Predicting random close packing of binary hard-disk mixtures via third-virial-based parameters

This paper proposes a simple and accurate method to predict the random close packing fraction of binary hard-disk mixtures by introducing a parameter based on the reduced third virial coefficient, which captures three-body correlations and yields a near-linear relationship that outperforms existing models across various size ratios and compositions.

The Gist

Imagine you are trying to pack a suitcase for a trip. If you only have identical socks, you can stack them neatly, but there will always be some empty air between them. Now, imagine you have a mix of socks, shoes, and bulky sweaters. If you just throw them in randomly, you might get a decent amount of stuff in, but you could probably fit more if you knew the secret trick of how to arrange different sizes together.

In the world of physics, scientists study this exact problem, but instead of suitcases, they look at hard disks (like flat coins) and hard spheres (like marbles). They want to know: What is the absolute maximum amount of "stuff" you can cram into a space if you just shake it until it's as tight as possible, without arranging it in a perfect crystal pattern? This limit is called Random Close Packing (RCP).

For a long time, scientists have tried to predict this limit for mixtures of different sizes. Here is a simple breakdown of what this new paper does:

1. The Old Way: A Rough Guess

Previously, a scientist named Brouwers proposed a formula to guess how much space a mixture of big and small disks would take up. Think of this like a recipe that says, "If you mix big and small cookies, here is how much space they will fill."

The problem? The recipe worked okay for cookies that were slightly different sizes, but if you mixed tiny crumbs with giant cookies, the recipe started to fail. It couldn't perfectly predict the "tipping point" where the mixture gets jammed.

2. The New Idea: The "Three-Way Handshake"

The authors of this paper, Andrés Santos and Mariano López de Haro, realized that the old recipe was missing a crucial ingredient: how three objects interact at once.

Imagine a crowded dance floor:

• Two-body interaction: Two people bumping into each other is easy to understand.
• Three-body interaction: Imagine three people trying to dance in a small circle. If Person A is big, Person B is small, and Person C is medium, the way they squeeze together creates a unique "hole" or "gap" that two people alone couldn't create.

The authors introduced a new parameter based on the Third Virial Coefficient. In plain English, this is a mathematical way of measuring those complex three-way interactions. They call this new parameter $\mu$ (mu).

3. The "Magic Line"

Here is the cool discovery: When they plotted their data using this new "three-way handshake" parameter, something magical happened.

• The Old Way: If you plotted the packing density for different mixtures, the data points were scattered all over the place, like a messy cloud.
• The New Way: When they used their new parameter, all those scattered points collapsed into a single, straight line.

It's as if they found a universal "translation key." No matter if you are mixing 10% small disks with 90% big ones, or 50/50, or if the size difference is huge or tiny, they all fall onto the same simple line. This means the relationship is nearly linear and universal.

4. Why This Matters

This isn't just about coins on a table. This concept applies to:

• Concrete and Ceramics: Mixing sand, gravel, and cement to make the strongest, most compact building material.
• Pharmaceuticals: Packing pills of different sizes into a bottle efficiently.
• Biology: Understanding how cells or chloroplasts pack inside a plant cell.
• Traffic and Granular Materials: How cars or sand grains jam together.

The Takeaway

The authors didn't just tweak an old formula; they found a deeper geometric truth. By focusing on how three particles constrain each other (the "three-way handshake"), they created a simple, accurate tool that works for almost any mixture of sizes.

They even showed that this tool can be extended to mixtures with infinite variations in size (like a smooth gradient from tiny dust to large rocks), not just simple two-size mixtures.

In short: They found a new "ruler" that measures the complexity of a messy pile of objects, allowing us to predict exactly how tightly we can pack them, no matter how different the sizes are. It turns a chaotic puzzle into a simple, straight line.

Technical Summary

1. Problem Statement

The paper addresses the challenge of predicting the Random Close Packing (RCP) fraction ($\phi_{\text{mixt}}$) for binary mixtures of hard disks (HD). While RCP is a fundamental concept in condensed matter physics, granular media, and colloids, its analytical determination in polydisperse systems remains an open problem.

• Context: Previous attempts, such as Brouwers' geometric approach (2023/2025) and Zaccone's equilibrium-based model (2025), have provided approximations. However, these models often fail to collapse simulation data across a wide range of size ratios ($q = \sigma_L/\sigma_S$) and compositions, particularly as the size disparity increases.
• Gap: Existing models rely on geometric parameters or compressibility factors that do not fully capture the complex three-body correlations and excluded-area constraints inherent in jammed disordered states.

