Please copy and paste this embed script to where you want to embed

CHAPTER 1 Generalities and Basics of Fluidization

I. INTRODUCTION Fluidization is a unit operation, and through this technique a bed of particulate solids, supported over a fluid-distributing plate (often called the grid), is made to behave like a liquid by the passage of the fluid (gas, liquid, or gas–liquid) at a flow rate above a certain critical value. In other words, it is the phenomenon of imparting the properties of a fluid to a bed of particulate solids by passing a fluid through the latter at a velocity which brings the fixed or stationary bed to its loosest possible state just before its transformation into a fluidlike bed. A. Fluidlike Behavior Let us consider the various situations that could prevail in a bed of particulate solids. When there is no fluid flow in the bed, it remains in a static condition and the variation in pressure across the bed height is not proportional to its height, unlike in a liquid column. When a fluid such as a gas or a liquid is allowed to percolate upward through the voidage of a static bed, the structure of the bed remains unchanged until a velocity known as the minimum fluidization velocity is reached; at this velocity, drag force, along with buoyant force, counteracts the gravitational force. In this situation, the bed just attains fluidlike properties. In other words, a bed that maintains an uneven surface in a static, fixed, or defluidized state now has an even or horizontal surface (Figure 1.1a). A heavy object that would rest on the top of a static bed would now sink; likewise, a light object would now tend to float. The pressure would now vary proportional to the height, like a liquid column, and any hole made on the vessel or column would allow the solid to flow like a liquid. All these features are depicted in Figure 1.1.

© 1999 by CRC Press LLC

Figure 1.1 Examples of fluidlike behavior of fluidized bed relative to fixed bed.

B. Fluidization State 1. Gas/Liquid Flow The fluid under consideration which flows upward can be either a gas, a liquid, or both. In general, liquid fluidized beds are said to have a smooth or homogeneous or particulate nature of fluidization. The bed expands depending on the upward liquid flow rate, and due to this expansion the bed can become much higher than its initial or incipient height. In contrast, a gas fluidized bed is heterogeneous or aggregative or bubbling in nature and its expansion is limited, unlike what happens in a liquid fluidized bed. It is seldom possible to observe particulate fluidization in a gas fluidized bed and aggregative fluidization in a liquid fluidized bed. If the fluid

© 1999 by CRC Press LLC

flow regimes are such that a bed of particulate solids has a boundary defined by a surface, then it has solid particles densely dispersed in the fluid stream. In other words, a dense-phase fluidized bed is achieved. When the surface is not clearly defined at a particular velocity, the solid particles are likely to be carried away by the fluid. This situation corresponds to a dilute or a lean phase. The situation where solid particles are entrained by the fluid flow corresponds to a state called pneumatic transport. As the velocity of the liquid in a liquid fluidized bed is increased, homogeneous or particulate fluidization with smooth expansion occurs, followed by hydraulic transport of particles at a velocity equal to or greater than the particle terminal velocity. In the gas fluidized bed, bubbling is predominant. The minimum bubbling velocity is the velocity at which the bubbles are just born at the distributor. The bubbling bed at velocities greater than the minimum bubbling velocity tends to slug, especially in a deep and/or narrow column, and the slugging is due to the coalescence of bubbles. When the bubbles coalesce and grow as large as the diameter of the column, a slug is initiated. Now solids move above the gas slug like a piston, and they rain through the rising slugs. Here the gas–solid contact is poor. The slugging regime, through a transition point, attains a turbulent condition of the bed, and this process is often termed fast fluidization. Pneumatic transport of solid particles by the gas stream occurs at and above the particle terminal velocity. Liquid and gas fluidized beds for various gas flow rates are illustrated in Figure 1.2.

Figure 1.2 Liquid and gas fluidized beds at various operating velocities.

© 1999 by CRC Press LLC

2. Onset of Fluidization Estimation of the onset of the fluidization velocity is essential because it is the most important fundamental design parameter in fluidization. This velocity determines the transition point between the fixed bed and the fluidized bed. In a fixed bed, solid particles remain in their respective fixed positions while the fluid percolates through the voids in the assemblage of particles. In such a situation, the fluid flow does not affect or alter the voidage or the bed porosity. As the fluid flow is increased, the bed pressure drop increases. At a certain stage, the bed pressure drop reaches a maximum value corresponding to the bed weight per unit area (W/A); at this stage, a channelfree fluidized bed at the ideal condition is obtained. 3. Situation at the Onset of Fluidization Let us examine the situation at the onset of fluidization; the corresponding fluidization velocity is also referred to as the incipient velocity. When this velocity is just attained, the fixed bed of particles exists in is loosest possible condition without any appreciable increase in bed volume or bed height. In such a condition, the bed weight less the weight equivalent to buoyancy is just balanced by the drag due to the upward flow of the fluid. In other words, the distributor plate or the grid which supports the bed of solids does not experience any load under this condition. The velocity at which the bed is levitated, achieving a state of fluidlike behavior just at the transition of a fixed to a fluidized bed, is called the minimum fluidization velocity. The transition from a fixed to a fluidized bed may not be the same for increasing and decreasing direction of the fluid flow; this is particularly so when the fluid is a gas. 4. Bed Pressure Drop We will now examine the various situations that can occur in a bed pressure drop versus superficial velocity plot for a typical gas–solid system. In its true sense, the superficial velocity is the net volume of fluid crossing a horizontal (empty) plane per unit area per unit time. This superficial velocity is many times smaller than the interstitial fluid velocity inside the bed. Nevertheless, superficial velocity is considered because of convenience and ease of measurement. When a gas passes through a fixed bed of particulate solid, the resistance to its flow, in addition to various hydrodynamic parameters, depends on the previous history of the bed, that is, whether the bed under consideration is a well-settled bed or a well-expanded and just settled bed. In a well-settled bed, the important structural parameter, the bed voidage (), is relatively low, and thus the pressure drop obtained initially by passing the gas upward is of a relatively high magnitude, as depicted by line A–B in Figure 1.3. This figure is similar to one depicted by Zenz and Othmer1 and Barnea and Mednick.2 At point B, a transition from a fixed bed to a fluidized bed starts, and this prevails up to point C. The bed pressure drop beyond C for a fluidized bed remains unchanged in an ideal case even though the superficial velocity (U) is increased. The bed pressure drop beyond point D, which corresponds to the

© 1999 by CRC Press LLC

Figure 1.3 Variation in bed pressure drop (∆P) with superficial velocity (U).

particle terminal velocity for a monosized bed of solids, is no longer constant, and it increases in a manner similar to an empty column. This is so because the particles are carried away from the bed or are completely entrained when the superficial velocity equals the particle terminal velocity. In this situation, the bed voidage () is unity, or the volume fraction of solid particles (1 – ) is zero. Line EDF corresponds to the pressure drop in the empty column. When the upward flow of gas is gradually decreased, the bed pressure drop assumes its original path on a pressure drop versus velocity plot as long as the bed continues to be in a state of fluidization. Retracing the path, as shown in Figure 1.3 by line DCG, indicates that the pressure drop obtained for a fixed bed during its settling is lower than that obtainable for increasing upward flow of gas. Point C on line GCD is the transition point between a fixed and a fluidized bed, and the velocity corresponding to this is the minimum fluidization velocity. It may be observed that during increasing flow through the bed, there is no distinct point marking the transition, except that zone which corresponds to B–C with a peak. The relatively high value of the pressure drop (∆P) along line AB in a fixed bed when the flow is in the laminar regime compared to line GC of an expanded settled bed is due to low bed permeability. Let us now look at the fluid dynamic aspects at points A and B. For laminar flow conditions, most correlations for the pressure drop give: ∆P α U(1 – )2/3

(1.1)

At point A, the bed porosity () corresponds to the static bed value (0) (i.e., = 0), and at point B, the bed porosity is equal to the minimum fluidization value (mf) (i.e., = mf). From Equation 1.1, it follows that:

© 1999 by CRC Press LLC

(∆P)/[U(1 – )2/3] = Constant

(1.2)

and this is true if is not altered for a fixed bed. However, as indicated in Figure 1.1, points A and B correspond to a fixed bed but to different values; the proportionality constant in Equation 1.2 is thus altered. In view of this, Barnea and Mednick2 cautioned against the use of any unmodified fixed bed correlation for predicting the minimum fluidization velocity and also pointed out that point C in Figure 1.3 is the limiting condition for a fixed bed and point B for a fluidized bed. Because the particle concentration and its randomness affect the pressure drop, any attempt to use a fixed-bed pressure drop correlation for the purpose of predicting its minimum fluidization velocity will lead to erroneous results. C. Advantages of Fluidized Bed 1. A high rate of heat and mass transfer under isothermal operating conditions is attainable due to good mixing. 2. A fluidlike behavior facilitates the circulation between two adjacent reactors (e.g., catalytic cracking and regenerator combination). 3. There is no moving part, and hence a fluidized bed reactor is not a mechanically agitated reactor. For this reason, maintenance costs can be low. 4. The reactor is mounted vertically and saves space. This aspect is particularly important for a plant located at a site where the land cost is high. 5. A continuous process coupled with high throughput is possible. 6. No skilled operator is required to operate the reactor. 7. The fluidized bed is suitable for accomplishing heat-sensitive or exothermic or endothermic reactions. 8. The system offers ease of control even for large-scale operation. 9. Excellent heat transfer within the fluidized bed makes it possible to use low-surface-area heat exchangers inside the bed. 10. Multistage operations are possible, and hence the solids residence time as well as the fluid residence time can be adjusted to desired levels. D. Disadvantages of Fluidized Bed 1. Fine-sized particles cannot be fluidized without adopting some special techniques, and high conversion of a gaseous reactant in a single-stage reactor is difficult. 2. The hydrodynamic features of a fluidized bed are complex, and hence modeling and scaleup are difficult. 3. Generation of fines due to turbulent mixing, gas or liquid jet interaction at the distributor site, and segregation due to agglomeration result in undesirable products. 4. Elutriation of fines and power consumption due to pumping are inevitable.

© 1999 by CRC Press LLC

5. Sticky materials or reactions involving intermediate products of a sticky nature would defluidize the bed. 6. Limits on the operating velocity regime and on the choice of particle size range are disadvantages of fluidization. Fluidization of friable solids requires careful attention to avoid loss of fines formed due to attrition. 7. Highly skilled professionals in this area are needed for design and scaleup. 8. Erosion of immersed surfaces such as heat-exchanger pipes may be severe. 9. Reactions that require a temperature gradient inside the reactor cannot be accomplished in a fluidized bed reactor.

II. PROPERTIES OF PARTICLES AND THE GRANULAR BED A particulate solid has several properties which, in addition to the density, size, shape, and distribution, also play a key role in determining stationary bed or fixed bed properties. The roughness and the voidage associated with the particles should also be considered in fluid–particle interactions. A. Particles 1. Size Particulate materials or granular solids, whether manufactured or naturally occurring, can never have the same particle size. In other words, particulate solids comprised of uniformly sized particles are very difficult to obtain unless they are sized, graded, or manufactured under extreme control of the operating conditions. Achieving particles close in size in a manufacturing process is not simple; only specific processes like shot powders and liquid-drop injection into a precipitating solution can yield powders of relatively uniform size, but then only in the initial period of large-size particle production. Hence, uniformly sized particles are obtained by several physical techniques such as sieving, sedimentation, microscopy, elutriation, etc. 2. Definition There are several ways to define particle size. For the purpose of powder characterization, it is not customary to define all the sizes as given by various mathematical functions. A particle size that is the diameter equivalent of a sphere is used along with a shape factor in many hydrodynamic correlations. The shape factor will be discussed and defined later in this section. Several definitions of the mean particle diameter are found in the literature, some of which are presented in the following discussion. If a sample of a powder of a given mass is constituted of particles of different sizes and if there are ni particles with a diameter dpi (i = 1 to N), then the diameter can be defined in a variety of ways.

© 1999 by CRC Press LLC

1. Arithmetic mean:

( (d ) = ∑

n d pi

∑

p AM

)

N i =1 i

(1.3)

N i =1

2. Geometric mean: (dp)GM = 3(dp1 × dp2 × … × dpn)0.5

(1.4)

3. Log geometric mean:

( )

log d p

∑ (n log d ) ∑ n

(1.5)

(∑ n ) [∑ (n d )]

(1.6)

=

GM

N i =1

i

pi

i

4. Harmonic mean:

(d )

p HM

=

N i =1 i

N i =1

i

pi

The harmonic mean diameter in principle is related to the particle surface area per unit weight, that is,

( ∑ n ) ⋅ π( d ) (∑ n ) ⋅ ρ π ( d )

2

N i =1 i

Total surface area = Weight

N i =1 i

p HM 3

p

= 6

p HM

6 1 ⋅ ρp dp

( )

(1.7) HM

where ρp is the density of the particle. 5. Mean diameter, dpm: 50% of particles > dpm

50% of particles < dpm

6. Length mean diameter, (dp)LM:

(d )

=

p LM

∑

n ⋅ d pi2

N i =1 i

∑

N i =1 pi

d

(1.8)

7. Surface mean diameter, (dp)SM:

(d )

p SM

=

(∑

N i=1 i

n d pi2

∑ n)

(1.9)

∑ n)

(1.10)

12

N i=1 i

8. Volume mean diameter, (dp)VM:

(d )

p VM

© 1999 by CRC Press LLC

=

(∑

N i =1 i

n d pi3

N i =1 i

13

9. Weight mean diameter, (dp)WM:

( ) dp

WM

=

∑

x dp =

N i =1 i

∑ ∑

N i =1 i

n d pi4

N i =1 i

n d pi3

(1.11)

where xi is the weight or volume percent of particles with diameter dpi. For example,

xi =

ni d pi3

∑

(1.12)

N i =1 i

n d pi3

10. Volume surface diameter, (dp)VS:

( ) dp

VS

=

6 = Sw ρ p

∑ ∑

N i =1 i

n d pi3

(1.13)

N i =1 i

n d pi2

where Sw is the surface area per unit weight of the particles:

Sw =

∑

N i =1 i wi

xs

and

xi =

ni d pi3

∑

N i =1 i

n d pi3

(1.14)

Zenz and Othmer1 gave an account of industries that are inclined to choose the particle diameter of their own interest and also showed that it is not necessarily meaningful to select any particle diameter as desired. Zabrodsky3 recommends the use of the harmonic mean diameter in fluidization studies. Geldart4 commented that the particle size accepted in packed and fluidized beds is the surface–volume diameter (dsv), defined as the diameter of a sphere that has the same ratio of the external surface area to the volume as the actual particle. For a powder that has a mixture of particle sizes, it is equivalent to the harmonic mean diameter. Another diameter is the volume diameter, which is the diameter of a sphere whose volume is the same as that of the actual particle, that is, dv = (6M/ρpπ)1/3

(1.15)

The particle size determined by sieve analysis is the average of the size or opening of two consecutive screens, and it may be referred to as dp. The surface–volume diameter for most sands5 is dsv = 0.87 dp. The ratio dsv/dp is expected to vary widely depending on the particle shape, and it may not be easy to determine this ratio experimentally for irregularly shaped particles. Based on calculations pertaining to

© 1999 by CRC Press LLC

18 regularly shaped solids, Abrahamsen and Geldart5 showed that if the volume diameter (dv) = 1.127 dp for a sphericity of 0.773, then dsv = 0.891 dp. 3. Sphericity Sphericity (Φs) is a parameter that takes into consideration the extent of the deviation of an actual particle from the spherical shape or degree of sphericity. It is defined as the ratio of the surface area of a sphere to that of the actual particle that has the same volume. Let us consider a sphere made up of clay with a surface area So. If the sphere is distorted by pressing it, the shape changes; the resultant clay mass has the same volume as it had originally, but now it has a different surface area, say Sv. The ratio of the original surface area (So) to the new surface area (S) for the distorted clay sphere is its sphericity. It can be mathematically defined6 as:

⎛S ⎞ Φs = ⎜ o ⎟ ⎝ Sv ⎠ constant volume

(1.16)

If the diameter of a particle (dp) is defined as the diameter of the sphere that has the same volume as the actual particle, then the shape factor (λ) is given by: 23

1 1 Vp = Φs = 5.205 S p λ

(1.17)

If j is the ratio of the particle volume to the volume of a sphere that has the same surface area as the particle, then λ j2/3 = 1

(1.18)

The sphericity of many regularly shaped particles can be estimated by analytical means, whereas it is not easy to estimate the sphericity value analytically for an irregularly shaped particle. Pressure drop correlations for fixed bed reactors incorporate the shape factor. Hence, these correlations are often used to obtain the shape factor for the particle using experimental data on pressure drop and the fluid–solid properties. The shape factor for a regularly shaped particle can be estimated readily. For example, the shape factor for a cube is 1.23. For cylinders, it is a function of the aspect ratio, and for rings it is dependent on the ratio of inside to outside diameter. It should be noted here that the hydraulic resistance or the bed pressure drop correlation used to estimate the shape factor should correspond to the laminar flow (Re < 10) regime and there should be no roughness effect. 4. Roughness The roughness of a particle obviously adds to the friction between particles, and it leads to an increase in bed porosity (loose packing) when the bed settles down. The increase in bed porosity in turn reduces resistance to fluid flow. In other words,

© 1999 by CRC Press LLC

the bed pressure drop in a bed of particles with rough surfaces is lower compared to one that has particles with smooth surfaces, which have a tendency to form a less dense or low-porosity bed. Leva7 estimated that the friction factor values for clay particles are 1.5 times higher than for glass spheres and 2.3 times higher than for fused MgO particles. The particle roughness has to be determined by measuring the friction factor and by comparing with standard reference plots for particles of various known roughness factors. B. Granular Bed 1. Bed Porosity or Voidage Bed porosity or voidage is affected by several parameters, such as the size, shape, size distribution, and roughness of the particles; the packing type; and the ratio of the particle diameter to the vessel diameter. Small or fine-sized particles have low settling or terminal velocities and low ratios of mass to surface area. Hence, fine particles, when poured into a vessel, settle slowly and create mass imbalance. As a result of these two factors, the bed has a tendency to form bridges or arches, which causes the bed voidage to increase. A bed with bridges or arches is not suitable for smooth fluidization. A dense bed of fine powders can be obtained by shaking, tapping, or vibrating the vessel. 2. Voidage and Packing Voidage for spherical particles depends on the type of packing and varies from 25.95% (rhombohedral packing) to 47.65% (square packing). For nonuniform angular particles, voidage can vary widely depending on the type of packing (i.e., whether it is loose, normal, or dense). A bed is in its loosest packed form when the bed material is wet charged (i.e., the material is poured into the vessel containing the liquid or the material and after pouring into an empty container is fluidized by a gas and then settled). The bed will be dense when the container or the column is vibrated, shaken, or tapped for a prolonged period. A representative variation of voidage () of packing with uniformly sized particle diameter (dp) for three packing conditions (viz., loose, normal, and dense) is shown in Figure 1.4. 3. Polydisperse System Furnas9 investigated experimentally the voidage of a binary system of varying particle size ratios. His studies showed that if the initial voidages of the individual components of the binary system are not the same, the voidage of the mixture will be less than the volumetric weighted average of the initial voidages. In a binary system of coarse and fine particles that have equal particle density and also equal voidage (), the volume fraction of the coarse particles is given as 1/(1 + ) at the condition of minimum voidage (i.e., maximum density). This low value of voidage is due to the fine particles that fill the interstices of the coarse material. A third component which is smaller than the second component and also finer relative to

© 1999 by CRC Press LLC

Figure 1.4 Voidage in uniformly sized and randomly packed beds. (From Brown, G.G., Katz, D., Foust, A.S., and Schneidewind, R., Unit Operation, John Wiley & Sons, New York, 1950, 77. With permission.)

the first component may be added to fill the interstices of the second component; theoretically, the process can go on for an infinite number of components. The resultant or the final volume fraction (m) of a solid mixture with n components is given by:

(1 – m ) = (1 +1 ) + (1 + ) + (1+ ) + … + (1+ ) 2

n

(1.19)

From the above expression, it is possible to obtain the percentage of each component required to produce the minimum voids by multiplying both sides by 100/(1 – m). 4. Container Effect The effect of particle roughness on voidage was discussed in the preceding section on particle roughness. Now we will consider the effect of the ratio of the particle size to the container/column diameter on bed voidage. Particle packing close to the walls is relatively less dense. Hence, for vessels, particularly when d/Dt (where d is the particle diameter and Dt is the vessel diameter) is large, the voidage contribution due to the wall effect is high, and the wall effect for particles away from it is insignificant for large-diameter vessels. Ciborowski’s10 data on bed porosity for various values of d/Dt corresponding to different vessel geometries and materials are shown in Table 1.1. It can be seen from these data that bed porosity increases as d/Dt increases.

© 1999 by CRC Press LLC

Table 1.1 Bed Porosities for Various Shapes of Packings d/D

Ceramic Spheres

Smooth Spheres

Smooth Cylinders

Raschig Rings

0.1 0.2 0.3 0.4 0.5

0.4 0.44 0.49 0.53 —

0.33 0.37 0.40 0.43 0.46

0.34 0.38 0.44 0.50 0.56

0.55 0.58 0.64 0.70 0.75

5. Important Properties of Particulate Solids a. Density There are in general three types of densities referred to in the literature: true, apparent, and bulk densities. True density is the weight of the material per unit volume, when the volume is considered free of pores, cracks, or fissures. The true density thus obtained gives the highest value of the density of a material. If, on the other hand, the particle volume is estimated by taking into consideration the intraparticle voids, then the density calculated on this basis is the apparent density. If ρt is the true or theoretical density and ρa is the apparent density for the same mass, then a simple relationship can be deduced between these two parameters. Let 0 be the intraparticle voids due to pores, cracks, etc.; Vsv the volume per unit mass of the particles in the absence of intraparticle voids; and Va the apparent volume per unit mass; then:

1 1 – V – Vsv ρ a ρt = I = a 1 Va ρa

(1.20)

Upon rearrangement of Equation 1.20, one obtains: ρa /ρt + I = 1

(1.21)

An equation for estimating ρa can also be rewritten as:

ρa =

1 V +

1 ρt

(1.22)

where V is the pore volume per unit mass. The bulk density (ρb) is obtained by considering the weight of granular solids packed per unit volume of a vessel or container. This density is less than the apparent density because the volume under consideration for the same mass of solid is increased in this case due to interparticle

© 1999 by CRC Press LLC

voids. If is the interparticle voidage for a granular bed of porous solids that have an intraparticle voidage I, then, as per the preceding procedure:

ρb + =1 ρa

(1.23)

Substituting for ρa from Equation 1.21 and modifying Equation 1.23, one gets:

ρb = (1 – I )(1 – ) ρt

(1.24)

The bulk density again depends on certain other parameters, such as the degree of packing. Packing, as discussed earlier, can be of three types and depends on the extent of tapping employed while compacting a granular bed of solids. The bulk density, when measured after compaction by tapping, is called the tap density. b. Angular Properties The angular properties that are significant in relation to studies on rheology of powders are 1. 2. 3. 4. 5.

Angle Angle Angle Angle Angle

of internal friction (αt ) of repose (θ) of wall friction (γ) of rupture (δ) of slide (ω)

III. GROUPING OF GAS FLUIDIZATION A. Hydrodynamics-Based Groups The type of fluidization specifically for gas fluidized beds is related to the properties of the gas and the solid. In a gas fluidized bed, the bubbles moving through the dense particulate phase have a strong influence on the quality of fluidization. Hence, it is important to define the types of fluidization with respect to the properties of the gas–solid system. 1. Geldart Groups In fluidization literature, most of the inferences are drawn from studies on one class of gas–solid system and then extrapolated to another group or class. This could have an adverse effect on scaleup and could result in the failure of the system. Much confusion and many contradictions in the published literature have

© 1999 by CRC Press LLC

been pointed out by Geldart,11 and these have been attributed to extending the data obtained on one powder to another powder. In view of this, Geldart12 classified powders that have similar properties into four groups and designated them by the letters A, B, C, and D. These groups are characterized by the difference in density between the fluidizing gas (i.e., air and the solid) and the mean size of the particles. A mapping of these groups is shown in Figure 1.5 for air fluidized beds. Of these

Figure 1.5 Geldart classification of powders. (From Geldart, D., Powder Technol., 7, 285, 1973. With permission.)

four groups, the two extreme groups are Group C, which is difficult to fluidize, and Group D, which is spoutable. The intermediate Groups A and B are suitable for the purpose of fluidization. Of these two groups, Group A powders have densephase expansion after minimum fluidization but prior to the commencement of bubbling, whereas Group B powders exhibit bubbling at the minimum fluidization velocity itself. Group A powders are often referred to as aeratable powders and Group B powders as sandlike powders. Detailed characteristics of powders that belong to the four groups are presented in Table 1.2. Geldart12 developed numerical criteria to differentiate Group A, B, and D powders. The numerical criteria for solid particle size (dp), density (ρs), and fluid density (ρf) are 1.

(ρs – ρf) dp ð 225

for Group A

(1.25)

Equation 1.25 is the boundary between Group A and Group B powders. 2.

© 1999 by CRC Press LLC

(ρs – ρf) dp2 Š 106

for Group D

(1.26)

Table 1.2 Geldart12 Classification of Powders Group A Example Particle size (dp), µm Density (ρs), kg/m3 Expansion Bed collapse rate Mixing Bubbles

Slugs

B

Cracking catalyst 30–100

Sand 40 < dp < 500

1) Split and recoalesce frequently Rise velocity > interstitial gas velocity For freely bubbling bed, rise velocity (30–40 cm/s) of small bubble ( interstitial gas velocity Size increases linearly with bed height and excess gas velocity No evidence Cloud-to-bubble-volume ratio not negligible Slugs at high velocity of gas, rise along wall and no evidence of breakdown

Group C Example Particle size (dp), µm Density (ρs), kg/m3 Expansion Bed collapse rate Mixing

Bubbling/ fluidization

Finer 1400 Solid particles are spoutable; hence expansion is similar to spouted bed Fastest of all groups because of dense or large size of particles Particle mixing as well as heat Solid mixing is relatively poor; high transfer between a surface and particle momentum and little particle bed are poorer than Group A and contact minimize agglomeration; gas B velocity in dense phase is high, and hence backmixing of dense-phase gas is less As the interparticle forces are Bubbles form at 5 cm above the greater than the force exerted by distributor fluid, the powder lifts as slug in Bubbles of similar size to those of small-diameter column or channel; Group B are possible at same bed hence bubbling is absent or not height and excess gas flow rate; reported largest bubbles rise slower than intertitial gas, and hence gas enters Agglomeration due to excessive the bubble base and comes out at the electrostatic force Fluidization is generally possible by top using agitator or vibrator to break the channels Electrostatic charges removed by using conductive solids or solids with graphite coating or column wall with oxide coating

© 1999 by CRC Press LLC

Equation 1.26 is the boundary between Group B and Group D powders, where density is expressed in grams per cubic centimeter and the particle diameter (dp) in micrometers. No numerical criterion or equation for the boundary line between Group A and Group C powders was proposed. The classification of Geldart groups has been well recognized and is often referred to in the literature, even though several other criteria based on similar conceptual premises were proposed later. 2. Molerus Groups In the Geldart classification of powders, one could observe that as the particle density (i.e., the difference in the densities of the solid and the gas [air]) decreases, the boundaries separating the groups shift toward larger particle sizes. The transition between the groups, more specifically from Group C to Group A or from Group A to Group B, is not observed to occur at sharp boundaries. Molerus13 proposed some criteria to group the powders by taking into account the interparticle force as well as the drag exerted by the gas on the particle. The powder classification diagram of Geldart,12 superimposed with the criteria of Molerus,13 results in a mapping of powder groups in the manner depicted in Figure 1.6, wherein sharp boundaries can be seen during the transition from one group to another.

Figure 1.6 Powder classification of Geldart as modified by Molerus. (From Molerus, O., Powder Technol., 33, 81, 1982. With permission.)

1. Criterion for transition from Group A to Group B:

(ρ

© 1999 by CRC Press LLC

s

– ρg

)

πd p3 g 6 Fe

= K1

(1.27)

The constant K1 is dependent on the nature of the powder (i.e., soft or hard). Adhesive or cohesive force should be estimated or known for these powders. 2. Criterion for transition from Group C to Group A:

(ρ

s

)

– ρ g d p3 g Fe

= K2

(1.28)

where K2 is again dependent on the nature of the powders (soft or hard), and hence two straight lines with a narrow band or strip result in the diagram. The slope of the straight line obtained for the transition either from Group C to Group A or from Group A to Group B, when plotted on a log scale, is –3, and so the lines are parallel. 3. Criterion for transition from Group B to Group D: (ρs – ρg) dp g = K3

(1.29)

The constant K3 for a stable spouted bed of sand14 (dp = 600 µm and ρs = 2600 kg/m3) is 15.3 N m–3. It should be noted that in the criteria proposed by Molerus, the effect of consolidation of the bed due to gravitational load prior to fluidization, the effect of humidity, and the electrostatic effect are neglected. For this reason, the criteria developed have some limitations. However, the mapping of powder groups which also satisfy the Geldart criteria is simple and exhibits a clear transition at the boundaries. 3. Clark et al. Groups The two-dimensional mapping of the powder groups as proposed by Geldart12 and Molerus13 suffers from no provision for identification by numerical means that would be useful for computer analysis. Clark et al.15 devised a method of representing the powder groups of Geldart by certain dimensionless numbers. The following ranges of numerical values for dimensionless numbers that represent different powder groups have been proposed: Powder number @ < 1.5 1.5 < @ < 2.5 2.5 < @ < 3.5 3.5 < @ < 4.5

Powder type C A B D

(1.30) (1.31) (1.32) (1.33)

where the powder number (@) that fits with the two-dimensional mapping of the Geldart and the Molerus groups is complex.

© 1999 by CRC Press LLC

The powder number that fits with the lines in the Geldart mapping12 is given by:

{

[

]

}

@ = 1.221 log d p – 2 log(1 + 0.112( ∆ρ) –1.5 ) – 0.0885

[

+ 0.901(log( ∆ρ) + 1.5 log d p ) – 0.526

[1 + tan h( – c1 )]

1 + tan h c2 4

[

1 + tan h1 2

]

(1.34)

(

)

]

+ 0.667 log( ∆ρ) + 2 log d p – 0.5

1 + tan h( – c2 ) 2

where

c1 =

c2 =

1000 – ( ∆ρ)d p3 2

(1.35)

150

14, 000 – ( ∆ρ)d p3 2

(1.36)

3000

It can be seen that the equation or the model developed to match the Geldart mapping is rather cumbersome and does not fit very well. The powder number for Molerus powder mapping13 is relatively simple and is expressed as:

[

]

) 1 + tan2 hC

(

@ = 0.454 log( ∆ρ) + 3 log d p – 0.395 – 0.9 log d p + 2.45

3

(1.37)

where

[

]

C3 = ( ∆ρ)d p3 – 1.73 × 10 7 2 × 10 6

(1.38)

The equation(s) for the powder number developed by Clark et al.15 are dimensionally inconsistent, and the units for particle diameter and particle density are in micrometers and grams per cubic centimeter, respectively. Nevertheless, it has the convenience of determining the powder group without having a recourse to a mapping of powder classification. Clark et al.15 recommend the development of correlations that relate the powder number to such fundamental parameters as the minimum fluidization velocity and the minimum bubbling velocity, especially for Group A and B powders. 4. Dimensionless Geldart Groups So far, gas fluidization powder grouping has been discussed on the basis of the two-dimensional powder mapping proposed by Geldart12 and Molerus13 and the numerical representation suggested by Clark et al.15 In view of the difficulties associated with these schemes, Rietema16 attempted to remap powder classification using dimensionless numbers, namely, the Archimedes number and the cohesion

© 1999 by CRC Press LLC

number (C/ρsgdp), where C is the cohesion constant. The dimensionless representation of the Geldart classification for powders belonging to Groups A, B, and C is shown in Figure 1.7.

Figure 1.7 Dimensionless representation of Geldart powders. (From Clark, N.N., Van Egmond, J.W., and Turton, R., Powder Technol., 56, 225, 1968. With permission.)

As can be seen from Figure 1.7, the voidage of the packed bed influences the transition from Group A to Group C. The transition or boundary between Group A and Group B powder behavior occurs when the fluidization number, N (= ρ3s dp4 g2/µ2f EM), is

N A– B =

150(1 – 0 )

2

20 (3 – 2 0 )

(1.39)

where EM is the elasticity modulus of the powder bed (N/m2). For transition from Group A and Group C behavior, one has: NA-C = Ar (Ncoh)–1

(1.40)

where Ncoh is the cohesion number (C/ρs gdp). The parameter C is a cohesion constant of powder (N/m2). The boundary between Group D and any other group powder behavior occurs when: Ar (ρg/ρs) > 60,000

© 1999 by CRC Press LLC

(1.41)

One point to note is that the viscosity has a profound effect on the boundary between Group A and Group C powder behavior. Fine cracking catalysts (dp Ý 30 µm), when fluidized by nitrogen or neon gas, behave like Group A powders (i.e., show homogeneous bed expansion), but the behavior changes to that of Group C powders when fluidization is effected by hydrogen. As yet, there is no theory or explanation available for the effect of viscosity on Group C powders. B. Hydrodynamics- and Thermal-Properties-Based Groups The classification scheme for powder characterization is generally based on the hydrodynamic characteristics or the fluidization properties of various powders with air at ambient temperature. Saxena and Ganzha17 pointed out that a particle that is larger according to hydrodynamics classification can be smaller in terms of thermal properties. For example, the Geldart12 criterion for Group D powders in the case of sand fluidized by air at ambient temperature and pressure leads to a particle size greater than 0.63 mm, and this size obviously falls in the large category according to hydrodynamics classification. On the other hand, they could still be considered to be small particles in terms of thermal characteristics. For example, for large particles, the heat transfer coefficient to the immersed surface increases with particle size. In other words, heat transfer is controlled by the gas-convective component. In the case of fine particles, the heat transfer coefficient decreases with an increase in particle diameter. A size of 1 mm has been proposed12 to be the demarcation between small and large particle sizes. In view of this, Saxena and Ganzha17 suggested that a powder should be classified by considering both hydrodynamic and thermal characteristics. The Archimedes number (Ar) along with the Reynolds number at minimum fluidization (Remf) have been considered for powder grouping due to the fact that Remf and the Nusselt number at its maximum value are unique functions of Ar. Saxena and Ganzha17 classified powders into three groups and demonstrated the validity of such a classification by considering heat transfer data for sand and clay at ambient temperature and pressure. This classification takes into consideration the heat transfer correlations and the models for large particles. The particle groups and the grouping criterion according to Saxena and Ganzha17 are given in Table 1.3. The quantity Ψ is given by:

⎡ µ 2g Ψ=⎢ ⎢ ρg g ρs – ρg ⎣

(

)

⎤ ⎥ ⎥ ⎦

13

The two subgroups shown in Table 1.3 depict the clear transition between Groups I and II as well as II and III. The classification criterion has a simple dimensionless form and is very useful for setting conditions while computing. However, it cannot identify powder types as originally conceived by Geldart.12

© 1999 by CRC Press LLC

Table 1.3 Criteria for Powder Classification Based on Hydrodynamics and Thermal Properties Group I IIA IIB III

Criteria or Relation 3.35 ð Ar ð 21,700 or 1.5 Ψ ð dp ð 27.9 Ψ 21,700 ð Ar ð 13,000 or 27.9 Ψ ð dp ð 50.7 Ψ 13,000 ð Ar ð 1.6 × 106 or 50.7 Ψ ð dp ð 117 Ψ Ar Š 1.6 × 106 or dp Š 117 Ψ

Ar = dp3 ρf (ρs – ρf)g/dp3 . Ψ = [µf2 /ρf (ρs – ρf)g]1/3. From Saxena, S.C. and Ganzha, V.L., Powder Technol., 39, 199, 1984. With permission.

C. Variables Affecting Fluidization In a fluidization bed, the fluidization medium is always comprised of solid particles. The fluidizing medium is any fluid: gas, liquid, or gas and liquid. In normal fluidization, the fluidizing fluid flows upward, thereby counteracting the gravitational force acting on the bed of particulate solid materials. The various parameters that influence the fluidization characteristics can be classified into two major groups comprised of independent variables and dependent variables. The properties of the fluid and the solid, pressure, and temperature are the major independent variables. The dependent variables include Van der Waals, capillary, electrostatic, and adsorption forces. The parameters that influence fluidization behavior can be depicted in a flow diagram (Figure 1.8).

Figure 1.8 Parameters affecting fluidization behavior.

D. Varieties of Fluidization Depending on the mode of operation and the flow regime, fluidization can be classified in several ways. For example, in normal fluidization, the gas flows upward and the particle density is greater than the fluid density. If a fluidized bed is

© 1999 by CRC Press LLC

homogeneous, it is often referred to as particulate. Heterogeneous fluidization is often bubbling or slugging in nature. Particulate fluidization is akin to liquid fluidization. Slugging occurs at high velocities in a gas fluidized bed with a narrowdiameter column and a deep bed. If the density of the particulate solids is less than that of the fluid, normal fluidization is not possible, and in such a case the fluid flow direction has to be reversed (i.e., the fluid has to flow downward). Such a situation prevails for fluidizing certain polymers. This type of fluidization is termed inverse fluidization. Fine powders are normally difficult to fluidize due to interparticle forces. In such a case, it is necessary to overcome the cohesive forces by some external forces in addition to the drag force exerted by the fluid flow. Agitation or vibration helps to overcome the interparticle forces. The fluidized bed in such a case is called a vibro fluidized bed. In all the above types of fluidization, gravitational force plays a key role. There is a minimum fluid flow required to overcome the gravitational force, and the flow rate required just for fluidization may be much more than that demanded by stoichiometry. This would finally result in wasting the unreacted fluid, which in turn increases the cost of recycling. Furthermore, it is not advisable to use such normal fluidization for a costly gas. One alternative is to use centrifugal fluidization. A new class of fluidized beds comes into play at very high velocities (much higher than or equal to the particle terminal velocity), where carryover or elutriation of the bed inventory occurs. Here the solids are to be recycled into the bed; such a system is called a circulating fluidized bed. Particulate solids fluidized by either gas or liquid consist of two phases and belong to the category of two-phase fluidization. A bed of solids fluidized by both gas and liquid is known as three-phase fluidization. Three-phase fluidization is more complex than two-phase fluidization and has additional classifications depending on the flow direction of the gas and the liquid. In general, they can flow either cocurrently or countercurrently. Figure 1.9 shows in a nutshell some common varieties of fluidization.

Figure 1.9 Some common varieties of fluidization.

© 1999 by CRC Press LLC

IV. HYDRODYNAMICS OF TWO-PHASE FLUIDIZATION A. Minimum Fluidization Velocity 1. Experimental Determination a. Pressure Drop Method This method of determining the minimum fluidization velocity (Umf) involves the use of data on the variation in bed pressure drop across a bed of particulate solids with fluid velocity. The trend in variation of the bed pressure drop with the superficial gas velocity was shown in Figure 1.3 and discussed in detail. Figure 1.10

Figure 1.10 Various experimental methods to determine minimum fluidization velocity: (A) pressure drop, (B) bed voidage, and (C) heat transfer.

depicts the various methods by which the minimum fluidization velocity can be determined. The plot in Figure 1.10A depicts bed pressure drop versus gas velocity. The transition point from the fixed bed to the fluidized bed is marked by the onset of constant pressure. This is also the point at which the increasing trend in the bed pressure drop (∆Pb) of a packed bed terminates. For an ideal case, gas flow reversal in the fluidized bed condition does not change the magnitude of ∆Pb. However, the value of ∆Pb is smaller when the bed starts settling during flow reversal compared to previous values obtained at the same velocity in the increasing flow direction. The pressure drop method is the most popular means of determining Umf experimentally.

© 1999 by CRC Press LLC

b. Voidage Method Bed expansion in a fixed bed is negligible. Hence, the bed voidage () remains constant. When a fixed bed is brought to the fluidized state, the bed voidage increases due to bed expansion. The onset of fluidization corresponds to the point where the voidage just starts increasing with the gas velocity (U). This is shown in Figure 1.10B. The bed voidage becomes constant and is equal to unity when the gas velocity corresponds to the particle terminal velocity. This method of determining Umf is not simpler than the bed pressure drop method, because the bed expansion cannot be accurately determined by any simple (i.e., visual) means. c. Heat Transfer Method The variation in the wall heat transfer coefficient (h) with gas velocity (U) forms the basis of one of the interesting methods of determining Umf. The wall heat transfer coefficient increases gradually in a fixed bed as the gas velocity is increased, and it suddenly shoots up at a particular velocity, indicating the onset of fluidization. The velocity that corresponds to the sudden increase in the wall heat transfer coefficient is the minimum fluidization velocity. The trend in the variation of h with U is shown in Figure 1.10C. In this method, another important gas velocity, the optimum gas velocity (Uopt), which corresponds to the maximum wall heat transfer coefficient (Umax), is also obtained. This method of determining Umf is more expensive than the two aforementioned methods, and it requires good experimental setup to measure the heat transfer data under steady-state conditions. For these reasons, this method is seldom used to determine Umf in fluidization engineering. 2. Theoretical Predictions The various methods, based on first principles, available in the literature for the theoretical prediction of the minimum fluidization velocity can be broadly classified into four groups. These methods are derived from (1) dimensional analysis, (2) the drag force acting on single/multiparticles, (3) the pressure drop in a fixed bed extendable up to incipient fluidization, and (4) a relative measure with respect to the terminal particle velocity. Although a significant body of literature pertaining to the above methods exists and numerous correlations are listed by various researchers, there has been no classification of these correlations. The aforementioned methods are briefly described and the correlations based on them are listed in the following sections. a. Dimensional Analysis (Direct Correlation) This conventional method of developing a correlation takes into consideration the physical properties of the fluid and the solid. In its most general form, the correlation is given as:

© 1999 by CRC Press LLC

(

U mf = K d pa ρs – ρ f

) ρ µ g (φ b

c f

d f

c

m n s mf

)(1 – ) mf

p

(1.42)

Correlations of the above type are presented in Table 1.4. It can be seen that the factor ( φ sm nmf ) is variable because of the dependence of mf on φs. However, many correlations have been proposed by assuming that the quantity K φ sm nmf (1 – mf)p is a constant. This assumption is valid only within a certain range of experimental parameters. This type of direct correlation has the inherent disadvantage of involving a dimensional constant (K) that changes depending on the system of units used for the variables in Equation 1.42. b. Drag Force Method In this method, a force balance is assumed at incipient fluidization. The force balance equation is Fg = αFD + FB

(1.43)

where Fg, FD, and FB are, respectively, the force due to gravity, drag, and buoyancy, and α is a correction or multiplication factor for a multiparticle system. Because the drag force varies depending on the flow range, there can be several correlations for the prediction of Umf by this method. Taking 2 FD = α(1 2)CDρ gU mf (π 4)d p2

(1.44)

and using appropriate expressions for Fg and FB for a particle of diameter dp fluidized by a fluid of density ρg, a general equation can be written as:

3 α ⋅ CD Re 2mf = Ar 4

(1.45)

The above equation can be solved if an appropriate value of drag coefficient (CD) is chosen, depending on the flow regime. For Stokes flow (Remf ð 0.1), CD = 24/Remf. Hence Equation 1.45 becomes

Re mf =

Ar 18α

(1.46)

For Newton’s flow regime (i.e., for turbulent flow condition), Remf > 500 and CD = 0.44. Hence,

Re 2mf =

© 1999 by CRC Press LLC

Ar 0.33α

(1.47)

Table 1.4 Correlation for Minimum Fluidization Velocity Based on Dimensional Analysis Method n (1 – )p A. General Form for Umf = Ko dpa (ρs – ρf)b ρfc µfd gc φsm emf mf m n p K = Ko φ emf (1 – emf)

Ref.

a

b

Exponential Factor d e K

c

m

n

p

Remarks a GF, SI, Remf < 5 GF, SI units

Leva18

1.82 0.94

–0.06

–0.88 1

7.169 × 10–4

—

—

—

Miller and Logwinuk19 Rowe and Yacono20 Davidson and Harrison21 Yacono22

2

0.1

–1

1

0.00125

—

—

—

1.92 0

0

0

0

0.00085

0

2

1

0

–1

1

0.0008

—

1.77 0

0

0

0

0.000512

0

0

0

–1

0

0

0.361

0

0

0

—

—

—

Umf = cm/s, dp = µm (ρs – ρf) = µsb (bulk density) SI units SI units

—

—

—

CGS units

—

—

—

SI units, Remf < 5

0.9

Baerg et al.23 1.23 1.23

Kumar and 1.34 0.78 –0.22 0.56 0.78 0.005 Sen Gupta24 Baeyens and 1.8 0.934 –0.066 –0.87 0.934 9.125 × 10–4 Geldart25 Leva et al.26 1.82 0.94 –0.06 — — 7.39

2.92 0.945 Umf = cm/s, dp = µm — — All SI units

B. General Form for Re mf Remf = K (dp3 ρf2 g/µf2 )x Ref. Riba27 Ballesteros28 Doichev and Akhmakov29 Thonglimp et al.30 a

K 1.54 12.56 1.08 7.54 1.95

× × × × ×

⎛ ρ s – ρf ⎞ ⎜ ⎟ ⎝ ρf ⎠

y

x 10–2 10–2 10–3 10–4 10–2

0.66 1.0523 0.947 0.98 0.66

y 0.7 0.66 + 1.0523 0.947 0.98 0.66

Remarks

Remf < 30 30 < Remf < 180

GF = general form.

