EPA-650/2-73-047
December 1973
Environmental Protection Technology Series
'
-
:
I
-------
EPA-650/2-73-047
ABSORPTION OF S02
INTO LIME SLURRIES:
ABSORPTION RATES AND KINETICS
by
J. Edward Vivian
Massachusetts Institute of Technology
Department of Chemical Engineering
Cambridge, Massachusetts 02139
Contract No. 68-02-0018 (Task 4)
ROAP No. 21ADE-21
Program Element No. 1AB013
EPA Project Officer: Charles J. Chatlynne
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
December 1973
-------
This report has been reviewed by the Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency, nor does
mention of trade names or commercial products constitute endorsement
or recommendation for use.
ii
-------
iii
ABSTRACT
The absorption of sulfur dioxide from sulfur dioxide-nitrogen gas
mixture into water and lime solutions and slurries was studied in a short
wetted wall column at approximately one atmosphere and 25 deg. C. The
mole per cent of sulfur dioxide in the gas was varied from 0.03 to 0.37.
In solutions of calcium hydroxide greater than 0.01 g.mole/liter and in
slurries the system appeared to be gas absorption accompanied by an
"infinitely" rapid irreversible second-order reaction in the liquid phase.
The only effect of the solid in the slurries was to increase the absorption
capacity. In water the system appeared to be gas absorption accompanied
by a fast reversible hydrolysis reaction in the liquid phase. The major
resistance to gas absorption at these concentration levels was found to
be in the gas phase and at the lowest concentrations studied the system
was gas phase controlled.
-------
iv
FOREWARD
The work reported here was carried out in accordance with Task Order
No. 4 under HEW Contract CPA-68-02-0018« The principal investigator was
J. Edward Vivian, Professor of Chemical Engineering. Research assistants
who carried out the laboratory work were W. D. Franklin, R. N. Hindy and
H. F. Jackson.
-------
TABLE OF CONTENTS
Page
SUMMARY 1
INTRODUCTION 3
THEORETICAL CONSIDERATIONS 5
APPARATUS and PROCEDURE 14
RESULTS and DISCUSSION 20
CONCLUSIONS and RECOMMENDATIONS 31
APPENDICES 33
A. DETAILS OF PROCEDURE 34
B. SAMPLE CALCULATIONS 38
C. SUMMARIZED DATA 50
D. NOMENCLATURE 55
E. LITERATURE CITED 57
-------
SUMMARY
Restrictions on the emission to the atmosphere of sulfur dioxide have
generated the need for effective methods for its removal from various waste
gas streams such as stack gas from power plants burning sulfur-bearing fuels.
One of the techniques under investigation is the absorption of the sulfur
dioxide into lime slurries in water. This absorption system involves gas
absorption accompanied by simultaneous chemical reactions in the liquid
phase, and because of the low solubility of calcium hydroxide in water
slurries are used in order to provide sufficient absorption capacity. This
study was concerned with the effect of the reactions in the liquid phase upon
the rate of absorption of sulfur dioxide by the slurries.
The experimental data were obtained with a short wetted wall column in
which sulfur dioxide in nitrogen flowed counter to a falling film of liquid.
Rates of absorption of sulfur dioxide into calcium hydroxide solutions and
slurries and into water were measured at approximately 25 deg. C and one
atmosphere total pressure. The concentration of sulfur dioxide in the gas
was varied from 0.03 to 0.37 mole percent. Liquid phase absorption coeffi-
cients were obtained by deducting the gas phase resistance from the overall
resistance derived from the experimental data. Comparison of these liquid
phase coefficients with data on liquid phase controlled physical absorption
permitted the calculation of the enhancement factor for the effect of liquid
phase reactions on the absorption rate.
At the concentration levels used in this study the major part of the
resistance to absorption both into slurries and into water was in the gas
phase, and at the lowest concentrations the process was entirely gas phase
controlled.
The absorption of sulfur dioxide into calcium hydroxide solution and
-------
slurries appeared to be a system of gas absorption accompanied by an "in-
finitely" rapid second-order irreversible chemical reaction involving
molecular (unhydrolysed) sulfur dioxide and hydroxide in solution. The rate
of solution of solid calcium hydroxide did not appear to influence the
absorption process at the interface. The effect of the solid in suspension
was to increase the capacity of the liquid phase to absorb sulfur dioxide.
This mechanism indicates the need in continuous contacting to provide hold-
up to allow for the dissolution of the solid in the slurry to maintain the
concentration of hydroxide near saturation.
The absorption of sulfur dioxide into water is a system of gas absorp-
tion accompanied by a fast reversible hydrolysis reaction. The reaction
is sufficiently rapid to maintain hydrolysis equilibrium in the liquid as
sulfur dioxide diffuses away from the interface. The enhancement factor
for this case becomes very large at low concentrations of sulfur dioxide
in the gas, and consequently, the system becomes gas phase controlled at
these concentration levels. At calcium hydroxide concentrations greater
than about 0.01 g.mole/liter the rate of absorption of sulfur dioxide into
these hydroxide solutions was somewhat greater than that observed for the
absorption into water under similar conditions.
Experimental work on the absorption of sulfur dioxide into slurries in
a continuous absorber such as a long wetted wall column or a packed bed of
spheres in which the liquid holdup undergoes repeated exposure to the gas
followed by bulk mixing is needed to supplement these short wetted wall data
to show the effect of the rate of dissolution of the solid on the absorption
process.
-------
INTRODUCTION
The absorption of sulfur dioxide into alkaline media is an important
process which is receiving considerable attention as a prime means of con-
trolling sulfur dioxide emission in stack gas from the combustion of sulfur-
bearing fossil fuels. The process has been widely used in the process
industries and in particular in chemical pulping, and the technology to
meet the requirements of these industries is available. However, the scale
of operations and particularly the concentration ranges of sulfur dioxide
in feed gas and exhaust gas encountered in emission control problems impose
new and more stringent requirements on this technology. For example, process
exhaust gas may be the feed gas for emission control processes, and these
processes are required to reduce sulfur dioxide levels to a few hundred
parts per million.
The absorption of sulfur dioxide from gaseous mixture into alkaline
media is a process involving gas absorption accompanied by simultaneous
chemical reactions in the liquid phase. Depending on the relative rates of
diffusion and reaction one or the other phenomenon may be the controlling
factor. An understanding of the interaction of diffusion and reaction is
necessary for design of equipment and also for operation of such processes
since operating conditions may affect the relative rates significantly.
Fundamental kinetic data are often not available, and furthermore since
such reactions fall in the category of "fast reactions" fundamental kinetic
data are difficult to obtain independently without the complicating factors
involved in the combined diffusion and reaction processes. Consequently
these kinetic data are generally determined in the laboratory by comparing
absorption rates with and without reaction under carefully controlled fluid
dynamic conditions. Laminar jets and short wetted wall columns have been
-------
useful for this work.
Several alkaline reactants can be used to absorb sulfur dioxide. Sodium
hydroxide, sodium carbonate, sodium sulfite and magnesium hydroxide are
effective reactants of commercial importance for emission control, providing
the process includes a chemical regeneration system. The only reactants for
sulfur dioxide absorption which appear to be economically justifiable in a
once-through process are calcium hydroxide and calcium carbonate. Since
plant investment and technological development will be considerably less for
chemical processing which does not include recovery operations, substantial
emphasis is being directed to the development of lime/limestone absorbers
for first generation plants for sulfur dioxide emission control. In view
of the low solubility in water of these reactants aqueous slurries must be
used to provide adequate absorption capacity. Thus in developing an absorp-
tion model for absorption of sulfur dioxide into these slurries not only is
consideration of diffusion and reaction in the liquid phase important but
also the rate of dissolution of the solid reactant.
