Detailed simulations of transport and nanoparticle formation in a hot-wall reactor

Detailed simulations of transport and nanoparticle formation in a hot-wall reactor presented by Patrick Wollny on 2016-11-28
Authors: Patrick Wollny, Johannes Sellmann, Hans Orthner, Hartmut Wiggers, Irenäus Wlokas, Andreas Kempf
Address: 900 Boylston St, Hynes, Level 2, Room 209, 02115, Boston, United States
Keywords: pyrolysis, chemical synthesis

Date: 2016-11-28 (13:45:00 - 14:00:00)


Detailed simulations of the transport phenomena and the of particle dynamics have been conducted for a hot wall reactor. Simulations are essential for the scaling of laboratory experiments to pilot and industrial plants, as the similarity parameters can rarely be preserved. Two modeling frameworks for the representation of the dispersed particle phase (Euler-Lagrange and Euler-Euler) are presented. Both models are compared and validated against data measured in a laboratory scale hot-wall reactor producing iron nanoparticles by pyrolysis of iron pentacarbonyl in a nitrogen atmosphere. A Monodisperse model1 for the population balance equation of the nanoparticles has been implemented into the open-source library OpenFOAM for both (Eulerian and Lagrangian) representations of the dispersed phase. Thermophoretic transport of the particle phase is described following Li and Wang2; particle diffusion is modeled as a diffusive flux or through a random walk. Concentrations of the gas phase are solved by transport equations, with a finite rate model describing chemical reactions. The Lagrangian and Eulerian implementations were individually tested against generic setups and later applied in simulations of a laboratory scale hot-wall reactor. The laminar flow field, simple reaction kinetics and the known wall temperature profile made the hot-wall reactor an ideal target for validation of the different models describing the physical effects and their impact on the major nanoparticle characteristics. The experimental setup allowed TEM probing and was additionally attached to a scanning mobility particle sizer system in parallel. The comparison of the experimental data with both simulation approaches shows a good agreement in the mean particle size. Buoyancy effects show a crucial impact on the alignment and recirculation strength of the three dimensional flow field. Thermophoresis influences notably the particle size and distribution in the reactor. Applying diffusive random walk, the Lagrangian approach is capable to model a particle size distribution. Furthermore, rare big particles released from the recirculation zone of the reactor can be captured leading to a more realistic particle distribution. Therefore, it is pointed out that the Euler-Lagrange approach is more suitable for modelling nanoparticle formation in this context. Nevertheless, the Lagrangian model requires a higher computational effort, which makes the Euler-Euler approach attractive for parameter studies. The work is supported by the European commission in the Horizon 2020 framework, project Nanodome (reference: 646121). 1 Kruis, F. Einar, et al. A simple model for the evolution of the characteristics of aggregate particles undergoing coagulation and sintering. Aerosol science and technology, 1993, 19. Jg., Nr. 4, S. 514-526. 2 LI, Zhigang; WANG, Hai. Thermophoretic force and velocity of nanoparticles in the free molecule regime. Physical Review E, 2004, 70. Jg., Nr. 2, S. 021205.

University of Duisburg-Essen
Carl-Benz Straße 199

This might also be of interest to you: