Under low reducing agent rate and high pulverized-coal injection rate operations, coke powder and unburnt char are generated in the blast furnace. These powders flow through the packed bed inside the furnace entrained by the gas flow. The powders accumulation in packed bed will deteriorate gas and liquid permeability and creates an unstable situation inside the blast furnace. Thus, it is important to quantitatively evaluate the accumulation rate on a distribution channel of coke powders. In this study, the behavior of the powder particles passing through the orifice formed by three coarse particles that osculate one another and compound in an equilateral triangle was numerically analyzed using the DEM-CFD. The results revealed the effect of drag force on the powder motion passing through the orifice. Moreover, the effect of the position of the powder passing through the orifice to pressure drop fluctuation inside the triangular prism is also analyzed.

Every year, Japan steelmaking industries produce about 100 million tones pig iron which consumes 12% of Japan’s total energy consumption. Moreover, this steelmaking process also contributes about 15% of the total CO₂ emissions.¹⁾ On average, 70% of the primary energy input to the integrated steelworks is used for the ironmaking process, thus the design of a high-efficiency blast furnace with an aim for the reduction of CO₂ emission and energy consumption is in high demand.^2,3)

As a method for reducing CO₂ in blast furnaces, a low-reducing agent ratio operation is being carried out to reduce the amount of coke used and reduce the carbon input to the blast furnace. At the same time, in order to reduce the cost of blast furnace operation, a large amount of pulverized coal is injected into the furnace using inexpensive pulverized coal as an auxiliary reducing agent. However, these operating methods increase the amount of coke powder and unburned char generated in the blast furnace. The generated powder flows through the packed bed in the blast furnace as a solid-gas two-phase flow, which affects the ventilation and liquid permeability in the blast furnace. It is important to quantitatively evaluate the distribution route of the generated powder in the blast furnace and the accumulation rate on that route because excessive accumulation of powder in the packed bed may cause operational malfunctions such as hanging and slips.

As the method for blockage prevention is very important in blast furnace operation, some studies regarding the inspections, experimental approaches, and numerical approaches already been done in the past years. For inspection approaches, Puttinger et al.^3,4,5) discussed the detection method for blockages inside the blast furnace and its relation to blast flow rate. Their works were divided into three series; classification of blockage events, signal processing of hot blast pressure data, and visual detection based on tuyere camera images. Takahashi et al.,^6,7) and Dong et al.,⁸⁾ conducted some experiments on macroscopic models, to obtain the blockage behavior of fine particles inside the packed bed in basic level experiments. Takahashi et al. then proposed a critical ratio of fine to coarse diameter greater than 0.12 for the blockage to occur. Various numerical approaches also had been done to predict the behavior of ore particles inside the blast furnace using DEM or DEM-CFD coupling analyses. Fukuda et al.⁹⁾ and Luo et al.,¹⁰⁾ used Discrete Element Method (DEM) to analyze the powder behavior in static flow inside the equilateral triangle system where numbers of fine particles are dropped simultaneously towards the orifice formed by the arrangement of three equilateral coarse particles. Zaki et al.,¹¹⁾ predicted granular mass discharge rate through different orifice shapes of the same opening area at the flat bottomed cylindrical silo. Ariyama et al.¹²⁾ and Zhou et al.¹³⁾ were discussed the development model of the fine particle inside the blast furnace using the DEM model, which contains the key phenomena including percolation velocity, residence time distribution, longitudinal and transverse dispersion, of cohesive fine particles in a packed bed. However, for a system like a blast furnace where the gas flow is strongly affected the fine particle arrangements and porosity distribution, the coupling analysis between DEM and Computational Fluid Dynamic (CFD) for gas flow is needed. Natsui et al.^14,15) Kurosawa et al.,¹⁶⁾ Matsuhashi et al.,¹⁷⁾ Kawai et al.,¹⁸⁾ Mahmoodi et al.,¹⁹⁾ Ishii et al.,²⁰⁾ Kikuchi et al.,²¹⁾ Narita et al.,²²⁾ and Lichtenegger et al.,²³⁾ were summarized the coupling analyses of DEM-CFD inside the fluidized bed in macroscopic level. Ueda et al.,²⁴⁾ proposed DEM-CFD coupling analysis on the relation between melting behavior of iron ore, to the shape of the cohesive zone and gas permeability. Numerical approach by Bambauer et al.,²⁵⁾ shows a strong correlation of porosity created by the mixing between the coke and ore to gas permeability. Miao et al.,²⁶⁾ observed the effect of raceway orientations, tuyere length, and solid extraction on solid flow patterns and flow and force structure of particles inside the blast furnace. Using DEM-CFD coupling analysis, Dong et al.²⁷⁾ was studied the shaft injected gas penetration behavior will change due to its collision with the upward gas flow.

It can be noticed that the previous studies of numerical analyses mainly focused on DEM-CFD coupling analysis in the macroscopic level.^{14,15,16,17,18,19,20,21,22,23,24,25,26,27)} In all these cases, the gas flow patterns inside the packed bed are the overall or proximate amount of gas flow. Meanwhile, in the microscopic level the analyses were mostly only focused on DEM approach,^{9,10,11,12,13)} such as the interactions of fine to coarse particle through their morphology characteristics and cohesive behavior. On the contrary, the powders are driven by the interstitial gas flow within the void space among packed particles. This interstitial gas flow pattern is largely affected by the void shape that changes complexly and continuously. Such a microscopic gas flow pattern is unable to be reflected in the macroscopic level analysis. Thus, it is also become important to investigate in detail by using the DEM-CFD model at the microscopic level.

In this research, microscopic behavior analysis has been performed to show the passing behavior of fine particles through the narrowest possible gap inside the packed bed or we call as an orifice, with the microscopic gas flow. Simple experiment also being conducted as a validation for the numerical model on fine particle passing time through the orifice. The interaction between gas flow and fine particles also been observed, where the fine particle affecting the characteristic of gas flow such as pressure drop fluctuations and gas flow patterns, while the existence of gas flow also affecting the passing behavior of the fine particle. Furthermore, two-phase flow gas-solid behavior leading to the initial clogging process at the microscopic level will also be observed using multiple fine particle dropping in the triangular prism model of the packed bed.

The moving particles are tracked using the discrete element method (DEM), in which each element is numerically analyzed on the condition where each particle is adequate the equation of motion and the force transfer of Newton’s third law of action and reaction. Powder particles flowing as a group consisting of a large number of particles in a moving bed or a fluidized bed move while repeating collision and friction with neighboring particles and walls. The equations of motion for the translational and rotational motions of a particle receiving a force from its surroundings are described as follows, based on Newton’s second law of motion given by the following Eqs. (1) and (2).   

Where m is particle mass [kg], v is particle velocity [m s⁻¹], t is time [s], g is gravity [m s⁻²], and I is the moment of inertia of the particle [kg m²], ω is the particle angular velocity [rad s⁻¹], and r is the particle radius [m]. The motion of the particle of interest is affected by force and gravity due to contact with surrounding objects. F_n, F_s, F_D, and M_r are inter-particle contact forces in the normal direction [N], contact forces in the tangential direction [N], drag force [N], and rotational resistivity [N m]. Subscripts i and j each determine the particle of interest and the particles that come into contact with this particle, respectively.

The normal direction and tangential direction are defined as the direction of the line that passes through the center of two interconnected particles, and the direction that is perpendicular to this line. Contact forces in the normal and tangential directions can be given by the following Eqs. (3) and (4). u and φ are the small displacements ([m] and [rad]) in normal and tangential direction because of the contact. K is spring constant [N m⁻¹], and η is damping constant [kg s⁻¹]. Under the slip condition, the tangential contact force is given by the following Eq. (5). The rotational motion resistance force is given by following Eq. (6) and incorporated in Eq. (2), where α and r_c are rotational friction coefficient [–] and radius of contact circle [m].   

In this study, gas-powder two-phase flow analysis was performed using thermal fluid and powder analysis software R-FLOW. The governing equations for the gas flow are the momentum conservation law (Navier-Stokes equation), Eq. (7) and the mass conservation law (continuity equation), Eq. (8).   

Here, the flow velocity is V [m s⁻¹], density is ρ [kg m⁻³], and F_p is force that applied to particles other than fluid [N]. The suffixes c and d are physical quantities related to the gas flow as continuous phase and the powder particles as dispersed phase, respectively. ε_c and (ε_d = 1 −ε_c) are the volume fractions of the continuous and disperse phases, and β is the coefficient of momentum exchange between the continuous and the disperse phases.

The Gidaspow drag model²⁹⁾ is adopted as the gas-powder interaction force in the two-phase flow analysis. The Gidaspow model is the combination between two drag model, which consist of the Ergun model²⁸⁾ which is use for the low void fraction region (ε_c≦0.8), and Wen-Yu model for high void fraction region (ε_c > 0.8),³⁰⁾ as shown in Eqs. (9) and (10). Here d_p is particle diameter [m]. The drag coefficient C_D is given as the function of Reynolds number as written in Eq. (11).   

In this simulation, the narrowest part of the flow path in the spherical packed bed is being discussed. It is obvious that the arrangement of the actual packed bed inside the blast furnace is more likely irregularly shaped and randomly packed, which sets complex flow paths among the packed material. However, it is considered that the possible narrowest gap in the packed bed is the opening of three coarse particles that osculate with one another and compound in an equilateral triangle. In this paper, the gap will be called as an orifice hereinafter. As mentioned, the packed particles are placed horizontally and form a small aperture or orifice in the middle by nudging each other. The powder particles are dropped onto the orifice by gravity, and the trajectories of the particles are numerically tracked. The calculation domain is set as a vertical triangular prism, of which the cross-section is an equilateral triangle comprised of orifice particles in the center, as shown in Fig. 1. This study is focused on the passing behavior of powder particles onto the orifice and its interactions with gas flow blown from the bottom of the packed bed.

Fig. 1.

Analysis model. (Online version in color.)

In this study, the packed powder that creates the orifice and the sides of the prism is treated as a solid wall. As the scale of the model, the packed particle diameter is 25 mm, the vertical triangular prism height is 70 mm, and the fine particle diameter is 2.5 mm. In this simulation, prism walls are treated as a non-slip wall. The analysis conditions are shown in Table 1. The input particles were set to 1 or 30 particles and the gas inflow velocity is 1.0 m s⁻¹. For this analysis, the friction coefficient is 0.2 [–] and the spring constant is 100 [N m⁻¹]. Before discussing the gas-solid flow characteristic inside the triangular prism, initial investigation using experimental and numerical approaches will be performed in the next section to determine the basic value of the coefficient of restitution and comparing the passing time of fine particle using numerical and experimental approach.

The validation process for numerical analysis is taken in two steps. The first step is to measure the passing time of fine particle through the orifice, continuing with the second step to determine the coefficient restitution using experimental apparatus to be applied in numerical analysis. The first step of validation is performed as follows. A schematic diagram of the experimental apparatus is shown in Fig. 2(a). The measuring device consists of a positioning device for setting the position of dropped fine particles into a triangular flow path. Consisting of three beads particles that osculate with one and another and compound in an equilateral triangle with an orifice in the middle, an optical sensor that detects the inflow and outflow of particles, and data logger that records the signal from the sensor. The transit time is defined by the time difference between the signals of the upper and lower sensors. The horizontal cross section of the triangular prism was set so that the inside is divided to 136 points as shown in Fig. 2(b), and the center of the particle was at the vertex of the triangle. From these 136 points, 6 points are selected and then named as point A (4.167; 4.333), B (5.83; 1.44), C (19.167; 1.44), D (20.835; 4.33), E (14.167; 15.877), and F (10.835; 15.877), where the coordinate of the nearest vertex to point A is (0.0; 0.0), (unit: mm), shown in Fig. 2(c). The distance between the laser sheets from the optical sensors is 51 mm and the length of the channel is 50 mm, as shown in Fig. 2(d). Three measurements are made per one point, and the average value of transit time is taken from every dropped position. The passing time from this experimental approach will be shown in Table 3.

Fig. 2.

(a) Experimental apparatus for passing time (b) orifice position (c) particle dropped positions (d) schematic for calculate passing time. (Online version in color.)

The second step is performed to determine coefficient restitution to be used in numerical analysis. This second step is performed as follows. A schematic diagram of the experimental apparatus is shown in Fig. 3(a). A fine glass spherical particle of which the diameter is 2.5 mm was dropped from above of which the height is 60 mm, onto the coarse spherical particle with the diameter of 25 mm. The particle trajectory was recorded by a high speed camera, and the velocity of the glass spheres before and after the collision was determined using image analysis software. Table 2 shows the experimental conditions.

Fig. 3.

(a) Experimental apparatus design (b) Coefficient of restitution results. (Online version in color.)

The coefficient of restitution was calculated from the ratio of velocity of the glass sphere after and before the collision. Figure 3(b) shows the coefficient of restitution obtained by the experiment, where the average value is 0.92 [–].

This obtained coefficient restitution is used for numerical analysis, mimicking the experimental model as mentioned in step one, Fig. 2. In numerical analysis, the glass bead is dropped from 6 coordinate positions; A, B, C, D, E, and F, similar to the coordinates of the experimental model. The comparison between the numerical analysis and the experimental result for the coefficient restitution of 0.92 [–] are shown in Table 3. For the simulation results, even though the transit times of positions B and E are larger than the other input positions, the average transit time of the numerical analysis with a coefficient of restitution of 0.92 [–] is 0.41 s, which is close to the experimental results. The difference in result may occur because some errors during experiments such as the precision of dropping location and shape distortion of both fine and coarse particles. However, because the average value between the numerical and experimental results are close, the numerical approach is reliable to be used for the next model. For the next analysis, the fine particle dropping position is set at point F, where the experimental and numerical results are the closest. The coefficient of restitution is set to 0.95 [–] with the aim to extend the retention time of fine particle above the orifice, while the friction coefficient and spring constant remain the same as mentioned in Table 1.

The analysis result of the fluid for single phase flow through the orifice of coarse particles will be discussed in this section. Figures 4 and 5 show the variation of flow fields after the gas is introduced from the bottom of the triangular prism. The side and the height of the triangular prism are 25 mm and 70 mm, the particle diameter is 25 mm, and the coarse particle center locates at the height of 35 mm. Updraft gas is introduced uniformly from the bottom at 1 m s⁻¹. The plane shows in these figures are located at 8.66 mm from the side face. Figure 4 shows the variation of the velocity field. Although the air is uniformly introduced from the bottom, the gas flow concentrates at the orifice aperture at t = 0.04 [s]. With this concentration of flow, a gas plume is formed and its velocity increases rapidly in this region and the downstream region. The maximum velocity reaches about 13 m s⁻¹, where the sudden increase of upward gas velocity is caused mainly because the area ratio between the aperture to the prism area is about 0.093 [–] or approximately 1:12. Figure 5 shows the variations of the pressure field. The pressure field rapidly decreases during the process of passing through the orifice, inverse to the velocity profile. The velocity profile and pressure distribution inside the system are changed immediately after the fluid is introduced at short intervals but reach a steady state in about 0.2 s under no fine particle condition.

Fig. 4.

Time variation of flow velocity.

Fig. 5.

Time variation of pressure distribution.

Figure 6 shows the trajectories of a single particle inside the prism through the coarse particle orifice, with and without upward gas flow. The particle drop position is located at point F. On the trajectories, the cross marks are plotted every 0.002 s. In both cases, the fine particle initially falls onto the right-front coarse particle. It bounces and collides with the prism walls and coarse particles several times then it comes into the region just above the aperture opening. In this region, the fine particle collides many times with coarse particles and finally passes the aperture. In these figures, remarkable differences can be seen in two regions. One is above the aperture and the other is beneath the coarse particles. In the former, horizontal motion is enhanced under the upward gas flow condition. In the latter, the fine particle trajectory is almost straight downward under no gas flow condition compared to the under upward gas flow condition.

