Chapter 8

Gas Bubbles in Fluidized Beds

© Cedric Briens May 12, 2025

1
 Introduction
In bubbling fluidized beds, gas bubbles affect:
• bed expansion
• solids entrainment from the bed
• heat transfer between bed and heat exchangers
• heat transfer between particles and gas
• gas backmixing
• solids backmixing
• mass transfer between gas and solid particles
• conversion in fluidized bed reactors

2
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Effect of bed internals on gas bubbles
9. Effect of solids circulation patterns on gas bubbles

3
4
5
https://www.youtube.com/watch?v=5xoeV-LOUrY 6
 Slow motion (1000 frames/s)

https://www.youtube.com/watch?v=xEUIAuT7jRI 7
 General behavior of the gas bubbles
• Bubbles are formed at U > Umf

• Bubbles grow by coalescence

• Bubbles carry solid particles in their wake:

– Bubbles responsible for most of the solids mixing
– Fluidized bed is "CSTR" for solids

8
 Effect of bubble properties on
conversion in fluidized bed reactors
• To react with solids or with other gases at the surface of
solid catalysts, the gas must come into contact with the
solids, which are mostly in the emulsion phase.
• Large bubbles result in lower conversion rates:
– Large bubbles are faster: less time available for gas transfer to the
emulsion phase
– Large bubbles result in a smaller interfacial area between the
bubbles and the emulsion phase
– Weaker recirculation currents between large bubbles and emulsion
phase: less gas exchanged
• Good knowledge of the bubble properties is essential to the
prediction of the conversion in fluidized bed reactors
9
Solids circulation patterns around bubbles

10
 Coalescence of bubbles

• Coalescence between a small and a large bubble:

– larger bubble rises faster than the small bubble,
– catches up,
– deflects smaller bubble,
– sucks it into its wake

• There cannot be side-by-side coalescence

• Bubbles are continuously splitting and re-coalescing:

– greatly enhances gas-solid contacting
11
 Effect of particle properties on bubble shape

• Gas bubbles take a variety of shapes ranging from

the nearly spherical to narrow, flat crescents

• There are no clear-cut rules to predict the effect of

particle properties on bubble shape
12
 Bubbles in lab and pilot plant reactors
large H/D small H/D
(H = 0.5 to 1.0 m)
slugging freely bubbling

Group A
Vg > > Umf

D= 0.05m 0.25m 0.50 m good grid poor grid

If Vg is not
much larger than
Group B
B Umf, then, the
Group gas bubbles also
Vg > > Umf stop growing
with group B
13
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Effect of bed internals on gas bubbles
9. Effect of solids circulation patterns on gas bubbles

14
 Invasive techniques

• Use an array of small fast response probes:

– measure a property which is different in a gas bubble

• Comparing signals from the various probes of the array:

→ bubble velocity, size and frequency

• Generally poor results, especially if Vg >> Umf

15
 Invasive techniques
Δt12 z
Single bubble UB =
t12
probe
signal

time
probe
Δz signal

Δt1 time

z
Chord length =  t1
t12

16
 Invasive techniques
Many bubbles

probe
Δz signal

time

17
Non-invasive technique:
X-rays or γ-rays

• 2-dimensional picture
• Projection of all bubbles
in the X-ray path
• Only suitable for small
lab columns

18
 Non-invasive technique:
3D tomography using X-rays or γ-rays

https://www.youtube.com/watch?v=rwr9KJG5sBc 19
Use of gas tracers to get
the bubble velocity

• Inject a pulse of radioactive tracer

• Detectors at 2 different heights

• Fastest gas through bed is bubble gas

→ Ub from times of first detected tracer

20
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Effect of bed internals on gas bubbles
9. Effect of solids circulation patterns on gas bubbles

21
 Rise velocity of single bubble
at Umf

Ubmf = K (g db)1/2

• K = 0.671 or 0.711

22
 Freely bubbling beds

For freely bubbling beds, an empirical correction

factor must be used to account for:
– dense phase velocity
– bubble interactions (bubbles speed up when they
coalesce)

23
Correlation from Hilligardt/Werther

Ub =  (Vg - Umf) + 0.711  (g db)0.5

Two correction factors:

 for powder group and height above grid

 for powder group and column diameter

24
 

group A: z/D< 1.0,  = 0.8

z/D> 1.0,  = 0.8 (not specified by Werther)

group B: z/D< 1.7,  = 0.67

1.7<z/D<4.0,  = 0.51(z/D)0.5
z/D>4.0,  = 1.0

25
 

group A: 0.05 m < D < 1.0 m,  = 3.2 D0.33

D > 1.0 m,  = 3.2

group B: 0.10 m < D < 1.0 m,  = 2.0 D0.5

D > 1.0 m,  = 2.0

26
 Correlation from Kunii/Levenspiel

Kunii/Levenspiel proposed the following correlation

based on Werther's data:

group A:
Ub = 1.55 [(Vg - Umf) + 14.1 (db + 0.005)] D0.32 + 0.711 [g db]0.5

group B:
Ub = 1.6 [(Vg - Umf) + 1.13 db0.5] D1.35 + 0.711 [g db]0.5

Do not use!
27
 Generalized Darton’s correlation
for bubble size
• Darton (1979) derived the following correlation from a
bubble coalescence model

