Application of J J -Integral to a Random Elastic Medium
Jan Eliáš, Josef Martinásek, Jia-Liang Le · 2025
Read Application of J J -Integral to a Random Elastic Medium by Jan Eliáš, Josef Martinásek, Jia-Liang Le (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-2501-03156.txt.
Introduction
Jan Elias, Josef Martinasek, and Jia-Liang Le extend a fracture tool to random materials in a 2025 paper. The J-integral estimates energy release at a crack tip, and it famously gives the same answer along any path. Real materials vary in stiffness from spot to spot, so the team modeled stiffness as a random field and simulated many samples.
Path independence survived for the average energy release rate. It failed for higher moments, like variance. Longer correlation in the stiffness field changed the rates too. Readers in fracture mechanics get clear limits for using the tool on uneven materials.
Catalog record
| CREATOR | Jan Eliáš, Josef Martinásek, Jia-Liang Le |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-03156
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Application of J J -Integral to a Random Elastic Medium Jan Eliáš, Josef Martinásek, Jia-Liang Le Abstract Abstract This study investigates the use of the J -integral to compute the statistics of the energy release rate of a random elastic medium. The spatial variability of the elastic modulus is modeled as a homogeneous lognormal random field. Within the framework of Monte Carlo simulation, a modified contour integral is applied to evaluate the first and second statistical moments of the energy release rate. These results are compared with the energy release rate calculated from the potential energy function. The comparison shows that, if the random field of elastic modulus is homogenous in space, the path independence of the classical J -integral remains valid for calculating the mean energy release rate. However, this path independence does not extend to the higher order statistical moments. The simulation further reveals the effect of the correlation length of the spatially varying elastic modulus on the energy release rate of the specimen. 1 Introduction Over the past century, linear elastic fracture mechanics (LEFM) has been developed into a complete subject that plays a pivotal role in a variety of engineering disciplines. The cornerstone concept in LEFM is the energy release rate proposed by Griffith [ 8 ] , which provided a definitive answer to the fundamental question on why structures with a preexisting crack would have a finite load carrying capacity even though the elastic stress at the crack tip is infinite. Within the framework of LEFM, various methods have been developed to calculate the energy release rate. Among them, the most prominent one is Rice’s J -integral [ 18 ] . The J -integral has a deep mathematical root, which can be traced back to the celebrated Noether’s work on the invariant variational principle [ 16 ] . Meanwhile, it has been well understood that this integral is closely related to the Eshelby tensor [ 6 , 5 , 10 , 2 ] , a fundamental concept leading to the derivation of configurational forces for a broad class of problems in solid mechanics [ 9 ] . It is also noted that the J -integral can be applied to nonlinear elasticity, which provides a good approximation to the yielding zone of plastic-hardening metals as long as there is no unloading [ 1 , 3 ] . The most salient property of the J -integral is its path independence. This property allows one to choose any convenient path to calculate the energy release rate. However, the path independence property is valid only if the material is homogeneous at least in the direction of the crack [ 18 ] . It has been shown that for non-homogeneous materials the original form of the J -integral needs to be modified in order to preserve the path independence property [ 4 , 10 , 12 ] . The modified J -integral has successfully been used to calculate the stress intensity factors for various non-homogeneous materials, such as the functionally graded materials [ 12 ] . Many engineering structures are composed of statistically homogeneous materials, in which the elastic properties could exhibit some degree of spatial randomness and at any point they share the same probability distribution. Some typical examples include concrete, rock, particulate composites, and ceramics. Due to the intrinsic randomness of material properties, the energy release rate of a specimen would be random. Quantifying the statistics of the energy release rate is an essential step for the estimation of the randomness of the fracture energy, which is a key material property for failure analysis. However, so far few studies have applied the J -integral to statistically homogeneous materials. Based on the theory of random fields, the spatial randomness of the elastic properties of statistically