Bubble Column Technology
Bubble Column Technology
Bubble Column Technology
KGaA, Weinheim
10.1002/14356007.b04 275
Bubble Columns 1
Bubble Columns
For other industrial reactors and their applications, see Stirred-Tanc and Loop Reactors, Tubular
Reactors, Fixed-Bed Reactors, Fluidized-Bed Reactors, Three-Phase Trickle-Bed Reactors,
Reaction Columns, Thin-Film Reactors, Metallurgical Furnaces, and Biochemical Engineering.
Peter Zehner, BASF Aktiengesellschaft, Ludwigshafen, Federal Republic of Germany
Matthias Kraume, BASF Aktiengesellschaft, Ludwigshafen, Federal Republic of Germany
1. Introduction . . . . . . . . . . . . . . . . 2
2. Bubble Columns and Modications 4
2.1. Design and Applications . . . . . . . . 4
2.2. Gas Distribution . . . . . . . . . . . . . 5
2.3. Flow Regimes . . . . . . . . . . . . . . . 6
2.4. Fluid Dynamics . . . . . . . . . . . . . . 7
2.5. Bubble Size . . . . . . . . . . . . . . . . 8
2.6. Bubble Rise Velocity . . . . . . . . . . 9
2.7. Dispersion of the Liquid Phase . . . . 9
2.8. Dispersion of the Gas Phase . . . . . . 10
2.9. Gas Holdup . . . . . . . . . . . . . . . . 10
2.10. Specic Interfacial Area . . . . . . . . 12
2.11. Volumetric Mass-Transfer
Coefcient . . . . . . . . . . . . . . . . . 13
2.12. Heat Transfer . . . . . . . . . . . . . . . 14
2.13. Slurry Bubble Columns . . . . . . . . 14
2.14. Airlift Loop Reactors . . . . . . . . . . 16
3. Downow Bubble Columns . . . . . . 18
3.1. Design and Applications . . . . . . . . 19
3.2. Operating Conditions
and Gas Holdup . . . . . . . . . . . . . 20
3.3. Mass Transfer . . . . . . . . . . . . . . . 21
4. Jet Loop Reactors . . . . . . . . . . . . 22
4.1. Design and Applications . . . . . . . . 23
4.2. Typical Dimensions . . . . . . . . . . . 26
4.3. Energy Balance . . . . . . . . . . . . . . 27
4.4. Mixing Behavior and Fluid Dynamics 27
4.5. Gas Holdup . . . . . . . . . . . . . . . . 28
4.6. Mass Transfer . . . . . . . . . . . . . . . 30
4.7. Three-Phase Loop Reactor . . . . . . 31
5. References . . . . . . . . . . . . . . . . . 31
Symbols (see also Principles of Chemi-
cal Reaction Engineering and Model Reac-
tors and Their Design Equations)
Variables
a specic interfacial area, m
1
A interfacial area, m
2
d diameter, m
d
h
diameter of holes, m
d
i
inner diameter of draft tube, m
d
n
nozzle diameter, m
D diffusion or dispersion coefcient, m
2
/s
D
G, L
diffusion coefcient of dissolved gas in
liquid, m
2
/s
e
M
energy dissipation rate per unit mass,
W/kg
e
n
jet power per unit volume, W/m
3
e
V
energy dissipation rate per unit volume,
W/m
3
f fraction of cross-sectional area
f
i
fraction of cross-sectional area of draft
tube
F cross-sectional area, m
2
F
i
cross-sectional area of draft tube, m
2
F
R
cross-sectional area of reactor, m
2
h height, m
h
R
height of gas liquid mixture, m
h
t
height of reactor, m
J
D
dispersion ow
k
L
liquid-phase mass-transfer coefcient,
m/s
P power, W
r radial distance from column axis, m
t time, s
u supercial velocity, m/s
v velocity, m/s
v
rG
relative velocity of bubble swarm in liq-
uid, m/s
v
rS
relative velocity of particle swarm in liq-
uid, m/s
V volume, m
3
V volumetric ow rate, m
3
/s
z axial coordinate, m
2 Bubble Columns
Greek symbols
heat-transfer coefcient, WK
1
m
2
volume fraction
G
gas holdup
drag coefcient of circulation ow
dynamic viscosity, kg m
1
s
1
kinematic viscosity, m
2
/s
density, kg/m
3
density difference between liquid and
gas, kg/m
3
S
density difference between liquid and
solids, kg/m
3
surface tension, N/m
mass concentration, kg/m
3
Subscripts
a annular space
b bubble
bS Sauter diameter
c, circ circulation
D downow
G gas phase
h hole
i inside draft tube
L liquid
max maximum value
min minimum value
M per unit mass
n nozzle
p particle
r relative
R upow, reaction mixture
slip slip
S solids
t reactor
V per unit volume
1. Introduction
Bubble columns are devices in which gas, in
the form of bubbles, comes in contact with liq-
uid. The purpose may be simply to mix the liq-
uid phase. Far more often, however, substances
are transferred from one phase to the other, for
example, when gaseous reactants are dissolved
in a liquid or when liquid reaction products
are stripped. Both processes can take place si-
multaneously. Achemical or biological reaction
nearly always proceeds in the liquid phase. De-
pending on the application, special measures to
intensify mass transfer between the two phases
may be useful, or the residence-time distribution
of one or both phases may be modied.
The liquid may also contain inert, cat-
alytically active, or reactive particles in sus-
pension. Oxidation, hydrogenation, chlorina-
tion, phosgenation, alkylation, and other pro-
cesses have long been performed in bubble-
column reactors in the chemical industry. In
1978, more than 10
7
t/a of chemical products
were made in bubble columns [1]. Since then,
marked growth has occurred. Industrial reac-
tors for high-tonnage products have capacities
of 100 300 m
3
. Larger bubble columns, with
capacities up to 3000 m
3
, are employed as fer-
menters for protein production from methanol.
The largest units (20 000 m
3
) are those for waste-
water treatment.
Scientic interest in bubble columns has in-
creased considerably in the past 10 15 years.
Up to the mid-1970s, only 10 to 20 publications
appeared annually; by the mid- to late 1980s, the
number had increased to 80 per year. This led
to the development of many empirical correla-
tions and theoretical models enabling the math-
ematical simulation of bubble- column reactors.
Some academic research groups and commer-
cial software developers have offered simulation
programs.
The mixing of a liquid and a gas having only
partial mutual solubility is one of the unit op-
erations in chemical technology. As Figure 1
shows, this operation takes one of three prin-
cipal forms. The simplest design is the bubble
column (Fig. 1 A) in which gas is fed into the
column at the bottom and rises in the liquid, es-
caping from it at the upper surface; the gas is
consumed to a greater or lesser extent (depend-
ing on the intensity of mass transfer and chem-
ical reaction). When the off-gas contains high
concentrations of valuable reactants, part of it
is recycled to the reactor. This recycle design,
however, lowers the concentration prole in the
bubble column and must be optimized from an
economic standpoint. In a simple bubble column
the liquid is led in either cocurrently or counter-
currently to the upward gas streamand has a long
residence time. The ow direction of the liquid
phase has little effect on the gas-phase residence
time, which is comparatively short. Thus, in the
Bubble Columns 3
Figure 1. Principal methods of gas liquid mixing
A) Bubble column; B) Downow bubble column; C) Jet loop reactor
simple column, the ow of gas is always from
bottom to top, and the stream can be made up of
both fresh and recycle gas.
Longer gas-phase residence times can be
achieved with the downow bubble column
shown in Figure 1 B. The liquid is pumped down
through the column at a velocity of more than
20 cm/s, so that gas let in at the top is entrained in
the ow and can even be held in a suspension-
like state until it has reacted completely. Usu-
ally, however, unconsumed gas is removed with
the liquid and separated. Special designs per-
mit phase separation inside the apparatus. The
downow bubble column is used mainly when
large liquid streams are to be contacted with
small gas streams and a short liquid residence
time is required. The necessary velocity cannot
always be obtained with the liquid inlet to the
reactor. Thus, like the gas in an ordinary bub-
ble column, the liquid in the downow bubble
column can be recycled. Typical applications
for downow bubble columns are the ozonation
of drinking water and the treatment of water in
swimming pools. A special use of such devices
in the evacuation and compression of gases has
also been reported [2].
In both types of column energy must be sup-
plied continuously to the two-phase system to
keep the liquid and gas mixed. Only in this way
can separation of the phases be counteracted or
reversed. In the rst case, the simple bubble col-
umn, this energy is supplied by the gas. In the
downow bubble column the energy is supplied
by the downowing liquid.
A different mechanism comes into play in
the jet loop reactor (Fig. 1 C). Here no net ow
of gas or liquid occurs along the column; in-
stead, an internal circulating ow is produced.
One way to achieve this is with a propeller, but
other approaches exist. In the most commonly
used type of loop reactor, the jet loop reactor,
the owis drivenbya high-velocityliquidjet. As
in the downow bubble column, gas is let in at
the top and dispersed by the jet energy. Bubbles
can be distributed throughout the reactor volume
only if the downward liquid ow velocity in the
internal tube is greater than the slip velocity of
the bubbles. Accordingly, a minimum power in-
put is required.
4 Bubble Columns
These three basic methods of dispersing gas
in liquid are generally not used in their pure
forms. The variety of problems in chemical and
biotechnical processes has led to many differ-
ent contacting devices that combine these basic
techniques.
Figure 2. Types of bubble-column reactors
A) Simple bubble column; B) Cascade bubble column with
sieve trays; C) Packed bubble column; D) Multishaft bubble
column; E) Bubble column with static mixers
2. Bubble Columns and
Modications
2.1. Design and Applications
Bubble columns are very adaptable gas liquid
contacting devices; possible designs are shown
in Figure 2. The simplest form of bubble col-
umn (Fig. 2 A) consists of a vertical tube with
no internals. Gas is fed in at the bottom while
liquid is led through the apparatus cocurrently
or countercurrently. This simple form is seldom
used in practice; instead, a number of modi-
cations are employed. The back-mixing of gas
and liquid phases in the simple bubble column
and the nonuniform distribution of gas bubbles
over the cross section can be reduced by the in-
stallation of trays (Fig. 2 B), packings (Fig. 2 C),
or shafts (Fig. 2 D). All these devices can oper-
ate either cocurrently or countercurrently. To set
up the most homogeneous possible bubble ow,
static mixer elements can also be placed in the
ascending ow section (Fig. 2 E).
Figure 3. Hydroformylation of propene
a) Stripping zone; b) Reaction zone
Hydroformylation. The hydroformylation
of propene is carried out in simple bubble
columns. The reaction is homogeneously cat-
alyzed by rhodium complexes. Usually the
propene and the CO/H
2
gas mixture are let in
at the bottom of the reactor. Incompletely re-
acted gas, saturated with the reaction product,
exits the reactor. The hydroformylation product
is separated fromthe gas streamby condensation
and forwarded to downstream processing, while
the gas is recycled to the reactor. Because the
heat of reaction cannot be completely removed
by evaporative cooling using the enthalpy of va-
porization of the product, the bubble column is
also equipped with an external cooling loop.
