A Study of the Hydrodynamics Behavior of Cylindrical Gas-Solid Fluidized Beds for pharmaceutical material “ Paracetamol “

The hydrodynamics behavior of gas solid fluidized beds is complex and it should be analyzed and understood due to its importance in the design and operating of the units. The effect of column inside diameter and static bed height on the minimum fluidization velocity, minimum bubbling velocity, fluidization index, minimum slugging velocity and slug index have been studied experimentally and theoretically for three cylindrical columns of 0.0762, 0.15 and 0.18 m inside diameters and 0.05, 0.07 and 0.09 m static bed heights .The experimental results showed that the minimum fluidization and bubbling velocities had a direct relation with column diameter and static bed height .The minimum slugging velocity had an inverse relation with static bed height and a direct one with column diameter .There was no agreement between the experimental and calculated values of Umb for Di=0.0762m ,this was a result to the assumption used in the correlation development. The fluidization index values were around 1 in all cases and that proved that the material is of Geldart type B. There was not a significant dependence of fluidization index and slug index on static bed height and column diameter.


Minimum Fluidization Velocity (Umf):
It is a very important parameter when the hydrodynamics of fluidized column is characterized. It can be defined as the gas superficial velocity at which the bed is fluidized. At this point the pressure drop is equal to the bed weight per unit cross sectional area. Minimum fluidization velocity depends on material properties, the geometry of the bed and gas properties and it is usually obtained experimentally. (Ramos, et al., 2002) , mentioned to the most common equations used to calculate the theoretical minimum fluidization velocity: for Rep < 20 (1) for Rep ˃ 1000 (2) It can be noted from the equations that the column diameter and static bed height are not included among the parameters even thought they have a large influence on minimum fluidization velocity, so the effect of each them should be studied experimentally. (Gunn, et al., 1997), investigated gas-solid cylindrical fluidized beds .They used glass beads of(100-500 )ϻm diameters ,column diameters of 89, 290 mm and bed heights of (20,30,40 ,50)cm , the results showed that there was not a significant variation in the Umf when the static bed height was increased . ( Hilal, et al., 2001), studied the effect of bed diameter on Umf for two cylindrical beds (0.29, 0.089) m and they concluded that the Umf decreased with an increase in the bed diameter. ( Ramos, et al., 2002), studied 2D rectangular gas-solid fluidized bed filled with different diameters of glass ballotoni ranges (160-700) ϻm and for different static bed heights (2,4,8,16,20,40,60) cm ,they found that the Umf decreased with increased bed width and increased when the bed height increased . (Akhil, et al ., 2010), used spherical particles in cylindrical gas-solid fluidized bed, they noticed that the Umf increased as the column diameter is reduced, or the bed height is increased. ( Escudero, 2010), studied the effect of material density and static bed height on the Umf for ground walnut shells, ground corncob and glass beads in a 3D cylindrical fluidized bed at different static bed heights and found that the Umf approximately remained constant. (Sarker, et al., 2012), investigated the most effective parameters on gas -solid fluidized beds, these are bed diameter, static bed height and particle shape. They used laboratory scale systems having 3.5 and 12.4 cm inside column diameter and a static bed height ranges 2 to 7 cm, they observed that the Umf decreased by decreasing the column diameter and there was not a significant dependence of Umf on bed height .( Yupeng, et al., 2017), studied the effect of column inner diameter of (8,12,and 16) mm and different static bed heights for Geldart B particles behavior in micro fluidized beds ,they concluded that both increasing static bed height and decreasing the diameter of the column increases the minimum fluidization velocity. Therefore, based on the mentioned literatures the bed diameter and static bed height are among the parameters those affect on Umf in gas -solid fluidized beds.

