🔬 materials science

How glass breaks -- Damage explains the difference between surface and fracture energies in amorphous silica

This study utilizes multi-scale simulations combining molecular dynamics and phase-field modeling to demonstrate that the discrepancy between surface and fracture energies in amorphous silica arises primarily from damage diffusion over a 16–23 Å range around the crack path, rather than plastic deformation.

Original authors: Gergely Molnár, Etienne Barthel

Published 2026-02-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Gergely Molnár, Etienne Barthel

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 Big Mystery: Why Does Glass Break So "Expensively"?

Imagine you have a perfect sheet of glass. To break it, you have to snap the tiny atomic bonds holding it together. Physics tells us that the energy required to snap these bonds (creating two new surfaces) should be exactly equal to the energy you put in to break the glass.

However, for decades, scientists have been puzzled by a mystery: When they actually break glass in the lab, it takes five times more energy than the math says it should.

It's like trying to tear a piece of paper. You expect it to take a certain amount of force to rip the fibers. But in reality, it feels like you have to pull five times harder. Where is that extra energy going?

For a long time, scientists guessed that the glass was "squishing" or bending (plasticity) right at the tip of the crack, like soft clay, and that this squishing was soaking up the extra energy. But glass is supposed to be brittle and hard, not squishy. So, this explanation didn't quite fit.

The New Discovery: It's Not "Squishing," It's "Unraveling"

In this paper, the researchers used powerful computer simulations to watch exactly what happens to the atoms inside silica glass (the main ingredient in window glass) as a crack forms. They found that the old idea of "squishing" was wrong.

Instead, they discovered that the extra energy is used to unravel the microscopic structure of the glass around the crack.

Here is the breakdown of what they found:

1. The Two Types of Energy

The researchers separated the total energy of breaking the glass into two distinct buckets:

Bucket A: The Surface Energy (The "Snap")
This is the energy needed to actually sever the bonds and create the two new, empty surfaces of the crack.
- Analogy: Think of this like snapping a single dry twig. It takes a specific, small amount of force to break the wood fibers. This happens in a very thin layer, just a few atoms wide.
- What they found: This energy accounts for only about 20% of the total energy used to break the glass.
Bucket B: The Damage Energy (The "Unraveling")
This is the energy used to mess up the structure of the glass around the crack, but not necessarily break the surface yet.
- Analogy: Imagine a woven basket. If you pull a thread to break the basket, you don't just snap the thread. The whole weave around the tear gets distorted, stretched, and loosened. The "damage" spreads out in a fuzzy cloud around the tear.
- What they found: This "fuzzy cloud" of damage extends about 20 Angstroms (roughly 20 times the width of a single atom) away from the crack. This accounts for the remaining 80% of the energy.

2. What is actually happening inside the glass?

The researchers looked closely at the atomic structure to see what was changing in that "fuzzy cloud."

The Surface (The Snap): At the very edge of the crack, the silicon atoms lose their neighbors. Their "coordination number" (how many friends they are holding hands with) drops. This is the actual breaking of the surface.
The Damage Zone (The Unraveling): A bit further away from the crack, the atoms are still holding hands, but the shape of the rings they form is changing. In glass, atoms form little loops or rings. As the crack approaches, these rings get distorted, stretched, or broken, even though the surface hasn't opened yet.

The Key Insight: The extra energy isn't being lost to the glass getting "soft" or "plastic" (like bending a metal paperclip). Instead, the energy is being spent on rearranging the internal architecture of the glass. The glass is essentially "unraveling" its own structure before it finally snaps.

How They Proved It

The scientists didn't just guess; they built a digital model of glass and used a technique called "Phase-Field modeling."

The Analogy: Imagine trying to measure the size of a cloud. You can't see the exact edge, so you use a camera with a slightly blurry lens (coarse-graining) to see the general shape. They used this method to measure the "damage cloud" around the crack.
The Result: They calculated the energy of the "damage cloud" and the "surface snap" separately. When they added them up, the total matched the experimental data perfectly. They confirmed that the "damage" (the unraveled rings) is the main reason glass requires so much extra energy to break.

The Bottom Line

Glass is brittle, but it's not as simple as just snapping a stick.

When glass breaks, it creates a zone of structural chaos around the crack tip. This zone is about 20 atoms wide. The energy required to create this chaotic, distorted zone is four times larger than the energy required to actually make the crack surface.

This explains the long-standing mystery: The "extra" energy isn't wasted on plasticity (squishing); it's spent on unraveling the microscopic web that holds the glass together. This discovery helps us understand that even in the hardest, most brittle materials, there is a complex, diffuse process happening right before the break.

