(439f) Large Eddy Simulations of Titania Nanoparticle Evolution in a Turbulent Flame with Detailed Chemistry Conference: AIChE Annual MeetingYear: 2009Proceeding: 2009 AIChE Annual MeetingGroup: Particle Technology ForumSession: Applications of Engineered Structured Particulates Time: Wednesday, November 11, 2009 - 5:20pm-5:43pm Authors: Sung, Y. "., University of Texas at Austin Fox, R. O., Iowa State University Mehta, M., Iowa State University Raman, V., University of Texas at Austin Pratsinis, S. E., Swiss Federal Institute of Technology, Particle Technology Laboratory, ETH Zurich Titania (TiO2) nanoparticles have widespread practical use in pigments, photocatalysis, and as catalyst supports. Each year, over 3 million tons of titania nanoparticles are produced, for which gas-phase combustion based synthesis has been a dominant method due to its advantage in large-scale production. In this flame-based synthesis, titanium chloride (TiCl4) vapor is oxidized in a pre-existing turbulent jet flame, which then forms solid nanoparticles through interactions with the gas-phase species. The final size distribution of the nanoparticles is dictated by the nonlinear interaction of turbulence, gas-phase flame and titanium chloride chemistry, and titania particle evolution. In order to accurately compute this complex process, it is crucial to account for these interactions using detailed physical models. For this, we present a computational approach based on the large eddy simulation (LES) methodology, where the gas-phase titanium chloride evolution and methane combustion are represented using very detailed chemical kinetics. In particular, the transformation of titanium chloride to the solid primary nucleating particle is represented using 30 species and 51 reactions. The evolution of the titania number density function is tracked using the quadrature method of moments (QMOM) approach. For validation purposes, this detailed computational model is first compared against an experimental setup. The effect of flame configuration on the final product distribution is then discussed.