One great advantage of the process is that the
product is recovered from the reaction mixture
without additional separation operations which
would damage the expensive catalyst system.
The close coupling between the product and the
recycle gas necessary to discharge it (i.e., a cer-
tain quantity of gas is required for product dis-
charge for thermodynamic reasons), however,
presents some problems. First, the gas ow rate
causes a high gas holdup, which reduces the re-
action volume and thus decreases the productiv-
Bubble Columns 5
Figure 4. Oxidation of montan waxes in cascade bubble columns
a) Cascade bubble-column reactors; b) Separators; c) Final purication of wax oxidate; d) Off-gas treatment
ity of the reactor. Second, large bubbles occur,
which limit the delivery of gaseous reactants to
the liquid phase in the reactor. For these reasons,
recycle gas is admitted to the bubble column at
two levels (Fig. 3) [3]. About half of the recy-
cle gas is fed via the bottom sparger to disperse
reactants into the overlying reaction zone. The
remaining recycle gas is let in via the top sparger,
which lies slightly below the liquid surface, to
facilitate separation of the reaction product. Fi-
nally, the CO/H
2
reactant streamis fed at various
levels to supply CO that has been consumed by
the reaction in the liquid phase.
Oxidation of Montan Waxes. Bubble
columns are used in a cascade when a narrow
residence-time distribution is required, for ex-
ample, to prevent or limit undesired consecutive
reactions. Reducing back-mixing (i.e., a narrow
residence-time distribution) is also useful when
reaction-engineering considerations dictate that
the gas must be fed to various points in the reac-
tor or when a liquid reactant must be degraded
to the greatest extent possible.
Montan waxes from brown coal must be
deresinied, oxidatively bleached, and esteried
(optional) [4], [5]. Oxidation of the waxes con-
sists of several consecutive reactions; the rst
three steps (oxidation of resins and dark- colored
substances, saponication of montan waxes, ox-
idation of wax alcohols) are desirable, whereas
the fourth (oxidative degradation of wax acids)
is not. The residence-time distribution in the re-
actor must be controlled so that the desired re-
actions go as far as possible without the unde-
sirable reaction occurring to any marked extent.
Oxidation is performed in four cascaded bub-
ble columns connected in series (Fig. 4). In the
rst bubble column, the crude wax for bleach-
ing is metered in along with half of the required
amount of chromic acid. Air is supplied to en-
hance mixingof the reactants. The spent chromic
acid is separated from the wax downstream of
both the rst and the second bubble columns.
Another 25 % of the total acid required is added
to the second and third columns. The reaction
preferably takes place at 100 125
C and 1 5
bar, with a residence time of 1 3 h for the en-
tire cascade. The enthalpy of reaction is removed
by partial evaporation of the water contained in
the chromic acid. After exiting the fourth bub-
ble column, the oxidizedproduct, spent acid, and
off-gas are separated in two separators.
2.2. Gas Distribution
Usually, the gas is dispersed to create small bub-
bles and distribute themuniformly over the cross
section of the equipment to maximize the in-
tensity of mass transfer. The formation of ne
bubbles is especially desirable in coalescence-
hindered systems and in the homogeneous ow
6 Bubble Columns
regime (Section 2.3). In principle, however, sig-
nicant mass transfer can be obtained at the
gas distributor through a high local energy-
dissipation density [6], [7].
In most cases, gas bubbles are generated by
pores or holes or in the shear zone of a liquid
jet. Figure 5 shows typical forms of static
gas spargers, in which bubble formation occurs
without any additional energy supplied from
outside. The simplest of these devices, the dip
tube (Fig. 5 A), only gives an acceptably uni-
form gas distribution over the cross section at
some distance above the sparger. Perforated
plates (Fig. 5 B) and perforated ring spargers
(Fig. 5 C) are more effective. Both of these re-
quire a certain minimumgas owrate to achieve
uniform distribution and prevent the liquid from
getting into the sparger [810]. Very ne bub-
bles can be generated by the use of porous plates
(Fig. 5 D), but their pores are susceptible to foul-
ing, and this type of sparger is seldom used in
full-scale equipment.
Figure 5. Static gas spargers
A) Dip tube; B) Perforated plate; C) Perforated ring sparger;
D) Porous plate
Dynamic spargers offer an alternative to the
static types. They use the power of a liquid
jet to disperse gas in a zone of high energy-
dissipation rate [1113]. Figure 6 illustrates sev-
eral frequently used dynamic gas spargers. The
simple two-phase jet nozzle alone (Fig. 6 A) or
with momentum-transfer tube (Fig. 6 B) is not
able to simultaneously disperse gas and suck in
the gas stream. This can be achieved, however,
with the ejector jet nozzle (Fig. 6 C), the ejec-
tor (Fig. 6 D), and the Venturi tube (Fig. 6 E). In
nozzle selection the ratio of the gas liquid vol-
umetric ow rates must always be considered.
Commonvalues lie between0.5and2. However,
much higher values can be achieved in special
cases with momentum-transfer tubes [12].
2.3. Flow Regimes
The upward motion of bubbles gives rise to three
distinct ow regimes. The crucial quantity for a
ow regime is the supercial gas velocity. The
homogeneous ow regime is marked by a nar-
row bubble-size distribution, and bubbles are
distributed relatively uniformly over the cross
section of the apparatus. This regime extends
to supercial gas velocities of 0.03 0.08 m/s,
depending on the gas liquid system and gas
sparger type.
The uniform distribution of gas bubbles van-
ishes at higher gas rates, and a highly turbulent
ow structure appears. In this heterogeneous or
churn-turbulent owregime, large bubbles or ag-
glomerates of bubbles formand travel upward at
highvelocity(see Section2.6), mainlyinthe axis
of the column. The circulating ow that results
may be so vigorous that bubbles of a size corre-
sponding to that in the homogeneous regime are
actually transported downward in the zone near
the column wall (see Section 2.4).
In the small-diameter columns often used as
laboratory equipment, slug ow occurs at high
gas ow rates. Large bubbles are stabilized by
the column wall and take on the characteristic
slug shape.
The relationship between supercial gas ve-
locity and reactor diameter is illustrated by the
ow map of Figure 7 [14]. The broad transition
regions are due to the effects of the gas distrib-
utor, the gas liquid system, and the liquid rate.
A knowledge of the ow regime is particularly
important because it strongly affects the produc-
tivity of bubble- column reactors.
Bubble Columns 7
Figure 6. Dynamic gas spargers
Figure 7. Flow regimes in bubble columns
2.4. Fluid Dynamics
Rising gas bubbles entrain liquid in their wakes.
As a rule, this upward ow of liquid is much
greater than the net liquid ow rate. Because of
continuity, regions therefore exist in which the
liquid is predominantly moving downward.
Many theoretical and experimental studies
have described the ow behavior of the liquid
phase [15]. The circulation velocity is given as
a function of supercial gas velocity, column
diameter, gas holdup, bubble diameter and rise
velocity, viscosity of the liquid, and dispersion
height. Published analyses deal with both lami-
nar liquid circulation, which is only of theoreti-
cal interest [1618], and turbulent ow, to which
the following discussion is devoted. For exam-
ple, Miyauchi and coworkers use a force bal-
ance over an annular, axially symmetrical vol-
ume element to obtain the velocity prole shown
in Figure 8, [19]. Calculation of the velocities,
however, requires knowledge of the gas holdup
as a function of radial position.
Models of circulation velocity based on en-
ergy balances, in contrast, assume a cell struc-
ture in the bubble column similar to that shown
inFigure 9[20], [21]. Inslender bubble columns,
both calculations and experimental results show
that the height of the circulation cells h
c
is equal
to the apparatus diameter d
t
[20], [22]. Joshi
and Sharma take into account the energy input
due to gas compression and energy losses by
dissipation in the wakes of the rising bubbles,
as well as liquid transport across the liquid sur-
face (hydraulic pump), thus obtaining a veloc-
ity prole over the cross section. Hills [23] and
Kojima and coworkers [24] have determined ve-
locity proles experimentally in bubble columns
8 Bubble Columns
having diameters of 0.14 and 5.5 m, respectively
(Fig. 10). For the mean circulation velocity
L, c
in bubble columns with additional liquid rate u
L
,
Joshi and Sharma [20] give the expression
v
L,c
= 1.4
3
_
gd
t
_
u
G
G
u
L
1
G
G
v
rG
_
(2.1)
where
G
is the gas holdup (+ for countercur-
rent, for cocurrent). Zehner, using a force
balance, arrives at a similar relation for the mean
circulation velocity [21]:
v
L,c
=
3
_
1
2.5
L
gd
t
u
G
(2.2)
The velocity proles derived from the models
and, in particular, the mean velocities enable cal-
culation of the essential uid-dynamic parame-
ters in bubble columns [20], [21], [25].
Figure 8. Radial distribution of liquid velocity in a bubble
column
2.5. Bubble Size
Analysis of bubble size in bubble columns must
distinguish between bubble-size distribution just
after bubble formation at the sparger and size
distribution further away from the distributor.
Because of breakupandcoalescence of the rising
bubbles, the two distributions can differ signif-
icantly. Since the efciency of bubble columns
depends chiey on bubbles far from the gas dis-
tributor,the following discussion only concerns
these.
Figure 9. Cell structure in bubble columns
Two basic methods photography and probe
techniques exist for determining bubble size;
however, they do not lead to identical results.
Both methods are subject to certain limitations
in view of the marked bubble selection that
may occur (i.e., not all bubble sizes can be de-
tected) [26], [27]. In particular, any measure-
ment method only leads to realistic results if the
ow is homogeneous (i.e., a narrow bubble-size
distribution is found). As yet, no method can
be recommended for the measurement of large
bubbles in the heterogeneous ow regime.
If bubbles are generated in a region of high
turbulence (as with dynamic gas spargers), the
following formula [28] can be used to describe
the Sauter diameter d
bS
(mean bubble diame-
ter, calculated from the volume to surface ratio)
[29], [30].
d
bS
=
2
e
0.4
M
_
L
_
0.6
0.5
G
_
L
_
0.25
(2.3)
This formula is based on Kolmogorovs theory
of isotropic turbulence.
Bubble Columns 9
When static gas spargers are used, the bubble
diameter is only weakly dependent on gas veloc-
ity. Descriptive correlations [3134] are applica-
ble only to the systems and sparger geometries
for which they were obtained; a generally valid
description of bubble size does not yet exist. The
maximum bubble diameter d
b, max
can be used
for purposes of estimation [27], [35]. For low-
viscosity liquids, the maximumbubble diameter
is given by
d
b,max
=
3
_
g
L
(2.4)
where is the surface tension. For the water air
system, d
b, max
=8 mm. Larger bubbles have
a high probability of being unstable and thus
breaking up. The Sauter diameter for real distri-
butions is between 40 and 60 % of the largest
stable bubble diameter. This estimate is not,
however, applicable to the heterogeneous ow
regime due to the binodal bubble-size distribu-
tion in this regime.
Figure 10. Calculated radial proles of liquid velocity in
bubble columns [20]
2.6. Bubble Rise Velocity
In the homogeneous ow regime, bubbles of al-
most uniform size and shape rise in the form of
a swarm distributed uniformly over the column
cross section. When the regime changes, larger
bubbles or agglomerates of bubbles form in ad-
dition to the bubbles already present [36], [37].
