Constitutive Theory of Superelastic Materials

In the field of solid mechanics, traditional linear elasticity—governed by Hooke's Law—is sufficient for describing the behavior of metals and ceramics under small strains. However, when dealing with materials such as rubbers, elastomers, and various biological soft tissues, these linear assumptions fail. These materials exhibit hyperelasticity: the ability to undergo massive, non-linear deformations and still return to their original shape upon unloading.

Unlike linear elastic materials, hyperelastic materials are characterized by a highly non-linear stress-strain relationship and can sustain strains reaching hundreds or even thousands of percent. To accurately model such behavior, we must move beyond infinitesimal strain theory and adopt a framework based on finite deformation theory and thermodynamic potentials.

Kinematics of Finite Deformation

To describe the behavior of a material undergoing large displacements and large strains, we must establish a rigorous kinematic description that maps the material from its initial (reference) state to its deformed (current) state.

1. The Deformation Gradient Tensor

The fundamental quantity in finite deformation is the deformation gradient tensor, $\mathbf{F}$. It defines the mapping from a material point in the reference configuration ($\mathbf{X}$) to its position in the current configuration ($\mathbf{x}$):
$$\mathbf{F} = \frac{\partial \mathbf{x}}{\partial \mathbf{X}}$$
The determinant of this tensor, $J = \det(\mathbf{F})$, represents the ratio of the deformed volume to the initial volume. For many hyperelastic materials, such as vulcanized rubber, the volume change during deformation is negligible. Consequently, these are often modeled as incompressible materials, where $J = 1$.

2. The Right Cauchy-Green Deformation Tensor

A significant challenge in large deformation mechanics is distinguishing between actual material stretching and rigid-body rotations. To ensure material objectivity (frame invariance), we utilize the Right Cauchy-Green deformation tensor, $\mathbf{C}$:
$$\mathbf{C} = \mathbf{F}^T \mathbf{F}$$
$\mathbf{C}$ is a symmetric, positive-definite tensor that captures only the stretching components of the deformation. For isotropic materials, the constitutive response depends solely on the invariants of $\mathbf{C}$:

  • $\text{I}_1 = \text{tr}(\mathbf{C})$
  • $\text{I}_2 = \frac{1}{2} [(\text{tr}\mathbf{C})^2 - \text{tr}(\mathbf{C}^2)]$
  • $\text{I}_3 = \det(\mathbf{C}) = J^2$

The Thermodynamic Framework of Hyperelasticity

The defining characteristic of a hyperelastic material is that its stress state is derived from a scalar potential known as the Strain Energy Density Function ($W$). This function represents the elastic energy stored per unit of reference volume.

1. The Strain Energy Density Function

For an isotropic hyperelastic material, the energy density $W$ is a function of the deformation invariants:
$$W = W(\text{I}_1, \text{I}_2, \text{I}_3)$$
This approach ensures that the material's response is consistent with the laws of thermodynamics, specifically the requirement that the work done during a closed loading-unloading cycle is zero.

2. Deriving Stress from Energy

The relationship between the stored energy and the resulting stress is established through differentiation. The Second Piola-Kirchhoff stress tensor ($\mathbf{S}$), which acts in the reference configuration, is obtained by:
$$\mathbf{S} = 2 \frac{\partial W}{\partial \mathbf{C}}$$
In practical engineering applications, we are typically interested in the Cauchy stress ($\boldsymbol{\sigma}$), also known as the "true stress," which represents the force per unit area in the current, deformed configuration. The Cauchy stress is related to the Second Piola-Kirchhoff stress via the push-forward transformation:
$$\boldsymbol{\sigma} = \frac{1}{J} \mathbf{F} \mathbf{S} \mathbf{F}^T$$
For the specific case of incompressible materials ($J=1$), the stress expression must account for the fact that the pressure is not uniquely determined by the deformation alone. We introduce an indeterminate hydrostatic pressure $p$:
$$\boldsymbol{\sigma} = -p\mathbf{I} + 2 \mathbf{F} \frac{\partial W}{\partial \mathbf{C}} \mathbf{F}^T$$

Classical Hyperelastic Constitutive Models

Depending on the complexity of the material and the expected strain range, different mathematical forms of $W$ are employed.

1. The Neo-Hookean Model

Derived from the statistical mechanics of polymer chains, the Neo-Hookean model is the simplest hyperelastic formulation. It assumes that the energy depends only on the first invariant:
$$W = \frac{\mu}{2}(\text{I}_1 - 3)$$
where $\mu$ is the shear modulus. While computationally efficient and effective for small to moderate strains (typically < 30%), it fails to account for the "stiffening" effect observed in elastomers at higher stretches.

2. The Mooney-Rivlin Model

To capture more complex shear behavior, the Mooney-Rivlin model incorporates the second invariant:
$$W = C_{10}(\text{I}1 - 3) + C{01}(\text{I}_2 - 3)$$
This model provides a significantly better fit for medium-range deformations and is widely used in industrial applications involving rubber components.

3. The Ogden Model

The Ogden model departs from the invariant-based approach and instead expresses the energy density in terms of the principal stretches ($\lambda_1, \lambda_2, \lambda_3$):
$$W = \sum_{i=1}^N \frac{\mu_i}{\alpha_i} (\lambda_1^{\alpha_i} + \lambda_2^{\alpha_i} + \lambda_3^{\alpha_i} - 3)$$
By adjusting the parameters $\mu_i$ and $\alpha_i$, the Ogden model can be tuned to fit experimental data with extreme precision, making it the gold standard for simulating very large strains (> 700%).

Practical Application: Uniaxial Tension Analysis

To illustrate the transition from theory to physical reality, consider the uniaxial stretching of an incompressible Neo-Hookean material.

  1. Kinematics: If the material is stretched in the $X_1$ direction by a stretch ratio $\lambda$, the incompressibility constraint ($J=1$) dictates that the transverse stretches must be $\lambda_2 = \lambda_3 = 1/\sqrt{\lambda}$.
  2. Invariants: The first invariant becomes $\text{I}_1 = \lambda^2 + \frac{2}{\lambda}$.
  3. Stress Derivation: Using the constitutive relation $\boldsymbol{\sigma} = -p\mathbf{I} + \mu \mathbf{B}$ (where $\mathbf{B} = \mathbf{F}\mathbf{F}^T$ is the left Cauchy-Green tensor) and applying the boundary condition that lateral stresses are zero ($\sigma_{22} = \sigma_{33} = 0$), we find the hydrostatic pressure $p = \mu/\lambda$.
  4. Resulting Stress: The true axial stress is:
    $$\sigma_{11} = \mu \left( \lambda^2 - \frac{1}{\lambda} \right)$$
    This equation clearly demonstrates the non-linear growth of stress as the stretch $\lambda$ increases, a hallmark of hyperelastic behavior.

Summary

The constitutive theory of hyperelastic materials provides a robust mathematical bridge between microscopic energy storage and macroscopic mechanical response. The logical progression follows a clear hierarchy:
Deformation Gradient ($\mathbf{F}$) $\rightarrow$ Invariants ($\text{I}_n$) $\rightarrow$ Strain Energy Density ($W$) $\rightarrow$ Cauchy Stress ($\boldsymbol{\sigma}$).

Choosing the appropriate model is a balance between computational cost and required accuracy:

  • Neo-Hookean: Best for preliminary analysis and low-strain regimes.
  • Mooney-Rivlin: Ideal for medium-strain applications and shear-dominated loading.
  • Ogden: Necessary for high-fidelity simulations involving extreme deformations.