Atomic-scale finite element method in multiscale computation with applications to carbon nanotubes

B. Liu*, H. Jiang, Y. Huang, S. Qu, M. F. Yu, K. C. Hwang

*Corresponding author for this work

Research output: Contribution to journalArticle

116 Citations (Scopus)

Abstract

We have developed an accurate atomic-scale finite element method (AFEM) that has exactly the same formal structure as continuum finite element methods, and therefore can seamlessly be combined with them in multiscale computations. The AFEM uses both first and second derivatives of system energy in the energy minimization computation. It is faster than the standard conjugate gradient method which uses only the first order derivative of system energy, and can thus significantly save computation time especially in studying large scale problems. Woven nanostructures of carbon nanotubes are proposed and studied via this new method, and strong defect insensitivity in such nanostructures is revealed. The AFEM is also readily applicable for solving many physics related optimization problems.

Original languageEnglish (US)
Article number035435
JournalPhysical Review B - Condensed Matter and Materials Physics
Volume72
Issue number3
DOIs
StatePublished - Jul 15 2005

Fingerprint

Carbon Nanotubes
Carbon nanotubes
finite element method
carbon nanotubes
Finite element method
Nanostructures
Derivatives
conjugate gradient method
optimization
Conjugate gradient method
energy
Physics
continuums
Defects
physics
sensitivity
defects

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics

Cite this

@article{093747f73b734d42b0e6abaf6d1c6557,
title = "Atomic-scale finite element method in multiscale computation with applications to carbon nanotubes",
abstract = "We have developed an accurate atomic-scale finite element method (AFEM) that has exactly the same formal structure as continuum finite element methods, and therefore can seamlessly be combined with them in multiscale computations. The AFEM uses both first and second derivatives of system energy in the energy minimization computation. It is faster than the standard conjugate gradient method which uses only the first order derivative of system energy, and can thus significantly save computation time especially in studying large scale problems. Woven nanostructures of carbon nanotubes are proposed and studied via this new method, and strong defect insensitivity in such nanostructures is revealed. The AFEM is also readily applicable for solving many physics related optimization problems.",
author = "B. Liu and H. Jiang and Y. Huang and S. Qu and Yu, {M. F.} and Hwang, {K. C.}",
year = "2005",
month = "7",
day = "15",
doi = "10.1103/PhysRevB.72.035435",
language = "English (US)",
volume = "72",
journal = "Physical Review B - Condensed Matter and Materials Physics",
issn = "1098-0121",
publisher = "American Physical Society",
number = "3",

}

Atomic-scale finite element method in multiscale computation with applications to carbon nanotubes. / Liu, B.; Jiang, H.; Huang, Y.; Qu, S.; Yu, M. F.; Hwang, K. C.

In: Physical Review B - Condensed Matter and Materials Physics, Vol. 72, No. 3, 035435, 15.07.2005.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Atomic-scale finite element method in multiscale computation with applications to carbon nanotubes

AU - Liu, B.

AU - Jiang, H.

AU - Huang, Y.

AU - Qu, S.

AU - Yu, M. F.

AU - Hwang, K. C.

PY - 2005/7/15

Y1 - 2005/7/15

N2 - We have developed an accurate atomic-scale finite element method (AFEM) that has exactly the same formal structure as continuum finite element methods, and therefore can seamlessly be combined with them in multiscale computations. The AFEM uses both first and second derivatives of system energy in the energy minimization computation. It is faster than the standard conjugate gradient method which uses only the first order derivative of system energy, and can thus significantly save computation time especially in studying large scale problems. Woven nanostructures of carbon nanotubes are proposed and studied via this new method, and strong defect insensitivity in such nanostructures is revealed. The AFEM is also readily applicable for solving many physics related optimization problems.

AB - We have developed an accurate atomic-scale finite element method (AFEM) that has exactly the same formal structure as continuum finite element methods, and therefore can seamlessly be combined with them in multiscale computations. The AFEM uses both first and second derivatives of system energy in the energy minimization computation. It is faster than the standard conjugate gradient method which uses only the first order derivative of system energy, and can thus significantly save computation time especially in studying large scale problems. Woven nanostructures of carbon nanotubes are proposed and studied via this new method, and strong defect insensitivity in such nanostructures is revealed. The AFEM is also readily applicable for solving many physics related optimization problems.

UR - http://www.scopus.com/inward/record.url?scp=33749233507&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749233507&partnerID=8YFLogxK

U2 - 10.1103/PhysRevB.72.035435

DO - 10.1103/PhysRevB.72.035435

M3 - Article

VL - 72

JO - Physical Review B - Condensed Matter and Materials Physics

JF - Physical Review B - Condensed Matter and Materials Physics

SN - 1098-0121

IS - 3

M1 - 035435

ER -