In the intermediate flow regime, there can be as many as six empirical correlations for CD, as proposed by Morse.31 The drag force on a single sphere situated in an infinite expanse of fluid, according to Schiller and Naumann,32 is

FD =

© 1999 by CRC Press LLC

πd p2 ρ gU gc

(3 Re + 0.45 Re

–0.313

)

(1.48)

for 0.001 < Re < 1000. The correlation given by Equation 1.48 fairly covers all three flow regimes and thus can be used in Equation 1.43 to predict Re at Umf conditions for most gas–solid systems (i.e., for Group A to C particles of Geldart’s classification). The factor α in Equation 1.43 has been found to be a function of voidage () by Wen and Yu.33 They found experimentally that for particulate fluidization: α = f() = –4.7

(1.49)

Using Equations 1.48 and 1.49 and the appropriate expressions for FB and Fg in Equation 1.43, the general form of the correlation for obtaining Remf in the range 0.001–1000 is

2.7 Re1mf.687 – 18 Re mf – 4.7 Ar = 0

(1.50)

for 0.818 < ρf (kg/m3) < 1135, 6 × 10–4 < dp (µm) < 25 × 10–2, 1060 < ρs (kg/m3) < 11,250, 1 < µ (CP) < 1501, 244 × 10–5 < (dp/D) < 10–1. Assuming (mf = 0.42, Wen and Yu33 proposed a simplified form of the above correlation as:

159 Re1mf,687 – 1060 Re mf – Ar = 0

(1.51)

The solution for Remf using Equation 1.50 or 1.51 is not straightforward. On the other hand, Remf can be obtained more easily by using appropriate expressions for different flow conditions as given by Equations 1.46 and 1.47. Using appropriate correlations33 for FD and δ, the following relationship can be arrived at:

⎛U ⎞ α=⎜ t⎟ ⎝U⎠

β

(1.52)

where β = 1 for laminar flow (0.001 ð Re ð 2), 1.4 for intermediate flow (2 < Re < 500), and 2.0 for turbulent flow (Re > 500). It may be recalled here that Richardson and Zaki34 experimentally obtained the relationship

Ut = –n U

(1.53)

which indicated that α is a function of voidage only. In light of the above analysis of the development of an expression for Re at Umf conditions, a variety of correlations can be obtained, depending on the range of Re and the corresponding equation for the drag coefficient. Most expressions available for predicting CD are based on experimental results pertaining to spherical particles, and no single correlation is able to cover the entire (laminar to turbulent) range of flow conditions. The right

© 1999 by CRC Press LLC

choice of CD for a nonspherical particle is often difficult. More recently, Haider and Levenspiel35 presented explicit equations for predicting CD for spherical and nonspherical particles. For spherical particles:

CD =

(

)

24 0.4251 1 + 0.1806 Re 0.6459 + + Re 1 6880 .95 Re) (

(1.54)

for Re ð 2.6 × 105. For nonspherical, isometric particles:

CD =

[

]

24 1 + 8.1716 (exp– 4.0655 φ s ) Re Re

[

]

(1.55)

+ 73.69 Re exp ( –5.0748 φ s ) Re + 5.378 exp(6.2122 φ s )

for Re ð 25,000. A more accurate correlation with rms Ý 3.1, as given by Turton and Levenspiel,36 is cumbersome. Equation 1.55 is considered to be a simple form wtih an rms error of 5%. The drag coefficient in the case of a porous spherical particle falls outside the ambit of the above type of correlations. Masliyah and Polikar37 reported from their experimental findings that the drag experienced by a porous sphere is less than that experienced by an impermeable sphere of the same diameter and bulk density. However, at high values of the Reynolds number, the effect of inertia on a porous sphere is found to be greater than that on a comparable impermeable sphere. The drag coefficient proposed by Masliyah and Polikar37 for a porous sphere is given by:

CD = 24

[

Ω K – K log Re 1 + K5 Re ( 6 7 10 ) Re

]

(1.56)

where

Ω=

[

2 ψ 2 1 – (tan h ψ ) ψ

[

]

2 ψ + 3 1 – (tan h ψ ) ψ 2

]

and Ψ = d p

k pe

The values of K5, K6, and K7 are 0.1315, 0.82, and 0.05, respectively (for 0.1 < Re < 7), and 0.0853, 1.093, and 0.105, respectively (for 7 < Re < 120). A list of correlations based on the drag force principle or with similar forms is given in Table 1.5. c. Pressure Drop Method The most general form of expression for the pressure drop through a fixed bed of particulate solids can be given as:

© 1999 by CRC Press LLC

Table 1.5 Correlations for Remf Based on Drag Force Principle or Similar Form as Derived from Drag Force Principle Ref.

Remarksa

K2 A. General Form for Low Remf Remf = K2 Ar 8.1 × 10–3

Rowe and Henwood38 Frantz39 Wen and Yu33 Davidson and Harrison21 Davies and Richardson40 Pillai and Raja Rao41

GF

1.065 × 10–3 –4.7/18 0.00081

Remf < 32, GF 0.001 < Remf < 2, GF GF

7.8 × 10–4

GF

7.01 × 10–4

Re < 20, GF

B. General Form for All Ranges of Re mf Remf = [(4/3) (Ar/α)/CD]1/2, where α = e–4.7 Schiller and Naumann32 Morse31

CD = 3/Remf + 0.45 Remf–0.313 CD = C1/Re + C2/Re2 + C3 C1

Haider and Levenspiel35

0.001 < Remf < 1000, for spherical particle For spherical particle C2

C3

Remf Range

24 22.73

0 0.0903

0 3.69

29.1667 46.5

3.8889 116.67

1.222 0.6167

98.33 148.62

2,778.0 4.75 × 104

0.3644 0.357

490.546 1,662.5 0 See Equation 1.54

57.87 × 104 5.4167 × 106 0

0.46 0.5191 0.44

Remf ð 0.1 0.1 ð Remf ð 1 1 ð Remf ð 10 10 ð Remf ð 100 100 ð Remf ð 1,000 1,000 ð Remf ð 5,000 5,000 ð Remf ð 10,000 104 Remf ð 5 × 104 Remf Š 5 × 104

Remf < 2.6 × 105 for spherical particle Remf < 25,000 for nonspherical particle, φ Š 0.67

See Equation 1.55

a

GF = general formula.

(∆P H ) = f (ρ U 2) 6 (1 – ) ( Φ d ) b

g

2 g

3

s

p

(1.57)

where the bed friction factor ( f ) is a function of bed voidage () and the particle Reynolds number (Re). At the incipient fluidization condition, taking ∆Pb /H = (ρs – ρg) (1 – mf)g, Equation 1.57 can be transformed to:

© 1999 by CRC Press LLC

Ar =

1 Re 2mf fk φ s 3mf

(1.58)

Ergun42 proposed that the friction factor (fk) for randomly packed beds can be expressed as:

fk =

150(1 – ) + 1.75 φ s Re

(1.59)

Substituting Equation 1.59 in Equation 1.58 at Umf condition and rearranging:

(

)

1.75 150 Re 2mf + 2 3 1 – mf ⋅ Re mf = Ar 3 φ s mf φ s mf

(1.60)

Wen and Yu33 were the first to use this type of correlation and to solve it for Remf. In order to arrive at a suitable solution, Wen and Yu collected the data for mf and φs and made the following approximations:

1 – mf φ 2s 3mf

⯝11

and

1 ⯝14 φ s 3mf

(1.61)

The Wen and Yu33 correlation expressed using Remf and Ar is

24.5 Re 2mf + 1650 Re mf = Ar

(1.62)

The above correlation has been found to fit experimental data well in the range 0.001 < Remf < 4000, and this form of correlation is cited in the literature frequently. The Wen and Yu33 correlation has been refined to increase the accuracy of prediction, and several correlations of such type have been introduced by altering the coefficients of Remf2 and Remf. The friction factor (ff) for a fluidized or sedimenting suspension can be expressed after Davidson and Harrison21 as: ff = 3(ρs – ρf)g/SρfU2

(1.63)

For a packed bed one has:

∆P ⎞ fp = ⎛ ⎝ SH 1 – ⎠

(ρ U f

2

2

)

(1.64)

Usisng Carman’s proposed equation43 for ∆P with a constant of 5, one obtains:

© 1999 by CRC Press LLC

⎡ 5(1 – )2 2 ⎤ 1 2 fp = ⎢ ⋅ S µ H U⎥ ⋅ ⋅ 2 3 ⎣ ⎦ SH 1 – ρ f U

(1.65)

The above friction factors have been evaluated, and correlations in terms of Rel based on a linear dimension analogous to the hydraulic mean diameter, i.e., /(1 – )S, have been reported as: ff = 3.36/ReL

for ReL 2 × 10 4

(1.103)

For elevated pressures and temperatures, the method of Yang et al.86 should be used; it is based on graphical methods using (Re2 · CD)1/3 versus (Re · CD)1/3 plots for various mf values that are functions of pressure and temperature. Having selected an appropriate correlation for Remf and then for Ret as outlined, it is possible to generate a plot of Ret/Remf versus Ar. Richardson59 utilized a similar

© 1999 by CRC Press LLC

kind of technique and generated a plot using the Ergun equation for spherical particles and taking mf = 0.4 to predict Remf. The following correlations were employed to predict Ret: Ar = 18 Ret Ar = 18 Ret + 2.7 Ret1.87 Ar = 1/3 Ret2

for Ar < 3.6

(1.104)

for 3.6 < Ar < 105

(1.105)

for Ar > 105

(1.106)

His plots for mf = 0.38, 0.4, and 0.42 fit well with experimental data and show that: Ret/Remf = constant

for Ar < 1

(1.107)

where the constant = 64 for mf = 0.42 = 78 for mf = 0.4 = 92 for mf = 0.38 and Ret/Remf = Ý10

(1.108)

at different values of mf (0.38, 0.4, 0.42) for Ar > 105.

V. FLOW PHENOMENA A. Particulate and Aggregative Fluidization Particulate or smooth fluidization, which is normally exhibited by a liquid–solid system, should be distinguished from aggregative- or heterogeneous-type fluidization that prevails in a gas–solid system. There is not much work available on this subject to enable one to distinguish between particulate and aggregate fluidization in a precise manner. Wilhem and Kwauk87 proposed a dimensionless group to explain the quality of fluidization as follows: Frmf < 0.13 for smooth or particulate fluidization > 1.3 for aggregative or bubbling fluidization

(1.109) (1.110)

where Frmf is the Froude number (Umf2 /dpg). It may be recalled here that Equation 1.62 for the determination of Umf, when used for inertial flow or Newton’s flow regime, results in the equation

© 1999 by CRC Press LLC

Re 2mf =

Ar 24.5

(1.111)

which can be simplified to:

⎛ ρs – ρ f ⎞ Frmf = 0.0408⎜ ⎟ ⎝ ρf ⎠

(1.112)

For most gases ρf < 10 kg/m3, and for liquids ρf > 500 kg/m3. For glass beads of ρs < 2500 kg/m3, when fluidized by gas, ρs/ρf > 250, and the values obtained are Frmf Š 10.1, Hence, the glass bead–gas fluidized bed system is expected to behave like an aggregative fluidization. Applying a similar rule for a water–glass bead system, one gets ρs/ρf > 2500/1000 and Frmf ⯝ 0.0612 (which is less than 0.13). Thus the system exhibits smooth fluidization. In the above examples, it is possible to obtain an expression for the Froude number directly by using the densities ρs and ρf for the flow regime dominated by the inertial force, more precisely for Rep > 1000. For Rep < 20, the viscous force is dominant and the expression for Frmf is obtained in terms of the density ratio (ρs – ρf )/ρf and Remf:

Frmf =

ρs – ρ f 1 ⋅ Re mf ⋅ 1650 ρf

(1.113)

For example, a quantitative determination of Frmf when Remf = 20 can be carried out using Equation 1.113. For the glass bead–air system (ρs = 2500 kg/m3 and ρf = 1.2 kg/m3) when Remf = 20:

Frmf =

20 ⎛ 2500 ⎞ ⋅ – 1 = 0.0121(2082.3) = 25.2 ⎠ 1650 ⎝ 1.2

Because Frmf > 1.3, an aggregative type of fluidization occurs for the glass bead–air system. Similarly, for the glass bead–water system (ρf = 1000 kg/m3), when Remf = 20: Frmf = 0.018 Because Frmf < 0.13, smooth fluidization is possible with the glass bead–water system. The very fact that the quality of fluidization cannot be assessed in terms of a single dimensionless group like the Froude number has prompted researchers to evaluate the quality of fluidization by using a larger number of dimensionless groups. Romero and Johanson88 proposed four dimensionless groups to assess the quality of fluidization:

© 1999 by CRC Press LLC

Frmf ,

Re mf ,

ρs – ρ f ρf

and

Hmf Dt

The quality of fluidization is related to its stability, which would be affected by bubbling. An increase in each of the four dimensionless groups will lead to poor quality or unstable or bubbling fluidization. From the experimental findings, the criterion for grouping the two modes of fluidization may be assessed by the product of the four dimensionless groups:

⎛ ρs – ρ f Hmf ⎞ < 100 ⋅ ⎜ Frmf ⋅ Re mf ⋅ ρ Dt ⎟⎠ ⎝ f

(1.114)

for particulate or smooth fluidization and:

⎛ ρs – ρ f Hmf ⎞ > 100 ⋅ ⎜ Frmf ⋅ Re mf ⋅ ρ Dt ⎟⎠ ⎝ f

(1.115)

for aggregative or bubbling fluidization. Classifying the mode of fluidization in terms of the product of four dimensionless groups seems to be superior to using the single Froude group because almost all variables are considered in the former. However, there is no theoretical proof available to validate the aforementioned criterion; therefore, the criterion for identifying the fluidization mode can be regarded merely as a useful rule of thumb. B. Regimes of Fluidization 1. Bubbling Bed In a gas fluidized bed, right from the onset of fluidization, a change in the fluidization regime or the bed behavior would occur with an increase in the gas flow rate. The first fluidization regime prevails when bubbles are formed in the bed from the attainment of a minimum gas velocity designated as the minimum bubbling velocity. Bubbles once formed in the bed start rising, grow in size, coalesce, reach the bed surface, and finally erupt. In a tall column with a small diameter, the bubbles coalesce as they rise up and form slugs whose size or diameter would be the same as the column diameter. Generally, the slugs are followed by a piston of solids; once carried up to the top surface of the bed, the solids rain down along the column wall. When a fluidized bed of a larger diameter is used for a bubbling fluidized bed, the formation of slugs can be avoided or minimized. Further, with a large-diameter fluidization vessel and shallow beds, bubbles have less of a chance to coalesce. The resulting bubbling fluidized bed would then have bubbles of more or less uniform size. The bubbling point varies depending on the powder group (as classified by

© 1999 by CRC Press LLC

Geldart12). Bubbling in a fine powder system is quite different from one comprised of coarse particles. With fine powders, the bed may expand uniformly without any bubble formation, and particulate fluidization can occur until the first bubble is formed at a gas velocity known as the minimum bubbling velocity (Umb). Experimental measurement of this velocity is not easy and requires careful assessment of the bed behavior. Just at the minimum bubbling point, the expanded particulate bed will show a decrease in its height without a change in the overall voidage of the emulsion phase. However, the structure of the bed is changed. The simplest correlation proposed by Geldart89 for the minimum bubbling velocity (Umb) is Umb = Kmb dsv

(1.116)

where Kmb = 100 if Umb and dsv are expressed in CGS units and dsv = 1/Σxidsvi is the particle surface/volume diameter (m–1). Abrahamsen and Geldart5 determined the ratio of the minimum bubbling velocity (Umb) to the minimum fluidization velocity (Umf) as:

(

0.126 0.523 exp 0.716 Ff U mb 2300ρ g µ = 0.934 U mf d p–0.8 g 0.934 ρs – ρ g

(

)

)

(1.117)

It is, in general, observed that the addition of fines ( 100

(1.136)

For Rep < 100, the contribution of the convective component is negligible and Nup = 2 when Rep tends to zero. The gas velocity relative to a moving particle is less, and this leads to a lower value of Nup in a fluidized bed than in a packed bed. Measurements of the temperatures of individual particles and of the gas are not possible by using a bare thermocouple, which can measure only the mean temperature of the emulsion packet. Therefore, experimental determination of Nup values is difficult. Two types of experimental techniques were proposed by Kunii and Levenspiel58 to predict the fluid–particle temperature, and these techniques are briefly discussed below. a. Steady State In this method, the heat lost by the gas stream is assumed to be equal to the heat gained by the solids. For example, the heat balance for a section of a bed of height dH is –CpgUgρgdTg = hgp(Tg – Tp)dH (Heat lost by gas)

(Heat gained by solids)

(1.137)

The value of the heat transfer coefficient (hgp) as determined by the above equation is only an apparent value, if axial conduction is not considered. In order to arrive at the true value, the right-hand side of the equation, that is, the heat gained by the solids, should be reduced by an amount equivalent to the axial conduction loss. b. Unsteady State In this method, the heat lost by the gas stream over a differential section of the fluid–solid mixture is considered to be the same as the heat accumulated by the solids. The final equation which could be used to estimate hgp by this method is

© 1999 by CRC Press LLC

ln( ∆T ∆Ti ) = –

C pgU g ρ g A ⎡ hgp H ⎤ ⎥t ⎢1 – exp W C ps ⎢⎣ C pgU g ρ g ⎥⎦

(1.138)

Experimental investigations by Chen and Pei126 on fluid-to-particle heat transfer in fluidized beds of mixed-size particles showed that the addition of coarse particles to a fine-particle system improves the fluidization and hence increases the fluid–particle heat transfer coefficient. The most effective concentration of the coarse particles is about 25% and the optimum particle diameter ratio is Ý7. The correlation proposed by Chen and Pei126 to predict fluid–particle heat transfer is jH (1 – ) = 0.000256 Ar0.4 Pr2/3

(1.139)

for 10 < Ar < 105 and 0.1 < Rep < 50 where jH = Colburn’s factor = Nu/Re Pr1/3. 2. Bed–Wall Heat Transfer The overall bed–wall heat transfer coefficient (hw) is more useful for design purposes than the local heat transfer coefficient. Heat transfer measured at various locations or at the same location at different times usually fluctuates. It is thus necessary to be able to estimate the local heat transfer coefficient in order to develop a suitable model that could help in predicting the overall mean value. In a gas fluidized bed, hw in principle increases with increasing gas velocity until it reaches a maximum value; thereafter hw falls with further increase in gas velocity. The trend in the plot of hw versus U is the same as that shown in Figure 1.10C. Thus, in a gas fluidized bed, at gas velocities beyond Umf, the plot of hw versus U has two sections: (1) a rising branch up to hmax, corresponding to a velocity designated as Uopt, and (2) a falling branch beyond U = Uopt. C. Models Various models have been proposed to explain the variation in hw with gas velocity, and this topic was reviewed in detail by Gutfinger and Abuaf,127 Bartel,128 and Saxena and Gabor.129 Among these models, three distinct groups may be identified, as indicated in the following sections. 1. Film Model A thin fluid film adjacent to the heat transfer wall is considered to offer the main resistance to heat transfer and this resistance is reduced by the moving particles which scour away the fluid film. Steady-state heat transfer is assumed. This model/hypothesis was supported by Van Heerden et al.130 and Ziegler and Brazelton.131 However, the results based on this model were subsequently found to be in poor agreement with many other experimental data. The drawback of this model lies in its failure to consider the properties of solid particles, which play a vital role in the transport of heat in a fluidized bed.

© 1999 by CRC Press LLC

2. Modified Film Model Transient conduction of heat between the surface and the particles (single/multiparticles)131–134 is assumed in a particulate type of fluidization. The solid particle properties are considered, and the heat transfer rate is assumed to change depending on the mobility of the solid particles and their concentration on the heat transfer surface. As this model cannot be applied to a bubbling bed, a third type of model, known as the packet model, has emerged. 3. Emulsion Packet Model A packet of emulsion (a mixture of solid and gas corresponding to the incipient state of fluidization) contacts the heat transfer surface periodically, undergoes transient conduction during its stay, moves away (making way for a new packet), and transports the sensible heat to the bulk of the bed. Unsteady-state approach and surface renewable rate of the packets are the main features of this packet. Mickley and Fairbanks135 were the first to propose a model of this kind. The emulsion packet model does not consider the individual properties of the gas and the solid. Instead, it considers the properties of the homogenous gas–solid mixture at the incipient state. Botterill and Williams,136 in their single-particle model, considered the individual properties of the gas and the solid. The single-particle model has the limitation of a short contact time during which heat cannot penetrate into a second particle; for this reason, this model has been extended to a two-particle layer by Botterill and Butt137 and further to a chain of particles of unlimited length by Gabor.138 D. Predictions of Heat Transfer Coefficient There are numerous correlations available in the literature for predicting the wall heat transfer coefficient, and these correlations were developed under different conditions. Zabrodsky,139 therefore, recommended that care be taken to use these correlations only under the conditions for which they were proposed. The applicability of any mechanism warrants detailed knowledge of the transient contact characteristics between the emulsion and the heat transfer surface on a local basis. The literature on this subject is scant. Selzer and Thomson140 pointed out that a model should clearly explain the heat penetration depth and the contact time criteria. They also stressed the need to test the applicability of these models for different types of distributors. A detailed listing of all the available correlations to predict the wall heat transfer coefficient can be found in the literature. In general, the heat transfer coefficient is a function of (1) the properties of the gas and the solid (ρg, Kg, Cpg, µg, ρs, Cps, Ks), (2) the relevant geometric factors (dp, Dt), (3) the fluidization parameters (θD, fL, p, b), (4) the system variables (U, Umf), (5) the location of the heat transfer surface (immersed) inside the bed, and (6) the type and design of the grid/distributor plates.

© 1999 by CRC Press LLC

1. Additive Components The wall heat transfer coefficient, also called the convective heat transfer coefficient, is generally constituted of two additive components: the particle-convective component (hpc) and the gas-convective component (hgc). The particle-convective component arises due to the motion or mobility of the particulate solids over the heat transfer surface and is more pronounced for fine particles than for coarse particles. The gas-convective component is due to the gas percolating through the bed and to gas bubbles. The gas-convective component is negligible in fine particle fluidization but is predominant in coarse or heavy particle fluidization; the reason is the requirement for high flow rates to achieve fluidization in the latter case. The overall convective heat transfer is given by: hc = hpc + hgc

(1.140)

Prediction of hpc is difficult due to the complex hydrodynamics of fluidization and the interaction of the lean and the dense phases over the heat transfer surface. a. Particle-Convective Component The particle-convective component (hpc), in the absence of any film resistance, is hp and should be evaluated from the knowledge of the instantaneous heat transfer coefficient (hi) and the age distribution of the packet of emulsion over the transfer surface. Mickely and Fairbanks135 proposed a model to evaluate hp as:

hp =

1 τ

t

∫ h dt 0

(1.141)

i

where the instantaneous heat transfer coefficient (hi) is given by: hi = (KeρeCps/πt)

(1.142)

and the packet residence time (τ) is given by:

⎤ ⎡ ⎥ ⎢ dpg ⎥ τ = 0.44 ⎢ 2 ⎢ ⎞ ⎥ ⎛ U 2 ⎢ U mf ⎜ – 1⎟ ⎥ ⎢ ⎠ ⎥ ⎝ U mf ⎥⎦ ⎢⎣

0.14

dp Dt

(1.143)

If the conditions correspond to slugging, the packet residence time (τ) over a short vertical surface of length L is

© 1999 by CRC Press LLC

τ = L/(U – Umf)

(1.144)

Under all practical conditions, the heat transfer is effective only when the packet contacts the surface. For the fraction of surface which is shrouded by gas bubbles (fb), hp is negligible. Hence, hpc, in the absence of any film resistance, can be expressed as: hpc = hp(1 – fb)

(1.145)

where

(

) ⎤⎥

⎡U 2 U U – A mf mf fb = 0.33⎢ dpg ⎢ ⎣

0.14

(1.146)

⎥ ⎦

The accuracy of the determination of the average value of hpc can be improved to better than 5% by taking into consideration the film resistance (1/hf) which is in series with the packet resistance (1/hp). When this is done, the particle-convective component of heat transfer (hpc) is given by:

hpc =

1

( ) ( ) 1 hp + 1 h f

(1 – f )

(1.147)

b

The film heat transfer coefficient (hf) is given by: hf = mcKg/dp

(1.148)

where mc has varying values; the range of these values under different experimental conditions was enumerated by Xavier and Davidson.141 They also suggested that mc = 6 may be taken for design purposes. In order to evaluate hpc, it is necessary to know the effective thermal conductivity of the particulate bed (Ke), which is the sum of the effective thermal conductivity of the bed when there is no fluid flow ( Keo ) and the conductivity due turbulent diffusion ( Ket ). Thus,

Ke = Keo + Ket

(1.149)

Keo can be evaluated142,143 by using the expression Keo ⎛ K s ⎞ = K g ⎜⎝ K g ⎟⎠

© 1999 by CRC Press LLC

0.28– 0.757

log10 –0.057 log10

Ks Kg

(1.150)

The contribution of turbulent heat diffusion ( Ket ) in determining the effective thermal conductivity is given by:

Ket K g = (α1 , β1 ) Re p Pr

(1.151)

Here, α1 is the ratio of the mass velocity of the fluid in the direction of heat flow to the superficial mass velocity in the direction of fluid flow. The parameter β1 is the ratio of the distance between the two adjacent particles to the mean particle diameter. The product α1β1 for small values of d/D is equal to 0.1 for normal packing of spheres. Hence, the effective conductivity of the particulate phase or the emulsion packet is

Ke = Keo + 0.1 ρ g C pg d pU mf

(1.152)

where the value of Umf can be predicted as per the correlations presented in Tables 1.4–1.6. The particle-convective heat transfer coefficient can be calculated from Equation 1.147 by using the appropriate expression for hp (Equations 1.141–1.146 and Equation 1.148). b. Gas-Convective Component Because the interphase gas-convective component (hgc) cannot be determined directly, Baskakov et al.145 proposed from an analogy to mass transfer that:

Nu gc =

hgc d p Kg

= 0.009 Ar 1 2 Pr 1 3

(1.153)

For most gases, Pr is constant and for the inertial flow regime Remf α Ar1/2. Hence, Nugc α Remf. In view of this, it has been assessed that the value of hmf obtained at Umf is approximately the same as that of hgc. However, for a bubbling fluidized bed at a high fluid flow rate, the equation hgc = hmf(1 – B) + hBB

(1.154)

can be approximated to hgc Ý hmf. Botterill and Denloye146 proposed the following correlation for predicting hgc by taking into account the heater length (L):

hgc dL K g = 0.3 Ar 0.39

for 10 3 < Ar < 2 × 10 6

(1.155)

c. Radiative Component The radiative heat transfer coefficient (hr) between a fluidized bed at a temperature Tb and a heater surface like an immersed tube at a temperature Ts is

© 1999 by CRC Press LLC

(

)

hr = qr (Tb – Ts ) = σ s Ebs Tb2 + Ts2 (Tb + Ts )

(1.156)

where qr is the radiant heat flux, σs is the Stefan–Boltzmann constant (5.67 × 10–8 W/m2 K4), and Ebs is the generalized emissivity factor given by: Ebs = (1/Eb + 1/Es – 1)

(1.157)

The generalized emissivity factor depends on the shape, disposition, and emissivity of the radiating and receiving bodies. Radiation plays a major role in heat transfer when the temperature of the radiating body is above 800–900°C. The convective heat transfer coefficient increases due to the component hr, which is significantly high when the temperature of the radiating body is above 1000°C. In the case of many a fluidized bed combustion, the bed temperature (Tb) itself is around 1000°C. At high temperatures or at high values of the difference between Tb and Ts, it is important to evaluate the intermediate mean temperature at which the physical properties of the fluid and the solid must be determined. The intermediate temperature147 (Te) is given by:

Tb – Te Tb – Tw = 0.5 Rr 0.5 Rr + Rw

(1.158)

Rr is the mean thermal resistance of the packet, that is,

Rr =

πt Ke C ps ρ B

where Cs = Cp (1 – mf), ρB is the bulk density of the packet = ρbmf, and Rw is the contact resistance of the zone near the wall and equals δw/Kew. 2. Overall Heat Transfer Coefficient The overall heat transfer coefficient is the sum of the individual convective components that are due to bubbles (hb), particles (hpc), interphase gas (hgc), and radiant heat transfer (hr). Thus, the overall or total heat transfer coefficient (h) is given by: h = hb fb + (hpc + hgc) (1 – fb) + hr

(1.159)

In the case of a dense fluidized bed where there are no bubbles and the fraction of the immersed surface covered by bubbles (fb) is zero, the total heat transfer coefficient (h) is given by: h = hpc + hgc + hr

© 1999 by CRC Press LLC

(1.160)

For large particles, due to the requirement of a high fluidization velocity, it can be assumed that hb Ý hgc, so that: h = (1 – fb) hpc + hgc + hr

(1.161)

In the case of solid particles whose emissivity lies in the range of 0.3–0.6, Baskakov148 proposed the following expression for the overall heat transfer coefficient under maximum condition:

(

)

hmax = K g 0.85 Ar 0.19 + 0.006 Ar 0.5 Pr 0.33 d p + 7.3 E p Es Ts3

(1.162)

E. Heat Transfer to Immersed Surfaces The immersed surfaces inside a fluidized bed can be of any object meant for transferring the heat either from the bed or to the bed, depending on whether the bed is to be cooled or heated. Generally, tubes or tube banks are immersed in largescale fluidized beds for exchanging the heat. Leva18 reported that, except at high flow rates of the fluidizing fluid/gas (that is, for Umf >> 1), the heat transfer rate obtainable with immersed surfaces is fourfold higher than that between the bed and the external wall. There have been many investigations1,7,149–156 into this subject. The heat transfer coefficients relevant to heat transfer to immersed objects are broadly of two types: local heat transfer coefficients and total or overall heat transfer coefficients. Because the latter coefficients are useful for design purposes, this topic is given special emphasis in the following discussion. 1. Vertical Surfaces Studies135 on vertically immersed electrical heaters have shown that the heat transfer coefficients are smaller than those corresponding to horizontal surfaces under similar experimental conditions. This has been attributed to the small size of the heater over which large bubbles pass during their upward flow. The correlations often cited in the literature for predicting the wall-to-bed heat transfer coefficient for vertically immersed surfaces are those of Vreedenberg157,158 and Wender and Cooper.159 2. Horizontal Surfaces Horizontally immersed tubes in a fluidized bed are exposed to more cross-flow of solids than are vertical tubes, and thus relatively higher heat transfer rates could be possible with these. However, a single horizontal tube poses some problems, mainly due to the hydrodynamic environment around a single horizontal tube whose top portion is piled with stagnant solid particles that reduce the local heat transfer rate at the top. On the other hand, the bottom portion of the tube becomes shrouded with rising bubbles, creating a solid free zone. This condition is opposite that prevailing at the top surface of the tube. Lateral parts of the tubes are frequently

© 1999 by CRC Press LLC

contacted by solid particles, and hence the maximum local heat transfer occurs around this lateral region. Many reports are available in the literature160–162 on the measurement of local heat transfer coefficients. In a gas fluidized bed, the minimum occurs at the top or the upward face of a horizontally immersed tube (i.e., in the downstream side of the flow). The maximum162,162 has been found to occur on the lateral side, which tends to shift toward the downward direction in a large-diameter tube. In a liquid fluidized bed, unlike a gas fluidized bed, a uniform temperature around a horizontally immersed tube was observed by Blenke and Neukirchem.163 In order to improve the heat transfer or to arrive at a uniform temperature around the tube, the use of tube bundles is recommended. Heat transfer from a horizontal tube is lowered by the presence of unheated tubes placed on one side or both sides of a heated tube. For a horizontally immersed loop type of heater, heat transfer is poor if it is a single tube.164 This has been attributed to the stable void of gas between the tubes of a loop-type heater, which would disappear in the case of tube bundles.165 Generally, it has been observed164,165 that with a well-immersed tube bundle, the heat transfer obtained is independent of the location of the bundle inside the fluidized bed. There are numerous correlations157,167–172 available in the literature for predicting the total heat transfer coefficient for an immersed horizontal tube. The correlation proposed by Vreedenberg157,158 is widely accepted. Subsequently, the need to incorporate the volumetric heat capacity of the solid and an appropriate correction for extending the application of the correlation to tube bundles were suggested by Saxena and Grewal.172 F. Effects of Operating Variables 1. Effect of Velocity a. Heat Transfer Coefficient versus Velocity The effect of velocity on heat transfer was discussed in Section IV. It was also mentioned that there exists an optimum velocity at which the heat transfer coefficient is maximum (hmax). A typical plot depicting the heat transfer coefficient as a function of the fluidizing velocity is shown in Figure 1.23 at a bed temperature of 600°C. A list of the variables that influence the heat transfer in a gas fluidized bed is given in Table 1.11. Among the various variables, bed temperature and pressure are important in studies on advanced fluidized bed reactors for coal combustion or fluidized bed boilers. These effects have not been examined in depth for gas–solid reactions in mineral or materials processing. b. Flow Regime Effect Molerus185 recently gave a detailed account of the effects of gas velocity on the heat transfer coefficient for three different flow regimes: laminar, intermediate, and turbulent. A typical plot presented by Molerus is shown in Figure 1.24 for the

© 1999 by CRC Press LLC

Figure 1.23 Heat transfer coefficient as a function of the fluidizing velocity. (dp = 0.55 mm, Ho = 57–60 cm. For U < Umf , Tp = 43–52°C, Tb = 600°C; for U > Umf , Tp = 98–134°C.) (From Draijer, W., in Fluidized Bed Combustion, Radovanovic, M., Ed., Hemisphere Publishing, Washington, D.C., 1986, 211. With permission.)

powders classified by Geldart.12 Curve 1 in Figure 1.24 corresponds to the laminar flow regime (i.e., laminar flow Ar number, Al = [(dp3 g)0.5(ρs – ρg)/µg = 159]); this curve also corresponds to a typical Geldart Group A powder for which bubbles flow at low velocity. Hence, hmax is attained only at a high velocity, about three times greater than that obtainable with Group B powders (especially in the intermediate flow regime). Curve 2 is for the intermediate flow regime (e.g., 103 ð Ar ð 105) and is typical of Group B powders for which the advantage of unlimited bubble causes h to rise steeply just after Umf and brings about the attainment of hmax at 2Umf. In the intermediate flow regime, both particle- and gas-convective components are important in determining the overall heat transfer coefficient. Curve 3 in Figure 1.24 corresponds to the turbulent flow regime (Ar Š 105), where gas-convective heat transfer is predominant. Hence, the transition from a fixed bed to a fluidized bed does not have a significant effect on the h values in this case. Borodulya et al.186 reported that the maximum heat transfer coefficient increases exponentially at elevated temperatures and varies linearly at high pressures. 2. Optimum Velocity The occurrence of a maximum in the heat coefficient could be anticipated because of the opposing effects of the particle velocity and the bed voidage, both of which increase with the fluidizing velocity. The attainment of a maximum heat transfer coefficient and a corresponding optimum velocity has been confirmed by numerous research workers, and their results show that the maximum heat transfer depends mainly on fluid–solid properties. A list of the correlations that can be applied to gas–solid fluidized beds to predict the maximum heat transfer coefficient and the corresponding optimum fluidizing gas velocity is given in Table 1.12. Although there are many correlations for Numax or hmax, they are not, in principle, functions of the optimum velocity because Uopt is also a function of gas–solid properties. Figure 1.25 shows the plot of the variation in Reopt with the Galileo number for a wide range of values of the latter (Archimedes number and Galileo number are the

© 1999 by CRC Press LLC

Table 1.11 Influence of Various Parameters on Heat Transfer in Gas Fluidized Bed Sl. no.

Variable

1

Fluid velocity (U)

2

Particle diameter

3

Thermal conductivity of solid (Kp) Specific heat of solid (Cp) Specific heat of fluid (Cf)

4 5

6

Thermal conductivity of fluid (Kf)

7

Fluid bed height (H)

8

Fluidized grid zone

9

Fluidized bed diameter

10

Length of heat transfer surface (L) Heat transfer tube diameter (dt) Vertical versus horizontal heat transfer tubes

11 12

13 a

b

Tube bundles Vertical

Horizontal

14

Bed temperature

15

Bed pressure

© 1999 by CRC Press LLC

Influences

Ref.

