Collective motion of chemically powered nanomotors

Abstract

Self-propelled chemically powered synthetic micron- and nano-scale motors are being intensively studied because of the wide range of potential applications that exploit their directed motion. The collective behavior of chemically powered sphere-dimer motors made from linked catalytic and noncatalytic spheres in a quasi-two-dimensional confined geometry is studied using a coarse-grain microscopic dynamical model. Chemical reactions at the catalytic spheres that convert fuel to product generate forces that couple to solvent degrees of freedom as a consequence of momentum conservation in the microscopic dynamics. The collective behavior of the many-body system is influenced by direct intermolecular interactions among the motors, chemotactic effects due to chemical gradients, hydrodynamic coupling, and thermal noise. Segregation into high and low density phases and globally homogeneous states with strong fluctuations are investigated as functions of the motor characteristics. Factors contributing to this behavior are discussed in the context of active Brownian models. Like their larger micron-scale counterparts, Angstrom-scale chemically powered motors use asymmetric catalytic activity to produce self-generated concentration gradients that lead to directed motion. Unlike their micron-scale counterparts, Angstrom-scale motors are dominated by fluctuations, molecular reorientation often occurs on picosecond time scales limiting the regime where ballistic motion dominates, and solvent sizes are often comparable to those of the motor, so structural effects come into play and solvent depletion forces are very strong. These new features are studied using full molecular dynamics simulations of an Angstrom-scale sphere-dimer motor. The properties of an Angstrom-scale sphere-dimer motor confined between two planar walls separated by distances of tens of nanometers are investigated. The simulations were carried out on many tightly-coupled graphics processing units (GPU) using two massively parallel codes, Nano-dimer and Angstrom-dimer that have been published under a free-software license.