These aggregates rise at a markedly higher ve-
locity than the small bubbles. Figure 11 shows
measured velocities for large and small bubbles
[36]. Large bubbles rst appear at a supercial
gas velocity of ca. 0.03 m/s. The formation of
large bubbles, however, depends strongly on the
type of sparger used. With sintered plates, for ex-
ample, larger bubbles do not appear at gas rates
lower than ca. 0.1 m/s. As shown in Figure 11,
large bubbles have a rise velocity that is four or
more times larger than small ones. Thus, most
of the gas transport in the heterogeneous ow
regime is accomplished by large bubbles. In this
regime, the quantity of gas transported by small
bubbles remains constant, whereas the quantity
transported by large bubbles increases linearly
with gas velocity. This relationship applies to co-
alescing and coalescence-hindered gas liquid
systems.
Figure 11. Velocities of rising bubbles for the system
water air
Reactor: d
t
=0.44 m, h
t
=5 m; Gas distributor: perforated
plate (d
h
=3 mm)
2.7. Dispersion of the Liquid Phase
Because of the large-scale circulation ows,
back-mixing occurs in both phases. The result-
ing dispersion owJ
D
is usually governed by an
10 Bubble Columns
equation analogous to Ficks rst lawfor molec-
ular diffusion. For the one-dimensional case of
axial dispersion, which is generally sufcient for
a description, follows
J
D
= D
L
dc
dz
(2.5)
The dispersion coefcient D
L
is essentially a
function of the supercial gas velocity and the
column diameter (e.g., see [38]). Flow direc-
tion or liquid velocity does not show any ef-
fect, provided the supercial liquid velocity re-
mains within the range common in industry
(u
L
<0.03 m/s). The dispersion coefcient can
be estimated fairly accurately on the basis of
uid-dynamic models. For example, Joshi and
Sharma [20] and Zehner [21] give dispersion
coefcients derived from the mean circulation
velocity. Each of these formulas gives a good de-
scription of the experimentally determined dis-
persion coefcients known from the literature.
By way of example, Figure 12 compares experi-
mental results reported by various workers with
the theoretical relation derived by Zehner:
D
L
=
d
t
v
L,c
2
=
d
t
2
3
_
1
2.5
L
gd
t
u
G
(2.6)
The equation emphasizes that D
L
strongly de-
pends on column diameter.
Figure 12. Liquid-phase dispersion coefcient measured by
various authors [39]
2.8. Dispersion of the Gas Phase
Due to the large-scale circulation ow both the
liquid and gas phases are dispersed. Further-
more, the formation of large and small bub-
bles, coalescence, and breakup result in addi-
tional dispersion in the gas phase. Whereas the
gas phase in a bubble column with a smaller
diameter ows with virtually no back-mixing,
large units behave more like stirred tanks. The
gas-phase dispersion coefcient depends more
strongly on gas velocity and column diameter
than does that of the liquid phase. For this rea-
son, the degree of axial gas mixing is especially
relevant for scale-up when the gas phase is ex-
pected to show strong concentration variations.
Many formulas in the literature describe the
dispersion coefcient as a function of different
independent variables. A particularly suitable
formula is [40]:
D
G
= 5 10
4
_
u
G
G
_
3
d
1.5
t
(2.7)
This formula is not, however, dimensionally ho-
mogeneous (D
G
in cm
2
/s, u
G
in cm/s, d
t
in cm),
and the gas holdup must be known. By contrast,
the equation
G
D
G
=0.2d
t
u
G
L
gd
t
u
G
v
3
rG
(2.8)
derived by Zehner and Schuch is dimension-
ally correct [41]. However, more recent mea-
surements [42] have shown that this correla-
tion must be modied in the heterogeneous ow
regime (u
G
5 cm/s) because the proportional-
ity is somewhat different:
G
D
G
(u
G
d
t
)
1.65
2.9. Gas Holdup
Gas holdup is one of the most important operat-
ing parameters because it not only governs phase
fraction and gas-phase residence time but is also
crucial for mass transfer between liquid and gas.
Gas holdup depends chiey on gas owrate, but
also to a great extent on the gas liquid system
involved. Accordingly, many correlations that
Bubble Columns 11
have been published only apply to the systems
investigated.
Gas holdup is dened as the volume of the
gas phase divided by the total volume of the dis-
persion:
G
=
V
G
V
G
+V
L
(2.9)
The relationship between gas holdup and gas ve-
locity is generally described by the proportion-
ality
G
u
n
G
In the homogeneous ow regime, n is close to
unity. When large bubbles are present, the expo-
nent decreases, i.e., the gas holdup increases less
than proportionally to the gas owrate (Fig. 13).
The higher the contribution of large bubbles to
the total gas holdup, the smaller is the exponent
n. In the fully developed heterogeneous ow
regime, n nally takes on values between 0.4
and 0.7, depending on the gas liquid system.
Figure 13. Gas holdup and fraction of large bubbles (sys-
tem: water air; gas distributor: perforated plate d
h
=3 mm)
The effect of low liquid velocities u
L
on gas
content is generally negligible. At high ow
rates the gas holdup decreases in cocurrent sys-
tems because gas bubbles pass through the col-
umn more quickly. In contrast, the gas holdup
rises in countercurrent systems; this can lead to
extremely high gas holdup, especially in down-
ow bubble columns [43].
Above 0.1 m, the reactor diameter is of sec-
ondary importance for gas holdup, as measure-
ments onunits havingdiameters between0.1and
5.50 m show [34], [36], [44], [45].
The effects of physical properties on gas
holdup are exceedingly complex. Increasing the
viscosity of the liquid phase leads to increased
bubble coalescence and thus a decrease in gas
holdup. Above ca. 50 mPa s, however, the gas
holdup remains constant [46]. Although surface
tension is not very important for the gas holdup,
a change in coalescence behavior may have last-
ing effects. When gas-phase residence times are
long and gas distribution is obtained with per-
forated or sintered plates, the presence of salts
or alcohols that counteract coalescence has lit-
tle effect [44]. In contrast, gas holdup increases
markedly in systems sparged by two-phase noz-
zles when coalescence is hindered [47]. Such
behavior can be attributed to a small-bubble gas
holdup higher than that in coalescing systems,
whereas the content of large bubbles is identi-
cal [36]. Small bubbles formed under high shear
stresses in the region near the two-phase nozzle
cannot recombine so the gas holdup increases
signicantly with this type of gas distributor.
The relation of Akita and Yoshida [48] is
suitable for estimating the gas holdup and is
based on the investigation of numerous systems:
G
(1
G
)
4
c
1
=
_
gd
2
t
_
1
8
_
gd
3
t
L
_
1
12
_
u
G
gd
t
_
(2.10)
For pure liquids and nonpolar solutions the con-
stant c
1
is 0.2, for electrolyte solutions it is
0.25. However, reliable results cannot be ex-
pected for systems that have not been investi-
gated in this study. The effects of reactor pres-
sure ongas holduphave not beenfullyexplained.
Although some authors nd no effect between
1 and 16 bar [49], others nd that gas holdup
increases with pressure in systems with small
sparging holes (d
h
1 mm) or with sintered
plates [44], [5052]. Transition from the homo-
geneous to the heterogeneous regime occurs at
higher gas ow rates as pressure increases.
Gas holdup is generally a function of posi-
tion in the bubble column. Axial proles of gas
holdup show a zone near the gas distributor in
whichthe holdupincreases tothe value that char-
acterizes the following equilibrium zone. The
gas holdup at the top of the column, in the zone
12 Bubble Columns
of bubble breakup, is markedly higher than the
equilibrium value [45].
Gas holdup also depends on radial position.
The prole shows gradients only near the wall in
the homogeneous owregime [23], [45], [53]. In
contrast, a parabolic radial gas holdup distribu-
tion appears in the heterogeneous regime [23],
[37], [45], as a consequence of the preferential
rising of large bubbles or agglomerates of bub-
bles in the axis of the column. Figure 14 shows
radial proles of gas holdup at various gas ow
rates in the water air system.
Figure 14. Radial proles of local gas holdup
d
t
=0.45 m; h
t
=6.2 m; h =3.03 m (at measurement point);
perforated plate d
h
=1 mm
2.10. Specic Interfacial Area
The area of the gas liquid interface is one of
the most important process parameters. Espe-
cially at high reaction rates (e.g., when a bubble
column is employed as an absorber), the interfa-
cial area becomes a crucial factor in equipment
sizing. Like gas holdup, interfacial area depends
on the geometry, operating conditions, and gas
liquid system. Gas holdup and interfacial area
per unit volume are related as
a=
A
V
R
=
6
G
d
bS
(2.11)
where V
R
is the volume of the reaction mixture
and d
bS
is the mean bubble diameter (Sauter di-
ameter, Section 2.5). As Figure 15 shows, the
interfacial area increases with increasing gas
ow rate. An exception occurs when a porous-
plate sparger is used; like gas holdup, interfacial
area decreases on transition to the heterogeneous
ow regime and then approaches the same val-
ues observed with perforated plates. The growth
in interfacial area with increasing gas velocity is
always greater in the homogeneous than in the
heterogeneous ow regime. The reason lies in
the formation of large bubbles in the heteroge-
neous regime: the interfacial area of large bub-
bles per unit volume is markedly lower than that
of smaller ones.
Figure 15. Specic interfacial area as a function of super-
cial gas velocity
a) d
t
=0.102 m; b) d
t
=0.29 m; c) d
t
=0.14 m; d) d
t
=0.1 m;
Porous plate; - Perforated plate
The specic interfacial areas attainable in
various gas liquid reactors can be compared on
the basis of power input P per unit volume [29].
Experimental values can be described by the re-
lation
a=k
_
P
V
R
_
m
n
G
(2.12)
The exponent m is between 0.4 and 1 [54]. The
plot in Figure 16 enables a direct comparison
to be made between reactors with respect to the
energy required to produce a given interfacial
area.
Bubble Columns 13
Figure 16. Specic interfacial area as a function of specic
power input [55]
a) Stirred tank; b) Bubble column with porous plate; c) Bub-
ble column; d) Bubble column with two-phase jet nozzle
( jet loop reactor); e) Packed column; f ) Bubble column
with injector nozzle
2.11. Volumetric Mass-Transfer
Coefcient
The mass transfer between the gas and the liq-
uid phase in a bubble column can be described in
most cases by the volumetric mass-transfer coef-
cient k
L
a, which is the liquid-phase masstrans-
fer coefcient k
L
multiplied by the specic in-
terfacial area. Gas-phase resistance can usually
be neglected, so k
L
a gives an adequate descrip-
tion. To determine the mass-transfer rate, how-
ever, the driving concentration difference must
be known which in turn requires a knowledge of
mixing behavior in the gas and the liquid phase.
In industrial units (d
t
>1 m), estimates can be
based on the assumption of complete mixing in
both liquid and gas phases.