In this investigation an attempt was made to determine the effect of
the liquid phase reactions of sulfur dioxide with lime slurries upon the rate
of absorption of sulfur dioxide under known fluid dynamic and diffusion con-
ditions with a view to determining the rate limiting steps. A short wetted
wall column was used and the laboratory work was limited to reagent chemicals
and to concentrations of sulfur dioxide in the gas phase of interest in the
emission control program.
-------
THEORETICAL CONSIDERATIONS
Sulfur dioxide dissolves in water and undergoes hydrolysis to form a
solution of molecular unhydrolysed SO. and hydrogen, bisulfite and sulfite
ions. Sulfurous acid (H SO.) is believed not to exist in significant amount.
Equilibrium exists between the various species in solution and between the
unhydrolysed molecular SO and the partial pressure of SO in the gas phase.
The equilibrium between SO. in the gas and the unhydrolysed molecular SO.
follows Henry's law and can be written in the form:
CA - HP
where C is the unhydrolysed molecular SO- concentration, g.mole/liter, and
p is the partial pressure in atmospheres in the gas phase. The equilibrium
established in the liquid phase for the hydrolysis of the dissolved SO. to
form bisulfite can be written in the form:
[HSO "][H+]
The dissociation of HSO ~ is very small unless the hydrogen ion
concentration is suppressed by the presence of alkali. Consequently the
total concentration C of sulfur dioxide in solution can be represented by
the sum of the concentrations of the unhydrolysed molecular SO., C , and
t+ A
the bisulfite ions, C_:
E
CT - CA*CE
Combining this with the expression for Henry's law to substitute for the
molecular SO concentration and the expression for the dissociation con-
stant to substitute for the bisulfite concentration and noting that the
concentration of hydrogen and bisulfite ion concentrations are essentially
equal an expression for the gas-liquid equilibrium for sulfur dioxide in
-------
water is obtained:
CT - Hp +
The importance of this expression lies in the fact that it permits the
evaluation of the Henry's law constant. This constant is required to
predict the interfacial concentration of unhydrolysed molecular S0_ in
an absorption process.
Several investigators have studied the equilibrium solubility of
sulfur dioxide in water. The work of Johnston and Leppla (2) is repre-
sentative of these studies and has been used in this work to derive
the Henry's law constant. The data are shown in Figure 1 where C
has been plotted against ^p so that the slope of the isotherm is equal
to the Henry's law constant, H, and the intercept is equal to /K H.
The constants are given as a function of temperature in Figure 2. Equa-
tions for these functions are:
In H » 317T5'2 - 10.4523
In K - 2-3%2'6 - 11.929
where H is given in g.moles/liter-atm, K in g.moles/liter and T in degrees
Kelvin. It is to be noted that the values of K derived from these gas-
liquid equilibrium data agree closely with values reported in the literature.
The model assumed in this investigation for the absorption of sulfur
dioxide into water is based on diffusion of SO from the bulk of the gas
phase at a partial pressure p across a gas phase resistance to the gas-
liquid interface where the partial pressure p. establishes an equilibrium
concentration CA. of unhydrolysed molecular SO. in the liquid interface.
Molecular SO is removed from the interface by diffusion as molecular SO..
and by reaction with water. Depending on the relative rates of diffusion
and reaction, liquid phase equilibrium between the species in solution at
-------
7
- I
-
u~l:. _i
;
*
I/ I ";':^
/
>
-------
£
i
!_-
| ----- !
"t
_i i i
3*1
3-2
3-3
1000 / T
37
3-3 3-
1WO/T frft. A)"
-------
any given distance from the interface may or may not be maintained.
Eigen, Kustin and Maass (1_) report data on the hydrolysis reaction:
S02 + H20 t H+ + HS03-
which indicate the forward rate as written to have a pseudo-first order
rate constant in excess of 10 sec . Relative to ordinary diffusion
rates this is a very fast reaction, and it is expected that the various
species will be approximately in equilibrium during diffusion away froin
the interface.
In cases involving absorption with simultaneous chemical reaction it
has been convenient to retain the usual forms of the absorption rate
equation and to allow for the effects of the associated chemical reactions
by an enhancement factor. Thus the rate equation becomes:
2
where N is the solute flux (g.moles/sec-cm ) and fe * is the physical
A I»
absorption coefficient dependent on physical properties and fluid dynamics
of the system. The driving-force (CAi~ Cft) is the difference between the-
concentration at the interface and in the bulk liquid of the specie enter-
ing the liquid phase, in this case unhydrolysed molecular SO , and is
the enhancement factor dependent on the relative rates of diffusion and
reaction. For a limited number of absorption models can be estimated
from properties of the system.
Applying the concept of equilibrium among all species during the
diffusion of SO_ away from the interface in the absorption of SO. into
water the enhancement factor was found to be:
where D and D are the diffusivities of molecular SO. and bisulfite ion
A £ 2
-------
10
respectively, and K. is the equilibrium constant for the reaction of SO.
with water. The model used here to derive this expression for was based
on film theory. The corresponding expression for penetration theory was
not available, but the two theories are not expected to differ signifi-
cantly in this case.
To model the absorption of SO into lime slurry the reaction of SO-
A A
with hydroxyl ion and the dissolution of the lime particles must be
included. At very low concentrations hydroxyl ions will have negligible
effect on increasing the rate of absorption over the rate into water.
However, as the hydroxyl ion concentration is increased eventually the
rate of absorption will be governed by the reaction of SO with hydroxyl
ion to produce bisulfite and sulfite ions. These reactions are expected
to be very rapid. For the case where there is no sulfite in the initial
absorbent liquid, hydroxide will diffuse from the bulk toward the inter-
face as SO. in one or more forms diffuses away from the interface. This
is a situation typical of a fast irreversible (homogeneous) second-order
reaction in the liquid phase. The overall effect is:
S02(aq) + Ca(OH)2 + CaS03 + H2O
The fact that CaSO, has very limited solubility and will precipitate will
have only a secondary effect on the rate at which SO.(aq) is removed from
the interface. The presence of solid Ca(OH). in the slurry however may
affect the rate depending on its rate of dissolution.
If the rate of dissolution is relatively slow, the absorption process
during a single exposure will be similar to the infinitely rapid irrever-
sible reaction case involving reacting species in solution. The
-------
11
where C_ is the saturated calcium hydroxide concentration in the bulk
15
liquid.
On the other hand if the rate of dissolution of the calcium hydroxide
particles is sufficiently rapid relative to the rate at which hydroxide
in solution is consumed by the absorbed SO. so that a concentration of
hydroxide close to saturation is maintained, the reaction accompanying
the absorption will appear to be pseudo-first order. The <|>-values will
lie along the 45 degree line on the Peacemen-van Krevelen plot, or in
other words the fe * product will be a constant independent of flow rate
and equal to /fe D C where feds the second-orderireaofcion rate constant.
Thus experiments measuring the rate of absorption of SO. into lime
slurries at different flow rates (which varies fe *) should give an indic-
ii
ation of whether the rate of dissolution of the solid is rapid enough to
increase the overall rate of absorption above that expected for absorp-
tion into the homogeneous liquid phase or whether the advantage in the
presence of the solid is only to increase the capacity of the liquid
phase to dissolve and react with SO.. In the latter case, holding time
between surface exposures ,to complete reaction and saturate the liquid
will be required to use the slurry reactant effectively.
Since sulfur dioxide has appreciable solubility in water and under-
goes rapid reactions in the liquid phase, there will be significant gas
phase resistance to absorption. To reduce the data on SO. absorption to
obtain data on liquid phase phenomena an estimate of the gas phase resis-
tance must be made and deducted from the overall resistance. The short
wetted wall column was used to study gas phase controlled absorption by
Behrmann (8) and by Schoenburg (9). Liquid flow rate had negligible
effect on the gas phase coefficient, and the results for the gas phase
were correlated by reporting the Sherwood number, N_. , as a function of
-------
12
the Schmidt number, N , and Reynolds number, N as shown in Figure 3.
