Self-propelled particles


Self-propelled particles, also referred to as self-driven particles, are terms used by physicists to describe autonomous agents, which convert energy from the environment into directed or persistent random walk. Natural systems which have inspired the study and design of these particles include walking, swimming or flying animals. Other biological systems include bacteria, cells, algae and other micro-organisms. Generally, self-propelled particles often refer to artificial systems such as robots or specifically designed particles such as swimming Janus colloids, bimetallic nanorods, nanomotors and walking grains. In the case of directed propulsion, which is driven by a chemical gradient, this is referred to as chemotaxis, observed in biological systems, e.g. bacteria quorum sensing and ant pheromone detection, and in synthetic systems, e.g. enzyme molecule chemotaxis and enzyme powered hard and soft particles.

Overview

Self-propelled particles interact with each other, which can lead to the emergence of collective behaviours. These collective behaviours mimic the self-organization observed with the flocking of birds, the swarming of bugs, the formation of sheep herds, etc.
To understand the ubiquity of such phenomena, physicists have developed a number of self-propelled particles models. These models predict that self-propelled particles share certain properties at the group level, regardless of the type of animals in the swarm. It has become a challenge in theoretical physics to find minimal statistical models that capture these behaviours.

Examples

Biological systems

Most animals can be seen as SPP: they find energy in their food and exhibit various locomotion strategies, from flying to crawling. The most prominent examples of collective behaviours in these systems are fish schools, birds flocks, sheep herds, human crowds. At a smaller scale, cells and bacteria can also be treated as SPP. These biological systems can propel themselves based on the presence of chemoattractants. At even smaller scale, molecular motors transform ATP energy into directional motion. Recent work has shown that enzyme molecules will also propel themselves. Further, it has been shown that they will preferentially move towards a region of higher substrate concentration, a phenomenon that has been developed into a purification technique to isolate live enzymes. Additionally, microparticles, vesicles, and even macroscale sheets can become self-propelled when they are functionalized with enzymes. The catalytic reactions of the enzymes direct the particles or vesicles based on corresponding substrate gradients.

Artificial systems

There is a distinction between wet and dry systems. In the first case the particles "swim" in a surrounding fluid; in the second case the particles "walk" on a substrate.
Active colloidal particles, dubbed nanomotors, are the prototypical example of wet SPP. Janus particles are colloidal particles with two different sides, having different physical or chemical properties. This symmetry breaking allows, by properly tuning the environment, for the motion of the Janus particle. For instance, the two sides of the Janus particle can induce a local gradient of, temperature, electric field, or concentration of chemical species. This induces motion of the Janus particle along the gradient through, respectively, thermophoresis, electrophoresis or diffusiophoresis. Because the Janus particles consume energy from their environment, the resulting motion constitutes an irreversible process and the particles are out of equilibrium.
  • The first example of an artificial SPP on the nano or micron scale was a gold-platinum bimetallic nanorod developed by Sen and Mallouk. In a solution of hydrogen peroxide, this "nanomotor" would exhibit a catalytic oxidation-reduction reaction, thereby inducing a fluid flow along the surface through self-diffusiophoresis. A similar system used a copper-platinum rod in a bromine solution. A recent study demonstrated the control of the positions and orientations of these active nanorods under confined microfluidic nozzles using ultrasound.
  • Another Janus SPP was developed by coating half of a polystyrene bead with platinum. These were used to direct the motion of catalytic motors when they were close to a solid surface. These systems were able to move the active colloids using geometric constraints. It was also found that metal–organic framework -based Janus micromotors can function as light-powered active colloids, capable of autonomous propulsion and enabling real-time surface-enhanced Raman sensing in chemically complex environments.
  • Another example of a Janus SPP is an organometallic motor using a gold-silica microsphere. Grubb's catalyst was tethered to the silica half of the particle and in solution of monomer would drive a catalytic polymerization. The resulting concentration gradient across the surface would propel the motor in solution.
  • Another example of an artificial SPP are platinum spinner microparticles that have controllable rotations based on their shape and symmetry. By utilizing multidirectional magnetic fields, the trajectories of these particles can also be directed into specific patterns.
  • Another example is biphasic Janus oil droplets which shows self propelled motion.
  • Several other examples are described in the nanomotor-specific page.
Walking grains are a typical realization of dry SPP: The grains are milli-metric disks sitting on a vertically vibrating plate, which serves as the source of energy and momentum. The disks have two different contacts with the plate, a hard needle-like foot in the front and a large soft rubber foot in the back. When shaken, the disks move in a preferential direction defined by the polar symmetry of the contacts. This together with the vibrational noise result in a persistent random walk.

