In search of constitutive conditions in isotropic hyperelasticity: polyconvexity versus true-stress-true-strain monotonicity

This paper demonstrates that neither polyconvexity nor true-stress-true-strain monotonicity alone guarantees physically reasonable behavior in isotropic hyperelasticity, suggesting that while their combination is a promising solution to Truesdell's Hauptproblem, no global strain-energy function satisfying both conditions has yet been identified.

The Gist

Imagine you are a master architect designing a building made of a special, stretchy material. Your goal is to write a set of rules (a "constitutive law") that predicts exactly how this material will behave when you pull, push, or twist it. You want to be sure that your rules never predict something impossible, like the material suddenly snapping back when you pull it harder, or behaving wildly differently just because you rotated the building.

In the world of physics, this is the "Hauptproblem" (Main Problem): How do we write these rules so they are mathematically sound and physically realistic?

This paper explores two famous sets of rules that scientists have proposed to solve this problem. The authors, Wollner, Holzapfel, and Neff, act like detectives testing these rules against each other. They ask: "If a material follows Rule A, does it automatically follow Rule B?"

Here is the breakdown of their investigation using simple analogies.

The Two Contenders

1. Polyconvexity (The "Mathematical Safety Net")
Think of Polyconvexity as a strict mathematical safety net. It's a rule that ensures the building won't collapse into a mathematical black hole (where solutions don't exist). It's very popular in computer simulations because it's easy to check.

• The Promise: If you use this rule, the math works out, and the material won't do weird, impossible things in the equations.
• The Catch: The authors found that just because a material passes this "safety net" test, it doesn't mean it behaves like a real, sensible material in every situation.

2. TSTS-M++ (The "Common Sense Monotony")
Think of TSTS-M++ (True-Stress-True-Strain Monotonicity) as a rule of "Common Sense." It says: "If you pull the material harder, the force required to pull it should keep increasing. If you twist it more, the resistance should keep increasing." It's like stretching a rubber band; it should get harder to stretch the further you go, not suddenly get easier.

• The Promise: This rule guarantees that the material behaves predictably in specific tests, like pulling it straight or twisting it.
• The Catch: This rule is also not a magic bullet. A material can follow this rule and still behave strangely in other ways.

The Investigation: Testing the Rules

The authors set up two specific challenges to see if one rule could replace the other.

Challenge 1: The Stretching Test (Uniaxial Extension)

• The Scenario: Imagine pulling a block of material straight out, like taffy.
• The Question: If a material follows the "Mathematical Safety Net" (Polyconvexity), will it always get harder to pull as you stretch it?
• The Result: No. The authors built a specific mathematical model (a "fake material") that passed the Polyconvexity test perfectly. However, when they simulated pulling it, the force required to stretch it went up, then suddenly went down before going up again.
• The Analogy: It's like a car that is mathematically guaranteed to be safe, but when you press the gas pedal, it speeds up, then suddenly slows down on its own, then speeds up again. That's not how a real car (or a real material) should behave.
• Conclusion: Polyconvexity alone is not enough to guarantee "Common Sense" behavior when stretching.

Challenge 2: The Twisting Test (Simple Shear)

• The Scenario: Imagine sliding the top of a deck of cards sideways while holding the bottom still. This is "shear."
• The Question: If a material follows the "Common Sense" rule (TSTS-M++), will it always get harder to twist as you twist it more?
• The Result: No. The authors built another "fake material" that followed the Common Sense rule perfectly. But when they simulated twisting it, the resistance went up, then dropped, then went up again.
• The Analogy: Imagine a door hinge that gets harder to push open, then suddenly becomes loose and easy to push, then gets hard again. This violates the "Mathematical Safety Net" (specifically a condition called Legendre-Hadamard ellipticity, which ensures stability).
• Conclusion: Common Sense (TSTS-M++) alone is not enough to guarantee the mathematical stability required for twisting.

The Big Picture: The Missing Link

The authors conclude that neither rule is strong enough on its own.

• You need Polyconvexity to ensure the math is stable (no wild oscillations in twisting).
• You need TSTS-M++ to ensure the material behaves sensibly when stretched (force always increases with stretch).

The Ultimate Goal: The "Holy Grail" of this field is to find a single set of rules that satisfies both conditions at the same time for all possible deformations.

• Current Status: The authors tried very hard to find this "perfect material" but couldn't find one that works globally (for all stretches and twists).
• Partial Success: They did find some "chain-limited" solutions. Think of these as materials that behave perfectly, but only up to a certain limit (like a rubber band that works great until it reaches a specific length, at which point the rules break down).

Summary for the General Audience

This paper is a reality check for scientists designing materials. It says: "Don't just rely on one mathematical trick to ensure your material model is good."

• If you only check for Mathematical Safety (Polyconvexity), your material might act weird when you stretch it.
• If you only check for Common Sense (TSTS-M++), your material might act unstable when you twist it.

To truly solve the problem of modeling ideal elastic materials, we likely need a combination of both rules. However, finding a single formula that satisfies both perfectly for every possible situation remains an unsolved mystery, though the authors have provided new tools and partial answers to help future researchers crack the code.

