A computational framework for rheologically complex thermo-visco-elastic materials

Pietro Lenarda,Marco Paggi
DOI: https://doi.org/10.48550/arXiv.2107.08796
2021-07-09
Abstract:Fractional calculus has been proved to be very effective in representing the visco-elastic relaxation response of materials with memory such as polymers. Moreover, in modelling the temperature dependency of the material functions in thermo-visco-elasticity, the standard time-temperature superposition principle is known to be ineffective in most of the cases (thermo-rehological complexity). In this work, a novel finite element formulation and numerical implementation is proposed for the simulation of transient thermal analysis in thermo-rehologically complex materials. The parameters of the visco-elastic fractional constitutive law are assumed to be temperature dependent functions and an internal history variable is introduced to track the changes in temperature which are responsible for the phase transition of the material. The numerical approximation of the fractional derivative is employed via the so called Grünwald-Letnikov approximation. The proposed model is used to numerically solve some test cases related to relaxation and creep tests conducted on a real polymer (Etylene Vynil Acetate), which is used as the major encapsulant of solar cells in photovoltaics.
Computational Physics,Soft Condensed Matter
What problem does this paper attempt to address?
The problem that this paper attempts to solve is that in thermodynamically complex materials, the traditional time - temperature superposition principle (TTS) cannot effectively describe the viscoelastic behavior of materials. Specifically, for materials such as Ethylene - Vynil - Acetate (EVA), their internal microstructures will undergo phase transitions as the temperature changes, resulting in significant differences in material properties at different temperatures. Such materials are called thermo - rheologically complex materials. The paper proposes a new finite element formulation. By introducing an internal variable related to time and temperature history, it simulates the behavior of these materials in transient thermal analysis. This method uses fractional - order derivatives to describe the viscoelastic response of materials and assumes that the parameters of the fractional - order viscoelastic constitutive law change with temperature. To handle the numerical approximation of fractional - order derivatives, the paper adopts the Grünwald - Letnikov approximation method. Specifically, the main contributions of the paper include: 1. **New constitutive model**: A new fractional - order viscoelastic constitutive model is proposed, in which material parameters (such as the fractional - order index \(\alpha\) and the pre - factor \(A\)) change with temperature. 2. **Temperature - dependent internal variable**: An internal variable \(\tau(t, T)\) is introduced to track the material phase transitions caused by temperature changes. 3. **Numerical approximation of fractional - order derivatives**: The Grünwald - Letnikov approximation method is used to numerically solve the fractional - order derivatives, thereby achieving accurate simulation of the viscoelastic response. 4. **Numerical verification**: The effectiveness of the proposed model is verified through a series of numerical experiments. In particular, the relaxation modulus of EVA materials at different temperatures is simulated and compared with the experimental results. In conclusion, this paper aims to provide a more accurate method to simulate the viscoelastic behavior of thermo - rheologically complex materials, especially when temperature changes cause material phase transitions.