🔬 condensed matter

Elastoplastic Modelling of Cyclic Shear Deformation of Amorphous Solids

This paper presents an energy-landscape-based elasto-plastic model that uses mesoscopic sub-volumes and finite element methods to successfully replicate various complex deformation phenomena in amorphous solids under both uniform and cyclic shear.

Original authors: Pushkar Khandare, Srikanth Sastry

Published 2026-02-10

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Pushkar Khandare, Srikanth Sastry

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The "Jiggling Jelly" Mystery: How Amorphous Solids Break

Imagine you have a giant bowl of jelly. If you push it gently, it wobbles and bounces back. If you push it hard, it might tear or slide apart.

Now, imagine that jelly isn't just one smooth substance, but is actually made of millions of tiny, microscopic "LEGO blocks" that are all jammed together. These blocks aren't neatly stacked like a real LEGO set; they are shoved in randomly, like a messy junk drawer. This is what scientists call an amorphous solid (like glass, certain metals, or even foams).

This paper, written by researchers in India, uses a clever computer model to figure out exactly how these "messy blocks" behave when you shake, squeeze, or repeatedly jiggle them.

1. The "Energy Landscape": The Valley and the Hill

To understand the model, imagine every tiny block lives in a small valley.

The Stable State: As long as the block stays in its valley, it’s happy and behaves like a solid.
The "Snap": If you push the block too hard, it climbs up the side of its valley, reaches the rim, and POPS! It falls into a completely different valley. This "pop" is what scientists call plastic deformation—it’s the moment the material stops being bouncy and starts permanently changing shape.

The researchers found that "well-annealed" materials (materials that were cooled very slowly and are very stable) live in very deep, steep valleys. "Poorly annealed" materials (cooled quickly) live in shallow, slippery valleys.

2. Uniform Squeezing vs. The "Jiggle Test"

The researchers tested two ways of stressing the material:

The Big Squeeze (Uniform Shear): This is like pushing the whole bowl of jelly at once. If the jelly is "messy" (poorly annealed), it flows smoothly like honey. If it is "well-organized" (well-annealed), it resists for a long time and then suddenly SNAPS, creating a "shear band"—a single, violent tear that runs through the material.
The Jiggle Test (Cyclic Shear): This is like shaking the bowl back and forth repeatedly. This is where things get weird.

3. The "Trenching" and the "Moving Band"

When you jiggle the material, two strange things can happen:

The Pinned Tear (Trenching): Imagine a crack forms in the jelly, but instead of moving, it gets stuck in one spot. Around the crack, the jelly becomes incredibly "stiff" and "deep," acting like a protective trench that keeps the crack from moving. The researchers call this trenching.
The Wandering Tear: In other cases, the crack doesn't stay put. It wanders around the bowl like a lost traveler, eventually visiting every single part of the jelly. This "wandering" actually helps the material settle into a more stable state.

4. Why does this matter? (The "Fatigue" Problem)

Have you ever noticed how a paperclip doesn't break the first time you bend it, but if you bend it back and forth twenty times, it suddenly snaps? That is fatigue.

The researchers found they could actually predict how many "jiggles" it would take before the material finally gave up and failed. They discovered that the "damage" builds up in a very specific mathematical pattern, almost like a countdown timer ticking toward zero.

The Big Picture

By building this digital "jelly bowl," the scientists are learning how to design better materials. If we can understand exactly how these microscopic "pops" and "wandering cracks" work, we can create stronger glasses, more durable metals for airplanes, and better materials for technology that won't snap unexpectedly when they are shaken or stressed.

Technical Summary: Elastoplastic Modelling of Cyclic Shear Deformation of Amorphous Solids

1. Problem Statement

Amorphous solids (e.g., metallic glasses, foams, and gels) exhibit complex mechanical responses under stress, including stress overshoots, shear banding, and a transition from ductile to brittle behavior depending on their preparation history (annealing level). While uniform shear is well-studied, the behavior of these materials under cyclic shear—which involves repeated loading and unloading—presents a much richer and more complex phenomenology. Specifically, the mechanisms behind the yielding transition, the "threshold energy" that separates poorly annealed from well-annealed samples, and the dynamic processes leading to fatigue failure remain areas of active investigation. The goal of this research is to develop a mesoscopic model capable of capturing these diverse phenomena.