Like gas holdup and interfacial area, k
L
a also
depends on the gas ow rate, type of sparger,
and gas liquid system. The mass-transfer coef-
cient and the gas rate are again proportional to
one another:
k
L
au
n
G
where n can be between 0.7 and 0.92 [31], [56
59]. Mass-transfer coefcients two- to threefold
higher can be achieved in the homogeneous ow
regime if a porous plate is used as sparger instead
of a perforated plate (Fig. 17). In the heteroge-
neous regime, however, the effect of the sparger
is negligible.
Figure 17. Mass-transfer coefcients in bubble columns
According to experimental results, the col-
umn diameter above about 15 cm has no ef-
fect on mass-transfer coefcient. Some correla-
tions nonetheless include reactor diameter [31],
[57], [60]. Akita and Joshida [31] state that
the value of the column diameter used for cal-
culation should not be increased beyond 0.6 m.
Based on this premise, their correlation for k
L
a
is
k
L
ad
2
t
D
G,L
=
0.6
_
L
D
G,L
_
0.5
_
gd
2
t
_
0.62
_
gd
3
t
2
L
_
0.31
1.1
G
(2.13)
andhas the best experimental support. The mass-
transfer coefcient increases in coalescence-
hindered systems [54], [61]. This increase de-
pends on the system and the concentration
of coalescence-hindering substance. The maxi-
mumgain in mass-transfer coefcient due to the
presence of electrolytes, however, is only 30 %.
14 Bubble Columns
Figure 18. Examples of heat exchanger inserts
A) Parallel single tubes; B) Tube bundle in cross ow; C) Longitudinal tube bundle; D) Tube spiral or helix; E) Draft tube
with jacket
2.12. Heat Transfer
In many cases, heat must be removed when oper-
ating bubble columns. A particularly simple so-
lution is to utilize the latent heat of vaporization
of the liquid phase for heat removal, although
this is not always feasible. Inaddition, manypos-
sibilities exist for heat transfer through heated or
cooled surfaces, as shown in Figure 18. In this
way, up to ca. 30 m
2
/m
3
of heat-transfer area can
be installed in a bubble column.
The turbulent ow generated by rising gas
bubbles increases heat transfer even at low gas
rates (Fig. 19). The increase in heat-transfer co-
efcient , with gas throughput is markedly
greater in the homogeneous than in the hetero-
geneous regime.
The heat-transfer coefcient does not depend
on the column diameter, type of sparger, or coa-
lescence behavior of the gas liquid system.
Two distinct concepts are used to describe
the heat-transfer coefcient at the wall. Whereas
Kast [62] and Deckwer [63] consider ra-
dial ow and heat transported by it, Joshi and
coworkers [64] and Zehner [21] use circulation
velocities derived by them for physical model-
ing. Here withthe followingrelationfor the heat-
transfer coefcient can be derived [21]:
= 0.18 (1
G
)
_
2
L
L
c
pL
v
2
L,c
l
L
_
1/2
(2.14)
where
l=d
b
_
6
G
_
1/3
and
L, c
is calculated by Equation (2.2). On the
whole, these two approaches correlate well with
literature data. Heat transfer in bubble columns
with heat-exchange internals has not been in-
tensively studied [6569]. For tube bundles ar-
ranged in an axial direction (Fig. 18 C), the heat-
transfer coefcient increases with increasing
tube pitch and decreases when the free cross-
sectional area increases [68], [69]. A similar re-
lationship is found for a tube bundle arranged in
cross ow (Fig. 18 B), but here a marked effect
of liquid throughput occurs [67].
The installation of tube bundles leads to an
overall change in uid dynamics and thus in
mixing behavior. For example, tubes installed
in cross ow hinder ow in the longitudinal di-
rection and thus reduce dispersion in the liquid
phase [70]. In contrast, the arrangement of heat-
transfer surfaces in the ow direction leads to
more intense mixing of the liquid phase by in-
tensifying circulation [71], [72].
2.13. Slurry Bubble Columns
Solid particles are present in bubble columns in
a wide variety of processes; they must be held
Bubble Columns 15
in suspension by the rapid liquid circulation al-
ready discussed. The presence of the solid phase
in a slurry bubble column means that all pro-
cess parameters behave differently, andina more
complicated way, than in a two-phase bubble
column.
Figure 19. Heat-transfer coefcient at reactor wall
d
t
=0.196 m; h
t
=6.20 m; liquid velocity u
L
=1.2 cm/s
The minimum gas velocity necessary to hold
the solids in suspension increases as the con-
centration and density of the particles increase.
The increment depends, however, on the phys-
ical properties of the solid and liquid phases.
Many empirical equations for the critical gas ve-
locity for complete suspension show a marked
increase with increasing single-particle settling
velocity. Equation (2.15) [73] can be used for
design purposes:
u
G,min
v
rS
= 0.801
_
L
_
0.6
_
S
S
_
0.146
_
gd
t
v
rS
_
0.24
_
_
1 + 807
_
g
4
L
3
_
0.578
_
_
(2.15)
where v
rS
is the relative settling velocity of the
particle swarm in the liquid. At low solids con-
centration (<10 wt %) and low settling veloc-
ity of the particles the gas holdup is nearly un-
changed [74]. In contrast, the gas holdup de-
creases at higher settling velocities with increas-
ing solids concentration [7577]. The strength
of this effect differs from one ow regime to an-
other. The decrease is particularly marked when
an increase in solids content leads to a change
from the homogeneous to the heterogeneous
regime. On the other hand, in the heterogeneous
regime the reduction in gas holdup is only slight
with increasing solids content. Yasunishi and
coworkers [78] verify and recommend the gas
holdup relation of Koide and coworkers [79]
G
(1
G
)
4
= (2.16)
0.277
_
u
G
_
0.918
_
g
4
L
3
_
0.252
1 + 4.35
_
S
S
_
0.748
_
L
_
0.881
_
d
t
u
G
L
_
0.168
for a wide range of solids concentrations and liq-
uid properties. For coalescence-hindered aque-
ous electrolyte solutions a coefcient 0.364must
be used instead of 0.277 in Equation (2.16).
The mixing behavior of the liquid and the
gas phase is very similar to that in the two-
phase bubble column. The axial solids distri-
bution can be described by a one-dimensional
sedimentation dispersion model. The solids
dispersion coefcient is generally lower than the
corresponding liquid dispersion coefcient [80].
The difference between the two values increases
rapidly with increasing settling velocity of par-
ticles. For small solid particles (v
rS
<0.01 m/s),
the effect of supercial gas velocity on the axial
solids concentration prole is negligible above
the minimum gas velocity for suspension [81].
The specic interfacial area declines contin-
uously with increasing solids content [76], [82].
This phenomenon can be explained by the for-
mation of larger gas bubbles, due to the pres-
ence of solid particles that lead to the observed
decrease in gas holdup. If, however, very ne
particles are used in aqueous electrolyte solu-
tions (i.e., systems with hindered coalescence),
the interfacial areas produced do not differ from
those in the two-phase system [83].
The effect of solids on the volumetric mass-
transfer coefcient depends largely on particle
properties, solids content, and physical proper-
ties of the liquid. At lowsolids concentration (up
16 Bubble Columns
to ca. 3 5 wt %), the mass-transfer coefcient
matches that of the two-phase bubble column.
A higher solids content, like increasing parti-
cle size, leads to a drop in k
L
a relative to the
solid-free condition [78], [79], [84], [85]. For
very ne particles (d
P
<36 m), however, k
L
a
also decreases with decreasing particle diameter
[86]. Overall, the behavior of k
L
a is governed
by the interfacial area per unit volume because
the change in k
L
is generally small.
The mass-transfer coefcient between liquid
and solid increases roughly as the fourth root of
the gas ow rate, decreases with increasing liq-
uid viscosity and particle diameter, and becomes
partly independent of these factors at high gas
rates. Two distinct models can be used for the
mathematical description, but they lead to simi-
lar values if the solids are completely suspended.
Values reported by various authors are compared
in Figure 20 [87].
Figure 20. Liquid solid mass-transfer coefcient esti-
mated with ve different correlations [87]
L
=10
6
m
2
/s;
L
=1000 kg/m
3
; D
G, L
=10
9
m
2
/s;
-
S
/
L
=1.5;
S
/
L
=0.3
Hydrogenation of Benzene. In the IFP pro-
cess, benzene is hydrogenated to cyclohexane in
a slurry bubble column [88] (Cyclohexane,
Chap. 4.1.). This process was used to produce
1.8 10
6
t of cyclohexane in 23 plants world-
wide in 1991.
Figure 21 shows the slurry bubble column (a)
in which benzene is hydrogenated on suspended
Raney nickel. The hydrogen-rich gas that is let in
at the bottomof the main reactor provides hydro-
gen for the reaction and also strips the product
cyclohexane out of the reactor. Thus, the pro-
cess can be operated without the need for ex-
pensive equipment to separate the product from
the catalyst. Inthe external coolingloop, the high
heat of reaction is removed and suspension of the
solid catalyst is assisted by the circulating liq-
uid stream. Complete conversion of benzene is
accomplished in a xed-bed reactor (b) installed
downstream on the gas side.
Figure 21. Hydrogenation of benzene to cyclohexane
a) Main reactor; b) Secondary reactor; c) Steam
drumh
t
=10 m; d
t
=2.5 m; T =195
G
=
u
G
v
L
v
rG
(3.2)
The relative velocity v
rG
for coalescing aqueous
systems takes on values in the range 0.2 m/s
v
rG
0.3 m/s. The liquid velocity v
L
can be
calculated from the liquid rate u
L
and the gas
holdup
G
:
v
L
=
u
L
1
G
(3.3)
Hence the gas holdup can be expressed as
G
=
u
G
u
L
1
G
v
rG
(3.4)
If gas and liquid rates are given, Equation (3.4)
can be solved for the gas holdup:
G
=
B
2
_
_
+
_
_
2
B
_
2
u
G
v
rG
+11
_
_
where
B=
u
L
+u
G
v
rG
1 (3.5)
Finally, Equation (3.4) can be written so as to
yield the liquid rate:
Bubble Columns 21
u
L
= (1
G
)
_
u
G
G
+v
rG
_
(3.6)
A ow chart can then be derived for the
water air system(v
rG
=0.23 m/s). InFigure 27,
the gas velocity is plotted versus the supercial
liquid velocity, with the gas holdup
G
as a pa-
rameter. For
G
>0.35, ow is in the hetero-
geneous regime. In this regime a highly turbu-
lent two-phase owdevelops, resulting in strong
mixing of liquid and gas. In the extreme case,
gas accumulates at the top and can propagate
throughout the reactor fromthere. Kulkarni re-
ports somewhat different results [111], possibly
because of less uniform gas feeding.
Figure 27. Flow chart for downow bubble columns
Gas velocity is plotted as a function of liquid velocity,
with gas holdup as parameter, for the water air system
(v
rG
=0.23 m/s) calculated with Equation (3.6).
For systems with hindered coalescence, de-
termination of these values and other process
parameters normally requires experimentation
[105].
3.3. Mass Transfer
The following statements are applicable only to
coalescing systems. In the homogeneous ow
regime, all bubbles in the downow bubble col-
umn are almost equal in size. Diameters of
3 mmd
b
4 mm are observed [112]. A slight
decrease in bubble size with increasing liquid
ow rate and rising pressure has been reported
[108].