These data were used to estimate the gas phase resistance in this work.
Additivity of resistances as demonstrated by Mayr (2/4) was assumed in
computing the liquid phase resistance. Thus the additivity equation
appropriate for this work can be written:
(KC'G
where (K )_ and (fe )_ are the overall and individual gas phase coeffic-
C G C G
ients, k * is the liquid phase physical absorption coefficient, P and
p are the total gas pressure and molal gas density and H is Henry's
law constant. It should be noted that as increases the controlling
resistance will shift to the gas phase, and at low liquid phase resis-
tance calculation of the small difference between the overall and the
individual gas phase resistance will be subject to difficulties due
to precision of measurements.
-------
NSh
Ni/3
Sc
ou
30
20
10
o WATER VAPORIZATION
METHANOL DESORPTION
x AMMONIA ABSORPTION
0
i V*"
X*x
o
/
o
0
o
yv
^
0
'/
so
o
ft
/
1000
2000
3000
5000
7000
FIGURE 3
-------
14
APPARATUS AND PROCEDURE
In the study of the relatively rapid processes at the gas-liquid
interface which occur in gas absorption the development of suitable
experimental techniques to obtain reproducible data under known experi-
mental conditions was an important problem. Several laboratory devices
have been used with varying success each with its particular advantages
and disadvantages.
Two of these devices, the laminar jet developed by Pigford and
Scriven and the short wetted wall column developed by Vivian and Peace-
man perform closely according to theoretical predictions. The laminar
jet gives highly reproducible data and is relatively easy to operate.
However, it is limited to the use of a pure gas phase (i.e., no gas
phase resistance), and therefore, it could not be applied for this
study involving absorption of SOQ at very low partial pressures.
The short wetted wall column has been used to study both liquid
and gas phase controlled absorption. It provides surface exposures
of the same order of magnitude as many commercial types of absorbers,
and penetration theory can be used to analyse data obtained with it.
As originally designed, it was not suitable to operate with a slurry
system. However, a special feed system was designed and constructed
which allowed its operation with slurries, and all the data reported
here were obtained with a short wetted wall column.
Operating a relatively small flow system with slurries presents
considerable difficulty, and in this case particularly with maintenance
of uniform slurry concentration at the column. By use of a circulating
system with a short downward-sloping drawn-off through a small-bore
flow meter directly to the head of the short wetted wall column data with
-------
15
a fair degree of reproducibility could be obtained.
The overall apparatus is shown schematically in Figure 4. Slurry
was prepared by mixing distilled water and reagent grade calcium hydroxide
in the feed tank in which an agitator maintained the hydroxide in suspen-
sion. A pump circulated the slurry through a small diameter line at high
velocity to an open surge tank mounted above the column. From the surge
tank the slurry flowed directly back to the feed tank. Flow of slurry
to the column was drawn from the line to the surge tank through the cali-
brated small-bore flow meter. Slurry discharged from the column was
discharged to waste except when liquid samples were being collected.
Liquid samples were collected under N_ to reduce possible oxidation of SO..
The gas phase was prepared by mixing in a baffled section of the gas
line metered flows of N. and SO- supplied by cylinders. The N2 flow meter
was calibrated by use of a calibrated dry test meter and the SO. flow meter
was calibrated by analysis of the mixed gas streams. Essentially all of
the data were taken at a Reynolds number of about 3000. This was chosen
as a balance between the need to minimize gas phase resistance and the
need to maintain an undisturbed liquid layer on the column wall.
The short wetted wall column is shown in outline in Figure 5, and is
the same construction as described elsewhere (8).
A special head for the column was constructed to avoid dead spots
where slurry solids could collect and to give a uniform liquid layer on
the wall. After considerable experimentation the design shown in Figure 6
was developed. All of the data reported here were obtained with the use
of this head.
Run procedure consisted of continuous operation of the flow systems
to achieve steady-state after which the liquid phase to and from the column
was sampled and pertinent data on operating conditions were read. Only
-------
Flowmeter
Flowmeter
N.
A A
N.
Gas
Saturator
Flowmeter
A
SO,
sww
Column
Gas
Heater
Gas
Mixer
Liquid
Sampler
ABSORPTION
Sampler
FIGURE 4
SCHEMATIC DIAGRAM OF APPARATUS
OF SO2 IN SLURRIES IN SHORT WETTED WALL
Liquid
Sampler
Liquid
Supply
Tank
Pump
COLUM N
-------
17
TEFLON
TEFLON
0-RING
STAINLESS
STEEL
MOOD:
SHORT WETTED WALL COLUMN ASSEMBLY
-------
18
Tef Ion
Head - Piece
7
Glass Column
FIGURE
-------
19
the outlet liquid analyses have been reported here since, in general,
the total alkali content of both samples agreed within operating pre-
cision.
The procedure used in runs with calcium hydroxide solutions and
with water was essentially the same.
In spite of the high velocity circulating liquid system and short
draw off to the column, entirely satisfactory operation with a calcium
carbonate slurry was not obtained before the laboratory work was termin-
ated. Further modification of the liquid feed system would have been
required to obtain satisfactory data.
Details of the analytical methods and equipment calibration are
given in Appendex A.
-------
20
RESULTS AND DISCUSSION
The rate of absorption of S0_ into lime slurry was calculated from
the volumetric flow rate of the slurry and the concentration of total SO
(i.e., unhydrolysed SO , bisulfite and sulfite ion) in the slurry leaving
the short wetted wall column. An overall coefficient, (K ) expressed as
C G
cm/sec, was obtained by dividing the rate of absorption by the interfacial
area of the column, the znolal density of the gas and the mole fraction of
S02 in the entering gas. The data are summarized in Appendix C. In order
to interpret these data it was necessary to distinguish between the com-
ponents of the overall resistance to absorption.
The gas phase resistance was estimated from previous work on the
short wetted wall column (£,9.) , and assuming additivity of resistances
as demonstrated by Mayr (2^,4) the liquid phase resistance was calculated
by deducting the gas phase resistance from the overall resistance. This
liquid phase resistance was assumed to be the result of kinetic and
diffusional effects, and the liquid phase coefficient is reported in
Appendix C as the product <(> fe * where fe* is the absorption coefficient
l< L
without chemical reaction and <|> is the enhancement factor to account for
the effect of chemical reaction in the liquid phase. The dependence of
the absorption coefficient without chemical reaction, k,*» on liquid rate
and fluid properties is well understood, and by estimating fe from
previous work on the short wetted wall column such as that of Mayr (4_)
the value of was obtained.
It should be noted that for the absorption conditions encountered
in this work, the liquid phase resistance accounted for only a small
fraction of the total resistance with the result that ordinary precision
of measurements introduced considerable uncertainty in the small differences
-------
21
representing the liquid phase resistances. This uncertainty directly
affected the calculated values of the enhancement factor. Consequently
the value of these data lie not in their absolute value but in their
order of magnitude and directional trends in supporting theoretical
considerations. In the runs with the lowest SO concentration the cal-
£
culation of a liquid phase resistance had no significance within the
precision of the experimental data, and consequently/ no conclusion can
be drawn from those runs except that in fact the absorption process
becomes gas phase controlled as the concentration of SO. in the gas
decreases.
The absorption of SO into lime slurries and solutions appears to
be a case of absorption accompanied by very fast chemical reactions.
