MATERIALS AND METHODS

Cold model experiments in a Small Bubbling Fluidized Bed (SBFB) were conducted using compressed air.

Figure 1 and 2 show the cold model experimental setup. The SBFB consists of a nozzle distributor plate and was supported on a stand. The main bed was made from transparent perspex. The overall height of the SBFB is 877.5 mm. The main bed has an inner diameter of 74 and 600 mm height. The nozzle distributor plate has five small nozzles that protrude up from the surface of the plate. Each nozzle has four holes of 1.6 mm diameter to allow gas to flow. The distributor plate was made from stainless steel. Figure 3 shows the 3-D diagram of the nozzle distributor plate.

The SBFB is connected to a 2.5 hp air compressor model HD47C to provide a compressed air flow and a bag filter to capture any of the bed material leaving the reactor.

Fig. 1:	Schematic diagram of the experimental setup

Fig. 2:	Photograph of the cold model experimental setup

A simple U-tube manometer was used to determine the pressure drop of the material. Figure 4 shows the 2-D design of the small bubbling fluidized bed.

The 1000 μm of CaO mixed with 350 μm sand was used as the bed material. These particles sizes were sieved using an Endecott’s multi-layer test sieve shaker. The 350 μm sand particles size has been selected in this experimental because according to Geldart (1973) and Yates (1983), the particles size of sand in the range of 150-500 μm (the Group B powders) gives well fluidization. In the experiment, for each run, 240 g constant weight of CaO was used and mixed with 226, 421, 496 and 808 g of sand to obtain the CaO-sand mixture percentages. The experiment started when the mixtures of 240 and 226 g were put into the SBFB. Then the air compressor was set to 2 bar using pressure regulator and the air flow in term of the volume flow rate of 5 L min^-1 was passed to the SBFB controlled by a rotameter. During the experiment the fluidization behaviors in term of bed expansion height and the pressure drop inside the SBFB were recorded. Some of the fine particles found in the bag filter was weighed and recorded at the end of the experiment.

Fig. 3:	3-D diagram of the nozzle distributor plate

Fig. 4:	The design of the SBFB. All dimensions in mm

The experiment was repeated at different CaO mixture percentages (70, 60, 50, 40%), air volume flow rate (5, 15, 25, 35, 45, 55 L min^-1) and air intake pressure (2, 3, 4, 5, 6 bar).

RESULT AND DISCUSSION

Table 1 shows the expected minimum fluidization velocity (u_mf) obtained for every percentage mixture of 1000 μm CaO and air pressure. All the expected minimum fluidization velocity was calculated based on the Ergun Equation (Wen, 2003; McCabe et al., 2004) using:

(2)

where, μ_g is dynamic viscosity of the carrier gas (kg m^-1 sec), ρ_g is density of the carrier gas (kg m^-3), d_s is solid particles size (m) and Ar_numb is Archimedes number.

The u_mf is needed to separate the particles of CaO and sand from each other and float in the SBFB. As shown in Fig. 5, under the constant air pressure of 2 bars, the bed pressure drops for different CaO-sand percentages were not similar.

Table 1:	Expected minimum fluidization velocity

Fig. 5:	Effect of 1000 μm CaCO3 mixture percentages, air volume flow rate and air pressure on the pressure drop (2 bars)

Fig. 6:	Effect of 1000 μm CaCO3 mixture percentages, air volume flow rate and air pressure on the pressure drop (3 bars)

Fig. 7:	Effect of 1000 μm CaCO3 mixture percentages, air volume flow rate and air pressure on the pressure drop (4 bars)

However the constant bed pressure drop was obtained when the u_mf reached for every CaO-sand percentages (between 0.05-0.08 m sec^-1). In Fig. 6, when the u_mf reached 0.04-0.06 m sec^-1, the constant bed pressure drop was obtained for all percentage mixture of 1000 μm CaO under 3 bars constant air pressure. The similar trends can also be seen in Fig. 7-9. When the u_mf reached 0.05-0.06 m sec^-1, the constant bed pressure drop was obtained for all constant air pressure 4, 5 and 6 bars. The comparison between the expected minimum fluidization velocity (Table 1) using Ergun Equation (Wen, 2003; McCabe et al., 2004) and the experimentally (Fig. 5-9) were approximately same.

When the superficial fluid velocity was gradually increased, the bed pressure drop gradually increased for all percentage mixture of 1000 μm CaO and air pressure. Then the bed pressure drop levels off and no longer increases as the superficial velocity is increased because it has reached the u_mf. This due to the upward force exerted by the fluid on the particles was sufficient to balance the net weight of the bed and the particles begin to separate from each other and float in the fluid. As the velocity is increased further, the bed continues to expand in height, but the bed pressure drop stays constant.

