Fluctuation induced network patterns in active matter with spatially correlated noise

Sebastian Fehlinger, Kai Cui, Arooj Sajjad, Heinz Koeppl, Benno Liebchen

Abstract

Fluctuations play a central role in many fields of physics, from quantum electrodynamics to statistical mechanics. In active matter physics, most models focus on thermal fluctuations due to a surrounding solvent. An alternative but much less explored noise source can occur due to fluctuating external fields, which typically feature certain spatial correlations. In this work, we introduce a minimal model to explore the influence of spatially correlated but temporally uncorrelated noise on the collective behavior of active particles. We find that specifically in chiral active particles such fluctuations induce the formation of network patterns, which neither occur for spatially (uncorrelated) thermal noise, nor in the complete absence of fluctuations. These networks show (i) a percolated structure, (ii) local alignment of the contained particles, but no global alignment, and (iii) hardly coarsen. We perform a topological data analysis to systematically characterize the topology of the network patterns. Our work serves as a starting point to explore the role of spatially correlated fluctuations and presents a route towards noise-induced phenomena in active matter.

Introduction

Fluctuations are ubiquitous in physical systems, often viewed as disruptive forces that hinder order. In the study of active matter—systems composed of self-propelled units like bacteria or synthetic colloids—noise is typically modeled as "white," where random perturbations are spatially and temporally independent. However, real-world environments often involve fluctuating external fields with inherent spatial correlations. This research investigates how such spatially correlated noise influences the collective behavior of chiral active particles, revealing that noise can act as a structural architect rather than a mere source of disorder.

Methods: A Minimal Model of Chiral Active Matter

The study employs a modified Vicsek model to simulate $N=2500$ self-propelled particles. Two key features differentiate this model from standard frameworks:

  1. Chirality ($\omega$): Particles possess an intrinsic tendency to rotate, a common feature in biological and synthetic active systems.
  2. Spatially Correlated Noise: Particle orientation is influenced by a random field $\phi_\alpha(\vec{r}, t)$ with a correlation length controlled by the parameter $\alpha$.

To rigorously characterize the resulting patterns, the authors utilized Topological Data Analysis (TDA). By applying persistent homology, they quantified the connectivity ($H_0$) and the presence of loops or "holes" ($H_1$) within the particle distributions. This approach provides a robust metric for "network-ness" that goes beyond simple density fluctuations, allowing for a precise definition of the noise-induced state.

Results: Noise-Induced Network Formation

The primary discovery of this work is the emergence of percolated network patterns under specific conditions: high chirality ($\omega^* \geq 3.0$) and significant spatial noise correlation ($\alpha \geq 3.0$). In the absence of noise or in the presence of standard uncorrelated noise, these particles typically form small, isolated clusters or remain in a disordered gas-like state.

Evolution of network patterns under correlated noise vs. no noise.

The mechanism behind this formation is identified as "dimer stability." In a correlated noise field, adjacent particles receive similar random "kicks," allowing them to remain aligned for significantly longer periods than they would under independent fluctuations. This stability facilitates the growth of elongated, string-like structures. As these chains grow, a geometric branching transition occurs—where the likelihood of side-attachment matches end-elongation—leading to the formation of a complex, percolated network that spans the entire system.

Quantitative and Topological Characterization

Quantitative metrics confirm the distinct nature of these noise-induced states. The maximum cluster mass ($M^{max}$) and radius ($R^{max}$) grow significantly over time only when noise is spatially correlated.

Quantitative metrics of cluster growth and hole formation.

Topological analysis reveals that these networks are characterized by a high number of persistent holes. The largest voids in the noise-induced network are approximately 2.5 times larger than those found in deterministic simulations. Crucially, these patterns are robust, persisting even when additional thermal noise or positional noise is introduced to the system. Unlike other active matter structures that may coarsen over time, these networks reach a stable, non-equilibrium steady state.

Conclusion and Outlook

This work demonstrates a novel class of fluctuation-driven phenomena where spatially correlated noise induces large-scale structural order in active matter. These findings suggest that environmental fluctuations in real-world systems—such as fluctuating light or magnetic fields—could play a functional role in organizing active components. Future research may explore the impact of temporal correlations and hydrodynamic interactions, providing a path toward experimental verification in systems like chiral colloids or magnetic rollers.

Summary updated

Abstract

Fluctuations play a central role in many fields of physics, from quantum electrodynamics to statistical mechanics. In active matter physics, most models focus on thermal fluctuations due to a surrounding solvent. An alternative but much less explored noise source can occur due to fluctuating external fields, which typically feature certain spatial correlations. In this work, we introduce a minimal model to explore the influence of spatially correlated but temporally uncorrelated noise on the collective behavior of active particles. We find that specifically in chiral active particles such fluctuations induce the formation of network patterns, which neither occur for spatially (uncorrelated) thermal noise, nor in the complete absence of fluctuations. These networks show (i) a percolated structure, (ii) local alignment of the contained particles, but no global alignment, and (iii) hardly coarsen. We perform a topological data analysis to systematically characterize the topology of the network patterns. Our work serves as a starting point to explore the role of spatially correlated fluctuations and presents a route towards noise-induced phenomena in active matter.
BibTeX
@inproceedings{fehlinger2025fluctuation,
  author = {Sebastian Fehlinger and Kai Cui and Arooj Sajjad and Heinz Koeppl and Benno Liebchen},
  title = {{Fluctuation induced network patterns in active matter with spatially correlated noise}},
  booktitle = {Soft},
  year = {2025},
  doi = {10.1039/d5sm00245a},
}
Figure