(399b) A Computational Study of the Various Flow Regimes in Pneumatic Conveying of Granular Materials

Gas-solid systems are commonly encountered in the chemical and petrochemical, food and mineral processing and pharmaceutical industries. Their applications include fluid catalytic cracking, drying operations, mixing and granulation and the transport of granular material and fine powders through pipelines. In particular, the pneumatic transport of granular material is a common operation frequently employed to transport solid particles from one location to another. Depending on the system geometry, gas velocities and material properties of the solid particles to be transported, such transportation processes can take place in different modes usually referred to as dense or dilute-phase conveying.

Numerical modeling of pneumatic conveying and other gas-solid systems plays an important role in improving our understanding of such systems. One of the commonly used approaches to pneumatic conveying modeling is the Eulerian/Lagrangian method where particles are tracked in a Lagrangian frame of reference either individually or as groups with identical properties known as parcels (Tashiro et al., 1997; Huber and Sommerfeld, 1998). An alternative approach has been Computational Fluid Dynamics (CFD) with two-fluid continuum models to represent the gas and solid phases as two interpenetrating continua (Levy, 2000). Further, the technique of particle dynamics simulation has also been widely used for investigations of granular and gas-solid systems. In particular, the Discrete Element Method (DEM) originally developed by Cundall and Strack (1979) for describing the mechanical behavior of assemblies of discs and spheres, has been successfully applied by many research workers in various areas of engineering interests. Tsuji et al. (1992) carried out numerical simulations of horizontal pneumatic conveying of solid particles using DEM coupled with CFD and showed that particles moved in the form of plugs in the conveying pipe. Several research workers have also applied the same approach of combining DEM with CFD to the simulation of two-dimensional fluidized beds (Tsuji et al., 1993; Xu and Yu, 1997; Mikami et al., 1998; Kaneko et al., 1999; Rhodes et al., 2001). Li and Mason (2000) used the same approach to model heat transfer between gas, solid particles and pipe wall in a pneumatic conveying system. Han et al. (2003) simulated the flow of salt particles through a dilute phase pneumatic conveying system to predict particle attrition and breakage.

An understanding of the differences in physics between the various flow regimes found in pneumatic conveying of granular material may be important to actual industrial or commercial applications with regards to the optimality of operation, ease of control and extent of damage inflicted on the solid particles as well as the conveying lines. Despite the large number of work reported on gas-solid systems, there have been relatively fewer attempts at modeling the various flow regimes in vertical and horizontal pneumatic conveying systems. The ability to predict the flow behaviors of both gas and solid phases during a typical pneumatic conveying operation or the modes in which the transportation would take place remains limited. As such, the objective of this study is to apply the technique of combining DEM with CFD to the numerical simulation of pneumatic conveying of granular material in both vertical and horizontal pipes. The emphasis has been on reproducing computationally the different types of solid flow patterns and behaviors observed experimentally under different operating conditions.

The geometry of the pneumatic conveying system and type of particles used in the present simulations were based on the experimental work of Rao et al. (2001) and Zhu et al. (2003) so that a meaningful comparison between the simulation and experimental outputs can be made. The gas velocities considered in this study were in the ranges 14 m/s ? 24 m/s and 10 m/s ? 30 m/s for the vertical and horizontal pneumatic conveying simulations respectively because these would include all the flow regimes observed via Electrical Capacitance Tomography measurements by Rao et al. (2001) and Zhu et al. (2003) for the two systems. The numbers of particles used were 500, 1000, 1500 and 2000 corresponding to overall solid concentrations (defined as the overall volume fraction of particles divided by the volume fraction of particles at maximum packing, 0.65) of 0.08, 0.16, 0.24 and 0.32 respectively. The flow patterns from the numerical simulations were then compared qualitatively with the experimental observations of Rao et al. (2001) and Zhu et al. (2003).

