CD A U.S. Environmental Protection Agency Industrial Environmental Research
I—I •» Office of Research and Develonment I aboratnrv
i_.'.o. r_iiviruniiit r i u ii'i,. i lull "yt.'nuy IIUIUSULJI ciiviiuui oitii CDA fiAA/7 77 AOC
Office of Resear-h and Development L.iboratory CrM-OWU/ I - 11 'V£9
Research Triangle Park, North Carolina 27711 MflFCh 1977
ANALYSIS AND SIMULATION
OF RECYCLE SO2-LIME SLURRY
IN TCA SCRUBBER SYSTEM
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. iocioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
EPA-600/7-77-026
March 1977
ANALYSIS AND SIMULATION
OF RECYCLE SO2-LIME SLURRY
IN TCA SCRUBBER SYSTEM
by
C.Y. Wen and Fred K. Fong
West Virginia University
Department of Chemical Engineering
Morgantown, West Virginia 26506
Grant No. R800781-03-0
Program Element No. EHE624A
EPA Project Officer R.H. Borgwardt
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, N.C. 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
-------
Figures
Tables
Abbreviations and Symbols
CONTENTS
Page
1. Introduction ....................... 1
1.1 Processes for Desulfurizing Flue Gas ......... 1
1.2 Description of the Wet Lime/ Limes tone
Scrubbing Systems ................... 2
1.3 Objective of this Study ............... 5
2. Literature Review ..................... 6
3. Pressure Drop in the Turbulent Contacting Absorber .... 10
3.1 Model of Pressure Drop in TCA Scrubber ........ 10
3.2 Simulation of the Pressure Drop Across TCA Scrubber. . 20
4. Absorption of S02 and CO^ in TCA ............. 25
4.1 S02 absorption — Effect of Pressure Drop
on the S02 Absorption Efficiency ........... 25
4.2 C02 Absorption .................... 29
5. Simulation and Design of Recycle Lime Scrubbing System. . . 43
5.1 Material Balances ................... 43
5.2 Prediction of the Concentrations of Magnesium and
Chloride in the Scrubbing Slurry ........... 46
5.3 Simulation ...................... 49
5.4 Design of Lime Slurry FGD Systems Using TCA Scrubbers 53
5.5 Effect of Variations in the Parameters of the Models
for SO, and CO. Absorptions on the Outcome of the
Simulation ...................... 59
6. Conclusion and Discussion ................. 70
Bibliography ............................ 73
Appendix
A. Computer Program ..................... 7?
B. Plots of the Mass Transfer Coefficient for C02
Asorption vs. Various Average pH Values ......... 103
ill
-------
List of Figures
Number Page
1.1 Schematic Arrangement of the TVA Scrubber System .... 3
1.2 Schematic Arrangement of the EPA Pilot Scrubber
System 4
3.1 Schematic of TVA Shawnee Three-Bed TCA 11
3.2 Schematic of the EPA/RTP Research TCA Scrubber 12
3.3 An Idealized Stage of a Turbulent Contacting Absorber. . 13
3.4 Effect of Gas Velocity and Liquid Velocity on Pressure
Drop for TCA 14
3.5 Effect of Gas Velocity and Liquid Velocity on Pressure
Drop for TCA 15
3.6 Pressure Drop Across the Grids as a Function of Gas
and Liquid Mass Velocity in the Shawnee TCA Operated
Without Packing Spheres 17
3.7 Modified Pressure Drop for the Packing Section,
AP-, as a Function of Gas and Liquid Flow Rates .... 21
3.8 Comparison of the Predicted and Observed Pressure
Drop Across the TCA Scrubber 24
4.1 Effect of Pressure Drop on the SO- Removal Efficiency. . 27
4.2 Scrubber and Idealization of the Concentration
Profile of Carbon Dioxide 35
4.3 Effect of Slurry pH and Flow Rate on the Overall
Mass Transfer Coefficient for CO- Absorption
into Recycled Lime Slurries in a TCA Scrubber 42
5.1 A Simplified Lime-TCA Flue Gas Desulfurization .... 44
5.2 Concentrations of Magnesium and Chloride in the
Recycle Slurry 48
5.3 Flow Diagram for the Simulation of Wet Lime
Scrubbing Process 50
iv
-------
List of Figures (Con't)
Number page
5.4 Comparison Between the Observed and Calculated
SO- Removal Efficiencies in TCA Lime Slurry Scrubber ... 51
5.5 Comparison of the Observed and Calculated pH at
the Entrance and Exit of the TCA Scrubber ......... 52
5.6 Simulation of Lime Slurry TCA Scrubber Indicating
Maximum Flue Gas Flow Rates ............... 55
5.7 Operating Lines for Lime TCA Scrubber at Various
Equivalent Packing Heights ............... 56
5.8 Effect of Variations of the Mass Transfer Coefficient,
k?a, for S0~ Absorption on the System Operating Lines . . 61
&
5.9 Effect of Variations of the Mass Transfer Coefficient,
k|a, for S02 Absorption on the System Operating Lines . . 62
5.10 Effect of Variations of the Overall Mass Transfer
Coefficient, (Kra)Cn * ^or ^2 AbsOTPtion on
System Operating Lines .................. 63
5.11 Effect of Variations of the pH Effect Factor, A , in the
Packing Section for S02 Absorption on the System
Operating Lines ..................... 64
5.12 Effect of Variations of the pH Effect Factor, AS, in the
Spray Section f6r S02 Absorption on the System
Operating Lines ..................... 65
5.13 Effect of Variations of the Magnesium Effect Factor,
A , in the Packing Section on the System Operating Lines . 66
5.14 Effect of Variations of the Magnesium Effect Factor,
A in the Spray Section on the System Operating Lines . . 67
5.15 Effect of Variations of the Coefficient, a, on the
Exponential Term Relating to the Inlet Partial Pressure
of S02, exp (oPS(J ) ................... 68
B Effect of Slurry pH and Flow Rate on the Overall Mass
Transfer Coefficient for C02 Absorption into Recycled
Lime Slurries in TCA Scrubber .............. 104
-------
List of Tables
Number
3.1 Summary of Equations Necessary for Simulating
the Pressure Drop Across a TCA 22
3.2 Range of Data Used in Testing the Validity of
the Pressure Drop Correlation 23
4.1 Summary of Equations Necessary for Simulating
the SC>2 Absorption of the TVA Shawnee TCA and
Spray Column and the EPA In-House TCA 26
5.1 Sensitivity of Parameter Accuracy on the S0_
Removal Efficiency 69
VI
-------
List of Abbreviations and Symbols
A Cross sectional area of the TCA, m.
a Interfacial area per unit volume of bed, m2/m .
A , A Pre-exponential factor in the expression for the
ratio of the gas to liquid film mass transfer
resistance for the packed and spray sections,
dimensionless.
C Concentration of the total carbon dioxide in the slurry,
gmole/liter.
Cabs C02 absorbed per unit volume of scrubbing slurry,
gmole/liter.
C° Concentration of H2C03 in the bulk liquor phase,
gmole/liter.
C^° Concentration of l^COj in the slurry at the interface
of gas-liquid film, gmole/liter.
Caf Lime fed rate, gmole/sec.
Cx Concentration of magnesium or chloride in the liquor
phase, gmole/liter.
Cz Concentration of total carbon dioxide at a height of Z,
gmole/liter.
D Equivalent diameter for free sectional area, meter.
d Diameter of each hole in the grid, m.
dp Diameter of the packing sphere, meter.
F Feed rate of magnesium or chloride into the system,
gmole/sec.
F£ o Lime feed rate, gmole/sec.
3.
f Fraction opening of the grid, dimensionless.
6 Gas flow rate based on the cross-sectional area of the
scrubber at 0°C, m/sec.
G Gas flow rate based on the cross-sectional area of the
scrubber, Kg/m2sec.(=1.2946 G)
-------
g Gravitational acceleration, m/sec.