• The correlation is:

db = 0.54 g-0.2 (Vg-Umf)0.4 [z - h0e + 4 (A0e)1/2]0.8
with:
A0e = [db0e / 1.64]2.5 g0.5 / (Vg - Umf)

• Always check that:

db < dbmax

28
 Initial effective bubble size (db0e)
and its effective formation height (h0e)

1) Porous plate distributors

2) Perforated distributors:
– Bubbling regime
– Jetting regime

29
 Size of bubbles
formed at porous plates

Darton: db0e = db0 = K g-0.2 (Vg - Umf)0.4

h0e = 0

K between 0.03 and 0.09 depending on correlation

30
Bubbling at perforated gas distributor
time

31
 Bubbling
perforated plates or spargers
• All equations are of the type:
db0e = db0 =  g-0.2 (Vg - Umf)0.4 (A/Nh)0.4
h0e = 0

•  ranges from 0.87 to 1.64, depending on source

• Note that the distributor is supposed to operate

properly (e.g. no defluidized zones)
32
Jetting at perforated gas distributor
time

db0e z = h0e

z=0

33
Expansion cycle of a pulsating vertical gas jet
 Jetting distributors

Zenz showed that:

db0e  Ljet / 2

• for a downward facing jet:

h0e = (distributor height) - Ljet

• for an upward facing jet:

h0e = (distributor height) + Ljet

Ljet is the penetration depth of the gas jet 34

 Bubbling or Jetting ?

• Larger bubbles take less energy to form, per

unit of gas volume

• The most stable regime is the one which

gives the largest bubble size db0e, thus
minimizing the energy expenditure

35
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Effect of bed internals on gas bubbles
9. Effect of solids circulation patterns on gas bubbles

36
Slugging Regime

37
Slugging Regime
Group A Group B

"bullet" flat-nosed wall

slugs slugs slugs 38
 Slugging flow criteria

Criteria that must be satisfied for slugging to occur:

1) D < dbmax
2) Vg < minimum velocity for turbulent regime
3) db (calculated for a freely bubbling bed) > 0.4 D

39
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Effect of bed internals on gas bubbles
9. Effect of solids circulation patterns on gas bubbles

40
 Maximum bubble size dbmax
3 factors can limit the bubble size:

1. Column diameter: slugging

2. Coalescence limitation: db,coal < db0e (Nb0e)1/3

3. Maximum stable bubble size: only for Group A powders

Group A powder : d b max = Minimum of ( D, d b max,coal , d b max,stable )

Group B powder : d b max = Minimum of ( D, d b max,coal )

41
Maximum bubble size due to
coalescence limitation
 3  
volume balance : d b max,coal = N b0e  d3b0e 
6 6 
1
( Nb0e ) d b0e
d b max,coald b=max,coal
3

N b0e , d b0e
42
 Mechanisms which limit
bubble growth in group A powders

• Roof instabilities

• Solids sucked in from wake

(Dynamic equilibrium between coalescence

and splitting)
43
Roof instabilities

44
Solids sucked in from wake

45
 Prediction of
maximum stable bubble size

• Grace proposed:
dbmax,stable corresponds to:
Ubr = 0.7 (g dbmax,stable )0.5 = Ut
Ut for dp = 2.7 dpsm

• His selection of the characteristic dp is

purely empirical
• It is the best correlation proposed so far
46
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Beds of large particles (group D)
9. Effect of bed internals on gas bubbles
10. Effect of solids circulation patterns on gas bubbles
47
Differences between groups A and B

• Group A powders: db reduced when:

– the average dp is reduced or
– fines are added to the powder

• Group B powders:
– db not affected by relatively small changes in
particle size distribution

48
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Effect of bed internals on gas bubbles
9. Effect of solids circulation patterns on gas bubbles

49
Effect of bed internals on gas bubbles
• Bed internals (diplegs, heat exchanger tubes, baffles) have
large effects on gas bubbles
• Arrays of vertical rods or tubes:
– Bubbles split if center-to- center pitch between 33% and 90% of db
– Otherwise bubbles are stabilized by tubes, grow bigger and faster
• Arrays of horizontal tubes:
– Split bubbles with db much larger than tube diameter
– Bubbles recombine unless there are several successive tube rows
– Impede solids circulation and can lead to temperature gradients
• Horizontal perforated plates and variations
50
 Chapter Outline
1. Introduction
2. Qualitative Observations
3. Measurement of gas bubble properties
4. Prediction of bubble properties in freely bubbling beds
5. Slugging Regime
6. Maximum bubble size
7. Differences between groups A and B
8. Beds of large particles (group D)
9. Effect of bed internals on gas bubbles
10. Effect of solids circulation patterns on gas bubbles
51
Effect of solids circulation patterns
on gas bubbles

al an

Centr

nulus
52
Effect of solids circulation patterns
on gas bubbles

53