🔬 materials science

Quantifying translational and bond-orientational order metrics in hyperuniform and nonhyperuniform many-particle systems

This paper introduces a bond-orientational order metric $\tau_O$ to complement the translational order metric $\tau_T$ , demonstrating through analysis of hard-particle, random sequential addition, and stealthy hyperuniform systems that while bond-orientational order is generally subdominant to translational order, the two metrics are positively correlated and reveal distinct, preparation-dependent structural trajectories.

Original authors: Anirban Mukherjee, Salvatore Torquato

Published 2026-09-10

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Anirban Mukherjee, Salvatore Torquato

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

In the physical world, matter organizes itself in a spectrum that stretches from the total chaos of a gas to the rigid perfection of a crystal. Between these two extremes lies a vast landscape of materials where particles are neither completely random nor perfectly ordered. Understanding exactly where a material sits on this spectrum is a fundamental challenge for physicists and materials scientists. For decades, researchers have relied on different tools to measure these states. Some tools measure how well particles line up in straight rows, a property known as translational order. Others measure how particles align their neighbors in specific geometric patterns, such as forming hexagons, which is called bond-orientational order. The problem has been that these two types of order were measured using different languages, making it difficult to compare them directly or to see how they evolve together as a material changes.

A new study by Anirban Mukherjee and Salvatore Torquato at Princeton University introduces a unified way to measure both types of order simultaneously. The researchers developed two new metrics, or numerical scores, that treat translational alignment and geometric bonding on the same statistical footing. By applying these metrics to a wide variety of simulated particle systems, they mapped out how order develops in different materials. Their work reveals that while these two forms of order usually grow together, they do not always grow at the same speed, and the way a material is prepared can leave a distinct fingerprint on its internal structure.

The researchers tested their new metrics on three very different types of particle systems to see how the scores behaved. First, they looked at equilibrium systems, which are collections of particles that have settled into a stable state, like hard spheres floating in a fluid or arranged in a crystal. Second, they examined non-equilibrium systems created by a process called random sequential addition, where particles are dropped one by one into a container until no more can fit, a method that mimics how some materials jam without ever forming a crystal. Third, they studied a special class of "stealthy" systems that are designed to suppress density fluctuations in a way that is neither fully random nor fully crystalline.

In the equilibrium fluids, where particles move freely but interact, the researchers found that the score for geometric bonding remained much lower than the score for translational alignment. However, as the particles were packed more tightly together, the geometric score began to rise faster than the translational one, particularly in two-dimensional systems. This suggests that as a liquid approaches the point where it might freeze, the particles start organizing their neighbors into specific shapes before they lock into a rigid grid. In three-dimensional systems, this effect was much weaker, indicating that the path to order depends heavily on the dimensionality of the space.

When the researchers looked at the random sequential addition packings, a different picture emerged. Even as the particles became densely packed, the geometric bonding score stayed very low compared to the translational score. Crucially, the path traced by these jammed systems in the new measurement space was distinct from the path taken by the equilibrium fluids. This means that two materials could have the same amount of translational order but different amounts of geometric order, simply because they were prepared in different ways. The study shows that the history of how a material was made leaves a permanent mark on its structural signature.

In the special "stealthy" systems, which are designed to be hyperuniform—meaning they suppress large-scale density fluctuations in a unique way—the researchers observed a similar trend. As the systems were tuned to become more ordered, the geometric score increased relative to the translational score, but it remained subordinate until the system reached a critical threshold where it would transition into a fully ordered state. Throughout all the simulations, the two scores were positively correlated, meaning that as one increased, the other increased as well. However, the ratio between them varied significantly depending on the type of material and its preparation.

The findings suggest that these two metrics can serve as a powerful coordinate system for mapping the complex transitions of matter. By plotting a material's position on a graph defined by these two scores, scientists can distinguish between different phases of matter that might look similar under traditional analysis. For instance, the metrics can help identify when a liquid is beginning to develop the local geometric order that precedes crystallization, or when a jammed material is stuck in a disordered state. The researchers propose that this approach could be used to guide the design of new materials with specific structural properties, to better understand how glasses form, and to classify the complex states found in driven, non-equilibrium systems. The work does not claim to solve the mystery of order in all materials, but it provides a clearer, more consistent language for describing how particles arrange themselves in the space between chaos and perfection.

Technical Summary: Quantifying Translational and Bond-Orientational Order Metrics in Hyperuniform and Nonhyperuniform Many-Particle Systems

Problem Statement
Quantifying the degree of order or disorder in many-particle systems remains a fundamental challenge in physics, materials science, and mathematics. While long-range translational and bond-orientational order are defining characteristics of perfect crystals and quasicrystals, intermediate systems span a continuous spectrum from complete disorder to perfect order. Existing scalar order metrics typically address only one aspect: either positional (translational) order or spatially averaged local orientational order. Furthermore, conventional metrics often lack a unified statistical footing, making it difficult to compare the development of these two distinct forms of order across different preparation protocols, dimensionalities, and proximity to phase transitions.