The simulation results obtained were in good agreement with previously reported experimental observations in terms of the types of flow patterns arising at different operating conditions used. In vertical pneumatic conveying, particles tend to be dispersed throughout the pipe at high gas velocities and low solid concentrations. On the other hand, particles tend to cluster together and move in the form of a dense plug when gas velocities are low or solid concentrations are high. These flow patterns have been referred to as the dispersed and plug flow regimes respectively. The solid concentration profile for dispersed flow was observed to be symmetrical and with a minimum near the center of the pipe while that for plug flow was almost flat. In horizontal pneumatic conveying, the simulations also show the presence of homogeneous or slug flow regimes where particles are distributed along the length of the pipe or packed together as a large cluster respectively. In addition, due to the effects of gravitational forces which cause particles to settle towards the bottom wall of the horizontal pipe, the stratified and moving dunes flow regimes where particles are transported by traction along the pipe wall are observed at low gas velocities and solid concentrations. The solid concentration profile for stratified flow was unsymmetrical with higher concentration near the lower wall of the pipe while that for slug flow was similar to the flat profile seen for plug flow in vertical pneumatic conveying. The various flow regimes and their corresponding operating conditions have been represented in the form of phase diagrams. In the range of gas velocity values where transition between two flow regimes might be taking place in a vertical pipe, hysteresis of the solid flow rates was observed to occur. The types of flow regimes obtained at the different operating conditions were observed to be insensitive to parameters used in the mathematical model within the range of values investigated in a sensitivity analysis of these parameters. However, the steady state solid flow rate showed a marginal decrease with increasing coefficient of friction. Solid particles with a low viscous contact damping coefficient or equivalently high coefficient of restitution have a low tendency to form large clusters and the plug flow regime normally observed in vertical pneumatic conveying may not exist under such conditions. The mathematical model used in the present study may also be extended to include the effects of electrostatics or particle attrition to investigate the influences of such effects in pneumatic conveying systems in subsequent studies. In particular, it may be possible to simulate the few other types of flow regimes mentioned earlier which have been observed in physical experiments with the incorporation of such effects.

References:

Cundall, P. A., and O. D. L. Strack, ?A discrete numerical model for granular assemblies,? Geotechnique, 29, 47 (1979).

Han, T., A. Levy, and H. Kalman, ?DEM simulation for attrition of salt during dilute-phase pneumatic conveying.? Powder Technol., 129, 92 (2003).

Huber, N., and M. Sommerfeld, ?Modelling and numerical calculation of dilute-phase pneumatic conveying in pipe systems,? Powder Technol., 99, 90 (1998).

Kaneko, Y., T. Shiojima, and M. Horio, ?DEM simulation of fluidized beds for gas-phase olefin polymerization,? Chem. Eng. Sci., 54, 5809 (1999).

Levy, A., ?Two-fluid approach for plug flow simulations in horizontal pneumatic conveying,? Powder Technol., 112, 263 (2000).

Li, J., and D. J. Mason, ?A computational investigation of transient heat transfer in pneumatic transport of granular particles,? Powder Technol., 112, 273 (2000).

Mikami, T., H. Kamiya, and M. Horio, ?Numerical simulation of cohesive powder behavior in a fluidized bed,? Chem. Eng. Sci., 53, 1927 (1998).

Rao, S. M., K. Zhu, C.-H. Wang, and S. Sundaresan, ?Electrical capacitance tomography measurements on the pneumatic conveying of solids,? Ind. Eng. Chem. Res., 40, 4216 (2001).

Rhodes, M. J., X. S. Wang, M. Nguyen, P. Stewart, and K. Liffman, ?Study of mixing in gas-fluidized beds using a DEM model,? Chem. Eng. Sci., 56, 2859 (2001).

Tashiro, H., X. Peng, and Y. Tomita, ?Numerical prediction of saltation velocity for gas-solid two-phase flow in a horizontal pipe,? Powder Technol., 91, 141 (1997).

Tsuji, Y., T. Kawaguchi, and T. Tanaka, ?Discrete particle simulation of two-dimensional fluidized bed,? Powder Technol., 77, 79 (1993).

Tsuji, Y., T. Tanaka, and T. Ishida, ?Lagrangian numerical simulation of plug flow of cohesionless particles in a horizontal pipe,? Powder Technol., 71, 239 (1992).

Xu, B. H., and A. B. Yu, ?Numerical simulation of the gas-solid flow in a fluidized bed by combining discrete particle method with computational fluid dynamics,? Chem. Eng. Sci., 52, 2785 (1997).

Zhu, K., S. M. Rao, C.-H. Wang, and S. Sundaresan, ?Electrical capacitance tomography measurements on vertical and inclined pneumatic conveying of granular solids,? Chem. Eng. Sci., 58, 4225 (2003).