“Effect of Apex Cone Shape on Fine Particle Classification of Gas-cyclone” Hideto Yoshida*, Yusuke Nishimura, Kunihiro Fukui and Tetsuya Yamamoto * Department of Chemical Engineering, Hiroshima University, 1-4-1, Kagamiyama Higashi-Hiroshima, 739-8527, Japan Corresponding Author Hideto Yoshida Department of Chemical Engineering, Hiroshima University, 1-4-1, Kagamiyama Higashi-Hiroshima, 739-8527, Japan Tel. +81+82-424-7853 FAX +81+82-424-5494 e-mail [email protected] 1 “Effect of Apex Cone Shape on Fine Particle Classification of Gas-cyclone” Key Words: Gas-cyclone, Particle separation, Cut size, Particle size, Sub-micron particles Abstract The purpose of this paper is to study the effects of apex cone shape on particle separation performance of gas-cyclones by experiment and CFD studies. It is found that the optimum apex cone angle is 70 deg. The minimum 50% cut size was obtained by use of this special apex cone. From the flow visualization method by use of soap foam, the upward flow and downward flow coexisted on the surface of this special apex cone. The clear interface between upward flow and downward flow was detected on the apex cone angle of 70 deg. The effect of the apex cone angle on particle separation performance decreases under high inlet velocity conditions, because most particles are moving in the area away from the apex cone. The particle separation performance and flow visualization results qualitatively supported the 3-dimensional CFD simulation based on the direct method. 2 Introduction Gas-cyclones are widely used in separation or size classification apparatus for gas-solid flows because of their simple structure and low cost. Recently, with improvements in the dimensions of various parts of cyclones, a fairly high level of precision in the classification of sub-micron order particles has become possible［1-3］. Recent industrial requirements of ceramics or metal powders require particle size distributions with a narrow standard deviation, because the physical properties of such classified particles are related to the performance of electrical, magnetic and chemical reactions. In particular, particles with a diameter of 0.1 to 1 μm are required for various powder handling processes. In order to classify a particle cut size in the size range of 0.1 to 1μm, forced centrifugal classifiers are not always effective because of particle erosion problems and the increased maintenance costs of such classifiers due to high revolutions. While Iinoya et al.［4］found that it is possible to classify powders even in the 0.4μm size range using special cyclones, it is generally difficult to change the cut size in a conventional cyclone separator. To solve this problem, Yoshida et al.［1, 5］ found that it is possible to change the cut size by the use of a moving circular guide plate at the cyclone inlet or by the use of an additional secondary flow injection method in the upper cylindrical part of the cyclone. By use of a special apex cone at the inlet part of the dust box and by using the blow-up method, it is reported that the cut size can be easily controlled from 1μm to 40μm size range［6］. In order to increase particle separation performance of a gas-cyclone, the effect of an apex cone used at the upper part of a dust box was examined and it is confirmed that rotational and upward fluid velocity components decreased with the apex cone set at the dust box［7］. However, the optimum shape of the apex cone in the gas-cyclone was not clear. The purpose of this paper is to find the optimum apex cone shape in a gas-cyclone by use of simulation and experimental method. By use of the flow visualization method near the apex 3 cone, the optimum apex cone angle was determined. Interesting and new conclusions that can be used to increase particle separation performance of gas-cyclones are obtained. 1 Experimental apparatus Figure 1 shows a schematic diagram of the gas-cyclone used in this study. The cyclone diameter was set at 72mm and each dimension was determined from experimental data that proved superior in classification performance［4］. The secondary flow injection method in the upper part of the cyclone as shown in Fig. 1 was used to improve classification performance ［5］. The performance of the cyclones were examined by use of the partial separation efficiency Δη defined by: m f D c c p  (1) m f D  m f D c c p s s p In the above equation, m and m are the masses of the collected particles in the coarse and c s fine sides, respectively. The particle size frequency distributions for each size range are indicated by f and f . Particle size distributions were measured by Laser-light scattering c s (Horiba Co., Ltd. LA-920). The experimental error in the partial separation efficiency was about 3 percent. A ring nozzle-type particle disperser (Nisshin Eng. Co., Ltd.) was installed after the screw feeder. The test particle used was Kanto Loam (JIS, Z8901, No. 11). The mass median diameter was 1.8μm and true density was 2900 kg/m3. The total flow rate, including secondary flow rate of 100 l/min. was changed from 500 to 700 l/min. and powder flow rate was set about 2.0 g/min. Figure 2 shows the dimensions of the test cyclone. In this study, various types of apex cones were used at the inlet part of the dust box. The cone angles examined were from 40 deg. to 80 deg. In general, the cone angle of 60 deg. is used in the gas-cyclone, but the optimum cone angle of the apex cone has not been clearly investigated. For each experiment, the spacing 4 between the bottom of the cylindrical part and the bottom of the apex cone was set to be 31.2 mm. Under this condition, the top position of the apex cone and the bottom of the cylindrical wall were set at nearly the same height. 