H* Concentration of free hydrogen ion, gmole/liter.
H Average concentration of free hydrogen ion in the
scrubber, gmole/liter.
K Equilibrium constant for H2C03, liter/gmole.
K' Proportional constant, Kg/mole.
K" Proportional constant defined as the product of K1
and M, liter/gmole.
Kga Overall gas side mass transfer coefficient for C02
absorption, gmole/m3atm sec.
k^a Gas side mass transfer coefficient for the packed
section, gmole/m^atm sec.
k|a Gas side mass transfer coefficient for the spray section,
gmole/m^atm sec.
L1 Liquid flow rate, liter/sec.
M Liquor content in the purge, based on the dry solid,
liter of liquid/Kg of solid.
MQQ Absorption rate of C02 in the scrubber, gmol/sec.
MSO Absorption rate of S02 in the scrubber, gmol/sec.
Nco Molar flux of C02 across the gas-liquid interface,
2 gmole/nrsec.
Ng Number of grids.
P Purge rate, liter/sec.
Pa Pascal, N nf2.
PT Total pressure, atm.
APb Pressure drop in the packing section, Pa.
&Pf Friction loss in the packing, Pa.
AP Pressure drop across the grids, Pa.
AP^ Liquid holdup in the packing section, Pa.
AP Static bed weight, Pa.
viii
-------
Normal pressure drop across the scrubber, Pa.
APg Pressure drop across the spray section, Pa.
APt Total pressure drop, Pa.
Partial pressure of (X^ in the bulk gas phase, atm.
Partial pressure of C02 at the gas-liquid interface, atm.
U2 » 1
*
PCO Partial pressure of C02 in the gas phase that is in
2 equilibrium with ^COj in the bulk liquid phase, atm.
Pgj-jn Inlet partial pressure of S02 in the bulk gas phase,
2 atm.
ps8Ut Outlet partial pressure of S02 in the bulk gas phase,
2 atm.
R^ Mean hydraulic radius defined as the ratio of cross-
section of the flow and the wetted perimeter, meter.
S Henry's law constant, atm/gmole-liter.
T Liquid temperature, degree Kelvin.
2
V Liquid flow rate, liter/m sec.
2
V Liquid mass velocity, Kg/m sec.
Z Height of the scrubber, meter.
Zo Total height of the scrubber, m.
Zp Height of the packing section in the TCA, m.
Zs Height of the spray section in the TCA, m.
Zpe Equivalent packing height, meter.
ZpT Total equivalent packing height, meter.
Aout (S)0ut + (C)
purge v"
A Magnesium correction factor for A_ and As, dimensionless.
-------
P Density of the packing sphere, Kg/m .
e Voidage of the static packed bed (packings are randamly
arranged), diraensionless.
p. Density of the slurry, Kg/m .
LI
PW Density of the water, Kg/m .
a Coefficient on the exponential term relating to the
inlet partial pressure of SO-, exp(oPcn . ), atnr1.
fc 2HJ ry 9 1H
-------
CHAPTER 1
INTRODUCTION
The combustion of coal is accompanied by the two major sources
of environmental pollution, sulfur dioxide and ash. Considerable
research and development efforts have been spent to make use of the
vast resources of coal available in the states and to meet the
enviromental limitations.
Two alternatives being considered for protecting the environ-
ment from the consequence of an extensive use of coal are:
(A) Reduction in sulfur dioxide and particulate emissions
from new and existing power plants and other facilities.
(B) Conversion of coal into a pollutant-free and usable
liquid or gas which is convenient to handle, transport, and
utilize in the final energy consumption stage.
1.1 Processes for Desulfurizing the Flue Gas
Of more than 50 gas desulfurization control concepts (Nelson
1974) which have been proposed and studied, the major routes can be
classified into three categories: amines, metal oxides,and alkaline
solutions. Complete descriptions are given by Woodies et al. (1973),
Shale et al. (1971), Strauss (1972), Berkowitz (1973), LaMantia et
al. (1973), Mcllrg et al. (1973), and Nannen et al. (1974).
The lime/limestone wet-scrubbing system is considered to be
one of the viable ways to reduce stack gas emission, because it
employs least expensive reactants, is less sensitive to
-------
operating conditions,and requires no complex control or regeneration
equipment.
1.2 Description of the Wet Lime/Limestone Scrubbing Systems
The Environmental Protection Agency (EPA) through its Office
of Research and Development and Control Systems Laboratory, sponsored
a program to test the wet lime and limestone scrubbing system for
removing sulfur dioxide and particulates from flue gases.
A pilot and a prototype wet scrubbing facility for removing
sulfur dioxide are set at EPA Research Triangle Park (RTP) and the
Tennessee Valley Authority (TVA) Shawnee Power Station respectively
to test the reliability and performance of the systems. The
schematics of these facilities are shown in Figure 1.1 and 1.2.
In these wet-scrubbing systems, the particulates (fly ash) are
captured by liquid droplets while sulfur dioxide is absorbed by
the scrubber into the lime/limestone slurry where it reacts with
the dissolved lime/limestone, forming calcium sulfite and calcium
sulfate (gypsum)•
The holding tank which receives the scrubber effluent provides
enough time for the precipitation of calcium sulfite and calcium
sulfate. The precipitants are then purged out by the vacuum filter.
For a limestone wet scrubbing system, fresh limestone slurry is fed
into the holding tank as the reactant.For a lime wet scrubbing
system, the unslaked residue of lime slurry is discarded and the
slurry is fed into the scrubber-effluent holding tank (Borgwardt
(1974b)).
2
-------
Scrubber
Height: 8.4 m.
Cross Section:
1.7ra. x.l.7m.
Flue Gas
Lime
rocess
Water
Hold
Tank
Figure 1.1: Schematic Arrangement of the TVA Scrubber System.
-------
Gas Out
Scrubber
Height: 2.8 m.
Diameter: 0.23m.
Flue Gas
°bo
Hold Tank
1.3 m3
V
LFilter
PURGE
1
0.15 m
Lime
Slurry
Figure 1.2: Schematic Arrangement of the EPA Pilot Scrubber System.
-------
1.3 Objective of This Study
In order to analyze data on the scrubbing of flue gas using
lime , a simulation model of the system is very useful. Such a model
can be used to predict the SC^ removal efficiency, lime utilization,and
pH values at the inlet and outlet slurries of the scrubber. The
known variables are the lime feed rate, the size of the scrubber,
the recycling liquid flow rate, the composition and the feed rate
of flue gas. The first objective therefore is to simulate the scrub-
ber hold-tank flue gas desulfurization (FGD) process by formulating
the mass balance equations across the scrubber and the whole system.
Each term in these equations, namely, the absorption rates for 862
and CO-, and the concentrations of total dissolved sulfur plus carbon
minus calcium in the inlet and outlet recycling liquors of the scrub-
ber must be determined separately.
The second objective is to determine the maximal rate of flue
gas that can be treated, as a function of liquid flow rate for a
specified lime feed and SO- removal efficiency, for a given size of
scrubber. The achievement of this objective can contribute greatly
to the effort which would be needed in the design of a scrubber unit
for a -specific application.
In this study, the wet scrubbing system with lime as the
recycled medium is simulated. Data are taken from the EPA inhouse
turbulent contacting absorber (TCA) at Research Triangle Park, North
Carolina and the TCA scrubber located at the TVA Shawnee Power
Station.
-------
CHAPTER 2
LITERATURE REVIEW
The turbulent bed contact absorber (TCA) with movable packing,
first described in 1959 for removing particulates from a dust laden
gas (Douglas et al. (1963)), has been recently used in desulfuriza-
tion of flue gas (Douglas et al 1964, Kielback et al. 1959). It
consists of large diameter uniform spheres of low density placed
between retaining grids sufficiently far apart to permit turbulent
and random motion of the packings. Hollow polyethylene, foam
polystyrene,and thermo-plastic rubber spheres have been found to be
satisfactory for this purpose. Because of the low density and the
counter-current flow of liquid and gas, TCA provides a state of
vigorous contacting between liquid and gas.