Minimum Bubbling Velocity (Umb) and Fluidizing Index (FI):
By Increasing the gas velocity larger than the Umf, the gas -solid system exhibit either bubbling or non-bubbling condition. The non-bubbling is known as homogeneous or particulate fluidization while the bubbling is referred as heterogeneous or aggregative fluidization. The upper limit of the gas velocity at Particulate condition corresponds to the first bubble appearance and the velocity at this point is called minimum bubbling velocity. (Abrahamsen, and Geldart, 1980).developed an equation for minimum bubbling velocity, it can be observed that the column diameter and static bed height are not among the equation parameters: (3) (Xiaotao, 1994), cleared that the Umb depends mainly on particles properties and it is equal to minimum fluidization velocity for Geldart B type. (Singh, and Roy, 2005), defined the minimum bubbling velocity as the gas superficial velocity where the first bubble appears. Also they mentioned that particulate fluidization lies between minimum bubbling velocity and minimum fluidization velocity. They defined an important parameter, the fluidization index (FI), it is a measure of the uniform expansion of the fluidized bed after bubbling. The fluidization index values range from 1 to 2 as a great value but in special cases it is more, the high values of FI means the bed capacity for holding gas is larger. It is an important parameter i.e., in fluidized bed reactor, for catalysts in industry means there is a plasticity and the bed can bent around the corners, contract and expand while the low value means the fluidization state is brittle where a small air velocity change cause a break from uniform fluidization state to a packed or bubbling bed state. FI can be expressed as:

FI =
They predicted a dimensionless relation for Umb in cylindrical columns for gas -solid systems at ambient conditions fluidized by air: (5) (Leu, and Tsai, 2009), studied the effect of static bed height for Plexiglas fluidized bed 1.8 m high and 10 cm bed inside diameter using type A material, they concluded that the Umb was independent on static bed height. (Escudero, et al., 2011) ,mentioned that for cylindrical bubbling fluidized beds, the Umb was independent on static bed height. (Panda, 2013) ,studied the Umf, Umb and FI for 3D tapered gas -solid fluidized bed with angles 7.36° and 9.28°, he used Limestone, Hematite and Dolomite with range of particle size 20-80 ϻm. He observed that they increase with increased static bed height. (Gajbliye, et al., 2013) ,studied the hydrodynamics and stability of a fluidized bed with respect to minimum bubbling velocity, minimum fluidization velocity and fluidization quality ,they used Geldart group B particles ,they named the fluidization quality for Umb /Umf and it was equal to one for this type. (Agu, et al., 2017) ,had an experimental study on samples of 150-450 ϻm limestone particles and two samples of 100-550 ϻm glass beads, they concluded that the bubbling velocity is determined when the first bubble was observed in the top of the fluidized bed also they mentioned that for a bed of Geldart B type with particles size range 100-1000 ϻm, the bubbles appear as soon as the fluidization is started.