Fig. 6.

Particle trajectory (a) stationary fluid. (b) gas flow of 1 m s⁻¹. (Online version in color.)

Figure 7 compares the variations of the velocity and the height of powder particles with and without upward gas flow. In these figures, the red and black lines represent the height position and velocity of fine particles, respectively. In static fluid, the velocity of fine particles shows linear jagged changes. This profile indicates that the downward acceleration and sudden velocity change due to the collision occur alternately. The downward acceleration is mainly affected by gravitational acceleration with less drag resistance from static gas. Contrarily, the effect of the drag force is significant under upward gas flow conditions. In Fig. 6(b), the velocity profile of dropping fine particles is curved around t =0.09, 0.15, 0.3, and 0.40 s. especially around 0.1 and 0.4 s, the upward acceleration occurs. During the time when the particle stays in the aperture region, the chance of contact between fine and coarse particles increases. Under the upward gas flow conditions, numerous fine-coarse particle collisions occur from 0.40 and 0.62 s. It is considered due to the strong upward gas flow in the aperture region because the fine particle velocity fluctuates around 0 m s⁻¹. As a result, the residence times of the fine particles in the orifice are 0.59 s for static fluid and become longer for upward flowing conditions, about 0.67 s.

Fig. 7.

Velocity and height of powder in (a) static fluid (b) gas flow 1 m s⁻¹.

The cumulative contact force is shown in Fig. 8. In the case of static fluid, the cumulative contact force is lower compared to the case of upward gas flow of 1.0 m s⁻¹. These two results indicate that more contact occurred between fine to coarse particles in the case with upward gas flow compared to the static fluid case. For both cases, we can see that cumulative contact force increases, wherein the case of static flow, it is noticeable that the accumulated contact force occurs in the stepwise configuration. In the case of upward gas flow, the change of cumulative contact force to time passage occurs in much smaller stepwise configuration passed between t = 0.42 to t = 0.6 [s], compared to the prior case. In the latter case, this phenomenon occurred when the powder entered the area near the orifice which was strongly affected by the high-speed airflow concentrated at the aperture area and rapidly bounced back and forward amongst the coarse particles before finally entering the orifice.

Fig. 8.

Cumulative contact force (a) static fluid (b) gas flow 1 m s⁻¹.

Figure 9 shows the variations of pressure drops inside the prism in upward flow with time. Figure 9(a) is the single phase flow without dropping fine particles and Fig. 9(b) is the two phase flow case with single fine particle dropping. The pressure drop in this figure was defined by the difference between average pressures in the cross-sectional areas at the bottom and the top of triangular prisms. Under the single phase flow, the pressure drop is almost constant at 84.4 Pa, and it can be treated as a steady system. In the case of two-phase flow, the pressure drop starts to fluctuate from t = 0.09 [s] and continues till t = 0.65 [s]. Fairly large pressure fluctuation can be divided into 4 periods, occurring around 0.09, 0.32, 0.40, and from 0.50 to 0.62 s. In the last period, the fine particle stays in the orifice region and blocks the major area of the orifice aperture, and generates a large pressure drop. The fine particle continues to move as indicated in Fig. 6(b) and cause large fluctuations, where during this period fine particles going back and forth among coarse particles before finally passing through the orifice. These fluctuations mainly occur because the orifice area becomes narrower during fine penetration which causes a temporary increase in fluid resistance and energy loss.

Fig. 9.

Pressure loss in triangular prism (a) single phase flow (b) two phase Flow.