This equipment has some advantages over a conventional gas-
liquid contactor with fixed packings. The motion of the packing
prevents•plugging and by-passing which may occur when gases and
liquids containing suspended solid particles are used. Rates of
heat and mass transfer have also been reported to increase owing
to the bed agitation (Douglas 1964). The use of TCA also permits
much greater gas and liquid velocities than are possible in con-
ventional scrubbers. Thus, a smaller TCA tower may be employed for a
given operation compared to other conventional scrubber.
-------
Result of the investigation by Gel'perin C1965) shows a strong
dependence of liquid holdup,bed expansion and pressure drop on the
liquid and gas flow rates. Blykher (1967) has also studied the pres-
sure drop in TCA. However, these investigators did not report the
pressure drop across a mobile bed, but the pressure drop
data reported included the contribution of the bottom grid with the
free cross-sectional area in one case as small as 19%.
Levsh (1968) studied the pressure drop of TCA with low density
packings and found that pressure drop increases linearly with the
gas flow rate. Tichy (1972) correlated the pressure drop data with
the well-known Fanning equations. A substantial wall effect was
observed by Tichy (1972) in his small scale scrubber (0.14-m. dia.).
Epstein (1975) has run a series of experiments by passing air
and sodium carbonate solution through a large TCA scrubber (1.7 m.
square) at TVA Shawnee power station to observe the pressure drop
without the presence of scaling. When lime or limestone slurry is
used, deposition of calcium sulfite and calcium sulfate on the wall
and grids of the scrubber takes place. An empirical correlation of
pressure drop was presented by Epstein (1975).
Liquid holdup was measured and correlated by Groeneveld (1967),
Chen and Douglas (1968), Barile and Meyer (1971), and Kito et al(1975),
-------
Groeneveld ( 1967 ) measured the U
-------
Recently, Borgwardt (1972b) and Epstein (1970) studied the
absorption of S02 by alkaline solutions and limestone/lime slurries
in TCA, venturi scrubbers^and marble bed scrubbers.Their data from the
TCA scrubber demonstrated a significant improvement in S02 removal
efficiency over the conventional packed tower. McMichael et al.
(1975) presented a mathematical model to describe the absorption
of S02 in TCA. Later, Fan (1975) successfully correlated the
data for TCA and spray tower units by considering the combined effects
of spray section and packing section which are affected by the
hydrodynamics, the magnesium concentration and the pH value of the
scrubber.
Kito et al. (1975) measured the gas-liquid interfacial area
and gas mass transfer coefficient in a TCA operated with stagnant
liquid. Further, the specific gas-liquid interfacial area in TCA
was investigated by Groeneveld (1967). According to his study, the
specified interfacial area was proportional to the liquid flow rate,and
increased with increasing gas flow rate. A slow increase was observed
below the flooding point and a rapid increase at the flooding point,
where the interfacial area reached a value of 200 m /m3.
-------
CHAPTERS
PRESSURE DROP IN THE TURBULENT CONTACTING ABSORBER
In this chapter, an analysis of pressure drop will be made on
data reported by Epstein (1975) . Since a large diameter scrubber
was used, the correlation obtained could be used for the design
of commercial sized scrubbers.
3.1 Model of Pressure Drop in TCA Scrubber
Typical TCA configurations to be considered in this study are
shown schematically in Figure 3.1 and 3.2.
The pressure drop data of TVA Shawnee scrubber (1.7 m. square)
for the air/water and sodium carbonate runs and four additional
limestone runs with 0.76 m. static packing height are summarized in
Figure 3.4 and 3.5. Data from the EPA small scale (0.23 m. dia.)
TCA scrubber with lime slurry as the scrubbing liquor are also
presented in Figure 3.4.
For the TCA operating with packing spheres, the column can be
divided into spray, packed, and grid sections as shown in Figure 3.3.
The following equation is proposed for determining the pressure drop
across the TCA scrubber:
APt = APg + APs + APb (3.1)
where APt is the total pressure drop in Pascals, Pa,
AP is the pressure drop across the grids, Pa,
APg is the pressure drop in the spray section, Pa,and
APfe is the pressure drop in the packing section, Pa.
10
-------
GAS OUT
CHEVRON
DEMISTER
INLET KOCH
TRAY WASH
LIQUOR
-------
GAS IN
GAS OUT
(To Demister)
o o
oo
oov
INLET SLURRY
MOBILE PACKING
SPHERES
RETAINING GRID
SO cm
Approx. Scale
EFFLUENT SLURRY
Figure 3.2: Schematic of the EPA/RTP Research TCA Scrubber
12
-------
EXPANDED
PACKING
HEIGHT
GRID
\
LIQUID
FLOW
PACKING
SPHERES
o-o-oo°c9(9o0
oooooooo
GRID
1
Z = HEIGHT OF SPRAY
S SECTION
^p = Un-EXPANDED HEIGHT
I OF PACKED SECTION
GAS FLOW
Figure 3.3:
An Idealized Stage of a Turbulent Contacting Absorber.
13
-------
4500 _
Epstein's Data (1975 a)
Packing
A
Liquid
4000 -
Screens Height (m) Flow (Kg/nTsec)
O 4 0.508 13.5
A 4 0.508 21.0
Q 4 0.508 27.5
3500
3000
2500
rt
Q.
C
•1-1
< 2000
i
ti
o
1 1500
V)
o>
Q.
2 1000
500
1
~ T o 0.762 27.5 • 1
1 /
Borgwardt'.s Data (1974 b) 1 /
~^> IB
4 screens X^ j 1 /
Z = 0.762 m ** ^ f ,
V = 31.5 Kg/m2sec j ' /
• /
1 /
T/ / j /
r x^ ''' *\ /°
x ^ i /
- . --^ ^'
Q """" xx
x- ^
^ -*
1 i i 1 I
.5 2.0 2.5 3.0 3.5 4.0
Gas Flow Rate, Kg/m2sec
Figure 3.4: Effect of Gas Velocity and Liquid Velocity on Pressure
Drop for TCA.
14
-------
in
o>
c
2500-T
2000
1500
H
a
1000
cd
4J
Epstein's Data (1975 a)
Packing Liquid
Screens Height (m) Flow Rate (Kg/m sec)
• 2 0.254 0.0
500
A 2
0 2
E 2
0.254
0.254
0.254
9.0
18.0
27.0
/
m
T
/
X
o
o- —
-A -AA'
0.0
Figure 3.5:
0.5
1
1
1
1
'
1.0
3.0
3.5
1.5 2.0 2.5
Gas Flow Rate, Kg/m2sec
Effect of Gas Velocity and Liquid Velocity on Pressure Drop for TCA.
4.0
-------
Effect of the Grid:
The pressure drop across the grid, AP as a function of liquid
and gas flow rates is shown in Figure 3.6. The following empirical
eauation is obtained when a first-order dependence on each of the
variables is assumed so as to approximate AP which is a small
part of the total pressure drop.
AP = 0.579 G V N (3.2)
" 2
where G is the superficial gas mass velocity, Kg/m sec,
* 2
V is the superficial liquid mass velocity, Kg/m sec,
and Ng is the number of grids.
Effect of the Spray Section:
The pressure drop of the spray section of the Shawnee TCA has been
correlated by Wen (1973) in terms of the gas and liquid flow rates
as
A1 17 *fl ft
AP = 1.79 G •* V Z (3.3)
S 5
* 2
where G is the gas mass velocity, Kg/m sec,
and Z is the height of spray section, m.
5
Equation (3.3) will be used in this study.
Effect of the Packing Section:
A TCA scrubber utilizes very low density packings. The gas flow
rate is increased in a TCA scrubber at a constant liquid flow rate
until the upward force of the gas flow balances the weight of the
packings plus the liquid holdup, which are both .equal to the total
pressure drop.
