Motion of driven and active particles in polymer networks

Understanding how small particles (such as proteins, bacteria, viruses and drugs) move through fluids in the human body is of great importance for biomedical applications: this applies both to preventing bacteria to reach human organs as well as drug delivery. Migrating through the body, these particles have to overcome physical barriers such as mucus layers which line many organs in the human body and consist of large biopolymers called mucins which form a dense network in a water background, similar as spaghetti but at much smaller scales. So far most of the experimental and theoretical studies focused on a random motion of particles, which get stuck in the network in case they are too large. One possibility to overcome such particle trapping is the use of magnetic fields which can ensure particle transport of magnetically functionalized drugs to specific target sites. Another promising approch are so-called active colloids which can perform autonomous motion, similar as biological microorganisms such as bacteria. Still, no clear overall understanding exists about the relevant physical forces how driven and active particles can be transported through biopolymer networks. In our project we perform computer simulations to uncover the physical mechanisms how otherwise trapped particles can be transported through polymer networks when they are magnetically driven or made active. We use a state-of-the-art simulation approach, called multiparticle collision dynamics, which is able to simulate faithfully the flows around the tagged particle; these simulations also include thermal noise, which is of great importance on the nano-scale. With this method at hand we will find out the microscopic details how the presence of driven or active particles influence the local propereties of the polymer network which is currently not possible with experimental techniques. In turn, we can see how the polymer network configuration around the particles may help or hinder their motion. We expect that there exists a critical driving or active force which is necessary to enable particles to migrate. We also suspect that the dynamics of the polymer cross-links, which connect individual polymers, play a central role in this scenario, in particular, when these links are allowed to form and to break-up dynamically, as this is the case for many real polymer networks. In the last part of the project we simulate the motion of particles in an externally imposed fluid flow to model fluid flows in the human body. We expect that the presence of such an external force will support and enhance the transport of particles which is possible to find out with our detailed numerical simulations. In this project we will closely collaborate with experimentalists in order to test our predictions in real applications.

The fundamental question how microorganisms such as bacteria, sperm cells or other "active particles" are able to move surprisingly efficient in complex environments in the human body such as in biofluids (blood, mucus,) is still poorly understood. The Lise Meitner project M 2458 investigated the physical locomotion mechanisms of such active particles and of externally driven (gravity, fluid flow) particles in polymer networks and microchannels. All in all, the results of the project helps to better understand how microorganisms move in the human body and in external fluid flows. Many of the obtained results have been published in different important scientific journals, including well-known interdisciplinary journals such as Nature Communications, Science Advances, or PNAS. Different numerical methods have been used to simulate fluid flows at the micron scale including Brownian fluctuations, such as the MPCD ("multi-particle collision dynamics") method in combination with molecular dynamics (MD) or Brownian dynamics (BD) simulations. On the one side the project focused on computer simulations unraveling fundamental locomotory behavior of active and externally driven particles in complex media and fluid flows. On the other side we investigated the dynamics of bacteria and driven particles together with international experimental collaborators. All in all our investigations revealed a multitude of interesting, significant, and partly surprising results. For example, it could be demonstrated that the specific locomotory swimming gait and the corresponding self-generated fluid flows of spherical microswimmers strongly determines their ability and efficiency to move through polymer networks. This is particularly interesting since in the absence of a polymer matrix the used model microswimmers show a swimming speed independent of the specific swimming mechanism. Furthermore, together with experimentalists from ESPCI Paris we could demonstrate the physical principles which enable bacteria to swim upstream in microchannels. We could also demonstrate how so-called chirality - or handedness - of bacterial flagella determine bacterial dynamics and how left-right symmetry of their trajectories can be broken. This is also of clinical relevance, related to the fact that bacteria are able to infect patients through upstream swimming in catheters. Our fundamental physical and hydrodynamic understanding of these phenomena enabled us to propose an idea how to mechanically design and pattern surfaces of catheters in order to prevent bacterial infections. Finally we showed via modern machine learning techniques using neural networks and reinforcement learning, how microswimmers can make decisions in a viscous environment and using sensory input on their swimming gaits in order to efficiently move towards nutrient sources.