A General Framework of Linear Elasticity Enhanced Multiscale Coupling Methods for Crystalline Defects

Yanbo Zhan, Yangshuai Wang, Hao Wang · 2025

Read A General Framework of Linear Elasticity Enhanced Multiscale Coupling Methods for Crystalline Defects by Yanbo Zhan, Yangshuai Wang, Hao Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-17164.txt.

Introduction

Simulating crystal defects at two scales at once is refined by Yanbo Zhan, Yangshuai Wang, and Hao Wang in a 2025 paper. Full atom models are exact near a defect but too slow for large samples. Far from the defect, a smooth stretch model is cheap and close enough, so coupling methods join the two zones.

Both zones are usually nonlinear, which keeps costs high. The new method swaps the far zone for a linear stretch model. Careful error analysis shows the cheaper method converges at the same rate as the classic one. Number tests back the proof. Materials teams gain a faster path to study cracks and dislocations in large samples.

Catalog record

CREATORYanbo Zhan, Yangshuai Wang, Hao Wang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE92 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-17164

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A General Framework of Linear Elasticity Enhanced Multiscale Coupling Methods for Crystalline Defects Yanbo Zhan, Yangshuai Wang, Hao Wang Abstract . The atomistic-to-continuum (a/c) coupling methods, also known as the quasicontinuum (QC) methods, are a important class of concurrent multisacle methods for modeling and simulating materials with defects. The a/c methods aim to balance the accuracy and efficiency by coupling a molecular mechanics model (also termed as the atomistic model) in the vicinity of localized defects with the Cauchy-Born approximation of the atomistic model in the elastic far field. However, since both the molecular mechanics model and its Cauchy-Born approximation are usually a nonlinear, it potentially leads to a high computational cost for large-scale simulations. In this work, we propose an advancement of the classic quasinonlocal (QNL) a/c coupling method by incorporating a linearized Cauchy-Born model to reduce the computational cost. We present a rigorous a priori error analysis for this QNL method with linear elasticity enhancement (QNLL method), and show both analytically and numerically that it achieves the same convergence behavior as the classic (nonlinear) QNL method by proper determination of certain parameters relating to domain decomposition and finite element discretization. More importantly, our numerical experiments demonstrate that the QNLL method perform an substantial improvement of the computational efficiency in term of CPU times. 1. Introduction Crystalline defects, such as vacancies, dislocations, and cracks, play a crucial role in determining mechanical behavior of the materials [ 32 , 28 , 45 , 43 , 7 , 50 , 6 ] . While the defect core undergoes significant lattice distortions, the far-field region can be effectively modeled as an elastic force field [ 4 , 8 , 5 , 31 ] . By leveraging this distinction, concurrent multiscale methods form an important class of methodology to simulating such systems. In order to achieve a balance between accuracy and computational efficiency, these methods decompose the computational domain into a small core region , which is modeled with detailed atomistic scale approaches such as molecular mechanics, and a large far-field region, which is often described by macroscopic continuum elasticity that can be further discretized by the finite element method to reduce the total degrees of freedoms (DoFs). Concurrent multiscale methods may date back to 1980’s and it was in [ 32 ] that the Cauchy-Born approximation was introduced to the methods to initiate the atomisti-to-continuum (a/c) coupling methods, which are also well known as the quasicontinuum (QC) methods among the engineering community [ 27 , 28 , 26 , 16 , 17 , 49 ] . Cauchy-Born approximation is essentially the continuum elasticity model derived the from a molecular mechanics model by certain homogenization process [ 24 , 13 , 25 , 19 , 54 ] . It is an important step in concurrent multiscale modeling to guarantee that the models of different scales produce the same result under elastic deformation, which is also known as the first level consistency of the method [ 11 ] . It was later discovered that the energy-based a/c coupling methods suffer from the so-called “ghost force” at the interface between the regions where models of different scales transit [ 41 ] . Such phenomenon arises due to the nonlocality of the molecular mechanics model and the Cauchy-Born approximation which is continuum and thus local in nature. This is also known as the second level of consistency [ 11 ] . Considerable effort has been made in the recent two decades for the development and the mathematical analysis of the consistent a/c coupling methods so that the interface error is controllable and will not ruin the overall accuracy of the solution. Shimokawa et al. first proposed the Quasi-Nonlocal (QNL) method [ 42 ] , which eliminates the ”ghost force” by reconstructing the energy of atoms near the interface. E, Lu and Yang proposed the Geometric Reconstruction (GRC) method [ 12 ] , which is used to eliminate the ”ghost force” in higher-dimensional cases. Ming and Yang, Dobson and Luskin conducted error analysis on the QNL method [ 29 , 10 ] . Ortner, Ortner and Wang performed a priori and a posteriori error analysis on the QNL and QNL with coarse-graining methods based on negative norms [ 34 , 37 ] . Ortner and Zhang proposed the two-dimensional geometric reconstruction-based consistent atomistic/continuum (GRAC) method [ 40 , 33 ] , which extends the QNL method to two-dimensional cases. We refer to [ 26 ] for a thorough review of the QNL method, and we will follow its theoretical framework to complete some of the proofs in this paper. The a/c methods improves the computational efficiency compared with the full molecular mechanics simulations in terms of the substantial reduced DoFs as a result of the finite element discretization of the continuum elasticity model. Therefore existing works concentrate mainly on the convergence rate for different a/c methods, which illustrate that for a given error how many DoFs are required or vice versa [ 2 , 30 , 15 , 20 ] (cf. Figure 2 , 3 and 4 in the current paper). However, the molecular mechanics model we employ in the core region is usually highly nonlinear and thus is the corresponding Cauchy-Born approximation in the elastic far field as well as the a/c coupling model in whole. The high nonlinearity of the a/c model may still results in an excessive computational cost, especially in large scale simulations, which motivates this study. The current work focuses on developing and analyzing an energy-based a/c coupling methods that integrates the classic (nonlinear) quasinonlocal (QNL) method [ 42 , 34 , 23 ] with the linearized Cauchy-Born approximation, which we name as the QNLL method. The purpose of such development is to enhance computational efficiency of the a/c method through the proper application of linearized model in the elastic far-field region while retain the same level of accuracy as the fully nonlinear coupling scheme. We then perform a rigorous a prior analysis of the QNLL method. We identify that the optimal rate of convergence of the method can be achieved by choosing certain parameters according to the regularity of the defect equilibrium. In particular, by balancing the lengths of the regions where different models apply and the size of the finite element discretization, we obtain a sharp error estimate of the QNLL method which retain the same order of convergence as that of the fully nonlinear QNL method. We validate our theoretical results through a series of numerical experiments. In addition, our numerical experiments demonstrate that QNLL achieve a substantial improvement of computational efficiency compared with the fully nonlinear QNL method in terms of CPU times. To set out our ideas in a clear way, we consider a one dimensional atomistic system with nearest and next-nearest neighbor multibody interactions. However, we note that the error analysis and balancing techniques proposed are general and are expected to be extended to higher dimensions problems and other fully nonlinear a/c coupling methods with finite interaction ranges [ 52 , 39 ] . This study serves as a proof of concept, laying the foundation for possible further extension, which are discussed in the conclusion. 1.1. Outline The paper is organized as follows. Section 2 introduces

Read the complete text