Minimum Slugging Velocity (Ums) , Bubbling Bed Index (BBI) and Slug Index
(Singh , and Roy, 2008), defined the minimum slugging velocity as the gas velocity at which the size of bubble equals to the bed diameter, at this velocity the bed is lifted by the gas and this phenomenon is called slug formation. Slugging increases entrainment problems and lowers the bed performance. They called the difference between Ums and Umb as the bubbling regime. They developed a Correlation for Ums prediction in cylindrical gas -solid fluidization systems. They used air at ambient temperature and non-spherical particles of diameter range 324-925 micron and density of 1500-4800 Kg/m 3 : They defined the Bubbling bed index (BBI) as the ratio of minimum slugging velocity to minimum bubbling velocity. BBI predicts the bubbling range for gas-solid fluidization system: Bubbling Bed Index=BBI= (7) (Ho, et al., 1983) ,used different beds of sand and glass particles (358-1112) ϻm, they observed that the Ums is independent on the static bed height and column diameter. (Yang, 2003) ,considered that the slug appears in fluidized beds when the static bed height (Hs) /the diameter of the bed (Di) is larger than 2 to ensure that the bubbles have the time to coalesce to larger bubbles and the slugging velocity is reached when the bubbles become 2/3 of the internal diameter of the bed. (Dimattia, et al., 2009) ,studied a bed of Geldart type D for coarse sand, 10mm glass beads, whole peens and long rice at different static bed heights, they mentioned that the Ums was independent on static bed height because those particles offer a low resistance to gas flow and the shape of the particles has the important role on Ums . ( Kong, et al., 2017) ,investigated the slugging velocity in 0.05 m inside bed diameter for Geldart A particles type; they concluded that for fine particles when the diameter is less than 100 ϻm, the Ums was independent on static bed height. (Agu, et al., 2017) ,used glass and limestone particles in a cylindrical column with 1.4m height and 10.4 cm internal diameter using air at ambient conditions as the fluidizing medium and static bed height of 52 ,58 and 64 cm ,they concluded that the transition from bubbling to slugging is related to bed height and the slug formation depends on aspect ratio ,which is defined as the ratio between bed height to column diameter, also they suggested that the Ums decreases with an increase in bed height up to hmf=60 Di 0.175 after this it increases . 196 studied 3D cylindrical bed of 0.084m diameter and 1.5m height, sand of mean particle diameter of 234.74ϻm was used .They used aspect ratio (Hs/Di) of 0.7, 1.0, 1.5, 2.0 and 2.5 .They concluded that as the static bed height increased the slugging velocity decreased because the turbulence increased .Also the Umf droped to a stable value, Umb remained constant and Ums decreased with the increase of static bed height.
(Agu, et al., 2017),defined a slug index which indicates the ease of bed slugging, the low values means the bed can easily slug. Slug index slightly depends on bed height: Bubbles and slugs flow influence the solid and gas interaction in fluidized beds, so identifying the velocities of bubbling and slugging are important in the study of the hydrodynamics behavior of fluidized columns.
The main objective of this work was to study the effect of column inside diameter and static bed height on the minimum fluidization velocity, minimum bubbling velocity, fluidization index, minimum slugging velocity, bubbling bed index and slug index and the experimental results were compared with the theoretical values by using the correlation equations.

MATERIALS and METHOD 2.1 Material and Chemicals
The material used as a bed was a pharmaceutical material, Paracetamol, its properties is shown in Table1. The density is 600 g /m 3 and a mean particle size of 1.06*10 -3 m. (Geldart, 1986) ,classified powders according to the relation between particle diameter and density difference between the solid and the fluidizing gas, Fig.1. The material is classified of type B according to Geldart classification and the bed behavior is identified according to this practical classification.  ( Cocco, et al., 2014) ,mentioned that Geldart group B particles fluidize easily and are widely used in unit operations because they cause few difficulties. Most pyrolysis and fluidized bed combustors units use particles of this type. They form bubbles at the onset fluidization so the minimum fluidization velocity almost equals to the minimum bubbling velocity. Also they cleared that this type allows the formation of large bubbles size nearly equal to 2/3 of the internal column diameter and that cause slug formation.

Method:
The experimental set up is shown in Fig.2 and it consists of the following parts: 1-Air Compressor: 2-Silica gel column: A column was filled with silica gel granules to adsorb any humidity exits in the air. The color of the silica was blue and became pink when it was saturated with moisture. 3-Calming section: A column was used to get a stable rotameters readings. 4-Rotameters: Two rotameters were used .They calibrated at 20°C and 0.981 bar abs. m 3 /hr. air. 5-Manometer. 6-Fluidization column: It consists of a cylindrical column of 0.0762, 0.15 and 0.18 m inside diameters and 1.5 m height and made of Perspex material. A calming section with Rachig rings was used at the air entrance to the column for uniform distribution of the air throughout the bed. Air distributor was fitted at the column bottom. A measured amount of the material was charged from the top of the fluidized bed ,the bed fluidized once and the particles allowed to settle down slowly, the desired static bed height was recorded ( The bed was fluidized once before each experiment to get the particles in a loosely state). Compressed air at ambient temperature was allowed to pass through the silica gel column, then through the calming column .The air passed through a porous plate distributor to the fluidization column and its velocity was increased until the minimum fluidization velocity was observed 198 experimentally and the height at minimum fluidization velocity was recorded. The air velocity was increased until the first bubble observed visually and the first bubble erupted from the bed free surface and the velocity recorded as the experimental minimum bubbling velocity then the air velocity was further increased till the slug formation started, the velocity recorded as experimental minimum slugging velocity. This procedure was repeated for each column diameter, three times for each bed height .Each experiment was repeated for several times and the average value was reported.

