TY - JOUR
T1 - A multiscale projection method for the analysis of carbon nanotubes
AU - Qian, Dong
AU - Wagner, Gregory J.
AU - Liu, Wing Kam
N1 - Funding Information:
The support of this project to QD from the Ohio Board of Regents is gratefully acknowledged. WKL acknowledges the support from Army Research Office (ARO), National Science Foundation (NSF), the NSF IGERT and NSF Summer Institute on nano mechanics and materials.
Copyright:
Copyright 2008 Elsevier B.V., All rights reserved.
PY - 2004/5/7
Y1 - 2004/5/7
N2 - The main objective of this paper is to develop a multiscale method for the analysis of carbon nanotube systems. The multiscale coupled governing equations are first derived based on a multiscale partition, which consists of a coarse scale component represented by meshfree approximation, and a fine scale component to be resolved from molecular dynamics. It is noted that the coarse scale representation is present in the entire domain of the problem, and coexists with the fine scale representation in the enriched region. Using the projection properties of the partition, the decoupled non-linear set of equations are obtained and solved iteratively using Newton's method. In this multiscale analysis, a virtual atom cluster (VAC) model is also proposed in the coarse scale treatment. This model does not involve a stress update scheme employing the Cauchy-Born hypothesis, which has been widely used in the crystal elasticity theory. Finally, to approximate a curved surface at the nanoscale, meshfree approximations are introduced to interpolate a single layer of atoms. Unlike traditional shell or continuum elements, the geometric constraint is only imposed in two dimensions. The high order continuity property of the meshfree shape functions guarantees an accurate description of the geometry and thus the energy of the atomic bond. The accuracy and efficiency of the method are illustrated in the post-buckling analysis of carbon nanotube structures. To our knowledge, this is the first multiscale post-buckling analysis presented for such a nanoscale system.
AB - The main objective of this paper is to develop a multiscale method for the analysis of carbon nanotube systems. The multiscale coupled governing equations are first derived based on a multiscale partition, which consists of a coarse scale component represented by meshfree approximation, and a fine scale component to be resolved from molecular dynamics. It is noted that the coarse scale representation is present in the entire domain of the problem, and coexists with the fine scale representation in the enriched region. Using the projection properties of the partition, the decoupled non-linear set of equations are obtained and solved iteratively using Newton's method. In this multiscale analysis, a virtual atom cluster (VAC) model is also proposed in the coarse scale treatment. This model does not involve a stress update scheme employing the Cauchy-Born hypothesis, which has been widely used in the crystal elasticity theory. Finally, to approximate a curved surface at the nanoscale, meshfree approximations are introduced to interpolate a single layer of atoms. Unlike traditional shell or continuum elements, the geometric constraint is only imposed in two dimensions. The high order continuity property of the meshfree shape functions guarantees an accurate description of the geometry and thus the energy of the atomic bond. The accuracy and efficiency of the method are illustrated in the post-buckling analysis of carbon nanotube structures. To our knowledge, this is the first multiscale post-buckling analysis presented for such a nanoscale system.
KW - Carbon nanotubes
KW - Meshfree method
KW - Multiscale analysis
KW - Nanomechanics
KW - Virtual atom cluster
UR - http://www.scopus.com/inward/record.url?scp=1842686601&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=1842686601&partnerID=8YFLogxK
U2 - 10.1016/j.cma.2003.12.016
DO - 10.1016/j.cma.2003.12.016
M3 - Article
AN - SCOPUS:1842686601
SN - 0045-7825
VL - 193
SP - 1603
EP - 1632
JO - Computer Methods in Applied Mechanics and Engineering
JF - Computer Methods in Applied Mechanics and Engineering
IS - 17-20
ER -