The pressure drop of the bed sections, AP, , is thus assumed to
D
have the following form:
16
-------
400
cfl
a.
•S
0.°° 30°
<
ft
P*
g 200
a
l/>
VI
£
*• 100
TJ
i
n
Data, of
Epstein (1975 a)
Liquid Flow
Screens Rate (Kg/mzsec)
- o 2
• 2
D 4
" • 4
1
13.0
26.0
13.0 ^^\^
26.0 ^-
- — "
• ^ ' —
. • _ . — • —
^- _^ - — - QO
- — -
o . - — '
II 1 1 1 1 1
0
0.5
1.0
3.5
4.0
1.5 2.0 2.5 3.0
Gas Flow Rate, Kg/m2sec
Figure 3.6: Pressure Drop Across the Grids as a Function of Gas and Liquid Mass Velocity
in the Shawnee TCA Operated Without Packing Spheres.
-------
AP. = AP, + AP + AP. (3.4)
o n p r
where AP, is the pressure drop due to the liquid holdup in the
packing section, Pa,
AP is the pressure drop due to the static bed weight, Pa,
and APf represents the pressure drop due to the friction loss as
the operating conditions are close to the flooding point,
Pa.
A few investigators, such as Barile et al. (1971), Chen et al.
(1968) and Kito et al. (1975), have correlated the results of the
liquid and gas holdups of a TCA scrubber. Their investigations have
shown that there is a strong dependence of liquid holdup on liquid
velocities,on the characteristics of the supporting grid, and on
the properties of the packing spheres.
Although Chen and Douglas (1968) presented a correlation for
the pressure drop in a TCA scrubber, the effects of the packing
properties and the characteristics of the supporting grid on the
liquid holdup were not considered. The correlation of Barile and
Meyer (1971) on the other hand is applicable only at the minimum
fluidization velocity.
Kito et al. (1975) studied the liquid holdup by considering
the effects of liquid velocity, diameter and density of the packing
spheres^ and the static height of the bed. Since their correlation
includes the effects of packing properties and characteristics of
the supporting grid, it will be used in this study. Their correla-
tion has the following form:
18
-------
APh = 0.024 (f £) ' (dp.)''84 CPp
-------
where AP. is obtained from Epstein's data while AP , AP AP. ,
t K s n
and AP are calculated from Equations (3.2), (3.3), (3.4), and (3.7)
respectively.
The values of APf obtained from Equation (3.9) are shown in
Figure 3.7 as a function of the liquid and gas mass velocities. The
result shows that AP- is negligible for all the gas mass velocities
tested in the experiments if the liquid mass velocity is less than
8.0 kg/m sec.
3.2 Simulation of the Pressure Drop across TCA Scrubber
Table 3.1 summarizes the correlations which can be used to
simulate the pressure drop across a TCA.Based on the model presented
in this study the pressure drop across the TVA Shawnee TCA can be
computed fairly accurately, in most cases within 10% accuracy. A
comparison of the calculated and observed pressure drop is shown
in Figure 3.8 for TCA units. Data for EPA's small scale TCA
scrubber are excluded since they varied over a large range at the
same gas and liquid velocity as shown in Figure 3.4.
20
-------
CO
Gi E
4000
3500
3000
2500
2000
1500
1000
500
0 0
Epsteinjs Data (1975 a,b)
Parameters: Gas Mass Velocity, Kg/m sec
10
20
30
40
Liquid Mass Velocity (Kg/m sec)
50
Figure 3.7: Pressure Drop for the Packed Section as a function of Gas and Liquid Flow Rates
-------
Table 3.1
Summary of Equations Necessary for Simulating The Pressure
Drop Across a TCA:
APt = AP + APg + APh + AP + APf (3.8)
AP = 0.579 G V NG (3.2)
AP = 1.79 G1-17 V0'6 Z (3.3)
5 S
>.024 (f £
+ 147.1 Z p /p (3.5)
p L w
APp = 5.06 Zp Pp (3.7)
AP is given in Figure 3.7
22
-------
Table 3.2
Range of Data Used in Testing the Validity of the Pressure Drop
Correlation
Data are for TVA Shawnee TCA Using Water-Soda Ash and Limestone as
Scrubbing Media (Epstein (1975 a,b))
Gas Flow Rate (kg/m2sec) 1.5 to 4.2
Liquid Flow Rate (kg/m2sec) 0 to 34
Packing Height (m) 0 to 0.76
Number of Grids (-) 0 to 6
Equivalent Diameter of the
scrubber (m) 0-084 to 0.0925
Packing diameter (m) 0.0097 to 0.029
Diameter of the Hole in
the Grid (m) 0.002 to 0.012
Packing Density (kg/m3) 170 to 1250
Fraction Free Opening of the
Grid (-) 0.5 to 0.84
23
-------
Epstein (1975 a) Epstein (1975 b)
Liquid Flow Rate
0 to 26
19 to 34
tvuu
3500
3000
to
o.
e
•H 2500
4-1
O.
<3
f£
^ 2000
£
3
l/J
8 1500
Q.
i— l
cd
•P
°
£ 1000
cd
|
cd
u
500
n
_^ivg/iu acuy w «.w «.« — -— -
Gas Flow Rate
(Kg/m2sec) 1.6 to 4.1 2.7 to 4.2
-Bed Height (m) 0 to .76 .38 to .57
Type of Grid Mesh Screen Bar Grid .
t~
/v
<'/,
/ / /
' / A^A
&i'+
>i&^
r^V
/ ^5w ^
XVfflPX
'jj?/ °
ifi$^^/
? i^V™/
f^^Cf
/Gnr/ Dotted lines indicate the bound
fQ& ' of ± 10% Error
cfoy^
p9v t>
f^^fvf , /
^
/Q^
^
/^ J 1 1 1 1 1
500 1000 1500 2000 2500
Measured Total Pressure Drop, APt, in Pa
3000
Figure 3.8: Comparison of the Predicted and Observed Pressure Drop
Across TCA Scrubber.
24
-------
CHAPTER 4
ABSORPTION OF SOg AND COg IN TCA
Flue gas containing SO- and CO- passes through the EPA scrubber.
Upon absorption of S02 and CO- from the flue gas, sulfite, sulfate
and carbonate salts of ca'lcium begin to form and the pH of the scrub-
bing lime slurry drops rapidly greatly reducing the amount of CO
absorbed.
In this chapter, the effect of pressure drop on S0_ scrubbing
efficiency and the effect of CO- absorption in flue gas scrubbing
are studied.
4.1 S02 Absorption—Effect of Pressure Drop on the S02 Absorption
Efficiency
McMichael et al. (1976) reported a procedure by which the sulfur
dioxide removal efficiency of a TCA scrubber can be calculated from
the specification of the scrubber characteristics and inlet slurry
composition. This procedure is based on the following equation;
p in s p
G_ SQ2 fcgaZs kga \
P ln out ' 330P in * 330P in (4 1}
T PS02 As e S02 A e b°2
AS p
Correlations for the parameters appearing in this equation are given
in Table 4.1.
Equation (4.1) can be simplified by converting the height of
spray section into an equivalent height of packing. The equivalent
packing height, Z e, is defined as the height of packing which gives
the same amount of absorption of SO- as the absorption in the spray
25
-------
Table 4.1
Summary of Equations Nessary for Simulating the SO- Absorption of the TVA Shawnee TCA and
Spray Column and the EPA In-House TCA (McMichael et al. 1976)
N>
kSa=0.1586G°-8V°-4
S
exp(-1.35 pH + 7.82)- 0.15
= 1.188 G°'47 V°-51
50.1 Mg
0
"
; for Mg * 350 ppm
A = 1.0 ; for Mg < 350 ppm
g
A"1 = -0.417 pH + 3.41 ; for pH > 6.0
A"1 = 0.308 ; for pH < 6.0
P
A = 2.2 x 107 Mg"2*065 ; for Mg * 3600 ppm
A = 1.0 ; for Mg < 3600 ppra
where the pH refers to the log mean hydrogen ion concentration across the scrubber.