Heat transfer increases above Umf up to an optimum velocity (Uopt) and then decreases Heat transfer coefficient (h) increases with fine-sized particle and decreases with coarse size; for large size of particle, h increases mainly due to increase in convective component in heat transfer at high fluid velocity No influence on h

173

h is proportional to Cpn where 0.25 < n < 0.8

176, 177

Data are contradictory at moderate pressure and velocity; at high pressures, h is increased by Cf or the volumetric specific heat Cf ρf h α Kfn , n = 0.5–0.66; as bed temperature increases, h increases, attributed mainly to increase in Kf For a well-developed fluidized bed and a wellimmersed heat transfer surface, h is not dependent on H Grid zone affects h depending on grid type and its free area; for low free area, h is relatively higher than high free area grid No qualitative or quantitative information available h is independent of L

178, 179

h increases with decrease in dt

179, 183

h for vertical tubes is 5–15% higher than for horizontal tubes; construction and technological condition decide the tube orientation

152, 180

Ratio of tube spacing to tube diameter affects h; when the ratio is reduced, hmax drops; for closely packed bundles, h reduces even by 35–50% Almost similar effect as vertical bundle; for same ratio of tube spacing to diameter, horizontal bundles occupy greater portion of bed cross-section Gas-convective component increases for small (dp < 0.5 mm) particle and decreases for coarse (dp > 0.5) particle; particle-convective component decreases; for 1-mm particle, net effect is a reduction Particle-convective component not affected; gas-convective component is enhanced and is proportional to square root of gas density

152

173

174, 175

180

159, 176, 181 182

152

152

184

141

Figure 1.24 Comparison of measurements and predictions for heat transfer versus superficial gas velocity (fluidizing agent helium). Curve 1: A powder, curve 2: B-powder laminar flow region, curve 3: intermediate Archimedes number, curve 4: turbulent flow region, curve 5: B-powder laminar flow region. (Al is laminar flow Archimedes number. Al = (dp3g)1/2(ρs – ρg)/µg.) (From Molerus, O., Powder Technol., 15, 70, 1992. With permission.)

same). The correlations for predicting Reopt are also given in Table 1.12. The best fit of experimental data was made by Sathiyamoorthy et al.182 for predicting the optimum Reynolds number (Reopt) with respect to the maximum heat transfer coefficient; the correlations proposed are Reopt = 0.00812 Ar0.868 Reopt = 0.13 Ar0.52

for 1 ð Ar < 1000

for 3000 ð Ar < 107

(1.163) (1.164)

The experimental results of Sarkits173 show a strong inverse dependence for hmax on the diameter of fine particles (i.e., for dp < 2.5 mm) and a weak direct dependence on the diameter of coarse particles (2.5 ð dp ð 4.5 mm). 3. Distributor Effects The gas–solid mixing in a fluidized bed is influenced by the distributor type and its design. Many research workers do not mention the types of distributors used in their studies. However, Baerg et al.188 used perforated plate-type and Sarkits173 used screen-type distributors. Saxena and Grewal201 found that the free open area of a distributor affects the maximum heat transfer coefficient and also the corresponding optimum gas velocity. Sathiyamoorthy et al.202 confirmed that the free area in a

© 1999 by CRC Press LLC

Table 1.12 Correlations for Predicting Maximum Heat Transfer Coefficient and Optimum Reynolds Number in Gas Fluidized Bed Correlations hmax = 33.7 ρ Kf dp Numax = 0.86 Ar0.2 0.2 s

0.6

–0.36

hmax = 239.5 [log(7.05 × 10–3 ρSB)]/dp Numax = 0.64 Ar0.22 Z/dt Numax = 0.0087 Ar0.42 Pr0.33(Cps/Cpf)0.45 Numax = 0.019 Ar0.5 Pr0.33(Cps/Cpf)0.1 Numax = 0.021 Ar0.4 Pr0.33(D/dp)0.13(Ho/dp)0.16 hw,max = ho(1 – ) [1 – exp(–p kf)]

Remarks — Reopt = 0.118 Ar0.5 ρSB = nonfluidized bed density Z = distance between axes of tubes, dt = heater diameter Reopt = 0.12 Ar0.5 (laminar region) Reopt = 0.66 Ar0.5 Reopt = 0.55 Ar0.5

0.8 Pr0.4dp–0.69 Numax = 0.0017 Reopt 0.423 Ar0.14 Pr1/8 Numax = Reopt Numax = 0.3 Ar0.2 Pr0.4

ho = constant, p = constant Reopt = 0.20 Ar0.52 Reopt = 0.09 Ar0.58 1 < Ar ð 220

Numax = 0.0843 Ar0.15

103 ð Ar ð 106

Numax = 1.304 Ar0.2 Pr0.3 Numax = 0.064 Ar0.4 Numax = 2.0 (1 – ) (Ar d12.7/dt)0.21 (Cps/Cpg)45.5 Ar–0.7 Numax = 0.85 Ar0.19 + 0.006 Ar0.5 Pr0.33 + dp/kg (7.3 σ EpEsTs3) Numax = α Arβ Pr1/3 (Cvs/Cvg)1/8

Reopt = 0.113 Ar0.53 5 × 104 ð Ar ð 5 × 108 75 < Ar < 20,000

Numax = 0.14 Ar0.3(dp/dt)0.2(ρp/ρo)–0.07 0.25 Pr0.33 (C /V )0.09 Numax = 0.99 Reopt vs vg

Numax = const. Fr0.42(α · β · Remf)0.5(Reopt/Remf)Pr0.3 1. Overall a. Numax = 0.13 Al0.6 α–1 Al ð 300 b. Numax = 0.54 Al0.34 α–1 Al Š 300

2. Particle convective a. Nupc = 0.69 (Al ∆ρ/ρg)0.1 α–1 b. Nupc = 9 α–1 3. Gas convective a. Nugc = 0.4527 Ar0.2323 Pr0.33 b. Nugc = 0.024 Ar0.4304 Pr0.33

© 1999 by CRC Press LLC

100 < dp < 5000 µm for horizontal tubes and plates For 20 < Ar < 20,000, α = 0.074, β = 020; for 2 × 104 < Ar < 107, α = 0.013, β = 0.37 Heat transfer between particle (p) and freely circulating object (o), φ = shape factor Reopt = 0.00812 Ar0.868 1 ð Ar ð 3000 α = (1 – mf)/mf , β = Cvs/Cvg, const. = 0.0755 when Fr = Umf2 /(g · dt ), const. = 0.0144 when Fr = Umf2 /(g · dt ) Al = laminar flow, Ar = Ðdp3g ∆ρ/µ, α = [1 + a + kg/(2Cpµ)], a = constant to take care of additional resistance due to thick and porous oxide layer over metal powder; 0 for ceramic, 0.5 for Cu

Ref. Zabrodsky123 Varygin and Martyushin187 Baerg et al.188 Gelprin et al.189

Sarkits173 Sarkits173 Traber et al.178 Jakob and Osberg190 Chechetkin191 Ruckenstein192 Richardson and Shakiri193 Botterill and Denloye146 Kim et al.194 Brodulya et al.186 Grewal and Saxena195 Baskakov and Panov196 Chen and Pei197

Palchenok and Tamarin198 Sathiyamoorthy et al.182 Sathiyamoorthy and Raja Rao199

Molerus185, 200

103 ð Ar ð 105 Ar Š 105

Molerus185, 200

103 ð Ar ð 105 Ar Š 105

Molerus185, 200

Figure 1.25 Variation of optimum Reynolds number (Reopt) with Galileo number (Ga) (Ga is same as the Archimedes number). (From Sathiyamoorthy, D., Sridhar Rao, C.H., and Raja Rao, M., Chem. Eng. J., 37, 149, 1988. With permission.)

multiorifice distributor affects the value of hmax, which increases as the distributor freeflow area is reduced. Many correlations for predicting Numax or hmax can be found in the literature, and a list of these is given in Table 1.12. A model based on first principles and incorporating all the possible parameters for the prediction of hmax has yet to evolve. A model for predicting the maximum heat transfer coefficient which incorporates the distributor parameters and defines the stable state of fluidization was developed by Sathiyamoorthy and Raja Rao,199 and the relevant correlations are given in Table 1.12. Although many correlations are available for predicting hmax, the choice of the most appropriate correlation depends on the specific requirements and the designer’s skill. G. Heat Transfer in Liquid Fluidized Beds 1. Differences with Gas–Solid Systems Few studies have been carried out on heat transfer in liquid fluidized beds, unlike gas fluidized beds. In contrast to the latter, heat transfer in liquid fluidized beds increases with increasing particle size. Liquid fluidization is generally particulate in nature. The mixing process and the heat transfer characteristics of a liquid fluidized

© 1999 by CRC Press LLC

bed are not exactly similar to those associated with a gas fluidized bed. For a given temperature difference, the quantity of heat transferred through a liquid (film) to a solid particle could be thousands-fold higher than what could be transferred through a gas film.203 2. Heat Transfer In a liquid fluidized bed, the temperature gradient extends more into the bed than it does in a gas fluidized bed. According to Wasmund and Smith,204 the ratio of bed resistance to total resistance increases with increasing solid concentration and decreases with increasing particle size for a constant bed porosity. They suggest that the particle-convective component and the effect of the thermal conductivity of the solid particles are negligible. Generally, it has been established that heat transfer in a liquid fluidized bed is influenced by bed voidage. At constant voidage, an increase in particle size increases heat transfer. On the basis of experimental data, Romani and Richardson205 proposed a correlation that has limited application. The mechanism of heat transfer in a liquid fluidized bed was analyzed by Krishnamurthy and Sathiyamoorthy206 with the aim of investigating the reason why the heat transfer rate should increase with increasing particle size in liquid fluidized beds. Krishnamurthy and Sathiyamoorthy206 tested a model and arrived at the conclusion that for fine-sized particles in a liquid fluidized bed, the film thickness on the heat transfer surface increases. In a typical case, it was found that the film thickness varies from 1.6 to 36% for just a twofold increase in particle size but when liquid velocity is less than 10 cm/s. Particle size seems to have a relatively greater influence compared to liquid velocity in altering the magnitude of the heat transfer rate. Although the applicability of the few available correlations for practical uses is limited, the correlation of Mishra and Farid207 may be used for purposes of prediction: Nu = 0.24 (1 – ) –2/3Re0.8 Pr0.33

(1.165)

Heat transfer in liquid fluidized beds has been shown to have enhanced rates208 (as much as fourfold) on pulsating the fluidization. H. Heat Transfer in Three-Phase Fluidized Bed 1. Heat Transfer Coefficient Heat transfer data on three-phase fluidized beds have not been published extensively and only limited information is available. Studies on the heat transfer between a fluidized bed and a heat transfer surface were reported by Østergaard,209 Viswanathan et al.,210 Armstrong et al.,211 Baker et al.,212 and Kato et al.213 In general, all these studies reveal that the extent of heat transfer increases with the gas flow rate and is higher than in corresponding gas–liquid (bubble column) or liquid–solid (two-phase fluidization) systems. Heat transfer coefficients obtained in an air–water–glass-bed system have been found to increase with the gas flow rate. Figure 1.26 shows the results on the

© 1999 by CRC Press LLC

dependence of the heat transfer coefficient on the gas flow rate, and it can be seen that the trend is similar to that associated with a bubble column or a bubbling fluidized bed. Heat transfer coefficients are very high in three-phase fluidized beds (or the order of 4000 W/m2 K) and are reported to be two to three times greater than those attainable even with efficient processes such as boiling and condensation.

Figure 1.26 Heat transfer coefficient in three-phase fluidized beds. Air–water–glass beads. Data of Armstrong et al.211 (From Armstrong, E.A., Baker, C.G.J., and Bergognou, M.A., in Fluidization Technology, Keairns, D.L., Ed., Hemisphere Publishing, New York, 1976, 453. With permission.)

2. Particle Size Effect Heat transfer increases for particle diameters greater than 1 mm (Armstrong et al.211). With smaller particles in a bed of 0.12 m ID, Kato et al.213,214 observed a local maximum followed by a minimum during the transition from a fixed bed to a fully fluidized bed. This type of behavior is attributed to the heterogeneity of fluidization with fine particles when working at low liquid velocities.215 No such behavior was reported by Kato et al.214 for beds of coarse particles. 3. Correlation The wall heat transfer coefficient can be estimated by using the correlation of Kato et al.,214 which is given as:

© 1999 by CRC Press LLC

⎡ d pU1ρ1 C pl µ l 1 ⎤ h ⋅ 1 = 0.44 ⎢ ⋅ ⎥ k1 1 – 1 ⎦ Kl 1 – 1 ⎣ µ1 dp

0.78

⎡ Ug ⎤ + 2⎢ ⎥ ⎢⎣ gd p ⎥⎦

0.17

(1.166)

This correlation is recommended for a wide range of liquid–phase Prandtl numbers. The literature on three-phase fluidized beds with bed internals such as cooling or heating coils/pipes and studies on high velocities is scant.

VIII. MASS TRANSFER A. Introduction Fluidized beds in general are considered to be in transition between the states of packed bed and pneumatic transport. As a result, the correlations for the transport processes in all three units (viz., packed bed, fluidized bed, pneumatic transport reactor) are almost similar in type. Mass transfer data or the driving forces in a fluidized process must be evaluated quantitatively to arrive at a sound design of the reactor. Processes that are carried out in a fluidized bed can be either surface area based (like coating and agglomeration) or volume based (like crystallization and freezing). The transport processes of the surface-area-dependent type are complex in the sense that data on the probability of particle-to-particle or particle-to-object interaction are difficult to obtain. However, in volume-based processes, adequate information on the transport steps is available. In a fluidized bed, the particles act as turbulence promoters to enhance mass transport and reduce the hydrodynamic boundary layer. In a heat transfer process, however, the particles perform an extra function by carrying the heat. The conventional approach of drawing an analogy between heat transfer and mass transfer is valid for a fluidized bed also due to the fact that the particle-to-particle or particle-to-object contact time is low and is not much different for the heat and mass transport steps. However, there is some complexity in fluidized beds in which segregation (i.e., aggregative or bubbling fluidization) occurs. B. Mass Transfer Steps The mass transport steps in a fluidized bed can be broadly regarded to be (1) transfer between fluidized bed and object or wall, (2) transfer between particle and fluid, and (3) transfer between segregated phases (e.g., lean to dense phase). 1. Mass Transfer Between Fluidized Bed and Object or Wall a. Correlations The jM factor in mass transfer studies is defined as: jM = St Sc2/3 = (Kgw/u) Sc2/3 = constant Re–p

© 1999 by CRC Press LLC

(1.167)

where Kgw is mass transfer between a fluidized bed and an object or wall. The above equation can be used for a particulate type of fluidized bed by considering the bed as having several channels formed by the irregular contour of the wall of the particles. In the specific case of mass transfer between a fluidized bed and the wall or an object, the transfer can be viewed to occur between the two walls of a channel, one made up of the particle layer itself and the other formed by the object. The hydraulic diameter of the channel (dH) is

dH =

d p 6 = S (1 – )

(1.168)

The velocity of the fluid through the channel is assumed to be proportional to the interstitial velocity (u) (i.e., U/). If the Reynolds number in Correlation 1.167 is expressed in terms of U and dH, then,

K gw U

Sc

23

⎡ U dp ρ ⎤ = constant ⎢ ⎥ ⎣ µ(1 – ) ⎦

–p

(1.169)

The constant and the exponent p in Equation 1.169 vary depending on the state and the type of the bed. Typical values are tabulated in Table 1.13. Table 1.13 Magnitude of Constant, Bed Voidage ( ), and the Exponent (m) of the Correlation (Kgw/U) Sc2/3 = Const [Udp /µ (1 – )]–p Ref.

Constant in Equation 1.169

Packed

223

0.35–0.45

0.7

0.4

Liquid fluidized bed Gas fluidized bed

224 225

0.5–0.9 0.5–0.95

0.43 0.7

0.38 0

Bed Type

m

Re

Sc

50–500

0.9–218 450–3,700 200–24,000 1,300 300–12,000 2.57

b. Influencing Parameters Most studies on mass transfer do not provide useful data on voidage (), and hence comparison of results is often difficult. The degree of particle turbulence in a gas or liquid fluidized bed is not the same even if an average dynamic similarity is achieved. In addition to particle mean velocity and porosity, data on turbulence are necessary for an accurate prediction of mass as well as heat transfer coefficients. Measuring techniques play a key role. For example, mass transfer data can be obtained by simple experimental techniques more easily in a liquid than in a gas fluidized bed. To obtain reproducible results, gas fluidized beds operated at incipient fluidization are preferred.

© 1999 by CRC Press LLC

c. Role of Voidage It is useful for the data on voidage to correspond to an optimum situation where mass and/or heat transfer rates are high. The effect of the operating velocity on attainment of the maximum heat transfer coefficient was explained in the preceding section, and various correlations for predicting Uopt were presented. By analogy, one can choose the value of Uopt for heat transfer for mass transfer also. However, data on opt are also required, as can be seen from Equation 1.169. The voidage at Uopt does not have a unique value. In fact, the occurrence of the maximum does not correspond to any particular velocity, as pointed out in the section on heat transfer. In other words, the maximum can occur over a range of U. The voidage226 at the optimum condition lies in the range 0.6–0.75. In many industrial practices, attainment of maximum conversion and minimization of effluent heat content and convective heat loss are more important than Uopt values. The various useful ranges of values for opt, Uopt/Umf, and Uopt/Ut and the range of values for the drag coefficient (CD) for a single spherical particle were discussed by Beek.227 These data are presented in Table 1.14. It can be seen from the table that the useful operating velocity for attaining the maximum mass (heat) transfer rate falls in the range of three to five times the minimum fluidization velocity, and these values are used in many designs as a rule of thumb. Table 1.14 Optimum Parameter for Mass Transfer for Various Drag Coefficients ( CD) of a Single Spherical Particle opt (0.65 ± 0.08) CD (0.8 ± 0.04)

Uopt/Umf 1/6

(5 + 0.5) CD 36

Uopt /Ut 3/4

(0.3 + 0.1) CD 0.6

CD 1/4

10

2. Mass Transfer Between Particle and Fluid a. Comparison of Mass Transfer from Single Particle and Fixed Bed to Fluid A comparison of correlations for mass transfer from a solid particle to a flowing fluid (Kpl) can be made from Table 1.15 for three cases: (1) a single sphere moving through a fluid, (2) a fixed bed, and (3) a liquid fluidized bed. For large particles (Reynolds number, Rep > 80), it can be generalized from the data given in Table 1.15 that: Kp, fixed bed > Kp, fluidized bed > Kp, single sphere

(1.170)

The constant 2 in all the correlations (Table 1.15) is due to molecular diffusion and represents the theoretical minimum when the fluid is stagnant (i.e., Rep = 0). For a fixed bed of fine solids, measurement of mass transfer parameters is difficult due to rapid attainment of equilibrium brought about by the large surface area of the particles. For industrial-scale liquid fluidized beds, particulate fluidization need not necessarily prevail.

© 1999 by CRC Press LLC

Table 1.15 Correlation Comparison for Particle-to-Fluid Mass Transfer Coefficient (Kp) Condition Single sphere to fluid

Fixed bed

Liquid fluidized bed

Correlations for Sherwood no. (Sh) 2 + 0.65Sc1/3 Re1/2 Re = dp Uo ρf /µ Sc = µ/ρf 2 + 1.85 C1/3 Rep1/2 2 + 1.5 Sc1/3 [(1 – )Rep]1/2

Remarks Particle moving through the fluid at relative velocity (Uo) For coarse solids (i.e., Rep > 80), Uo is superficial velocity of fluid For laboratory-size bed exhibiting particulate fluidization, 5 < Rep < 120 ð 0.84

Ref. 228

229

230

b. Complexity in Measurement In a gas fluidized bed, measurement of the mass transfer coefficient poses practical problems due to features like high surface area, bubble formation, backmixing, and gas bypassing. Attainment of equilibrium for a gas inside the bed is so rapid that the gas has to pass through a bed of a height equal to only a few particle diameters to reach equilibrium. The height required to reach equilibrium is often referred to as the height of a transfer unit. In order to measure the mass transfer coefficient in a gas fluidized bed, a very shallow bed is necessary because there is no bubble generation and gas backmixing or bypassing in shallow beds. It is not possible to attain truly ideal conditions in a fluidized bed. The mass transfer data of Rebnick and White231 and Kettenring et al.232 were analyzed by Kunii and Levenspiel;233 they showed that a sharp decrease in the Sherwood number occurs when the Reynolds number is lowered. The decrease in the Sherwood number is lower than that obtained under corresponding conditions in a fixed bed or a single sphere. It was found that the Sherwood number in a fluidized bed can be less than 2, a value which is the minimum for a single particle as well as a fixed bed (see Table 1.15). This could result from the hydrodynamic conditions which may not be conducive for mass transfer from particle to fluid. c. Correlations Richardson and Szekely234 proposed correlations for the Sherwood number in the case of a shallow fluidized bed of a height equivalent to five times the particle diameter. The correlations are Sh = 0.374 Re1.18

(1.171)

Sh = 2.01 Rep0.5

(1.172)

for 0.1 < Rep < 15 and

© 1999 by CRC Press LLC

for 15 < Rep < 250. It may be observed from these correlations (1.171 and 1.172) that the Sherwood number attains a value far below 2 for Rep values smaller than 15. According to Richardson and Szekely,234 the low values of Sh ( U fall back to the bed, and the other is the height TDH(F) above which entrainment remains constant. Although both types of TDH apparently seem to have the same value, this is not so in a true sense. Hence, the method of evaluating TDH should always be mentioned if a correlation is proposed for it. For Group D particles, Soroko et al.330 proposed the following correlation:

© 1999 by CRC Press LLC

TDH(C) = 1200 Ho Rep1.55Ar–1.1

(1.219)

for 15 < Rep < 300, 19.5 < Ar < 6.5 × 105, Ho < 0.5 m. For Group A particles, the correlation proposed by Horio et al.331 is TDH(F) = 4.47 Des1/2

(1.220)

where Des is the equivalent bubble diameter at the surface. Zenz and Weil296 provided a graph to compute TDH at various vessel diameters for different operating velocities; their predictions show that TDH(F) increases with increasing bed vessel diameter, and no correlation appears to account for this effect. 5. Some Useful Remarks We have seen that several mechanisms and methods have been made use of in connection with the evaluation of entrainment. The ultimate goal is to estimate the total entrainment outside the reactor so that post-reactor equipment can be properly designed for efficient gas–solid separation. One of the important applications of data on entrainment is prediction of the conversion in the freeboard region. The solids dispersed in the gas phase in the freeboard can exhibit significant conversion if the bed contains a major fraction of fines. When the reaction is highly exothermic, it is necessary to provide adequate cooling or heat control in the freeboard, and knowledge of solids entrainment is required for this purpose. From the preceding discussion on entrainment, a design engineer can appreciate that estimating the entrainment rate is essential and cannot be neglected in the final design of an efficient reactor. However, it is clear that much work remains to be carried out in this area. The subject of entrainment has been viewed mainly in the context of gas–solid systems. For liquid–solid systems, the mechanisms described earlier cannot automatically be assumed to be valid. In many mineral operation systems, liquid and gas fluidization are used for physical separation. The field of entrainment or elutriation has considerable relevance for such applications. In the context of three-phase fluidization, the subject of elutriation is in the process of evolving and is gaining importance in light of the biotechnological applications of this type of fluidization.

NOMENCLATURE A a aI A1, B1 Ap Ar B c1 , c2

area of cross-section (m2) constant in Equation 1.214 interfacial surface area (m2/m3) constants in Equation 1.68 projected surface area of the particle (m2) Archimedes number, dρ3s ρf (ρs – ρf)g/µ2g (–) discharge rate (kg/m2) constants in Equation 1.34

© 1999 by CRC Press LLC

C C1 C2 C3 C7 Cbo Cbt CD Cg Cgi Cl Cp Cpg Cpl Cps CS,E D d d* db dbo De Des DG dH DL dp (dp)AM (dp)GM (dp)HM (dp)LM (dp)SM (dp)VS (dp)WM dpi dpm dsv Dt dv E e Eb Ebs El EM Eo

cohesion constant (N/m2) inlet concentration (mol/l) outlet concentration (mol/l) constant in Equation 1.37 equilibrium concentration (mol/l) reactant concentration at the distance z = 0 (mol/m3) concentration in bubble phase at elevation t (mol/m3) drag coefficient (–) reactant concentration in the liquid (mol/l) concentration of the gas at the interphase (mol/l) reactant concentration in the liquid (mol/l) gas concentration in the emulsion phase (mol/m3) specific heat of gas at constant pressure (J/kg · K) specific heat of liquid (J/kg · K) specific heat of solid particle (J/kg · K) solid concentration at the exit of the column (kg) molecular diffusion coefficient (m2/s) diameter of a spherical particle (m) dimensionless particle diameter in Equation 1.73 bubble diameter (m) initial bubble diameter (m) equivalent diameter (m) equivalent bubble diameter at the surface (m) diffusion coefficient (m2/s) hydraulic diameter of the channel (m) diffusion coefficient in liquid (m2/s) particle diameter (m) arithmetic mean particle diameter (m) geometric mean particle diameter (m) harmonic mean particle diameter (m) length mean particle diameter (m) surface mean particle diameter (m) volume surface diameter (m) weight mean particle diameter (m) diameter of particles in ith section (m) mean diameter (m) surface volume diameter (m) diameter of tube or column or vessel (m) volume diameter (m) elutriation rate (kg/m2 · s) energy per unit mass (J/kg) emissivity factor of radiating/receiving bodies (–) generalized emissivity factor (–) liquid-phase dispersion coefficient (m2/s) elasticity modulus of powder bed (N/m2) entrainment just above the bed surface (kg/m2 · s)

© 1999 by CRC Press LLC

Ep Es Es F f FB fb FD Fe ff Ff Fg fk fL Frmf Frt fw g gc H h h hB hc hf hgc hgp Hgsl hi hmax hmax hmf Hmf Ho hp hpc hr hw i J* jfo jH jM K K1

emissivity of the particle (–) emissivity of the surface (–) emissivity factor due to shape (–) feed rate (kg/m2) bed friction factor (–) force due to buoyancy (N) fraction of surface shrouded by gas bubbles (–) drag force (N) adhesion force transmitted in particle contact (N) friction for fluidized bed (–) parameter dependent on fines fraction, Equation 1.117 (–) force due to gravity (N) friction factor for packed bed (–) fractional bubble-phase contact time (–) (= Umf2 /dpg), Froude number at minimum fluidization (–) Froude number based on Ut and Dt, Ut/Ð(gDt) (–) fraction of wake volume (–) gravitational constant (m2/s) conversion factor, gc = 1 kg · m/N · s2 = 32.2 lb · ft/lbf · s2 = 9.8 kg · m/kg · wt · s2 (–) height up to the bed surface (m) overall heat transfer coefficient (W/m2 · K) bed height below H (m) heat transfer coefficient in the bubble-covered surface (W/m2 · K) convective heat transfer coefficient (W/m2 · K) film heat transfer coefficient (W/m2 · K) gas-convective component of heat transfer (W/m2 · K) gas-to-particle heat transfer coefficient (W/m2 · K) height occupied by gas–solid–liquid phases (m) instantaneous heat transfer coefficient (W/m2 · K) maximum heat transfer coefficient (W/m2 · K) maximum heat transfer coefficient corresponding to Uopt (W/m2 · K) heat transfer coefficient at Umf (W/m2 · K) bed height at Umf (m) static bed height (m) particle-convective component of heat transfer coefficient (W/m2 · K) convective heat transfer coefficient due to particle (–) radiative component of heat transfer coefficient (W/m2 · K) bed-to-wall heat transfer coefficient (W/m2 · K) ith component (–) dimensionless fluid flux (–) fluid flux (superficial velocity) relative to particles (m/s) Colburn’s factor (= Nu/Re Pr1/3) Colburn j factor for mass transfer (–) constant in Equation 1.42 (–) constant in Equation 1.27 (–)

© 1999 by CRC Press LLC

K2 K3 k k5, k6, k7 Kc kc Kd KE kef keo kew KG kg Kg Kgw KI Kl kl Kla Kmb KN kp kpe Kpl ks ks KST Kw L M m me N Nb Ncoh Nf ni Nu Nugc Numax Nup p

constant in Equation 1.28 (–) constant in Equation 1.29 (–) thermal conductivity (W/m · K) constants in Equation 1.56 (–) overall mass transfer coefficient (m/s) effective thermal conductivity (W/m · K) mass transfer coefficient (m/s) elutriation rate constant (–) conductivity due to turbulent diffusion (–) effective thermal conductivity of the bed with no fluid (W/m · K) effective thermal conductivity near wall zone (W/m · K) average mass transfer coefficient (m/s) thermal conductivity of gas (W/m · K) gas-side mass transfer coefficient in a gas–liquid mass (–) mass transfer coefficient between fluidized bed (m/s) constant as defined by Equation 1.97 (–) liquid-side mass transfer coefficient in a gas–fluid mass transfer (m/s) thermal conductivity of liquid (W/m · K) volumetric mass transfer coefficient (s–1) constant in Equation 1.116 (–) Newton’s constant, Equation 1.96 (–) mass transfer from particle to particulate bed or spherical particle, Equation 1.170 (s) permeability (m2) mass transfer between particle and fluid (s) thermal conductivity of solid (W/m · K) solid–solid mass transfer coefficient (–) Stokes’ constant, Equation 1.93 (–) ratio of wake voidage to gas-phase voidage (–) length of vertical surface (m) mass of solid (kg) Henry constant (atm/mol fraction) (m/s) constant in Equation 1.148 (–) fluidization number (= ρ3s dp4 g2/µ2E), Equation 1.39 (–) number of bubbles (–) cohesion number (C/ρsgdp) (–) flux of the material transferred from the gas to the liquid phase (mol/s) number of particles of diameter dpi (m) Nusselt number based on h (–) Nusselt number for gas convection, Equation 1.153 (–) Nusselt number at hmax (–) Nusselt number based on particle diameter, dp, hdp/k (–) constant in Equations 1.167 and 1.169 (–)

© 1999 by CRC Press LLC

Pr Q q qr (Q – Qmf)c r* Re Rec Rel ReL Remd Remf Rems Reopt Rep Res Ret Retr Rr Rw S Sb Sc Sh Si So Sp Sv Sw St t Tb Te Tg Tp Ts Tw tz U u U* UA UAV Uc Uf

Prandtl number (Cp µ/kg) (–) volumetric exchange rate of gas between a single bubble and dense phase (m3/s) volumetric exchange rate due to convective flow of gas (m3/s) radiant heat flux (J/m2) critical excess flow of gas (m3/s) dimensionless particle size (–) particle Reynolds number, dpUfρf /µf (–) Reynolds number based on Uc (–) Reynolds number based on liquid velocity (Ul) (–) Reynolds number based on linear dimension (–) Reynold number based on Umd (–) Reynolds number at minimum fluidization velocity (Umf) (–) Reynolds number at the onset of slugging (–) Reynolds number based on Uopt (–) Reynolds number based on particle diameter (dp) (–) Reynolds number defined as 2j*r* (–) Reynolds number based on Ut (–) Reynolds number based on Utr (–) mean thermal resistance of the packet (m2 · K/W) contact resistance near the wall zone (m2 · K/W) surface area per unit volume (m2/m3) frontal surface area of a spherical bubble (m2) Schmidt number, µρg/D (–) Sherwood number, kpldp/D (–) surface area per unit weight of ith component (m2/kg) surface area of a spherical particle (m2) surface area of particle of diameter dp (m2) surface area of a nonspherical particle of constant volume (m2) surface area per unit weight of the particles (m2/kg) Stanton number, kgw /u (–) time (s) fluidized bed temperature (K) intermediate temperature (K) temperature of gas (K) temperature of particle (K) temperature of the surface (K) wall temperature (K) particle residence time (s) superficial gas velocity (m/s) fluid velocity (m/s) dimensionless particle velocity (Equation 1.72) (–) absolute bubble velocity (m/s) mean average bubble rise velocity (m/s) transition velocity from bubbling to turbulent (m/s) fluid velocity (m/s)

© 1999 by CRC Press LLC

Ug Ugl Ul ul ls Ulmf Umb Umd Umf Uop Uopt Up Ur Usl Ut Utr V V Vb Vf Vo Vp Vs Vsv W Ws W/A Wep x x Xe Xi XiB XiE XiF Xo z @

gas velocity (m/s) gas–liquid slip velocity (m/s) liquid velocity (m/s) particle fluid velocity in a suspension (m/s) minimum fluidization velocity for liquid in a liquid–solid (twophase) system (m/s) minimum bubbling velocity (m/s) velocity of transition from dense phase to dilute phase (m/s) minimum fluidization velocity (m/s) optimum gas velocity (m/s) optimum operating velocity corresponding to hmax (m/s) particle velocity (m/s) relative velocity (m/s) particle slip velocity (m/s) single-particle terminal velocity (m/s) transport velocity (m/s) particle velocity (m/s) pore volume per unit mass (m3/kg) volume of bubble (m3) volume of the floating bubble breakers (m3) particle velocity at the bed surface (m/s) volume of particle (m3) volume of solid particle (m3) volume per unit mass of the particle (m3/kg) weight (kg) solids entrained per unit area (kg/m2) bed weight per unit area (kg/m2) Weber number (= ρsUs2dp /σgl) (–) solids weight fraction (–) ratio of the solid holdup in the wake region to that in liquid fluidized bed region (–) exit mole fraction (–) volume fraction (m) mass fraction of ith component in discharged solid (–) mass fraction of ith component in elutriated solid (–) mass fraction of ith component in feed (–) inlet mole fraction (–) distance above distributor (m) powder number (Equations 1.34 and 1.37)

Greek Symbols α αf α1

correction factor in Equation 1.43 angle of internal fraction (°) ratio of the mass velocity of the fluid in the direction of heat flow to the superficial mass velocity in the direction of fluid flow (–)

© 1999 by CRC Press LLC

β β1 γ ∆ρ ∆ρσ ∆c ∆Clm ∆Pb ∆T ∆Ti δ δw B b g gl I l mf o opt p s θ θD λ µ µeff µf µg µl ρa ρB ρb ρbmf ρe ρf ρg ρl ρp ρs ρsz ρt

constant in Equation 1.52 (–) ratio of distance between two adjacent particles to the mean particle diameter (–) angle of wall friction (°) difference in density of solid and gas (kg/m3) pressure drop due to surface tension (Pa · s) concentration driving force (mol/m3) log mean concentration difference (mol/l) bed pressure drop (Pa · s) mean temperature difference (K) mean temperature difference across the ith section (K) angle of rupture (°) film thickness on the wall (m) bed voidage (–) voidage due to bubble phase (–) bubble volume fraction (–) holdup due to gas per unit volume of three-phase fluid bed (–) holdup of gas in gas–liquid column per unit volume of two-phase fluidized bed (–) intraparticle voids due to pores, cracks, etc. (kg/m3) liquid holdup (–) bed voidage at minimum fluidization velocity (–) static bed porosity (–) voidage at optimum velocity (Uopt) (–) solid volume fraction (–) solids volume fraction (–) angle of repose (°) dense-phase root square average residence time (s) shape factor (–) viscosity (Pa · s) effective viscosity (Pa · s) viscosity of fluid (Pa · s) viscosity of gas (Pa · s) viscosity of liquid (Pa · s) apparent density (kg/m3) bulk density of the bed material (kg/m3) bed density (kg/m3) bulk density of the bed at Umf (kg/m2) density of emulsion phase (kg/m3) density of fluid (kg/m3) density of gas (kg/m3) density of liquid (kg/m3) particle density (kg/m3) density of solid particle (kg/m3) solid holdup density at an elevation of z (kg/m3) theoretical density (kg/m3)

© 1999 by CRC Press LLC

σ φ ω

surface tension (N/m) sphericity of a particle (–) angle of slide (°)

REFERENCES 1. Zenz, F.A. and Othmer, D.F., Fluidization and Fluid Particle Systems, Van NostrandReinhold, Princeton, NJ, 1960, 113. 2. Barnea, E. and Mednick, L., Correlation for minimum fluidization velocity, Trans. Inst. Chem. Eng., 53, 278, 1975. 3. Zabrodsky, S.S., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, chap. 1. 4. Geldart, D., Estimation of basic particle properties for use in fluid particle process calculations, Powder Technol., 60, 1, 1990. 5. Abrahamsen, A.R. and Geldart, D., Behavior of gas fluidized beds of fine powders. I. Homogeneous expansion, Powder Technol., 26, 35, 1980. 6. Wadell, H., Volume shape and roundness of rock particles, J. Geol., 15, 443, 1982. 7. Leva, M., Fluidization, McGraw-Hill, New York, 1959, chap. 34. 8. Brown, G.G., Katz, D., Foust, A.S., and Schneidewind, R., Unit Operation, John Wiley & Sons, New York, 1950, 77. 9. Furnas, C.C., Grading aggregates. I. Mathematical relations for beds of broken solids of maximum density, Ind. Eng. Chem., 23, 1052, 1931. 10. Ciborowski, J., Fluidyzacja, PWT, Warsaw, 1957. 11. Geldart, D., The effect of particle size and size distribution on the behaviour of gas fluidized beds, Powder Technol., 6, 201, 1972. 12. Geldart, D., Types of fluidization, Powder Technol., 7, 285, 1973. 13. Molerus, O., Interpretation of Geldart’s Type A, B, C, and D powders by taking into account interparticle cohesion forces, Powder Technol., 33, 81, 1982. 14. Mathur, K.B., Spouted bed, in Fluidization, Davidson, J.F. and Harrison, D.H., Eds., Academic Press, London, 1971, chap. 17. 15. Clark, N.N., Van Egmond, J.W., and Turton, R., A numerical representation of Geldart’s classification, Powder Technol., 56, 225, 1968. 16. Rietema, K., Powders, what are they? Powder Technol., 37, 20, 1984. 17. Saxena, S.C. and Ganzha, V.L., Heat transfer to immersed surfaces in gas fluidized beds of large particles and powder characterization, Powder Technol., 39, 199, 1984. 18. Leva, M., Fluidization, McGraw-Hill, New York, 1959, 64. 19. Miller, Co. and Logwinuk, A.K., Fluidization studies of solid particles, Ind. Eng. Chem., 41, 1135, 1949. 20. Rowe, P.N. and Yacono, C., The distribution of bubble size in gas fluidized beds, Trans. Inst. Chem. Eng., 53, 59, 1975. 21. Davidson, J.F. and Harrison, D., Fluidized Particles, Cambridge Press, London, 1963. 22. Yacono, C., Ph.D. thesis, University of London, 1975. 23. Baerg, A., Klassen, J., and Gishler, P.E., Heat transfer in a fluidized solids beds, Can. J. Res., F28, 287, 1950. 24. Kumar, A. and Sen Gupta, P., Prediction of minimum fluidization velocity for multicomponent mixtures, Indian J. Technol., 12(5), 225, 1974.

© 1999 by CRC Press LLC

25. Baeyens, J. and Geldart, D., Predictive calculations of flow parameters in gas fluidized beds and fluidization behaviour of various powders, in Proc. Int. Symp. on Fluidization and Its Application, Toulouse, France, 1973, 263. 26. Leva, M., Shirai, T., and Wen, C.Y., La prevision due debut de la fluidization daus les lits solides granulaires, Genie Chim., 75(2), 33, 1956. 27. Riba, J.P., Expansion de couches fluidisees par des liqudes, Can. J. Chem. Eng., 55, 118, 1977. 28. Ballesteros, R.L., These de Docteur Ingenieur, Universite de Toulouse, France, 1980. 29. Doichev, K. and Akhmakov, N.S., Fluidization of polydisperse systems, Chem. Eng. Sci., 34, 1357, 1979. 30. Thonglimp, V., Higuily, N., and Leguerie, C., Vitesse minimale decouches fluidize par un gas, Powder Technol., 38, 233, 1984. 31. Morse, R.D., Fluidization of granular solids–fluid mechanics and quality, Ind. Eng. Chem., 41, 117, 1949. 32. Schiller, L. and Naumann, A.N., Z. Ver. Dtsch. Ing., 77, 318, 1935. 33. Wen, C.Y. and Yu, Y.H., Mechanics of fluidization, Chem. Eng. Prog. Symp. Ser., 62(2), 100, 1966. 34. Richardson, J.F. and Zaki, W.N., Sedimentation and fluidization. I, Trans. Inst. Chem. Eng., 32, 35, 1954. 35. Haider, A. and Levenspiel, O., Drag coefficient and terminal velocity of spherical and non-spherical particles, Powder Technol., 58, 63, 1989. 36. Turton, R. and Levenspiel, O., A short note on the drag correlation for sphere, Powder Technol., 47, 83, 1986. 37. Masliyah, J.B. and Polikar, M., Terminal velocity of porous sphere, Can. J. Chem. Eng., 58, 299, 1980. 38. Rowe, P.N. and Henwood, G.A., Drag forces in a hydraulic model of a fluidized bed. I, Trans. Inst. Chem. Eng., 39, 43, 1961. 39. Frantz, J.F., Minimum fluidization velocities and pressure drop in fluidized beds, Chem. Eng. Prog. Symp. Ser., 62, 21, 1966. 40. Davies, L. and Richardson, J.W., Gas interchange between bubbles and the continuous phase in a fluidized bed, Trans. Inst. Chem. Eng., 44(8), 293, 1966. 41. Pillai, B.C. and Raja Rao, M., Pressure drop and minimum fluidization velocities in air fluidized beds, Indian J. Technol., 9, 225, 1971. 42. Ergun, S., Fluid flow through packed columns, Chem. Eng. Prog., 48, 89, 1952. 43. Carman, P.C., Fluid flow through granular beds, Trans. Inst. Chem. Eng., 15, 150, 1937. 44. Bourgeois, P. and Grenier, P., The ratio of terminal velocity to minimum fluidization velocity for spherical particles, Can. J. Chem. Eng., 46, 325, 1968. 45. Ghosal, S.K. and Mukherjee, R.N., Momentum transfer in solid–liquid fluidized beds. 1. Prediction of minimum fluidization velocity, Br. Chem. Eng., 17, 248, 1972. 46. Saxena, S.C. and Vogel, G.J., The measurement of incipient fluidization velocities in a bed of coarse dolomite at high temperature and pressure, Trans. Inst. Chem. Eng., 55, 184, 1977. 47. Babu, S.P., Shah, B., and Talwalkar, A., Fluidization correlations for coal gasification materials–minimum fluidization velocity and fluidized bed expansion, AIChE Symp. Ser., 74, 176, 1978. 48. Richardson, J.F. and Jeromino, M.A.S., Velocity voidage relations for sedimentation and fluidization, Chem. Eng. Sci., 39, 1419, 1979. 49. Thonglimp, V., These Dr. Ingenier, Institut National Polytechnique de Toulouse, France, 1981.

© 1999 by CRC Press LLC

50. Chitester, D.C., Kornosky, R.M., Fan, L.S., and Danko, J.P., Characteristics of fluidization at high pressure, Chem. Eng. Sci., 39, 253, 1984. 51. Masaaki, N., Hamada, Y., Toyama, S., Fouda, A.E., and Capes, C.E., An experimental investigation of minimum fluidization velocity at elevated temperatures and pressures, Can. J. Chem. Eng., 63, 8, 1985. 52. Agarwal, P.K. and O’Neil, B.K., Transport phenomena in multi-particle system. I. Pressure drop and friction factors unifying the hydraulic radius and submerged object approaches, Chem. Eng. Sci., 43, 2487, 1988. 53. Satyanarayana, K. and Rao, P.G., Minimum fluidization velocity at elevated temperatures, Ind. Chem. Eng., 31, 79, 1989. 54. Grace, J.R., Fluidized bed hydrodynamics, in Handbook of Multiphase Systems, Hestroni, G., Ed., Hemisphere Press, New York, 1982, chap. 8.1. 55. Panigrahi, M.R. and Murty, J.S., A generalized spherical multiparticle model for particulate systems: fixed and fluidized beds, Chem. Eng. Sci., 46(7), 1863, 1991. 56. Fletcher, J.V., Deo, M.D., and Hanson, F.V., Reexamination of minimum fluidization velocity correlations applied to group B sands and coked sands, Powder Technol., 69, 147, 1992. 57. Lippens, B.C. and Mulder, J., Prediction of the minimum fluidization velocity, Powder Technol., 75, 67, 1993. 58. Kunii, D. and Levenspiel, O., Fluidization Engineering, John Wiley & Sons, New York, 1979, chap. 3. 59. Richardson, J.F., Incipient fluidization and particulate system, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 2. 60. Clift, R. and Gauvin, W.H., The motion of particle in turbulent gas streams, Chemeca 70, Butterworths, Australia, 1970, 14. 61. Concha, F. and Almendra, E.R., Settling velocities of particulate systems. 1. Settling velocities of individual spherical particles, Int. J. Miner. Process., 5, 349, 1979. 62. Concha, F. and Almendra, E.R., Settling velocities of particulate systems. 2. Settling velocities of suspensions of spherical particles, Int. J. Miner. Process., 6, 31, 1979. 63. Concha, F. and Barrientos, A., Settling velocities of particulate systems. 3. Power series expansion for the drag coefficient of sphere and prediction of the settling velocity, Int. J. Miner. Process., 9, 167, 1982. 64. Concha, F. and Barrientos, A., Settling velocities of particulate systems. 4. Settling of nonspherical isometric particles, Int. J. Miner. Process., 18, 297, 1986. 65. Concha, F. and Christiansen, A., Settling velocities of particulate systems. 5. Settling velocities of suspensions of particles of arbitrary shape, Int. J. Miner. Process., 18, 309, 1986. 66. Geldart, D., Estimation of particle properties for use in fluid particle process calculations, Powder Technol., 60, 1, 1990. 67. Pettyjohn, E.S. and Christiansen, E.B., Effect of particle shape on free settling rates of isometric particles, Chem. Eng. Prog., 44, 157, 1948. 68. McCabe, W.L. and Smith, J.C., Unit Operations of Chemical Engineering, McGrawHill, New York, 1967, chap. 7. 69. Rubey, W.W., Settling velocities of gravel, sand and silt particles, Am. J. Sci., 225, 325, 1933. 70. Dallavalle, J.M., Micromeritics, Pitman Publishing, New York, 1943. 71. Dallavalle, J.M., Micromeritics, The Technology of Fine Particles, 2nd ed., Pitman Publishing, New York, 1948, chap. 2.

© 1999 by CRC Press LLC

72. Lapple, C.E., Fluid and Particle Mechanics, University of Delaware Publ., Newark, 1951, 284. 73. Torobin, L.B. and Gauvin, W.H., Fundamental aspects of solid–gas flow. Parts 1 and 2, Can. J. Chem. Eng., 37, 129, 1959. 74. Olson, R., Essentials of Engineering Fluid Mechanics, International Text Book, Scranton, PA, 1961. 75. Barnea, E. and Mizrahi, J., Generalized approach to the fluid dynamics of particulate systems. I. General correlation for fluidization and sedimentation in solid multiparticle systems, Chem. Eng. J., 5, 171, 1973. 76. Clift, R., Grace, J.R., and Weber, M.E., Bubble, Drops and Particles, Academic Press, New York, 1978. 77. Flemmer, R.L.C. and Banks, C.L., On the drag coefficient of a sphere, Powder Technol., 48, 217, 1986. 78. Khan, A.R. and Richardson, J.F., Pressure gradient and friction factor for sedimentation and fluidization of uniform spheres in liquids, Chem. Eng. Sci., 45, 255, 1990. 79. Wallis, G.B., A simple correlation for fluidization and sedimentation, Trans. Inst. Chem. Eng., 55, 74, 1977. 80. Turton, R. and Clark, N.N., An explicit relationship to predict spherical particle terminal velocity, Powder Technol., 53, 127, 1987. 81. Geldart, D., Gas Fluidization Technology, Wiley, Chichester, 1986, chap. 6. 82. Lewis, E.W. and Bowerman, E.W., Fluidization of solid particles in liquid, Chem. Eng. Prog., 48, 603, 1952. 83. Lewis, W.K., Gilliland, E.R., and Bauer, W.C., Characteristics of fluidized particles, Ind. Eng. Chem., 41, 1104, 1949. 84. Chen, P. and Pei, D.C.T., Fluidization characteristics of fine particles, Can. J. Chem. Eng., 62, 464, 1984. 85. Lucas, A., Arnaldos, J., Casal, J., and Pulgjaner, L., Improved equation for the calculations of minimum fluidization, Ind. Eng. Chem. Process Des. Dev., 25, 426, 1986. 86. Yang, W.C., Chitester, D.C., Kornos, R.M., and Keairn, D.L., A generalized methodology for estimating minimum fluidization velocity at elevated pressure and temperature, AIChE J., 31, 1086, 1985. 87. Wilhelm, R.H. and Kwauk, M., Fluidization of solid nonvesicular particles, Chem. Eng. Prog., 44, 201, 1948. 88. Romero, J.B. and Johanson, L.N., Factors affecting fluidized bed quality, Chem. Eng. Prog. Symp. Ser., 58(38), 28, 1962. 89. Geldart, D., The fluidized bed as a chemical reactor: a critical review of the first 25 years, Chem. Ind., p. 1474, September 2, 1967. 90. Ho, T.C., Fan, L.T., and Walawender, W.P., The onset of slugging in gas–solid fluidized beds with large particles, Powder Technol., 35, 249, 1983. 91. Lanneau, K.P., Gas solids contacting in fluidized beds, Trans. Inst. Chem. Eng., 38, 125, 1960. 92. Rhodes, M. and Geldart, D., Measurement of axial solid flux variation in the riser of a circulating fluidized bed, in Circulating Fluidized Bed Technology II, Basu, P. and Large, P., Eds., Pergamon Press, Oxford, 1988, 155. 93. Lee, G.S. and Kim, S.D., Pressure fluctuations in turbulent fluidized beds, J. Chem. Eng. Jpn., 21, 515, 1988. 94. Lee, G.S. and Kim, S.D., Bed expansion characteristics and transition velocity in turbulent fluidized beds, Powder Technol., 62, 207, 1990.

