Cyclone separators are widely used in industrial applications. The separation efficiency and pressure drop are the most important parameters to evaluate the performance of processing system. In the simulations,the flow behavior of gas and particles within a square cyclone separator is simulated by means of computational fluid dynamics. The RNG k- ε model and the Reynolds stress model( RSM) are used to model gas turbulence. The flow behavior is examined in the term of tangential velocity components,static pressure and pressure drop contour plots for flow field and solid volume fraction. The effects of the turbulence model and solid volume fraction on the square cyclone are discussed. The results indicate that the pressure drop increases with the increase of solid volume fraction,and increase with the increase of inlet velocities for two turbulence models, moreover,the simulations results are compared with pressure field. For all runs,the RSM model gives a higher pressure drop compared to the RNG k- ε model. The RSM model provides well the forced vortex and free vortex,and captures better the phenomena occurring during intense vortex flow in the presence of walls within cyclone separators. The separation efficiency and pressure drop are the most important parameters to evaluate the performance of processing system. In the simulations, the flow behavior of gas and particles within a square cyclone separator is simulated by means of computational The dynamics of the RNG k-ε model and the Reynolds stress model (RSM) are used to model gas turbulence. The flow behavior is examined in the term of tangential velocity components, static pressure and pressure drop contour plots for flow field and solid volume fraction. The effects of the turbulence model and solid volume fraction on the square cyclone are discussed. The results that that the pressure drop increases with the increase of solid volume fraction, and increase with the increase of inlet velocities for two turbulence models, moreover, the simulations results are compared with pressure field. For all runs, the RSM model gives a higher pressure drop compared The RNM k- ε model. The RSM model provides well the forced vortex and free vortex, and captures better the phenomena occurring during intense vortex flow in the presence of walls within cyclone separators.