If the gas holdup is known, the specic inter-
facial area can be estimated as
a=
6
G
d
bS
which gives maximum values up to
a =450 700 m
1
. Figure 28 compares these
values to those measured in simple bubble
columns and packed columns [108] (referred
not to the total volume but to the liquid vol-
ume). This gure illustrates the advantages of
downow bubble columns, which have higher
a
L
values at low gas rates.
A liquid-phase mass-transfer coefcient
k
L
3.7 10
4
m/s has been calculated by
plotting the volumetric mass-transfer coefcient
k
L
a versus the volumetric gas holdup [110] and
the bubble size stated above. Somewhat lower
estimates, up to 3 10
4
m/s, have been re-
ported in [109], [112]. However, the mass-trans-
fer coefcient k
L
cannot be measured directly,
and large uncertainties are to be expected. The
volumetric mass-transfer coefcients k
L
a mea-
sured in bubble columns at equal gas holdups are
roughly the same. Because of the very different
volumetric gas ow rates in upow and down-
ow bubble columns, different gas conversions
are achieved; Figure 29 compares these gures
for sulte oxidation with air [110].
Data on the axial back-mixing of the
gas phase have been reported [111]. At gas
rates of u
G
=0.001 0.01 m/s, the dispersion
coefcient takes on constant high values:
D
G
0.2 m
2
/s (d
t
=0.025 m, u
L
=0.334 m/s,
G
=0.025 0.09).
Back-mixing of the liquid phase under com-
parable conditions is also markedly less than in
normal bubble columns [113].
Examples. Steiner and Herbrechtsmeier
studied the oxidation of sulte solutions with
air in the downow bubble column, nding a
twelvefold higher depletion than with a simple
bubble column (Fig. 29) [110].
A process for the absorption of gases con-
taining nitrogen oxides has been reported [114].
Virtually complete conversion can be obtained
through the use of a downow bubble column
3 m high.
22 Bubble Columns
Figure 28. Comparison of mass-transfer area per unit liquid volume for various sparged devices as a function of gas velocity,
with liquid velocity as parameter, according to [108]
Figure 29. Comparison of oxygen depletion in upow and downow bubble columns for sulte oxidation with air [110]
T =22
C; [SO
2
3
] =0.4 0.8 mol/L; [Co
3+
] =710
6
mol/L; pH=8.0
The degradation of organic contaminants by
ozonolysis is a well-known method of water
treatment. For economic and safety reasons, vir-
tually complete depletion of the ozone is desir-
able, which can be achieved in downowbubble
columns as reactors [115]. No danger of fouling
exists in these devices, and high liquid through-
puts can be handled. Figure 30 is a ow sheet
of the entire process. Only a single theoreti-
cal mass-transfer unit can be realized, but this
drawback can be overcome by the proposed use
of a reactor cascade [115]. The phases are led
through the cascade countercurrently.
4. Jet Loop Reactors
Jet loop reactors are among the most versatile
gas liquid contactors. The momentum of the
liquid jet issuing from the nozzle enhances in-
ternal circulation and opposes demixing of the
phases (distribution method Cin Fig. 1). The liq-
uid jet can be utilized to suck in, compress, and
disperse fresh or recycle gas (Figs. 31, 32, 33,
34). The liquid-jet data are important process
parameters. The liquid volumetric ow rate, ve-
locity, and power can be varied over wide ranges.
Finally, the size of the draft tube and the upper
Bubble Columns 23
Figure 30. Plant with downow bubble column for ozone treatment of water
a) Downow reactor; b) Pressurizing pump; c) Ozone generator; d) Compressor; e) Deozonizer
ow-reversal zone strongly affect uid dynam-
ics and gas separation. Other possible variations
are offered by the nozzle conguration (Figs. 32,
33, 34).
The four examples in Figure 31 illustrate op-
tions for the direction ow pattern phases. In all
cases the gas is incorporated into the liquid via
the nozzle located in the gas space. The liquid
jet entrains gas bubbles until the nozzle orice
is closed by the rising liquid surface. The in-
corporation of more gas submerges the nozzle
and blocks the gas inlet, the surface level then
drops again; this self-regulating mechanism en-
ables the gas holdup to be controlled.
A jet loop reactor (also called a gas-
circulation reactor) [116] without net gas or liq-
uid throughput is shown in Figure 31 A. Both
phases are let in at the top and discharged at the
top (the gas phase is consumed to a higher or
lower degree). This corresponds to dispersion
method C of Figure 1.
If the gas is under pressure, it can also be let
in at the bottom to intensify circulation. The re-
sult is a net gas rate, as in the bubble column
(Fig. 31 B); here, distribution methods A and C
of Figure 1 are combined.
Figure 31 Cshows a combination of sparging
methods B and C in Figure 1. The liquid is fed
at the top and discharged at the bottom. This jet
loop reactor has an additional net ow of liquid,
as in a downow bubble column.
A combination of all three distribution meth-
ods (bubble column, downow bubble column,
and jet loop) is shown in Figure 31 D. The pro-
cess characteristics of one spargingtype will pre-
dominate, depending on the selected gas and liq-
uid ow rates. At high gas ow rates, for exam-
ple, the liquidsurface level rises above the nozzle
orice. The liquid jet then no longer entrains gas
bubbles, serving only to drive the circulation and
disperse the bubbles. This versatile type of dis-
tribution can be further rened through variation
of the nozzle position and the use of self-priming
ejectors (Figs. 32, 33, 34). To evaluate a design,
the essential process parameters must be esti-
mated, which is not always feasible because of
the many possible variations. For the basic forms
shown in Figure 31, however, some information
can be derived from a power balance (Section
4.2).
4.1. Design and Applications
LoopReactors withDownowLiquidJets.
Figure 32 illustrates several loop reactors with
downward-pointing nozzles. These reactors are
even suitable for suspension catalysis. If the
pump or feedstream is cut off or fails, the noz-
zle drains clear, and the danger of the nozzle
being plugged by the suspended catalyst is thus
reduced. Another advantage is the long gas res-
24 Bubble Columns
idence time. From the inlet at the top, the gas
circulates through the loop at least once. More-
over, the devices are designed so that the gas
is internally recycled. This is important for the
complete conversion of gases containing little or
no inerts.
Figure 31. Options for phase routing in the gas-
circulation type of jet loop reactor
A, C) With surface gas sparging; B, D) With pressure sparg-
ing (with gas throughput); C, D) With bottom outlet for
liquid, inlet at top (with liquid throughput)
In the gas- circulation reactor of Figure 32 A
(see also Fig. 31) [116], the gas can also be let
in at the bottom of the reactor, independently
of the nozzle, if the gas is available at reactor
pressure. For a given jet power, this design of-
fers much higher gas holdups and better mass-
transfer performance (Sections 4.5 and 4.6). The
jet only has to supply the recycle gas. Another
marked increase (up to a factor of two) in the gas
holdup is achieved by installing a momentum-
transfer tube in the reactor. This can be sub-
merged (Fig. 33 B) or can extend above the liq-
uid surface (Fig. 33 C).
Figure 32. Types of jet loop reactors with downward liquid
jet
A) Gas-circulation reactor with pressure sparging; B) Gas-
circulation reactor without connected gas space; C) Gas-
circulation reactor with submerged nozzle; D) Compact
reactor with annular space for liquid injection
Bubble Columns 25
To prevent separation of the gas phase, the
nozzle can be built directly into the reactor top
(Fig. 32 B) [117]. Gas bubbles separating in this
zone are immediately entrained by the liquid jet
and redispersed into the circulating ow.
Figure 33. Options for spontaneous gas sparging from top
Surface sparging in gas-circulation reactor without (A) and
with (B) momentum-transfer tube; dip-tube sparging with-
out (C) and with (D) self-priming ejector
The submerged nozzle in Figure 32 C can ei-
ther accept pressurized gas fromoutside the sys-
tem[118] or suck the gas in. Internal gas recycle
is also possible. The suction of the nozzle can
be enhanced by applying a swirl to break up the
liquid jet, provided the liquid nozzle orice is
set back somewhat to the rim of the nozzle. The
ejectors and ejector nozzles discussed in Chap-
ter 2 are suitable for deeper submergence (see
also Fig. 34 B and C).
In the jet loop reactor proposed by R abiger
and coworkers, the liquid is fed in via an annu-
lar nozzle (Fig. 32 D) [119], [120]. Gas can be
drawn in via the center tube (ejector fashion) or
supplied under pressure (injector fashion). For
application of this reactor type in wastewater
treatment see [121], [122].
Figure 34. Design options for loop reactors with upward
liquid jet
A) With expanded head for gas separation; B) With gas re-
cycle via a self-priming ejector; C) With gas recycle via an
ejector jet nozzle and perforated plates; D) With external
recycle tube
26 Bubble Columns
Submerged nozzles supply the gas in the
loop ow. The penetration depth of the jet dif-
fers. The reactor is thus easier to start up and
the circulation ow can be build up in a sim-
pler way. This is particularly important in batch
processes. In principle, submerged nozzles rep-
resent an intermediate stage between surface
sparging (Fig. 32 A and B) and pressurized gas
sparging through nozzles at the bottom of the
reactor (Fig. 34).
In the gas- circulation reactor of Figure 33,
gas spargingtakes place throughthe free surface.
If a momentum-transfer tube is used (Fig. 33 B),
gas enters the circulating owat a greater depth.
This arrangement functions even when the liq-
uid level drops and the momentum-transfer tube
extends into the gas space (Fig. 33 C). In Fig-
ure 33 D, the tube is led outside the reactor, so
that gas can be delivered directly from outside
without any mixing with recycle gas. The last
two types of gas sparging (Fig. 33 C and D) are
also referred to as dip-tube sparging designs.
Jet Loop Reactors with Upow Liquid Jet
(Fig. 34). In the second major variant of the jet
loop reactor, the nozzle points upward. This de-
sign is closely related to the airlift loop reactor
(Section 2.14). The liquid jet mainly produces
smaller gas bubbles so that conversion of the
gas phase can be improved. On the other hand,
the circulation velocity is increased. At least a
portion of the gas is thereby driven through the
reactor faster than in airlift reactors, with the
possible result of lower conversion; this danger
exists particularly in the reactor of Figure 34 A,
which uses a jet nozzle. The widened separation
zone at the top can reduce gas recirculation or
cut it off altogether; the result is improved mix-
ing of the liquid phase because the circulation
velocity increases with increasing gas holdup
in the downcomer. Gas recycle is recommended
whenever part of the gas leaves the column un-
converted; it can be accomplished without a me-
chanical recycle-gas compressor if an ejector or
an ejector jet nozzle is employed (Fig. 34 B and
C). The ejector jet nozzle offers a higher com-
pression efciency than the ejector and produces
larger interfacial areas [12]. From the chemi-
cal reaction engineering standpoint the combi-
nation of a loop reactor with a series of perfo-
rated plates is very interesting (Fig. 34 C). The
lower part (loop reactor) exhibits the residence-
time distribution of a stirred tank. Back-mixing
is suppressed in the second section (perforated
plates). For higher-order reactions, the conver-
sion of both liquid and gaseous components can
be increased in this way. With pure gases, re-
cycle ensures adequate distribution because the
gas ow rate is sufcient for an even gas load of
the perforated plate.