In the first three periods, judging from fine particle trajectories, the pressure fluctuations occurred when the fine particle went across the upward gas plume formed above the aperture as shown in Fig. 4. Thus the duration of the fluctuations are shorter compared to the last period. After the initial collision, at t = 0.07 [s], the particle passes the gas plume at 0.09 s, and after that collides with coarse particles and passes again at t = 0.32 [s] and t = 0.4 [s]. This also can be observed from the moment where the velocity profile curves, as shown in Fig. 7(b), at the same time when the pressure inside the triangular prism fluctuated at t = 0.09, 0.32 and 0.40 s.

After 0.65 s, fluctuation continued for a while after the particle passed the orifice and positioned at the area just below the orifice before finally falling into the bottom of the triangular prism. Judging from this fact, it can conclude that the collision between fine and coarse particles itself does not affect the fluctuations of pressure drop, but after the collision the fine particle sometimes will pass the gas plume and block the aperture area of the orifice, causing the pressure inside the system to fluctuate.

To demonstrate the passing behavior of multiple fine particles, 30 fine particles are dropped on to the orifice at the same time. The geometry of the system is the same as the previous section, but 20 mm of additional height is added to the top of the upper section of the prism above the orifice. The fine particles are dispatched pseudo-randomly. The motions of 30 particles are tracked under the static fluid and upward gas flow (1.0 m s⁻¹) conditions.

Figure 10 shows the visualization of 30 fine particles dropped in an equilateral triangle system with an upward gas flow of 1 m s⁻¹. From the visualizations, it can be observed that fine particles do not immediately pass through the orifice, but first form an arch-like structure as shown in Fig. 10(h) around the orifice aperture. The arch-like structure is a blocked configuration of fine particles that formed the concave surface of the clogged dome. This arch-like structure occurred around t = 0.2 [s] to t = 0.3 [s] as depicted in Figs. 10(b) and 10(c), and will gradually break down because of the unstable condition between fine to fine and fine to coarse particles due to the existence of the upward gas flow. The upward gas flow causes the particles to budge all the time, which results in loose friction between particles and makes it difficult to maintain the arch-like structure. Finally, the arch-like structure will completely collapse as shown in Fig. 10(d), where fine particles gradually pass through the orifice. From the observation of this phenomenon, we can deduce that the presence of the upward gas flow will obstruct the emergence of blockages because of the arch-like structure become unstable due to loose friction between the particles. Another reason that complete blockage does not occur is because the diameter ratio between fine to coarse particles is less than the critical value of the diameter ratio proposed by Takahashi et al.⁷⁾ of 0.12 [–], where in this current study the diameter ratio is 0.1 [–].

Fig. 10.

Powders positions inside the bed at (a) initial condition (b) 0.2 s (c) 0.3 s (d) 0.4 s (e) 0.5 s (f) 0.740 s (g) 0.940 s (h) arch like structure. (Online version in color.)

Figure 11 shows the pressure drop inside the vertical triangular prism of multiple fine particles. Compared to the previous case with a single occurrence of dropping particles, pressure fluctuations already occurred since the beginning of the dropping process. In the case of single particles dropping, the possibilities of fine particles to collide and bounce to the area above the aperture is less compared to the case of multiple fine particles. This is because fine particles already preclude above the aperture since the beginning of the dropping process. In this case, fine particles congregated and collided back and forth and resulted in fluctuating changes in the area of the aperture above the orifice and caused the pressure drop to oscillate since the beginning of the dropping process. In the case of multiple fine particles, the average pressure drop also becomes higher, mainly because of the additional height above the orifice, where the vacuum area occurred in the upper section of the vertical triangular prism.

Fig. 11.

Pressure loss in triangular prism of 30 fine particles dropped. (Online version in color.)