© 1999 by CRC Press LLC

95. Yang, W.C., Criteria for choking in vertical pneumatic conveying lines, Powder Technol., 35, 143, 1983. 96. Yang, W.C., A mathematical definition of choking phenomenon and a mathematical model for predicting choking velocity and choking voidage, AIChE J., 21, 1013, 1975. 97. Østergaad, K., Studies of Gas–Liquid Fluidization, Danish Technical Press, Copenhagen, 1969. 98. Østergaad, K., Three phase fluidization, in Fluidization, Davidson, J.F. and Harrison, D. Eds., Academic Press, New York, 1971, chap. 18. 99. Østergaad, K., Three phase fluidization, in Recent Advances in the Engineering Analysis of Chemically Reacting Systems, Doraiswamy, L.K., Ed., Wiley Eastern Ltd., New Delhi, 1984, chap. 22. 100. Davidson, J.F., Three phase systems, in Chemical Reactor Theory — A Review, Lapidus, L. and Amundson, N.R., Eds., Prentice-Hall, Englewood Cliffs, NJ, chap. 10. 101. Kim, S.D. and Kim, C.H., Three phase fluidized beds: a review, Hwahak Konghak, 18(5), 313, 1980. 102. Wild, G., Saberian, M., Schwartz, J.L., and Charpentier, J.C., Gas–liquid–solid fluidized bed reactors, state of the art and industrial perspectives, Entropie, 106, 3, 1982; English translation in Int. Chem. Eng., 24(4), 639, 1984. 103. Epstein, N., Three phase fluidization: some knowledge gaps, Can. J. Chem. Eng., 59, 649, 1981. 104. Epstein, N., Hydrodynamics of three phase fluidization, in Handbook of Fluids in Motion, Cheremisinoff, N.P. and Gupta, R., Eds., Ann Arbor Science, Ann Arbor, MI, 1983, 1165. 105. Muroyama, K. and Fan, L.S., Fundamentals of gas–liquid–solid fluidization, AIChE J., 31(1), 1, 1985. 106. Fan, L.S., Gas–Liquid–Solid Fluidization Engineering, Butterworths, Boston, 1989. 107. Fan, L.S., Muroyama, K., and Chern, S.H., Hydrodynamics of inverse fluidization in liquid–solid and gas–liquid–solid systems, Chem. Eng. J., 24, 143, 1982. 108. Bhatia, V.K. and Epstein, N., Three phase fluidization; a generalized wake model, in Fluidization and Its Applications, Angelino, H., Couderc, J.P., Gibert, H., and Laguerie, Co., Cepadues Editions, Toulouse, France, 1974, 380. 109. Dakshinamurthy, P., Subrahmanyam, V., and Rao, J.N., Bed porosities in gas–liquid fluidization, Ind. Eng. Chem. Process Des. Dev., 10, 322, 1971. 110. Begovich, J.M. and Watson, J.S., Hydrodynamic characteristics of three phase fluidized beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 190. 111. Nikov, I., Grandjean, B.P.A., Carreau, P.J., and Paris, J., Viscosity effect in cocurrent three phase fluidization, AIChE J., 36(10), 1613, 1990. 112. Kato, Y., Uchida, K., and Morooka, S., Liquid hold-up and heat transfer coefficient between bed and wall in liquid–solid and gas–liquid–solid fluidized beds, Powder Technol., 28, 173, 1981. 113. Saberian-Broudjenni, M., Wild, G., Charpentier, J.C., Fortin, Y., Euzen, J.P., and Patoux, R., Contribution to the hydro-dynamics study of gas–liquid–solid fluidized bed reactors, Int. Chem. Eng., 27(3), 423, 1987. 114. Grace, J.R., Fluidization, in Handbook of Multiphase Systems, Hestroni, G., Ed., McGraw-Hill–Hemisphere, Washington, D.C., 9, 1982, 8.1.17. 115. Douglas, H.R., Snider, I.W.A., and Tomlinson, G.H., II, The turbulent contact absorber, Chem. Eng. Prog., 59(12), 85, 1963. 116. Visvanathan, C. and Leung, L.S., On the design of a fluidized bed scrubber, Ind. Eng. Chem. Process Des. Dev., 24, 677, 1985.

© 1999 by CRC Press LLC

117. Barile, R.G., Dengler, J.L., and Hertwig, T.A., Performance and design of a turbulent bed cooling tower, AIChE Symp. Ser., 70, 154, 1975. 118. Gelperin, N.I., Savchenko, V.I., and Grishko, V.Z., Some hydrodynamics laws of absorption apparatus packed with fluidized spheres, Theor. Found. Chem. Eng., 2, 65, 1968. 119. Kito, M., Tabei, K., and Murata, K., Gas and liquid holdups in mobile bed under the countercurrent flow of air and liquid, Ind. Eng. Chem. Process Des. Dev., 17, 568, 1978. 120. Wozniak, M., Pressure drop and effective interfacial area in a column with a mobile bed, Int. Chem. Eng., 17, 553, 1977. 121. Draijer, W., Heat transfer in fluidized bed-boilers, in Fluidzed Bed Combustion, Radovanovic, M., Ed., Hemisphere Publishing, Washington, D.C., 1986, 211. 122. Saxena, S.C., Grewal, N.S., Gabor, J.D., Zabrodsky, S.S., and Galershtein, D.M., Heat transfer between a gas fluidized bed and immersed tubes, in Advances in Heat Transfer, Vol. 14, Irvine, T.F. and Hartnett, J.P., Eds., Academic Press, New York, 1978, 149. 123. Zabrodsky, S.S., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, chap. 7–10. 124. Botterill, J.S.M., Fluid Bed Heat Transfer, Academic Press, London, 1975. 125. Ranz, W.E. and Marshall, W.R., Evaporation from drops. I, Chem. Eng. Prog., 48, 141, 1952. 126. Chen, P. and Pei, D.C.T., Fluid particle heat transfer in fluidized beds of mixed-sized particles, Can. J. Chem. Eng., 62, 469, 1984. 127. Gutfinger, C. and Abuaf, N., in Advances in Heat Transfer, Vol. 10, Hartnett, J.P. and Irvine, T.F., Jr., Eds., Academic Press, New York, 1974, 167. 128. Bartel, W.J., Heat transfer from a horizontal bundle of tubes in an air fluidized bed, Ph.D. thesis, Montana State University, Bozeman, 1971. 129. Saxena, S.C. and Gabor, J.D., Mechanism of heat transfer between a surface and a gas fluidized bed for combustion application, Prog. Energy Combust. Sci., 7(2), 73, 1981. 130. Van Heerden, C., Nobel, A.P., and van Krevelen, D., Studies on fluidization. I. The critical mass velocity, Chem. Eng. Sci., 1, 37, 1951. 131. Ziegler, E.N. and Brazelton, W.T., Mechanism of heat transfer to a fixed surface in a fluidized bed, Ind. Eng. Chem. Fund., 3(2), 94, 1964. 132. Botterill, J.S.M., The mechanism of heat transfer to fluidized beds, in Proc. Int. Symp. on Fluidization, Drinkenburg, A.A.H., Ed., Netherlands University Press, Amsterdam, 1967, 183. 133. Yamazaki, R. and Jimbo, G., Heat transfer between fluidized beds and heated surfaces, J. Chem. Eng. Jpn., 3(1), 44, 1970. 134. Wicke, E. and Fetting, F., Warmeubertragung in GaswirbelSchichten, Chem. Ing. Tech., 26, 301, 1954. 135. Mickley, H.S. and Fairbanks, D.F., Mechanism of heat transfer to fluidized beds, AIChE J., 1, 374, 1965. 136. Botterill, J.S.M. and Williams, J.R., The mechanism of heat transfer to fluidized beds, Trans. Inst. Chem. Eng., 41, 217, 1963. 137. Botterill, J.S.M. and Butt, M.H.D., Achieving high heat transfer rates in fluidized beds, Br. Chem. Eng., 13(7), 1000, 1968. 138. Gabor, J.D., Wall to bed heat transfer in fluidized beds and packed beds, Chem. Eng. Prog. Symp. Ser., 66(105), 76, 1970.

© 1999 by CRC Press LLC

139. Zabrodsky, S.S., Hydrodynamics and Heat Transfer, MIT Press, Cambridge, MA, 1966, chap. 8. 140. Selzer, V.W. and Thomson, W.J., Fluidized bed heat transfer — the packet theory revisited. AIChE Symp. Ser., 73(161), 29, 1977. 141. Xavier, A.M. and Davidson, J.F., Convective heat transfer in fluidized bed, in Fluidization, 2nd ed., Davidson, J.F., Clift, R., and Harrison, D., Eds., Academic Press, London, 1985, chap. 13. 142. Krupiczka, R., Analysis of thermal conductivity in granular materials, Int. Chem. Eng., 7, 122, 1967. 143. Ranz, W.E., Friction and transfer coefficients for single particles and packed beds, Chem. Eng. Prog., 48(3), 247, 1952. 144. Baron, T., Generalized graphical method for the design of fixed bed catalytic reactors, Chem. Eng. Prog., 48(3), 118, 1952. 145. Baskakov, A.P., Vitt, O.K., Kirakosyan, V.A., Maskaev, V.K., and Filippovsky, N.F., Investigation of heat transfer coefficient pulsations and of the mechanism of heat transfer from a surface immersed into a fluidized bed, in Proc. Int. Symp. on Fluidization and Its Applications, Cepadues Editions, Toulouse, France, 1974, 293. 146. Botterill, J.S.M. and Denloye, A.O.O., Bed to surface heat transfer in a fluidized bed of large particles, Powder Technol., 19, 197, 1978. 147. Gelprin, N.I. and Einstein, V.G., Heat transfer in fluidized beds, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 10. 148. Baskakov, A.P., Process of Heat and Mass Transfer in Fluidized Beds, Metallurgia, Moscow, 1978, 184. 149. Kunii, D. and Levenspiel, O., Eds., Fluidized Engineering, John Wiley & Sons, New York, 1968, chap. 9. 150. Botterill, J.S.M., Ed., Fluid Bed Heat Transfer, Academic Press, London, 1975, 81. 151. Zabrodsky, S.S., Ed., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, chap. 10. 152. Gelprin, N.I. and Einstein, V.G., Heat transfer in fluidized beds, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 10. 153. Grewal, N.S., Experimental and Theoretical Investigations of Heat Transfer Between a Gas Solid Fluidized Bed and Immersed Tubes, Ph.D. thesis, University of Illinois at Chicago, 1979. 154. Gutfinger, C. and Abuaf, N., Heat transfer in fluidized beds, in Advances in Heat Transfer, Vol. 10, Irvine, T.F., Jr. and Hartnett, J.P., Eds., Academic Press, New York, 1964, 167. 155. Zabrodsky, S.S., Antonishin, V.N., and Paranas, A.L., On fluidized bed surface heat transfer, Can. J. Chem. Eng., 54, 52, 1976. 156. Grewal, N.S. and Saxena, S.C., Effect of surface roughness on heat transfer from horizontal immersed tubes in a fluidized bed, Heat Transfer, 101(3), 397, 1979. 157. Vreedenberg, H., Heat transfer between a fluidized bed and a horizontal tube, Chem. Eng. Sci., 1(1), 52, 1958. 158. Vreedenberg, H., Heat transfer between a fluidized bed and verticle tube, Chem. Eng. Sci., 11, 274, 1960. 159. Wender, L. and Cooper, G.T., Heat transfer between fluidized bed and boundary surfaces: correlation data, AIChE J., 1(1), 15, 1958. 160. Baskakov, A.P., Berg, B.V., Vitt, O.K., Fillippovsky, N.F., Kiroskosyan, V.A., Goldobin, J.M., and Maskaev, V.K., Heat transfer to objects immersed in fluidized beds, Powder Technol., 8, 273, 1973.

© 1999 by CRC Press LLC

161. Dahloff, B. and Von Brachel, H., Warmeubergang, and horizontal cenem fluidat bett ange ordneten rohrbundelu, Chem. Ing. Tech., 40, 373, 1968. 162. Noak, R., Lokaler Warmeubergang, an horizontalen Fohren in Wirbelschichten, Chem. Ing. Tech., 42(6), 371, 1970. 163. Blenke, H. and Neukirchem. B., Gastaltung horizontaler rohrbundel in gas-wirbelschichtreaktorn kach warmetchnischen, Chem. Ing. Tech., 45(5), 307, 1973. 164. McClaren, J. and Williams, D.F., Combustion efficiency, sulphur retention and heat transfer in pilot fluidized bed combustion, J. Inst. Fuel, 42, 303, 1969. 165. Lese, J.K. and Kermode, R.I., Heat transfer from a horizontal tube to a fluidized bed in the presence of unheated tubes, Can. J. Chem. Eng., 50, 44, 1972. 166. Bartel, W.J. and Genetti, W.F., Heat transfer from a horizontal bundle of bare and finned tubes in an air fluidized bed, Chem. Eng. Prog. Symp., 69(128), 85, 1973. 167. Andeen, B.R. and Glicksman, L.R., Heat Transfer to Horizontal Tubes in Shallow Fluidized Beds, Heat Transfer Conf. Paper No. 76 H-67, ASME-AIChE., St. Louis, August 9–11, 1976. 168. Petrie, J.C., Freeby, W.A., and Buckham, J.A., In bed exchangers, Chem. Eng. Prog., 64(67), 45, 1968. 169. Zabrodsky, S.S., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, 270. 170. Genetti, W.E., Sehmall, R.A., and Grimmett, E.S., The effect of tube orientation on heat transfer with bare and finned tubes in a fluidized bed, Chem. Eng. Prog. Symp. Ser., 67(116), 90, 1971. 171. Ternovskaya, A.N. and Korenbrg, Y.G., Pyrite Kilning in a Fluidized Bed, Izd. Khimiya, Moscow, 1971. 172. Saxena, S.C. and Grewal, N.S., Heat transfer between a horizontal tube and a gas solid fluidized bed, Int. J. Heat Mass Transfer, 23, 1505, 1980. 173. Sarkits, V.B., Heat Transfer from Suspended Beds of Granular Materials to Heat Transfer Surfaces (Russian), Leningrad Technol. Inst., Lensovieta, 1959. 174. Miller, C.O. and Logwink, A.K., Fluidization studies of solid particles, Ind. Eng. Chem., 43, 1220, 1951. 175. Wicke, E. and Hedden, F., Stromungsformen and warmeubertragung in von luft aufgewirbelten schuttgutschichten, Chem. Ing. Tech., 24(2), 82, 1952. 176. Dow, W.M. and Jakob, M., Heat transfer between a vertical tube and a fluidized bed air mixture, Chem. Eng. Prog., 47, 637, 1951. 177. Gaffney, B.J. and Drew, T.B., Mass transfer from packing to organic solvents in single phase flow through a column, Ind. Eng. Chem., 42, 1120, 1950. 178. Traber, P.G., Pomaventsev, V.M., Mukhenov, I.P., and Sarkits, V.B., Heat transfer from a suspended layer of catalyst to the heat transfer surface, Zh. Prikl. Khim., 35, 2386, 1962. 179. Baskakov, A.P., High speed non oxidative heating and heat treatment in a fluidized bed, Metallurgia (Moscow), 1968. 180. Vreedenberg, H.A., Heat transfer between fluidized beds and vertically inserted tubes, J. Appl. Chem., 2(Suppl. issue), 26, 1952. 181. Brotz, W., Utersuchungen uber transportvorgange in durchstrometem gekontem Gut, Chem. Ing. Techn., 28, 165, 1956. 182. Sathiyamoorthy, D., Sridhar Rao, Ch., and Raja Rao, M., Effect of distrubutor on heat transfer from immersed surfaces in gas fluidized beds, Chem. Eng. J., 37, 149, 1988. 183. Baskakov, A.P. and Berg, B.V., Inzh. Fiz Zh. Akad. Nauk Belorussk SSR6, 8, 3, 1966. 184. Botterill, J.S.M., Teoman, Y., and Yüregir, K.R., Temperature effect on the heat transfer behaviour of gas fluidized beds, AIChE Symp. Ser., 77(208), 330, 1984.

© 1999 by CRC Press LLC

185. Molerus, O., Heat transfer in gas fluidized beds. 2. Dependence of heat transfer on gas velocity, Powder Technol., 15, 70, 1992. 186. Borodulya, V.A., Ganzh, V.A., and Podberlzsky, A.I., Heat transfer in a fluidized bed at high pressure, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 201. 187. Varygin, N.N. and Martyushin, I.G., A calculation of heat transfer surface area in fluidized bed equipment, Khim Mashinostr. (Moscow), 5, 6, 1959. 188. Baerg, A., Klassen, J., and Grishler, P.E., Heat transfer in fluidized solid beds, Can. J. Res. Sect. F, 28, 324, 1964. 189. Gelprin, N.I., Einstein, V.G., and Romanova, N.A., Khim. Prom. (Moscow), 5, 6, 1959. 190. Jakob, A. and Osberg, G.L., Effect of gas thermal conductivity on local heat transfer in a fluidized bed, Can. J. Chem. Eng., 35, 5, 1957. 191. Chechetkin, A.V., Vysokotemperature (High Temperature Carriers), Gosenergoizdat, Moscow, 1962. 192. Ruckenstein, E., Zh. Prikl. Khim., 35, 71, 1962; as cited by Gutfinger, C. and Abuaf, N., Heat transfer in fluidized beds, in Advances in Heat Transfer, Irvine T.F. and Hartnett, J.P., Eds., Academic Press, New York, 1964. 193. Richardson, J.F. and Shakiri, K.J., Heat transfer between a gas–solid fluidized bed and a small immersed surface, Chem. Eng. Sci., 34, 1019, 1979. 194. Kim, K.J., Kim, D.J., and Chun, K.S., Heat and mass transfer in fixed and fluidized bed reactors, Int. Chem. Eng., 8(5), 176, 1971. 195. Grewal, N.S. and Saxena, S.C., Maximum heat transfer coefficient between a horizontal tube and a gas–solid fluidized bed, Ind. Eng. Chem. Process Res. Dev., 20, 108, 1981. 196. Baskakov, A.P. and Panov, O.P., Comparison between measured maximum heat transfer coefficients to surfaces in fluidized beds, Inzh. Fiz. Zh., 45, 896, 1983. 197. Chen, P. and Pei, D.C.T., A model heat transfer between fluidized beds and immersed surfaces, Int. J. Heat Mass Transfer, 28(3), 675, 1985. 198. Palchenok, G.I. and Tamarin, A.I., Study of heat transfer between model particles and fluidized beds, Inzh. Fiz. Zh., 45, 427, 1983. 199. Sathiyamoorthy, D. and Raja Rao, M., Prediction of maximum heat transfer coefficient in gas fluidized bed, Int. J. Heat Mass Transfer, 25(5), 1027, 1992. 200. Molerus, O., Heat transfer in gas fluidized beds. I, Powder Technol., 70, 1, 1992. 201. Saxena, S.C. and Grewal, N.S., Effect of distributor design on heat transfer from an immersed horizontal tube in a fluidized bed, Chem. Eng. J., 18, 197, 1979. 202. Sathiyamoorthy, D., Sridhar Rao, Ch., and Raja Rao, M., Heat transfer in gas fluidized beds using multi-orifice distributors, Indian J. Technol., 25, 219, 1987. 203. Frantz, J.F., Fluid to particle heat transfer in fluidized beds, Chem. Eng. Prog., 57, 65, 1961. 204. Wasmund, B. and Smith, J.W., Wall to fluid heat transfer in liquid fluidized bed, Can. J. Chem. Eng., 45, 156, 1967. 205. Romani, M.N. and Richardson, J.F., Heat transfer from immersed surfaces to liquid fluidized beds, Lett. Heat Mass Transfer, 1, 55, 1974. 206. Krishnamurthy, N. and Sathiyamoorthy, D., Mechanism of heat transfer in liquid fluidized beds, in Proc. National Heat and Mass Transfer Conf., IISc Bangalore, India, Ind. Soc. Heat Mass Trans., Madras, 1987, 81. 207. Mishra, P. and Farid, M.M., Wall to bed heat transfer in liquid fluidized beds, Ind. Chem. Eng., 28(3), 30, 1986. 208. Hog, G.W. and Grimmett, E.S., Effects of pulsation on heat transfer in a liquid fluidized bed, Chem. Eng. Prog. Symp. Ser., 62(67), 51, 1966.

© 1999 by CRC Press LLC

209. Østergaard, K., Discussion in the paper of R. Turner, Fluidization, Soc. for the Chem. Ind., London, 1964, 58. 210. Viswanathan, S.A., Kakar, A.S., and Murti, P.S., Effect of dispersing bubbles into liquid fluidized beds on heat transfer holdup at constant bed expansion, Chem. Eng. Sci., 20, 903, 1964. 211. Armstrong, E.A., Baker, C.G.J., and Bergougnou, M.A., Heat transfer and hydrodynamics studies on three phase fluidized beds, in Fluidization Technology, Keairns, D.L., Ed., Hemisphere Publishing, New York, 1976, 453. 212. Baker, C.G.J., Armstrong, F.R., and Bergongnou, M.A., Heat transfer in three phase fluidized beds, Powder Technol., 21, 195, 1978. 213. Kato, Y., Uchida, K., Kago, T., and Morooka, S., Liquid holdup and heat transfer coefficient between bed and wall in liquid–solid and gas–liquid–solid fluidized beds, Powder Technol., 28, 173, 1981. 214. Kato, Y., Kago, T., and Morooka, S., Heat transfer from wall to packed and fluidized bed for gas–liquid co-current up-flow, Kagaku Kogaku Ronbunshu, 4, 328, 1978. 215. Ermakova, A., Ziganskin, G.K., and Slinko, M.G., Hydrodynamics of a gas–liquid reactor with a fluidized bed of solid matter, Theor. Found. Chem. Eng., 4, 84, 1970. 216. Phillippovski, N.F. and Baskakov, A.P., Investigation of the temperature field in a fluidized bed close to a heated plate and heat transfer between them, Int. Chem. Eng., 13, 5, 1953. 217. Botterill, J.S.M. and Desai, M., Limiting factors in gas fluidized bed, Powder Technol., 6(2), 231, 1972. 218. Knuth, M. and Weinspach, P.M., Experimentelle unter-Schchung, des warme and Stoffubergangs and die Partikeln einer wirbelschicht beider desubbimation, Chem. Ing. Tech., 48, 893, 1976. 219. Kharchenko, N.V. and Makhorin, K.E., The rate of heat transfer between a fluidized bed and an immersed body at high temperatures, Int. Chem. Eng., 4, 650, 1964. 220. Mersmann, A., Determination of fluidizing velocity in fluidized beds by heat transfer measurements, Chem. Ing. Tech., 38(10), 1095, 1966. 221. Mersmann, A., Heat transfer in fluidized beds, Chem. Ing. Tech., 39(56), 349, 1967. 222. Mersmann, A., Zum Warmeuvergang Vuischen dispersen zueiphasensysten unl senkrechten heizflachen in erdschwereseld, Verfahrenstechnik, 10(10), 641, 1976. 223. Thones, D. and Kramers, H., Mass transfer from spheres in various regular packings to a flowing fluid ionic mass transfer in the presence of fluidized solids, Chem. Eng. Sci., 8, 271, 1958. 224. Jaganadharaju, G.J.V. and Rao Venkata, C., Ionic mass transfer in the presence of fluidized solids, Indian J. Technol., 3, 201, 1965. 225. Zeigler, E.N. and Holmes, J.T., Mass transfer from fixed surfaces to gas fluidized beds, Chem. Eng. Sci., 21, 117, 1966. 226. Einstein, V.G. and Gelpin, N.I., Heat transfer between a fluidized bed and a surface, Int. Chem. Eng., 6, 67, 1966. 227. Beek, W.J., Mass transfer in fluidized beds, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 9. 228. Froessling, N., Gerland Beitr, Geophys. J., 52, 170, 1938. 229. Ranz, W.E., Friction and transfer coefficients for single particles and packed beds, Chem. Eng. Prog., 48, 247, 1952. 230. Fan, L.T., Yang, C.Y., and Wen, C.Y., Mass transfer in semi fluidized beds for solid liquid system, AIChE J., 6, 482, 1960. 231. Rebnick, W.E. and White, R.R., Mass transfer in systems of gas and fluidized solids, Chem. Eng. Prog., 45, 377, 1949.

© 1999 by CRC Press LLC

232. Kettenring, K.N., Manderfield, E.L., and Smith, J.M., Heat and mass transfer in fluidized systems, Chem. Eng. Prog., 46, 139, 1950. 233. Kunii, D. and Levenspiel, O., Fluidization Engineering, John Wiley & Sons, New York, 1969, 200. 234. Richardson, J.F. and Szekely, J., Mass transfer in a fluidized bed, Trans. Inst. Chem. Eng., 39, 212, 1961. 235. Wakao, N., Particle to fluid heat/mass transfer coefficients in packed bed catalytic reactors, in Recent Advances in the Engineering Analyses of Chemically Reacting Systems, Doraiswamy, L.K., Ed., Wiley Eastern Ltd., New Delhi, 1984, chap. 2. 236. Damronglerd, S., Couderc, J.P., and Angelino, H., Mass transfer in particulate fluidization, Trans. Inst. Chem. Eng., 53, 175, 1975. 237. Wakao, N. and Funazkri, T., Effect of fluid dispersion coefficients on particle to fluid mass transfer coefficients in packed beds — correlation of Sherwood number, Chem. Eng. Sci., 33, 1375, 1978. 238. Kunii, D. and Levenspiel, O., Bubbling bed model — model for flow of gas through a fluidized bed, Ind. Eng. Chem. Fundam., 7, 446, 1968. 239. Chiba, T. and Kobayashi, H., Gas exchange between the bubble and emulsion phases in gas fluidized beds, Chem. Eng. Sci., 25, 1375, 1970. 240. Murray, J.D., On the mathematics of fluidization. 1. Fundamental equations and wave propagation, J. Fluid Mech., 21, 465, 1965; 2. Steady motion of fully developed bubbles, J. Fluid Mech., 22, 57, 1965. 241. Partridge, B.A. and Rowe, P.N., Chemical reaction in a bubbling gas fluidized bed, Trans. Inst. Chem. Eng., 44, T335, 1966. 242. Stephens, G.K., Sinclair, R.J., and Potter, O.E., Gas exchange between bubbles and dense phase in a fluidized bed, Powder Technol., 1, 157, 1967. 243. Kunii, D. and Levenspiel, O., Effect of particles in bubbles on fluidized bed mass and heat transfer kintics, J. Chem. Eng. Jpn., 24(2), 183, 1991. 244. Yasui, G. and Johanson, L.N., Characteristics of gas packets in fluidized beds, AIChE J., 4, 445, 1958. 245. Whitehead, A.B. and Young, A.D., Fluidization performance in large scale equipment. I, in Proc. Int. Symp. on Fluidization, Drikenburg, A.A.H., Ed., Netherlands University Press, 1967, 284. 246. Park, W.H., Kang, W.K., Capes, C.E., and Osberg, G.L., The properties of bubbles in fluidized beds of conducting particles as measured by an electro resistivity probe, Chem. Eng. Sci., 24, 851, 1969. 247. Kobayashi, H., Arai, F., and Chiba, T., Behaviour bubbles in gas fluidized beds, Chem. Eng. Sci., 21, 239, 1967. 248. Kato, K. and Wen, C.Y., Bubble assemblage model for fluidized bed catalytic reactors, Chem. Eng. Sci., 24, 1351, 1969. 249. Chiba, T., Terashima, K., and Kobayashi, H.J., Lateral distribution of bubble sizes in two dimensional gas fluidized bed, Chem. Eng. Jpn., 8, 167, 1975. 250. Mori, S. and Wen, C.Y., Estimation of bubble diameter in gaseous fluidized bed, AIChE J., 21, 109, 1975. 251. Hirma, T., Ishida, M., and Shirai, T., Kagaku Kogaku Ronbunshu, 1, 272, 1975. 252. Rowe, P.N., Prediction of bubble size in a gas fluidized bed, Chem. Eng. Sci., 31, 285, 1976. 253. Darton, R.C., La Nauze, R.D., Davidson, J.F., and Harrison, D., Bubble growth due to coalescence in fluidized beds, Trans. Inst. Chem. Eng., 55, 274, 1977. 254. Werther, J., Mathematical modelling of fluidized bed reactors, Chem. Ing. Tech., 50, 850, 1978.

© 1999 by CRC Press LLC

255. Werther, J., Bubbles in gas fluidized beds. I and II, Trans. Inst. Chem. Eng., 52, 149, 1974. 256. Davidson, J.F. and Schuler, B.O.G., Bubble formation at an orifice in inviscid liquid, Trans. Inst. Chem. Eng., 38, 335, 1960. 257. Chiba, T., Terashima, K., and Kobayashi, H., Behaviour of bubbles in gas–solid fluidized beds; initial formation of bubbles, Chem. Eng. Sci., 27, 965, 1972. 258. Miwa, K., Mori, S., Kato, T., and Muchi, I., Behaviour of bubble in a gaseous fluidized bed, Int. Chem. Eng., 12, 187, 1972. 259. Fryer, C. and Potter, O.E., Experimental investigation of models for fluidized bed catalytic reactors, AIChE J., 22, 38, 1976. 260. Davis, L. and Richardson, J.F., Gas interchange between bubbles and continuous phase in a fluidized bed, Br. Chem. Eng., 12, 1223, 1967. 261. Toei, R., Matsuno, R., Nishitani, K., Miyagawa, H., and Komagawa, Y., Gas transfer between a bubble and the continuous phase in a gas fluidized bed, Int. Chem. Eng., 9, 358, 1969; Kogaku Kogaku, 32, 565, 1968. 262. Kobayashi, H., Arai, F., and Chiba, T., Kagaku Kogaku, 31, 239, 1967. 263. Miyauchi, T. and Morooka, S., Mass transfer rate between bubble and emulsion phase in fluid bed, J. Chem. Eng. Jpn., 33, 880, 1969. 264. Chiba, T. and Kobayashi, H., Gas exchange between the bubble and emulsion phases in gas fluidized beds, Chem. Eng. Sci., 25, 1375, 1970. 265. Calderbank, P.H., Pereira, J., and Burgss, M.M., The physical and mass transfer properties of bubbles in fluidized beds of electrically conducting particles, in Int. Conf., Pacific Grove, CA, 1976, 261. 266. Mori, S. and Muchi, I., Theoretical analysis of catalytic reaction in a fluidized bed, Kogaku Kogaku, 5, 251, 1972. 267. Davidson, J.F., Harrison, D., Darton, R.C., and LaNauze, R.D., in Chemical Reactor Theory — A Review, Lapidus, L. and Amundson, N.R., Eds., Prentice-Hall, Englewood Cliffs, NJ, 1977. 268. Alper, E., Wichtendahl, B., and Deckwer, W.D., Gas absorption mechanism in catalytic slurry reactors, Chem. Eng. Sci., 35, 217, 1980. 269. Østergaard, K. and Suchozebrski, W., Gas–liquid mass transfer in gas liquid fluidized beds, in Proc. Eur. Symp. React. Eng., Pergamon Press, Oxford, 1969, 21. 270. Østergaard, K. and Fosbol, P., Transfer of oxygen across the gas–liquid interface in gas–liquid fluidized beds, Chem. Eng. J., 3, 105, 1972. 271. Alvarez-Cuenca, M. and Nerenberg, M.A., The plug flow model for mass transfer in three phase fluidized beds and bubble columns, Can. J. Chem. Eng., 59, 739, 1981. 272. Dankwerts, P.W., Gas–Liquid Reactions, McGraw-Hill, New York, 1970, 15. 273. Geankopolis, C.J., Mass Transport Phenomena, Holt, Reinhart and Winston, New York, 1972, 316. 274. Sharma, M.M. and Danckwerts, P.V., Chemical method of measuring interfacial area and mass transfer coefficients in two-fluid systems, Br. Chem. Eng., 524, 208, 1970. 275. Gay, F.V., Holtman, C.B., and Sheppard, N.F., Mass Transfer in Three Phase Fluidized Beds, ORNL Report No. ORNL/MIT-283, Oak Ridge, TN, December 21, 1978. 276. Patwari, A.N., Nguyen-tien, K., Schumpe, A., and Deckwer, W.D., Three phase fluidized beds with viscous liquid: hydrodynamics and mass transfer, Chem. Eng. Commun., 40, 49, 1989. 277. Schumpe, A., Deckwer, W.D., and Nigam, K.D.P., Gas fluid mass transfer in three phase fluidized beds with viscous pseudoplastic liquids, Can. J. Chem. Eng., 67, 873, 1989.

© 1999 by CRC Press LLC

278. Tang, W.T. and Fan, L.S., Gas–liquid mass transfer in a three phase fluidized bed containing low density particles, Ind. Eng. Chem. Res., 29, 128, 1990. 279. Alvarez-Cuenca, M., Oxygen Mass Transfer in Bubble Columns and Three Phase Fluidized Beds, Ph.D. thesis, University of Western Ontario, Canada, 1979. 280. Alvarez-Cuenca, M., Oxygen mass transfer in three phase fluidized beds working at large flow rates, Can. J. Chem. Eng., 61, 5, 1983. 281. Muroyama, K. and Fan, L.S., Fundamentals of gas–liquid–solid fluidization, AIChE J., 31(1), 1, 1985. 282. Dhanuka, V.R. and Stepanek, J.B., Gas–liquid mass transfer in a three phase fluidized bed, in Fluidization, Grace, J.R. and Matson, G.M., Eds., Plenum Press, New York, 1980, 261. 283. Kim, S.D. and Chang, H.S., Hydrodynamics and bubble breakage in three phase fluidized beds, Hwahak Konghak, 17, 407, 1979. 284. Kim, J.O. and Kim, S.D., Bubble characteristics in three phase fluidized beds of floating bubble breakers, Particulate Sci. Technol., 5, 309, 1987. 285. Kim, S.D., Lee, Y.J., and Kim, J.O., Heat transfer and hydrodynamic studies on two and three phase fluidized beds of floating bubble breakers, Exp. Thermal Fluid Sci., 1, 237, 1988. 286. Kim, J.O. and Kim, S.D., Gas–liquid mass transfer in a three phase fluidized bed with floating bubble breakers, Can. J. Chem. Eng., 68, 368, 1990. 287. Morooka, S., Kusakabe, K., and Kato, Y., Mass transfer coefficient at the wall of a rectangular fluidized bed for liquid–solid and gas–liquid–solid systems, Kogaku Kogaku Ronbunshu, 5, 162, 1979; Int. Chem. Eng., 20, 433, 1980. 288. Batchelor, G.K., Mass transfer from small particles suspended in turbulent fluid, J. Fluid Mech., 98, 609, 1980. 289. Arters, D. and Fan, L.S., Liquid–solid mass transfer in a gas–liquid–solid fluidized bed, presented at AIChE Meeting, San Francisco, November 25–30, 1984. 290. Groshe, E.W., Analysis of gas fluidized solid systems by X-ray absorption, AIChE J., 1(3), 358, 1955. 291. Bakker, P.I. and Heertjes, P.M., Porosity measurement in fluidized beds, Br. Chem. Eng., 3, 240, 1958. 292. Fan, L.T., Chan, J.L., and Baile, R.C., Axial solid distribution in gas–solid fluidized beds, AIChE J., 8(2), 239, 1962. 293. Rowe, P.N., MacGillivray, H.J., and Chessman, D.J., Gas discharge from an orifice into a gas fluidized bed, Trans. Inst. Chem. Eng., 57, 194, 1979. 294. Yang, W.C. and Keairns, D.L., Design and operating parameters for a fluidized bed agglomerating combustor gasifier, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 208. 295. Yang, W.C. and Keairns, D.L., Estimating the jet penetration depth of multiple grid jets, Ind. Eng. Chem. Fundam., 18, 317, 1979. 296. Zenz, F.A. and Weil, N.A., A theoretical-empirical approach to the mechanism of particle entrainment from fluidized beds, AIChE J., 4, 472, 1958. 297. Shakova, N.A., Outflow of turbulent jets into a fluidized bed, Inzh. Fiz. Zh., 14, 6, 1968. 298. Basov, V.A., Markhevka, V.I., Melik-Akhnazarov, T.Kh., and Orochko, D.I., Investigation of the structure of a non-uniform fluidized bed, Int. Chem. Eng., 9, 263, 1969. 299. Merry, J.M.D., Penetration of vertical jets into fluidized beds, AIChE J., 21, 507, 1975. 300. Deole, N.R., Study of Jets in Three Dimensional Gas Fluidized Beds, M.S. thesis, West Virginia University, Morgantown, 1980.

© 1999 by CRC Press LLC

301. Yang, W., Jet penetration in a pressurised fluidized bed, Ind. Eng. Chem. Fundam., 2, 297, 1981. 302. Blake, T.R., Wen, C.Y., and Ku, C.A., The correlation of jet penetration measurements in fluidized beds using non-dimensional hydrodynamic parameters, AIChE Symp. Ser., 80(234), 42, 1984. 303. Kececioglu, I., Yang, W., and Keairns, D.L., Fate of solids fed pneumatically through a jet into a fluidized bed, AIChE J., 30, 99, 1984. 304. Wen, C.Y., Deole, N.R., and Chen, L.H., A study of jets in a three dimensional gas fluidized bed, Powder Technol., 31, 1975, 1982. 305. Kozin, V.E. and Baskakov, A.P., An investigation of the grid zone of a fluidized bed above cap-type distributors, Int. Chem. Eng., 8(2), 257, 1968. 306. Filla, M., Massimilla, L., and Vaccaro, S., Gas jets in fluidized beds and spouts: a comparison of experimental behaviour and models, Can. J. Chem. Eng., 61, 370, 1983. 307. Donadono, S., Maresca, A., and Massimilla, L., Gas injection in shallow beds of fluidized coarse solids, Ing. Chim. Ital., 16, 1, 1980. 308. Lofroy, G.A. and Davidson, J.F., The mechanics of spouted beds, Trans. Inst. Chem. Eng., 47, T120, 1969. 309. Littman, H., Morgan, M.H., III, Vukovic, D.V., Zdanski, F.K., and Graveie, Z.B., A method for predicting the relationship between the spout and inlet tube radii in a spouted bed at its maximum spoutable height, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 381. 310. Yang, W.C. and Keairns, D.L., Momentum dissipation and gas entrainment in a gas–solid two-phase jet in a fluidized bed, in Fluidization, Grace, J.R. and Matsen, J.M., Eds, Plenum Press, New York, 1980, 305. 311. Mamuro, T. and Hattori, H., Flow pattern of fluid spouted beds, J. Chem. Jpn., 1, 5, 1968. 312. Donsi, G., Massimilla, L., and Colantuonio, L., The dispersion of axisymmetric gas jets in fluidized beds, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 297. 313. Becker, H.A., An investigation of laws governing the spouting of coarse particles, Chem. Eng. Sci., 13, 245, 1961. 314. Lim, C.J. and Mathur, K.B., Modelling of particle movement in spouted beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 104. 315. Ching, Ho. T., Yutani, N., Fan, L.T., and Walawender, W.P., Stochastic modelling of bubble formation on the grid in a gas–solid fluidized bed, Can. J. Chem. Eng., 61, 654, 1983. 316. Hsiung, T.P. and Grace, J.R., Formation of bubbles at an orifice in fluidized beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, chap. 21. 317. Werther, J., Effect of gas distributor on the hydrodynamics of gas fluidized beds, Ger. Chem. Eng., 1, 166, 1978. 318. Brien, C.L., Bergougnou, M.A., and Baker, C.G.J., Grid leakage (weeping, dumping, particle back flow) in gas fluidized beds, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 413. 319. Wen, C.Y., Krishnan, R., Dutta, S., and Khosravi, R., Dead zone height near the grid region of fluidized bed, in Proc. 2nd Eng. Foundation Conf., Cambridge, England, 1978, 32. 320. Wen, C.Y. and Dutta, S., Research needs for the analysis, design and scale up of fluidized beds, AIChE Symp. Ser., 73(161), 2, 1977.

© 1999 by CRC Press LLC

321. Wen, C.Y., Krishnan, R., Khosravi, R., and Dutta, S., Dead zone height near the grid of fluidized beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, chap. 2–3. 322. Wen, C.Y., Krishnan, R., and Kalyanaraman, R., Particle mixing near the grid region of fluidized beds, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 405. 323. Leva, M. and Wen, C.Y., Elutriation, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, London, 1971, chap. 14. 324. Geldart, D., Elutriation, in Fluidization, Davidson, J.F., Clift, R., and Harrison, D., Eds., Academic Press, London, 1985, chap. 11. 325. Rowe, P.N. and Partridge, B.A., X-ray study of bubbles in fluidized beds, Trans. Inst. Chem. Eng., 43, T157, 1965. 326. Kunii, D. and Levenspiel, O., Fluidization Engineering, John Wiley & Sons, New York, 1979, chap. 10. 327. Kunii, D. and Levenspiel, O., Entrainment of solids from fluidized beds. 1. Hold up in the free board, 2. Operation of fast fluidized beds, Powder Technol., 61, 193, 1990. 328. Kraft, W.W., Ulrich, W., and O’Connor, W., in Fluidization, Othmer, D.F., Ed., Reinhold, New York, 1956, 194. 329. Lewis, W.K., Gilliland, E.R., and Lang, P.M., Entrainment from fluidized beds, Chem. Eng. Prog. Symp. Ser., 58(68), 65, 1962. 330. Soroko, V., Mikhalev, M., and Mukhlenov, I., Calculation of the minimum height of the space above the in fluidized bed contact equipment, Int. Chem. Eng. 9, 280, 1969. 331. Horio, M., Taki, A., Hsieh, Y.S., and Muchi, I., Elutriation and particle transport through the freeboard of gas–solid fluidized bed, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 509.

