Abstract In order to design tougher materials, it is crucial to understand the relationship between their composition and their resistance to fracture. To this end, we investigate the fracture toughness of usual sodium silicate glasses (NS) and complex calcium--silicate--hydrates (CSH), the binding phase of cement. Their atomistic structure is described in the framework of the topological constraints theory, or rigidity theory. We report an analogous rigidity transition, driven by pressure in NS and by composition in CSH. Relying both on simulated and available experimental results, we show that optimally constrained isostatic systems show improved fracture toughness. The flexible to stressed--rigid transition is shown to be correlated to a ductile-to-brittle transition, with a local minimum of the brittleness for isostatic system. This fracture toughening arises from a reversible molecular network, allowing optimal stress relaxation and crack blunting behaviors. This opens the way to the discovery of high-performance materials, designed at the molecular scale.
Introduction
In materials science, a persistent challenge is the inherent trade-off between hardness and toughness. While increasing atomic connectivity typically enhances stiffness, the material's resistance to fracture—its toughness—often follows a more complex, non-monotonic path. For instance, adding more cross-links to a molecular network might increase its hardness but simultaneously make it more brittle, leading to catastrophic failure. This study explores the molecular origins of these properties by applying Rigidity Theory to complex, heterogeneous materials: Sodium Silicate (NS) glasses and Calcium-Silicate-Hydrate (CSH), the primary binding phase in cement.
Methodology: Rigidity Theory and Molecular Dynamics
The researchers utilized Molecular Dynamics (MD) simulations to model the atomistic structures of NS and CSH. Central to their analysis is the concept of Topological Constraint Counting, which determines the number of constraints per atom ($n_c$). According to rigidity theory:
- Flexible networks ($n_c < 3$) have internal degrees of freedom.
- Stressed-rigid networks ($n_c > 3$) are over-constrained and "locked."
- Isostatic networks ($n_c \approx 3$) are optimally constrained, representing a transition state between flexibility and rigidity.
To evaluate fracture properties, the team simulated "Mode I" fracture by introducing a sharp crack into the models and applying incremental tensile strain. They calculated fracture toughness ($K_{Ic}$) and fracture energy ($G_c$), while also performing loading/unloading cycles to measure "unrecovered volume"—a proxy for structural irreversibility.
The Isostatic Window: Optimizing Fracture Toughness
The study reveals that fracture toughness is not a linear function of atomic connectivity. Instead, it reaches a maximum within the "isostatic window" (where $n_c$ is between 3.0 and 3.2). This behavior was observed in both NS glasses (where $n_c$ was tuned by pressure) and CSH (where $n_c$ was tuned by the Ca/Si ratio).
Quantitatively, the calculated $K_{Ic}$ for NS glass at ambient pressure was 0.61 MPa m$^{1/2}$, aligning well with experimental values (~0.7). Similarly, for CSH at a Ca/Si ratio of 1.71, the $K_{Ic}$ of 0.37 MPa m$^{1/2}$ falls within the expected range for cement paste. These results suggest that the isostatic state represents a "Goldilocks" zone for material performance.
Mechanisms of Toughening: Reversibility and Crack Blunting
The peak in toughness is intrinsically linked to the "stress-free" character of isostatic networks. Unlike flexible networks that deform plastically or stressed-rigid networks that are brittle and prone to sudden bond breaking, isostatic networks exhibit a unique capacity for reversible molecular rearrangement.
As shown in the analysis of unrecovered volume, isostatic systems show near-zero structural hysteresis. This reversibility allows for crack blunting: the tip of a propagating crack rounds out rather than remaining sharp. A blunted crack tip distributes stress over a larger volume, significantly increasing the energy required for the fracture to advance. This transition from flexible to stressed-rigid also marks a ductile-to-brittle transition, with the isostatic state providing a local minimum in brittleness.
Conclusion and Future Directions
This research demonstrates that rigidity theory provides a universal framework for understanding and designing the fracture properties of chemically diverse materials like glass and cement. By tuning the topological constraints to the isostatic point, it is possible to achieve superior fracture toughness through reversible stress relaxation.
While these findings provide a molecular-scale blueprint for high-performance materials, challenges remain in scaling these insights. Real-world materials like concrete possess macro-scale pores and aggregates that influence fracture behavior beyond the molecular "glue." Future work will need to address how these optimal networks age over time and how they interact with larger-scale structural features to ensure long-term durability in infrastructure.
Summary updated