Figures 12 and 13 show the number of particles in the stationary fluid, and in the upward gas flow, respectively. In this simulation, particle count exists in the range between 80 mm at the highest and 10 mm at the lowest. From these two figures, it can be observed that the passing time for fine particles in static flow is shorter compared to the case of updraft gas flow. In both cases, a sudden increase of contact force occurred at t = 0.1 [s], where at this time, the initial collision between coarse particles and fine particles occurred. After the initial collision, the contact force is suddenly increased again in the second collision at t = 0.15 [s], where fine particles bounced after the initial collision and collided with the coarse particles again. Finally, following the second collision, the contact force smoothly decreased without significant fluctuation. Gradual decreasing of contact force in the case of stationary fluid also can be clarified with the number of particles inside the triangular prism, where the number of particles are decreased smoothly with time changes. This means it is relatively easy for fine particles to pass through the orifice, resulting in less collision among particles as the time goes. In the case with updraft gas flow, after the sudden increase of contact force at the initial collision similar to the previous case, the contact force also decreases but in more fluctuated ways.

Fig. 12.

Number of particles and contact force in time passage for static flow.

Fig. 13.

Number of particles and contact force in time passage for upward gas flow.

Figures 14(a) and 14(b) shows the pressure distribution and the velocity profile inside of triangular prism at t = 0.1 [s] and gas velocity of 1 m s⁻¹ with the case of multiple fine particles, respectively. From the pressure distribution depicted in Fig. 14(a) it can be observed some local pressure drop or low pressure spotted above the orifice area because the existence of fine particle straitens the aperture of the orifice area. Furthermore, for the velocity profile depicted in Fig. 14(b), fluctuation in gas plume can also be observed, mainly because of the presence of fine particles narrowing the aperture of the orifice area. The effect of the drag force is significant under upward gas flow conditions to make it difficult for fine particles to pass through the orifice. The existence of updraft gas, makes the fine particles floating and restrained in the area above the orifice aperture and makes the fine particles unstable. This resulted in more collisions occurring before passing the orifice, compared to the case of stationary fluid. These statements can be clarified by observing the number of particles inside the triangular prism that decreased in a stepwise manner which resulted in a longer time passage for fine particles to pass through the orifice.

Fig. 14.

(a) Pressure distribution and (b) velocity profile of multiple fine particles at t = 0.1 [s].

This study performed some analysis of powder motion passing through the orifice consisting of the coarse particles with the actual microscopic gas flow. For single and multiple particle dropping, it is discovered that the residence time of fine particles above the aperture is longer when the upward gas flow is being injected. This longer residence time occurred because the upward gas drag force obstructed the fine particle path to pass through the orifice. In the case with singe particle dropping, the existence of fine particles did not necessarily influence pressure drop fluctuation. Pressure fluctuations inside the packed bed are largely affected when the particles stay in the main channel of the gas flow located in the aperture area above the orifice. However, when the fine particle is outside the aperture area, the pressure drop fluctuations are not significant. Pressure drop fluctuations also occurred when fine particles passing the orifice region and block the major area of the orifice aperture and caused the area to become narrower. These phenomena caused a temporary increase in fluid resistance and energy loss. Remarkable fluctuations of the pressure drop occurred in the case of multiple particle dropping, since the beginning of the dropping process of fine particles. These phenomena occurred because the possibilities of fine particle bounce and obstructing the aperture area above the orifice are higher due to the existence of a larger amount of particles.

A part of this work was performed under the Cooperative Research Program of “Network Joint Research Center for Materials and Devices.”