© 1999 by CRC Press LLC

View more...
I. INTRODUCTION Fluidization is a unit operation, and through this technique a bed of particulate solids, supported over a fluid-distributing plate (often called the grid), is made to behave like a liquid by the passage of the fluid (gas, liquid, or gas–liquid) at a flow rate above a certain critical value. In other words, it is the phenomenon of imparting the properties of a fluid to a bed of particulate solids by passing a fluid through the latter at a velocity which brings the fixed or stationary bed to its loosest possible state just before its transformation into a fluidlike bed. A. Fluidlike Behavior Let us consider the various situations that could prevail in a bed of particulate solids. When there is no fluid flow in the bed, it remains in a static condition and the variation in pressure across the bed height is not proportional to its height, unlike in a liquid column. When a fluid such as a gas or a liquid is allowed to percolate upward through the voidage of a static bed, the structure of the bed remains unchanged until a velocity known as the minimum fluidization velocity is reached; at this velocity, drag force, along with buoyant force, counteracts the gravitational force. In this situation, the bed just attains fluidlike properties. In other words, a bed that maintains an uneven surface in a static, fixed, or defluidized state now has an even or horizontal surface (Figure 1.1a). A heavy object that would rest on the top of a static bed would now sink; likewise, a light object would now tend to float. The pressure would now vary proportional to the height, like a liquid column, and any hole made on the vessel or column would allow the solid to flow like a liquid. All these features are depicted in Figure 1.1.

© 1999 by CRC Press LLC

Figure 1.1 Examples of fluidlike behavior of fluidized bed relative to fixed bed.

B. Fluidization State 1. Gas/Liquid Flow The fluid under consideration which flows upward can be either a gas, a liquid, or both. In general, liquid fluidized beds are said to have a smooth or homogeneous or particulate nature of fluidization. The bed expands depending on the upward liquid flow rate, and due to this expansion the bed can become much higher than its initial or incipient height. In contrast, a gas fluidized bed is heterogeneous or aggregative or bubbling in nature and its expansion is limited, unlike what happens in a liquid fluidized bed. It is seldom possible to observe particulate fluidization in a gas fluidized bed and aggregative fluidization in a liquid fluidized bed. If the fluid

© 1999 by CRC Press LLC

flow regimes are such that a bed of particulate solids has a boundary defined by a surface, then it has solid particles densely dispersed in the fluid stream. In other words, a dense-phase fluidized bed is achieved. When the surface is not clearly defined at a particular velocity, the solid particles are likely to be carried away by the fluid. This situation corresponds to a dilute or a lean phase. The situation where solid particles are entrained by the fluid flow corresponds to a state called pneumatic transport. As the velocity of the liquid in a liquid fluidized bed is increased, homogeneous or particulate fluidization with smooth expansion occurs, followed by hydraulic transport of particles at a velocity equal to or greater than the particle terminal velocity. In the gas fluidized bed, bubbling is predominant. The minimum bubbling velocity is the velocity at which the bubbles are just born at the distributor. The bubbling bed at velocities greater than the minimum bubbling velocity tends to slug, especially in a deep and/or narrow column, and the slugging is due to the coalescence of bubbles. When the bubbles coalesce and grow as large as the diameter of the column, a slug is initiated. Now solids move above the gas slug like a piston, and they rain through the rising slugs. Here the gas–solid contact is poor. The slugging regime, through a transition point, attains a turbulent condition of the bed, and this process is often termed fast fluidization. Pneumatic transport of solid particles by the gas stream occurs at and above the particle terminal velocity. Liquid and gas fluidized beds for various gas flow rates are illustrated in Figure 1.2.

Figure 1.2 Liquid and gas fluidized beds at various operating velocities.

© 1999 by CRC Press LLC

2. Onset of Fluidization Estimation of the onset of the fluidization velocity is essential because it is the most important fundamental design parameter in fluidization. This velocity determines the transition point between the fixed bed and the fluidized bed. In a fixed bed, solid particles remain in their respective fixed positions while the fluid percolates through the voids in the assemblage of particles. In such a situation, the fluid flow does not affect or alter the voidage or the bed porosity. As the fluid flow is increased, the bed pressure drop increases. At a certain stage, the bed pressure drop reaches a maximum value corresponding to the bed weight per unit area (W/A); at this stage, a channelfree fluidized bed at the ideal condition is obtained. 3. Situation at the Onset of Fluidization Let us examine the situation at the onset of fluidization; the corresponding fluidization velocity is also referred to as the incipient velocity. When this velocity is just attained, the fixed bed of particles exists in is loosest possible condition without any appreciable increase in bed volume or bed height. In such a condition, the bed weight less the weight equivalent to buoyancy is just balanced by the drag due to the upward flow of the fluid. In other words, the distributor plate or the grid which supports the bed of solids does not experience any load under this condition. The velocity at which the bed is levitated, achieving a state of fluidlike behavior just at the transition of a fixed to a fluidized bed, is called the minimum fluidization velocity. The transition from a fixed to a fluidized bed may not be the same for increasing and decreasing direction of the fluid flow; this is particularly so when the fluid is a gas. 4. Bed Pressure Drop We will now examine the various situations that can occur in a bed pressure drop versus superficial velocity plot for a typical gas–solid system. In its true sense, the superficial velocity is the net volume of fluid crossing a horizontal (empty) plane per unit area per unit time. This superficial velocity is many times smaller than the interstitial fluid velocity inside the bed. Nevertheless, superficial velocity is considered because of convenience and ease of measurement. When a gas passes through a fixed bed of particulate solid, the resistance to its flow, in addition to various hydrodynamic parameters, depends on the previous history of the bed, that is, whether the bed under consideration is a well-settled bed or a well-expanded and just settled bed. In a well-settled bed, the important structural parameter, the bed voidage (), is relatively low, and thus the pressure drop obtained initially by passing the gas upward is of a relatively high magnitude, as depicted by line A–B in Figure 1.3. This figure is similar to one depicted by Zenz and Othmer1 and Barnea and Mednick.2 At point B, a transition from a fixed bed to a fluidized bed starts, and this prevails up to point C. The bed pressure drop beyond C for a fluidized bed remains unchanged in an ideal case even though the superficial velocity (U) is increased. The bed pressure drop beyond point D, which corresponds to the

© 1999 by CRC Press LLC

Figure 1.3 Variation in bed pressure drop (∆P) with superficial velocity (U).

particle terminal velocity for a monosized bed of solids, is no longer constant, and it increases in a manner similar to an empty column. This is so because the particles are carried away from the bed or are completely entrained when the superficial velocity equals the particle terminal velocity. In this situation, the bed voidage () is unity, or the volume fraction of solid particles (1 – ) is zero. Line EDF corresponds to the pressure drop in the empty column. When the upward flow of gas is gradually decreased, the bed pressure drop assumes its original path on a pressure drop versus velocity plot as long as the bed continues to be in a state of fluidization. Retracing the path, as shown in Figure 1.3 by line DCG, indicates that the pressure drop obtained for a fixed bed during its settling is lower than that obtainable for increasing upward flow of gas. Point C on line GCD is the transition point between a fixed and a fluidized bed, and the velocity corresponding to this is the minimum fluidization velocity. It may be observed that during increasing flow through the bed, there is no distinct point marking the transition, except that zone which corresponds to B–C with a peak. The relatively high value of the pressure drop (∆P) along line AB in a fixed bed when the flow is in the laminar regime compared to line GC of an expanded settled bed is due to low bed permeability. Let us now look at the fluid dynamic aspects at points A and B. For laminar flow conditions, most correlations for the pressure drop give: ∆P α U(1 – )2/3

(1.1)

At point A, the bed porosity () corresponds to the static bed value (0) (i.e., = 0), and at point B, the bed porosity is equal to the minimum fluidization value (mf) (i.e., = mf). From Equation 1.1, it follows that:

© 1999 by CRC Press LLC

(∆P)/[U(1 – )2/3] = Constant

(1.2)

and this is true if is not altered for a fixed bed. However, as indicated in Figure 1.1, points A and B correspond to a fixed bed but to different values; the proportionality constant in Equation 1.2 is thus altered. In view of this, Barnea and Mednick2 cautioned against the use of any unmodified fixed bed correlation for predicting the minimum fluidization velocity and also pointed out that point C in Figure 1.3 is the limiting condition for a fixed bed and point B for a fluidized bed. Because the particle concentration and its randomness affect the pressure drop, any attempt to use a fixed-bed pressure drop correlation for the purpose of predicting its minimum fluidization velocity will lead to erroneous results. C. Advantages of Fluidized Bed 1. A high rate of heat and mass transfer under isothermal operating conditions is attainable due to good mixing. 2. A fluidlike behavior facilitates the circulation between two adjacent reactors (e.g., catalytic cracking and regenerator combination). 3. There is no moving part, and hence a fluidized bed reactor is not a mechanically agitated reactor. For this reason, maintenance costs can be low. 4. The reactor is mounted vertically and saves space. This aspect is particularly important for a plant located at a site where the land cost is high. 5. A continuous process coupled with high throughput is possible. 6. No skilled operator is required to operate the reactor. 7. The fluidized bed is suitable for accomplishing heat-sensitive or exothermic or endothermic reactions. 8. The system offers ease of control even for large-scale operation. 9. Excellent heat transfer within the fluidized bed makes it possible to use low-surface-area heat exchangers inside the bed. 10. Multistage operations are possible, and hence the solids residence time as well as the fluid residence time can be adjusted to desired levels. D. Disadvantages of Fluidized Bed 1. Fine-sized particles cannot be fluidized without adopting some special techniques, and high conversion of a gaseous reactant in a single-stage reactor is difficult. 2. The hydrodynamic features of a fluidized bed are complex, and hence modeling and scaleup are difficult. 3. Generation of fines due to turbulent mixing, gas or liquid jet interaction at the distributor site, and segregation due to agglomeration result in undesirable products. 4. Elutriation of fines and power consumption due to pumping are inevitable.

© 1999 by CRC Press LLC

5. Sticky materials or reactions involving intermediate products of a sticky nature would defluidize the bed. 6. Limits on the operating velocity regime and on the choice of particle size range are disadvantages of fluidization. Fluidization of friable solids requires careful attention to avoid loss of fines formed due to attrition. 7. Highly skilled professionals in this area are needed for design and scaleup. 8. Erosion of immersed surfaces such as heat-exchanger pipes may be severe. 9. Reactions that require a temperature gradient inside the reactor cannot be accomplished in a fluidized bed reactor.

II. PROPERTIES OF PARTICLES AND THE GRANULAR BED A particulate solid has several properties which, in addition to the density, size, shape, and distribution, also play a key role in determining stationary bed or fixed bed properties. The roughness and the voidage associated with the particles should also be considered in fluid–particle interactions. A. Particles 1. Size Particulate materials or granular solids, whether manufactured or naturally occurring, can never have the same particle size. In other words, particulate solids comprised of uniformly sized particles are very difficult to obtain unless they are sized, graded, or manufactured under extreme control of the operating conditions. Achieving particles close in size in a manufacturing process is not simple; only specific processes like shot powders and liquid-drop injection into a precipitating solution can yield powders of relatively uniform size, but then only in the initial period of large-size particle production. Hence, uniformly sized particles are obtained by several physical techniques such as sieving, sedimentation, microscopy, elutriation, etc. 2. Definition There are several ways to define particle size. For the purpose of powder characterization, it is not customary to define all the sizes as given by various mathematical functions. A particle size that is the diameter equivalent of a sphere is used along with a shape factor in many hydrodynamic correlations. The shape factor will be discussed and defined later in this section. Several definitions of the mean particle diameter are found in the literature, some of which are presented in the following discussion. If a sample of a powder of a given mass is constituted of particles of different sizes and if there are ni particles with a diameter dpi (i = 1 to N), then the diameter can be defined in a variety of ways.

© 1999 by CRC Press LLC

1. Arithmetic mean:

( (d ) = ∑

n d pi

∑

p AM

)

N i =1 i

(1.3)

N i =1

2. Geometric mean: (dp)GM = 3(dp1 × dp2 × … × dpn)0.5

(1.4)

3. Log geometric mean:

( )

log d p

∑ (n log d ) ∑ n

(1.5)

(∑ n ) [∑ (n d )]

(1.6)

=

GM

N i =1

i

pi

i

4. Harmonic mean:

(d )

p HM

=

N i =1 i

N i =1

i

pi

The harmonic mean diameter in principle is related to the particle surface area per unit weight, that is,

( ∑ n ) ⋅ π( d ) (∑ n ) ⋅ ρ π ( d )

2

N i =1 i

Total surface area = Weight

N i =1 i

p HM 3

p

= 6

p HM

6 1 ⋅ ρp dp

( )

(1.7) HM

where ρp is the density of the particle. 5. Mean diameter, dpm: 50% of particles > dpm

50% of particles < dpm

6. Length mean diameter, (dp)LM:

(d )

=

p LM

∑

n ⋅ d pi2

N i =1 i

∑

N i =1 pi

d

(1.8)

7. Surface mean diameter, (dp)SM:

(d )

p SM

=

(∑

N i=1 i

n d pi2

∑ n)

(1.9)

∑ n)

(1.10)

12

N i=1 i

8. Volume mean diameter, (dp)VM:

(d )

p VM

© 1999 by CRC Press LLC

=

(∑

N i =1 i

n d pi3

N i =1 i

13

9. Weight mean diameter, (dp)WM:

( ) dp

WM

=

∑

x dp =

N i =1 i

∑ ∑

N i =1 i

n d pi4

N i =1 i

n d pi3

(1.11)

where xi is the weight or volume percent of particles with diameter dpi. For example,

xi =

ni d pi3

∑

(1.12)

N i =1 i

n d pi3

10. Volume surface diameter, (dp)VS:

( ) dp

VS

=

6 = Sw ρ p

∑ ∑

N i =1 i

n d pi3

(1.13)

N i =1 i

n d pi2

where Sw is the surface area per unit weight of the particles:

Sw =

∑

N i =1 i wi

xs

and

xi =

ni d pi3

∑

N i =1 i

n d pi3

(1.14)

Zenz and Othmer1 gave an account of industries that are inclined to choose the particle diameter of their own interest and also showed that it is not necessarily meaningful to select any particle diameter as desired. Zabrodsky3 recommends the use of the harmonic mean diameter in fluidization studies. Geldart4 commented that the particle size accepted in packed and fluidized beds is the surface–volume diameter (dsv), defined as the diameter of a sphere that has the same ratio of the external surface area to the volume as the actual particle. For a powder that has a mixture of particle sizes, it is equivalent to the harmonic mean diameter. Another diameter is the volume diameter, which is the diameter of a sphere whose volume is the same as that of the actual particle, that is, dv = (6M/ρpπ)1/3

(1.15)

The particle size determined by sieve analysis is the average of the size or opening of two consecutive screens, and it may be referred to as dp. The surface–volume diameter for most sands5 is dsv = 0.87 dp. The ratio dsv/dp is expected to vary widely depending on the particle shape, and it may not be easy to determine this ratio experimentally for irregularly shaped particles. Based on calculations pertaining to

© 1999 by CRC Press LLC

18 regularly shaped solids, Abrahamsen and Geldart5 showed that if the volume diameter (dv) = 1.127 dp for a sphericity of 0.773, then dsv = 0.891 dp. 3. Sphericity Sphericity (Φs) is a parameter that takes into consideration the extent of the deviation of an actual particle from the spherical shape or degree of sphericity. It is defined as the ratio of the surface area of a sphere to that of the actual particle that has the same volume. Let us consider a sphere made up of clay with a surface area So. If the sphere is distorted by pressing it, the shape changes; the resultant clay mass has the same volume as it had originally, but now it has a different surface area, say Sv. The ratio of the original surface area (So) to the new surface area (S) for the distorted clay sphere is its sphericity. It can be mathematically defined6 as:

⎛S ⎞ Φs = ⎜ o ⎟ ⎝ Sv ⎠ constant volume

(1.16)

If the diameter of a particle (dp) is defined as the diameter of the sphere that has the same volume as the actual particle, then the shape factor (λ) is given by: 23

1 1 Vp = Φs = 5.205 S p λ

(1.17)

If j is the ratio of the particle volume to the volume of a sphere that has the same surface area as the particle, then λ j2/3 = 1

(1.18)

The sphericity of many regularly shaped particles can be estimated by analytical means, whereas it is not easy to estimate the sphericity value analytically for an irregularly shaped particle. Pressure drop correlations for fixed bed reactors incorporate the shape factor. Hence, these correlations are often used to obtain the shape factor for the particle using experimental data on pressure drop and the fluid–solid properties. The shape factor for a regularly shaped particle can be estimated readily. For example, the shape factor for a cube is 1.23. For cylinders, it is a function of the aspect ratio, and for rings it is dependent on the ratio of inside to outside diameter. It should be noted here that the hydraulic resistance or the bed pressure drop correlation used to estimate the shape factor should correspond to the laminar flow (Re < 10) regime and there should be no roughness effect. 4. Roughness The roughness of a particle obviously adds to the friction between particles, and it leads to an increase in bed porosity (loose packing) when the bed settles down. The increase in bed porosity in turn reduces resistance to fluid flow. In other words,

© 1999 by CRC Press LLC

the bed pressure drop in a bed of particles with rough surfaces is lower compared to one that has particles with smooth surfaces, which have a tendency to form a less dense or low-porosity bed. Leva7 estimated that the friction factor values for clay particles are 1.5 times higher than for glass spheres and 2.3 times higher than for fused MgO particles. The particle roughness has to be determined by measuring the friction factor and by comparing with standard reference plots for particles of various known roughness factors. B. Granular Bed 1. Bed Porosity or Voidage Bed porosity or voidage is affected by several parameters, such as the size, shape, size distribution, and roughness of the particles; the packing type; and the ratio of the particle diameter to the vessel diameter. Small or fine-sized particles have low settling or terminal velocities and low ratios of mass to surface area. Hence, fine particles, when poured into a vessel, settle slowly and create mass imbalance. As a result of these two factors, the bed has a tendency to form bridges or arches, which causes the bed voidage to increase. A bed with bridges or arches is not suitable for smooth fluidization. A dense bed of fine powders can be obtained by shaking, tapping, or vibrating the vessel. 2. Voidage and Packing Voidage for spherical particles depends on the type of packing and varies from 25.95% (rhombohedral packing) to 47.65% (square packing). For nonuniform angular particles, voidage can vary widely depending on the type of packing (i.e., whether it is loose, normal, or dense). A bed is in its loosest packed form when the bed material is wet charged (i.e., the material is poured into the vessel containing the liquid or the material and after pouring into an empty container is fluidized by a gas and then settled). The bed will be dense when the container or the column is vibrated, shaken, or tapped for a prolonged period. A representative variation of voidage () of packing with uniformly sized particle diameter (dp) for three packing conditions (viz., loose, normal, and dense) is shown in Figure 1.4. 3. Polydisperse System Furnas9 investigated experimentally the voidage of a binary system of varying particle size ratios. His studies showed that if the initial voidages of the individual components of the binary system are not the same, the voidage of the mixture will be less than the volumetric weighted average of the initial voidages. In a binary system of coarse and fine particles that have equal particle density and also equal voidage (), the volume fraction of the coarse particles is given as 1/(1 + ) at the condition of minimum voidage (i.e., maximum density). This low value of voidage is due to the fine particles that fill the interstices of the coarse material. A third component which is smaller than the second component and also finer relative to

© 1999 by CRC Press LLC

Figure 1.4 Voidage in uniformly sized and randomly packed beds. (From Brown, G.G., Katz, D., Foust, A.S., and Schneidewind, R., Unit Operation, John Wiley & Sons, New York, 1950, 77. With permission.)

the first component may be added to fill the interstices of the second component; theoretically, the process can go on for an infinite number of components. The resultant or the final volume fraction (m) of a solid mixture with n components is given by:

(1 – m ) = (1 +1 ) + (1 + ) + (1+ ) + … + (1+ ) 2

n

(1.19)

From the above expression, it is possible to obtain the percentage of each component required to produce the minimum voids by multiplying both sides by 100/(1 – m). 4. Container Effect The effect of particle roughness on voidage was discussed in the preceding section on particle roughness. Now we will consider the effect of the ratio of the particle size to the container/column diameter on bed voidage. Particle packing close to the walls is relatively less dense. Hence, for vessels, particularly when d/Dt (where d is the particle diameter and Dt is the vessel diameter) is large, the voidage contribution due to the wall effect is high, and the wall effect for particles away from it is insignificant for large-diameter vessels. Ciborowski’s10 data on bed porosity for various values of d/Dt corresponding to different vessel geometries and materials are shown in Table 1.1. It can be seen from these data that bed porosity increases as d/Dt increases.

© 1999 by CRC Press LLC

Table 1.1 Bed Porosities for Various Shapes of Packings d/D

Ceramic Spheres

Smooth Spheres

Smooth Cylinders

Raschig Rings

0.1 0.2 0.3 0.4 0.5

0.4 0.44 0.49 0.53 —

0.33 0.37 0.40 0.43 0.46

0.34 0.38 0.44 0.50 0.56

0.55 0.58 0.64 0.70 0.75

5. Important Properties of Particulate Solids a. Density There are in general three types of densities referred to in the literature: true, apparent, and bulk densities. True density is the weight of the material per unit volume, when the volume is considered free of pores, cracks, or fissures. The true density thus obtained gives the highest value of the density of a material. If, on the other hand, the particle volume is estimated by taking into consideration the intraparticle voids, then the density calculated on this basis is the apparent density. If ρt is the true or theoretical density and ρa is the apparent density for the same mass, then a simple relationship can be deduced between these two parameters. Let 0 be the intraparticle voids due to pores, cracks, etc.; Vsv the volume per unit mass of the particles in the absence of intraparticle voids; and Va the apparent volume per unit mass; then:

1 1 – V – Vsv ρ a ρt = I = a 1 Va ρa

(1.20)

Upon rearrangement of Equation 1.20, one obtains: ρa /ρt + I = 1

(1.21)

An equation for estimating ρa can also be rewritten as:

ρa =

1 V +

1 ρt

(1.22)

where V is the pore volume per unit mass. The bulk density (ρb) is obtained by considering the weight of granular solids packed per unit volume of a vessel or container. This density is less than the apparent density because the volume under consideration for the same mass of solid is increased in this case due to interparticle

© 1999 by CRC Press LLC

voids. If is the interparticle voidage for a granular bed of porous solids that have an intraparticle voidage I, then, as per the preceding procedure:

ρb + =1 ρa

(1.23)

Substituting for ρa from Equation 1.21 and modifying Equation 1.23, one gets:

ρb = (1 – I )(1 – ) ρt

(1.24)

The bulk density again depends on certain other parameters, such as the degree of packing. Packing, as discussed earlier, can be of three types and depends on the extent of tapping employed while compacting a granular bed of solids. The bulk density, when measured after compaction by tapping, is called the tap density. b. Angular Properties The angular properties that are significant in relation to studies on rheology of powders are 1. 2. 3. 4. 5.

Angle Angle Angle Angle Angle

of internal friction (αt ) of repose (θ) of wall friction (γ) of rupture (δ) of slide (ω)

III. GROUPING OF GAS FLUIDIZATION A. Hydrodynamics-Based Groups The type of fluidization specifically for gas fluidized beds is related to the properties of the gas and the solid. In a gas fluidized bed, the bubbles moving through the dense particulate phase have a strong influence on the quality of fluidization. Hence, it is important to define the types of fluidization with respect to the properties of the gas–solid system. 1. Geldart Groups In fluidization literature, most of the inferences are drawn from studies on one class of gas–solid system and then extrapolated to another group or class. This could have an adverse effect on scaleup and could result in the failure of the system. Much confusion and many contradictions in the published literature have

© 1999 by CRC Press LLC

been pointed out by Geldart,11 and these have been attributed to extending the data obtained on one powder to another powder. In view of this, Geldart12 classified powders that have similar properties into four groups and designated them by the letters A, B, C, and D. These groups are characterized by the difference in density between the fluidizing gas (i.e., air and the solid) and the mean size of the particles. A mapping of these groups is shown in Figure 1.5 for air fluidized beds. Of these

Figure 1.5 Geldart classification of powders. (From Geldart, D., Powder Technol., 7, 285, 1973. With permission.)

four groups, the two extreme groups are Group C, which is difficult to fluidize, and Group D, which is spoutable. The intermediate Groups A and B are suitable for the purpose of fluidization. Of these two groups, Group A powders have densephase expansion after minimum fluidization but prior to the commencement of bubbling, whereas Group B powders exhibit bubbling at the minimum fluidization velocity itself. Group A powders are often referred to as aeratable powders and Group B powders as sandlike powders. Detailed characteristics of powders that belong to the four groups are presented in Table 1.2. Geldart12 developed numerical criteria to differentiate Group A, B, and D powders. The numerical criteria for solid particle size (dp), density (ρs), and fluid density (ρf) are 1.

(ρs – ρf) dp ð 225

for Group A

(1.25)

Equation 1.25 is the boundary between Group A and Group B powders. 2.

© 1999 by CRC Press LLC

(ρs – ρf) dp2 Š 106

for Group D

(1.26)

Table 1.2 Geldart12 Classification of Powders Group A Example Particle size (dp), µm Density (ρs), kg/m3 Expansion Bed collapse rate Mixing Bubbles

Slugs

B

Cracking catalyst 30–100

Sand 40 < dp < 500

1) Split and recoalesce frequently Rise velocity > interstitial gas velocity For freely bubbling bed, rise velocity (30–40 cm/s) of small bubble ( interstitial gas velocity Size increases linearly with bed height and excess gas velocity No evidence Cloud-to-bubble-volume ratio not negligible Slugs at high velocity of gas, rise along wall and no evidence of breakdown

Group C Example Particle size (dp), µm Density (ρs), kg/m3 Expansion Bed collapse rate Mixing

Bubbling/ fluidization

Finer 1400 Solid particles are spoutable; hence expansion is similar to spouted bed Fastest of all groups because of dense or large size of particles Particle mixing as well as heat Solid mixing is relatively poor; high transfer between a surface and particle momentum and little particle bed are poorer than Group A and contact minimize agglomeration; gas B velocity in dense phase is high, and hence backmixing of dense-phase gas is less As the interparticle forces are Bubbles form at 5 cm above the greater than the force exerted by distributor fluid, the powder lifts as slug in Bubbles of similar size to those of small-diameter column or channel; Group B are possible at same bed hence bubbling is absent or not height and excess gas flow rate; reported largest bubbles rise slower than intertitial gas, and hence gas enters Agglomeration due to excessive the bubble base and comes out at the electrostatic force Fluidization is generally possible by top using agitator or vibrator to break the channels Electrostatic charges removed by using conductive solids or solids with graphite coating or column wall with oxide coating

© 1999 by CRC Press LLC

Equation 1.26 is the boundary between Group B and Group D powders, where density is expressed in grams per cubic centimeter and the particle diameter (dp) in micrometers. No numerical criterion or equation for the boundary line between Group A and Group C powders was proposed. The classification of Geldart groups has been well recognized and is often referred to in the literature, even though several other criteria based on similar conceptual premises were proposed later. 2. Molerus Groups In the Geldart classification of powders, one could observe that as the particle density (i.e., the difference in the densities of the solid and the gas [air]) decreases, the boundaries separating the groups shift toward larger particle sizes. The transition between the groups, more specifically from Group C to Group A or from Group A to Group B, is not observed to occur at sharp boundaries. Molerus13 proposed some criteria to group the powders by taking into account the interparticle force as well as the drag exerted by the gas on the particle. The powder classification diagram of Geldart,12 superimposed with the criteria of Molerus,13 results in a mapping of powder groups in the manner depicted in Figure 1.6, wherein sharp boundaries can be seen during the transition from one group to another.

Figure 1.6 Powder classification of Geldart as modified by Molerus. (From Molerus, O., Powder Technol., 33, 81, 1982. With permission.)

1. Criterion for transition from Group A to Group B:

(ρ

© 1999 by CRC Press LLC

s

– ρg

)

πd p3 g 6 Fe

= K1

(1.27)

The constant K1 is dependent on the nature of the powder (i.e., soft or hard). Adhesive or cohesive force should be estimated or known for these powders. 2. Criterion for transition from Group C to Group A:

(ρ

s

)

– ρ g d p3 g Fe

= K2

(1.28)

where K2 is again dependent on the nature of the powders (soft or hard), and hence two straight lines with a narrow band or strip result in the diagram. The slope of the straight line obtained for the transition either from Group C to Group A or from Group A to Group B, when plotted on a log scale, is –3, and so the lines are parallel. 3. Criterion for transition from Group B to Group D: (ρs – ρg) dp g = K3

(1.29)

The constant K3 for a stable spouted bed of sand14 (dp = 600 µm and ρs = 2600 kg/m3) is 15.3 N m–3. It should be noted that in the criteria proposed by Molerus, the effect of consolidation of the bed due to gravitational load prior to fluidization, the effect of humidity, and the electrostatic effect are neglected. For this reason, the criteria developed have some limitations. However, the mapping of powder groups which also satisfy the Geldart criteria is simple and exhibits a clear transition at the boundaries. 3. Clark et al. Groups The two-dimensional mapping of the powder groups as proposed by Geldart12 and Molerus13 suffers from no provision for identification by numerical means that would be useful for computer analysis. Clark et al.15 devised a method of representing the powder groups of Geldart by certain dimensionless numbers. The following ranges of numerical values for dimensionless numbers that represent different powder groups have been proposed: Powder number @ < 1.5 1.5 < @ < 2.5 2.5 < @ < 3.5 3.5 < @ < 4.5

Powder type C A B D

(1.30) (1.31) (1.32) (1.33)

where the powder number (@) that fits with the two-dimensional mapping of the Geldart and the Molerus groups is complex.

© 1999 by CRC Press LLC

The powder number that fits with the lines in the Geldart mapping12 is given by:

{

[

]

}

@ = 1.221 log d p – 2 log(1 + 0.112( ∆ρ) –1.5 ) – 0.0885

[

+ 0.901(log( ∆ρ) + 1.5 log d p ) – 0.526

[1 + tan h( – c1 )]

1 + tan h c2 4

[

1 + tan h1 2

]

(1.34)

(

)

]

+ 0.667 log( ∆ρ) + 2 log d p – 0.5

1 + tan h( – c2 ) 2

where

c1 =

c2 =

1000 – ( ∆ρ)d p3 2

(1.35)

150

14, 000 – ( ∆ρ)d p3 2

(1.36)

3000

It can be seen that the equation or the model developed to match the Geldart mapping is rather cumbersome and does not fit very well. The powder number for Molerus powder mapping13 is relatively simple and is expressed as:

[

]

) 1 + tan2 hC

(

@ = 0.454 log( ∆ρ) + 3 log d p – 0.395 – 0.9 log d p + 2.45

3

(1.37)

where

[

]

C3 = ( ∆ρ)d p3 – 1.73 × 10 7 2 × 10 6

(1.38)

The equation(s) for the powder number developed by Clark et al.15 are dimensionally inconsistent, and the units for particle diameter and particle density are in micrometers and grams per cubic centimeter, respectively. Nevertheless, it has the convenience of determining the powder group without having a recourse to a mapping of powder classification. Clark et al.15 recommend the development of correlations that relate the powder number to such fundamental parameters as the minimum fluidization velocity and the minimum bubbling velocity, especially for Group A and B powders. 4. Dimensionless Geldart Groups So far, gas fluidization powder grouping has been discussed on the basis of the two-dimensional powder mapping proposed by Geldart12 and Molerus13 and the numerical representation suggested by Clark et al.15 In view of the difficulties associated with these schemes, Rietema16 attempted to remap powder classification using dimensionless numbers, namely, the Archimedes number and the cohesion

© 1999 by CRC Press LLC

number (C/ρsgdp), where C is the cohesion constant. The dimensionless representation of the Geldart classification for powders belonging to Groups A, B, and C is shown in Figure 1.7.

Figure 1.7 Dimensionless representation of Geldart powders. (From Clark, N.N., Van Egmond, J.W., and Turton, R., Powder Technol., 56, 225, 1968. With permission.)

As can be seen from Figure 1.7, the voidage of the packed bed influences the transition from Group A to Group C. The transition or boundary between Group A and Group B powder behavior occurs when the fluidization number, N (= ρ3s dp4 g2/µ2f EM), is

N A– B =

150(1 – 0 )

2

20 (3 – 2 0 )

(1.39)

where EM is the elasticity modulus of the powder bed (N/m2). For transition from Group A and Group C behavior, one has: NA-C = Ar (Ncoh)–1

(1.40)

where Ncoh is the cohesion number (C/ρs gdp). The parameter C is a cohesion constant of powder (N/m2). The boundary between Group D and any other group powder behavior occurs when: Ar (ρg/ρs) > 60,000

© 1999 by CRC Press LLC

(1.41)

One point to note is that the viscosity has a profound effect on the boundary between Group A and Group C powder behavior. Fine cracking catalysts (dp Ý 30 µm), when fluidized by nitrogen or neon gas, behave like Group A powders (i.e., show homogeneous bed expansion), but the behavior changes to that of Group C powders when fluidization is effected by hydrogen. As yet, there is no theory or explanation available for the effect of viscosity on Group C powders. B. Hydrodynamics- and Thermal-Properties-Based Groups The classification scheme for powder characterization is generally based on the hydrodynamic characteristics or the fluidization properties of various powders with air at ambient temperature. Saxena and Ganzha17 pointed out that a particle that is larger according to hydrodynamics classification can be smaller in terms of thermal properties. For example, the Geldart12 criterion for Group D powders in the case of sand fluidized by air at ambient temperature and pressure leads to a particle size greater than 0.63 mm, and this size obviously falls in the large category according to hydrodynamics classification. On the other hand, they could still be considered to be small particles in terms of thermal characteristics. For example, for large particles, the heat transfer coefficient to the immersed surface increases with particle size. In other words, heat transfer is controlled by the gas-convective component. In the case of fine particles, the heat transfer coefficient decreases with an increase in particle diameter. A size of 1 mm has been proposed12 to be the demarcation between small and large particle sizes. In view of this, Saxena and Ganzha17 suggested that a powder should be classified by considering both hydrodynamic and thermal characteristics. The Archimedes number (Ar) along with the Reynolds number at minimum fluidization (Remf) have been considered for powder grouping due to the fact that Remf and the Nusselt number at its maximum value are unique functions of Ar. Saxena and Ganzha17 classified powders into three groups and demonstrated the validity of such a classification by considering heat transfer data for sand and clay at ambient temperature and pressure. This classification takes into consideration the heat transfer correlations and the models for large particles. The particle groups and the grouping criterion according to Saxena and Ganzha17 are given in Table 1.3. The quantity Ψ is given by:

⎡ µ 2g Ψ=⎢ ⎢ ρg g ρs – ρg ⎣

(

)

⎤ ⎥ ⎥ ⎦

13

The two subgroups shown in Table 1.3 depict the clear transition between Groups I and II as well as II and III. The classification criterion has a simple dimensionless form and is very useful for setting conditions while computing. However, it cannot identify powder types as originally conceived by Geldart.12

© 1999 by CRC Press LLC

Table 1.3 Criteria for Powder Classification Based on Hydrodynamics and Thermal Properties Group I IIA IIB III

Criteria or Relation 3.35 ð Ar ð 21,700 or 1.5 Ψ ð dp ð 27.9 Ψ 21,700 ð Ar ð 13,000 or 27.9 Ψ ð dp ð 50.7 Ψ 13,000 ð Ar ð 1.6 × 106 or 50.7 Ψ ð dp ð 117 Ψ Ar Š 1.6 × 106 or dp Š 117 Ψ

Ar = dp3 ρf (ρs – ρf)g/dp3 . Ψ = [µf2 /ρf (ρs – ρf)g]1/3. From Saxena, S.C. and Ganzha, V.L., Powder Technol., 39, 199, 1984. With permission.

C. Variables Affecting Fluidization In a fluidization bed, the fluidization medium is always comprised of solid particles. The fluidizing medium is any fluid: gas, liquid, or gas and liquid. In normal fluidization, the fluidizing fluid flows upward, thereby counteracting the gravitational force acting on the bed of particulate solid materials. The various parameters that influence the fluidization characteristics can be classified into two major groups comprised of independent variables and dependent variables. The properties of the fluid and the solid, pressure, and temperature are the major independent variables. The dependent variables include Van der Waals, capillary, electrostatic, and adsorption forces. The parameters that influence fluidization behavior can be depicted in a flow diagram (Figure 1.8).

Figure 1.8 Parameters affecting fluidization behavior.

D. Varieties of Fluidization Depending on the mode of operation and the flow regime, fluidization can be classified in several ways. For example, in normal fluidization, the gas flows upward and the particle density is greater than the fluid density. If a fluidized bed is

© 1999 by CRC Press LLC

homogeneous, it is often referred to as particulate. Heterogeneous fluidization is often bubbling or slugging in nature. Particulate fluidization is akin to liquid fluidization. Slugging occurs at high velocities in a gas fluidized bed with a narrowdiameter column and a deep bed. If the density of the particulate solids is less than that of the fluid, normal fluidization is not possible, and in such a case the fluid flow direction has to be reversed (i.e., the fluid has to flow downward). Such a situation prevails for fluidizing certain polymers. This type of fluidization is termed inverse fluidization. Fine powders are normally difficult to fluidize due to interparticle forces. In such a case, it is necessary to overcome the cohesive forces by some external forces in addition to the drag force exerted by the fluid flow. Agitation or vibration helps to overcome the interparticle forces. The fluidized bed in such a case is called a vibro fluidized bed. In all the above types of fluidization, gravitational force plays a key role. There is a minimum fluid flow required to overcome the gravitational force, and the flow rate required just for fluidization may be much more than that demanded by stoichiometry. This would finally result in wasting the unreacted fluid, which in turn increases the cost of recycling. Furthermore, it is not advisable to use such normal fluidization for a costly gas. One alternative is to use centrifugal fluidization. A new class of fluidized beds comes into play at very high velocities (much higher than or equal to the particle terminal velocity), where carryover or elutriation of the bed inventory occurs. Here the solids are to be recycled into the bed; such a system is called a circulating fluidized bed. Particulate solids fluidized by either gas or liquid consist of two phases and belong to the category of two-phase fluidization. A bed of solids fluidized by both gas and liquid is known as three-phase fluidization. Three-phase fluidization is more complex than two-phase fluidization and has additional classifications depending on the flow direction of the gas and the liquid. In general, they can flow either cocurrently or countercurrently. Figure 1.9 shows in a nutshell some common varieties of fluidization.

Figure 1.9 Some common varieties of fluidization.

© 1999 by CRC Press LLC

IV. HYDRODYNAMICS OF TWO-PHASE FLUIDIZATION A. Minimum Fluidization Velocity 1. Experimental Determination a. Pressure Drop Method This method of determining the minimum fluidization velocity (Umf) involves the use of data on the variation in bed pressure drop across a bed of particulate solids with fluid velocity. The trend in variation of the bed pressure drop with the superficial gas velocity was shown in Figure 1.3 and discussed in detail. Figure 1.10

Figure 1.10 Various experimental methods to determine minimum fluidization velocity: (A) pressure drop, (B) bed voidage, and (C) heat transfer.

depicts the various methods by which the minimum fluidization velocity can be determined. The plot in Figure 1.10A depicts bed pressure drop versus gas velocity. The transition point from the fixed bed to the fluidized bed is marked by the onset of constant pressure. This is also the point at which the increasing trend in the bed pressure drop (∆Pb) of a packed bed terminates. For an ideal case, gas flow reversal in the fluidized bed condition does not change the magnitude of ∆Pb. However, the value of ∆Pb is smaller when the bed starts settling during flow reversal compared to previous values obtained at the same velocity in the increasing flow direction. The pressure drop method is the most popular means of determining Umf experimentally.

© 1999 by CRC Press LLC

b. Voidage Method Bed expansion in a fixed bed is negligible. Hence, the bed voidage () remains constant. When a fixed bed is brought to the fluidized state, the bed voidage increases due to bed expansion. The onset of fluidization corresponds to the point where the voidage just starts increasing with the gas velocity (U). This is shown in Figure 1.10B. The bed voidage becomes constant and is equal to unity when the gas velocity corresponds to the particle terminal velocity. This method of determining Umf is not simpler than the bed pressure drop method, because the bed expansion cannot be accurately determined by any simple (i.e., visual) means. c. Heat Transfer Method The variation in the wall heat transfer coefficient (h) with gas velocity (U) forms the basis of one of the interesting methods of determining Umf. The wall heat transfer coefficient increases gradually in a fixed bed as the gas velocity is increased, and it suddenly shoots up at a particular velocity, indicating the onset of fluidization. The velocity that corresponds to the sudden increase in the wall heat transfer coefficient is the minimum fluidization velocity. The trend in the variation of h with U is shown in Figure 1.10C. In this method, another important gas velocity, the optimum gas velocity (Uopt), which corresponds to the maximum wall heat transfer coefficient (Umax), is also obtained. This method of determining Umf is more expensive than the two aforementioned methods, and it requires good experimental setup to measure the heat transfer data under steady-state conditions. For these reasons, this method is seldom used to determine Umf in fluidization engineering. 2. Theoretical Predictions The various methods, based on first principles, available in the literature for the theoretical prediction of the minimum fluidization velocity can be broadly classified into four groups. These methods are derived from (1) dimensional analysis, (2) the drag force acting on single/multiparticles, (3) the pressure drop in a fixed bed extendable up to incipient fluidization, and (4) a relative measure with respect to the terminal particle velocity. Although a significant body of literature pertaining to the above methods exists and numerous correlations are listed by various researchers, there has been no classification of these correlations. The aforementioned methods are briefly described and the correlations based on them are listed in the following sections. a. Dimensional Analysis (Direct Correlation) This conventional method of developing a correlation takes into consideration the physical properties of the fluid and the solid. In its most general form, the correlation is given as:

© 1999 by CRC Press LLC

(

U mf = K d pa ρs – ρ f

) ρ µ g (φ b

c f

d f

c

m n s mf

)(1 – ) mf

p

(1.42)

Correlations of the above type are presented in Table 1.4. It can be seen that the factor ( φ sm nmf ) is variable because of the dependence of mf on φs. However, many correlations have been proposed by assuming that the quantity K φ sm nmf (1 – mf)p is a constant. This assumption is valid only within a certain range of experimental parameters. This type of direct correlation has the inherent disadvantage of involving a dimensional constant (K) that changes depending on the system of units used for the variables in Equation 1.42. b. Drag Force Method In this method, a force balance is assumed at incipient fluidization. The force balance equation is Fg = αFD + FB

(1.43)

where Fg, FD, and FB are, respectively, the force due to gravity, drag, and buoyancy, and α is a correction or multiplication factor for a multiparticle system. Because the drag force varies depending on the flow range, there can be several correlations for the prediction of Umf by this method. Taking 2 FD = α(1 2)CDρ gU mf (π 4)d p2

(1.44)

and using appropriate expressions for Fg and FB for a particle of diameter dp fluidized by a fluid of density ρg, a general equation can be written as:

3 α ⋅ CD Re 2mf = Ar 4

(1.45)

The above equation can be solved if an appropriate value of drag coefficient (CD) is chosen, depending on the flow regime. For Stokes flow (Remf ð 0.1), CD = 24/Remf. Hence Equation 1.45 becomes

Re mf =

Ar 18α

(1.46)

For Newton’s flow regime (i.e., for turbulent flow condition), Remf > 500 and CD = 0.44. Hence,

Re 2mf =

© 1999 by CRC Press LLC

Ar 0.33α

(1.47)

Table 1.4 Correlation for Minimum Fluidization Velocity Based on Dimensional Analysis Method n (1 – )p A. General Form for Umf = Ko dpa (ρs – ρf)b ρfc µfd gc φsm emf mf m n p K = Ko φ emf (1 – emf)

Ref.

a

b

Exponential Factor d e K

c

m

n

p

Remarks a GF, SI, Remf < 5 GF, SI units

Leva18

1.82 0.94

–0.06

–0.88 1

7.169 × 10–4

—

—

—

Miller and Logwinuk19 Rowe and Yacono20 Davidson and Harrison21 Yacono22

2

0.1

–1

1

0.00125

—

—

—

1.92 0

0

0

0

0.00085

0

2

1

0

–1

1

0.0008

—

1.77 0

0

0

0

0.000512

0

0

0

–1

0

0

0.361

0

0

0

—

—

—

Umf = cm/s, dp = µm (ρs – ρf) = µsb (bulk density) SI units SI units

—

—

—

CGS units

—

—

—

SI units, Remf < 5

0.9

Baerg et al.23 1.23 1.23

Kumar and 1.34 0.78 –0.22 0.56 0.78 0.005 Sen Gupta24 Baeyens and 1.8 0.934 –0.066 –0.87 0.934 9.125 × 10–4 Geldart25 Leva et al.26 1.82 0.94 –0.06 — — 7.39

2.92 0.945 Umf = cm/s, dp = µm — — All SI units

B. General Form for Re mf Remf = K (dp3 ρf2 g/µf2 )x Ref. Riba27 Ballesteros28 Doichev and Akhmakov29 Thonglimp et al.30 a

K 1.54 12.56 1.08 7.54 1.95

× × × × ×

⎛ ρ s – ρf ⎞ ⎜ ⎟ ⎝ ρf ⎠

y

x 10–2 10–2 10–3 10–4 10–2

0.66 1.0523 0.947 0.98 0.66

y 0.7 0.66 + 1.0523 0.947 0.98 0.66

Remarks

Remf < 30 30 < Remf < 180

GF = general form.

