Particle formation in turbulent flows arises in a large range of industrial and environmental processes. Examples include the formation of soot particles in combustion engines, the production of nanoparticles via flame synthesis or liquid-phase reactors, or the evolution of aerosol particles in urban/non-urban environments. Simulations' can significantly improve the understanding and design/control of these processes but several difficulties arise when trying to model the underlying physics. These are mainly rooted in both the requirements of appropriate description of the particle's polydispersity and dynamics, as well as of the random fluctuations of the turbulence. In the preliminary part of this work, the turbulent effects on the particle's dynamics are discussed, showing how correlations of various orders appear in the main equation that describes them, the Population Balance Equation (PBE). The thesis elaborates a new approach for the simulation of particle formation in inhomogeneous turbulent flows, the PBE-PDF method, whose concept was recently introduced by [105]. An equation for the joint pdf of the number density and various scalars (such as the chemical species kinetically involved) is derived from fundamental equations that describe the class of considered problems. An algorithm for the numerical solution of the pdf equation is developed, based on Monte Carlo simulations. The method is discussed both at the level of the general concept of the stochastic simulations and at the level of the specific requirements of the PBE-PDF approach. The structure of the implementation is discussed in detail. Starting from empirical or theoretical kinetics, the PBE-PDF is applied to two different turbulent processes involving particle nucleation and growth. The first case considered involves the precipitation of barium sulphate crystals in a tubular reactor, while the second one involves dibutyl phthalate condensation in a free jet. For both cases the method allows for clear analysis and understanding of the particulate phase evolution, emphasis being put on the turbulent effects. Turbulence is found to have significant impact on the overall processes by spatially redistributing the intensity of the particles' mechanisms. Both processes were selected because previously studied experimentally [5][63], which allows the comparison of computed particles' size distribution with measured data. For the crystallization process the simulations' results can be directly compared, showing excellent agreements. Identified uncertainties in the experimental methods are discussed for the aerosol process, even if the computed distributions still show good agreement with the measured ones, particularly for intermediate molar fractions values.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:515197 |
| Date | January 2009 |
| Creators | Di Veroli, Giovanni Yohanan |
| Publisher | University of Manchester |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0016 seconds