Technical Summary: How Glass Breaks – Damage Explains the Difference Between Surface and Fracture Energies in Amorphous Silica

Problem Statement
In fracture mechanics, a fundamental discrepancy exists between the free surface energy ( $\gamma$ ), which quantifies the energy required to create new surfaces, and the fracture energy ( $G_c$ ), which represents the total energy dissipated during crack propagation. In ideal brittle materials, $G_c$ should equal $2\gamma$ . However, experimental measurements on silicate glasses consistently show that $G_c$ exceeds $2\gamma$ by a factor of approximately 5. While this excess energy is often attributed to plastic dissipation at the crack tip, this explanation remains controversial for amorphous silica. Evidence for plasticity is difficult to obtain experimentally at the atomic scale, and standard plasticity models fail to apply to silicates due to their lack of hardening. Consequently, the physical origin of this energy discrepancy in brittle amorphous silicates has remained unresolved.

Methodology
The authors employed a multiscale simulation approach combining atomic-scale molecular dynamics (MD) with continuum phase-field modeling to investigate fracture in amorphous silica.

Atomic Simulations: MD simulations were conducted using the BKS potential to model atomic interactions and Wolf's truncation for Coulombic forces. Samples were generated by random atom placement, equilibrated, and quenched to 10 $^{-5}$ K. Fracture propagation was simulated under quasi-static, athermal conditions using a K-field displacement boundary condition to drive crack growth.
Coarse-Graining: To bridge the atomic and continuum scales, a physically based Gaussian convolution technique was applied. This method generated homogeneous fields for displacement, strain, stress, and potential energy while conserving mass and energy.
Damage and Energy Quantification:
- Damage: Defined mechanically as the loss of stiffness, calculated by comparing actual elastic energy to undamaged strain energy.
- Free Surface Energy (FSE): Isolated by calculating the difference between the total potential energy and the elastic strain energy ( $\psi_{FSE} = \psi_{pot} - \psi_{\epsilon}$ ), effectively separating surface formation energy from elastic deformation.
- Phase-Field Calibration: A Finite Element Update (FEMU) scheme was used to fit the damage field obtained from MD simulations to a continuum phase-field model. This allowed the extraction of material parameters, specifically the fracture toughness ( $g_c$ ) and the internal length scale ( $l_c$ ), directly from atomic snapshots.
Structural Analysis: Local structural changes were quantified by tracking the evolution of ring structures (using the Le Roux & Jund algorithm) and the coordination number of silicon atoms.

Key Results

Diffuse Damage vs. Plasticity: The study demonstrates that the energy dissipation around the crack tip is primarily due to diffuse damage rather than plasticity. The damage zone extends approximately 16–23 Å from the crack path. Analysis of the stress-strain response and energy balance confirms that plastic dissipation and nonlinear elasticity are negligible; the material loses stiffness (damage) rather than retaining it (plasticity).
Separation of Energy Contributions: The authors successfully separated the total fracture energy into two distinct components:
- Free Surface Energy: Localized strictly at the surface, associated with bond breaking and atomic relaxation.
- Damage Energy: Extending deeper into the material (~20 Å), associated with structural rearrangements.
  The ratio of free surface energy to total fracture energy ( $2\gamma/g_c$ ) was calculated to be approximately 23%, consistent with experimental observations where surface energy accounts for only ~20% of the total energy.
Structural Signatures:
- Diffuse Damage: Correlates strongly with changes in the ring structure of the glass network, which has a broader spatial extent.
- Free Surface Energy: Correlates with changes in the coordination number of silicon atoms, which are more localized near the surface.
Validation: The fracture toughness values derived from the phase-field fit, the global energy balance, and the J-integral method were consistent. The calculated internal length scale ( $l_c \approx 12\text{--}20$ Å) aligns with the observed width of the damage zone.

Significance and Claims
The paper claims to provide a fully consistent framework for understanding fracture in amorphous silica by resolving the long-standing discrepancy between surface and fracture energies. The primary contribution is the identification of diffuse damage—distinct from plasticity—as the dominant mechanism for energy dissipation in brittle silicate glasses. By linking this damage to specific structural rearrangements (ring structure breaking) and quantifying its spatial extent, the study offers a structural explanation for the "brittle" yet energy-dissipative nature of silica fracture.

The authors assert that their approach validates the concept of diffuse damage used in phase-field fracture techniques, providing a direct demonstration of its physical basis at the atomic scale. Furthermore, the results suggest that traditional views of brittle fracture may overlook the subtle mechanical dissipation inherent in amorphous materials. The paper concludes modestly, noting that while these findings clarify the mechanism in silicates, further investigation is required to determine if these concepts apply to other amorphous systems, such as metallic glasses, or to dynamic and fatigue loading scenarios.