Fig. 8:	Effect of 1000 μm CaCO3 mixture percentages, air volume flow rate and air pressure on the pressure drop (5 bars)

Fig. 9:	Effect of 1000 μm CaCO3 mixture percentages, air volume flow rate and air pressure on the pressure drop (6 bars)

The similar trends on the bed pressure drops constant level in the bubbling fluidization bed also can be found from experimental results reported by Flamant et al. (1991), Svoboda and Hartman (1981), Guo et al. (2003) and Yates (1983).

Based on the similar experimental results of the minimum fluidization velocity and the constant bed pressure drop were obtained compared to other researchers, the following variables affecting the fluidization height were investigated:

Effect of the CaO percentages: The amount of CaO percentage in the mixture has an effect on the bed pressure drop and bed expansion height under constant pressure. With the decrease of the CaO percentages from 70, 60, 50 to 40%, the bed pressure drop ΔP in the SBFB increased under constant pressure of 2 bars as shown in Fig. 5. When the air pressure increased to 3 bars (Fig. 6), the similar trend was obtained where the bed pressure drop ΔP increased when the CaO-sand mixtures percentages decreased.

Fig. 10:	Bed expansion height results (2 bars)

These happened because of the increasing in the amount of mass of the solid particles (1000 μm CaO-350 μm sand) in every mixture. These similar trends can also be found in the Fig. 7-9.

In Fig. 10, its can be observed that with the decreased of the CaO percentages by 70, 60, 50 to 40, the bed expansion height increased between 0-14.5 cm when the superficial velocity increased from 0.02-0.215 m sec^-1 under 2 bars constant air pressure. This can be explained by the adhesive force between CaO particles such as the Van der Waals force, the electronic force and the liquid bridge existing in the powder that has been reduced because of the increased of the 350 μm sand particles in the mixture. Previous studies reported by Geldart (1973) and Yates (1983) showed that the particle size of sand in the range of 150-500 μm (Group B powders) with low density is in the range of 1500-4000 kg m^-3 has a better fluidization. Similar trends can also be found from the experimental results of Gao et al. (2009). However at early stages of the experiments for 70 and 60% CaO mixture, the bed expansion height is 0 cm, which means that these is no fluidization. This is because the superficial flow velocity has not reach the minimum superficial flow velocity where the upward force exerted by the fluid on the particles is not sufficient to force the weight of CaO-sand mixture of the bed. The adhesive force between CaO-sand mixture particles is still strong and the amount of 350 μm of sand is not enough to overcome it.

In Fig. 11-13, the similar trends can also be seen where the bed expansion height was 0 cm when the CaO-sand percentages mixture was 70 and 60% at low superficial velocity of 0.02-0.06 m sec^-1. These happened under the constant air pressure of 3, 4 and 5 bars. The superficial flow velocity has not reach the minimum superficial flow velocity where the upward force exerted by the fluid on the particles is not sufficient to force the weight of CaO-sand mixture of the bed.

Fig. 11:	Bed expansion height results (3 bars)

Fig. 12:	Bed expansion height results (4 bars)

Fig. 13:	Bed expansion height results (5 bars)

In Fig. 14, its can be observed that the bed expansion height increased between 0 to 14.5 cm when the superficial velocity increased from 0.02-0.215 m sec^-1 under 6 bars of the constant air pressure. The increased of the bed expansion height can be seen for all CaO-sand percentages mixture (70, 60, 50 to 40%). The adhesive force between CaO particles existing in the powder has been reduced due to the increased of the air pressure and the amount of sand particles in the mixture. The results obtained were similar with the explanation by Yates (1983).

Effect of the air volume flow rate and air pressure: Air volume flow rate and pressure are important operating parameters in fluidized bed.

Fig. 14:	Bed expansion height results (6 bars)

Fig. 15:	No rat holes and fluidization formed during operation (15 L min-1, 1000 μm, 70% CaO, 2 bars)

The 1000 μm CaO-sand mixture was tested with 5, 15, 25, 35, 45 and 55 L min^-1 flow rates at different air pressures to determine the fluidization behavior. As shown in Fig. 10-14, when the air volume flow rate increases (in term of superficial flow velocities 0.020 to 0.215 m sec^-1), the bed expansion height also increased at all air pressures except at the low superficial flow velocity less than 0.06 m sec^-1 at 2, 3, 4, 5 bars (70, 60% CaO mixture) and at 5 L min^-1 (0.02 m sec^-1) at all variables. At these conditions, no fluidizations were found. The reason is due to the strong interaction force that acts between two macroscopic bodies such as particle-particle and particle-wall that exist in the bed material. Furthermore the amount percentage of CaO is too much to be overcome by air volume flow rate and pressure. Figure 15 shows the one example of the pictures taken for this phenomenon.

CONCLUSIONS

The characteristics on fluidization behaviors of 1000 μm CaO-sand mixtures in the cold model experiment were determined. The investigations of the fluidization of the bed in term of the bed expansion height and the pressure drop have been listed as follows:

ACKNOWLEDGMENTS

The authors would like to express their appreciation to The Research University Grant Scheme of University Science Malaysia (Grant No. MEKANIK/811122) for providing financial support for this study.