2 Experimental results Figure 3 shows the experimental results of the classification performance with and withoutthe apex cone. In this case, the cone angle of 60 deg. was used in the experiment. It is found that the 50% cut size with the apex cone in use was smaller than that without the apex cone. The decrease of particle collection efficiency without the apex cone might be particle re-entrainment from the dust box due to strong upward and rotational fluid components remained in the upper part of the dust box. To examine the flow differences with and without the apex cones, acryl particles of about 2 mm were initially set on the bottom of a transparent dust box. From the flow observations shown in Fig. 3, the acryl particles are lifted and moved to various directions in the case without the apex cone, but the moving particles are not detected in the case with the apex cone. This fact indicates that the upward and rotating fluid components remained in the dust box in the case without the apex cone and the probability of the outgoing particles from the dust box is increased. In order to clear the flow difference in the dust box, the flow visualization using water supplied on the bottom of the transparent dust box was carried out. Figure 4 shows the experimental results of flow visualization with and without the apex cone. In case without the apex cone, violent changes in the liquid level was observed, but movement of the water surface is not detected in the case with the apex cone of 70 deg. The small change of water surface is detected with the apex cone of 40 deg. Then from the photographs shown in Fig. 4, high particle collection efficiency might be expected in the case with the apex cone of 70 deg. 5 The effect of apex cone angle on particle separation performance was examined and the experimental results are shown in Fig. 5. Figure 5-a and 5-b show the experimental results of particle separation performance for the main flow inlet velocity of 11 and 16.3 m/s, respectively. From the data of Fig. 5-a, the 50% cut size decreases as the apex cone angle increases. For the three cases, the minimum value of the 50% cut size was obtained for the apex cone angle of 70 deg. On the other hand, the 50% cut size is nearly equal under the high inlet velocity conditions as shown in Fig. 5-b. In this case, the apex cone angle of 70 deg. is better than the other cases, but the difference between each case is small. In order to find out the reasons why the experimental data shown in Fig. 5 was obtained, the numerical simulation and flow visualization method using soap foam around the apex cone region were carried out. 3 Numerical Simulation In order to compare particle separation performance of various apex cone angles, the CFD simulation would be effective to find out the difference of physical phenomena around the apex cone region. The three-dimensional Navier-Stokes equations and equations of particle motions were solved numerically. Table 1 summarizes the basic equations used in the simulation. The direct flow method, that solves Navier-Stokes Equations directly, was used to calculate the flow field in the cyclone ［8］. To calculate the non-linear convection terms, the artificial viscosity coefficient of 1/3 was selected to obtain a stable flow field ［9］. The control volume method was used in the simulation ［10］. Figure 6 shows the grid distributions for the various apex cone angles. In order to make clear the grid near wall region, the scales in horizontal and vertical directions are not the same. In order to calculate the partial separation efficiency for small particles, the radial grid spacing near the cylindrical, conical and the apex cone walls are designed to be as small as possible. The fluid boundary conditions on the wall, including the apex cone in the dust box, were set to 6 zero. Grid numbers in axial, radial and circumferential directions are 51, 44, and 53, respectively. The boundary-fitted curvilinear coordinate system was used in the simulation. In order to obtain reliable results in calculating small particle diameters, the radial grid spacing near the conical and cylindrical walls should be as small as possible. In the calculations, the equations of mono-sized particle motions were solved by the RKG method ［11］. It was assumed that particles are collected when they touch the wall surface and particle repulsion was not taken into consideration. The particle trajectories of mono-sized particles are calculated, then calculated result does not depend on feed particle size distribution. In the experiment, inlet dust concentration was very small, so the interaction between particles and fluid was ignored in the simulation. Details of the calculation method are reported in our previous paper ［12］. 4 Flow around the apex cone Figure 7 shows the fluid velocity vectors around the apex cone of the cone angle from 40 to 80 deg. For the apex cone angle of 40 deg., the upward velocity components are detected on the inclined surface of the apex cone. But for the apex cone angle of 70 deg., the critical line between upward flow and downward flow shown in the dotted line is detected and the downward flow velocity components are very small compared to the upward flow. The separation line between upward and downward flow is clearly found for the apex cone angle of 70 deg. The red circles in Fig.7 show the upward flow region. For the 70 deg. case, the upward flow region is observed in the upper half area, but the region is slightly increased for the 80 deg. case. Then it is considered that particles collected on the upper conical section might be easily moved to the central upward flow. In order to find out the critical line of the upward and downward flow on the inclined surface of the apex cone, visualization experiments were carried out by use of soap foam. Before running the experiments, all surfaces of the apex cone were 7 coated with white soap foam. Then inlet flow was introduced to the cyclone during 10 min., and the observation of the surface of the apex cone was carried out. Figure 8 shows the enlarged photograph near the apex cone of 70 deg. The velocity vectors calculated by CFD simulations are also included in the photograph. For the lower part of the apex cone, a small amount of soap foam remained, but the soap foam is not detected on the upper zone of the apex cone. From the flow vectors, the upper and lower