Monte Carlo Aggregation Code (MCAC) Part 2: Application to soot agglomeration, highlighting the importance of primary particles

During the agglomeration of nanoparticles and in particular, soot, a change in both the flow regime (from free molecular to near continuum) as well as the change of agglomeration regime (from ballistic to diffusive) is expected. However, these effects are rarely taken into account in numerical simulations of particle agglomeration and yet, they are suspected to have an important impact on the agglomeration kinetics, particle morphologies, and size distributions. This work intends to study these properties by using the Monte Carlo Aggregation Code (MCAC) presented in the preceding work (part 1), focusing on the physical impacts of varying the particle volume fraction and monomers size and polydispersity. The results show an important sensitivity of the kinetics of agglomeration, coagulation homogeneity, and agglomerate morphology to the size of monomers. First, for smaller monomer diameters, the agglomeration kinetic is enhanced and agglomerates are characterized by larger fractal dimensions. Second, for large monomer diameters, fractal dimensions down to 1.67 can be found being smaller than the classical 1.78 for Diffusion Limited Cluster Agglomeration (DLCA) mechanism. One important conclusion is that variation in time of both regimes has to be considered for a more accurate simulation of the agglomerate size distribution and morphology.

Monte Carlo Aggregation Code (MCAC) Part 1: Fundamentals

The application of Monte Carlo methods to simulate the agglomeration of suspended nanoparticles is currently limited to specific agglomeration regimes with reduced accuracy in terms of the particle's physical residence time. The definition of specific particles persistent distance, its corresponding time step and subsequent probabilities for particle displacements may improve the accuracy of this method. To solve these issues, a new persistent distance and its corresponding time step based on Langevin dynamics simulations are introduced. Additionally, a probability of particle displacements, not restricted to a specific agglomeration regime, is introduced. All the modifications are validated by comparison with Langevin dynamics simulations. Finally, the above mentioned modifications considerably improve the accuracy of Monte Carlo methods to predict the dynamics and agglomeration of suspended nanoparticles.