Pattern formation in nanoparticle suspensions: a Kinetic Monte Carlo approach

Various experimental settings that involve drying solutions or suspensions of nanoparticles often called nano-fluids have recently been used to produce structured nanoparticle layers. In addition to the formation of polygonal networks and spinodal-like patterns, the occurrence of branched structures has been reported. After reviewing the experimental results, the work presented in this thesis relies only on simulations. Using a modified version of the Monte Carlo model first introduced by Rabani et al. [95] the study of structure formation in evaporating films of nanoparticle solutions for the case that all structuring is driven by the interplay of evaporating solvent and diffusing nanoparticles is presented. The model has first been used to analyse the influence of the nanoparticles on the basic dewetting behaviour, i.e., on spinodal dewetting and on dewetting by nucleation and growth of holes. We focus, as well, on receding dewetting fronts which are initially straight but develop a transverse fingering instability. One can analyse the dependence of the characteristics of the resulting branching patterns on the driving effective chemical potential, the mobility and concentration of the nanoparticles, and the interaction strength between liquid and nanoparticles. This allows to understand the underlying fingering instability mechanism. We describe briefly how the combination of a Monte Carlo model with a Genetic Algorithm (GA) can be developed and used to tune the evolution of a simulated self-organizing nanoscale system toward a predefined nonequilibrium morphology. This work has presented evolutionary computation as a method for designing target morphologies of self-organising nano-structured systems. Finally, highly localised control of 2D pattern formation in colloidal nanoparticle arrays via surface inhomogeneities created by atomic force microscope (AFM) induced oxidation is presented and some simulations are shown. Furthermore, the model can be extended further, and by including the second type of nanoparticle, the binary mixture behaviour can be captured by simulations. We conclude that Kinetic Monte Carlo simulations have allowed the study of the processes that lead to the production of particular nanoparticle morphologies.