homogeneous materials can be mathematically represented by homogeneous random fields, and the statistics of the energy release rate can then be calculated through the Monte Carlo simulations. It is evident that, for each realization of the random fields of material properties, the classical J -integral is inapplicable since the material is inhomogeneous. On the other hand, it is noted that, in an average sense, the material is homogeneous. It is reasonable to expect that Rice’s J -integral might still be applicable for calculating the mean value of the energy release rate. However, this conjecture has not been proven mathematically. Furthermore, the consideration of spatial randomness of elastic modulus would naturally introduce the correlation length to the problem. The effect of this length scale on the overall statistics of the energy release rate including its mean value is largely not understood. To answer these questions, this study conducts theoretical and numerical investigations on the application of J -integral to a random elastic media, which is highly relevant to the analysis of fracture behavior of large-scale structures made of brittle heterogeneous materials. 2 Theoretical Consideration Consider two-dimensional analysis of a point-wise isotropic elastic medium with a preexisting crack. The elastic stiffness tensor at any local point is expressed by C_{ijkl}(x_{1},x_{2})=\mu(x_{1},x_{2})(\delta_{ik}\delta_{jl}+\delta_{il}\delta_{jk})+\lambda(x_{1},x_{2})\delta_{ij}\delta_{kl} , where x_{1},x_{2}= local coordinate, \mu,\lambda= spatially varying elastic constants, \delta_{mn}= Kronecker delta. The elastic constants \mu and \lambda are related to the Young modulus E and Poisson ratio \nu by \mu=E/2(1+\nu) and \lambda=E\nu/(1+\nu)(1-2\nu) . In this study, the spatial randomness of the elastic stiffness tensor is governed by a homogeneous random field of Young modulus E(x_{1},x_{2}) . The Poisson ratio is considered to be deterministic. Figure 1: Calculation of the energy release rate from an arbitrary domain containing the crack tip. Following [ 17 ] , let \Omega be a finite region enclosing the crack tip, and the boundary of region \Omega is denoted by \Gamma (Fig. 1 ). The energy release rate can be written by {\cal{G}}=\lim_{\Delta a\rightarrow 0}\frac{1}{\Delta a}\left\{\int_{\Gamma}(t_{i}^{0}+\Delta t_{i})\Delta u_{i}\mbox{d}s-\int_{\Omega}\left[\bar{U}(\epsilon_{mn}^{0}+\Delta\epsilon_{mn})-\bar{U}(\epsilon_{mn}^{0})\right]\mbox{d}x_{1}\mbox{d}x_{2}\right\} (1) where t_{i}^{0}= original traction on the boundary \Gamma at crack length a_{0} , \Delta t_{i}= traction increment as the crack extends by an amount of \Delta a , \Delta u_{i}= displacement increment on \Gamma , \epsilon_{mn}^{0}= original strain field, \Delta\epsilon_{mn}= strain increment, and \bar{U}=\frac{1}{2}C_{ijkl}\epsilon_{ij}\epsilon_{kl}= strain energy density. To evaluate Eq. ( 1 ), we introduce a moving coordinate system (X_{1},x_{2}) , in which the origin is attached to the crack tip. Therefore, we have X_{1}=x_{1}-a , where a= crack length, and the total differential of a physical quantity with respect to crack length a is given by \mbox{d}(\cdot)/\mbox{d}a=\partial(\cdot)/\partial a-\partial(\cdot)/\partial X_{1}=\partial(\cdot)/\partial a-\partial(\cdot)/\partial x_{1} . The line integral in Eq. ( 1 ) can then be written by \lim_{\Delta a\rightarrow 0}\frac{1}{\Delta a}\int_{\Gamma}(t_{i}^{0}+\Delta t_{i})\Delta u_{i}\mbox{d}s=\int_{\Gamma}t_{i}^{0}(X_{1},x_{2})|_{a_{0}}\left(\frac{\partial u(X_{1},x_{2})}{\partial a}\big|_{a_{0}}-\frac{\partial u(X_{1},x_{2})}{\partial X_{1}}\big|_{a_{0}}\right)\mbox{d}s (2) For the domain integral in Eq. ( 1 ), we first note that the difference in strain energy density is strictly due to the change in strain as the crack extends. However, when we perform the total differentiation of the strain energy density function, the term \partial\bar{U}/\partial x_{1} includes the change in strain energy density due to the spatial variation of the elastic constants. Therefore, we need to subtract this contribution from the total differentiation, which leads to \lim_{\Delta a\rightarrow 0}\frac{1}{\Delta a}\int_{\Omega}\left[\bar{U}(\epsilon_{mn}^{0}+\Delta\epsilon_{mn})-\bar{U}(\epsilon_{mn}^{0})\right]\mbox{d}x_{1}\mbox{d}x_{2}=\int_{\Omega}\left\{\frac{\partial\bar{U}(X_{1},x_{2})}{\partial a}\big|_{a_{0}}-\left[\frac{\partial\bar{U}(X_{1},x_{2})}{\partial X_{1}}\big|_{a_{0}}-\frac{1}{2}C_{ijkl,1}\epsilon_{ij}\epsilon_{kl}|_{a_{0}}\right]\right\}\mbox{d}X_{1}\mbox{d}x_{2} (3) where C_{ijkl,1}=\partial C_{ijkl}/\partial x_{1}=\partial C_{ijkl}/\partial X_{1} . Note that \partial\bar{U}/\partial a=\sigma_{ij}\partial\epsilon_{ij}/\partial a . With the strain-displacement relation \epsilon_{ij}=\frac{1}{2}(u_{i,j}+u_{j,i}) and the equilibrium equation \sigma_{ij,j}=0 , it is easy to show