Methodology
The authors propose a unified framework to quantify global translational and bond-orientational order using two-point correlation functions.

Translational Order Metric ( $\tau_T$ ): Defined as the squared $L^2$ norm of the total correlation function $h(r) = g_2(r) - 1$ . It is computed in both real space and Fourier space (via Parseval's theorem), capturing both short-range and long-wavelength correlations.
$\tau_T = \frac{1}{D^d} \int_{\mathbb{R}^d} h^2(r) \, dr$
where $D$ is a characteristic length scale. For a Poisson point process, $\tau_T = 0$ ; for perfect crystals, it diverges in the thermodynamic limit.
Bond-Orientational Order Metric ( $\tau_O$ ): Introduced as a novel analogue to $\tau_T$ , defined from the weighted total correlation function $h_f(r)$ . This function incorporates local orientational weights ( $f_j$ ) assigned to each particle based on its geometric neighbors (e.g., $\psi_6$ in 2D, $Q_{6m}$ in 3D).
$\tau_O = \frac{1}{D^d |f|_{ref}^4} \int_{\mathbb{R}^d} |h_f(r)|^2 \, dr$
The normalization by $|f|_{ref}^4$ (the weight magnitude in a perfect reference crystal) ensures invariance under uniform rescaling. Unlike $\tau_T$ , $\tau_O$ possesses a small, non-zero baseline for Poisson processes due to the geometric construction of neighbors.
Model Systems: The metrics were computed for three distinct classes of systems in 2D and 3D:
- Equilibrium Hard Particles: Hard disks (2D) and hard spheres (3D) along stable fluid branches (up to near-freezing) and stable crystal branches.
- Nonequilibrium RSA Packings: Random Sequential Addition packings from dilute regimes to just below saturation.
- Stealthy Hyperuniform (SHU) Ground States: Disordered 2D configurations with suppressed density fluctuations ( $S(k)=0$ for $0 < |k| \le K$ ), controlled by the stealthiness parameter $\chi$ .

Key Results

Positive Correlation: Across all sampled state points (excluding the Poisson limit), $\tau_T$ and $\tau_O$ are positively correlated; $\tau_O$ increases monotonically with $\tau_T$ as the control parameter (packing fraction $\phi$ or stealthiness $\chi$ ) increases.
Subdominance in Disordered Phases: In equilibrium fluids, RSA packings, and disordered SHU phases, bond-orientational order remains subdominant to translational order ( $\tau_O \ll \tau_T$ $τ_{O} ≪ τ_{T}$ ).
- Dimensionality Dependence: In 2D equilibrium fluids, the ratio $\tau_O/\tau_T$ increases substantially as the system approaches the freezing point, indicating a sharpening of orientational correlations prior to the liquid-hexatic transition. This effect is significantly weaker in 3D fluids.
- RSA vs. Equilibrium: At common packing fractions, equilibrium fluids and RSA packings trace distinct trajectories in the $(\tau_T, \tau_O)$ plane, revealing preparation-dependent structural differences. RSA systems show no analogous enhancement of orientational order near saturation.
Crystal Branches: Along the sampled crystal branches, $\tau_T$ and $\tau_O$ become comparable in magnitude ( $\tau_O/\tau_T \approx 0.5\text{--}1$ ). The ratio approaches unity as the system approaches the perfect crystal limit, consistent with the theoretical prediction that for perfect crystals, the metrics are proportional.
SHU Systems: Within the disordered SHU phase, $\tau_O$ remains subdominant, but the ratio $\tau_O/\tau_T$ increases as $\chi$ approaches the disorder-to-order threshold ( $\chi = 1/2$ ).

Significance and Claims
The primary contribution of this work is placing translational and bond-orientational order on a "common two-point statistical footing." By defining both metrics as functionals of their respective two-point correlation functions, the pair $(\tau_T, \tau_O)$ provides a collective structural coordinate system.

The authors claim that this framework:

Distinguishes Preparation Protocols: It reveals structural differences between equilibrium and nonequilibrium systems that may appear similar under conventional pair statistics.
Sensitivity to Ordering Thresholds: It detects the strengthening of orientational correlations near phase transitions (e.g., the 2D liquid-hexatic transition and the SHU disorder-to-order threshold) where translational order metrics alone might not capture the full picture.
Potential Applications: The authors suggest these metrics may serve as collective variables for mapping structural pathways in freezing, melting, glass formation, jamming, and self-assembly. They also propose their utility in enhanced sampling, phase classification, and inverse design, provided future work establishes the relationship between specific regions of the $(\tau_T, \tau_O)$ plane and macroscopic properties.

The paper maintains a modest stance, noting that the observed hierarchy $\tau_O < \tau_T$ is specific to the sampled systems and not a universal constraint (e.g., in hexatic phases, $\tau_O$ could exceed $\tau_T$ ). The metrics are presented as tools to quantify integrated correlation magnitudes and spatial persistence, rather than as unique identifiers of specific underlying structures.

Technical Summary: Quantifying Translational and Bond-Orientational Order Metrics in Hyperuniform and Nonhyperuniform Many-Particle Systems

More like this