A three-parameter structurally motivated robust constitutive model for isotropic incompressible unfilled and filled rubber-like materials (Q2671677)

In the paper the authors provide a comparison between the model proposed by the first author [Math. Mech. Solids 26, No. 11, 1660--1674 (2021; Zbl 07589910)] and the other three-parameters models, existing in the literature, for isotropic incompressible and hyper-elastic materials. Related to the results presented in the paper, we pay attention to the following features, pointed out by \textit{C. Truesdell} and \textit{W. Noll} [The nonlinear field theories of mechanics. Edited and with preface by Stuart S. Antman. 3rd edition. Berlin: Springer (2004; Zbl 1068.74002)]: \begin{itemize} \item[1.] The constitutive equations for isotropic incompressible hyper-elastic materials can be expressed by \begin{gather*} \mathbf{T} = -p \mathbf{ I} + 2 W_1 \mathbf{B} + 2 W_2 \mathbf{B}^{-1}, \quad \mathbf{B}= \mathbf{F} \mathbf{F}^T, \\ W_{j} = \frac{\partial W}{\partial I_{j}},\quad j =1,2, \quad I_1 = \mathrm{tr} \mathbf{B}, \quad \det\mathbf{B} =1, \end{gather*} where \(I_2\) is the second invariant of \(\mathbf{B}\), \(W = W(I_1, I_2)\) is the strain energy function, \(\mathbf{F}\) denotes the deformation gradient, \(\mathbf{B}\) is the left Cauchy-Green deformation tensor, and \(\mathbf{T}\) is the Cauchy stress tensor. \item[2.] The hydrostatic pressure, \(p\), remains undefined until an appropriate boundary value problem is formulated. \end{itemize} The Anssari-Benam's model is derived from a three-parameter strain energy function \(W\), which is dependent on the first invariant of \(\mathbf{B}\), only. The model is applicable to various types of isotropic rubber-like materials. The authors employed a fitting procedure to characterize the model parameters. Firstly, the one-dimensional relationships between the appropriate principal stresses, \(\mathbf{T}\), and stretches, \(\lambda\), (or between the shear stress and shear, \(\gamma\), i.e. the formula (24)) are associated with (1), when \(W_2=0\). Secondly, these deformation processes are simultaneously associated to certain data sets existing in the literature. Finally, the best fit is achieved by the minimization of the residual sum of square functions defined through the differences between the stress values given by the model and the experimental data, respectively. The minimization is performed with respect to data points. The material parameters for all considered models are determined and several comparative plots are presented in order to illustrate the modeling results for the associated one-dimensional deformation processes and the computed errors, respectively. The application of the proposed model to the biomechanics of soft tissues is also performed. The authors conclude that their ``proposed model most favorably capture the data-sets for all unfilled, filled rubber and polymer specimens considered.'' We remark that the data set is considered for the one-dimensional deformation processes only. Moreover, the authors do not say anything about the assigned values for pressure \(p\), in terms of the stretches, involved in the one-dimensional deformation processes. Concerning the formula (23) related to the pure shear, \({T}_{ps}\) ought to be replaced by the difference between two principal stresses, namely \(T_1 - T_2\), corresponding to the principal stretch \(\lambda^2\) and \(\frac{1}{\lambda^2}\), respectively. Here the principal stretches \(\lambda^2\) depends on the shear \(\gamma\), while the formula (24) holds for one-dimensional process only if the shear \(\gamma\) is small. (See the calculus and the following remark by Truesdell and Noll [loc. cit.] on pages 174 and 175, ``In the infinitesimal theory the shear stress suffices to produce shear\dots. In fact, normal stresses, negligible only in small shears, are required if a simple shear is to be effected.'') Consequently, even if the material parameters have been calibrated via the procedure mentioned above, which means that the models provide similar aspects of certain one-dimensional processes, the response of the isotropic incompressible hyper-elastic model may be completely different, when the boundary problems are solved, as the unknowns are the stretch tensor \(\mathbf{B}\) together with the pressure \(p\).