Figure 34 Dshows a reactor (analogous to the
airlift loop of Fig. 22 D) with external recircu-
lation. This device has an operational behavior
comparable to that of reactors with internal cir-
culation. Its advantages include better gas sepa-
ration and simpler heating or cooling facilities.
The accessible heat-transfer area is larger, and
a conventional heat exchanger can be integrated
directly into the loop.
The most important characteristics of jet
loop reactors with upward-pointing nozzles can
be summed up as follows:
1) They are particularly suitable for higher gas
throughputs or when the gas has a high con-
tent of inerts
2) They have larger interfacial areas than bub-
ble columns, especially with system with
hindered coalescence
3) They offer more intensive back-mixing than
bubble columns
4) They are less suitable for suspension cataly-
sis, because the nozzle can become plugged
4.2. Typical Dimensions
In jet loop reactors, height-to-diameter ratios
h
t
/d
t
of 5 to 20 are common. When several
nozzles are used along with internal tubes, ar-
bitrarily small height-to-diameter ratios can be
achieved. Values larger than 20 are also seen in
high-pressure operation and in pilot plant reac-
tors.
The optimal diameter d
i
of the internal tube
is dictated by the direction of ow. If ow in
the tube is downward, tubes with a diameter ra-
tio d
i
/d
t
of 0.2 0.5 are suggested [116], [119].
The narrower internal tubes have the principal
advantages of higher gas holdupandbetter mass-
transfer at low energy dissipation rate (up to
1 kW/m
3
). As jet power increases, the wider
tubes become better in these respects [119],
[123].
Bubble Columns 27
For upward ow, Blenke and coworkers de-
termined the optimal internal tube diameter as
d
i
/d
t
=0.59 [124]; this result applies to single-
and two-phase reactors.
A crucial parameter of the jet loop reactor
is the nozzle diameter d
n
. Common values are
in the range of d
n
/d
t
=0.02 0.1. For a given
jet power, large nozzles are more efcient than
smaller ones for coalescing systems. In systems
with hindered coalescence, in contrast, this re-
lationship can be reversed.
4.3. Energy Balance
In the general case of Figure 31 D (combina-
tion of bubble column, downow bubble col-
umn, and jet loop reactor), a total of ve me-
chanical power terms can be identied [125].
These can be referred to the reaction volume
V
R
=
4
d
2
t
h
R
(4.1)
Three types of power are delivered to the reac-
tor:
1) Jet thrust power e
thrust
per unit volume:
e
thrust
=f
i
L
v
2
n
v
L,i
h
R
_
d
n
d
t
_
2
(4.2)
where f
i
is the fraction of cross-sectional
area of the inner tube and v
n
is the nozzle
velocity
2) Power input per volume due to liquid ow
e
L
(as in a downow bubble column) :
e
L
= g
G
u
L
(4.3)
3) Power input per unit volume due to gas ow
e
G
(as in a bubble column; holds approxi-
mately only for small pressure changes):
e
G
g (1
G
) u
G
(4.4)
This power is transformed to heat by two
mechanisms:
1) Power dissipation per unit volume due to cir-
culation ow:
e
circ
= f
i
v
3
L,i
2h
R
(4.5)
2) Power dissipation per unit volume due to
friction between phases (slip power):
e
slip
= g
G
(1
G
) v
rG
(4.6)
A combination of these terms gives the bal-
ance equation:
e
thrust
+e
G
+e
L
+e
slip
+e
circ
= 0 (4.7)
In general, the quantities v
L, i
(i.e., e
circ
) and
G
(i.e., e
slip
) are unknown, so direct evaluation is
impossible.
4.4. Mixing Behavior and Fluid
Dynamics
The ow processes in jet loop reactors are par-
ticularly crucial for the mixing and residence-
time distribution of both phases. The investiga-
tion of the relationship between mixing time and
liquid circulation time showed that complete ho-
mogenization of the liquid phase requires ca. ten
passes [126].
The residence-time distribution of the liquid
in the jet loop reactor has been investigated both
experimentally and theoretically [127], [128].
Values of dispersion coefcients in the liquid are
presented in [118], [128]. The crucial parameter
for mixing is the internal circulation ow gen-
erated by the liquid let in through the nozzle.
In certain cases, this quantity can be estimated
from the energy balance (Eq. 4.7).
The residence-time distribution of the dis-
persed gas phase can be found in [129]. The
dispersion coefcients are not substantially dif-
ferent from those of the liquid phase.
Fluid Dynamics of Single-Phase Flow. The
simplest formulation is that of single-phase ow.
For e
G
=0, e
slip
=0, e
L
=0, the following holds
for liquid circulation ow
v
L,i
d
t
=
_
2
f
i
v
n
d
n
(4.8)
The drag coefcient of circulation ow can
be found in [118], [130132]. Typical values
for bottom gas feed are =4; for top gas feed,
0.25 2.
28 Bubble Columns
Fluid Dynamics of Two-Phase Flow. In
two-phase ow the momentum balance can be
solved for the circulation ow v
L, i
only if the
gas holdup
G
is known, which is not normally
the case. Only in the gas- circulation reactor
(Fig. 32 A, B) can the gas holdup be denitely
specied within a certain range. Under the as-
sumption of low gas holdup
G
, for example,
the following implicit relation is reported [123]:
(v
n
d
n
)
2
=
_
f
i
2
v
2
L,i
+
L
gh
R
G
(v
rG
u
L
) u
G
v
L,i
_
d
2
t
(4.9)
The resistance coefcients obtainedfor single-
phase owcan be used to a good approximation.
Figure 35 shows a logarithmic plot of Equation
(4.9) for u
L
=0 and u
G
=0. The liquid velocities
in the two-phase regime are always lower than
in single-phase ow. If the velocity goes belowa
minimum value, the ow becomes unstable and
stops, as shown by the nonlinear behavior of the
curves at lowvelocities. Gas sparging then takes
place only in the upper part of the internal tube.
These features are illustrated in Figure 35. The
minimumowvelocities v
L, i, min
are a factor of
v
L,i
d
t
_
2
f
i
v
n
d
n
1
The minimum ow velocity can also be ob-
tained from Equation (4.9):
v
L,i,min
=
3
_
/
L
f
i
gh
R
[
G
(v
rG
u
L
) u
G
] (4.10)
At highgas velocities u
G
or whenthe nozzle is at
the bottom of the reactor (Fig. 34 A), gas holdup
cannot be freely selected. Instead, it adjusts it-
self as a function of uid-dynamic conditions.
At present, ow velocities cannot be calculated
in advance.
Bohners measurements for the jet loop re-
actor are plotted in Figure 36 in the form
v
L,c
d
t
=f (v
n
d
n
,u
G
)
[133], [134]. Two regions can be identied. At
low jet velocities v
n
, the jet loop behaves like
an airlift loop reactor (Section 2.14). The liquid
circulation velocity is almost independent of jet
velocity. Only at larger values does v
L, c
increase
linearly with v
n
as in single-phase ow [133
135]. Investigations on radial velocity proles
in loop reactors are found in [135].
Figure 35. Circulation ow rate in gas-circulation reactor
as a function of jet conditions and gas holdup, calculated
with Equation (4.9)
Assumptions: =2.5; v
rG
=0.23 m/s; Geometric dimen-
sions: d
t
=0.3 m; d
h
=0.015 m; h
t
=2.0 m; f
i
=0.25
Figure 36. Measured circulation ow in pressure-sparged
jet loop reactor as a function of jet conditions, with gas ve-
locity as parameter [133]
d
t
=0.29 m; h
t
=2.0 m; f
i
=0.35
4.5. Gas Holdup
In the gas-circulation reactor (Figs. 31 and 32 A
and B), the gas holdup can be set arbitrarily
Bubble Columns 29
within certain limits. For a given jet velocity v
n
,
the maximum values
G, max
can be calculated
from the power input by the liquid jet
P
n
=
L
2
v
3
n
F
n
where F
n
=
4
d
2
n
(4.11a)
or the jet power per unit volume
e
n
=
P
n
V
R
=
L
2
v
3
n
h
R
_
d
n
d
t
_
2
(4.11b)
The equations reported by Tebel and Zehner
are presented here in simplied form [123].
The maximum specic energy dissipation rate
caused by the slip between the two phases
(Eq. 4.6) is given by
e
slip,max
=
4
3
1.5
_
2
f
i
e
n
d
n
d
t
(4.12)
which can be solved as follows for the maximum
gas holdup:
G,max
=
2
3
1.5
_
2
f
i
2e
n
d
n
d
t
gv
rG
=
2
3
1.5
_
2
f
i
2P
n
d
n
d
t
gv
rG
V
R
(4.13)
Equation (4.13) is compared with measured val-
ues in Figure 37. The lowest gas holdups are ob-
tained in the coalescing systemwater air. Even
tiny amounts of methanol reduce the tendency
to coalesce. The primary gas bubbles generated
by the jet retain approximately the same size
over the whole apparatus because they do not
coalesce. This reduces their slip velocity v
rG
so
that much higher gas holdups and interfacial ar-
eas can be obtained.
Figure 37. Comparison of measured and calculated maxi-
mum gas holdup as a function of specic jet power in gas-
circulation reactor
System: water air with added methanol; measured values
after [123]; calculations with Equation (4.13)
d
t
=0.14 m; h
t
=1.32 m; d
i
=0.055 m; d
n
=4.9 mm
More complex relationships apply if pres-
surized gas is let in at the bottom of the gas-
circulation reactor. The gas holdup then sets it-
self, analogouslytobehavior ina bubble column,
an airlift loop reactor (Sections 2.9 and 2.14),
Figure 38. Gas holdup measured in the pressure-sparged jet loop reactor [133] versus jet conditions at various gas velocities
h
t
=2 m; d
t
=0.29 m; f
i
=0.33
30 Bubble Columns
or the ordinary loop reactor (Fig. 34 A). Typical
measured values for this type of gas sparging are
presented in Figure 38. As the gas rate increases,
gas holdup increases rapidly. The jet velocity, by
contrast, has relatively little effect.
Zehner and Thelen obtained the expres-
sion
e
slip
2e
n
d
n
d
t
=f
_
e
G
2e
n
d
n
d
t
_
(4.14)
for jet loop reactors with pressurized gas sparg-
ing [125]. Figure 39 shows the gas holdup for
different reactor types and sparging types, based
on this relation. To within measurement er-
ror, the same values are obtained for the gas-
circulation reactor with pressurized gas sparging
(Fig. 31 B) and for the jet loop reactor (Fig. 34).