Technical Summary

Technical Summary: In search of constitutive conditions in isotropic hyperelasticity: polyconvexity versus true-stress-true-strain monotonicity

Problem Statement
The paper addresses the "Hauptproblem" of finite elasticity: identifying appropriate constitutive constraints for the strain-energy function $W$ that ensure physically reasonable material behavior in ideal isotropic hyperelasticity. While the mathematical merits of polyconvexity are well-established (ensuring existence of minimizers and quasiconvexity), its mechanical consequences are less clear. Conversely, the true-stress-true-strain monotonicity (TSTS-M++) condition, recently rediscovered, ensures monotonic stress-strain trajectories in specific deformation modes but lacks the broad mathematical stability guarantees of polyconvexity. The central question is whether either condition alone is sufficient to guarantee physically reasonable behavior, or if a combination is required, and if such a combination can be realized globally.

Methodology
The authors employ a rigorous analytical approach based on the theory of isotropic hyperelasticity. Key methodological elements include:

• Invariant Representation: The strain-energy function is parametrized using a specific set of invariants $K_i$ (square roots of the principal invariants of the left Cauchy-Green tensor $\mathbf{B}$), which simplifies the representation of constitutive inequalities compared to principal stretches.
• Derivation of Sufficient Conditions: The paper derives new, explicit sufficient conditions for both polyconvexity and TSTS-M++ in terms of the derivatives of the energy function with respect to the invariants $K_1, K_2, K_3$.
• Explicit Counter-Examples: To test the sufficiency of these conditions, the authors construct specific families of strain-energy functions. These functions are designed to satisfy one condition (e.g., polyconvexity) while explicitly violating the expected physical behavior associated with the other (e.g., monotonicity in uniaxial extension).
• Analytical Proofs: The study relies on symbolic differentiation, spectral decomposition, and the analysis of ordinary differential equations (specifically Lemma 5.15) to prove properties of the constructed energy functions without resorting to large-scale numerical computation, except for visualization.

Key Contributions and Results
The paper provides definitive answers to four "challenge questions" posed by Neff et al. (2024) regarding the interplay between polyconvexity and TSTS-M++:

1. Insufficiency of Polyconvexity in Uniaxial Extension (Challenge ii & iii):

   • Compressible Case: The authors construct a polyconvex strain-energy function (Eq. 5.37) that yields a non-monotonic Cauchy stress response in unconstrained uniaxial extension. Specifically, the stress $\sigma_{11}$ reaches a maximum and then decreases as the stretch $\lambda_1$ increases, despite the function being polyconvex and satisfying the Legendre-Hadamard ellipticity condition.
   • Incompressible Case: While a counter-example was not found, the authors prove that any incompressible strain-energy function satisfying the sufficient conditions for polyconvexity proposed by Ball (1976) automatically results in a monotonic true-stress response. This significantly narrows the search space for a potential counter-example but does not prove impossibility.

2. Insufficiency of TSTS-M++ in Simple Shear (Challenge iv):

   • The authors construct a strain-energy function (Eq. 5.69) that satisfies TSTS-M++ globally but exhibits a non-monotonic Cauchy shear stress response in simple shear.
   • This result implies that the function is not rank-one convex (and thus not polyconvex), as rank-one convexity is proven to be a necessary condition for monotonic shear stress in simple shear (Proposition 5.13).

3. Chain-Limited Solutions (Challenge i):

   • The authors attempt to find a single function satisfying both polyconvexity and TSTS-M++ globally. They are unable to identify such a function defined over the entire domain $GL^+(3)$.
   • However, they successfully construct three families of "chain-limited" strain-energy functions (Propositions 5.2, 5.4, 5.5) that satisfy both conditions within restricted domains of deformation (e.g., limited by line, area, or volume element averages).

4. New Sufficient Conditions:

   • The paper establishes explicit sufficient conditions for polyconvexity (Theorem 3.1) and TSTS-M++ (Theorem 4.4) in terms of the invariants $K_i$. These conditions highlight the structural differences between the two constraints, particularly regarding the convexity requirements on the energy function and its derivatives.

Significance and Claims
The paper concludes that neither polyconvexity nor TSTS-M++ is sufficient on its own to guarantee a physically reasonable material response for ideal isotropic elasticity.

• Polyconvexity ensures stability (Legendre-Hadamard ellipticity) and monotonic shear stress but fails to guarantee monotonic stress in uniaxial extension.
• TSTS-M++ ensures monotonic stress in uniaxial extension but fails to guarantee monotonic shear stress (and thus rank-one convexity).

The authors suggest that a viable solution to Truesdell's Hauptproblem likely requires a combination of both conditions. However, they explicitly state that no isotropic strain-energy function has been identified to date that satisfies both constraints globally. The construction of such a function remains an open problem. The paper posits that the derived sufficient conditions for both properties may be mutually exclusive in the global setting, though they can be reconciled in restricted (chain-limited) domains. The work serves as a cautionary note for constitutive modeling, particularly in data-driven approaches (e.g., neural networks) where polyconvexity is often used as the sole stability constraint, potentially overlooking the physical requirement of monotonic stress-strain trajectories in extension.