RESULTS and DISCUSSION
The experiments were carried out for measuring the minimum fluidization velocity, minimum bubbling velocity, fluidization index, particulate fluidization range, minimum slugging velocity, bubbling bed index and slug index. A pharmaceutical material "Paracetamol" was used. The variables were the column inside diameter 0.0762, 0.15 and 0.18m and static bed height 0.05, 0.07 and 0.09m. The results are shown in Figs.

3-14 and Tables 2 and 3.
It can be noted from Table2. and Figs. 3 and 4 that the Umf increased by the increasing of the static bed height for the same column diameter and that agreed with (Akil, et al., 2010) , (Ram, 2013) and (Panda,2013), this can be explained due to the drag force affecting on the particles during the process of fluidization. The friction of the wall opposes the weight of the bed so increasing the bed height means increasing the bed weight and that increases the friction enhancement which causes to increase the Umf .Also the Umf increased with Journal of Engineering Volume 26 May 2020 Number 5 199 increasing column diameter for the same static bed height and that agreed with (Sarker, et al.,2012), this can be explained that by increasing column diameter and for the same bed height means larger amount of the material and that needs more force and larger velocity to raise the bed to reaches the minimum fluidization velocity. (Yupeng, et al., 2017) ,explained that the bed height increase cause to increase the wall effects: the boundary wall effect and the particles wall friction which are contributed, the second one effect mainly on the particles causing higher values of Umf. Also the particle wall friction and column diameter are coupled and can not be separated, they are responsible for the Umf variation. The gas superficial velocity was further increased until the first bubble was noticed at the surface of the fluidized bed and the velocity recorded as the experimental Umb as shown in Table 2., and Figs. 5, 6 and 7. The calculated values were obtained by using Eq. (5). Fig.5 shows a comparison between the experimental and calculated values of Umb, it can be observed that for Di= 0.0762m there was no agreement between them, a very good agreement for Di=0.15m, and a fairly one for D=0.18m. This diversion can be explained due to one of the assumptions in the correlation development was the particles are non-spherical . (Singh, andRoy, 2005, 2008) ,explained that the reason of the difference is due to the particle shape and its effect on drag and Umf then on Umb (There was not another correlation for sphere particles including the bed height and column diameter in the literatures to recalculate the Umb). (Singh, and Roy, 2005) ,whom predicted the correlation equation comment that" It is always better to rely on experimental values than on the theoretical from the correlations". From Fig.6 and Fig.7 it can be concluded that the Umb increased by increasing the static bed height and column diameter and that agreed with (Ram, 2013) and (Panda, 2013). The increase in Umf and subsequently in Umb can be explained due to the increase of frictional interaction between the particles and the wall, for the same static bed height and larger diameter that means bigger weight and that increases the drag forces and wall friction which cause to an increase in Umb. It was observed that directly after the Umf observation, the bubbling velocity was noticed where many bubbles appeared above the distributor plate, they became larger and moved upwards through the bed. Also the values of Umf and Umb are approximately the same and that agreed with (Agu, et al., 2017) and (Cocco, et al., 2014) ,that the behavior of this kind of material was according to Geldart classification and can be considered of kind B. Also that agreed well with (Jaiswal, et al., 2018) appear as soon as the bed fluidized where ho/Di for the three bed heights and column diameters equal to (0.66, 0.92, 1.2), (0.3, 0.47, 0.6) and (0.28, 0.39, 0.5) respectively.  There was a short period of particulate fluidization (Umb -Umf) with a range of (0.01-0.004 m/s) shown in Table 2. , this period disappeared in bubbling fluidized beds where nearly the Umb = Umf . The fluidization index was calculated using Eq. (4) and the results are shown in Table 2. and Figs. 8 and 9, it is obvious that there was not a large dependence on static bed height and column diameter also the values were around 1, these values mean the bed has a less capacity to hold gases and a large possibility to form bubbles for a small increase in air velocity above the minimum fluidization velocity .It can be concluded that for materials of Geldart type B, the fluidization index was nearly unity and that agreed well with (Gajbliye,  et   For further increase in gas velocity, slug started to form, it was noted visually with the structure of rounded-nosed, that agreed with (Agu, et al., 2017) , for easy fluidized materials, the material used was considered of this type, the slug had that structure and it can be noted visually for narrow (shallow) beds. The experimental and calculated values of Ums are shown in Table 3. and Figs.10, 11 and 12.  . (6). The difference can be explained due to an assumption used in the development of the correlation and that was the particles are non-spherical. (Singh, andRoy,2005,2008) ,explained that the reason of the difference is due to the particle shape and its effect on drag forces. (Dimattia, et al., 2009) ,concluded that the shape of the particles has an important role on Ums. A good agreement between the experimental results and the calculated values of Ums was obtained for the three column diameters and static bed heights when another equation have been used, Eq. (9), to calculate the Ums , this equation developed by (Geldart, and Baeyens, 1974). The results are shown in Table3. and From Table 3. and Fig. 11, it can be observed that Ums had an inverse relation with static bed height and that agreed with (Jaiswal, et al., 2018) ,that when the static bed height increased, the slugging velocity decreased because the turbulence increased. According to (Agu, et al. ,2017) ,the values of hmf =60 Di 0.175 were( 86,96.4 and 99.5 cm ) if the hmf value exceeded those values for each individual column diameter ,the Ums increased with bed height ( The 60 Di 0.175 value means the height of the fluidized bed at conditions of minimum fluidization for a flow of stable slug) ,experimentally hmf for each diameter for the three bed heights were (6,7.5 and 10), ( 5.6, 7.5, 10) and ( 5.4, 7.3, 9.2) cm they did not exceeded the upper limit and that means there was a stable slug and the Ums decreased with bed height . Also it can be noted that the lines were almost parallel to each other, so it can be concluded that the relation between the Ums and static bed height is linear and that agreed with (Agu, et al., 2017).
From Table 3. and Fig.12 it can be observed that the Ums had a direct relation with column diameter, this can be explained that bubbles first formed at the distributor then they rise and became larger so if the gas velocity increased the bubbles grow in size and become large enough to occupy nearly the whole diameter of the column then the slug starts to form. By increasing the column diameter and for the same height which means larger weight of the material, the wall friction increases and that increases the resistance to air flow and cause to an increase in Ums .  hs=0.09m Figure.12 The relation between Ums and column diameter for different static bed heights.