These values are tted well by the correlation
e
slip
2e
n
d
n
d
t
= 1.5
_
e
G
2e
n
d
n
d
t
_
0.8
(4.13a)
which can be solved directly for the gas holdup:
G
= 1.5
_
2P
n
d
n
d
t
gv
rG
V
R
_
0.2
_
u
G
v
rG
_
0.8
(4.13b)
Figure 39. Relationship between slip power
slip
, jet power
n
, and gas compression power
G
for determination of gas
holdup
a) Surface sparging in gas-circulation reactor; b) Transition
between surface and pressure sparging; c) Pressure sparging
in gas-circulation reactor and normal jet loop reactor
4.6. Mass Transfer
The mean bubble diameter in a sparged reactor is
always the result of distribution and coalescence
processes. In the easily coalescing water air
system, the bubble diameter corresponds essen-
tially to the local energy dissipation rate, a quan-
tity that is distributed very unevenly over the vol-
ume in jet loop reactors. Zones of particularly
high energy-dissipation rate include the imme-
diate actionregionof the liquidjet and, toa lesser
degree, the regions of loop ow reversal. This is
the reasonwhythe smallest bubbles are observed
near the jet in the compact reactor (see Fig. 32 D)
[120]. As Figure 40shows, bubble size decreases
from ca. 3 mm to almost 2 mm with increasing
energy dissipation rate P
n
/V
R
. In the other much
larger regions, the bubbles quickly coalesce to
bigger (3 4 mm) units. Jet loop reactors accord-
inglydonot feature smaller air-in-water bubbles,
on average, than bubble columns or downow
bubble columns operated in the homogeneous
ow regime. For a given gas holdup, compara-
ble mass-transfer should therefore be expected.
The relationships among the mean bubble diam-
eter (Sauter diameter) d
bS
, the gas holdup
G
,
the specic interfacial area a, and the volumetric
mass-transfer coefcient k
L
a have been pointed
out in Section 3.3. These considerations lead to
the following guideline values for the water air
system:
Figure 40. Bubble sizes in the compact reactor (system:
water air)
d
t
=0.1 m; h
t
/d
t
=5.9; d
n
=5 mm; d
i
/d
t
=0.6
Usual specic energy dissipation rates:
P
n
/V
R
1 10 kW/m
3
Mean bubble diameter (Sauter diameter):
d
bS
3.5 mm
Maximum gas holdup:
G
6 30 %
Bubble Columns 31
Special interfacial area: a 100 600 m
1
Volumetric mass-transfer coefcient:
k
L
a 0.04 0.2 s
1
If liquid mixtures and ionic or detergent so-
lutions exhibit a noncoalescing behavior, much
higher gas holdups and smaller bubble diameters
(signicantly less than 1 mm) can be achieved.
The volumetric mass-transfer coefcients and
specic interfacial areas behave in a similar way;
in systems with hindered coalescence they may
be a factor of 5 to 10 higher than the water air
values. Acompilation of volumetric mass-trans-
fer coefcients in various types of sparged ap-
paratus can be found in [136].
Figure 41. Effect of solids on gas holdup in the gas-
circulation reactor
System: 1 wt % NaCl air glass spheres; d
p
=0.075
0.15 mm
4.7. Three-Phase Loop Reactor
When solids are suspended in sparged loop re-
actors the same engineering considerations are
necessary as in slurry bubble columns (Section
2.13). Many workers have studied the effect of
solid particles on uiddynamics and mass-trans-
fer performance [76], [137139].
The uid-dynamic principles of solid liquid
systems are comparable to those of gas liquid
systems (Sections 4.3 and 4.4) [132], [140].
These considerations have been extended to
threephase systems so that theoretical models
are available for complex multiphase ow[120],
[123], [141]. In principle, the solid phase is ac-
counted for by a supplemental energy term for
the slip power dissipation of the particle swarm
(Section 4.3). How interactions between phases
inuence individual slip velocities is not clear.
For example, K urten and coworkers examined
the maximum possible gas holdup in a gas-
circulation reactor with gas sparging at the sur-
face [76]. As Figure 41 shows, the solids content
S
has a strong effect that cannot be accounted
for merely by the additional slip power dissipa-
tion of ne particles. Instead, the slurry must be
assumed to yield larger gas bubbles because of
its higher apparent viscosity. The higher slip ve-
locity of these larger bubbles might then explain
the marked dependence on solids concentration.
R abiger [138] and Wachsmann [139] also
studied the effect of solids. At constant energy
dissipation rate and constant volumetric gas ow
rate, the dependences are roughly similar to
those found in [76]. Only for low solids and gas
holdups does a slight increase in gas holdup oc-
cur relative to two-phase systems.
This has been explained by bubble breakup
by large solid particles [142]. The effect should
not occur below the critical Weber number
We=
S
d
S
v
2
b
= 3
However, this statement partially contradicts
R abigers results [138]. Technically, these dis-
crepancies are insignicant. Up to a solids con-
centrationof 10 vol %, nomajor differences exist
in gas holdup and mass-transfer between normal
and three-phase jet loop reactors. Higher volu-
metric particle concentrations do not, however,
normally occur in jet loop reactors.
5. References
1. H. Gerstenberg, Chem.-Ing.-Tech. 47 (1975)
no. 5, 209.
2. M. Bobik, Chem.-Ing.-Tech. 53 (1981) MS
922.
3. BASF, EP 254 180, 1991 (P. Zehner et al.).
4. Hoechst, DE-OS 2 855 263, 1978 (A. Riedel et
al.).
5. R. Steiner, Chem. Eng. Process. 21 (1987)
1 8.
6. W.-D. Deckwer, J. Hallensleben, M. Popovic,
Can. J. Chem. Eng. 58 (1980) 190.
7. H. Buchholz, Dissertation, Universit at
Hannover 1979.
8. K. Ruff, T. Pilhofer, A. Mersmann,
Chem.-Ing.-Tech. 48 (1976) no. 9, 759 764.
9. G. Neubauer, T. Pilhofer, Chem.-Ing.-Tech. 50
(1978) no. 2, 115 116.
32 Bubble Columns
10. Mersmann, A., Ger. Chem. Eng. (Engl.
Transl.) 1 (1978) 1.
11. P. Kr otsch, Fortschr. Verfahrenstech. Abt. D
16 (1978) 389.
12. P. Zehner, Chem.-Ing.-Tech. 47 (1975) no. 5,
209.
13. H. Gerstenberg, Chem.-Ing.-Tech. 51 (1979)
no. 3, 208 216.
14. Y. T. Shah, B. G. Kelkar, S. P. Godbole, W. D.
Deckwer, AIChE J. 28 (1982) 353 379.
15. J. B. Joshi, Y. T. Shah, Chem. Eng. Commun.
11 (1981) 165 199.
16. K. Rietema, S. P. P. Ottengraph, Trans. Inst.
Chem. Eng. 48 (1970) T 54.
17. J. R. Crabtree, J. Bridgwater, Chem. Eng. Sci.
24 (1969) 1755.
18. S. M. Bhavraju, T. W. F. Russel, H. W. Blanch,
AIChE J. 24 (1978) 454.
19. K. Ueyama, T. Miyauchi, AIChE J. 25 (1979)
258 266.
20. J. B. Joshi, M. M. Sharma, Trans. Inst. Chem.
Eng. 57 (1979) 244.
21. P. Zehner, Ist. Chem. Eng. 26 (1986) no. 1,
22 35.
22. A. L ubbert: Mass Transfer with Chemical
Reactions in Multiphase Systems, NATO
Advanced Study Institute, Cesme/Izmir, Aug.
10 11, 1981.
23. J. H. Hills, Trans. Inst. Chem. Eng. 52 (1974)
1 9.
24. E. Kojima et al., J. Chem. Eng. Jpn. 13 (1980)
16.
25. M. Kraume, P. Zehner, German/Japanese
Symposium Bubble Columns, Schwerte 1988,
preprints.
26. R. Buchholz, K. Sch ugerl, Eur. J. Appl.
Microbiol. Biotech. 6 (1979) 301 315.
27. E. Bla, Chem.-Ing.-Tech. 60 (1988) no. 12,
935 947.
28. P. H. Calderbank, Chem. Eng. 45 (1976) CE
209.
29. O. Nagel, B. Hegner, H. K urten,
Chem.-Ing.-Tech. 50 (1978) no. 12, 934 944.
30. H. Unno, I. Inoue, Chem. Eng. Sci. 35 (1980)
1571.
31. K. Akita, F. Yoshida, Ind. Eng. Chem. Proc.
Des. Dev. 13 (1974) 84 91.
32. T. Miyahara, Y. Matsuba, T. Takahashi, Int.
Chem. Eng. 23 (1983) 517 523.
33. T. Miyahara, N. Naga, T. Takahashi, Int.
Chem. Eng. 23 (1983) 524 531.
34. K. Koide et al., Chem. Eng. Jpn. 12 (1979)
98 104.
35. A. Mersmann, Chem.-Ing.-Tech. 61 (1989),
no. 2, 97 104.
36. H. Wezorke, Dissertation, Universit at
Dortmund 1986.
37. R. Beinhauer, Dissertation, TU Berlin 1971.
38. Y. T. Shah, G. J. Stiegel, M. M. Sharma, AIChE
J. 24 (1978) 369 400.
39. M. Kraume, P. Zehner, Chem.-Ing.-Tech. 61
(1989) no. 4, 332 333.
40. K.-H. Mangartz, T. Pilhofer, VT,
Verfahrenstech. 14 (1980) no. 1, 40 44.
41. P. Zehner, G. Schuch, Ger. Chem. Eng. (Engl.
Transl.) 8 (1985) 282 289.
42. G. Blasey, Dissertation, Universit at Dortmund
1987.
43. P. Herbrechtsmeier, R. Steiner,
Chem.-Ing.-Tech. 52 (1980) 468.
44. W. Hikita et al., Chem. Eng. J. 20 (1980)
59 67.
45. R. Nottenk amper, A. Steiff, P. M. Weinspach,
Ger. Chem. Eng. 6 (1983) 147 155.
46. H. F. Bach, T. Pilhofer, Ger. Chem. Eng. (Engl.
Transl.) 1 (1978) 270. H. Buchholz,
R. Buchholz, H. Niebesch utz, K. Sch ugert, Eur.
J. App. Microbiol. Biotechnol. 6 (1978) 115.
47. K. Sch ugerl, J. L ucke, U. Oels, Adv. Biochem.
Eng. 7 (1977) 1. G. Keitel, U. Onken,
Chem.-Ing.-Tech. 54 (1982) no. 3, 262 263.
48. K. Akida, F. Yoshida, Ind. Eng. Chem. Proc.
Des. Dev. 13 (1974) 84 91.
49. H. K olbel, E. Borchers, H. Langemann,
Chem.-Ing.-Tech. 33 (1961) no. 10, 668 675.
W.-D. Deckwer, Y. Luisi, A. Zaidi, M. Ralek,
Ind. Eng. Chem. Proc. Des. Dev. 10 (1980)
699.
50. M. H. Oyevaar, K. R. Westerterp, Chem. Eng.
Process. 25 (1989) 85 98.
51. M. H. Oyevaar, R. Bos, K. R. Westerterp,
Chem. Eng. Sci. 46 (1991) 1217 1231.
52. P. M. Wilkinson, L. v. Dierendonck, Chem.
Eng. Sci. 45 (1990) 2309 2315.