-------
-— •"
D,
N
«J
a tm
^— ^
^
c e
•H O
CO
(X
5.0
4.0
^— -*X
c CM 3.0
•H O
CO
0
CO
co o n
-------
section at a height of Z . Thus,
«»
kSa ! + B 2
-, e <• C -. A
ZP = ZS ( P ) _ E _ (4.2)
kga A 330PSOin
.5 ^
Equation (4.1) can then be written as
. X" "»" ZPT
5- In - — - - * — — - .- (4.3)
where ZpT = Zp + Zpe
Equation (4.1) was derived from the data of flue gas desulfuriza-
tion (FGD) in TCA and spray columns (see McMichael (1976)) without
regard to the effect of the pressure drop across the columns. It is
known that as the scale is formed the pressure drop across the column
increases and the sulfur dioxide removal efficiency increases. Pre-
sumably, this is due to the increase in the interfacial area available
for mass transfer as a result of an increase in the liquid hold-up.
Equation (4.3) can be corrected to take into account the effect of
pressure drop across the scrubber on the sulfur dioxide absorption
efficiency. Figure 4.1 shows this effect based on Borgwardt's data
(1974b,d). From Figure 4.1 the scrubber equation (Equation (4.3)) can
be revised to include the pressure drop effect as follows:
28
-------
p ln p
S°? AP 1'1 kaS Z
The normal pressure drop without scale formation, AP , (Equation
(4.4)) is difficult to calculate as discussed in Chapter 3. The
values of AP,, used in calculation of the data points in Figure 4.1
and in formulation of Equation (4.4) were based on the lowest pressure
drops reported by Borgwardt (1974 b,d) at a given set of flow
conditions.
4.2 C02 Absorption
In TCA, carbon dioxide is absorbed from flue gas into the
scrubbing medium, lime slurry, and in turn is precipitated in the
holding tank as CaCO according to the following reaction:
O
CO. + CaO = CaCO_
Calcium carbonate is then purged out from this wet-scrubbing system
with solid calcium sulfite and calcium sulfate as a waste. As a
result of the recarbonation of lime and precipitation of calcium
carbonate, the utilization of lime, is reduced.
The lime utilization is defined as follows:
Utilization of lime = ™}es *>* absorbed
moles CaO fed
moles S02 absorbed
S02 absorbed + C0_ absorbed + slaking loss
In the desulfurization scrubbing system, the absorption of carbon
dioxide into the scrubbing medium is undesirable since the precipitation
29
-------
of calcium carbonate lowers the usage of lime. For the purpose of
designing a scrubber,the utilization of lime must be predicted. This
in turn requires that a model for the absorption of carbon dioxide
from flue gas be developed.
4.2.1 C0_ Absorption in the absence of SO-
The absorption of carbon dioxide into an aqueous alkaline solution,
such as KOH and NaOH, is a process which has been studied by a
number of investigators.
When carbon dioxide is absorbed in the absence of S02 in an
alkaline solution, it reacts according to the following reactions
as proposed by Payne and Dodge (1932).
C02(g) = C02(l) (a)
C02(l) + H20 = H2C03 (b)
H2C03 = H+ + HC03" (c)
HC03" = H+ + C03= (d)
H* + OH" = H20 (e)
C02 + OH" = HC03" (f)
HC03" + OH" = C03~~ + H,0 (g)
C02 + 20H~ = C03= + H20 (h)
Although the ionic reactions are known to be very rapid, the
rates of the other reactions are not well known.
Various assumptions as to which of these reactions may be con-
trolling lead to different mechanism of the absorption process.
Hatta (1928) assumed that the reaction (a), (f), and (g) are con-
trolling and that (g) is much more rapid than (f).
30
-------
Eucken and Grutznei1 (1927) concluded that the following
reaction:
CO 2 + 20H~ = C03~~ + H20 (h)
was the major reaction.
Tepe and Dodge (1943) have reported their experimental study
of the absorption of carbon dioxide by sodium hydroxide solutions
in a 0.15 m.-diameter column filled to a height of 0.91 TO. with
0.0127 m.(0.5 in.) carbon Rashig rings. The overall mass transfer
coefficient K.,a was found to be a function of concentration of
b
sodium hydroxide. Changes in the gas flow rate was found to have
a negligible effect on K.,a. The value of K,,a increased in propor-
u u
tion to the liquid temperature.
A comprehensive investigation of CO- in an alkali solution has
been carried out by Nijsing (1969) . He used two different absorbers,
a laminar jet and a wetted-wall column, to study its mechanism. The
gas phase was pure CO at pressure from 20 KPa (0.2 atm) to
101 kPa (1 atm) and the liquid phase consisted of concentrated
hydroxide solutions (0.5 to 2.0 gmole/liter).
From the studies cited above, Astarita.. (1967) has drawn the
following conclusions:
(1) The absorption of CO-into an alkaline solution is a process
of chemical absorption.
(2) The overall absorption coefficient is rather insensitive
to the gas flow rate, which clearly indicates liquid-
side mass transfer control.
31
-------
(3) The overall gas absorption coefficient increases with
liquid flow rate.
(4) The overall gas absorption coefficient increases with
increase in the bulk-liquid concentration of the reacting
solute, namely of OH ion.
However, the absorption of CO, in the liquid medium accompanied with
the absorption of S02 has not been investigated to date.
4.2.2 GO, Absorption in the presence of S02
The absorption of C0_ into recycled lime slurries in TCA
scrubbers is a complex problem for several reasons. Firstly, the liquid
film mass transfer coefficients for TCA scrubbers have not been
reported. This fact makes the analysis of CO- absorption difficult
in that mass transfer and chemical effects cannot be isolated from
each other. Secondly., upon absorption, C02 hydrates can participate in several
reactions. Absorption of CO. into recycled lime slurries, which
contain various sulfur, magnesium, chlorine and carbon compounds,
is not simple f even though the studies on absorption of C0_ into
water or sodium hydroxide solutions have been well documented in
the literature. And finally, detailed data on the absorption of 00- into
recycled lime slurries in TCA scrubbers have not been reported.
Only qualitative information on the inlet and outlet streams of
the scrubber and the scrubber-hold tank system is available.
In this section experimental data reported by Borgwardt
(1974b,d) , oto a TCA scrubber at the Research Triangle Park,
are analyzed. A mathematical model describing the absorption of CO
32
-------
from flue gas in the presence of SO- into lime slurry in a TCA
scrubber is developed.
Model of CO- Absorption in TCA:
Both SO- and CO- are absorbed in a flue gas scrubber by the
recycling lime slurry. It is a competing reaction for a common
liquid phase reactant, OH . The pH values along the TCA scrubber
may decrease more drastically compared to the case of CO- absorption
in the absence of S02.Based on the experimental observation reported
by Borgwardt (1974b,d), the reduction in the value of pH is usually
from about 10 'at the top to about 5.0 at the bottom of a TCA.
The significant drop in pH value as a result of the presence of
SO- in flue gas greatly reduces the rate of absorption of CO- or
the value of its mass transfer coefficient. In addition, the
following assumptions are made in developing this model!
(1) Negligible precipitation of calcium carbonate in the scrubber.
(2) Negligible slaking loss which was experimentally shown to be
only about 5% of the total lime feed.
(3) In the bulk liquid phase, the following reaction is at
equilibrium:
H2C03 = H+ + HC03"
The concentration of carbonate ion is appreciable only when pH
is greater than 10.