Furthermore in the presence of excess hydroxyl ion the reactions appear
irreversible. Theory indicates that under these conditions an approxi-
mation for the enhancement factor is given by:
where C is the concentration of reactant B in the bulk of the liquid,
B
CA. is the concentration of solute A in the liquid at the interface and
v is a stoichiometric factor giving the number of molecules of B which
react with A. D. and D are the respective diffusivities. In the system
A D
under study here, C_ is the calcium hydroxide in solution and CA. is the
B J-
molecular SO in solution at the interface. Cft. is assumed to be in
equilibrium with the gas phase concentration at the interface. The
stoichiometric factor is 1, assuming that the reactions of importance in
the presence of relatively high hydroxyl ion concentration are:
SO.(aq) + OH" -» HSO,"
* J
-------
22
and HS03~ + OH~ "*" S03~ + H2°
which can be written as an overall reaction to indicate the relative
quantities of diffusing species:
S02(aq) + Ca(OH)2 -* CaS03 + H2O
The data obtained for the absorption into lime slurries and solutions
are plotted in Figure 7 to show the relationship of -1 to the ratio
CR/C .. In the runs with the slurries C was assumed to be saturation
*5 /\1 B
at the liquid temperature. The line through the data is drawn with a
slope of 1 as suggested by theory. While the data are somewhat scattered
due to difficulties with precision of data they tend to substantiate the
assumption for these conditions of absorption accompanied by very rapid
second-order irreversible chemical reaction.
The ratio C_/CA. is based only on a small range of C from half
saturation to saturation. The value of CA. however, varies widely due
to changes in gas phase concentration and gas phase driving force. The
CaSO, formed had very low solubility and precipitated. In the runs with
clear Ca(OH). solutions this precipitate was readily visible in the column.
These experiments indicate that the presence of solid calcium
hydroxide had little effect on the absorption rate which depended only
on the calcium hydroxide concentration in solution and not on slurry
concentration. Even though a highly reactive analytical grade of calcium
hydroxide was used to prepare the slurries, no effect of solid concentra-
tion is evident. Apparently the rate of solution of the solid was not
sufficiently rapid to influence the processes occurring at the interface
during the 0.1 sec (approx) exposure time.
These rate measurements were made for a single exposure of the
surface of the liquid layer, initially of uniform equilibrium composition.
Repeated exposure would, of course, deplete the reactant in solution
-------
fooo
-------
24
unless it were replenished by dissolution of the solid in the slurry.
Thus the rate of dissolution of the solid assumed importance in main-
taining the capacity of the liquid to absorb S0_ rather than partici-
pating in reactions at or near the interface. In view of this mech-
anism it should be possible to rank the behavior of various lime slurries
by observing their rates of solution in simple liquid phase measurements
without the necessity of characterizing them in actual absorption mea-
surements which are difficult to make. The mechanism also indicates the
desirability of providing adequate liquid holdup per unit interfacial
area to allow dissolution of solid and re-saturation of the liquid phase
between surface exposures. To perform satisfactorily, the rate of
dissolution per unit volume of slurry multiplied by the volume of holdup
per unit interfacial area must be equal to or greater than the rate of
absorption per unit interfacial area.
The short wetted wall column is well suited for investigating
absorption phenomena which occur at or near the interface and which
conform approximately to the surface renewal or penetration theory model
of gas absorption. Exposure times are relatively short, ranging from
about 0.05 to 0.2 seconds. Penetration depths are relatively small
compared to the thickness of the liquid layer. Consequently bulk pheno-
mena cannot interact to any great extent with what occurs in the liquid
at the interface until mixing occurs following the surface exposure.
In the system of interest here the absorbed SO reacted with
hydroxide in solution near the interface and depleted the hydroxide
concentration in that region. Solution of solid hydroxide proceeded in
the bulk liquid after mixing with the depleted surface liquid. To observe
the overall effect of interfacial and bulk processes an experimental
system which provides for repeated surface exposure and associated bulk
-------
25
reaction volume is required in addition to the short wetted wall column.
For example, rate data obtained with turbulent slurry film flow in a
long wetted wall column or with slurry flow over a packed bed of spheres
would probably be more representative of the overall process of absorp-
tion into calcium hydroxide slurry than the data obtained with the
short wetted wall, and such data combined with the fundamental homo-
geneous rate data obtained with the short wetted wall column would
allow analysis of the effect of the processes in the bulk liquid on the
absorption rate.
Data were obtained for the absorption of SO. into water in the
short wetted wall column over the same concentration range as used in
the lime slurry studies. These data were taken to indicate the perfor-
mance of the column and establish the experimental procedure without the
complicating factors due to the calcium hydroxide. In addition they
provide an indication of the effect of hydrolysis of SO in water on the
rate of absorption.
The rate of the hydrolysis of SO. in water has been studied (1.*6),
and the data obtained indicate only that it is a very fast reaction. The
reaction is:
S02(aq) + H20 t H* + HS03~
The few qualitative data available give rates so high that the maintenance
of equilibrium in water during diffusion of SO. away from the interface
would be expected. On this assumption mass balances on the rate of removal
of various forms of S0_ from the interface into the bulk of the liquid
allow the calculation of the enhancement of the absorption rate due to the
hydrolysis reaction. The absorption rate can be written as:
NA ' * feL*
-------
26
where, as before, CA>, C is the unhydrolysed molecular SO in solution.
The effect of the hydrolysis reaction was shown to be:
<)> = 1 +
D
A
/C/C
AAi
where K is the equilibrium constant for the hydrolysis reaction, and D
and D are the diffusivities of unhydrolysed and hydrolysed SO_ respec-
Jo Z
tively.
When the diffusivities of species A and B are nearly equal as in
the SO -water system (5) , and the concentration of SO. in the bulk, C ,
« ""^ £ A
is negligible relative to the concentration at the interface CA., the
expression for the enhancement factor reduces to:
4> * i -f
The data for the absorption of SO. into water are plotted in Figure 8
to show the relationship of <|>-1 to /K^/C^.. The line as indicated repre-
sents the theoretically derived relation for an assumed infinitely rapid
hydrolysis reaction. Although the data are somewhat scattered due to the
effect of precision of measurements in deriving the liquid phase resistance,
the trend of the data substantiates the theory based on a very rapid
hydrolysis reaction.
It is instructive to compare the hydrolysis reaction of SO. and water
with that of Cl. and water. The two systems have many similarities, and
the Cl_ system has been extensively studied. Both solutes react directly
with water to form hydrolysis products and establish a reaction equilibrium.
The evidence seems to indicate that dissolved SO reacts directly with
water to form an essentially completely dissociated product consisting of
hydrogen and bisulfite ions. Dissolved Cl_ reacts directly with water to
form hypochlorous acid and hydrogen and chloride ions. The rate of the
-------
FIGURE 3
ro
-------
28
chlorine hydrolysis has been measured by at least three techniques which
agree surprisingly closely. The rate is rapid but not as rapid as SO
hydrolysis appears to be.
The presence of hydrogen ions inhibits the hydrolysis of both Gl-
and SO . Under such conditions, absorption follows the laws of physical
absorption, and the driving-force for mass transfer is the unhydrolysed
molecular solute. At low hydrogen and hydroxyl ion concentrations data
on the absorption of Cl_ follow a pseudo-first order rate equation. At
hydroxyl ion concentrations above 0.01 N Cl hydrolysis involves hydroxyl
ions and appears to be second-order and irreversible. It is likely that
SO follows a similar pattern in its reactions with water and alkali
solutions, although apparently the hydrolysis reaction at medium and low
pH appears to be very much more rapid than is the case with Cl_.
At high concentrations of SO. in the gas and liquid the reaction
&
appears reversible and the rates of absorption approach physical absorp-
tion. As the concentration of SO_ in solution and in the gas is decreased,
the enhancement factor increases as the interfacial concentration CA.
decreases with the result that at very low concentrations the absorption
process will become controlled entirely by the gas phase resistance even
for absorption into water.