53. A. Serizawa, I. Kataoka, I. Michiyoshi, Int. J.
Multiphase Flow 2 (1975) 235 246.
54. W.-D. Deckwer, Reaktionstechnik in
Blasens aulen, Salle und Sauerl ander, Aarau
1985.
55. U. Oels, J. L ucke, R. Buchholz, K. Sch ugerl,
Ger. Chem. Eng. (Engl. Transl.) 1 (1978) 115.
56. W.-D. Deckwer, R. Burchhart, G. Zoll, Chem.
Eng. Sci. 29 (1974) 2177.
57. Y. Kawase, B. Halard, M. Moo-Young, Chem.
Eng. Sci. 42 (1987) 1609 1617.
58. H. Hikita et al., Chem. Eng. J. 22 (1981)
61 69.
59. J. S. Cho, N. Wakao, J. Chem. Eng. Jpn. 21
(1980) 576 581.
Bubble Columns 33
60. S. P. Godbole, A. Schumpe, Y. T. Shah, N. L.
Carr, AIChE J. 30 (1984) 213 220.
61. K. Akita, Int. Chem. Eng. 29 (1989) 127 135.
62. W. Kast, Int. J. Heat Mass Transfer 5 (1962)
329 336.
63. W.-D. Deckwer, Chem. Eng. Sci. 35 (1980)
1341 1346.
64. J. B. Joshi et al., Chem. Eng. Commun. 6
(1980) 257 271.
65. A. N. Khoze et al., J. Eng. Phys. 20 (1971)
759 763.
66. L. S. Akselrod, N. I. Vorotnikova, A. A.
Kozlov, Heat Transfer-Sov. Res. 8 (1976)
25 33.
67. R. Wendt, Dissertation, Universit at Dortmund
1983.
68. H.-J. Korte, Dissertation, Universit at
Dortmund 1987.
69. H.-J. Korte, A. Steiff, P.-M. Weinspach,
German/Japanese Symposium Bubble
Columns, Schwerte 1988, preprints.
70. R. Wendt, A. Steiff, P.-M. Weinspach,
Chem.-Ing.-Tech. 55 (1983) no. 10, 796 797.
71. Y. T. Shah, C. A. Ratway, H. G. McIlvried,
Trans. Inst. Chem. Eng. 56 (1978) 107 112.
72. K. Bernemann, Dissertation, Universit at
Dortmund 1989.
73. K. Koide, T. Yasuda, S. Iwamoto, E. Fukuda,
J. Chem. Eng. Jpn. 16 (1983) 7 12.
74. S. Kara et al., Ind. Eng. Chem. Proc. Des. Dev.
21 (1982) 584 594.
75. Y. Kato et al., Ist. Chem. Eng. 13 (1979)
562 567.
76. H. K urten, P. Zehner, Ger. Chem. Eng. (Engl.
Transl.) 2 (1979) 220 227.
77. T. Sauer, Dissertation, Universit at Paderborn
1986.
78. A. Yasunishi, M. Fukuma, K. Muroyama, J.
Chem. Eng. Jpn. 19 (1986) 444 449.
79. K. Koide, A. Takazawa, M. Komura, H.
Matsunaga, J. Chem. Eng. Jpn. 17 (1984)
459 466.
80. T. Matsumoto, N. Hidaka, S. Morooka, AIChE
J. 35 (1989) 1701 1709.
81. H. M urtz, Dissertation, Universit at Dortmund
1990.
82. M. Fukuma, K. Muroyama, A. Yasunishi, J.
Chem. Eng. Jpn. 20 (1987) 28.
83. E. Sada, H. Kumazawa, C.-H. Lee, H.
Nakamura, Ind. Eng. Chem. Proc. Des. Dev.
26 (1987) 112.
84. K. Nguen-Tien, H. Patwari, A. Schumpe,
W.-D. Deckwer, AIChE J. 31 (1985) 194.
85. H.-G. Springmann, A. Steiff, P.-M. Weinspach,
Chem.-Ing.-Tech. 63 (1991) 162 163.
86. J. Kratochvil, K. Winkler, J. Zahradnik: Coll.
Czech. Chem. Commun. 50 (1985) 48 60.
87. K. Muroyama, German/Japanese Symposium
Bubble Columns, Schwerte 1988, preprints.
88. Cyclohexane, Hydrocarbon Processing
1977, no. 11, 144.
89. U. Onken, P. Weiland, Adv. Biotechnol.
Processes 1 (1983) 67 95.
90. M. Y. Chisti, Airlift Bioreactors, Elsevier
Applied Science, London 1989.
91. P. Verlaan, J. C. Vos, K. vant Riet: J.
Chem. Technl. Biotechnol. 45 (1989)
109 121.
92. R. A. Bello, C. W. Robinson, M. Moo-Young,
Can. J. Chem. Eng. 40 (1984) 573 577.
93. P. Weiland, U. Onken, Chem.-Ing.-Tech. 52
(1980) 986 987.
94. P. Weiland, Ger. Chem. Eng. (Engl. Transl.) 7
(1984) 374 385.
95. Y. Kawase, M. Moo-Young, Chem. Eng.
Commun. 40 (1986) 67 83.
96. T. Reith, Ph. D. Thesis, Universit at Delft 1968.
97. W. Hirner, H. Blenke, VT, Verfahrenstech. 11
(1977) no. 5, 297 303.
98. K. Koide, H. Sato, S. Iwamoto, J. Chem. Eng.
Jpn. 16 (1983) 407 413.
99. K. Koide et al., J. Chem. Eng. Jpn. 16 (1983)
413 419.
100. K. Nakao et al., German/Japanese Symposium
Bubble Columns, Schwerte 1988, preprints.
101. K. Sch ugerl, Bioreaktionstechnik, vol. 2, Salle
und Sauerl ander, Aarau 1991.
102. P. K. Ouyoung, M. Y. Chisti, M. Moo-Young,
Chem. Eng. Res. Des. 67 (1989) 451 456.
103. G. M uller, G. Sell, DECHEMA-Monogr. 86
(1980) 589.
104. S. J. Heijnen et al., Chem. Eng. Technol. 13
(1990) 202 208.
105. Y. Bando et al., J. Chem. Eng. Jpn. 21 (1988)
no. 6, 607 612.
106. A. Ohkawa, Y. Shiokawa, N. Sakai, J. Chem.
Eng. Jpn. 18 (1985) no. 6, 172 174.
107. A. Ohkawa, Y. Shiokawa, N. Sakai, N. Imai, J.
Chem. Eng. Jpn. 18 (1985) no. 6, 466 469.
108. P. Herbrechtsmeier, H. Sch afer, R. Steiner,
Chem.-Ing.-Tech. 56 (1984) MS 1221.
109. P. Herbrechtsmeier, H. Sch afer, R. Steiner,
Chem.-Ing.-Tech. 53 (1981) MS 884.
110. P. Herbrechtsmeier, R. Steiner,
Chem.-Ing.-Tech. 50 (1978) no. 12, 944 947.
111. A. Kulkarni, Y. T. Shah, Chem. Eng. Commun.
28 (1984) 311 326.
112. A. Ohkawa et al., J. Chem. Eng. Jpn. 20
(1987) no. 1, 99 101.
34 Bubble Columns
113. W. J. Albrecht, G. Schulze-Pillot,
Chem.-Ing.-Tech. 50 (1987) no. 22.
114. R. Steiner, Chem. Eng. Process. 21 (1987)
1 8.
115. R. Steiner, DECHEMA-Monogr. 94 (1983)
164 197.
116. H. K urten, P. Zehner, Annual Conference of
Chemical Engineering, Straburg, Oct. 1980.
117. BASF, EP 0 087 670, 1983 (H. H. Schuster, H.
Dreher, J. Hambrecht).
118. H. Blenke in: Loop Reactors, Advances in
Biochemical Engineering 13, Springer-Verlag,
Berlin 1979.
119. N. R abiger, A. Vogelpohl, Chem.-Ing.-Tech.
53 (1983) MS 1111.
120. N. R abiger, Habilitationsschrift, Technische
Universit at, Clausthal 1988; Praxiswissen
thermische Verfahrenstechnik, Verlag T
UV
Rheinland, K oln 1988.
121. H. Loosen, ZFL 40 (1989) no. 9, 494 496.
122. B. Lohrengel, A. Vogelpohl, Chem.-Ing.-Tech.
62 (1990) MS 1856.
123. K. H. Tebel, P. Zehner, Chem. Eng. Technol.
12 (1989) 274 280.
124. H. Blenke, K. Bohner, W. Hirner,
Verfahrenstechnik (Mainz) 3 (1969) no. 10,
444 452.
125. P. Zehner, H.-G. Thelen:
Mehrphasenstr omung, Vortrag auf dem
GVC-Fachausschu, Bremen 1991.
126. A. Jaramillo, I. Adler, A. Fiechter, Third
European Congress on Biotechnology,
Munich, Sept. 10 14, 1984.
127. J. Lehnert, Verfahrenstechnik (Mainz) 6
(1972) 58.
128. H.-J. Warnecke, M. Weidenbach, J. Pr uss, H.
Langemann, Chem.-Ing.-Tech. 59 (1987)
no. 6, 496 499.
129. H.-J. Warnecke, D. Vaupel, I. Pr uss, H.
Langemann, Chem.-Ing.-Tech. 61 (1989
no. 10, 815 817.
130. H. Blenke, K. Bohner, S. Schuster,
Chem.-Ing.-Tech. 37 (1965) no. 3, 289 294.
131. H. Blenke, K. Bohner, W. Pfeifer,
Chem.-Ing.-Tech. 43 (1971) no. 1 2, 10 17.
132. P. Zehner, Habilitationsschrift, Universit at
Kaiserslautern 1988; Fortschr.-Ber. VDI Z.
Reihe 3, 1988, no. 160.
133. K. Bohner, Dissertation, Universit at Stuttgart
1971.
134. K. Bohner, H. Blenke, Verfahrenstechnik
(Mainz) 6 (1972) no. 2, 50 57.
135. D. K. Jain et al., Can. J. Chem. Eng. 68 (1990)
1047 1051.
136. H.-J. Warnecke, M. Geisend orfer, D.
Ch. Hempel, Chem. Eng. Technol. 11 (1988)
306 311.
137. L.-S. Fan, S.-J. Hwang, A. Masuura, Chem.
Eng. Sci. 39 (1984) no. 12, 1677 1688.
138. N. R abiger, Chem.-Ing.-Tech. 57 (1985) no. 3,
248 249.
139. U. Wachsmann, N. R abiger, A. Vogelpohl,
VDI-Ber. 607 (1986) 1537 1566. U.
Wachsmann, N. R abiger, A. Vogelpohl, Ger.
Chem. Eng. (Engl. Transl.) 8 (1985) 411 418.
140. P. Zehner, Chem.-Ing.-Tech. 62 (1980) MS
850.
141. K. H. Tebel, P. Zehner, Annual Meeting of the
Process Engineers, Straburg, Sept 17 19,
1986.
142. J. C. Lee, 4th Proc. European Symp., Chem.
React. Eng., 1971, p. 27.
Bubble Memories Information Storage Materials
Bufng Abrasives
Building Bricks Construction Ceramics