272
M. Michaud et al.
direct numerical simulation is possible up to Re < 4000 in simple reaction geometries (T-mixers [6, 7]). Driving force for solid formation is the chemical potential of
the key component, i.e. the supersaturation given as the ratio of the actual concentration and the equilibrium concentration. Particle formation can be either reactionor mixing-controlled. The former leads to a rather uniform distribution of components in the reactor, whereas the latter leads even in simple geometries such as Tor Y-mixers to more widely distributed component distributions. For flowsheet simulation, however, simple approaches are implemented while sufficient accuracy of
predictions must be guaranteed. In most industrial applications, a well-defined, often
narrow property distribution is targeted.
The topic of the current chapter is the development of such a dynamic flowsheet
simulation module using a well-known moment method for predicting the mean size
and shape of precipitated particles. The model covers mixing, activity-based supersaturation build-up even in systems of complex hydrochemistry, and solid formation
processes. Briesen et al. provided fundamentals to simulate independent particle
properties simultaneously [8]. To expand on this work, a generalized modelling tool
is developed, which is capable of describing the evolution of multi-component and
multiphase particle systems [1, 3, 5], which will then be implemented into a simulation framework. Additionally, our model is used to determine unknown or difficultto-measure material parameters such as surface energy or intermediate products.
The bivariate model formulation allows the prediction of multiple particle properties. In particular, we apply the model to core-shell QDs and to the formation of
needle-shaped crystals. Both systems are examples of particles with two dimensions
influencing the final product property. The model architecture represents a modular
micro-reaction plant consisting of a T-mixer for nucleation and a subsequent vessel for defined particle growth. The setup was characterized with respect to mixing
efficiency and residence time distribution.
The model allows transient predictions of a large number of different multivariate
precipitation processes in dependence of the underlying mixing and hydrochemistry
[9]. Precipitation processes are categorized into mixing- or reaction-controlled systems enabling efficient and problem-specific calculations. The computational effort
for these calculations is kept low by using appropriate numerical simplifications,
such as the DQMOM, which allows the calculation of multiple disperse properties
while keeping the relative error within reasonable bounds. Finally, the individual
sub-models are combined into one single module and coupled with a solver, which
can be integrated into the Dyssol framework [10].
2 Model Architecture
The developed model is best described as a generalized population balance approach
for the subsequent or parallel formation of multiple solid phases with up to two dimensions including agglomeration and ripening. Solid formation is described via the
M. Michaud et al.
direct numerical simulation is possible up to Re < 4000 in simple reaction geometries (T-mixers [6, 7]). Driving force for solid formation is the chemical potential of
the key component, i.e. the supersaturation given as the ratio of the actual concentration and the equilibrium concentration. Particle formation can be either reactionor mixing-controlled. The former leads to a rather uniform distribution of components in the reactor, whereas the latter leads even in simple geometries such as Tor Y-mixers to more widely distributed component distributions. For flowsheet simulation, however, simple approaches are implemented while sufficient accuracy of
predictions must be guaranteed. In most industrial applications, a well-defined, often
narrow property distribution is targeted.
The topic of the current chapter is the development of such a dynamic flowsheet
simulation module using a well-known moment method for predicting the mean size
and shape of precipitated particles. The model covers mixing, activity-based supersaturation build-up even in systems of complex hydrochemistry, and solid formation
processes. Briesen et al. provided fundamentals to simulate independent particle
properties simultaneously [8]. To expand on this work, a generalized modelling tool
is developed, which is capable of describing the evolution of multi-component and
multiphase particle systems [1, 3, 5], which will then be implemented into a simulation framework. Additionally, our model is used to determine unknown or difficultto-measure material parameters such as surface energy or intermediate products.
The bivariate model formulation allows the prediction of multiple particle properties. In particular, we apply the model to core-shell QDs and to the formation of
needle-shaped crystals. Both systems are examples of particles with two dimensions
influencing the final product property. The model architecture represents a modular
micro-reaction plant consisting of a T-mixer for nucleation and a subsequent vessel for defined particle growth. The setup was characterized with respect to mixing
efficiency and residence time distribution.
The model allows transient predictions of a large number of different multivariate
precipitation processes in dependence of the underlying mixing and hydrochemistry
[9]. Precipitation processes are categorized into mixing- or reaction-controlled systems enabling efficient and problem-specific calculations. The computational effort
for these calculations is kept low by using appropriate numerical simplifications,
such as the DQMOM, which allows the calculation of multiple disperse properties
while keeping the relative error within reasonable bounds. Finally, the individual
sub-models are combined into one single module and coupled with a solver, which
can be integrated into the Dyssol framework [10].
2 Model Architecture
The developed model is best described as a generalized population balance approach
for the subsequent or parallel formation of multiple solid phases with up to two dimensions including agglomeration and ripening. Solid formation is described via the