(4) The change in the partial pressure of CO- (approximately 30 Pa
(or 0.0003 atm)) in the bulk gas phase can be ignored since the
amount of CO absorbed by the scrubbing liquid is very small
compared to the CO- content in the bulk gas phase. Even though
33
-------
the change in the partial pressure of C02 in the bulk gas phase is
Mia 11, the CO 2 absorption can not be ignored since it is critical in
determining the calcium utilization^
A schematic drawing of the scrubber and a depiction of the
driving forces for carbon dioxide transfer across the gas liquid
interface are shown in Figure 4.2. For this sketch, the following
nomenclature are employed.
V = liquid flow rate, liter/m sec
Z = height of the scrubber, m
dZ = differential tower height, m
C = concentration of the total carbon dioxide in the slurry,
gmol/liter,
C. = concentration, C, at the inlet slurry of the scrubber,
gmol/liter,
C = concentration, C, at the outlet slurry of the scrubber,
gmol/liter,
P__ = partial pressure of carbon dioxide in the bulk gas phase,
uu2
atm,
PCO i ~ Part*al pressure of carbon dioxide at the gas- liquid
interface, atm .
The molar flux of carbon dioxide across the gas-liquid interface
of the scrubber can be written in terms of the overall gas phase resis-
tance.
N002 = V CPC02 - Pco"2> (4.5)
where N_0 is the molar flux of carbon dioxide across the gas-liquid
2
interface, gmol/m sec,
34
-------
tn
C0
^
G
1 ,
V
I
t
G
1
V
t
r
dZ
t
Z =
W
v, c,
in
-C, Z+flZ
v
V)
(a) Flow Diagram of Scrubber
Gas Phase
Slurry Phase
' 9?s (Liquid
lFllm \P11«
p r°
Fco2' c *
I
(b) Concentration profile-
Figure 4.2: Scrubber and Idealization of the Concentration Profile of Carbon Dioxide.
-------
KQa is the overall gas phase resistance, gmole/atm.sec.m
Pm is the partial pressure of carbon dioxide in the bulk
^2
gas phase, atm,
and P * is the partial pressure of carbon dioxide in the gas
LiO ^
phase that is in equilibrium with H-CO- in the bulk
liquid phase, atm;
*
For a dilute system P™ can be defined by the Henry's Law as
Pco2 - s c° 2 (4'6)
where S is the Henry's Law constant, atm/gmole/ liter
C° is the concentration of H9CO in the bulk liquor phase,
£• j
gmole/ liter.
Astarita (1967) concluded in his work that the absorption of
C0_ into alkali solution is a liquid phase controlled reaction. Thus,
P°, may be related to the interfacial concentration of H_CO by the
CU A £• J
expression
>co2 • s ci° t4-"
where C.° is the interfacial concentration of H_CO, in gmole/liter;
1 i 3
The Henry's Law constant, S, as a function of temperature has
been given by Lowell (1970) as
S = EXP (11.215 - ') (4.8)
where T is the liquid temperature in degrees Kelvin.
The molar flow of carbon dioxide through the gas-liquid interface
can be obtained in terms of liquid phase concentration by substituting
Equations (4.6) and (4.7) into Equation (4.5).
Noo2 ' V s (ci° - C^ «-V
36
-------
By applying the mass balance on carbon dioxide across a differen
tial height of the scrubber, dZ, the rate of CCL absorption can be
written
V+ KGaS (C.° - C") = 0 (4.10)
2 3
where a is the interfacial area per unit volume of bed, m /m .
Since equilibrium between carbon dioxide and the bulk liquor
phase is assumed and since the concentration of the carbonate ion
is negligible, the following reaction is the key reaction that takes
place in the bulk solution:
H2C03 = H+ + HC03"
Hence, the concentration of bicarbonic acid is given by
K (H CO )
HCO " = - 1—2- (4.11)
(H*)
where K is the equilibrium constant in liter/ gmol.
This equilibrium constant, K, has been given by Lowell (1970) as a
function of temperature as follows:
K = EXP (- IY^- - 0.075506 T + 34.183) (4.12)
Thus, the total carbon concentration, C, in the liquor can be obtained
in terms of H-CO, and H by employing Equation (4.11)
£f O
C = HC0 + HC0
3
K (H CO )•
= H CO
2
37
-------
After rearranging,
H rn - (C) (H ) ,.
H2C03 - K + (H+) (4
In terms of the nomenclature defined in Equation (4.6), Equation (4.13)
is rewritten as
K + H+
Substitution of Equation (4.14) into Equation (4.10) gives
Ka s Ka S
(4.15)
H
The overall mass transfer coefficient, K_a, in this equation is a
function of pH value, liquid flow rate, and partial pressure of CO-.
When integrating Equation (4.15), KQa and H are taken as constant
and are designated as K_a and H respectively.
K a S
H
The boundary conditions associated with Equation (4.16) are
i 7 - n r — p
i. z - o c - cout
(4.17)
Since the magnitude of the change in the partial pressure of
carbon dioxide along the tower is in an order of 0.0003 atm, Pb
co2
can be treated as a constant. This, in turn, assures the constancy
of C^0 in Equation (4.16). The integration, from ZQ to Z, of
38
-------
Equation (4.16) satisfying Equation (4.17) is
V „ C
C = (C,°in - C^ £ + 1)) EXP(-| (Z - Z ))
H V(* + 1) °
H
+ C ° C=+ 1) (4.18)
1 H
where Cz is the concentration of total carbon dioxide at a height of
Z, gmol/liter.
Based on the assumption (1) carbon dioxide absorbed per unit
volume of scrubbing medium can be obtained
cabs = cout Cin
= C,Z=0 - C,Z=ZQ
(4.19)
where Ca^ is CO- absorbed per unit volume of scrubbing medium,
gmole/liter;
Substituting Equation (4.18) into Equation (4.19) gives
VS K
= (1 - EXPC- -TT Zo» 'V Cz+ 1)
VP + 1) * H
H
- Cin) (4.20)
In this equation, C. is the concentration of total carbon dioxide
in the effluent of the holding tank and is very small compared to the
other term, C.° (i + 1). Equation (4.20) can be rewritten in a
1 H
simpler formI
Kra S ,,
C u = (1 - EXP( jT^ Z0))(Z + !) C-° C4'21^
V(— + 1) H
H
39
-------
The interfacial concentration of H2C03, C.°, can be obtained by use
of Equation (4.7)
C.° = P^/S (4.22)
Substituting Equation (4.22) into (4.21) leads to
Cabs ' « - EX"f- -T7 Zo»
H
This equation is used to obtain the amount of carbon dioxide
absorbed from the flue gas. K^a and H can be estimated by the method
described in the following section.
Mass Transfer Coefficients;
Equation (4.23) can be arranged to give
~S . (4.24)
-
In the case of lime slurries the pH variation across the
scrubber is substantial, ranging from 8.0 at the
inlet to 4.8 at the outlet. With this large change in pH it is
not reasonable to assume that the inlet slurry pH characterizes
the behavior of the scrubber as usually done by previous investiga-
tors. As shown in Appendix B,it has been found that the overall mass
transfer coefficient of the lime scrubbing system can be correlated
fairly accurately by the model developed in this chapter when the
I
mean hydrogen ion concentration is calculated based on the following
arithmetic mean pH value:
4C
-------
(4.25)
H = EXP (-2.3
The results are shown in Figure 4.3. The data were correlated
by the following equation
KGa = V""J EXP(22.3 + 11.11 piy (4.26)
The power of V is apparently much higher than the usual 0.7 power for
a packing tower. This may be explained as follows!
Since Borgwardt's data (1974 b,d) available for the investiga-
tion of C02 absorption were obtained from a rather small scale
scrubber (0.229 m.-dia), a study of its pressure drop is conducted
in Chapter 3. Figure 3.4 in Chapter 3 displays the pressure drop
reported by Borgwardt (1974 b,d) along with Epstein's data from a
large TCA scrubber. As seen in Figure 3.4, the data were possibly-
in a regime near the loaSing point due to the steep slopes obser-
ved.
