A Computational Study of Particulate Colloidal Systems

Particulate colloids exist in a broad range of natural, biological and industrial systems. To examine the particle aggregation behaviour in these systems, and to propose feasible aggregation control strategies, a novel computational model is developed based on the Eulerian-Lagrangian framework. Computational fluid dynamics techniques are utilised to simulate the flow field of the base fluid, and discrete phase modelling approach is used to compute the particle trajectories, with both long-range and contact particle-particle interactions considered. Two-way coupling is adopted to model particle-fluid interactions. Three studies of different colloidal systems are conducted using the developed model. In the first study, the morphology of gold nanoparticles is analysed at various particle volume fractions and base fluid pH levels. It is observed that small isolated particle clusters are formed at low particle volume fractions, whilst a more complex and interconnect particle network is formed at higher particle volume fractions. It is also found that larger but more compact particle aggregates are formed under acidic conditions (pH = 3.5), compared with neutral and basic conditions (pH = 6.7 and 9.4). The second study investigates the particle structures in magnetic nanofluids under static external magnetic fields with constant direction and magnitude. In the absence of the magnetic field, a randomly-oriented particle network is formed. With the application of external magnetic fields, distinct chain-like particle structures are observed parallel to the direction of the applied magnetic fields. The formation of the chain-like structures is more rapid, and the chain-like structures formed are thicker at higher magnitudes of the magnetic field. This observation is consistent between the two-dimensional and three-dimensional simulations. A novel particle aggregation control strategy is proposed in the third study via the application of alternating and rotating magnetic fields. It is demonstrated that the aggregation of magnetic-responsive particles is promoted under rotating magnetic fields, especially at higher rotating frequencies. It is also demonstrated that complete particle disaggregation can be achieved under alternating magnetic fields. Lastly, it is concluded that the computational model developed is a breakthrough in numerically predicting the particle aggregation behaviour in colloidal systems, and can be extended to many other studies.