An atomistic derivation of von-Kármán plate theory (Q2167930)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | An atomistic derivation of von-Kármán plate theory |
scientific article |
Statements
An atomistic derivation of von-Kármán plate theory (English)
0 references
1 September 2022
0 references
The purpose of the paper is to establish the von Karman plate theory from 3D and purely atomistic models with classical particle interaction. The authors start describing the atomistic models considering an open, bounded, connected, and nonempty set. \(S\subset \mathbb{R}^{2}=\mathbb{R}^{2}\times \{0\}\subset \mathbb{R}^{3}\), with Lipschitz boundary. They introduce a small parameter \(\varepsilon >0\) which describes the interatomic distance, the lattice \(\varepsilon \mathbb{Z}^{3}\), the number \(\nu \) of atom layers, the thickness \(h=(\nu -1)\varepsilon \) of the film, and sequences \( h_{n},\varepsilon _{n},\nu _{n},n\in \mathbb{N}\), such that \(\varepsilon _{n},h_{n}\rightarrow 0\). The macroscopic reference region is \(\Omega _{n}=S\times (0,h_{n})\) and the reference atoms of the film are \(\Lambda _{n}=\overline{\Omega }_{n}\cap \varepsilon _{n}\mathbb{Z}^{3}\). The atomic interaction energy for a deformation map \(w:\Lambda _{n}\rightarrow \mathbb{R }^{3}\) is written as \(E_{\mathrm{atom}}(w)=\sum_{x\in \Lambda _{n}^{\prime }}W(x, \overrightarrow{w}(x))\), where \(W(x,\cdot ):\mathbb{R}^{3\times 8}\rightarrow \lbrack 0,\infty )\) only depends on those \(\overrightarrow{w} _{i}\) with \(x+\varepsilon _{n}z^{i}\in \Lambda _{n}\). The authors consider a general representation of the energy functions \(W\) as combinations of cell \( W_{\mathrm{cell}}\) and surface \(W_{\mathrm{surf}}\) energies and they give examples of such energies. The energy contributions associated to body forces \(f_{n}:\Lambda _{n}\rightarrow \mathbb{R}^{3}\) are written as \(E_{\mathrm{body}}(w)=\sum_{x\in \Lambda _{n}}w(x)\cdot f_{n}(x)\), assuming that \(f_{n}\) does not depend on \( x_{3}\), that is \(f_{n}(x)=0\) in an atomistic neighborhood of the lateral boundary, and the overall energy is written as the sum \(E_{n}(w)=\frac{ \varepsilon _{n}^{3}}{h_{n}}(E_{\mathrm{atom}}(w)+E_{\mathrm{body}}(w))\). The main result of the paper establishes the \(\Gamma \)-convergence of the overall energy, the authors distinguishing the cases \(\nu _{n}\rightarrow \infty \) and \(\nu _{n}=\nu \).\ In the first (thin film) case, the sequence \(h_{n}^{4}E_{n}\) \( \Gamma \)-converges to \(E_{vK}\) with \(E_{vK}(u,v,R^{\ast })=\int_{S}(\frac{1}{ 2}Q_{2}(G_{1}(x^{\prime }))+\frac{1}{24}Q_{2}(G_{2}(x^{\prime }))+f(x^{\prime })\cdot v(x^{\prime })R^{\ast }e_{3})dx^{\prime }\), where \( G_{1}(x^{\prime })=sym\nabla ^{\prime }u(x^{\prime })+\frac{1}{2}\nabla ^{\prime }v(x^{\prime })\otimes \nabla ^{\prime }v(x^{\prime })\) and \( G_{2}(x^{\prime })=-(\nabla ^{\prime })^{2}v(x^{\prime })\). In the second (ultrathin film) case, the sequence \(h_{n}^{4}E_{n}\) \(\Gamma \)-converges to a limit energy \(E_{vK}^{(\nu )}\), with an explicit expression quite similar to the previous one, but much more complicated. Here the authors assume that there is a \(c_{0}>0\) such that \(W_{\mathrm{cell}}(A)\geq c_{0}\mathrm{dist}^{2}(A;SO(3)Z)\), for all \(A\in \mathbb{R} ^{3\times 8}\) satisfying \(\sum _{i=1}^{8}A_{.i}=0\). In further results, the authors replace this hypothesis now assuming that \(W_{\mathrm{cell}}(A)=W_{\mathrm{cell}}(-A)\), there is some neighborhood \(U\) of \(O(3)Z\) and a \(c_{0}>0\) such that \( W_{\mathrm{cell}}(A)\geq c_{0}\mathrm{dist}^{2}(A;SO(3)Z)\), for all \(A\in U\) satisfying \( \sum_{i=1}^{8}A_{.i}=0\) and \(W_{\mathrm{cell}}(A)\geq c_{0}\) for all \(A\notin U\) satisfying \(\sum_{i=1}^{8}A_{.i}=0\), \(\nu _{n}^{5}\varepsilon _{n}^{2}\rightarrow 0\), and \(f_{n}=0\), and they prove quite similar \(\Gamma \)-convergence properties of \(E_{n}\). For the proof of the \(\Gamma \)-convergence results, the authors need a modification scheme which guarantees that the rigidity and displacements of boundary cells can be controlled in terms of the displacements, respectively, rigidity of inner cells. They use a piecewise affine interpolation scheme which associates the exact average of atomic positions to the center and to the faces of lattice cells. They finally use compactness results.
0 references
discrete to continuum
0 references
gamma-convergence
0 references
dimension reduction
0 references
elastic plate
0 references
von-Kármán plate theory
0 references