In the intermediate flow regime, there can be as many as six empirical correlations for CD, as proposed by Morse.31 The drag force on a single sphere situated in an infinite expanse of fluid, according to Schiller and Naumann,32 is

FD =

© 1999 by CRC Press LLC

πd p2 ρ gU gc

(3 Re + 0.45 Re

–0.313

)

(1.48)

for 0.001 < Re < 1000. The correlation given by Equation 1.48 fairly covers all three flow regimes and thus can be used in Equation 1.43 to predict Re at Umf conditions for most gas–solid systems (i.e., for Group A to C particles of Geldart’s classification). The factor α in Equation 1.43 has been found to be a function of voidage () by Wen and Yu.33 They found experimentally that for particulate fluidization: α = f() = –4.7

(1.49)

Using Equations 1.48 and 1.49 and the appropriate expressions for FB and Fg in Equation 1.43, the general form of the correlation for obtaining Remf in the range 0.001–1000 is

2.7 Re1mf.687 – 18 Re mf – 4.7 Ar = 0

(1.50)

for 0.818 < ρf (kg/m3) < 1135, 6 × 10–4 < dp (µm) < 25 × 10–2, 1060 < ρs (kg/m3) < 11,250, 1 < µ (CP) < 1501, 244 × 10–5 < (dp/D) < 10–1. Assuming (mf = 0.42, Wen and Yu33 proposed a simplified form of the above correlation as:

159 Re1mf,687 – 1060 Re mf – Ar = 0

(1.51)

The solution for Remf using Equation 1.50 or 1.51 is not straightforward. On the other hand, Remf can be obtained more easily by using appropriate expressions for different flow conditions as given by Equations 1.46 and 1.47. Using appropriate correlations33 for FD and δ, the following relationship can be arrived at:

⎛U ⎞ α=⎜ t⎟ ⎝U⎠

β

(1.52)

where β = 1 for laminar flow (0.001 ð Re ð 2), 1.4 for intermediate flow (2 < Re < 500), and 2.0 for turbulent flow (Re > 500). It may be recalled here that Richardson and Zaki34 experimentally obtained the relationship

Ut = –n U

(1.53)

which indicated that α is a function of voidage only. In light of the above analysis of the development of an expression for Re at Umf conditions, a variety of correlations can be obtained, depending on the range of Re and the corresponding equation for the drag coefficient. Most expressions available for predicting CD are based on experimental results pertaining to spherical particles, and no single correlation is able to cover the entire (laminar to turbulent) range of flow conditions. The right

© 1999 by CRC Press LLC

choice of CD for a nonspherical particle is often difficult. More recently, Haider and Levenspiel35 presented explicit equations for predicting CD for spherical and nonspherical particles. For spherical particles:

CD =

(

)

24 0.4251 1 + 0.1806 Re 0.6459 + + Re 1 6880 .95 Re) (

(1.54)

for Re ð 2.6 × 105. For nonspherical, isometric particles:

CD =

[

]

24 1 + 8.1716 (exp– 4.0655 φ s ) Re Re

[

]

(1.55)

+ 73.69 Re exp ( –5.0748 φ s ) Re + 5.378 exp(6.2122 φ s )

for Re ð 25,000. A more accurate correlation with rms Ý 3.1, as given by Turton and Levenspiel,36 is cumbersome. Equation 1.55 is considered to be a simple form wtih an rms error of 5%. The drag coefficient in the case of a porous spherical particle falls outside the ambit of the above type of correlations. Masliyah and Polikar37 reported from their experimental findings that the drag experienced by a porous sphere is less than that experienced by an impermeable sphere of the same diameter and bulk density. However, at high values of the Reynolds number, the effect of inertia on a porous sphere is found to be greater than that on a comparable impermeable sphere. The drag coefficient proposed by Masliyah and Polikar37 for a porous sphere is given by:

CD = 24

[

Ω K – K log Re 1 + K5 Re ( 6 7 10 ) Re

]

(1.56)

where

Ω=

[

2 ψ 2 1 – (tan h ψ ) ψ

[

]

2 ψ + 3 1 – (tan h ψ ) ψ 2

]

and Ψ = d p

k pe

The values of K5, K6, and K7 are 0.1315, 0.82, and 0.05, respectively (for 0.1 < Re < 7), and 0.0853, 1.093, and 0.105, respectively (for 7 < Re < 120). A list of correlations based on the drag force principle or with similar forms is given in Table 1.5. c. Pressure Drop Method The most general form of expression for the pressure drop through a fixed bed of particulate solids can be given as:

© 1999 by CRC Press LLC

Table 1.5 Correlations for Remf Based on Drag Force Principle or Similar Form as Derived from Drag Force Principle Ref.

Remarksa

K2 A. General Form for Low Remf Remf = K2 Ar 8.1 × 10–3

Rowe and Henwood38 Frantz39 Wen and Yu33 Davidson and Harrison21 Davies and Richardson40 Pillai and Raja Rao41

GF

1.065 × 10–3 –4.7/18 0.00081

Remf < 32, GF 0.001 < Remf < 2, GF GF

7.8 × 10–4

GF

7.01 × 10–4

Re < 20, GF

B. General Form for All Ranges of Re mf Remf = [(4/3) (Ar/α)/CD]1/2, where α = e–4.7 Schiller and Naumann32 Morse31

CD = 3/Remf + 0.45 Remf–0.313 CD = C1/Re + C2/Re2 + C3 C1

Haider and Levenspiel35

0.001 < Remf < 1000, for spherical particle For spherical particle C2

C3

Remf Range

24 22.73

0 0.0903

0 3.69

29.1667 46.5

3.8889 116.67

1.222 0.6167

98.33 148.62

2,778.0 4.75 × 104

0.3644 0.357

490.546 1,662.5 0 See Equation 1.54

57.87 × 104 5.4167 × 106 0

0.46 0.5191 0.44

Remf ð 0.1 0.1 ð Remf ð 1 1 ð Remf ð 10 10 ð Remf ð 100 100 ð Remf ð 1,000 1,000 ð Remf ð 5,000 5,000 ð Remf ð 10,000 104 Remf ð 5 × 104 Remf Š 5 × 104

Remf < 2.6 × 105 for spherical particle Remf < 25,000 for nonspherical particle, φ Š 0.67

See Equation 1.55

a

GF = general formula.

(∆P H ) = f (ρ U 2) 6 (1 – ) ( Φ d ) b

g

2 g

3

s

p

(1.57)

where the bed friction factor ( f ) is a function of bed voidage () and the particle Reynolds number (Re). At the incipient fluidization condition, taking ∆Pb /H = (ρs – ρg) (1 – mf)g, Equation 1.57 can be transformed to:

© 1999 by CRC Press LLC

Ar =

1 Re 2mf fk φ s 3mf

(1.58)

Ergun42 proposed that the friction factor (fk) for randomly packed beds can be expressed as:

fk =

150(1 – ) + 1.75 φ s Re

(1.59)

Substituting Equation 1.59 in Equation 1.58 at Umf condition and rearranging:

(

)

1.75 150 Re 2mf + 2 3 1 – mf ⋅ Re mf = Ar 3 φ s mf φ s mf

(1.60)

Wen and Yu33 were the first to use this type of correlation and to solve it for Remf. In order to arrive at a suitable solution, Wen and Yu collected the data for mf and φs and made the following approximations:

1 – mf φ 2s 3mf

⯝11

and

1 ⯝14 φ s 3mf

(1.61)

The Wen and Yu33 correlation expressed using Remf and Ar is

24.5 Re 2mf + 1650 Re mf = Ar

(1.62)

The above correlation has been found to fit experimental data well in the range 0.001 < Remf < 4000, and this form of correlation is cited in the literature frequently. The Wen and Yu33 correlation has been refined to increase the accuracy of prediction, and several correlations of such type have been introduced by altering the coefficients of Remf2 and Remf. The friction factor (ff) for a fluidized or sedimenting suspension can be expressed after Davidson and Harrison21 as: ff = 3(ρs – ρf)g/SρfU2

(1.63)

For a packed bed one has:

∆P ⎞ fp = ⎛ ⎝ SH 1 – ⎠

(ρ U f

2

2

)

(1.64)

Usisng Carman’s proposed equation43 for ∆P with a constant of 5, one obtains:

© 1999 by CRC Press LLC

⎡ 5(1 – )2 2 ⎤ 1 2 fp = ⎢ ⋅ S µ H U⎥ ⋅ ⋅ 2 3 ⎣ ⎦ SH 1 – ρ f U

(1.65)

The above friction factors have been evaluated, and correlations in terms of Rel based on a linear dimension analogous to the hydraulic mean diameter, i.e., /(1 – )S, have been reported as: ff = 3.36/ReL

for ReL 2 × 10 4

(1.103)

For elevated pressures and temperatures, the method of Yang et al.86 should be used; it is based on graphical methods using (Re2 · CD)1/3 versus (Re · CD)1/3 plots for various mf values that are functions of pressure and temperature. Having selected an appropriate correlation for Remf and then for Ret as outlined, it is possible to generate a plot of Ret/Remf versus Ar. Richardson59 utilized a similar

© 1999 by CRC Press LLC

kind of technique and generated a plot using the Ergun equation for spherical particles and taking mf = 0.4 to predict Remf. The following correlations were employed to predict Ret: Ar = 18 Ret Ar = 18 Ret + 2.7 Ret1.87 Ar = 1/3 Ret2

for Ar < 3.6

(1.104)

for 3.6 < Ar < 105

(1.105)

for Ar > 105

(1.106)

His plots for mf = 0.38, 0.4, and 0.42 fit well with experimental data and show that: Ret/Remf = constant

for Ar < 1

(1.107)

where the constant = 64 for mf = 0.42 = 78 for mf = 0.4 = 92 for mf = 0.38 and Ret/Remf = Ý10

(1.108)

at different values of mf (0.38, 0.4, 0.42) for Ar > 105.

V. FLOW PHENOMENA A. Particulate and Aggregative Fluidization Particulate or smooth fluidization, which is normally exhibited by a liquid–solid system, should be distinguished from aggregative- or heterogeneous-type fluidization that prevails in a gas–solid system. There is not much work available on this subject to enable one to distinguish between particulate and aggregate fluidization in a precise manner. Wilhem and Kwauk87 proposed a dimensionless group to explain the quality of fluidization as follows: Frmf < 0.13 for smooth or particulate fluidization > 1.3 for aggregative or bubbling fluidization

(1.109) (1.110)

where Frmf is the Froude number (Umf2 /dpg). It may be recalled here that Equation 1.62 for the determination of Umf, when used for inertial flow or Newton’s flow regime, results in the equation

© 1999 by CRC Press LLC

Re 2mf =

Ar 24.5

(1.111)

which can be simplified to:

⎛ ρs – ρ f ⎞ Frmf = 0.0408⎜ ⎟ ⎝ ρf ⎠

(1.112)

For most gases ρf < 10 kg/m3, and for liquids ρf > 500 kg/m3. For glass beads of ρs < 2500 kg/m3, when fluidized by gas, ρs/ρf > 250, and the values obtained are Frmf Š 10.1, Hence, the glass bead–gas fluidized bed system is expected to behave like an aggregative fluidization. Applying a similar rule for a water–glass bead system, one gets ρs/ρf > 2500/1000 and Frmf ⯝ 0.0612 (which is less than 0.13). Thus the system exhibits smooth fluidization. In the above examples, it is possible to obtain an expression for the Froude number directly by using the densities ρs and ρf for the flow regime dominated by the inertial force, more precisely for Rep > 1000. For Rep < 20, the viscous force is dominant and the expression for Frmf is obtained in terms of the density ratio (ρs – ρf )/ρf and Remf:

Frmf =

ρs – ρ f 1 ⋅ Re mf ⋅ 1650 ρf

(1.113)

For example, a quantitative determination of Frmf when Remf = 20 can be carried out using Equation 1.113. For the glass bead–air system (ρs = 2500 kg/m3 and ρf = 1.2 kg/m3) when Remf = 20:

Frmf =

20 ⎛ 2500 ⎞ ⋅ – 1 = 0.0121(2082.3) = 25.2 ⎠ 1650 ⎝ 1.2

Because Frmf > 1.3, an aggregative type of fluidization occurs for the glass bead–air system. Similarly, for the glass bead–water system (ρf = 1000 kg/m3), when Remf = 20: Frmf = 0.018 Because Frmf < 0.13, smooth fluidization is possible with the glass bead–water system. The very fact that the quality of fluidization cannot be assessed in terms of a single dimensionless group like the Froude number has prompted researchers to evaluate the quality of fluidization by using a larger number of dimensionless groups. Romero and Johanson88 proposed four dimensionless groups to assess the quality of fluidization:

© 1999 by CRC Press LLC

Frmf ,

Re mf ,

ρs – ρ f ρf

and

Hmf Dt

The quality of fluidization is related to its stability, which would be affected by bubbling. An increase in each of the four dimensionless groups will lead to poor quality or unstable or bubbling fluidization. From the experimental findings, the criterion for grouping the two modes of fluidization may be assessed by the product of the four dimensionless groups:

⎛ ρs – ρ f Hmf ⎞ < 100 ⋅ ⎜ Frmf ⋅ Re mf ⋅ ρ Dt ⎟⎠ ⎝ f

(1.114)

for particulate or smooth fluidization and:

⎛ ρs – ρ f Hmf ⎞ > 100 ⋅ ⎜ Frmf ⋅ Re mf ⋅ ρ Dt ⎟⎠ ⎝ f

(1.115)

for aggregative or bubbling fluidization. Classifying the mode of fluidization in terms of the product of four dimensionless groups seems to be superior to using the single Froude group because almost all variables are considered in the former. However, there is no theoretical proof available to validate the aforementioned criterion; therefore, the criterion for identifying the fluidization mode can be regarded merely as a useful rule of thumb. B. Regimes of Fluidization 1. Bubbling Bed In a gas fluidized bed, right from the onset of fluidization, a change in the fluidization regime or the bed behavior would occur with an increase in the gas flow rate. The first fluidization regime prevails when bubbles are formed in the bed from the attainment of a minimum gas velocity designated as the minimum bubbling velocity. Bubbles once formed in the bed start rising, grow in size, coalesce, reach the bed surface, and finally erupt. In a tall column with a small diameter, the bubbles coalesce as they rise up and form slugs whose size or diameter would be the same as the column diameter. Generally, the slugs are followed by a piston of solids; once carried up to the top surface of the bed, the solids rain down along the column wall. When a fluidized bed of a larger diameter is used for a bubbling fluidized bed, the formation of slugs can be avoided or minimized. Further, with a large-diameter fluidization vessel and shallow beds, bubbles have less of a chance to coalesce. The resulting bubbling fluidized bed would then have bubbles of more or less uniform size. The bubbling point varies depending on the powder group (as classified by

© 1999 by CRC Press LLC

Geldart12). Bubbling in a fine powder system is quite different from one comprised of coarse particles. With fine powders, the bed may expand uniformly without any bubble formation, and particulate fluidization can occur until the first bubble is formed at a gas velocity known as the minimum bubbling velocity (Umb). Experimental measurement of this velocity is not easy and requires careful assessment of the bed behavior. Just at the minimum bubbling point, the expanded particulate bed will show a decrease in its height without a change in the overall voidage of the emulsion phase. However, the structure of the bed is changed. The simplest correlation proposed by Geldart89 for the minimum bubbling velocity (Umb) is Umb = Kmb dsv

(1.116)

where Kmb = 100 if Umb and dsv are expressed in CGS units and dsv = 1/Σxidsvi is the particle surface/volume diameter (m–1). Abrahamsen and Geldart5 determined the ratio of the minimum bubbling velocity (Umb) to the minimum fluidization velocity (Umf) as:

(

0.126 0.523 exp 0.716 Ff U mb 2300ρ g µ = 0.934 U mf d p–0.8 g 0.934 ρs – ρ g

(

)

)

(1.117)

It is, in general, observed that the addition of fines ( 100

(1.136)

For Rep < 100, the contribution of the convective component is negligible and Nup = 2 when Rep tends to zero. The gas velocity relative to a moving particle is less, and this leads to a lower value of Nup in a fluidized bed than in a packed bed. Measurements of the temperatures of individual particles and of the gas are not possible by using a bare thermocouple, which can measure only the mean temperature of the emulsion packet. Therefore, experimental determination of Nup values is difficult. Two types of experimental techniques were proposed by Kunii and Levenspiel58 to predict the fluid–particle temperature, and these techniques are briefly discussed below. a. Steady State In this method, the heat lost by the gas stream is assumed to be equal to the heat gained by the solids. For example, the heat balance for a section of a bed of height dH is –CpgUgρgdTg = hgp(Tg – Tp)dH (Heat lost by gas)

(Heat gained by solids)

(1.137)

The value of the heat transfer coefficient (hgp) as determined by the above equation is only an apparent value, if axial conduction is not considered. In order to arrive at the true value, the right-hand side of the equation, that is, the heat gained by the solids, should be reduced by an amount equivalent to the axial conduction loss. b. Unsteady State In this method, the heat lost by the gas stream over a differential section of the fluid–solid mixture is considered to be the same as the heat accumulated by the solids. The final equation which could be used to estimate hgp by this method is

© 1999 by CRC Press LLC

ln( ∆T ∆Ti ) = –

C pgU g ρ g A ⎡ hgp H ⎤ ⎥t ⎢1 – exp W C ps ⎢⎣ C pgU g ρ g ⎥⎦

(1.138)

Experimental investigations by Chen and Pei126 on fluid-to-particle heat transfer in fluidized beds of mixed-size particles showed that the addition of coarse particles to a fine-particle system improves the fluidization and hence increases the fluid–particle heat transfer coefficient. The most effective concentration of the coarse particles is about 25% and the optimum particle diameter ratio is Ý7. The correlation proposed by Chen and Pei126 to predict fluid–particle heat transfer is jH (1 – ) = 0.000256 Ar0.4 Pr2/3

(1.139)

for 10 < Ar < 105 and 0.1 < Rep < 50 where jH = Colburn’s factor = Nu/Re Pr1/3. 2. Bed–Wall Heat Transfer The overall bed–wall heat transfer coefficient (hw) is more useful for design purposes than the local heat transfer coefficient. Heat transfer measured at various locations or at the same location at different times usually fluctuates. It is thus necessary to be able to estimate the local heat transfer coefficient in order to develop a suitable model that could help in predicting the overall mean value. In a gas fluidized bed, hw in principle increases with increasing gas velocity until it reaches a maximum value; thereafter hw falls with further increase in gas velocity. The trend in the plot of hw versus U is the same as that shown in Figure 1.10C. Thus, in a gas fluidized bed, at gas velocities beyond Umf, the plot of hw versus U has two sections: (1) a rising branch up to hmax, corresponding to a velocity designated as Uopt, and (2) a falling branch beyond U = Uopt. C. Models Various models have been proposed to explain the variation in hw with gas velocity, and this topic was reviewed in detail by Gutfinger and Abuaf,127 Bartel,128 and Saxena and Gabor.129 Among these models, three distinct groups may be identified, as indicated in the following sections. 1. Film Model A thin fluid film adjacent to the heat transfer wall is considered to offer the main resistance to heat transfer and this resistance is reduced by the moving particles which scour away the fluid film. Steady-state heat transfer is assumed. This model/hypothesis was supported by Van Heerden et al.130 and Ziegler and Brazelton.131 However, the results based on this model were subsequently found to be in poor agreement with many other experimental data. The drawback of this model lies in its failure to consider the properties of solid particles, which play a vital role in the transport of heat in a fluidized bed.

© 1999 by CRC Press LLC

2. Modified Film Model Transient conduction of heat between the surface and the particles (single/multiparticles)131–134 is assumed in a particulate type of fluidization. The solid particle properties are considered, and the heat transfer rate is assumed to change depending on the mobility of the solid particles and their concentration on the heat transfer surface. As this model cannot be applied to a bubbling bed, a third type of model, known as the packet model, has emerged. 3. Emulsion Packet Model A packet of emulsion (a mixture of solid and gas corresponding to the incipient state of fluidization) contacts the heat transfer surface periodically, undergoes transient conduction during its stay, moves away (making way for a new packet), and transports the sensible heat to the bulk of the bed. Unsteady-state approach and surface renewable rate of the packets are the main features of this packet. Mickley and Fairbanks135 were the first to propose a model of this kind. The emulsion packet model does not consider the individual properties of the gas and the solid. Instead, it considers the properties of the homogenous gas–solid mixture at the incipient state. Botterill and Williams,136 in their single-particle model, considered the individual properties of the gas and the solid. The single-particle model has the limitation of a short contact time during which heat cannot penetrate into a second particle; for this reason, this model has been extended to a two-particle layer by Botterill and Butt137 and further to a chain of particles of unlimited length by Gabor.138 D. Predictions of Heat Transfer Coefficient There are numerous correlations available in the literature for predicting the wall heat transfer coefficient, and these correlations were developed under different conditions. Zabrodsky,139 therefore, recommended that care be taken to use these correlations only under the conditions for which they were proposed. The applicability of any mechanism warrants detailed knowledge of the transient contact characteristics between the emulsion and the heat transfer surface on a local basis. The literature on this subject is scant. Selzer and Thomson140 pointed out that a model should clearly explain the heat penetration depth and the contact time criteria. They also stressed the need to test the applicability of these models for different types of distributors. A detailed listing of all the available correlations to predict the wall heat transfer coefficient can be found in the literature. In general, the heat transfer coefficient is a function of (1) the properties of the gas and the solid (ρg, Kg, Cpg, µg, ρs, Cps, Ks), (2) the relevant geometric factors (dp, Dt), (3) the fluidization parameters (θD, fL, p, b), (4) the system variables (U, Umf), (5) the location of the heat transfer surface (immersed) inside the bed, and (6) the type and design of the grid/distributor plates.

© 1999 by CRC Press LLC

1. Additive Components The wall heat transfer coefficient, also called the convective heat transfer coefficient, is generally constituted of two additive components: the particle-convective component (hpc) and the gas-convective component (hgc). The particle-convective component arises due to the motion or mobility of the particulate solids over the heat transfer surface and is more pronounced for fine particles than for coarse particles. The gas-convective component is due to the gas percolating through the bed and to gas bubbles. The gas-convective component is negligible in fine particle fluidization but is predominant in coarse or heavy particle fluidization; the reason is the requirement for high flow rates to achieve fluidization in the latter case. The overall convective heat transfer is given by: hc = hpc + hgc

(1.140)

Prediction of hpc is difficult due to the complex hydrodynamics of fluidization and the interaction of the lean and the dense phases over the heat transfer surface. a. Particle-Convective Component The particle-convective component (hpc), in the absence of any film resistance, is hp and should be evaluated from the knowledge of the instantaneous heat transfer coefficient (hi) and the age distribution of the packet of emulsion over the transfer surface. Mickely and Fairbanks135 proposed a model to evaluate hp as:

hp =

1 τ

t

∫ h dt 0

(1.141)

i

where the instantaneous heat transfer coefficient (hi) is given by: hi = (KeρeCps/πt)

(1.142)

and the packet residence time (τ) is given by:

⎤ ⎡ ⎥ ⎢ dpg ⎥ τ = 0.44 ⎢ 2 ⎢ ⎞ ⎥ ⎛ U 2 ⎢ U mf ⎜ – 1⎟ ⎥ ⎢ ⎠ ⎥ ⎝ U mf ⎥⎦ ⎢⎣

0.14

dp Dt

(1.143)

If the conditions correspond to slugging, the packet residence time (τ) over a short vertical surface of length L is

© 1999 by CRC Press LLC

τ = L/(U – Umf)

(1.144)

Under all practical conditions, the heat transfer is effective only when the packet contacts the surface. For the fraction of surface which is shrouded by gas bubbles (fb), hp is negligible. Hence, hpc, in the absence of any film resistance, can be expressed as: hpc = hp(1 – fb)

(1.145)

where

(

) ⎤⎥

⎡U 2 U U – A mf mf fb = 0.33⎢ dpg ⎢ ⎣

0.14

(1.146)

⎥ ⎦

The accuracy of the determination of the average value of hpc can be improved to better than 5% by taking into consideration the film resistance (1/hf) which is in series with the packet resistance (1/hp). When this is done, the particle-convective component of heat transfer (hpc) is given by:

hpc =

1

( ) ( ) 1 hp + 1 h f

(1 – f )

(1.147)

b

The film heat transfer coefficient (hf) is given by: hf = mcKg/dp

(1.148)

where mc has varying values; the range of these values under different experimental conditions was enumerated by Xavier and Davidson.141 They also suggested that mc = 6 may be taken for design purposes. In order to evaluate hpc, it is necessary to know the effective thermal conductivity of the particulate bed (Ke), which is the sum of the effective thermal conductivity of the bed when there is no fluid flow ( Keo ) and the conductivity due turbulent diffusion ( Ket ). Thus,

Ke = Keo + Ket

(1.149)

Keo can be evaluated142,143 by using the expression Keo ⎛ K s ⎞ = K g ⎜⎝ K g ⎟⎠

© 1999 by CRC Press LLC

0.28– 0.757

log10 –0.057 log10

Ks Kg

(1.150)

The contribution of turbulent heat diffusion ( Ket ) in determining the effective thermal conductivity is given by:

Ket K g = (α1 , β1 ) Re p Pr

(1.151)

Here, α1 is the ratio of the mass velocity of the fluid in the direction of heat flow to the superficial mass velocity in the direction of fluid flow. The parameter β1 is the ratio of the distance between the two adjacent particles to the mean particle diameter. The product α1β1 for small values of d/D is equal to 0.1 for normal packing of spheres. Hence, the effective conductivity of the particulate phase or the emulsion packet is

Ke = Keo + 0.1 ρ g C pg d pU mf

(1.152)

where the value of Umf can be predicted as per the correlations presented in Tables 1.4–1.6. The particle-convective heat transfer coefficient can be calculated from Equation 1.147 by using the appropriate expression for hp (Equations 1.141–1.146 and Equation 1.148). b. Gas-Convective Component Because the interphase gas-convective component (hgc) cannot be determined directly, Baskakov et al.145 proposed from an analogy to mass transfer that:

Nu gc =

hgc d p Kg

= 0.009 Ar 1 2 Pr 1 3

(1.153)

For most gases, Pr is constant and for the inertial flow regime Remf α Ar1/2. Hence, Nugc α Remf. In view of this, it has been assessed that the value of hmf obtained at Umf is approximately the same as that of hgc. However, for a bubbling fluidized bed at a high fluid flow rate, the equation hgc = hmf(1 – B) + hBB

(1.154)

can be approximated to hgc Ý hmf. Botterill and Denloye146 proposed the following correlation for predicting hgc by taking into account the heater length (L):

hgc dL K g = 0.3 Ar 0.39

for 10 3 < Ar < 2 × 10 6

(1.155)

c. Radiative Component The radiative heat transfer coefficient (hr) between a fluidized bed at a temperature Tb and a heater surface like an immersed tube at a temperature Ts is

© 1999 by CRC Press LLC

(

)

hr = qr (Tb – Ts ) = σ s Ebs Tb2 + Ts2 (Tb + Ts )

(1.156)

where qr is the radiant heat flux, σs is the Stefan–Boltzmann constant (5.67 × 10–8 W/m2 K4), and Ebs is the generalized emissivity factor given by: Ebs = (1/Eb + 1/Es – 1)

(1.157)

The generalized emissivity factor depends on the shape, disposition, and emissivity of the radiating and receiving bodies. Radiation plays a major role in heat transfer when the temperature of the radiating body is above 800–900°C. The convective heat transfer coefficient increases due to the component hr, which is significantly high when the temperature of the radiating body is above 1000°C. In the case of many a fluidized bed combustion, the bed temperature (Tb) itself is around 1000°C. At high temperatures or at high values of the difference between Tb and Ts, it is important to evaluate the intermediate mean temperature at which the physical properties of the fluid and the solid must be determined. The intermediate temperature147 (Te) is given by:

Tb – Te Tb – Tw = 0.5 Rr 0.5 Rr + Rw

(1.158)

Rr is the mean thermal resistance of the packet, that is,

Rr =

πt Ke C ps ρ B

where Cs = Cp (1 – mf), ρB is the bulk density of the packet = ρbmf, and Rw is the contact resistance of the zone near the wall and equals δw/Kew. 2. Overall Heat Transfer Coefficient The overall heat transfer coefficient is the sum of the individual convective components that are due to bubbles (hb), particles (hpc), interphase gas (hgc), and radiant heat transfer (hr). Thus, the overall or total heat transfer coefficient (h) is given by: h = hb fb + (hpc + hgc) (1 – fb) + hr

(1.159)

In the case of a dense fluidized bed where there are no bubbles and the fraction of the immersed surface covered by bubbles (fb) is zero, the total heat transfer coefficient (h) is given by: h = hpc + hgc + hr

© 1999 by CRC Press LLC

(1.160)

For large particles, due to the requirement of a high fluidization velocity, it can be assumed that hb Ý hgc, so that: h = (1 – fb) hpc + hgc + hr

(1.161)

In the case of solid particles whose emissivity lies in the range of 0.3–0.6, Baskakov148 proposed the following expression for the overall heat transfer coefficient under maximum condition:

(

)

hmax = K g 0.85 Ar 0.19 + 0.006 Ar 0.5 Pr 0.33 d p + 7.3 E p Es Ts3

(1.162)

E. Heat Transfer to Immersed Surfaces The immersed surfaces inside a fluidized bed can be of any object meant for transferring the heat either from the bed or to the bed, depending on whether the bed is to be cooled or heated. Generally, tubes or tube banks are immersed in largescale fluidized beds for exchanging the heat. Leva18 reported that, except at high flow rates of the fluidizing fluid/gas (that is, for Umf >> 1), the heat transfer rate obtainable with immersed surfaces is fourfold higher than that between the bed and the external wall. There have been many investigations1,7,149–156 into this subject. The heat transfer coefficients relevant to heat transfer to immersed objects are broadly of two types: local heat transfer coefficients and total or overall heat transfer coefficients. Because the latter coefficients are useful for design purposes, this topic is given special emphasis in the following discussion. 1. Vertical Surfaces Studies135 on vertically immersed electrical heaters have shown that the heat transfer coefficients are smaller than those corresponding to horizontal surfaces under similar experimental conditions. This has been attributed to the small size of the heater over which large bubbles pass during their upward flow. The correlations often cited in the literature for predicting the wall-to-bed heat transfer coefficient for vertically immersed surfaces are those of Vreedenberg157,158 and Wender and Cooper.159 2. Horizontal Surfaces Horizontally immersed tubes in a fluidized bed are exposed to more cross-flow of solids than are vertical tubes, and thus relatively higher heat transfer rates could be possible with these. However, a single horizontal tube poses some problems, mainly due to the hydrodynamic environment around a single horizontal tube whose top portion is piled with stagnant solid particles that reduce the local heat transfer rate at the top. On the other hand, the bottom portion of the tube becomes shrouded with rising bubbles, creating a solid free zone. This condition is opposite that prevailing at the top surface of the tube. Lateral parts of the tubes are frequently

© 1999 by CRC Press LLC

contacted by solid particles, and hence the maximum local heat transfer occurs around this lateral region. Many reports are available in the literature160–162 on the measurement of local heat transfer coefficients. In a gas fluidized bed, the minimum occurs at the top or the upward face of a horizontally immersed tube (i.e., in the downstream side of the flow). The maximum162,162 has been found to occur on the lateral side, which tends to shift toward the downward direction in a large-diameter tube. In a liquid fluidized bed, unlike a gas fluidized bed, a uniform temperature around a horizontally immersed tube was observed by Blenke and Neukirchem.163 In order to improve the heat transfer or to arrive at a uniform temperature around the tube, the use of tube bundles is recommended. Heat transfer from a horizontal tube is lowered by the presence of unheated tubes placed on one side or both sides of a heated tube. For a horizontally immersed loop type of heater, heat transfer is poor if it is a single tube.164 This has been attributed to the stable void of gas between the tubes of a loop-type heater, which would disappear in the case of tube bundles.165 Generally, it has been observed164,165 that with a well-immersed tube bundle, the heat transfer obtained is independent of the location of the bundle inside the fluidized bed. There are numerous correlations157,167–172 available in the literature for predicting the total heat transfer coefficient for an immersed horizontal tube. The correlation proposed by Vreedenberg157,158 is widely accepted. Subsequently, the need to incorporate the volumetric heat capacity of the solid and an appropriate correction for extending the application of the correlation to tube bundles were suggested by Saxena and Grewal.172 F. Effects of Operating Variables 1. Effect of Velocity a. Heat Transfer Coefficient versus Velocity The effect of velocity on heat transfer was discussed in Section IV. It was also mentioned that there exists an optimum velocity at which the heat transfer coefficient is maximum (hmax). A typical plot depicting the heat transfer coefficient as a function of the fluidizing velocity is shown in Figure 1.23 at a bed temperature of 600°C. A list of the variables that influence the heat transfer in a gas fluidized bed is given in Table 1.11. Among the various variables, bed temperature and pressure are important in studies on advanced fluidized bed reactors for coal combustion or fluidized bed boilers. These effects have not been examined in depth for gas–solid reactions in mineral or materials processing. b. Flow Regime Effect Molerus185 recently gave a detailed account of the effects of gas velocity on the heat transfer coefficient for three different flow regimes: laminar, intermediate, and turbulent. A typical plot presented by Molerus is shown in Figure 1.24 for the

© 1999 by CRC Press LLC

Figure 1.23 Heat transfer coefficient as a function of the fluidizing velocity. (dp = 0.55 mm, Ho = 57–60 cm. For U < Umf , Tp = 43–52°C, Tb = 600°C; for U > Umf , Tp = 98–134°C.) (From Draijer, W., in Fluidized Bed Combustion, Radovanovic, M., Ed., Hemisphere Publishing, Washington, D.C., 1986, 211. With permission.)

powders classified by Geldart.12 Curve 1 in Figure 1.24 corresponds to the laminar flow regime (i.e., laminar flow Ar number, Al = [(dp3 g)0.5(ρs – ρg)/µg = 159]); this curve also corresponds to a typical Geldart Group A powder for which bubbles flow at low velocity. Hence, hmax is attained only at a high velocity, about three times greater than that obtainable with Group B powders (especially in the intermediate flow regime). Curve 2 is for the intermediate flow regime (e.g., 103 ð Ar ð 105) and is typical of Group B powders for which the advantage of unlimited bubble causes h to rise steeply just after Umf and brings about the attainment of hmax at 2Umf. In the intermediate flow regime, both particle- and gas-convective components are important in determining the overall heat transfer coefficient. Curve 3 in Figure 1.24 corresponds to the turbulent flow regime (Ar Š 105), where gas-convective heat transfer is predominant. Hence, the transition from a fixed bed to a fluidized bed does not have a significant effect on the h values in this case. Borodulya et al.186 reported that the maximum heat transfer coefficient increases exponentially at elevated temperatures and varies linearly at high pressures. 2. Optimum Velocity The occurrence of a maximum in the heat coefficient could be anticipated because of the opposing effects of the particle velocity and the bed voidage, both of which increase with the fluidizing velocity. The attainment of a maximum heat transfer coefficient and a corresponding optimum velocity has been confirmed by numerous research workers, and their results show that the maximum heat transfer depends mainly on fluid–solid properties. A list of the correlations that can be applied to gas–solid fluidized beds to predict the maximum heat transfer coefficient and the corresponding optimum fluidizing gas velocity is given in Table 1.12. Although there are many correlations for Numax or hmax, they are not, in principle, functions of the optimum velocity because Uopt is also a function of gas–solid properties. Figure 1.25 shows the plot of the variation in Reopt with the Galileo number for a wide range of values of the latter (Archimedes number and Galileo number are the

© 1999 by CRC Press LLC

Table 1.11 Influence of Various Parameters on Heat Transfer in Gas Fluidized Bed Sl. no.

Variable

1

Fluid velocity (U)

2

Particle diameter

3

Thermal conductivity of solid (Kp) Specific heat of solid (Cp) Specific heat of fluid (Cf)

4 5

6

Thermal conductivity of fluid (Kf)

7

Fluid bed height (H)

8

Fluidized grid zone

9

Fluidized bed diameter

10

Length of heat transfer surface (L) Heat transfer tube diameter (dt) Vertical versus horizontal heat transfer tubes

11 12

13 a

b

Tube bundles Vertical

Horizontal

14

Bed temperature

15

Bed pressure

© 1999 by CRC Press LLC

Influences

Ref.