2. Methodology

The authors propose a new, simple, and accurate approach based on the reduced third virial coefficient ($\bar{B}_3$) of the mixture.

• Core Hypothesis: The RCP fraction depends nearly linearly on a parameter $\mu$ derived from the mixture's third virial coefficient. This coefficient effectively captures three-body correlations and local triplet constraints, which are dominant in determining packing efficiency at jamming.

• The Proposed Model:
  The authors replace previous formulations with the following linear relationships:

  1. Linear dependence on $\mu$:
     $$\phi_{\text{mixt}} = \phi_{\text{mono}} + \mu(1 - \phi_{\text{mono}})$$
  2. Linear dependence on $\lambda$:
     $$\frac{\phi_{\text{mixt}}}{1 - \phi_{\text{mixt}}} = \frac{\lambda}{1 - \phi_{\text{mono}}} - 1$$
     where $\lambda = 1/(1-\mu)$.

  Here, $\phi_{\text{mono}}$ is the RCP fraction of the monodisperse system (taken as $0.844$). The parameter $\mu$ is defined as:
  $$\mu \equiv \frac{b_3 - 1 - (\bar{B}_3 - 1)m_2}{b_3 - 3}$$

  • $b_3 \approx 3.128$ is the reduced third virial coefficient for a monodisperse 2D hard-disk system.
  • $m_2$ is the second reduced moment of the size distribution.
  • $\bar{B}_3$ is the reduced third virial coefficient for the binary mixture, calculated using an exact integral involving the intersection area of three disks of varying sizes.

• Theoretical Basis: The formulation is structurally consistent with the "surplus" equation-of-state framework for fluid mixtures, linking the jamming state to low-order virial information.

3. Key Contributions

1. Novel Parameterization: Introduction of a universality parameter ($\mu$) based explicitly on the third virial coefficient, moving beyond purely geometric or second-order moment approximations.
2. Data Collapse: Demonstration that plotting simulation data against $\mu$ (or $\lambda$) results in a near-perfect linear collapse for binary mixtures with size ratios $q = 1.4, 1.7, 2,$ and $3$.
3. Extension to Polydispersity: The framework is naturally extended to continuous size distributions (e.g., log-normal distributions). The authors derive the expression for $\bar{B}_3$ in polydisperse systems and show that $\mu$ varies smoothly with the dispersity parameter $s$, remaining physically bounded ($0 \le \mu \le 1$).
4. Comparative Superiority: The paper provides a rigorous comparison showing that the third-virial-based approach outperforms both Brouwers' geometric model and Zaccone's equilibrium model, particularly for larger size ratios ($q \ge 2$).

4. Results

• Binary Mixtures:
  • Brouwers' Model: Accurate for small size ratios ($q=1.4$) but deviates significantly as $q$ increases.
  • Zaccone's Model: Performs reasonably well for $q=1.4$ but underestimates $\phi_{\text{mixt}}$ for larger $q$.
  • Proposed Model: Shows excellent agreement with Monte Carlo simulation data across all tested size ratios. The linear relationship between $\phi_{\text{mixt}}$ and $\mu$ holds robustly, confirming the "near-universality" of RCP in these systems.
• Polydisperse Mixtures:
  • For log-normal distributions, the parameter $\mu$ transitions smoothly from 0 (monodisperse limit, $s \to 0$) to 1 (infinite dispersity, $s \to \infty$).
  • The model predicts that $1-\mu$ decays approximately exponentially with $s^2$, ensuring the predicted packing fraction never exceeds the physical limit of 1.
• Physical Interpretation: The results suggest that three-body correlations (encoded in $\bar{B}_3$) are the dominant geometric factor governing the efficiency of disk arrangement in jammed states, more so than simple size disparity metrics.

5. Significance

• Predictive Power: The method offers a compact, analytical framework to predict RCP fractions without requiring computationally expensive simulations for every new mixture composition.
• Universality: It reinforces the idea that low-order virial coefficients can capture the essential physics of dense disordered packings, bridging the gap between equilibrium fluid theory and non-equilibrium jamming states.
• Future Directions: The authors suggest this framework could be extended to non-spherical particles (e.g., ellipses, spherocylinders) where the third virial coefficient would incorporate orientational excluded-volume effects. This opens a path for understanding RCP in more complex soft matter and granular systems.

In conclusion, the paper establishes a robust, third-virial-based parameter that significantly improves the accuracy of RCP predictions for binary and polydisperse hard-disk mixtures, providing a unified theoretical tool for understanding packing in disordered systems.