Symmetry breaking

Symmetry breaking is a necessary condition for SPPs, as there must be a preferential direction for moving. However, the symmetry breaking may not come solely from the structure itself but from its interaction with electromagnetic fields, in particular when taken into account retardation effects. This can be used for the phototactic motion of even highly symmetrical nanoparticles. In 2021, it was experimentally shown that completely symmetric particles experience a net thermophoretic force when illuminated from a given direction. For self-propelled enzyme molecules, symmetry breaking can also arise from diffusion and kinetic asymmetry.

Swirlons

In 2020, researchers from the University of Leicester reported a hitherto unrecognised state of self-propelled particles — which they called a "swirlonic state". The swirlonic state consists of "swirlons", formed by groups of self-propelled particles orbiting a common centre of mass. These quasi-particles demonstrate a surprising behaviour: In response to an external load they move with a constant velocity proportional to the applied force, just as objects in viscous media. Swirlons attract each other and coalesce forming a larger, joint swirlon. The coalescence is an extremely slow, decelerating process, resulting in a rarified state of immobile quasi-particles. In addition to the swirlonic state, gaseous, liquid and solid states were observed, depending on the inter-particle and self-driving forces. In contrast to molecular systems, liquid and gaseous states of self-propelled particles do not coexist.

Typical collective behaviour

Both biological swarms and bioinspired artificial self-propelled particle systems can exhibit collective behaviour. Typical collective motion generally includes the formation of self-assembled structures, such as clusters and organized assemblies.
The prominent and most spectacular emergent large scale behaviour observed in assemblies of SPP is directed collective motion. In that case all particles move in the same direction. On top of that, spatial structures can emerge such as bands, vortices, asters, moving clusters.
For self-propelled particles, one example is that peroxide-TiO2 micromotors exhibited programmable swarming behaviors under blue light, including reversible expansion–contraction and directional collective movement. Chemically powered silver phosphate colloidal micromotors also display collective oscillations and synchronized transitions between clustered and dispersed states.
Another class of large scale behaviour, which does not imply directed motion is either the spontaneous formation of clusters or the separation in a gas-like and a liquid-like phase, an unexpected phenomenon when the SPP have purely repulsive interaction. This phase separation has been called Motility Induced Phase Separation.

Examples of modelling

The modeling of SPP was introduced in 1995 by Tamás Vicsek et al. as a special case of the Boids model introduced in 1986 by Reynolds. In that case the SPP are point particles, which move with a constant speed. and adopt the average direction of motion of the other particles in their local neighborhood up to some added noise.
Simulations demonstrate that a suitable "nearest neighbour rule" eventually results in all the particles swarming together or moving in the same direction. This emerges, even though there is no centralised coordination, and even though the neighbours for each particle constantly change over time.
Since then a number of models have been proposed, ranging from the simple active Brownian particle to detailed and specialized models aiming at describing specific systems and situations. Among the important ingredients in these models, one can list
  • Self-propulsion: in the absence of interaction, the SPP speed converges to a prescribed constant value
  • Body interactions: the particles can be considered as points like in the Vicsek model. Alternatively, one can include an interaction potential, either attractive or repulsive. This potential can be isotropic or not to describe spherical or elongated particles.
  • Body orientation: for those particles with a body-fixed axis, one can include additional degrees of freedom to describe the orientation of the body. The coupling of this body axis with the velocity is an additional option.
  • Aligning interaction rules: in the spirit of the Vicsek model, neighboring particles align their velocities. Another possibility is that they align their orientations.
One can also include effective influences of the surrounding; for instance the nominal velocity of the SPP can be set to depend on the local density, in order to take into account crowding effects.
Self-propelled particles can also be modeled using on-lattice models, which offer the advantage of being simple and efficient to simulate, and in some cases, may be easier to analyze mathematically. On-lattice models such as BIO-LGCA models have been used to study physical aspects of self-propelled particle systems as well as specific questions related to real active matter systems.