Nanoparticle (NP) aggregation plays a crucial role in NP synthesis, which is increasingly relevant due to the favourable surface-to-volume properties of NPs. In molten salt reactors (MSRs), insoluble solid fission products (SFPs) form metallic nanoparticles that can aggregate and
...
Nanoparticle (NP) aggregation plays a crucial role in NP synthesis, which is increasingly relevant due to the favourable surface-to-volume properties of NPs. In molten salt reactors (MSRs), insoluble solid fission products (SFPs) form metallic nanoparticles that can aggregate and deposit on internal reactor components, reducing heat-transfer performance and complicating maintenance. Although the classical DLVO theory describes the aggregation kinetics, its predictive use is limited by the large number of system-specific parameters involved. Moreover, literature indicates that aggregation behaviour is typically examined qualitatively and on a case-by-case basis, lacking a general theory to relate different systems.
This thesis presents the development of a novel general non-dimensional framework that is based on DLVO theory. The framework reduces the DLVO potential parameters to two key coefficients, representing the relative strength and relative interaction of the electric double layer (EDL). Their combination uniquely defines the shape of the potential and enables classification of aggregation regimes: rapid aggregation, barrier-limited aggregation, or stable dispersion. The framework is applied to both sphere–sphere and sphere–plate geometries to capture bulk aggregation and sedimentation.
The framework is implemented and used in coarse grained (CG) molecular dynamics (MD) simulations to establish a first quantitative relation between the two key coefficients and aggregation kinetics. NPs are represented as spherical particles interacting through the dimensionless DLVO potential and the molten salt medium modelled using neutral beads interacting through a Lennard-Jones (LJ) potential. A parameter sweep across the non-dimensional coefficients quantifies the aggregation time and final cluster size with statistical certainty. The results demonstrate that the dimensionless coefficients are able to predict the aggregation behaviour. For MSR conditions, the coefficients place the system deep into the primary-minimum aggregation regime, indicating that metallic NPs will aggregate regardless of surface charge.
The developed framework allows for a first prediction on aggregation behaviour. Future research should focus on refining the CG simulation model, extending it to three dimensions and validating the results with experimental measurements. Additional insight is acquired by exploring concentration-dependent aggregation and extending the sphere-sphere framework beyond the near-field approximation.