• 1)  Ministry of the Environment, Japan: National Greenhouse Gas Inventory Report of Japan, National Institute for Environmental Studies, Ministry of the Environment, Japan Ibaraki, (2013), 2. 4, 3. 2.
• 2)   T.  Ariyama and  M.  Sato: ISIJ Int., 46 (2006), 1736.
• 3)   S.  Puttinger and  H.  Stocker: ISIJ Int., 59 (2019), 466.
• 4)   S.  Puttinger and  H.  Stocker: ISIJ Int., 59 (2019), 474.
• 5)   S.  Puttinger and  H.  Stocker: ISIJ Int., 59 (2019), 481.
• 6)   H.  Takahashi,  H.  Kawai,  M.  Kobayashi and  T.  Fukui: Tetsu-to-Hagané, 92 (2006), 996 (in Japanese).
• 7)   H.  Takahashi,  H.  Kawai,  M.  Kobayashi and  T.  Fukui: ISIJ Int., 45 (2005), 1386.
• 8)   X. F.  Dong,  D.  Pinson,  S. J.  Zhang,  A. B.  Yu and  P.  Zulli: Powder Technol., 149 (2004), 1.
• 9)   M.  Fukuda,  J.  Suzuki,  H.  Kawai and  H.  Nogami: ISIJ Int., 55 (2015), 1291.
• 10)   Z. G.  Luo,  H.  Zhou,  T.  Zhang,  Y.  You,  L. J.  Zhang,  Z. S.  Zou and  Y. S.  Shen: Powder Technol., 314 (2017), 102.
• 11)   M.  Zaki and  M. S.  Siraj: Powder Technol., 354 (2019), 641.
• 12)   T.  Ariyama,  S.  Natsui,  T.  Kon,  S.  Ueda,  S.  Kikuchi and  H.  Nogami: ISIJ Int., 54 (2014), 1457.
• 13)   H.  Zhou,  S.  Wu,  M.  Kou,  Z.  Luo,  Z.  Zou and  Y.  Shen: ISIJ Int., 58 (2018), 43.
• 14)   S.  Natsui,  S.  Ueda,  H.  Nogami,  J.  Kano,  R.  Inoue and  T.  Ariyama: ISIJ Int., 51 (2011), 51.
• 15)   S.  Natsui,  T.  Kikuchi and  R. O.  Suzuki: Metall. Mater. Trans. B, 45 (2014), 2395.
• 16)   H.  Kurosawa,  S.  Matsuhashi,  S.  Natsui,  T.  Kon,  S.  Ueda,  R.  Inoue and  T.  Ariyama: ISIJ Int., 52 (2012), 1010.
• 17)   S.  Matsuhashi,  H.  Kurosawa,  S.  Natsui,  T.  Kon,  S.  Ueda,  R.  Inoue and  T.  Ariyama: ISIJ Int., 52 (2012), 1990.
• 18)   H.  Kawai and  H.  Takahashi: ISIJ Int., 44 (2004), 1140.
• 19)   B.  Mahmoodi,  S. H.  Hosseini and  G.  Ahmadi: Particuology, 43 (2019), 171.
• 20)   J.  Ishii,  R.  Murai,  I.  Sumi,  Y.  Yongxiang and  R.  Boom: ISIJ Int., 57 (2017), 1531.
• 21)   S.  Kikuchi,  T.  Kon,  S.  Ueda,  S.  Natsui,  R.  Inoue and  T.  Ariyama: ISIJ Int., 55 (2015), 1313.
• 22)   Y.  Narita,  H.  Mio,  T.  Orimoto and  S.  Nomura: ISIJ Int., 57 (2017), 429.
• 23)   T.  Lichtenegger and  S.  Pirker: Powder Technol., 325 (2018), 698.
• 24)   S.  Ueda,  T.  Kon,  H.  Kurosawa,  S.  Natsui,  T.  Ariyama and  H.  Nogami: ISIJ Int., 55 (2015), 1232.
• 25)   F.  Bambauer,  S.  Wirtz,  V.  Scherer and  H.  Bartusch: Powder Technol., 334 (2018), 53.
• 26)   Z.  Miao,  Z.  Zhou,  A. B.  Yu and  Y.  Shen: Powder Technol., 314 (2017), 542.
• 27)   Z.  Dong,  Q.  Xue,  H.  Zuo,  X.  She,  J.  Li and  J.  Wang: ISIJ Int., 56 (2016), 1588.
• 28)   S.  Ergun: Chem. Eng. Prog., 48 (1952), 89.
• 29)   D.  Gidaspow: Multiphase Flow and Fluidization, Continuum and Kinetic Theory Descriptions, Academic Press, New York, (1994), 1.
• 30)   C. Y.  Wen and  Y. H.  Yu: Chem. Eng. Prog., Symp. Ser., 62 (1966), 100.