Inverse Design of a Graphene-Based Quantum Transducer via Neuroevolution

Kevin Ryczko; Pierre Darancet; Isaac Tamblyn

doi:10.1021/acs.jpcc.0c06903

Inverse Design of a Graphene-Based Quantum Transducer via Neuroevolution

Kevin Ryczko^*, Pierre Darancet^*, Isaac Tamblyn^*

^*Corresponding author for this work

MRC - Materials Research Center

Research output: Contribution to journal › Article › peer-review

8 Scopus citations

Abstract

We introduce an inverse design framework based on artificial neural networks, genetic algorithms, and tight-binding calculations, capable to optimize the very large configuration space of nanoelectronic devices. Our non-linear optimization procedure operates on trial Hamiltonians through superoperators controlling growth policies of regions of distinct doping. We demonstrate that our algorithm optimizes the doping of graphene-based three-terminal devices for valleytronics applications, monotonously converging to synthesizable devices with high merit functions in a few thousand evaluations (out of â23800 possible configurations). The best-performing device allowed for a terminal-specific separation of valley currents with â96% (â94%) K(K) valley purity. Importantly, the devices found through our non-linear optimization procedure have both a higher merit function and higher robustness to defects than the ones obtained through geometry optimization.

Original language	English (US)
Pages (from-to)	26117-26123
Number of pages	7
Journal	Journal of Physical Chemistry C
Volume	124
Issue number	48
DOIs	https://doi.org/10.1021/acs.jpcc.0c06903
State	Published - Dec 3 2020

Funding

K.R. and I.T. acknowledge funding from the Natural Sciences and Engineering Research Council of Canada. Work at the National Research Council was carried out under the auspices of the AI4Design Program. K.R. and I.T. acknowledge Compute Canada and the National Energy Research Scientific Computing Center for computational resources. This material is based upon work supported by Laboratory Directed Research and Development (LDRD) funding from the Argonne National Laboratory, provided by the Director, Office of Science, of the U.S. Department of Energy under contract no. DE-AC02-06CH11357. Use of the Center for Nanoscale Materials, an Office of Science user facility, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract no. DE-AC02-06CH11357.

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
General Energy
Physical and Theoretical Chemistry
Surfaces, Coatings and Films

Access to Document

10.1021/acs.jpcc.0c06903

Cite this

@article{fad1d7d0be9f420eb61cdab194db8165,

title = "Inverse Design of a Graphene-Based Quantum Transducer via Neuroevolution",

abstract = "We introduce an inverse design framework based on artificial neural networks, genetic algorithms, and tight-binding calculations, capable to optimize the very large configuration space of nanoelectronic devices. Our non-linear optimization procedure operates on trial Hamiltonians through superoperators controlling growth policies of regions of distinct doping. We demonstrate that our algorithm optimizes the doping of graphene-based three-terminal devices for valleytronics applications, monotonously converging to synthesizable devices with high merit functions in a few thousand evaluations (out of {\^a}23800 possible configurations). The best-performing device allowed for a terminal-specific separation of valley currents with {\^a}96% ({\^a}94%) K(K) valley purity. Importantly, the devices found through our non-linear optimization procedure have both a higher merit function and higher robustness to defects than the ones obtained through geometry optimization.",

author = "Kevin Ryczko and Pierre Darancet and Isaac Tamblyn",

note = "Funding Information: K.R. and I.T. acknowledge funding from the Natural Sciences and Engineering Research Council of Canada. Work at the National Research Council was carried out under the auspices of the AI4Design Program. K.R. and I.T. acknowledge Compute Canada and the National Energy Research Scientific Computing Center for computational resources. This material is based upon work supported by Laboratory Directed Research and Development (LDRD) funding from the Argonne National Laboratory, provided by the Director, Office of Science, of the U.S. Department of Energy under contract no. DE-AC02-06CH11357. Use of the Center for Nanoscale Materials, an Office of Science user facility, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract no. DE-AC02-06CH11357. Publisher Copyright: {\textcopyright} 2020 American Chemical Society.",

year = "2020",

month = dec,

day = "3",

doi = "10.1021/acs.jpcc.0c06903",

language = "English (US)",

volume = "124",

pages = "26117--26123",

journal = "Journal of Physical Chemistry C",

issn = "1932-7447",

publisher = "American Chemical Society",

number = "48",

}

TY - JOUR

T1 - Inverse Design of a Graphene-Based Quantum Transducer via Neuroevolution

AU - Ryczko, Kevin

AU - Darancet, Pierre

AU - Tamblyn, Isaac

N1 - Funding Information: K.R. and I.T. acknowledge funding from the Natural Sciences and Engineering Research Council of Canada. Work at the National Research Council was carried out under the auspices of the AI4Design Program. K.R. and I.T. acknowledge Compute Canada and the National Energy Research Scientific Computing Center for computational resources. This material is based upon work supported by Laboratory Directed Research and Development (LDRD) funding from the Argonne National Laboratory, provided by the Director, Office of Science, of the U.S. Department of Energy under contract no. DE-AC02-06CH11357. Use of the Center for Nanoscale Materials, an Office of Science user facility, was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract no. DE-AC02-06CH11357. Publisher Copyright: © 2020 American Chemical Society.

PY - 2020/12/3

Y1 - 2020/12/3

N2 - We introduce an inverse design framework based on artificial neural networks, genetic algorithms, and tight-binding calculations, capable to optimize the very large configuration space of nanoelectronic devices. Our non-linear optimization procedure operates on trial Hamiltonians through superoperators controlling growth policies of regions of distinct doping. We demonstrate that our algorithm optimizes the doping of graphene-based three-terminal devices for valleytronics applications, monotonously converging to synthesizable devices with high merit functions in a few thousand evaluations (out of â23800 possible configurations). The best-performing device allowed for a terminal-specific separation of valley currents with â96% (â94%) K(K) valley purity. Importantly, the devices found through our non-linear optimization procedure have both a higher merit function and higher robustness to defects than the ones obtained through geometry optimization.

AB - We introduce an inverse design framework based on artificial neural networks, genetic algorithms, and tight-binding calculations, capable to optimize the very large configuration space of nanoelectronic devices. Our non-linear optimization procedure operates on trial Hamiltonians through superoperators controlling growth policies of regions of distinct doping. We demonstrate that our algorithm optimizes the doping of graphene-based three-terminal devices for valleytronics applications, monotonously converging to synthesizable devices with high merit functions in a few thousand evaluations (out of â23800 possible configurations). The best-performing device allowed for a terminal-specific separation of valley currents with â96% (â94%) K(K) valley purity. Importantly, the devices found through our non-linear optimization procedure have both a higher merit function and higher robustness to defects than the ones obtained through geometry optimization.

UR - http://www.scopus.com/inward/record.url?scp=85096577515&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85096577515&partnerID=8YFLogxK

U2 - 10.1021/acs.jpcc.0c06903

DO - 10.1021/acs.jpcc.0c06903

M3 - Article

AN - SCOPUS:85096577515

SN - 1932-7447

VL - 124

SP - 26117

EP - 26123

JO - Journal of Physical Chemistry C

JF - Journal of Physical Chemistry C

IS - 48

ER -

Inverse Design of a Graphene-Based Quantum Transducer via Neuroevolution

Abstract

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this