A number of S02-water runs (No. 52-57) were made with very low S0_
flow rates (the lowest used in these experiments), and it is suspected that
the flow meter did not operate reliably at this end of the scale. Consider-
able difficulty was experienced in this range during the calibration runs.
It is not known whether this was due to the float sticking or to analytical
procedures. However, the absorption data indicated that the SO. concentra-
tion in the gas calculated from the flow rates is too low since the overall
gas phase coefficient cannot be greater than the individual gas phase co-
-------
29
efficient. If it is assumed that the liquid phase offered negligible resis-
tance in runs 52-57 so that the overall (K )_ would be equal to (fe )_, the
C G C G
flowrates of SO would be approximately 1.5 times the rates indicated in
these runs. This appears to be within the precision of the calibration
in this range of the flowmeter. Consequently these data serve only to
substantiate the prediction that SO absorption at very dilute gas concen-
trations becomes essentially gas phase controlled.
A few runs were made using a CaCO, slurry. These are reported in
Table 3, Appendix C. Great difficulty was encountered in maintaining a
stable slurry flow due to the settling and plugging characteristics of
the CaCO. particles. Unfortunately, time did not permit resolution of
these operating problems. Definitive conclusions cannot be drawn from
this phase of the work.
Comparison of the magnitude of the enhancement factors for absorption
into water and into alkaline solution reveals that for the lower hydroxyl
concentration range (below 0.01 g.mole per liter) the enhancement is about
the same as for water. Further work is required to indicate the actual
effect of hydroxyl ion concentration/ although these data indicate that
above 0.01 g.mole Ca(OH) per liter the enhancement is expected to increase
with alkaline concentration. In this behavior, SO. absorption into water
and alkaline solutions appears to be similar to the Cl.-water system.
-------
30
CONCLUSIONS AND RECOMMENDATIONS
The distribution of resistance to the absorption of SO into lime
slurry in the short wetted wall column depended on the gas phase concen-
tration. At low concentrations (0.3 mole per cent SO } the major resis-
tance was found to be in the gas phase, and at concentrations below 0.1
mole per cent the absorption process was essentially gas phase controlled.
The absorption of SO2 into lime slurry appeared to be a system of
gas absorption accompanied by an "infinitely" rapid second-order irreversible
chemical reaction involving unhydrolysed molecular SO and hydroxide in
£
solution. The rate of solution of solid calcium hydroxide did not influ-
ence the absorption process at the interface during the short exposure time
characteristic of the short wetted wall column. The effect of the solid
calcium hydroxide was to increase the capacity of the liquid phase to ab-
sorb SO.. Adequate holdup of liquid would be required in a continuous
absorber to allow sufficient dissolution of calcium hydroxide to maintain
the hydroxide in solution near saturation.
The absorption of SO into water is a system of gas absorption accom-
panied by a fast reversible hydrolysis reaction. The reaction is suffi-
ciently rapid to maintain hydrolysis equilibrium in the liquid as SO
diffuses away from the interface. In agreement with theory the enhancement
factor varied approximately inversely with the square root of the gas phase
interfacial concentration, and the system became gas phase controlled at
concentrations of SO. in the gas phase below 0.1 mole per cent.
The rates of absorption of SO into water and into calcium hydroxide
solutions of about 0.01 g.mole per liter were about the same, and the rate
into hydroxide solutions increased slightly with increase in hydroxide
concentration in solution.
-------
31
Significant data on the absorption of S0» into calcium carbonate
slurries could not be obtained with the equipment available.
Experimental work on the absorption of SO in slurries in a long
column or packed bed is needed in addition to studies on equipment such as
the short wetted wall column in order to investigate the overall effect
of interfacial and bulk liquid reactions since the dissolution process does
not appear to affect the processes in the region of the interface.
-------
32
APPENDIX
-------
33
APPENDIX A
DETAILS OF PROCEDURE
1. Gas Flow Rate and Composition
The nitrogen-sulfur dioxide gas mixture feed to the column was pre-
pared by mixing metered streams of nitrogen and sulfur dioxide. The
rotameter used to meter the nitrogen gas from the supply cylinder was
calibrated on air using a calibrated gas test meter. The small correction
for variable conditions and from air to nitrogen was made as shown on
Figure 9.
The sulfur dioxide rotameter was calibrated by determining analytic-
ally the dilution of the sulfur dioxide in a measured stream of nitrogen.
The sulfur dioxide was feed into the nitrogen gas stream in the 1-inch
diameter glass line to the short wetted wall column. The gas mixture was
sampled at a point about six feet downstream from the mixing-point and a
series of baffles to promote mixing. The gas sample was taken in an evacu-
ated Gaillard flask of calibrated volume (approximate volume was 300 cm ).
The vacuum system was connected directly to the sample point and the Gaillard
flask. A mercury menometer was used to measure the pressure in the Gaillard
flask before and after admitting the gas sample to the flask.
The flask with the sample was cooled in an ice bath to reduce the pressure
sufficiently to allow introduction of 10 cm of 0.01 N KIO^ solution, 0.2 cm
of 20% KI solution and 2 cm3 of 0.1 N HC1. The KIO3 oxidized the SC»2 to
SO, or SO, to SO. and the remainder reacted with hydrogen ion to form
I_ which was titrated with standard thiosulfate solution. The m. equivalents
of iodate less the m. equivalents of thiosulfate required for the-I gave
the m. equivalents of SO. in the sample. The ratio of the m. moles of SO_
M ft
to the m. moles of gas in the Gaillard flask gave the mole fraction of SO. in
-------
34
the gas sample. By material balance the sulfur dioxide flow rate was
calculated from the metered nitrogen flow rate to calibrate the sulfur
dioxide rotameter. When the gas phase was not analyzed during the absorp-
tion runs, the gas composition was calculated from the metered flow rates
of the nitrogen and sulfur dioxide.
2. Liquid Flow System
The liquid system consisted of a polyethylene feed storage container
(approx. 30 gallon capacity) in which the slurries of hydrated lime and
calcium carbonate were prepared by mixing the dry powder with distilled
water. A centrifugal pump circulated the slurry through a loop of 12-mm
glass tubing which extended about four feet above the column with the return
leg discharging into the feed storage container. A stirrer in the feed
storage container was used in addition to the recirculation to maintain
the slurry solids in suspension. A side stream through a small glass line
was taken from the upper part of the recirculating loop through a capillary
flowmeter and fed directly to the head-piece of the short wetted wall column.
The flowmeter was calibrated by direct measurement of the flow.
Liquid effluent from the column which was not needed for sampling for
analysis was discharged to the sewer.
3. Liquid Sampling and Analysis
In runs in which a slurry was used as absorbent samples of the effluent
liquid were taken from a sampling point immediately below the wetted wall
column in flasks purge with N_. Operating difficulties in handling the
slurry absorbent often interferred with steady operation and to check these
conditions duplicate samples were taken ten to twenty minutes after the
first set of samples.
In the runs using water as absorbent/ the effluent liquid sample stream
flowed continuously through a twenty-five ml pipette. After steady state
-------
35
had been established the pipette was removed from the line, drained to
the mark and the measured sample volume introduced under the surface of
a known amount of KIO solution.
The effluent liquid was analyzed for calcium hydroxide, carbonate
and sulfite. To determine the hydroxide and sum of the carbonate and
sulfite concentrations, a 25 ml sample of the slurry was titrated with
0.2 N HC1 solution to the phenolphthalein (PH) and methylorange (MO)
end points.