Groeneveld (1967) observed a rapid increase in the specific
point, and from the observation of Groeneveld (1967), it is not sur-
prising that the value of KQa increased in proportion to the 6.7
power of the liquid rate under the experimental conditions corres-
ponding to Figure 4.3.
Although Equation (4.26) shows that the overall mass transfer
coefficient is extremely sensitive to the liquid flow rate, the
magnitude of the C02 absorption compared to S02 absorption does not
vary greatly because of the extremely low solubility of C02 in the
scrubbing slurries.
41
-------
to
+J
g
•H
u
s
0)
8
fn
0)
1/1
2
0>
o
JJ 10
in
I
%:
-1
\ ^
8
10
-2
5.0
A A"
Borgwardt's Data
Liquid Flow Rate, liter/m sec
• 30.0
A 36.2
5.5
6.0
6.5
7.0
7.5
Average Slurry pH = (pHin + PHout)/2
Figure 4.3: Effect of Slurry pH and Flow Rate on the Overall Mass Transfer
Coefficient for CCL Absorption into Recycled Lime Slurries in a
TCA Scrubber.
-------
CHAPTER 5
SIMULATION AND DESIGN OF RECYCLED LIME
SCRUBBING SYSTEM
A simplified schematic of a scrubber-hold tank system is shown
in Figure 5.1. Flue gas loaded with SCL passes counter-currently to
recycled lime slurry flowing downward in a TCA scrubber. At the top
of the scrubber the slurry has a pH value in the range of 6.0 to 10.0.
The slurry effluent from the scrubber passes to the hold tank where
lime and make up water are added, and the high pH of the scrubbing
slurry is recovered. The solid loading in the scrubbing slurry is
approximately 10%. A portion of this slurry is fed to the solid
separation system from which a waste sludge is discharged. The clear
liquor produced in this step is recycled to the system.
In essence, the SO- absorbed from the flue gas is converted
to calcium sulfite and sulfate which are extremely insoluble in the
slurry. The precipitation of calcium sul'fate is one of the problem
areas in lime scrubbing in that calcium sulfate forms a hard, stubborn
coating on the process equipment. Methods of preventing sulfate
precipitation include reducing oxidation of sulfite to sulfate and
operating the system in the sulfate unsaturated mode. In this mode
calcium sulfate is incorporated into the crystal structure of calcium
sulfite. Borgwardt (1974 b) has discussed the sulfate unsaturated mode
of operation in lime scrubbing systems.
5.1 Material Balances
Given the inlet conditions of the slurry and flue gas, setting
the operating parameters for the scrubber and neglecting the small
43
-------
out
so
TCA
SCRUBBER
S0
Hold Tank
FILTER
WASTE SLUDGE
Figure 5.1:A Simplified Lime-TCA Flue Gas Desulfurization
System.
44
-------
change in the magnesium concentration across the scrubber, Equation
(4.1) (or equivalently Equation (4.4) represents a relationship between
two unknowns: 1) S02 removal efficiency and 2) outlet slurry pH.
To provide a. second relationship between these variables, McMichael
et al. (1975) proposed that sulfur and carbon balances around the
scrubber could be subtracted from the calcium balance to give
where A. and A are the concentration of total dissolved sulfur
in out
plus carbon minus the concentration of total dissolved calcium in the
inlet and outlet slurry of the scrubber, respectively, in gmole/liter.
It is observed that the C02 absorption and the "A" concentration can
be related to the pH in lime systems through equilibrium calculations
(Nelson (1974)). Equation (5.1) becomes
"id/1' •
-------
Equation (5.4) ignores the losses which experimentally have been found
to be only about 5% of the total lime feed.
A comment is needed at this point to explain why the rate of CO
absorption can be ignored in the development of Equation (5.2) but
not in Equation (5.4). Equation (5.2) was developed to provide
estimates of the SO. removal efficiency. Since the rate of absorption
of SO- is much larger than that of C02, we may neglect the C02
contribution in Equation (5.2) for convenience and still obtain
reasonable estimates. However, in Equation (5.4), the C02 absorption
cannot be ignored because it is critical in determination of the
calcium utilization.
The equilibrium calculations used in Equations (5.1) and (5.2)
are those of Nelson (1974). In this calculation several variables
must be specified. These include pH, CO. partial pressure and the
total concentration of dissolved magnesium and chloride. The total
sulfate concentration must also be specified. The procedure for
the specification of the total concentrations of magnesium and
chloride in the slurry are discussed in the following section.
5.2 Prediction of the Concentrations of Magnesium and Chloride in
the Scrubbing Slurry
The flue gas may contain chlorine compounds which are absorbed
in slurry of the TCA scrubbers. Magnesium is fed to the system in
the form of magnesium oxide along with lime. The concentration of
chloride and magnesium will build up and will not reach a steady state
46
-------
until the losses from the purge can be balanced.
Since chloride and magnesium have high solubilities in the slurry,
it may be assumed that their concentrations in the solid phase of
purge from the system are very low. Based on this assumption, the
following mass balance is formulated
P M Cx = F (5.5)
where P is the solid purge rate, Kg/sec,
M is the liquor content in the purge, based on the dry solid,
liter of liquid/Kg, of solid,
C is the concentration of magnesium or chloride in the
Ji
liquor phase, gmole/liter,
and F is the feed rate of magnesium or chloride into this system,
gmole/sec;
If we make an assumption that the solid purge rate is proportional to
the lime fed (Fr»0 (gmol/sec)] and since more than 95% of calcium in
the feed will be purged out as solid, Equation (5.5) can be written
in the following form
F " K'M FCaO Cx C5'6D
= K" FCaO Cx C5'7'
where K1 is the proportionality constant defined by Equation (5.6),
Kg/gmol,
K" is defined as the product of K and M, liter/gmol.
Rearranging Equation (5.7),
—E-=K»C C5.8)
FCaO X
A plot of C vs. p^— is shown in Figure 5.2. The line can be corre-
x CaO
lated by the least square fit of data.
CC1 or Mg ' 2'6 FCaO C5'95
47
-------
.3
00
Vi
0)
0)
r-l
o
§
•i
u
.2
,1
A Concentration of Chloride in TCA Slurry
O Concentration of Magnesium in TCA Slurry
D Concentration of Magnesium in Venturi Scrubber Slurry
CC1 or Mg " 2<6 FCaO
0 .01 .02 .03 .04 .05 .06 .07 .08
Figure 5.2: (Magnesium content in the lime fed + MgQ fed rate) or .(Cl fed rate), gmole/sec
Lime feed rate, gmole/sec
Concentrations of Magnesium and Chloride in the Recycle Slurry
.09
-------
Equation (5.9) can be used to predict the concentration of magnesium
in the slurry of a TCA and the venturi scrubber as shown in Figure
5.2.
5.3 Simulation
Using the procedures outlined above,the operation of a scrubber-
hold tank FGD system using lime slurry can be simulated according to
the following sequence of steps:
1) Specify values of the independent variables,such as PSCH*
PCO , gas and slurry flow rates, lime feed rate, size of the scrubber,
etc.
2) Assume the inlet pH to the scrubber.
3) Calculate SC^ removal efficiency and the outlet pH of the
scrubber by solving Equations (4.4) and (5.2) simultaneously.
4) Use Equation (4.23) to calculate the C02 absorption.
5) Determine whether Equation (5.4) is satisfied. If it is not,
assume another inlet pH to the scrubber and proceed from Step 3. If
Equation (5.4) is satisfied, then the simulation is complete.
The flow diagram for this simulation is given in Figure 5.3.