Heat transfer increases above Umf up to an optimum velocity (Uopt) and then decreases Heat transfer coefficient (h) increases with fine-sized particle and decreases with coarse size; for large size of particle, h increases mainly due to increase in convective component in heat transfer at high fluid velocity No influence on h

173

h is proportional to Cpn where 0.25 < n < 0.8

176, 177

Data are contradictory at moderate pressure and velocity; at high pressures, h is increased by Cf or the volumetric specific heat Cf ρf h α Kfn , n = 0.5–0.66; as bed temperature increases, h increases, attributed mainly to increase in Kf For a well-developed fluidized bed and a wellimmersed heat transfer surface, h is not dependent on H Grid zone affects h depending on grid type and its free area; for low free area, h is relatively higher than high free area grid No qualitative or quantitative information available h is independent of L

178, 179

h increases with decrease in dt

179, 183

h for vertical tubes is 5–15% higher than for horizontal tubes; construction and technological condition decide the tube orientation

152, 180

Ratio of tube spacing to tube diameter affects h; when the ratio is reduced, hmax drops; for closely packed bundles, h reduces even by 35–50% Almost similar effect as vertical bundle; for same ratio of tube spacing to diameter, horizontal bundles occupy greater portion of bed cross-section Gas-convective component increases for small (dp < 0.5 mm) particle and decreases for coarse (dp > 0.5) particle; particle-convective component decreases; for 1-mm particle, net effect is a reduction Particle-convective component not affected; gas-convective component is enhanced and is proportional to square root of gas density

152

173

174, 175

180

159, 176, 181 182

152

152

184

141

Figure 1.24 Comparison of measurements and predictions for heat transfer versus superficial gas velocity (fluidizing agent helium). Curve 1: A powder, curve 2: B-powder laminar flow region, curve 3: intermediate Archimedes number, curve 4: turbulent flow region, curve 5: B-powder laminar flow region. (Al is laminar flow Archimedes number. Al = (dp3g)1/2(ρs – ρg)/µg.) (From Molerus, O., Powder Technol., 15, 70, 1992. With permission.)

same). The correlations for predicting Reopt are also given in Table 1.12. The best fit of experimental data was made by Sathiyamoorthy et al.182 for predicting the optimum Reynolds number (Reopt) with respect to the maximum heat transfer coefficient; the correlations proposed are Reopt = 0.00812 Ar0.868 Reopt = 0.13 Ar0.52

for 1 ð Ar < 1000

for 3000 ð Ar < 107

(1.163) (1.164)

The experimental results of Sarkits173 show a strong inverse dependence for hmax on the diameter of fine particles (i.e., for dp < 2.5 mm) and a weak direct dependence on the diameter of coarse particles (2.5 ð dp ð 4.5 mm). 3. Distributor Effects The gas–solid mixing in a fluidized bed is influenced by the distributor type and its design. Many research workers do not mention the types of distributors used in their studies. However, Baerg et al.188 used perforated plate-type and Sarkits173 used screen-type distributors. Saxena and Grewal201 found that the free open area of a distributor affects the maximum heat transfer coefficient and also the corresponding optimum gas velocity. Sathiyamoorthy et al.202 confirmed that the free area in a

© 1999 by CRC Press LLC

Table 1.12 Correlations for Predicting Maximum Heat Transfer Coefficient and Optimum Reynolds Number in Gas Fluidized Bed Correlations hmax = 33.7 ρ Kf dp Numax = 0.86 Ar0.2 0.2 s

0.6

–0.36

hmax = 239.5 [log(7.05 × 10–3 ρSB)]/dp Numax = 0.64 Ar0.22 Z/dt Numax = 0.0087 Ar0.42 Pr0.33(Cps/Cpf)0.45 Numax = 0.019 Ar0.5 Pr0.33(Cps/Cpf)0.1 Numax = 0.021 Ar0.4 Pr0.33(D/dp)0.13(Ho/dp)0.16 hw,max = ho(1 – ) [1 – exp(–p kf)]

Remarks — Reopt = 0.118 Ar0.5 ρSB = nonfluidized bed density Z = distance between axes of tubes, dt = heater diameter Reopt = 0.12 Ar0.5 (laminar region) Reopt = 0.66 Ar0.5 Reopt = 0.55 Ar0.5

0.8 Pr0.4dp–0.69 Numax = 0.0017 Reopt 0.423 Ar0.14 Pr1/8 Numax = Reopt Numax = 0.3 Ar0.2 Pr0.4

ho = constant, p = constant Reopt = 0.20 Ar0.52 Reopt = 0.09 Ar0.58 1 < Ar ð 220

Numax = 0.0843 Ar0.15

103 ð Ar ð 106

Numax = 1.304 Ar0.2 Pr0.3 Numax = 0.064 Ar0.4 Numax = 2.0 (1 – ) (Ar d12.7/dt)0.21 (Cps/Cpg)45.5 Ar–0.7 Numax = 0.85 Ar0.19 + 0.006 Ar0.5 Pr0.33 + dp/kg (7.3 σ EpEsTs3) Numax = α Arβ Pr1/3 (Cvs/Cvg)1/8

Reopt = 0.113 Ar0.53 5 × 104 ð Ar ð 5 × 108 75 < Ar < 20,000

Numax = 0.14 Ar0.3(dp/dt)0.2(ρp/ρo)–0.07 0.25 Pr0.33 (C /V )0.09 Numax = 0.99 Reopt vs vg

Numax = const. Fr0.42(α · β · Remf)0.5(Reopt/Remf)Pr0.3 1. Overall a. Numax = 0.13 Al0.6 α–1 Al ð 300 b. Numax = 0.54 Al0.34 α–1 Al Š 300

2. Particle convective a. Nupc = 0.69 (Al ∆ρ/ρg)0.1 α–1 b. Nupc = 9 α–1 3. Gas convective a. Nugc = 0.4527 Ar0.2323 Pr0.33 b. Nugc = 0.024 Ar0.4304 Pr0.33

© 1999 by CRC Press LLC

100 < dp < 5000 µm for horizontal tubes and plates For 20 < Ar < 20,000, α = 0.074, β = 020; for 2 × 104 < Ar < 107, α = 0.013, β = 0.37 Heat transfer between particle (p) and freely circulating object (o), φ = shape factor Reopt = 0.00812 Ar0.868 1 ð Ar ð 3000 α = (1 – mf)/mf , β = Cvs/Cvg, const. = 0.0755 when Fr = Umf2 /(g · dt ), const. = 0.0144 when Fr = Umf2 /(g · dt ) Al = laminar flow, Ar = Ðdp3g ∆ρ/µ, α = [1 + a + kg/(2Cpµ)], a = constant to take care of additional resistance due to thick and porous oxide layer over metal powder; 0 for ceramic, 0.5 for Cu

Ref. Zabrodsky123 Varygin and Martyushin187 Baerg et al.188 Gelprin et al.189

Sarkits173 Sarkits173 Traber et al.178 Jakob and Osberg190 Chechetkin191 Ruckenstein192 Richardson and Shakiri193 Botterill and Denloye146 Kim et al.194 Brodulya et al.186 Grewal and Saxena195 Baskakov and Panov196 Chen and Pei197

Palchenok and Tamarin198 Sathiyamoorthy et al.182 Sathiyamoorthy and Raja Rao199

Molerus185, 200

103 ð Ar ð 105 Ar Š 105

Molerus185, 200

103 ð Ar ð 105 Ar Š 105

Molerus185, 200

Figure 1.25 Variation of optimum Reynolds number (Reopt) with Galileo number (Ga) (Ga is same as the Archimedes number). (From Sathiyamoorthy, D., Sridhar Rao, C.H., and Raja Rao, M., Chem. Eng. J., 37, 149, 1988. With permission.)

multiorifice distributor affects the value of hmax, which increases as the distributor freeflow area is reduced. Many correlations for predicting Numax or hmax can be found in the literature, and a list of these is given in Table 1.12. A model based on first principles and incorporating all the possible parameters for the prediction of hmax has yet to evolve. A model for predicting the maximum heat transfer coefficient which incorporates the distributor parameters and defines the stable state of fluidization was developed by Sathiyamoorthy and Raja Rao,199 and the relevant correlations are given in Table 1.12. Although many correlations are available for predicting hmax, the choice of the most appropriate correlation depends on the specific requirements and the designer’s skill. G. Heat Transfer in Liquid Fluidized Beds 1. Differences with Gas–Solid Systems Few studies have been carried out on heat transfer in liquid fluidized beds, unlike gas fluidized beds. In contrast to the latter, heat transfer in liquid fluidized beds increases with increasing particle size. Liquid fluidization is generally particulate in nature. The mixing process and the heat transfer characteristics of a liquid fluidized

© 1999 by CRC Press LLC

bed are not exactly similar to those associated with a gas fluidized bed. For a given temperature difference, the quantity of heat transferred through a liquid (film) to a solid particle could be thousands-fold higher than what could be transferred through a gas film.203 2. Heat Transfer In a liquid fluidized bed, the temperature gradient extends more into the bed than it does in a gas fluidized bed. According to Wasmund and Smith,204 the ratio of bed resistance to total resistance increases with increasing solid concentration and decreases with increasing particle size for a constant bed porosity. They suggest that the particle-convective component and the effect of the thermal conductivity of the solid particles are negligible. Generally, it has been established that heat transfer in a liquid fluidized bed is influenced by bed voidage. At constant voidage, an increase in particle size increases heat transfer. On the basis of experimental data, Romani and Richardson205 proposed a correlation that has limited application. The mechanism of heat transfer in a liquid fluidized bed was analyzed by Krishnamurthy and Sathiyamoorthy206 with the aim of investigating the reason why the heat transfer rate should increase with increasing particle size in liquid fluidized beds. Krishnamurthy and Sathiyamoorthy206 tested a model and arrived at the conclusion that for fine-sized particles in a liquid fluidized bed, the film thickness on the heat transfer surface increases. In a typical case, it was found that the film thickness varies from 1.6 to 36% for just a twofold increase in particle size but when liquid velocity is less than 10 cm/s. Particle size seems to have a relatively greater influence compared to liquid velocity in altering the magnitude of the heat transfer rate. Although the applicability of the few available correlations for practical uses is limited, the correlation of Mishra and Farid207 may be used for purposes of prediction: Nu = 0.24 (1 – ) –2/3Re0.8 Pr0.33

(1.165)

Heat transfer in liquid fluidized beds has been shown to have enhanced rates208 (as much as fourfold) on pulsating the fluidization. H. Heat Transfer in Three-Phase Fluidized Bed 1. Heat Transfer Coefficient Heat transfer data on three-phase fluidized beds have not been published extensively and only limited information is available. Studies on the heat transfer between a fluidized bed and a heat transfer surface were reported by Østergaard,209 Viswanathan et al.,210 Armstrong et al.,211 Baker et al.,212 and Kato et al.213 In general, all these studies reveal that the extent of heat transfer increases with the gas flow rate and is higher than in corresponding gas–liquid (bubble column) or liquid–solid (two-phase fluidization) systems. Heat transfer coefficients obtained in an air–water–glass-bed system have been found to increase with the gas flow rate. Figure 1.26 shows the results on the

© 1999 by CRC Press LLC

dependence of the heat transfer coefficient on the gas flow rate, and it can be seen that the trend is similar to that associated with a bubble column or a bubbling fluidized bed. Heat transfer coefficients are very high in three-phase fluidized beds (or the order of 4000 W/m2 K) and are reported to be two to three times greater than those attainable even with efficient processes such as boiling and condensation.

Figure 1.26 Heat transfer coefficient in three-phase fluidized beds. Air–water–glass beads. Data of Armstrong et al.211 (From Armstrong, E.A., Baker, C.G.J., and Bergognou, M.A., in Fluidization Technology, Keairns, D.L., Ed., Hemisphere Publishing, New York, 1976, 453. With permission.)

2. Particle Size Effect Heat transfer increases for particle diameters greater than 1 mm (Armstrong et al.211). With smaller particles in a bed of 0.12 m ID, Kato et al.213,214 observed a local maximum followed by a minimum during the transition from a fixed bed to a fully fluidized bed. This type of behavior is attributed to the heterogeneity of fluidization with fine particles when working at low liquid velocities.215 No such behavior was reported by Kato et al.214 for beds of coarse particles. 3. Correlation The wall heat transfer coefficient can be estimated by using the correlation of Kato et al.,214 which is given as:

© 1999 by CRC Press LLC

⎡ d pU1ρ1 C pl µ l 1 ⎤ h ⋅ 1 = 0.44 ⎢ ⋅ ⎥ k1 1 – 1 ⎦ Kl 1 – 1 ⎣ µ1 dp

0.78

⎡ Ug ⎤ + 2⎢ ⎥ ⎢⎣ gd p ⎥⎦

0.17

(1.166)

This correlation is recommended for a wide range of liquid–phase Prandtl numbers. The literature on three-phase fluidized beds with bed internals such as cooling or heating coils/pipes and studies on high velocities is scant.

VIII. MASS TRANSFER A. Introduction Fluidized beds in general are considered to be in transition between the states of packed bed and pneumatic transport. As a result, the correlations for the transport processes in all three units (viz., packed bed, fluidized bed, pneumatic transport reactor) are almost similar in type. Mass transfer data or the driving forces in a fluidized process must be evaluated quantitatively to arrive at a sound design of the reactor. Processes that are carried out in a fluidized bed can be either surface area based (like coating and agglomeration) or volume based (like crystallization and freezing). The transport processes of the surface-area-dependent type are complex in the sense that data on the probability of particle-to-particle or particle-to-object interaction are difficult to obtain. However, in volume-based processes, adequate information on the transport steps is available. In a fluidized bed, the particles act as turbulence promoters to enhance mass transport and reduce the hydrodynamic boundary layer. In a heat transfer process, however, the particles perform an extra function by carrying the heat. The conventional approach of drawing an analogy between heat transfer and mass transfer is valid for a fluidized bed also due to the fact that the particle-to-particle or particle-to-object contact time is low and is not much different for the heat and mass transport steps. However, there is some complexity in fluidized beds in which segregation (i.e., aggregative or bubbling fluidization) occurs. B. Mass Transfer Steps The mass transport steps in a fluidized bed can be broadly regarded to be (1) transfer between fluidized bed and object or wall, (2) transfer between particle and fluid, and (3) transfer between segregated phases (e.g., lean to dense phase). 1. Mass Transfer Between Fluidized Bed and Object or Wall a. Correlations The jM factor in mass transfer studies is defined as: jM = St Sc2/3 = (Kgw/u) Sc2/3 = constant Re–p

© 1999 by CRC Press LLC

(1.167)

where Kgw is mass transfer between a fluidized bed and an object or wall. The above equation can be used for a particulate type of fluidized bed by considering the bed as having several channels formed by the irregular contour of the wall of the particles. In the specific case of mass transfer between a fluidized bed and the wall or an object, the transfer can be viewed to occur between the two walls of a channel, one made up of the particle layer itself and the other formed by the object. The hydraulic diameter of the channel (dH) is

dH =

d p 6 = S (1 – )

(1.168)

The velocity of the fluid through the channel is assumed to be proportional to the interstitial velocity (u) (i.e., U/). If the Reynolds number in Correlation 1.167 is expressed in terms of U and dH, then,

K gw U

Sc

23

⎡ U dp ρ ⎤ = constant ⎢ ⎥ ⎣ µ(1 – ) ⎦

–p

(1.169)

The constant and the exponent p in Equation 1.169 vary depending on the state and the type of the bed. Typical values are tabulated in Table 1.13. Table 1.13 Magnitude of Constant, Bed Voidage ( ), and the Exponent (m) of the Correlation (Kgw/U) Sc2/3 = Const [Udp /µ (1 – )]–p Ref.

Constant in Equation 1.169

Packed

223

0.35–0.45

0.7

0.4

Liquid fluidized bed Gas fluidized bed

224 225

0.5–0.9 0.5–0.95

0.43 0.7

0.38 0

Bed Type

m

Re

Sc

50–500

0.9–218 450–3,700 200–24,000 1,300 300–12,000 2.57

b. Influencing Parameters Most studies on mass transfer do not provide useful data on voidage (), and hence comparison of results is often difficult. The degree of particle turbulence in a gas or liquid fluidized bed is not the same even if an average dynamic similarity is achieved. In addition to particle mean velocity and porosity, data on turbulence are necessary for an accurate prediction of mass as well as heat transfer coefficients. Measuring techniques play a key role. For example, mass transfer data can be obtained by simple experimental techniques more easily in a liquid than in a gas fluidized bed. To obtain reproducible results, gas fluidized beds operated at incipient fluidization are preferred.

© 1999 by CRC Press LLC

c. Role of Voidage It is useful for the data on voidage to correspond to an optimum situation where mass and/or heat transfer rates are high. The effect of the operating velocity on attainment of the maximum heat transfer coefficient was explained in the preceding section, and various correlations for predicting Uopt were presented. By analogy, one can choose the value of Uopt for heat transfer for mass transfer also. However, data on opt are also required, as can be seen from Equation 1.169. The voidage at Uopt does not have a unique value. In fact, the occurrence of the maximum does not correspond to any particular velocity, as pointed out in the section on heat transfer. In other words, the maximum can occur over a range of U. The voidage226 at the optimum condition lies in the range 0.6–0.75. In many industrial practices, attainment of maximum conversion and minimization of effluent heat content and convective heat loss are more important than Uopt values. The various useful ranges of values for opt, Uopt/Umf, and Uopt/Ut and the range of values for the drag coefficient (CD) for a single spherical particle were discussed by Beek.227 These data are presented in Table 1.14. It can be seen from the table that the useful operating velocity for attaining the maximum mass (heat) transfer rate falls in the range of three to five times the minimum fluidization velocity, and these values are used in many designs as a rule of thumb. Table 1.14 Optimum Parameter for Mass Transfer for Various Drag Coefficients ( CD) of a Single Spherical Particle opt (0.65 ± 0.08) CD (0.8 ± 0.04)

Uopt/Umf 1/6

(5 + 0.5) CD 36

Uopt /Ut 3/4

(0.3 + 0.1) CD 0.6

CD 1/4

10

2. Mass Transfer Between Particle and Fluid a. Comparison of Mass Transfer from Single Particle and Fixed Bed to Fluid A comparison of correlations for mass transfer from a solid particle to a flowing fluid (Kpl) can be made from Table 1.15 for three cases: (1) a single sphere moving through a fluid, (2) a fixed bed, and (3) a liquid fluidized bed. For large particles (Reynolds number, Rep > 80), it can be generalized from the data given in Table 1.15 that: Kp, fixed bed > Kp, fluidized bed > Kp, single sphere

(1.170)

The constant 2 in all the correlations (Table 1.15) is due to molecular diffusion and represents the theoretical minimum when the fluid is stagnant (i.e., Rep = 0). For a fixed bed of fine solids, measurement of mass transfer parameters is difficult due to rapid attainment of equilibrium brought about by the large surface area of the particles. For industrial-scale liquid fluidized beds, particulate fluidization need not necessarily prevail.

© 1999 by CRC Press LLC

Table 1.15 Correlation Comparison for Particle-to-Fluid Mass Transfer Coefficient (Kp) Condition Single sphere to fluid

Fixed bed

Liquid fluidized bed

Correlations for Sherwood no. (Sh) 2 + 0.65Sc1/3 Re1/2 Re = dp Uo ρf /µ Sc = µ/ρf 2 + 1.85 C1/3 Rep1/2 2 + 1.5 Sc1/3 [(1 – )Rep]1/2

Remarks Particle moving through the fluid at relative velocity (Uo) For coarse solids (i.e., Rep > 80), Uo is superficial velocity of fluid For laboratory-size bed exhibiting particulate fluidization, 5 < Rep < 120 ð 0.84

Ref. 228

229

230

b. Complexity in Measurement In a gas fluidized bed, measurement of the mass transfer coefficient poses practical problems due to features like high surface area, bubble formation, backmixing, and gas bypassing. Attainment of equilibrium for a gas inside the bed is so rapid that the gas has to pass through a bed of a height equal to only a few particle diameters to reach equilibrium. The height required to reach equilibrium is often referred to as the height of a transfer unit. In order to measure the mass transfer coefficient in a gas fluidized bed, a very shallow bed is necessary because there is no bubble generation and gas backmixing or bypassing in shallow beds. It is not possible to attain truly ideal conditions in a fluidized bed. The mass transfer data of Rebnick and White231 and Kettenring et al.232 were analyzed by Kunii and Levenspiel;233 they showed that a sharp decrease in the Sherwood number occurs when the Reynolds number is lowered. The decrease in the Sherwood number is lower than that obtained under corresponding conditions in a fixed bed or a single sphere. It was found that the Sherwood number in a fluidized bed can be less than 2, a value which is the minimum for a single particle as well as a fixed bed (see Table 1.15). This could result from the hydrodynamic conditions which may not be conducive for mass transfer from particle to fluid. c. Correlations Richardson and Szekely234 proposed correlations for the Sherwood number in the case of a shallow fluidized bed of a height equivalent to five times the particle diameter. The correlations are Sh = 0.374 Re1.18

(1.171)

Sh = 2.01 Rep0.5

(1.172)

for 0.1 < Rep < 15 and

© 1999 by CRC Press LLC

for 15 < Rep < 250. It may be observed from these correlations (1.171 and 1.172) that the Sherwood number attains a value far below 2 for Rep values smaller than 15. According to Richardson and Szekely,234 the low values of Sh ( U fall back to the bed, and the other is the height TDH(F) above which entrainment remains constant. Although both types of TDH apparently seem to have the same value, this is not so in a true sense. Hence, the method of evaluating TDH should always be mentioned if a correlation is proposed for it. For Group D particles, Soroko et al.330 proposed the following correlation:

© 1999 by CRC Press LLC

TDH(C) = 1200 Ho Rep1.55Ar–1.1

(1.219)

for 15 < Rep < 300, 19.5 < Ar < 6.5 × 105, Ho < 0.5 m. For Group A particles, the correlation proposed by Horio et al.331 is TDH(F) = 4.47 Des1/2

(1.220)

where Des is the equivalent bubble diameter at the surface. Zenz and Weil296 provided a graph to compute TDH at various vessel diameters for different operating velocities; their predictions show that TDH(F) increases with increasing bed vessel diameter, and no correlation appears to account for this effect. 5. Some Useful Remarks We have seen that several mechanisms and methods have been made use of in connection with the evaluation of entrainment. The ultimate goal is to estimate the total entrainment outside the reactor so that post-reactor equipment can be properly designed for efficient gas–solid separation. One of the important applications of data on entrainment is prediction of the conversion in the freeboard region. The solids dispersed in the gas phase in the freeboard can exhibit significant conversion if the bed contains a major fraction of fines. When the reaction is highly exothermic, it is necessary to provide adequate cooling or heat control in the freeboard, and knowledge of solids entrainment is required for this purpose. From the preceding discussion on entrainment, a design engineer can appreciate that estimating the entrainment rate is essential and cannot be neglected in the final design of an efficient reactor. However, it is clear that much work remains to be carried out in this area. The subject of entrainment has been viewed mainly in the context of gas–solid systems. For liquid–solid systems, the mechanisms described earlier cannot automatically be assumed to be valid. In many mineral operation systems, liquid and gas fluidization are used for physical separation. The field of entrainment or elutriation has considerable relevance for such applications. In the context of three-phase fluidization, the subject of elutriation is in the process of evolving and is gaining importance in light of the biotechnological applications of this type of fluidization.

NOMENCLATURE A a aI A1, B1 Ap Ar B c1 , c2

area of cross-section (m2) constant in Equation 1.214 interfacial surface area (m2/m3) constants in Equation 1.68 projected surface area of the particle (m2) Archimedes number, dρ3s ρf (ρs – ρf)g/µ2g (–) discharge rate (kg/m2) constants in Equation 1.34

© 1999 by CRC Press LLC

C C1 C2 C3 C7 Cbo Cbt CD Cg Cgi Cl Cp Cpg Cpl Cps CS,E D d d* db dbo De Des DG dH DL dp (dp)AM (dp)GM (dp)HM (dp)LM (dp)SM (dp)VS (dp)WM dpi dpm dsv Dt dv E e Eb Ebs El EM Eo

cohesion constant (N/m2) inlet concentration (mol/l) outlet concentration (mol/l) constant in Equation 1.37 equilibrium concentration (mol/l) reactant concentration at the distance z = 0 (mol/m3) concentration in bubble phase at elevation t (mol/m3) drag coefficient (–) reactant concentration in the liquid (mol/l) concentration of the gas at the interphase (mol/l) reactant concentration in the liquid (mol/l) gas concentration in the emulsion phase (mol/m3) specific heat of gas at constant pressure (J/kg · K) specific heat of liquid (J/kg · K) specific heat of solid particle (J/kg · K) solid concentration at the exit of the column (kg) molecular diffusion coefficient (m2/s) diameter of a spherical particle (m) dimensionless particle diameter in Equation 1.73 bubble diameter (m) initial bubble diameter (m) equivalent diameter (m) equivalent bubble diameter at the surface (m) diffusion coefficient (m2/s) hydraulic diameter of the channel (m) diffusion coefficient in liquid (m2/s) particle diameter (m) arithmetic mean particle diameter (m) geometric mean particle diameter (m) harmonic mean particle diameter (m) length mean particle diameter (m) surface mean particle diameter (m) volume surface diameter (m) weight mean particle diameter (m) diameter of particles in ith section (m) mean diameter (m) surface volume diameter (m) diameter of tube or column or vessel (m) volume diameter (m) elutriation rate (kg/m2 · s) energy per unit mass (J/kg) emissivity factor of radiating/receiving bodies (–) generalized emissivity factor (–) liquid-phase dispersion coefficient (m2/s) elasticity modulus of powder bed (N/m2) entrainment just above the bed surface (kg/m2 · s)

© 1999 by CRC Press LLC

Ep Es Es F f FB fb FD Fe ff Ff Fg fk fL Frmf Frt fw g gc H h h hB hc hf hgc hgp Hgsl hi hmax hmax hmf Hmf Ho hp hpc hr hw i J* jfo jH jM K K1

emissivity of the particle (–) emissivity of the surface (–) emissivity factor due to shape (–) feed rate (kg/m2) bed friction factor (–) force due to buoyancy (N) fraction of surface shrouded by gas bubbles (–) drag force (N) adhesion force transmitted in particle contact (N) friction for fluidized bed (–) parameter dependent on fines fraction, Equation 1.117 (–) force due to gravity (N) friction factor for packed bed (–) fractional bubble-phase contact time (–) (= Umf2 /dpg), Froude number at minimum fluidization (–) Froude number based on Ut and Dt, Ut/Ð(gDt) (–) fraction of wake volume (–) gravitational constant (m2/s) conversion factor, gc = 1 kg · m/N · s2 = 32.2 lb · ft/lbf · s2 = 9.8 kg · m/kg · wt · s2 (–) height up to the bed surface (m) overall heat transfer coefficient (W/m2 · K) bed height below H (m) heat transfer coefficient in the bubble-covered surface (W/m2 · K) convective heat transfer coefficient (W/m2 · K) film heat transfer coefficient (W/m2 · K) gas-convective component of heat transfer (W/m2 · K) gas-to-particle heat transfer coefficient (W/m2 · K) height occupied by gas–solid–liquid phases (m) instantaneous heat transfer coefficient (W/m2 · K) maximum heat transfer coefficient (W/m2 · K) maximum heat transfer coefficient corresponding to Uopt (W/m2 · K) heat transfer coefficient at Umf (W/m2 · K) bed height at Umf (m) static bed height (m) particle-convective component of heat transfer coefficient (W/m2 · K) convective heat transfer coefficient due to particle (–) radiative component of heat transfer coefficient (W/m2 · K) bed-to-wall heat transfer coefficient (W/m2 · K) ith component (–) dimensionless fluid flux (–) fluid flux (superficial velocity) relative to particles (m/s) Colburn’s factor (= Nu/Re Pr1/3) Colburn j factor for mass transfer (–) constant in Equation 1.42 (–) constant in Equation 1.27 (–)

© 1999 by CRC Press LLC

K2 K3 k k5, k6, k7 Kc kc Kd KE kef keo kew KG kg Kg Kgw KI Kl kl Kla Kmb KN kp kpe Kpl ks ks KST Kw L M m me N Nb Ncoh Nf ni Nu Nugc Numax Nup p

constant in Equation 1.28 (–) constant in Equation 1.29 (–) thermal conductivity (W/m · K) constants in Equation 1.56 (–) overall mass transfer coefficient (m/s) effective thermal conductivity (W/m · K) mass transfer coefficient (m/s) elutriation rate constant (–) conductivity due to turbulent diffusion (–) effective thermal conductivity of the bed with no fluid (W/m · K) effective thermal conductivity near wall zone (W/m · K) average mass transfer coefficient (m/s) thermal conductivity of gas (W/m · K) gas-side mass transfer coefficient in a gas–liquid mass (–) mass transfer coefficient between fluidized bed (m/s) constant as defined by Equation 1.97 (–) liquid-side mass transfer coefficient in a gas–fluid mass transfer (m/s) thermal conductivity of liquid (W/m · K) volumetric mass transfer coefficient (s–1) constant in Equation 1.116 (–) Newton’s constant, Equation 1.96 (–) mass transfer from particle to particulate bed or spherical particle, Equation 1.170 (s) permeability (m2) mass transfer between particle and fluid (s) thermal conductivity of solid (W/m · K) solid–solid mass transfer coefficient (–) Stokes’ constant, Equation 1.93 (–) ratio of wake voidage to gas-phase voidage (–) length of vertical surface (m) mass of solid (kg) Henry constant (atm/mol fraction) (m/s) constant in Equation 1.148 (–) fluidization number (= ρ3s dp4 g2/µ2E), Equation 1.39 (–) number of bubbles (–) cohesion number (C/ρsgdp) (–) flux of the material transferred from the gas to the liquid phase (mol/s) number of particles of diameter dpi (m) Nusselt number based on h (–) Nusselt number for gas convection, Equation 1.153 (–) Nusselt number at hmax (–) Nusselt number based on particle diameter, dp, hdp/k (–) constant in Equations 1.167 and 1.169 (–)

© 1999 by CRC Press LLC

Pr Q q qr (Q – Qmf)c r* Re Rec Rel ReL Remd Remf Rems Reopt Rep Res Ret Retr Rr Rw S Sb Sc Sh Si So Sp Sv Sw St t Tb Te Tg Tp Ts Tw tz U u U* UA UAV Uc Uf

Prandtl number (Cp µ/kg) (–) volumetric exchange rate of gas between a single bubble and dense phase (m3/s) volumetric exchange rate due to convective flow of gas (m3/s) radiant heat flux (J/m2) critical excess flow of gas (m3/s) dimensionless particle size (–) particle Reynolds number, dpUfρf /µf (–) Reynolds number based on Uc (–) Reynolds number based on liquid velocity (Ul) (–) Reynolds number based on linear dimension (–) Reynold number based on Umd (–) Reynolds number at minimum fluidization velocity (Umf) (–) Reynolds number at the onset of slugging (–) Reynolds number based on Uopt (–) Reynolds number based on particle diameter (dp) (–) Reynolds number defined as 2j*r* (–) Reynolds number based on Ut (–) Reynolds number based on Utr (–) mean thermal resistance of the packet (m2 · K/W) contact resistance near the wall zone (m2 · K/W) surface area per unit volume (m2/m3) frontal surface area of a spherical bubble (m2) Schmidt number, µρg/D (–) Sherwood number, kpldp/D (–) surface area per unit weight of ith component (m2/kg) surface area of a spherical particle (m2) surface area of particle of diameter dp (m2) surface area of a nonspherical particle of constant volume (m2) surface area per unit weight of the particles (m2/kg) Stanton number, kgw /u (–) time (s) fluidized bed temperature (K) intermediate temperature (K) temperature of gas (K) temperature of particle (K) temperature of the surface (K) wall temperature (K) particle residence time (s) superficial gas velocity (m/s) fluid velocity (m/s) dimensionless particle velocity (Equation 1.72) (–) absolute bubble velocity (m/s) mean average bubble rise velocity (m/s) transition velocity from bubbling to turbulent (m/s) fluid velocity (m/s)

© 1999 by CRC Press LLC

Ug Ugl Ul ul ls Ulmf Umb Umd Umf Uop Uopt Up Ur Usl Ut Utr V V Vb Vf Vo Vp Vs Vsv W Ws W/A Wep x x Xe Xi XiB XiE XiF Xo z @

gas velocity (m/s) gas–liquid slip velocity (m/s) liquid velocity (m/s) particle fluid velocity in a suspension (m/s) minimum fluidization velocity for liquid in a liquid–solid (twophase) system (m/s) minimum bubbling velocity (m/s) velocity of transition from dense phase to dilute phase (m/s) minimum fluidization velocity (m/s) optimum gas velocity (m/s) optimum operating velocity corresponding to hmax (m/s) particle velocity (m/s) relative velocity (m/s) particle slip velocity (m/s) single-particle terminal velocity (m/s) transport velocity (m/s) particle velocity (m/s) pore volume per unit mass (m3/kg) volume of bubble (m3) volume of the floating bubble breakers (m3) particle velocity at the bed surface (m/s) volume of particle (m3) volume of solid particle (m3) volume per unit mass of the particle (m3/kg) weight (kg) solids entrained per unit area (kg/m2) bed weight per unit area (kg/m2) Weber number (= ρsUs2dp /σgl) (–) solids weight fraction (–) ratio of the solid holdup in the wake region to that in liquid fluidized bed region (–) exit mole fraction (–) volume fraction (m) mass fraction of ith component in discharged solid (–) mass fraction of ith component in elutriated solid (–) mass fraction of ith component in feed (–) inlet mole fraction (–) distance above distributor (m) powder number (Equations 1.34 and 1.37)

Greek Symbols α αf α1

correction factor in Equation 1.43 angle of internal fraction (°) ratio of the mass velocity of the fluid in the direction of heat flow to the superficial mass velocity in the direction of fluid flow (–)

© 1999 by CRC Press LLC

β β1 γ ∆ρ ∆ρσ ∆c ∆Clm ∆Pb ∆T ∆Ti δ δw B b g gl I l mf o opt p s θ θD λ µ µeff µf µg µl ρa ρB ρb ρbmf ρe ρf ρg ρl ρp ρs ρsz ρt

constant in Equation 1.52 (–) ratio of distance between two adjacent particles to the mean particle diameter (–) angle of wall friction (°) difference in density of solid and gas (kg/m3) pressure drop due to surface tension (Pa · s) concentration driving force (mol/m3) log mean concentration difference (mol/l) bed pressure drop (Pa · s) mean temperature difference (K) mean temperature difference across the ith section (K) angle of rupture (°) film thickness on the wall (m) bed voidage (–) voidage due to bubble phase (–) bubble volume fraction (–) holdup due to gas per unit volume of three-phase fluid bed (–) holdup of gas in gas–liquid column per unit volume of two-phase fluidized bed (–) intraparticle voids due to pores, cracks, etc. (kg/m3) liquid holdup (–) bed voidage at minimum fluidization velocity (–) static bed porosity (–) voidage at optimum velocity (Uopt) (–) solid volume fraction (–) solids volume fraction (–) angle of repose (°) dense-phase root square average residence time (s) shape factor (–) viscosity (Pa · s) effective viscosity (Pa · s) viscosity of fluid (Pa · s) viscosity of gas (Pa · s) viscosity of liquid (Pa · s) apparent density (kg/m3) bulk density of the bed material (kg/m3) bed density (kg/m3) bulk density of the bed at Umf (kg/m2) density of emulsion phase (kg/m3) density of fluid (kg/m3) density of gas (kg/m3) density of liquid (kg/m3) particle density (kg/m3) density of solid particle (kg/m3) solid holdup density at an elevation of z (kg/m3) theoretical density (kg/m3)

© 1999 by CRC Press LLC

σ φ ω

surface tension (N/m) sphericity of a particle (–) angle of slide (°)

REFERENCES 1. Zenz, F.A. and Othmer, D.F., Fluidization and Fluid Particle Systems, Van NostrandReinhold, Princeton, NJ, 1960, 113. 2. Barnea, E. and Mednick, L., Correlation for minimum fluidization velocity, Trans. Inst. Chem. Eng., 53, 278, 1975. 3. Zabrodsky, S.S., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, chap. 1. 4. Geldart, D., Estimation of basic particle properties for use in fluid particle process calculations, Powder Technol., 60, 1, 1990. 5. Abrahamsen, A.R. and Geldart, D., Behavior of gas fluidized beds of fine powders. I. Homogeneous expansion, Powder Technol., 26, 35, 1980. 6. Wadell, H., Volume shape and roundness of rock particles, J. Geol., 15, 443, 1982. 7. Leva, M., Fluidization, McGraw-Hill, New York, 1959, chap. 34. 8. Brown, G.G., Katz, D., Foust, A.S., and Schneidewind, R., Unit Operation, John Wiley & Sons, New York, 1950, 77. 9. Furnas, C.C., Grading aggregates. I. Mathematical relations for beds of broken solids of maximum density, Ind. Eng. Chem., 23, 1052, 1931. 10. Ciborowski, J., Fluidyzacja, PWT, Warsaw, 1957. 11. Geldart, D., The effect of particle size and size distribution on the behaviour of gas fluidized beds, Powder Technol., 6, 201, 1972. 12. Geldart, D., Types of fluidization, Powder Technol., 7, 285, 1973. 13. Molerus, O., Interpretation of Geldart’s Type A, B, C, and D powders by taking into account interparticle cohesion forces, Powder Technol., 33, 81, 1982. 14. Mathur, K.B., Spouted bed, in Fluidization, Davidson, J.F. and Harrison, D.H., Eds., Academic Press, London, 1971, chap. 17. 15. Clark, N.N., Van Egmond, J.W., and Turton, R., A numerical representation of Geldart’s classification, Powder Technol., 56, 225, 1968. 16. Rietema, K., Powders, what are they? Powder Technol., 37, 20, 1984. 17. Saxena, S.C. and Ganzha, V.L., Heat transfer to immersed surfaces in gas fluidized beds of large particles and powder characterization, Powder Technol., 39, 199, 1984. 18. Leva, M., Fluidization, McGraw-Hill, New York, 1959, 64. 19. Miller, Co. and Logwinuk, A.K., Fluidization studies of solid particles, Ind. Eng. Chem., 41, 1135, 1949. 20. Rowe, P.N. and Yacono, C., The distribution of bubble size in gas fluidized beds, Trans. Inst. Chem. Eng., 53, 59, 1975. 21. Davidson, J.F. and Harrison, D., Fluidized Particles, Cambridge Press, London, 1963. 22. Yacono, C., Ph.D. thesis, University of London, 1975. 23. Baerg, A., Klassen, J., and Gishler, P.E., Heat transfer in a fluidized solids beds, Can. J. Res., F28, 287, 1950. 24. Kumar, A. and Sen Gupta, P., Prediction of minimum fluidization velocity for multicomponent mixtures, Indian J. Technol., 12(5), 225, 1974.

© 1999 by CRC Press LLC

25. Baeyens, J. and Geldart, D., Predictive calculations of flow parameters in gas fluidized beds and fluidization behaviour of various powders, in Proc. Int. Symp. on Fluidization and Its Application, Toulouse, France, 1973, 263. 26. Leva, M., Shirai, T., and Wen, C.Y., La prevision due debut de la fluidization daus les lits solides granulaires, Genie Chim., 75(2), 33, 1956. 27. Riba, J.P., Expansion de couches fluidisees par des liqudes, Can. J. Chem. Eng., 55, 118, 1977. 28. Ballesteros, R.L., These de Docteur Ingenieur, Universite de Toulouse, France, 1980. 29. Doichev, K. and Akhmakov, N.S., Fluidization of polydisperse systems, Chem. Eng. Sci., 34, 1357, 1979. 30. Thonglimp, V., Higuily, N., and Leguerie, C., Vitesse minimale decouches fluidize par un gas, Powder Technol., 38, 233, 1984. 31. Morse, R.D., Fluidization of granular solids–fluid mechanics and quality, Ind. Eng. Chem., 41, 117, 1949. 32. Schiller, L. and Naumann, A.N., Z. Ver. Dtsch. Ing., 77, 318, 1935. 33. Wen, C.Y. and Yu, Y.H., Mechanics of fluidization, Chem. Eng. Prog. Symp. Ser., 62(2), 100, 1966. 34. Richardson, J.F. and Zaki, W.N., Sedimentation and fluidization. I, Trans. Inst. Chem. Eng., 32, 35, 1954. 35. Haider, A. and Levenspiel, O., Drag coefficient and terminal velocity of spherical and non-spherical particles, Powder Technol., 58, 63, 1989. 36. Turton, R. and Levenspiel, O., A short note on the drag correlation for sphere, Powder Technol., 47, 83, 1986. 37. Masliyah, J.B. and Polikar, M., Terminal velocity of porous sphere, Can. J. Chem. Eng., 58, 299, 1980. 38. Rowe, P.N. and Henwood, G.A., Drag forces in a hydraulic model of a fluidized bed. I, Trans. Inst. Chem. Eng., 39, 43, 1961. 39. Frantz, J.F., Minimum fluidization velocities and pressure drop in fluidized beds, Chem. Eng. Prog. Symp. Ser., 62, 21, 1966. 40. Davies, L. and Richardson, J.W., Gas interchange between bubbles and the continuous phase in a fluidized bed, Trans. Inst. Chem. Eng., 44(8), 293, 1966. 41. Pillai, B.C. and Raja Rao, M., Pressure drop and minimum fluidization velocities in air fluidized beds, Indian J. Technol., 9, 225, 1971. 42. Ergun, S., Fluid flow through packed columns, Chem. Eng. Prog., 48, 89, 1952. 43. Carman, P.C., Fluid flow through granular beds, Trans. Inst. Chem. Eng., 15, 150, 1937. 44. Bourgeois, P. and Grenier, P., The ratio of terminal velocity to minimum fluidization velocity for spherical particles, Can. J. Chem. Eng., 46, 325, 1968. 45. Ghosal, S.K. and Mukherjee, R.N., Momentum transfer in solid–liquid fluidized beds. 1. Prediction of minimum fluidization velocity, Br. Chem. Eng., 17, 248, 1972. 46. Saxena, S.C. and Vogel, G.J., The measurement of incipient fluidization velocities in a bed of coarse dolomite at high temperature and pressure, Trans. Inst. Chem. Eng., 55, 184, 1977. 47. Babu, S.P., Shah, B., and Talwalkar, A., Fluidization correlations for coal gasification materials–minimum fluidization velocity and fluidized bed expansion, AIChE Symp. Ser., 74, 176, 1978. 48. Richardson, J.F. and Jeromino, M.A.S., Velocity voidage relations for sedimentation and fluidization, Chem. Eng. Sci., 39, 1419, 1979. 49. Thonglimp, V., These Dr. Ingenier, Institut National Polytechnique de Toulouse, France, 1981.

© 1999 by CRC Press LLC

50. Chitester, D.C., Kornosky, R.M., Fan, L.S., and Danko, J.P., Characteristics of fluidization at high pressure, Chem. Eng. Sci., 39, 253, 1984. 51. Masaaki, N., Hamada, Y., Toyama, S., Fouda, A.E., and Capes, C.E., An experimental investigation of minimum fluidization velocity at elevated temperatures and pressures, Can. J. Chem. Eng., 63, 8, 1985. 52. Agarwal, P.K. and O’Neil, B.K., Transport phenomena in multi-particle system. I. Pressure drop and friction factors unifying the hydraulic radius and submerged object approaches, Chem. Eng. Sci., 43, 2487, 1988. 53. Satyanarayana, K. and Rao, P.G., Minimum fluidization velocity at elevated temperatures, Ind. Chem. Eng., 31, 79, 1989. 54. Grace, J.R., Fluidized bed hydrodynamics, in Handbook of Multiphase Systems, Hestroni, G., Ed., Hemisphere Press, New York, 1982, chap. 8.1. 55. Panigrahi, M.R. and Murty, J.S., A generalized spherical multiparticle model for particulate systems: fixed and fluidized beds, Chem. Eng. Sci., 46(7), 1863, 1991. 56. Fletcher, J.V., Deo, M.D., and Hanson, F.V., Reexamination of minimum fluidization velocity correlations applied to group B sands and coked sands, Powder Technol., 69, 147, 1992. 57. Lippens, B.C. and Mulder, J., Prediction of the minimum fluidization velocity, Powder Technol., 75, 67, 1993. 58. Kunii, D. and Levenspiel, O., Fluidization Engineering, John Wiley & Sons, New York, 1979, chap. 3. 59. Richardson, J.F., Incipient fluidization and particulate system, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 2. 60. Clift, R. and Gauvin, W.H., The motion of particle in turbulent gas streams, Chemeca 70, Butterworths, Australia, 1970, 14. 61. Concha, F. and Almendra, E.R., Settling velocities of particulate systems. 1. Settling velocities of individual spherical particles, Int. J. Miner. Process., 5, 349, 1979. 62. Concha, F. and Almendra, E.R., Settling velocities of particulate systems. 2. Settling velocities of suspensions of spherical particles, Int. J. Miner. Process., 6, 31, 1979. 63. Concha, F. and Barrientos, A., Settling velocities of particulate systems. 3. Power series expansion for the drag coefficient of sphere and prediction of the settling velocity, Int. J. Miner. Process., 9, 167, 1982. 64. Concha, F. and Barrientos, A., Settling velocities of particulate systems. 4. Settling of nonspherical isometric particles, Int. J. Miner. Process., 18, 297, 1986. 65. Concha, F. and Christiansen, A., Settling velocities of particulate systems. 5. Settling velocities of suspensions of particles of arbitrary shape, Int. J. Miner. Process., 18, 309, 1986. 66. Geldart, D., Estimation of particle properties for use in fluid particle process calculations, Powder Technol., 60, 1, 1990. 67. Pettyjohn, E.S. and Christiansen, E.B., Effect of particle shape on free settling rates of isometric particles, Chem. Eng. Prog., 44, 157, 1948. 68. McCabe, W.L. and Smith, J.C., Unit Operations of Chemical Engineering, McGrawHill, New York, 1967, chap. 7. 69. Rubey, W.W., Settling velocities of gravel, sand and silt particles, Am. J. Sci., 225, 325, 1933. 70. Dallavalle, J.M., Micromeritics, Pitman Publishing, New York, 1943. 71. Dallavalle, J.M., Micromeritics, The Technology of Fine Particles, 2nd ed., Pitman Publishing, New York, 1948, chap. 2.

© 1999 by CRC Press LLC

72. Lapple, C.E., Fluid and Particle Mechanics, University of Delaware Publ., Newark, 1951, 284. 73. Torobin, L.B. and Gauvin, W.H., Fundamental aspects of solid–gas flow. Parts 1 and 2, Can. J. Chem. Eng., 37, 129, 1959. 74. Olson, R., Essentials of Engineering Fluid Mechanics, International Text Book, Scranton, PA, 1961. 75. Barnea, E. and Mizrahi, J., Generalized approach to the fluid dynamics of particulate systems. I. General correlation for fluidization and sedimentation in solid multiparticle systems, Chem. Eng. J., 5, 171, 1973. 76. Clift, R., Grace, J.R., and Weber, M.E., Bubble, Drops and Particles, Academic Press, New York, 1978. 77. Flemmer, R.L.C. and Banks, C.L., On the drag coefficient of a sphere, Powder Technol., 48, 217, 1986. 78. Khan, A.R. and Richardson, J.F., Pressure gradient and friction factor for sedimentation and fluidization of uniform spheres in liquids, Chem. Eng. Sci., 45, 255, 1990. 79. Wallis, G.B., A simple correlation for fluidization and sedimentation, Trans. Inst. Chem. Eng., 55, 74, 1977. 80. Turton, R. and Clark, N.N., An explicit relationship to predict spherical particle terminal velocity, Powder Technol., 53, 127, 1987. 81. Geldart, D., Gas Fluidization Technology, Wiley, Chichester, 1986, chap. 6. 82. Lewis, E.W. and Bowerman, E.W., Fluidization of solid particles in liquid, Chem. Eng. Prog., 48, 603, 1952. 83. Lewis, W.K., Gilliland, E.R., and Bauer, W.C., Characteristics of fluidized particles, Ind. Eng. Chem., 41, 1104, 1949. 84. Chen, P. and Pei, D.C.T., Fluidization characteristics of fine particles, Can. J. Chem. Eng., 62, 464, 1984. 85. Lucas, A., Arnaldos, J., Casal, J., and Pulgjaner, L., Improved equation for the calculations of minimum fluidization, Ind. Eng. Chem. Process Des. Dev., 25, 426, 1986. 86. Yang, W.C., Chitester, D.C., Kornos, R.M., and Keairn, D.L., A generalized methodology for estimating minimum fluidization velocity at elevated pressure and temperature, AIChE J., 31, 1086, 1985. 87. Wilhelm, R.H. and Kwauk, M., Fluidization of solid nonvesicular particles, Chem. Eng. Prog., 44, 201, 1948. 88. Romero, J.B. and Johanson, L.N., Factors affecting fluidized bed quality, Chem. Eng. Prog. Symp. Ser., 58(38), 28, 1962. 89. Geldart, D., The fluidized bed as a chemical reactor: a critical review of the first 25 years, Chem. Ind., p. 1474, September 2, 1967. 90. Ho, T.C., Fan, L.T., and Walawender, W.P., The onset of slugging in gas–solid fluidized beds with large particles, Powder Technol., 35, 249, 1983. 91. Lanneau, K.P., Gas solids contacting in fluidized beds, Trans. Inst. Chem. Eng., 38, 125, 1960. 92. Rhodes, M. and Geldart, D., Measurement of axial solid flux variation in the riser of a circulating fluidized bed, in Circulating Fluidized Bed Technology II, Basu, P. and Large, P., Eds., Pergamon Press, Oxford, 1988, 155. 93. Lee, G.S. and Kim, S.D., Pressure fluctuations in turbulent fluidized beds, J. Chem. Eng. Jpn., 21, 515, 1988. 94. Lee, G.S. and Kim, S.D., Bed expansion characteristics and transition velocity in turbulent fluidized beds, Powder Technol., 62, 207, 1990.