parts of the apex cone correspond to the upward and downward flow regions, respectively. The critical line of the soap foam might be considered to be the separation line of the upward and downward flows. Figure 9 shows the flow visualization photographs of soap foam with various kinds of apex cones. Also included in each photograph are the flow vectors calculated by CFD simulations. The area of soap foam remaining on the apex cone becomes maximum for the cone angle of 70 deg. The region without the soap foam corresponds to the upward flow and the region with soap foam corresponds to the downward flow. Then it is considered that upward fluid and particles of downward flow direction can be clearly separated with the apex cone at an angle of 70 deg. The particles collected on the upper region of the apex cone might be moved to the upward flow and not be collected. The probability of these escaping particles will be a minimum for the case of the apex angle of 70 deg. The fluid flow rates into the dust box for various apex cone angles are important. The maximum flow rate entering the dust box is obtained for the apex cone angle of 60 deg. Figure 10 shows the relation between 50% cut size and apex cone angle. It is found that 50% cut size indicated the minimum value for the apex cone angle of 70 deg. The calculated results indicated by the dotted line also shows the minimum 50% cut size for the cone angle of 60～65 deg. These results of high collection efficiency for cone angle of 70 deg. agree with the photographs shown in Fig.9. The particle trajectories calculated by CFD simulation are shown in Fig. 11. The initial starting positions of particles are set to the same points and three types of the apex cones are 8 used in each case. For the cone angle of 70 deg., the 1.3μm diameter particles are collected on the lower part of the apex cone. The particle collection place shown in Fig. 11-b corresponds to the down-flow region of the apex cone surface shown in Fig. 7-d. On the other hand, all particles are not collected for the apex cone angles of 40 and 80 deg. Due to the strong upward flow near these apex cone surface, the probability of particle collection will be decreased. For the apex cone angle of 40 deg., the particles entering into the lower part of the dust box move out of the dust box. The purpose of the apex cone set in the dust box is to decrease the fluid flow component, but the simulation results show that the apex cone angle of 40 deg. is ineffective. Figure 12 shows the particle separation performance for the three kinds of apex cones. The minimum cut size is obtained for the apex cone angle of 70 deg. The 50% cut size decreases as the apex cone angle increases. The calculation results by use of CFD simulation is also shown and simulation results agree with the experimental results qualitatively. The difference between the experimental results and calculations might be the selection of the flow Reynolds number used in CFD simulation. To obtain the converged solution, the flow Reynolds number in CFD simulation selected was 3000. In the experiment, the flow Reynolds number was greater than 2 x 104 and the fluid viscosity effect might be overestimated in the CFD simulation. It is difficult to obtain the converged solution with flow Reynolds number greater than 3000. In Fig.12, the reason of difference between calculation and experiment might be flow Reynolds between them. When the solution of high Reynolds number calculation is obtained, the difference between experiment and calculation might be small. The results shown in Fig. 12, indicating high separation performance are obtained for the apex cone angle of 70 deg., agree with the data shown in Fig. 10. 5 Effect of apex cone angle under the high inlet flow condition 9 The previous section examines the particle separation performance under low inlet velocity condition. This section examines the separation performance under the condition of high inlet velocity. Figure 13 shows the relation between 50% cut size and the apex cone angle under cyclone inlet velocity of 16.3 m/s. It is found that 50% cut size does not depend on the apex cone angle. The numerical results shown in the dotted line also indicated the same results. In order to find out the reason obtained in Fig. 13, the flow visualization experiments by use of soap foam under high inlet velocity condition was also carried out. Figure 14 shows the experimental results taken from various apex cone angles. The general patterns are nearly equal to the data shown in Fig. 9. The area of soap foam is maximum for the cone angle of 70 deg. The region without the soap foam corresponds to the upward flow and the region with soap foam corresponds to the downward flow. Then it is considered that particles moving near this apex cone surface will be affected by the upward and downward flows. Figure 15 shows the results of particle trajectories obtained by use of CFD simulation in the high inlet velocity condition. Figures 15-a and 15-b correspond to the particle diameter of 1.3 and 1.0μm, respectively. For the 1.3μm particles, the particles are collected on the cylindrical wall region and are not affected by the flow difference near the apex cone. On the other hand, for the 1.0μm particles, the particles move to the lower zone of the cylindrical wall due to the high inlet flow velocity. But all particles are moving in places away from certain distance of the apex cone. Then, from the trajectory calculation shown in Figs. 15-a and b, the particles are not strongly affected by the flow difference near the apex cone. Under high velocity condition, the particles tend to move near the conical and cylindrical wall section due to the strong centrifugal effect. Then particles entering into the dust box tend to move away from the apex cone. When particles move near the apex cone region, separation performance will be affected by changes in the apex cone angle, but from the CFD simulation under the high inlet velocity condition, particles move away from certain distance of the apex cone. 10