Journal of Engineering
From Table 3, it can be observed that the bubbling bed index (BBI) calculated from Eq. (7) had the range of 1.9-2.4 and it decreased with increasing static bed height and there was a slightly dependence on it because by increasing the bed height ,the minimum bubbling velocity slightly increased ,so according to Eq. (7) the BBI decreased . The slug index calculated by using Eq. (8) and From Table 3. and Figs. 13 and 14 the slug index also decreased with increased bed height and they had nearly the same values 1.9-2.4 and that agreed well with (Agu, et al. , 2017). Both of them decreased with column diameter but there were not a significant effect.

Figure. 13
The relation between slug index and static bed height for different column diameters. Figure.14 The relation between slug index and column diameter for different static bed heights.

CONCLUSIONS
-The minimum fluidization velocity nearly equals to the minimum bubbling velocity for this kind of material, the material considered of type B according to Geldart classification of particles. Umf has a direct relation with column diameter and static bed height.
-The minimum bubbling velocity increased with an increase in static bed height and column diameter.
-The fluidization index values were around 1 in all cases and that ensures the Paracetamol material is of Geldart type B.
-The minimum slugging velocity had an inverse relation with static bed height and a direct one with column diameter.
-The bubbling bed index and the slug index slightly depend on static bed height and there was not a significant effect for both of them with column diameter change.

BBI= bubbling bed index, dimensionless
Di= column inside diameter, m d p = mean particle diameter, m