The hydroxide concentration in g.mole/liter is given by:
/ran m. eg. HC1 (to PH) - m. eg. (PH to MO)
(On) » 'm " . * "'- -'-
2 2 (ml sample)
The carbonate plus sulfite concentration in g.mole/liter is given:
where PH refers to the phenolphlthalein endpoint and MO refers to the
methylorange endpoint. To determine the sulfite and dissolved sulfur
dioxide concentration, an effluent liquid sample (25 ml) was added to a
titration flask containing five or ten ml of 0.01 N KIO- to which had been
added 0.2 to 0.4 ml of 20% KI and two to four ml of 1.0 N HC1, The HC1
added in excess dissolved the insolubles in the slurry and released I
from the excess KIO_ not required to oxidize SO and SO,. The I was
titrated with standardized sodium thiosulfate solution to the starch endpoint.
The oxidizable SO and SO in m. moles/liter is given by:
JL 3
ten a. cr. \ = (m. eg. KIO3 - m. eg. thiosulfate) 10
(S02 + S03 ) 2 (ml sample)
-------
36
APPENDIX B
SAMPLE CALCULATIONS
Sample calculations are given for the data recorded for tabulated
Run No. 1, Appendix C except as noted otherwise.
Nitrogen Flowrate
Rotameter reading 70.5
Gas temperature 14.1 °C (287.2 °K)
Gas pressure (abs) 770 mm Hg
From Figure 9, Q/erpn\ = 2.37 cu.ft./min.
QN » (0.638)(2.37)/(770/287.2)
- 2.48 SCFM
Sulfur dioxide Flowrate
Rotameter reading 4.0
Gas temperature 24.5 °C (297.6 °K)
Gas pressure (abs) 759.1 mm Hg
cu.fl
s-3,
From Figure 10, Q/ornr..sa 0.78 x 10 cu.ft./min.
(5TD)
Qso - (0.626)(0.78 x 10 J)/(759.1/297.6)
0.78 x 10~3 SCFM
Gas Analysis (inlet)
Mole fr. S02 = QSo2/<°-N + °-S02>
0.78 x 10~3/2.48
- 0.315 x 10~3
Liquid Flowrate
Flowmeter reading 201 mm HO
Liquid temperature 23.2 °C
From Figure 11,Flowrate =325 cm /min.
Diameter of column =3.18 cm
Flowrate - ., ,Q. -32.5 cm /min-cm
71 (j . La )
-------
37
$ 2
S
-------
U)
00
-------
//
to
J£>t>
260
foo
-------
40
Liquid analysis (outlet)
Sample (A) volume 25oQ cm
Titration with 0=2 N HC1 solution
to phenolphthalein endpoint 6.54 m.eq.
to methyl orange endpoint 0,175 m.eq.
total 6,715 m.eq.
ca,OH,2 cone . ''
= 0.1273 g.mole/liter
CaCO 4- CaSO cone « 0.175/25.0
« Oo0070 g0mole/liter
Sample (B) volume 25 oO cm
KIO, solution added 0»1987 m.eq0
Titration with Na S_O solution 0.1882 m.eq.
KIO. solution used for SO oxidation
- 0 = 1987 - 0«,1882
= 0.0105 m.eq,
0.0105 __3
CaS03 cone = { x 10
0=210 m. mole/liter
CaCO cone = 0.00679 gdnole/liter
Kate of Absorption of SO2
(Liquid flowrate)(Outlet CaSOg cone)
(Surface area)
(32o5) (0,210)
(60) (4.5) (1000)
fi
25.3 x 10 numole SO /sec-cm
-------
41
Overall Gas Phase Coefficients
By definition:
NA 2
K « j g.mole/sec-cm -atm
(p - p*) 10
Na
Ky
g.mole/sec-cm
(y - y*) 10
NA
(K } » - - cm/sec
PG(y - y*) i
-------
42
Gas phase mass transfer studies in short wetted wall column (9) indicate
that:
-^- . = 18.5 at Nn = 2980
(N0 )1/3 RS
sc
The N for SO in N was calculated from data given by Sherwood and Reid,
"Properties of Gases and Liquid" McGraw-Hill Book Co., Inc.
Diffusivity, DP = 0.122 cm -atm/sec at 273 °K
At 298 °K and 1 atm, D - 0.139 cm2/sec
Gas density, p = 0.00114 g./cm
Gas viscosity, y =» 0.018 x 10 g./sec-cm
.". N » 1.136
Sc
k d
and N - -- - 18.5 (1.136)1/3 - 19.3
ih * (19.3) (0.139)
(fec,G . - __ - = o.84 cm/sec
Gas phase resistance = .\ . = 1.190 at N 2980
In this range of N , (k ) is approximately proportional to the N .
Liquid Phase Coefficients
By definition the liquid phase coefficient for physical absorption
is defined by:
NA " feL* (Ci - C)
where C is the concentration of the absorbed specie at the interface and in
the bulk liquid. When a chemical reaction occurs in the liquid phase in-
volving the absorbed specie, the rate of absorption is also governed by the
unreacted specie driving force. The enhancement of the rate due to the
-------
43
reaction is allowed for by a factor , must be defined for the various kinetic schemes
encountered. The hydrolysis of SO in water can be written as:
S0_(aq) J H+ + HSO ~
£ *3
where SO_(aq) refers to unhydrolysed molecular SO in solution in water.
This reaction is reversible and is believed to be very rapid (I) so that
during a diffusion process, equilibrium is maintained between reactants
and products.
If C. is the molecular SO concentration and C_ is the bisulfite
A £ E
ion concentration, the liquid phase equilibrium for SO^ absorbed into water
is given by :
Theoretical analysis assuming film theory for simplicity gives the enhance
ment factor, , in terms of the equilibrium concentrations:
and the rate of absorption is given by:
N - fe_* (C . - C )
A L Al A
If D - D , substituting the expression for in the rate equation gives:
Ct A
-------
44
where CTi and CT refer to the total SO in solution (i.e., SO (aq) plus
HSO ~) . Furthermore if the concentration o
zero, the enhancement factor, , reduces to
HSO ~) . Furthermore if the concentration of total SO in the bulk is
4> = (1
Thus becomes very large at low partial pressures of SO,, since CA.
This of course shifts the controlling resistance to the gas phase.
Estimation of the value of the enhancement factor, cj>, for the absorp-
tion of SO into calcium hydroxide and calcium carbonate solutions and
slurries is complicated by the fact that the H+ ion concentration is not
coupled directly with the bisulfite ion concentration. Consequently,
calculation of the <£ factor is dependent on specification of the H+ or OH~
ion concentration. These cases have not yet been worked out, and to analyse
the experimental data for absorption into calcium hydroxide slurries it will
be assumed that the molecular SO concentration at the interface was in
equilibrium with the SO concentration in the gas phase at the interface.
On this basis the product, k *, can be calculated from the data. can
L
then be estimated using correlations for physical absorption such as Mayr's
equation (4) for k *.
