Nanoparticle deposition in tubes: statistical study using the response surface methodology Edson R. Nucci , Simone A. Pozza, Andre N.C. Barros, José Renato Coury Federal University of São Carlos, Chemical Engineering Department, Rodovia Washington Luís (SP-310), km 235 São Carlos - SP - Brazil CEP 13565-905 ABSTRACT: The aim of the present work was to investigate the particle loss to walls during the flow of the nanoparticles through ducts. Operational conditions to investigate this loss during the flow through circular tubes were determined with nine parameters under consideration: aerosol flow rate, the tube diameter and the tube length. The influence of these three variables was studied using a 2 3 full factorial design and axial points with a total of 18 experiments. Response surface methodology was adopted and an empirical linear polynomial model constructed. The results showed that total number of particles depend on the tube length and diameter, whilst the particle median diameter was strongly dependent on the tube diameter. Keywords: factorial design, particle deposition, nanoparticles INTRODUCTION The sizing of nanoparticles present in aerosols, inevitably, involves sampling and transport of the solid suspension, usually through tubing. In the last years, there is an increasing advance in the technological applications of nanoparticles. This interest is due to the enhanced optical, mechanical, magnetic and electrical properties associated with their small size (CHEN et al., 1998). These interesting characteristics, nevertheless, can be undesirable when these particles are released into the environment (HOSOKAWA et al, 2007). Extensive experimental and computational studies related to particle transport in duct have been reported in the literature (e.g. ZHANG and LESSMANN, 1997; ZHANG and AHMADI, 2000; ZHAO and WU, 2006). However, attention devoted to nanoparticle tarnsport is still scarce (AROUCA et al., 2009; FEITOSA et al., 2009; SHAMS et al. 2009). The aim of the present work was to investigate the particle loss to walls during the flow of the nanoparticles through circular tubes using a factorial experimental design. The effects of three variables (aerosol flow rate, the tube diameter and the tube length) on nanoparticles deposition in tubes were simultaneously investigated using a 2 3 full factorial design leading to 18 sets of experiments. The choice of variables was made based on particle flow during aerosol sampling. The application of statistical design was used to determine the optimum operating conditions for the system. This technique is very useful for processes in which several variables must be considered simultaneously (DIAS et al., 2001). MATERIAL AND METHODS The test aerosol was generated by atomization of a sodium chloride (NaCl) 0.5 g.L -1 solution in a particle-free air stream. The solution was sprayed to form a cloud of droplets and, with the immediate evaporation of the solvent in the droplet, solid particles of salt remained suspended in the air stream. The aerosol was then directed through a desiccating column with silica gel (diffusion dryer) to ensure complete drying and removal of excess moisture, followed by electrostatic charge neutralization (with Kr-85). After that, the