Explosive dissemination and flow of nanoparticles Veeraya Jiradilok b , Dimitri Gidaspow a, ⁎ , Jalesh Kalra a , Somsak Damronglerd b , Suchaya Nitivattananon b a Department of Chemical and Environmental Engineering, Illinois Institute of Technology, Chicago, IL 60616, United States b Department of Chemical Technology, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand Received 9 August 2005; received in revised form 20 December 2005; accepted 21 December 2005 Available online 6 March 2006 Abstract Flow properties of 10 nm silica particles were determined in a two-story circulating fluidized bed riser. The pressure drop has a minimum at a velocity of 0.3 m/s. Particle concentrations were measured with a gamma ray densitometer. The solid fluxes were measured with a suction probe. From these measurements the nanoparticle viscosity was estimated. The measured viscosity is close to an estimate obtained from kinetic theory, assuming Brownian motion of nanoparticles. The viscosity and the previously measured solid stress modulus were used in a multiphase CFD code to study the behavior of explosive dissemination of mixtures of nanoparticles and micron size particles. The dissemination process was divided into two steps: early-time hydrodynamics and dissemination into an atmosphere. In the early-time hydrodynamic step the particles were accelerated by means of a high pressure and high temperature gas from a plastic explosive. When the device containing the particles broke, the early-time hydrodynamic velocities, concentrations, pressure and temperatures were used as the initial conditions for the dissemination step. This study shows how to use CFD to design a dissemination device that will prevent the overheating of a mixture of particles to be disseminated. The computed phenomena were similar to the experimental observations. The nanoparticles formed a cloud with a vortex ring structure for dissemination of small micron size particles and nanoparticles. For the dissemination of 100 μm aluminum and 10 nm silica particles, there was no vortex ring structure. As expected, the larger particles settled on the ground. The computed ground concentrations can be used to compare the model with observations, such as the covering of ground by dust after volcanic eruptions. © 2006 Elsevier B.V. All rights reserved. Keywords: Circulating fluidized bed; Computational fluid dynamics; Particular viscosity; Aerosol; Pyrotechnics 1. Introduction Nanoparticles have some unique flow and dispersion properties that make them useful for a number of applications [1]. Fumed nanoparticle silica has long been used as a flow agent to disperse sticky particles such as TNT. Recently, we [2,3,1,4,5] have shown that many nanoparticles fluidize without the formation of bubbles. Instead of forming bubbles upon the increase of gas velocity, like the FCC particles used to convert oil into gasoline, they keep expanding upon an increase of gas velocity. If the silica particles used here are coated with catalysts using chemical vapor deposition techniques, they may be an alternative to the conventional FCC particles used in the oil industry. Here we demonstrate that the silica nanoparticles can be circulated in a pilot plant type circulating fluidized bed. We have also used this apparatus to estimate the nanoparticle viscosity that is needed as an input into CFD models. Then we demonstrate that our CFD models [6] can be used to predict the dispersion of nanoparticles that are of interest in forming smoke and other obscurants [7]. Computational fluid dynamics (CFD) explosive dissemina- tion started at IIT about 20 years ago [8]. The process was divided into 2 steps: early-time hydrodynamics and dissemina- tion into an atmosphere or a bag. In early-time hydrodynamics, pressure wave propagation was computed in a dissemination device containing powder. When the dissemination device broke, the flow of powder into atmosphere began using the Powder Technology 164 (2006) 33 – 49 www.elsevier.com/locate/powtec ⁎ Corresponding author. E-mail address: gidaspow@iit.edu (D. Gidaspow). 0032-5910/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.powtec.2005.12.020