~ L
The equilibrium relation connecting gas phase concentration and
molecular SO. in the liquid is Henry's law:
Combining this with the liquid phase rate equation:
NA ' * kL
-------
45
we have
NA - feL* HP (y± - y*)
Applying additivity of resistances, we get:
(KC>G PG fe * » 0.334 cm/sec
The physical absorption correlating equation due to Mayr is:
fe */-*-= 1.759
L V/DA
-------
46
Data for Run 3:
z - 4=5 cm
D - 1,753 x 10~5 cm/sec at 25 °C
A
T = 32c5 g./min-cm
feT* .If- '- 1.759(32.5)°'354
A
= 6.03 (cm/sec)z
~- - 0.505 x 103
A
fe * « 0.0119 cm/sec
Then
A = (4> ki.*) 0.334
0119 " 28.1
Enhancement factor for absorption into water
(Data for Run 33)
(fe ) = 0.84 cm/sec
(K ) - 0.504 cm/sec
C G
P « 1.0 atm
p_ = 40.5 x 10 g.mole/cm
G
T - 23.4 °C (296.5 °K)
F » 25.0 g./rain-cm
Henry's law constant
- io-4523
H = 1.302 g. mole/1 iter-atm
= 1.302 x 10 g. mole/cm -atm
-------
47
[ i JL_1
0.504 0.84 J
1 m (1.302 x 10~3)(1.0)
*L 40.5 x 10"6}
fe * BE 0.0392 cm/sec
L
~ - 0.532 x 103 (sec/cm)'
A
1.759(25.0)°*354 - 5.48 (cm/sec)^
fe * - 0.0103 cm/sec
0.0392
3*81
Theoretical (j>
K
(J) « 1 +
H ** 1.302 x 10 g. mole/cm -atm
y m 3.16 x 10"3
K - 0.019 g. mole/liter
Applying additivity,
y± - (3.11 x 10"3) (1 - ^ ) - (1.53 x
CAi - (1.302 x 10"3) (1.0) (1.53 x lo"3)
- 2.00 x 10~ g. mole/can
0.019 x IP"3
2.00 x 10
-------
48
APPENDIX C
SUMMARIZED DATA
-------
TABLE I
Absorption of SO2 into Lime Slurries
Run
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Flowrate
SCFM
CM
2
2.48
2.48
2.43
2.49
2.49
2.48
2.43
2.45
3.37
2.48
2.42
2.50
2.43
2.43
2.44
2 Flowrate
SCFM x 10 3
o
W
0.78
0.78
9.02
8.82
8.69
9.02
9.30
8.82
8.92
9.02
9.02
8.82
2.58
2.49
2.40
1
b g
n 75
(0
13
2980
2980
2950
3030
3030
3020
2960
2980
4080
3010
2940
3040
2930
2930
2950
n N
n to
H «"!
id M o
0> X
« l-l
id g
CD g
0.315
0.315
3.71
3.54
3.48
3.63
3.83
3.60
2.64
3.64
3.73
3.51
1.06
1.025
0.982
tal Pressure
ironPg
£
759.1
759.1
759.1
756.8
756.8
756.8
758.8
758.8
765.2
765.2
765.2
765.2
760.0
760.0
760.0
quid Temperature
°C
3
25.6
26.3
26.4
25.5
26.9
26.4
32.3
31.4
27.9
29.0
28.7
28.5
28.4
28.3
28.4
quid Flowrate
cra^/min-can
H
32.5
18.5
32.5
24.0
11.0
35.0
21.5
7.65
22.0
31.5
8.35
21.7
33.0
22.8
10.8
OUTLET
LIQUID ANALYSIS
O^X X V.
(00) 0) (MO)
H H ~ H
3 i 8na si
o *
0.210
0.486
0.988
1.337
2.854
0.730
1.390
3.748
1.524
1.006
2.742
1.294
0.298
0.474
0.890
3 »
0.0068
0.0054
0.0346
0.0383
0.0399
0.0385
0.0354
0.0323
0.0191
0.0074
0.0071
0.0079
0.0081
0.0071
0.0047
id
o o>
0.1273
0.1263
0.0716
0.0772
0.0644
0.1924
0.1944
0.1664
0.0984
0.0510
0.0416
0.0495
0.1222
0.1194
0.1098
ABE.
RATE
\o
o
X
a M
25.3
33.3
118.9
118.8
116.3
94.6
110.7
106.2
124.2
117.4
84.8
104.0
36.4
40.0
35.6
0
04)
^^ n
OX
1.97
2.60
0.787
0.825
0.821
0.637
0.705
0.719
1.139
0.781
0.550
0.717
0.831
0.945
0.878
*
-S-
-
-
0.661
1.054
0.986
0.115
0.169
0.265
0.007
0.485
0.080
0.213
-
-
^QL
-
-
5.5
79.8
117.4
9.2
15.1
34.4
0.7
38.8
9.9
19.7
-
-
-------
TABLE II
Absorption of SO- into Lime Solutions
Run
16
17
18
19
20
21
22
23
24
25
26
27
MS
W «C|
H CO
sT
2.40
2.46
2.43
2.39
2.40
2.40
2.43
2.42
2.41
2.43
2.43
2.40
* x
H a
rica
o
en
8.95
6.01
8.55
5.97
3.33
0.80
8.61
6.87
5.12
3.37
8.58
6.83
k
5
rH
fa 8)
M
U Z
id
u
2920
2980
2950
2900
290.0
2900
2950
2940
2920
2940
2950
2910
(0 Oi
HO
n to
f*4 *f^
tA ^4 ^^
2«MH
ex
01 -«
o
H
33.0
33.0
33.0
32.8
33.0
32.8
26.8
26.8
26.8
26.8
20.0
20.0
ABE.
RATE
OUTLET *
LIQUID ANALYSIS °
OM * *3 X
o \ \ x. .
M «
H
1!
0.828
0.652
0.734
0.524
0.238
0.048
0.734
0.414
0.518
0.352
1.096
1.068
0)
H
8wa
3 CT>
0.0014
0.0013
0.0013
0.0013
0.0006
0.0005
0.0011
0.0009
0.0011
0.0009
0.0016
0.0017
NO) O
§1 i
Si a
N
B U
Q) W S
0.0138 101.2 0.666
0.0139 80.4 0.806
0.0091 89.7 0.626
0.0093 64.0 0.629
0.0090 29.1 0.514
0.0090
5.8 0.428
0.0118 72.8 0.505
0.0114 41.1 0.356
0.0115 51,4 0.560
0.0112 34.9 0.617
0.0055 81.2 0.565
0.0048 79.1 0.682
V
0.150
0.870
0.112
0.112
0.059
0.040
0.055
0.026
0.075
0.104
0.078
0.183
"Si
12.8
73.1
9.4
10.2
5.0
3.4
5.0
2.5
6.9
9.5
7.9
17.9
t/i
o
-------
TABLE III
Absorption of SO. into Calcium Carbonate Slurries
Run
28
29
30
31
5
2s
gs
H W
fa
IN
Z
2.38
2.43
2.44
2.43
din
4* 0
<-t a
01 en
o
01
8.95
9.00
2.71
0.80.
H
fa 01
M
0) Z
BJ
O
2890
2950
2950
2930
n n
rtO
n ca
1-1 «»>
* fc o
0 X
0)f-l
m g
o 6
3.76
3.70
1.11
0.33
res sure
Hg
1
a
g
(760)
767.2
767.2
767.2
Temperature
C
TJ
I
H
26.2
28.9
28.7
28.6
Plowrate
dLn-cm
o x
&1
rt
32.5
22.9
22.9
23.0
OUTLET
LIQUID ANALYSIS
IN J J J
OX. X X
31
O '
EH a
0.714
0.340
0.200
0.200
§1
nj .
U Of
-
-
_
ABS.
RATE
§
U
fl)
n
85.9
28.9
17.0
17.0
«-« n
o x.