© 1999 by CRC Press LLC

95. Yang, W.C., Criteria for choking in vertical pneumatic conveying lines, Powder Technol., 35, 143, 1983. 96. Yang, W.C., A mathematical definition of choking phenomenon and a mathematical model for predicting choking velocity and choking voidage, AIChE J., 21, 1013, 1975. 97. Østergaad, K., Studies of Gas–Liquid Fluidization, Danish Technical Press, Copenhagen, 1969. 98. Østergaad, K., Three phase fluidization, in Fluidization, Davidson, J.F. and Harrison, D. Eds., Academic Press, New York, 1971, chap. 18. 99. Østergaad, K., Three phase fluidization, in Recent Advances in the Engineering Analysis of Chemically Reacting Systems, Doraiswamy, L.K., Ed., Wiley Eastern Ltd., New Delhi, 1984, chap. 22. 100. Davidson, J.F., Three phase systems, in Chemical Reactor Theory — A Review, Lapidus, L. and Amundson, N.R., Eds., Prentice-Hall, Englewood Cliffs, NJ, chap. 10. 101. Kim, S.D. and Kim, C.H., Three phase fluidized beds: a review, Hwahak Konghak, 18(5), 313, 1980. 102. Wild, G., Saberian, M., Schwartz, J.L., and Charpentier, J.C., Gas–liquid–solid fluidized bed reactors, state of the art and industrial perspectives, Entropie, 106, 3, 1982; English translation in Int. Chem. Eng., 24(4), 639, 1984. 103. Epstein, N., Three phase fluidization: some knowledge gaps, Can. J. Chem. Eng., 59, 649, 1981. 104. Epstein, N., Hydrodynamics of three phase fluidization, in Handbook of Fluids in Motion, Cheremisinoff, N.P. and Gupta, R., Eds., Ann Arbor Science, Ann Arbor, MI, 1983, 1165. 105. Muroyama, K. and Fan, L.S., Fundamentals of gas–liquid–solid fluidization, AIChE J., 31(1), 1, 1985. 106. Fan, L.S., Gas–Liquid–Solid Fluidization Engineering, Butterworths, Boston, 1989. 107. Fan, L.S., Muroyama, K., and Chern, S.H., Hydrodynamics of inverse fluidization in liquid–solid and gas–liquid–solid systems, Chem. Eng. J., 24, 143, 1982. 108. Bhatia, V.K. and Epstein, N., Three phase fluidization; a generalized wake model, in Fluidization and Its Applications, Angelino, H., Couderc, J.P., Gibert, H., and Laguerie, Co., Cepadues Editions, Toulouse, France, 1974, 380. 109. Dakshinamurthy, P., Subrahmanyam, V., and Rao, J.N., Bed porosities in gas–liquid fluidization, Ind. Eng. Chem. Process Des. Dev., 10, 322, 1971. 110. Begovich, J.M. and Watson, J.S., Hydrodynamic characteristics of three phase fluidized beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 190. 111. Nikov, I., Grandjean, B.P.A., Carreau, P.J., and Paris, J., Viscosity effect in cocurrent three phase fluidization, AIChE J., 36(10), 1613, 1990. 112. Kato, Y., Uchida, K., and Morooka, S., Liquid hold-up and heat transfer coefficient between bed and wall in liquid–solid and gas–liquid–solid fluidized beds, Powder Technol., 28, 173, 1981. 113. Saberian-Broudjenni, M., Wild, G., Charpentier, J.C., Fortin, Y., Euzen, J.P., and Patoux, R., Contribution to the hydro-dynamics study of gas–liquid–solid fluidized bed reactors, Int. Chem. Eng., 27(3), 423, 1987. 114. Grace, J.R., Fluidization, in Handbook of Multiphase Systems, Hestroni, G., Ed., McGraw-Hill–Hemisphere, Washington, D.C., 9, 1982, 8.1.17. 115. Douglas, H.R., Snider, I.W.A., and Tomlinson, G.H., II, The turbulent contact absorber, Chem. Eng. Prog., 59(12), 85, 1963. 116. Visvanathan, C. and Leung, L.S., On the design of a fluidized bed scrubber, Ind. Eng. Chem. Process Des. Dev., 24, 677, 1985.

© 1999 by CRC Press LLC

117. Barile, R.G., Dengler, J.L., and Hertwig, T.A., Performance and design of a turbulent bed cooling tower, AIChE Symp. Ser., 70, 154, 1975. 118. Gelperin, N.I., Savchenko, V.I., and Grishko, V.Z., Some hydrodynamics laws of absorption apparatus packed with fluidized spheres, Theor. Found. Chem. Eng., 2, 65, 1968. 119. Kito, M., Tabei, K., and Murata, K., Gas and liquid holdups in mobile bed under the countercurrent flow of air and liquid, Ind. Eng. Chem. Process Des. Dev., 17, 568, 1978. 120. Wozniak, M., Pressure drop and effective interfacial area in a column with a mobile bed, Int. Chem. Eng., 17, 553, 1977. 121. Draijer, W., Heat transfer in fluidized bed-boilers, in Fluidzed Bed Combustion, Radovanovic, M., Ed., Hemisphere Publishing, Washington, D.C., 1986, 211. 122. Saxena, S.C., Grewal, N.S., Gabor, J.D., Zabrodsky, S.S., and Galershtein, D.M., Heat transfer between a gas fluidized bed and immersed tubes, in Advances in Heat Transfer, Vol. 14, Irvine, T.F. and Hartnett, J.P., Eds., Academic Press, New York, 1978, 149. 123. Zabrodsky, S.S., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, chap. 7–10. 124. Botterill, J.S.M., Fluid Bed Heat Transfer, Academic Press, London, 1975. 125. Ranz, W.E. and Marshall, W.R., Evaporation from drops. I, Chem. Eng. Prog., 48, 141, 1952. 126. Chen, P. and Pei, D.C.T., Fluid particle heat transfer in fluidized beds of mixed-sized particles, Can. J. Chem. Eng., 62, 469, 1984. 127. Gutfinger, C. and Abuaf, N., in Advances in Heat Transfer, Vol. 10, Hartnett, J.P. and Irvine, T.F., Jr., Eds., Academic Press, New York, 1974, 167. 128. Bartel, W.J., Heat transfer from a horizontal bundle of tubes in an air fluidized bed, Ph.D. thesis, Montana State University, Bozeman, 1971. 129. Saxena, S.C. and Gabor, J.D., Mechanism of heat transfer between a surface and a gas fluidized bed for combustion application, Prog. Energy Combust. Sci., 7(2), 73, 1981. 130. Van Heerden, C., Nobel, A.P., and van Krevelen, D., Studies on fluidization. I. The critical mass velocity, Chem. Eng. Sci., 1, 37, 1951. 131. Ziegler, E.N. and Brazelton, W.T., Mechanism of heat transfer to a fixed surface in a fluidized bed, Ind. Eng. Chem. Fund., 3(2), 94, 1964. 132. Botterill, J.S.M., The mechanism of heat transfer to fluidized beds, in Proc. Int. Symp. on Fluidization, Drinkenburg, A.A.H., Ed., Netherlands University Press, Amsterdam, 1967, 183. 133. Yamazaki, R. and Jimbo, G., Heat transfer between fluidized beds and heated surfaces, J. Chem. Eng. Jpn., 3(1), 44, 1970. 134. Wicke, E. and Fetting, F., Warmeubertragung in GaswirbelSchichten, Chem. Ing. Tech., 26, 301, 1954. 135. Mickley, H.S. and Fairbanks, D.F., Mechanism of heat transfer to fluidized beds, AIChE J., 1, 374, 1965. 136. Botterill, J.S.M. and Williams, J.R., The mechanism of heat transfer to fluidized beds, Trans. Inst. Chem. Eng., 41, 217, 1963. 137. Botterill, J.S.M. and Butt, M.H.D., Achieving high heat transfer rates in fluidized beds, Br. Chem. Eng., 13(7), 1000, 1968. 138. Gabor, J.D., Wall to bed heat transfer in fluidized beds and packed beds, Chem. Eng. Prog. Symp. Ser., 66(105), 76, 1970.

© 1999 by CRC Press LLC

139. Zabrodsky, S.S., Hydrodynamics and Heat Transfer, MIT Press, Cambridge, MA, 1966, chap. 8. 140. Selzer, V.W. and Thomson, W.J., Fluidized bed heat transfer — the packet theory revisited. AIChE Symp. Ser., 73(161), 29, 1977. 141. Xavier, A.M. and Davidson, J.F., Convective heat transfer in fluidized bed, in Fluidization, 2nd ed., Davidson, J.F., Clift, R., and Harrison, D., Eds., Academic Press, London, 1985, chap. 13. 142. Krupiczka, R., Analysis of thermal conductivity in granular materials, Int. Chem. Eng., 7, 122, 1967. 143. Ranz, W.E., Friction and transfer coefficients for single particles and packed beds, Chem. Eng. Prog., 48(3), 247, 1952. 144. Baron, T., Generalized graphical method for the design of fixed bed catalytic reactors, Chem. Eng. Prog., 48(3), 118, 1952. 145. Baskakov, A.P., Vitt, O.K., Kirakosyan, V.A., Maskaev, V.K., and Filippovsky, N.F., Investigation of heat transfer coefficient pulsations and of the mechanism of heat transfer from a surface immersed into a fluidized bed, in Proc. Int. Symp. on Fluidization and Its Applications, Cepadues Editions, Toulouse, France, 1974, 293. 146. Botterill, J.S.M. and Denloye, A.O.O., Bed to surface heat transfer in a fluidized bed of large particles, Powder Technol., 19, 197, 1978. 147. Gelprin, N.I. and Einstein, V.G., Heat transfer in fluidized beds, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 10. 148. Baskakov, A.P., Process of Heat and Mass Transfer in Fluidized Beds, Metallurgia, Moscow, 1978, 184. 149. Kunii, D. and Levenspiel, O., Eds., Fluidized Engineering, John Wiley & Sons, New York, 1968, chap. 9. 150. Botterill, J.S.M., Ed., Fluid Bed Heat Transfer, Academic Press, London, 1975, 81. 151. Zabrodsky, S.S., Ed., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, chap. 10. 152. Gelprin, N.I. and Einstein, V.G., Heat transfer in fluidized beds, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 10. 153. Grewal, N.S., Experimental and Theoretical Investigations of Heat Transfer Between a Gas Solid Fluidized Bed and Immersed Tubes, Ph.D. thesis, University of Illinois at Chicago, 1979. 154. Gutfinger, C. and Abuaf, N., Heat transfer in fluidized beds, in Advances in Heat Transfer, Vol. 10, Irvine, T.F., Jr. and Hartnett, J.P., Eds., Academic Press, New York, 1964, 167. 155. Zabrodsky, S.S., Antonishin, V.N., and Paranas, A.L., On fluidized bed surface heat transfer, Can. J. Chem. Eng., 54, 52, 1976. 156. Grewal, N.S. and Saxena, S.C., Effect of surface roughness on heat transfer from horizontal immersed tubes in a fluidized bed, Heat Transfer, 101(3), 397, 1979. 157. Vreedenberg, H., Heat transfer between a fluidized bed and a horizontal tube, Chem. Eng. Sci., 1(1), 52, 1958. 158. Vreedenberg, H., Heat transfer between a fluidized bed and verticle tube, Chem. Eng. Sci., 11, 274, 1960. 159. Wender, L. and Cooper, G.T., Heat transfer between fluidized bed and boundary surfaces: correlation data, AIChE J., 1(1), 15, 1958. 160. Baskakov, A.P., Berg, B.V., Vitt, O.K., Fillippovsky, N.F., Kiroskosyan, V.A., Goldobin, J.M., and Maskaev, V.K., Heat transfer to objects immersed in fluidized beds, Powder Technol., 8, 273, 1973.

© 1999 by CRC Press LLC

161. Dahloff, B. and Von Brachel, H., Warmeubergang, and horizontal cenem fluidat bett ange ordneten rohrbundelu, Chem. Ing. Tech., 40, 373, 1968. 162. Noak, R., Lokaler Warmeubergang, an horizontalen Fohren in Wirbelschichten, Chem. Ing. Tech., 42(6), 371, 1970. 163. Blenke, H. and Neukirchem. B., Gastaltung horizontaler rohrbundel in gas-wirbelschichtreaktorn kach warmetchnischen, Chem. Ing. Tech., 45(5), 307, 1973. 164. McClaren, J. and Williams, D.F., Combustion efficiency, sulphur retention and heat transfer in pilot fluidized bed combustion, J. Inst. Fuel, 42, 303, 1969. 165. Lese, J.K. and Kermode, R.I., Heat transfer from a horizontal tube to a fluidized bed in the presence of unheated tubes, Can. J. Chem. Eng., 50, 44, 1972. 166. Bartel, W.J. and Genetti, W.F., Heat transfer from a horizontal bundle of bare and finned tubes in an air fluidized bed, Chem. Eng. Prog. Symp., 69(128), 85, 1973. 167. Andeen, B.R. and Glicksman, L.R., Heat Transfer to Horizontal Tubes in Shallow Fluidized Beds, Heat Transfer Conf. Paper No. 76 H-67, ASME-AIChE., St. Louis, August 9–11, 1976. 168. Petrie, J.C., Freeby, W.A., and Buckham, J.A., In bed exchangers, Chem. Eng. Prog., 64(67), 45, 1968. 169. Zabrodsky, S.S., Hydrodynamics and Heat Transfer in Fluidized Beds, MIT Press, Cambridge, MA, 1966, 270. 170. Genetti, W.E., Sehmall, R.A., and Grimmett, E.S., The effect of tube orientation on heat transfer with bare and finned tubes in a fluidized bed, Chem. Eng. Prog. Symp. Ser., 67(116), 90, 1971. 171. Ternovskaya, A.N. and Korenbrg, Y.G., Pyrite Kilning in a Fluidized Bed, Izd. Khimiya, Moscow, 1971. 172. Saxena, S.C. and Grewal, N.S., Heat transfer between a horizontal tube and a gas solid fluidized bed, Int. J. Heat Mass Transfer, 23, 1505, 1980. 173. Sarkits, V.B., Heat Transfer from Suspended Beds of Granular Materials to Heat Transfer Surfaces (Russian), Leningrad Technol. Inst., Lensovieta, 1959. 174. Miller, C.O. and Logwink, A.K., Fluidization studies of solid particles, Ind. Eng. Chem., 43, 1220, 1951. 175. Wicke, E. and Hedden, F., Stromungsformen and warmeubertragung in von luft aufgewirbelten schuttgutschichten, Chem. Ing. Tech., 24(2), 82, 1952. 176. Dow, W.M. and Jakob, M., Heat transfer between a vertical tube and a fluidized bed air mixture, Chem. Eng. Prog., 47, 637, 1951. 177. Gaffney, B.J. and Drew, T.B., Mass transfer from packing to organic solvents in single phase flow through a column, Ind. Eng. Chem., 42, 1120, 1950. 178. Traber, P.G., Pomaventsev, V.M., Mukhenov, I.P., and Sarkits, V.B., Heat transfer from a suspended layer of catalyst to the heat transfer surface, Zh. Prikl. Khim., 35, 2386, 1962. 179. Baskakov, A.P., High speed non oxidative heating and heat treatment in a fluidized bed, Metallurgia (Moscow), 1968. 180. Vreedenberg, H.A., Heat transfer between fluidized beds and vertically inserted tubes, J. Appl. Chem., 2(Suppl. issue), 26, 1952. 181. Brotz, W., Utersuchungen uber transportvorgange in durchstrometem gekontem Gut, Chem. Ing. Techn., 28, 165, 1956. 182. Sathiyamoorthy, D., Sridhar Rao, Ch., and Raja Rao, M., Effect of distrubutor on heat transfer from immersed surfaces in gas fluidized beds, Chem. Eng. J., 37, 149, 1988. 183. Baskakov, A.P. and Berg, B.V., Inzh. Fiz Zh. Akad. Nauk Belorussk SSR6, 8, 3, 1966. 184. Botterill, J.S.M., Teoman, Y., and Yüregir, K.R., Temperature effect on the heat transfer behaviour of gas fluidized beds, AIChE Symp. Ser., 77(208), 330, 1984.

© 1999 by CRC Press LLC

185. Molerus, O., Heat transfer in gas fluidized beds. 2. Dependence of heat transfer on gas velocity, Powder Technol., 15, 70, 1992. 186. Borodulya, V.A., Ganzh, V.A., and Podberlzsky, A.I., Heat transfer in a fluidized bed at high pressure, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 201. 187. Varygin, N.N. and Martyushin, I.G., A calculation of heat transfer surface area in fluidized bed equipment, Khim Mashinostr. (Moscow), 5, 6, 1959. 188. Baerg, A., Klassen, J., and Grishler, P.E., Heat transfer in fluidized solid beds, Can. J. Res. Sect. F, 28, 324, 1964. 189. Gelprin, N.I., Einstein, V.G., and Romanova, N.A., Khim. Prom. (Moscow), 5, 6, 1959. 190. Jakob, A. and Osberg, G.L., Effect of gas thermal conductivity on local heat transfer in a fluidized bed, Can. J. Chem. Eng., 35, 5, 1957. 191. Chechetkin, A.V., Vysokotemperature (High Temperature Carriers), Gosenergoizdat, Moscow, 1962. 192. Ruckenstein, E., Zh. Prikl. Khim., 35, 71, 1962; as cited by Gutfinger, C. and Abuaf, N., Heat transfer in fluidized beds, in Advances in Heat Transfer, Irvine T.F. and Hartnett, J.P., Eds., Academic Press, New York, 1964. 193. Richardson, J.F. and Shakiri, K.J., Heat transfer between a gas–solid fluidized bed and a small immersed surface, Chem. Eng. Sci., 34, 1019, 1979. 194. Kim, K.J., Kim, D.J., and Chun, K.S., Heat and mass transfer in fixed and fluidized bed reactors, Int. Chem. Eng., 8(5), 176, 1971. 195. Grewal, N.S. and Saxena, S.C., Maximum heat transfer coefficient between a horizontal tube and a gas–solid fluidized bed, Ind. Eng. Chem. Process Res. Dev., 20, 108, 1981. 196. Baskakov, A.P. and Panov, O.P., Comparison between measured maximum heat transfer coefficients to surfaces in fluidized beds, Inzh. Fiz. Zh., 45, 896, 1983. 197. Chen, P. and Pei, D.C.T., A model heat transfer between fluidized beds and immersed surfaces, Int. J. Heat Mass Transfer, 28(3), 675, 1985. 198. Palchenok, G.I. and Tamarin, A.I., Study of heat transfer between model particles and fluidized beds, Inzh. Fiz. Zh., 45, 427, 1983. 199. Sathiyamoorthy, D. and Raja Rao, M., Prediction of maximum heat transfer coefficient in gas fluidized bed, Int. J. Heat Mass Transfer, 25(5), 1027, 1992. 200. Molerus, O., Heat transfer in gas fluidized beds. I, Powder Technol., 70, 1, 1992. 201. Saxena, S.C. and Grewal, N.S., Effect of distributor design on heat transfer from an immersed horizontal tube in a fluidized bed, Chem. Eng. J., 18, 197, 1979. 202. Sathiyamoorthy, D., Sridhar Rao, Ch., and Raja Rao, M., Heat transfer in gas fluidized beds using multi-orifice distributors, Indian J. Technol., 25, 219, 1987. 203. Frantz, J.F., Fluid to particle heat transfer in fluidized beds, Chem. Eng. Prog., 57, 65, 1961. 204. Wasmund, B. and Smith, J.W., Wall to fluid heat transfer in liquid fluidized bed, Can. J. Chem. Eng., 45, 156, 1967. 205. Romani, M.N. and Richardson, J.F., Heat transfer from immersed surfaces to liquid fluidized beds, Lett. Heat Mass Transfer, 1, 55, 1974. 206. Krishnamurthy, N. and Sathiyamoorthy, D., Mechanism of heat transfer in liquid fluidized beds, in Proc. National Heat and Mass Transfer Conf., IISc Bangalore, India, Ind. Soc. Heat Mass Trans., Madras, 1987, 81. 207. Mishra, P. and Farid, M.M., Wall to bed heat transfer in liquid fluidized beds, Ind. Chem. Eng., 28(3), 30, 1986. 208. Hog, G.W. and Grimmett, E.S., Effects of pulsation on heat transfer in a liquid fluidized bed, Chem. Eng. Prog. Symp. Ser., 62(67), 51, 1966.

© 1999 by CRC Press LLC

209. Østergaard, K., Discussion in the paper of R. Turner, Fluidization, Soc. for the Chem. Ind., London, 1964, 58. 210. Viswanathan, S.A., Kakar, A.S., and Murti, P.S., Effect of dispersing bubbles into liquid fluidized beds on heat transfer holdup at constant bed expansion, Chem. Eng. Sci., 20, 903, 1964. 211. Armstrong, E.A., Baker, C.G.J., and Bergougnou, M.A., Heat transfer and hydrodynamics studies on three phase fluidized beds, in Fluidization Technology, Keairns, D.L., Ed., Hemisphere Publishing, New York, 1976, 453. 212. Baker, C.G.J., Armstrong, F.R., and Bergongnou, M.A., Heat transfer in three phase fluidized beds, Powder Technol., 21, 195, 1978. 213. Kato, Y., Uchida, K., Kago, T., and Morooka, S., Liquid holdup and heat transfer coefficient between bed and wall in liquid–solid and gas–liquid–solid fluidized beds, Powder Technol., 28, 173, 1981. 214. Kato, Y., Kago, T., and Morooka, S., Heat transfer from wall to packed and fluidized bed for gas–liquid co-current up-flow, Kagaku Kogaku Ronbunshu, 4, 328, 1978. 215. Ermakova, A., Ziganskin, G.K., and Slinko, M.G., Hydrodynamics of a gas–liquid reactor with a fluidized bed of solid matter, Theor. Found. Chem. Eng., 4, 84, 1970. 216. Phillippovski, N.F. and Baskakov, A.P., Investigation of the temperature field in a fluidized bed close to a heated plate and heat transfer between them, Int. Chem. Eng., 13, 5, 1953. 217. Botterill, J.S.M. and Desai, M., Limiting factors in gas fluidized bed, Powder Technol., 6(2), 231, 1972. 218. Knuth, M. and Weinspach, P.M., Experimentelle unter-Schchung, des warme and Stoffubergangs and die Partikeln einer wirbelschicht beider desubbimation, Chem. Ing. Tech., 48, 893, 1976. 219. Kharchenko, N.V. and Makhorin, K.E., The rate of heat transfer between a fluidized bed and an immersed body at high temperatures, Int. Chem. Eng., 4, 650, 1964. 220. Mersmann, A., Determination of fluidizing velocity in fluidized beds by heat transfer measurements, Chem. Ing. Tech., 38(10), 1095, 1966. 221. Mersmann, A., Heat transfer in fluidized beds, Chem. Ing. Tech., 39(56), 349, 1967. 222. Mersmann, A., Zum Warmeuvergang Vuischen dispersen zueiphasensysten unl senkrechten heizflachen in erdschwereseld, Verfahrenstechnik, 10(10), 641, 1976. 223. Thones, D. and Kramers, H., Mass transfer from spheres in various regular packings to a flowing fluid ionic mass transfer in the presence of fluidized solids, Chem. Eng. Sci., 8, 271, 1958. 224. Jaganadharaju, G.J.V. and Rao Venkata, C., Ionic mass transfer in the presence of fluidized solids, Indian J. Technol., 3, 201, 1965. 225. Zeigler, E.N. and Holmes, J.T., Mass transfer from fixed surfaces to gas fluidized beds, Chem. Eng. Sci., 21, 117, 1966. 226. Einstein, V.G. and Gelpin, N.I., Heat transfer between a fluidized bed and a surface, Int. Chem. Eng., 6, 67, 1966. 227. Beek, W.J., Mass transfer in fluidized beds, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, New York, 1971, chap. 9. 228. Froessling, N., Gerland Beitr, Geophys. J., 52, 170, 1938. 229. Ranz, W.E., Friction and transfer coefficients for single particles and packed beds, Chem. Eng. Prog., 48, 247, 1952. 230. Fan, L.T., Yang, C.Y., and Wen, C.Y., Mass transfer in semi fluidized beds for solid liquid system, AIChE J., 6, 482, 1960. 231. Rebnick, W.E. and White, R.R., Mass transfer in systems of gas and fluidized solids, Chem. Eng. Prog., 45, 377, 1949.

© 1999 by CRC Press LLC

232. Kettenring, K.N., Manderfield, E.L., and Smith, J.M., Heat and mass transfer in fluidized systems, Chem. Eng. Prog., 46, 139, 1950. 233. Kunii, D. and Levenspiel, O., Fluidization Engineering, John Wiley & Sons, New York, 1969, 200. 234. Richardson, J.F. and Szekely, J., Mass transfer in a fluidized bed, Trans. Inst. Chem. Eng., 39, 212, 1961. 235. Wakao, N., Particle to fluid heat/mass transfer coefficients in packed bed catalytic reactors, in Recent Advances in the Engineering Analyses of Chemically Reacting Systems, Doraiswamy, L.K., Ed., Wiley Eastern Ltd., New Delhi, 1984, chap. 2. 236. Damronglerd, S., Couderc, J.P., and Angelino, H., Mass transfer in particulate fluidization, Trans. Inst. Chem. Eng., 53, 175, 1975. 237. Wakao, N. and Funazkri, T., Effect of fluid dispersion coefficients on particle to fluid mass transfer coefficients in packed beds — correlation of Sherwood number, Chem. Eng. Sci., 33, 1375, 1978. 238. Kunii, D. and Levenspiel, O., Bubbling bed model — model for flow of gas through a fluidized bed, Ind. Eng. Chem. Fundam., 7, 446, 1968. 239. Chiba, T. and Kobayashi, H., Gas exchange between the bubble and emulsion phases in gas fluidized beds, Chem. Eng. Sci., 25, 1375, 1970. 240. Murray, J.D., On the mathematics of fluidization. 1. Fundamental equations and wave propagation, J. Fluid Mech., 21, 465, 1965; 2. Steady motion of fully developed bubbles, J. Fluid Mech., 22, 57, 1965. 241. Partridge, B.A. and Rowe, P.N., Chemical reaction in a bubbling gas fluidized bed, Trans. Inst. Chem. Eng., 44, T335, 1966. 242. Stephens, G.K., Sinclair, R.J., and Potter, O.E., Gas exchange between bubbles and dense phase in a fluidized bed, Powder Technol., 1, 157, 1967. 243. Kunii, D. and Levenspiel, O., Effect of particles in bubbles on fluidized bed mass and heat transfer kintics, J. Chem. Eng. Jpn., 24(2), 183, 1991. 244. Yasui, G. and Johanson, L.N., Characteristics of gas packets in fluidized beds, AIChE J., 4, 445, 1958. 245. Whitehead, A.B. and Young, A.D., Fluidization performance in large scale equipment. I, in Proc. Int. Symp. on Fluidization, Drikenburg, A.A.H., Ed., Netherlands University Press, 1967, 284. 246. Park, W.H., Kang, W.K., Capes, C.E., and Osberg, G.L., The properties of bubbles in fluidized beds of conducting particles as measured by an electro resistivity probe, Chem. Eng. Sci., 24, 851, 1969. 247. Kobayashi, H., Arai, F., and Chiba, T., Behaviour bubbles in gas fluidized beds, Chem. Eng. Sci., 21, 239, 1967. 248. Kato, K. and Wen, C.Y., Bubble assemblage model for fluidized bed catalytic reactors, Chem. Eng. Sci., 24, 1351, 1969. 249. Chiba, T., Terashima, K., and Kobayashi, H.J., Lateral distribution of bubble sizes in two dimensional gas fluidized bed, Chem. Eng. Jpn., 8, 167, 1975. 250. Mori, S. and Wen, C.Y., Estimation of bubble diameter in gaseous fluidized bed, AIChE J., 21, 109, 1975. 251. Hirma, T., Ishida, M., and Shirai, T., Kagaku Kogaku Ronbunshu, 1, 272, 1975. 252. Rowe, P.N., Prediction of bubble size in a gas fluidized bed, Chem. Eng. Sci., 31, 285, 1976. 253. Darton, R.C., La Nauze, R.D., Davidson, J.F., and Harrison, D., Bubble growth due to coalescence in fluidized beds, Trans. Inst. Chem. Eng., 55, 274, 1977. 254. Werther, J., Mathematical modelling of fluidized bed reactors, Chem. Ing. Tech., 50, 850, 1978.

© 1999 by CRC Press LLC

255. Werther, J., Bubbles in gas fluidized beds. I and II, Trans. Inst. Chem. Eng., 52, 149, 1974. 256. Davidson, J.F. and Schuler, B.O.G., Bubble formation at an orifice in inviscid liquid, Trans. Inst. Chem. Eng., 38, 335, 1960. 257. Chiba, T., Terashima, K., and Kobayashi, H., Behaviour of bubbles in gas–solid fluidized beds; initial formation of bubbles, Chem. Eng. Sci., 27, 965, 1972. 258. Miwa, K., Mori, S., Kato, T., and Muchi, I., Behaviour of bubble in a gaseous fluidized bed, Int. Chem. Eng., 12, 187, 1972. 259. Fryer, C. and Potter, O.E., Experimental investigation of models for fluidized bed catalytic reactors, AIChE J., 22, 38, 1976. 260. Davis, L. and Richardson, J.F., Gas interchange between bubbles and continuous phase in a fluidized bed, Br. Chem. Eng., 12, 1223, 1967. 261. Toei, R., Matsuno, R., Nishitani, K., Miyagawa, H., and Komagawa, Y., Gas transfer between a bubble and the continuous phase in a gas fluidized bed, Int. Chem. Eng., 9, 358, 1969; Kogaku Kogaku, 32, 565, 1968. 262. Kobayashi, H., Arai, F., and Chiba, T., Kagaku Kogaku, 31, 239, 1967. 263. Miyauchi, T. and Morooka, S., Mass transfer rate between bubble and emulsion phase in fluid bed, J. Chem. Eng. Jpn., 33, 880, 1969. 264. Chiba, T. and Kobayashi, H., Gas exchange between the bubble and emulsion phases in gas fluidized beds, Chem. Eng. Sci., 25, 1375, 1970. 265. Calderbank, P.H., Pereira, J., and Burgss, M.M., The physical and mass transfer properties of bubbles in fluidized beds of electrically conducting particles, in Int. Conf., Pacific Grove, CA, 1976, 261. 266. Mori, S. and Muchi, I., Theoretical analysis of catalytic reaction in a fluidized bed, Kogaku Kogaku, 5, 251, 1972. 267. Davidson, J.F., Harrison, D., Darton, R.C., and LaNauze, R.D., in Chemical Reactor Theory — A Review, Lapidus, L. and Amundson, N.R., Eds., Prentice-Hall, Englewood Cliffs, NJ, 1977. 268. Alper, E., Wichtendahl, B., and Deckwer, W.D., Gas absorption mechanism in catalytic slurry reactors, Chem. Eng. Sci., 35, 217, 1980. 269. Østergaard, K. and Suchozebrski, W., Gas–liquid mass transfer in gas liquid fluidized beds, in Proc. Eur. Symp. React. Eng., Pergamon Press, Oxford, 1969, 21. 270. Østergaard, K. and Fosbol, P., Transfer of oxygen across the gas–liquid interface in gas–liquid fluidized beds, Chem. Eng. J., 3, 105, 1972. 271. Alvarez-Cuenca, M. and Nerenberg, M.A., The plug flow model for mass transfer in three phase fluidized beds and bubble columns, Can. J. Chem. Eng., 59, 739, 1981. 272. Dankwerts, P.W., Gas–Liquid Reactions, McGraw-Hill, New York, 1970, 15. 273. Geankopolis, C.J., Mass Transport Phenomena, Holt, Reinhart and Winston, New York, 1972, 316. 274. Sharma, M.M. and Danckwerts, P.V., Chemical method of measuring interfacial area and mass transfer coefficients in two-fluid systems, Br. Chem. Eng., 524, 208, 1970. 275. Gay, F.V., Holtman, C.B., and Sheppard, N.F., Mass Transfer in Three Phase Fluidized Beds, ORNL Report No. ORNL/MIT-283, Oak Ridge, TN, December 21, 1978. 276. Patwari, A.N., Nguyen-tien, K., Schumpe, A., and Deckwer, W.D., Three phase fluidized beds with viscous liquid: hydrodynamics and mass transfer, Chem. Eng. Commun., 40, 49, 1989. 277. Schumpe, A., Deckwer, W.D., and Nigam, K.D.P., Gas fluid mass transfer in three phase fluidized beds with viscous pseudoplastic liquids, Can. J. Chem. Eng., 67, 873, 1989.

© 1999 by CRC Press LLC

278. Tang, W.T. and Fan, L.S., Gas–liquid mass transfer in a three phase fluidized bed containing low density particles, Ind. Eng. Chem. Res., 29, 128, 1990. 279. Alvarez-Cuenca, M., Oxygen Mass Transfer in Bubble Columns and Three Phase Fluidized Beds, Ph.D. thesis, University of Western Ontario, Canada, 1979. 280. Alvarez-Cuenca, M., Oxygen mass transfer in three phase fluidized beds working at large flow rates, Can. J. Chem. Eng., 61, 5, 1983. 281. Muroyama, K. and Fan, L.S., Fundamentals of gas–liquid–solid fluidization, AIChE J., 31(1), 1, 1985. 282. Dhanuka, V.R. and Stepanek, J.B., Gas–liquid mass transfer in a three phase fluidized bed, in Fluidization, Grace, J.R. and Matson, G.M., Eds., Plenum Press, New York, 1980, 261. 283. Kim, S.D. and Chang, H.S., Hydrodynamics and bubble breakage in three phase fluidized beds, Hwahak Konghak, 17, 407, 1979. 284. Kim, J.O. and Kim, S.D., Bubble characteristics in three phase fluidized beds of floating bubble breakers, Particulate Sci. Technol., 5, 309, 1987. 285. Kim, S.D., Lee, Y.J., and Kim, J.O., Heat transfer and hydrodynamic studies on two and three phase fluidized beds of floating bubble breakers, Exp. Thermal Fluid Sci., 1, 237, 1988. 286. Kim, J.O. and Kim, S.D., Gas–liquid mass transfer in a three phase fluidized bed with floating bubble breakers, Can. J. Chem. Eng., 68, 368, 1990. 287. Morooka, S., Kusakabe, K., and Kato, Y., Mass transfer coefficient at the wall of a rectangular fluidized bed for liquid–solid and gas–liquid–solid systems, Kogaku Kogaku Ronbunshu, 5, 162, 1979; Int. Chem. Eng., 20, 433, 1980. 288. Batchelor, G.K., Mass transfer from small particles suspended in turbulent fluid, J. Fluid Mech., 98, 609, 1980. 289. Arters, D. and Fan, L.S., Liquid–solid mass transfer in a gas–liquid–solid fluidized bed, presented at AIChE Meeting, San Francisco, November 25–30, 1984. 290. Groshe, E.W., Analysis of gas fluidized solid systems by X-ray absorption, AIChE J., 1(3), 358, 1955. 291. Bakker, P.I. and Heertjes, P.M., Porosity measurement in fluidized beds, Br. Chem. Eng., 3, 240, 1958. 292. Fan, L.T., Chan, J.L., and Baile, R.C., Axial solid distribution in gas–solid fluidized beds, AIChE J., 8(2), 239, 1962. 293. Rowe, P.N., MacGillivray, H.J., and Chessman, D.J., Gas discharge from an orifice into a gas fluidized bed, Trans. Inst. Chem. Eng., 57, 194, 1979. 294. Yang, W.C. and Keairns, D.L., Design and operating parameters for a fluidized bed agglomerating combustor gasifier, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 208. 295. Yang, W.C. and Keairns, D.L., Estimating the jet penetration depth of multiple grid jets, Ind. Eng. Chem. Fundam., 18, 317, 1979. 296. Zenz, F.A. and Weil, N.A., A theoretical-empirical approach to the mechanism of particle entrainment from fluidized beds, AIChE J., 4, 472, 1958. 297. Shakova, N.A., Outflow of turbulent jets into a fluidized bed, Inzh. Fiz. Zh., 14, 6, 1968. 298. Basov, V.A., Markhevka, V.I., Melik-Akhnazarov, T.Kh., and Orochko, D.I., Investigation of the structure of a non-uniform fluidized bed, Int. Chem. Eng., 9, 263, 1969. 299. Merry, J.M.D., Penetration of vertical jets into fluidized beds, AIChE J., 21, 507, 1975. 300. Deole, N.R., Study of Jets in Three Dimensional Gas Fluidized Beds, M.S. thesis, West Virginia University, Morgantown, 1980.

© 1999 by CRC Press LLC

301. Yang, W., Jet penetration in a pressurised fluidized bed, Ind. Eng. Chem. Fundam., 2, 297, 1981. 302. Blake, T.R., Wen, C.Y., and Ku, C.A., The correlation of jet penetration measurements in fluidized beds using non-dimensional hydrodynamic parameters, AIChE Symp. Ser., 80(234), 42, 1984. 303. Kececioglu, I., Yang, W., and Keairns, D.L., Fate of solids fed pneumatically through a jet into a fluidized bed, AIChE J., 30, 99, 1984. 304. Wen, C.Y., Deole, N.R., and Chen, L.H., A study of jets in a three dimensional gas fluidized bed, Powder Technol., 31, 1975, 1982. 305. Kozin, V.E. and Baskakov, A.P., An investigation of the grid zone of a fluidized bed above cap-type distributors, Int. Chem. Eng., 8(2), 257, 1968. 306. Filla, M., Massimilla, L., and Vaccaro, S., Gas jets in fluidized beds and spouts: a comparison of experimental behaviour and models, Can. J. Chem. Eng., 61, 370, 1983. 307. Donadono, S., Maresca, A., and Massimilla, L., Gas injection in shallow beds of fluidized coarse solids, Ing. Chim. Ital., 16, 1, 1980. 308. Lofroy, G.A. and Davidson, J.F., The mechanics of spouted beds, Trans. Inst. Chem. Eng., 47, T120, 1969. 309. Littman, H., Morgan, M.H., III, Vukovic, D.V., Zdanski, F.K., and Graveie, Z.B., A method for predicting the relationship between the spout and inlet tube radii in a spouted bed at its maximum spoutable height, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 381. 310. Yang, W.C. and Keairns, D.L., Momentum dissipation and gas entrainment in a gas–solid two-phase jet in a fluidized bed, in Fluidization, Grace, J.R. and Matsen, J.M., Eds, Plenum Press, New York, 1980, 305. 311. Mamuro, T. and Hattori, H., Flow pattern of fluid spouted beds, J. Chem. Jpn., 1, 5, 1968. 312. Donsi, G., Massimilla, L., and Colantuonio, L., The dispersion of axisymmetric gas jets in fluidized beds, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 297. 313. Becker, H.A., An investigation of laws governing the spouting of coarse particles, Chem. Eng. Sci., 13, 245, 1961. 314. Lim, C.J. and Mathur, K.B., Modelling of particle movement in spouted beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, 104. 315. Ching, Ho. T., Yutani, N., Fan, L.T., and Walawender, W.P., Stochastic modelling of bubble formation on the grid in a gas–solid fluidized bed, Can. J. Chem. Eng., 61, 654, 1983. 316. Hsiung, T.P. and Grace, J.R., Formation of bubbles at an orifice in fluidized beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, chap. 21. 317. Werther, J., Effect of gas distributor on the hydrodynamics of gas fluidized beds, Ger. Chem. Eng., 1, 166, 1978. 318. Brien, C.L., Bergougnou, M.A., and Baker, C.G.J., Grid leakage (weeping, dumping, particle back flow) in gas fluidized beds, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 413. 319. Wen, C.Y., Krishnan, R., Dutta, S., and Khosravi, R., Dead zone height near the grid region of fluidized bed, in Proc. 2nd Eng. Foundation Conf., Cambridge, England, 1978, 32. 320. Wen, C.Y. and Dutta, S., Research needs for the analysis, design and scale up of fluidized beds, AIChE Symp. Ser., 73(161), 2, 1977.

© 1999 by CRC Press LLC

321. Wen, C.Y., Krishnan, R., Khosravi, R., and Dutta, S., Dead zone height near the grid of fluidized beds, in Fluidization, Davidson, J.F. and Keairns, D.L., Eds., Cambridge University Press, Cambridge, 1978, chap. 2–3. 322. Wen, C.Y., Krishnan, R., and Kalyanaraman, R., Particle mixing near the grid region of fluidized beds, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 405. 323. Leva, M. and Wen, C.Y., Elutriation, in Fluidization, Davidson, J.F. and Harrison, D., Eds., Academic Press, London, 1971, chap. 14. 324. Geldart, D., Elutriation, in Fluidization, Davidson, J.F., Clift, R., and Harrison, D., Eds., Academic Press, London, 1985, chap. 11. 325. Rowe, P.N. and Partridge, B.A., X-ray study of bubbles in fluidized beds, Trans. Inst. Chem. Eng., 43, T157, 1965. 326. Kunii, D. and Levenspiel, O., Fluidization Engineering, John Wiley & Sons, New York, 1979, chap. 10. 327. Kunii, D. and Levenspiel, O., Entrainment of solids from fluidized beds. 1. Hold up in the free board, 2. Operation of fast fluidized beds, Powder Technol., 61, 193, 1990. 328. Kraft, W.W., Ulrich, W., and O’Connor, W., in Fluidization, Othmer, D.F., Ed., Reinhold, New York, 1956, 194. 329. Lewis, W.K., Gilliland, E.R., and Lang, P.M., Entrainment from fluidized beds, Chem. Eng. Prog. Symp. Ser., 58(68), 65, 1962. 330. Soroko, V., Mikhalev, M., and Mukhlenov, I., Calculation of the minimum height of the space above the in fluidized bed contact equipment, Int. Chem. Eng. 9, 280, 1969. 331. Horio, M., Taki, A., Hsieh, Y.S., and Muchi, I., Elutriation and particle transport through the freeboard of gas–solid fluidized bed, in Fluidization, Grace, J.R. and Matsen, J.M., Eds., Plenum Press, New York, 1980, 509.

© 1999 by CRC Press LLC

Thank you for interesting in our services. We are a non-profit group that run this website to share documents. We need your help to maintenance this website.

To keep our site running, we need your help to cover our server cost (about $400/m), a small donation will help us a lot.