2. Methodology

The authors develop an Elastoplastic Model (EPM) based on an energy-landscape framework. Unlike simpler models that use a single yield stress, this model incorporates the inherent heterogeneity of amorphous solids through the following components:

Mesoscopic Sub-volumes (Mesoblocks): The solid is modeled as an $L \times L$ grid of mesoscopic blocks. Each block occupies a "mesostate" (a local energy minimum) characterized by an energy $E_0$ and a plastic strain $\gamma_0$ .
Energy Landscape & Stability: Each mesostate has a stability range in strain. As strain increases, a mesoblock remains linearly elastic until it reaches its stability limit, at which point it undergoes a discontinuous transition to a new mesostate. The stability range is coupled to the energy: deeper (lower energy) states have larger stability ranges.
Finite Element Method (FEM): To account for non-local elastic interactions, the authors use FEM to solve the linear elasticity problem. This allows the model to treat plastic events as extended regions rather than point defects and facilitates the incorporation of heterogeneous elastic moduli.
Statistical Prescription: The density of states (DOS) is modeled as a Gaussian distribution. The degree of annealing is controlled by the "parent temperature" ( $T_p$ ), which determines the initial distribution of mesostate energies and plastic strains.
Plastic Increment Rules: The model tests different protocols for how a block chooses its new plastic strain after a transition (Uniform, Soft Glassy Rheology (SGR), and Maximal).

3. Key Contributions

Development of a High-Fidelity EPM: The integration of an energy-landscape-based single-site model with a full FEM framework provides a robust tool for studying complex deformation histories.
Sensitivity Analysis of Model Parameters: The paper identifies that the qualitative success of the model in reproducing known yielding diagrams is highly sensitive to the choice of the DOS mean ( $\mu_{DOS}$ ) and the plastic strain increment rule.
Identification of "Trenching": The authors uncover a phenomenon where shear bands remain "pinned" in certain parameter regimes, surrounded by highly stable states (trenches), a feature that mirrors observations in molecular dynamics (MD) simulations.
Scaling Laws for Failure: The model provides analytical insights into the power-law divergence of failure times and the growth of shear bands.

4. Results

Uniform Shear: The model successfully reproduces the brittle-to-ductile crossover (well-annealed samples show stress overshoots and shear bands, while poorly annealed samples show homogeneous flow) and the Bauschinger effect (mechanical memory where the material softens in the reverse loading direction).
Cyclic Shear Yielding:
- Annealing vs. Yielding: Poorly annealed samples undergo "mechanical annealing," evolving toward a universal threshold state. Well-annealed samples remain stable until a critical amplitude is reached, after which they follow a universal "yielded" curve.
- Yielding Diagram: The model reproduces the qualitative features of the yielding diagram, including the change in behavior across a threshold energy.
- Dynamic Phenomena: The model captures the non-monotonic evolution of local yield rates and energies. It also demonstrates a power-law divergence of the number of cycles needed to reach an absorbing state or to fail.
Fatigue Failure: The model shows that accumulated damage (the area of the stress-strain loop) follows a power-law scaling with the failure time ( $\tau_f$ ), consistent with recent atomistic studies.

5. Significance

This work bridges the gap between simple mean-field elastoplastic models and computationally expensive molecular dynamics simulations. By demonstrating that specific model choices (like the "Maximal" increment rule and a shifted Gaussian DOS) are required to capture the correct physics, the authors provide a roadmap for designing more accurate EPMs. The discovery of the "trenching" effect and the investigation of the intermediate coexistence regime around the yield point offer new avenues for understanding how localized plastic events organize into macroscopic failure in amorphous materials.