0.561
0.192
0.376
1.265
0.082
0.012
0.034
7.6
1.1
3.1
-------
TABLE IV
Run
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
VI
N2 Flowrate
SCFM
2.41
2.41
2.39
2.39
2.39
2.41
2.35
2.39
2.44
2.44
2.43
2.43
2.43
2.43
2.44
2.44
2.40
2.47
2.47
2.47
2.41
2.39
2.41
2.39
2.32
7..^%
S(>2 Flowrate
SCFM x 10 3
7.60
7.50
7.45
7.60
7.50
7.45
2.70
2.70
2.70
2.70
2.70
2.70
2.65
2.65
8.85
8.70
8.80
8.80
9.20
9.05
0.80
0.70
0.90
0.90
0.95
^.^s
|
D^ flj
M
10 S3
n)
O
2910
2910
2880
2880
2880
2910
2840
2880
2950
2950
2930
2930
2930
2930
2950
2950
2900
2980
2980
2980
2910
2880
2910
2880
2800
7.ft%fc
Absorption of
Gas Analysis
mole f r . SO0
x 103 2
Total Pressure
mmHg
3.16 765
3.11 765
3.11 765
3.18 765
3.14 765
3.10 765
1.15 765
1.13 765
1.11
1.11
1.11
1.11
1.09 760
1.09 760
3.63 760
3.56 760
3.66 760
3.56 760
3.73 760
3.66 760
0.331 766
0.293 766
0.373 766
0.376 766
0.410 766
a.^% 16*
S02 into
£-
O
F
H
23.2
23.4
23.6
23.6
24.3
24.3
25.4
25.5
25.5
24.8
24.8
24.8
26.2
26.1
28.5
27.6
28.2
28.0
26.3
26.1
25.0
25.0
25.0
25.0
25.3
T.S.I
Water
01
%
u
f-t §
0^0
38
25.0
25.0
13.0
13.0
41.2
41.0
25.6
25.6
32.4
32.4
20.2
20.2
10.6
10.6
33.8
33.5
20.2
20.2
9.5
9.5
33.0
33.0
20.2
20.2
10.2
va.a
OUTLET
LIQUID
ANALYSIS
rn
ui O
H h
3 gs
g «P
0.599
0.690
1.174
1.195
0.615
0.517
0.365
0.343
0.278
0.286
0.444
0.415
0.705
0.705
0.608
0.602
0.860
0.874
1.818
1.850
0.150
0.153
0.312
0.308
0.570
a.&tfi
ABS.
RATE
vo
o
-I
X
55.5
63.8
56.6
57.5
93.8
78.5
34.6
32.5
33.4
34.3
33.2
31.1
27.7
27.7
76.4
74.5
64.1
65.7
64.2
65.3
18.3
18.7
23.4
23.1
21.3
«.-T
O 0)
~» n
n
0.425
0.504
0.440
0.438
0.718
0.610
0.725
0.693
0.725
0.745
0.720
0.675
0.613
0.613
0.507
0.504
0.433
0.433
0.415
0.430
1.33
1.54
1.51
1.49
1.25
V--i»
V
0.0267
0.0392
0.0287
0.0285
0.1570
0.0708
0.1795
0.1340
0.1795
0.2231
0.1704
0.1166
0.0772
0.0772
0.0472
0.0461
0.0330
0.0330
0.0285
0.0300
-
-
-
_
_
_
2.59
3.81
3.43
3.41
11.89
5.40
16.03
11.96
14.71
18.59
16.87
11.54
9.27
9.27
3.66
3.63
3.08
3.08
3.56
3.75
_
_
_
_
_
_
in
-------
53
APPENDIX D
NOMENCLATURE
C Concentration in liquid phase, g.mole/cm ; subscript A refers to
molecular (unhydrolysed) so., B to calcium hydroxide, E to bisul-
fite ion, i to interfacial concentration.
2
D Diffusivity, cm /sec; subscript A refers to molecular (unhydrolysed)
SO , B to calcium hydroxide, E to bisulfite ion.
d Column diameter, cm
G gas rate, g.mole/sec-cm
H Henry's law constant, g.mole/cm -atm
K Hydrolysis constant for SO in water, g.mole/cm
(K ) Overall coefficient based on gas phase concentration units, cm/sec.
C G
(fe ) Gas phase coefficient, cm/sec.
c G
k * Liquid phase (physical absorption) coefficient/ cm/sec.
L
L Liquid rate, cm /sec-cm
N Schmidt number, y/pD
Sc
N . Sherwood number, k d/D
Sn c
N Reynolds number, 4F/y or 4Qp/irdu
Re
P Total pressure, atm
p Partial pressure, atm; superscript * refers to partial pressure
in equilibrium with bulk liquid phase.
Q Gas flow rate, cm /sec or cu.ft./min.
R Gas constant, 82.1 cm -atm/g.mole - degree K.
-------
54
T Temperaturep degree K«
y mole fraction in gas phase
Z Column length? cm
Greek Symbols
F Liquid rate; g°/sec-cm
y Viscosity,, g/sec-cm
p Gas density, go/cm ; subscript G refers to molal gas density,
g0mole/cm
4> Enhancement factor (dimensionless)
-------
55
APPENDIX E
LITERATURE CITED
1. Eigen, M., K. Kustin and G. Maass, Z. physich. Chem. N.F. 30^130 (1961)
2. Brian, P. L= T., J. E» Vivian and So T. Mayr, Ind. Eng. Chem.
Fundamentals, 10, 75 (1971).
3. Johnstone, H. F. and P. W. Leppla, J. Amer= Chem. Soc., 56, 2233 (1934)
4. Mayr, S. T., Sc.D» Thesis, Dept. of Chem. Eng., M.I.T. (1970).
5o Peaceman, D. W., Sc.D. Thesis, Dept. of Chem. Eng., M.I.T. (1951).
6, Phipps, R. L., S.B. Thesis, Depto of Chem. Eng., M.I.T. (1947).
7. Spalding, C. W., A.I.Ch.E., Jl, _8_, 685 (1962).
80 Vivian, J. E. and Behrmann, W. C,, A.I.Ch.E., Jl, 11, 656 (1965).
9. Vivian, J. E. and T. Schoenberg, A.I.Ch.E., Jl, 14, 986 (1968).
-------
[BIBLIOGRAPHIC
SHEET
DATA
1. Report No.
EPA-650/2-73-047
"* Title and Subtitle
Absorption of SO2 into Lime Slurries: Absorption
Rat es and Ki net i cs
S.^ecipient's Accession No.
"5. Report Date
December 1973
6.
Author(s)
J. Edward Vivian
8. Performing Organization Kept.
No.
Performing Organization Name and Address
Massachusetts Institute of Technology
Department of Chemi cal Engineering
Cambridge, Massachusetts 02139
10. Ptoject/Task/Work Unit No.
ROAP 21ADE-21
11. Contract/Grant No.
68-02-0018 (Task 4)
12. Sponsoring Organization Name and Address
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, North Carolina 27711
13. Type of Report & Period
Covered
Final
14.
15. Supplementary Notes
stractsThe report gives results of a study of the absorption of SO2 from an SO2/N2
gas mixture into water and lime solutions and slurries (in a short wetted wall
column at approximately 1 atmosphere and 25°C). The mole per cent of SO2 in the gas
Was varied from 0.03 to 0. 37. In solutions of calcium hydroxide greater than 0. 01 g
mole/liter and in slurries, the system appeared to be gas absorption accompanied by
an "infinitely" rapid irreversible second-order reaction in the liquid phase. The only
effect of the solid in the slurries was to increase the absorption capacity. In water,
the system appeared to be gas absorption accompanied by a fast reversible hydrolysis
reaction in the liquid phase. The major resistance to gas absorption at these concen-
tration levels was found to be in the gas phase. At the lowest concentrations studied,
the system was gas-phase controlled.
Keywords and Document Analysis. 17o. Descriptors
Hydrolysis
Kinetics
' Key Words and L
Air Pollution
Flue Gases
Des ulfur ization
Sulfur Dioxide
Absorption
Calcium Oxides
Slurries
Solutions
17b. Identifiers/Open-Ended Terms
Air Pollution Control
Flue Gas Cleaning
I7c. COSATI Field/Group 7D, 13B
'8. Availability Statement
Unlimited
19.. Security Class (This
Report)
UNCLASSIFIED
20. Security Class (This
Page
UNCLASSIFIED
21. No. of Pages
61
22. Price
NTIS-35 (REV. 3-72)
THIS FORM MAY BE REPRODUCED
U5COMM-DC M852-P72
------- |