the active chalcogenide material. For instance, by tuning its composition, a change of several tens of degrees of the crystallization temperature can be achieved. This dramatic change of physical properties offers wide opportunities to tailor the memory performance to specific applications and/or to improve its scalability. However, this optimization procedure cannot be left to mere trial-and-error, but requires a detailed physical understanding of the material properties. In this respect, atomistic simulations can provide crucial insights to aid the experimental activity.Aim: Based on our previous first-principles simulations of phase-change materials, we plan to develop empirical interatomic potentials with ab-initio accuracy by fitting large ab-initio databases within a neural network scheme. These neural network potentials will allow performing very accurate simulations at a fraction of the computational costs of conventional ab-initio MD simulations enabling to study systems and problems which are otherwise inaccessible. In particular, we aim at identifying the mechanism of formation of crystalline seeds, the corresponding activation energy and its dependence on composition which controls the stability of the amorphous phase and thus the reliability of phase-change memories.Impact: The development of this method will allow simulating with ab-initio accuracy systems of an unprecedented size, very close to actual device sizes. We hope that these simulations will complement and illuminate experimental research of better performing materials for non volatile memories, leading to faster progress.
-