doi:10.1016/S0020-7683(02)00382-7
Copyright © 2002 Elsevier Science Ltd. All rights reserved.
Elasticity of DLCA model gels with loops
Hang-Shing Ma
,
, a, Jean-H. Prévostb and George W. Schererc
a Department of Chemical Engineering, Princeton University, Princeton, NJ 08544, USA
b Department of Civil and Environmental Engineering, Princeton University, Princeton, NJ 08544, USA
c Department of Civil and Environmental Engineering & Princeton Materials Institute, Princeton University, Princeton, NJ 08544, USA
Received 8 April 2002;
revised 20 May 2002.
Available online 28 August 2002.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Aggregation is a common natural phenomenon, but the structure–property relationship of the resulting porous, fractal gels is not well understood. An earlier study using the diffusion-limited cluster–cluster aggregation model revealed that loop structure is lacking in the model gels to account for their mechanical properties. The dangling bond deflection model was then developed to address loop formation during the aggregation process. This article describes the finite element method implemented to measure the various moduli of the resulting gel structure, which was modeled as a network of linearly elastic beams. The well-known empirical correlation in gels––the power–law scaling of Young's modulus with relative densities––was reproduced, and the scaling exponent of about 3.6, which is consistent with the experimental results, was captured in the analysis. About 70% of the total strain energy in the network came from bending of the beams. The contrast in the scaling exponent as compared to the open-cell foam model is attributed to the change of connectivity in the gel network when the density of the aggregate is adjusted.
Author Keywords: Aerogel; Sol–gel; Diffusion-limited cluster–cluster aggregation with dangling bond deflection; Finite element method; Elasticity; Beam theory; Fractal; Modulus–density scaling
Fig. 1. Effect of network size b on bulk modulus K for DLCADEF networks of bond compliance factor F=100 and various densities ρ (the number underneath the symbols denote the total number of samples).
Fig. 2. Scaling of bulk modulus K with relative density ρ at various bond compliance factor F of the DLCADEF networks (data of ρ=0.018 are for illustration only; regression shows mK=3.58 for data of F=100).
Fig. 3. Dependence of Poisson's ratio νp on relative density ρ at various bond compliance factor F of the DLCADEF networks.
Fig. 4. Scaling of Young's modulus E with relative density ρ at various bond compliant factor F of the DLCADEF networks (regression shows m=3.55 for data of F=100).
Fig. 5. (a) Two-dimensional DLCADEF network (node represents particle and link represents bond), (b) with solid line representing compressed state and broken line representing unstrained state.
Fig. 6. Contribution of different modes of deformation of the beams to the total strain energy (and modulus) of the DLCADEF network.
Table 1. Scaling exponents of modulus-density relationships for DLCADEF model gels of various bond compliance factor F under different forms of deformation