This simulation procedure has been applied to the Borgwardt's data
»
(1974 b>°0 .which were obtained from the- TCA scrubber- hold -tank system
utilizing lime slurry to desulfurize flue gas. The results of
simulating Borgwardt's data are given in Figures 5.4 and 5.5. It
can be seen from the Figure 5.4 that the S02 removal efficiency can
be predicted within 5% of the observed value. As shown in Figure 5.5
the inlet and outlet pH of the scrubbing slurry cannot be predicted as
49
-------
Read Data
<•
Calculate
\ Functions
A
Calculate the
mass transfer
ycoefficients for
SO
Guess
pHin = 7.5
(Calculate pHout,
302 and S02
absorbed
check
assumed SC>2
bs = calculate
value
Guess
pH in
Calculate
Utilization
Write Report
Figure 5.3: Flow Diagram for the Simulation of Wet Lime Scrubbing
Process.
50
-------
100
0)
-------
0)
3
r—I
at
a
•o
cd
3
O
iH
d
11
10
8
O PH at Scrubber Inlet
~ A pH at Scrubber Outlet
Op Dotted lines indicated
the .bound of ±10% error
- A
5678
Observed pH Value
10 11
Figure 5.5: Comparison of the Observed and Calculated pH
at the Entrance and Exit of the TCA Lime Scrubber.
52
-------
accurately as the removal efficiency.
In preparing Figures (5.4) and (5.5) the calculated concentra-
tions of chloride and magnesium were used in the equilibrium calcula-
tions. The concentration of the components are calculated by
employing Equation (5.9) assuming that the precipitation of the
magnesium and chlorine can be ignored. It is also assumed that
at the steady state, all chlorine in the flue gas is eventually
absorbed in the scrubber and can be balanced by the losses from
the purge. The sulfate concentration in the slurry is assumed as
the saturated value. Since the degree of saturation of the observed
sulfate concentration varies on average around its saturated value to
within 20%, the assumption of saturated sulfate is quite acceptable.
5.4 Design of Lime Slurry FGD Systems Using TCA Scrubbers
Using the simulation procedure discussed in the previous section
charts can be prepared with which lime slurry FGD system utilizing
spray or TCA scrubbers can be designed. In the remainder of this study
it will be assumed that there is no chlorine or magnesium in the
slurry to keep complications at ^ minimum.However, the methods
discussed here can easily be applied to cases where the scrubbing
slurry contains chlorine or magnesium compounds. The sulfate
concentrations in the slurry are determined by assuming sulfate
saturation. In a commercial application the sulfate levels will
probably be maintained only slightly below saturation because of the
expense of adding magnesium to the system to reduce the saturation.
Therefore, the saturated concentration of sulfate should be a
53
-------
reasonable approximation to the sulfate concentration in an actual
system.
The procedure, by which design charts can be constructed, is
given below:
1) Specify S02 removal efficiency, slurry flow rate, lime feed
rate, equivalent packing height, inlet S02 partial pressure and
pressure drop above the normal pressure drop.
2) Assume a gas flow rate.
3) Compute the S(>2 removal efficiency by the five step simulation
procedure given in the previous section.
4) If the S02 removal efficiency calculated in Step (3) does
not agree with the S02 removal efficiency specified in Step (1),
go to Step (2). If the calculated and specified S02 removal
efficiencies agree reasonably well, then the simulation is complete.
The result of carrying out this procedure is shown in Figure 5.6.
It can be seen from this figure that for a given lime feed rate
there exists a maximum in the amount of gas that can be treated.
The locus of these maximum points is presented in Figure 5.7 as
solid lines. Thus for a given slurry flow rate and equivalent
packing height the maximum amount of gas that can be treated (at
90% efficiency) and the lime stoichiometry can be read directly
from Figure 5.7.
A significant observation concerning the operation of a lime
scrubbing FGD system, which can be drawn from Figure 5.7 is that
for a given gas treatment rate a decrease in the specific flow rate
54
-------
in
o
CD
CO
9
a
0)
•p
a)
K
o
iH
fe
2.8
2.7
2.6
2.5
2.4
Parameters Lime Feed,
mo
m sec
Packing Height =0.95 m
SO2 inlet S02 outlet
2430 ppm 243 ppm
^-^
No Cl~ and MgO Fed
Average Pressure Drop = 2370 Pa (9.5 in. of
water)
27 28 29 30
Liquid Flow Rate,
31 32
liter
2
m sec
33
34
Figure 5.6:
Simulation of Lime Slurry TCA Scrubber Indicating Maximum
Flue Gas Flow Rates.
-------
tn
CO
B
o
0)
s
0)
•JJ
W
cd
O
0
3
rH
Pn
2.9
2.8
Syst^^L moles of lime fed
_ StoicTieometry - moies of SOQ absorbed
2.6
2.4
SOQ Inlet
ft
2430 ppm
outlet 243 ppm
Equivalent Packing Height
Average Total Pressure Drop
= 2370 Pa (9.5m of water)
Liquid Flow Rate,
m sec
33
Figure 5.7: Operating Lines for Lime TCA Scrubber at Various Equivalent Packing Heights
-------
of the lime slurry in the TCA scrubber decreases the "system stoicliio-
metry"Ci.e..increases lime utilization) but increases the equivalent
packing height necessary to achieve the prescribed S02 removal
efficiency (e.g., 90% in Figure 5.7. Thus decreasing the lime
slurry circulation rate decreases operating costs (increases lime
utilization) but increases capital costs (i.e., larger scrubber
volume). It can be seen from this observation that the possibility
of having an optimal slurry circulation rate exists. This optimal
rate will be also dependent upon the effect of gas and liquid flow
rates on the pressure drop across the TCA (or the power requirements).
The power requirements will be minimized at low values of the slurry
and gas flow rates.
In the design of TCA FGD systems,operating costs will likely be
more important than the capital cost requirements of the TCA scrubber.
Hence, the preferred region of operation of a lime scrubbing FGD
system, which uses a TCA scrubber, will be at the lower left hand
corner of Figure 5.7.
For purposes of illustration, suppose that the economic evalua-
tion is carried out on the lime-TCA FGD system and that the optimal
2
liquid flow rate and stoichiometry are found to be 27.8 liter/m sec and
1.12,respectively. Thus,from Eigure 5.7,the gas rate should be
2.4 m/sec and the equivalent packing height should be 0.88 m. For
a scrubber treating the flue gas of a 50 MW train of a power station
(89,000 SCF/min or 42.01 m3sec.) a scrubber diameter of 4.72 meters
would be necessary. The actual height of the scrubber would have
to be determined by mechanical considerations; however,the equivalent
57
-------
height of packing represented by the spray sections plus the height
of the unexpanded packing pieces in the TCA would have to total to
0.88 meters.
58
-------
5.5 Effect of Variations in the Parameters of the Models for SO^and
COg Absorptions on the Outcome of the Simulation
In the simulation procedure, the parameters, such as the mass
transfer coefficients for the C02 and S02 absorptions, were correlated
from experimental data (Borgwardt 1974b,d) by the least square method
as a function of flow characteristics, pH values and the concentration
of magnesium in the recycling slurries. Thus, it is expected that the
result of the simulation may deviate from the actual performance of the
scrubber due to the uncertainty of the parameters. In this section,
the variations in the results of the simulation due to the changes
in the model parameters are studied. The parameters investigated
include:
1) the mass transfer coefficient, k^a, for S02 absorption in
4>
the packing section.
g
2) the mass transfer coefficient, k a, for S02 absorption in
the spray section.
3) the overall mass transfer coefficient, O^a^ , for C02
absorption.
4) the magnesium effect factor, A , in the packing section.
5) the magnesium effect factor, AS, in the spray section.
6) the pH effect factor, A, in the packing section.
7) the pH effect factor, Ag. in the spray section.
8) the coefficient, a, on the exponential term relating to
the inlet partial pressure of S02, exp(aPso in)-
59
-------
Perturbation of a given parameter at ±10% of the estimated value
was made to test the sensitivity of the model performance. The ±10%
perturbation of each parameter is chosen because it is the approximate
average error of the correlation value of each of the parameters. The
remaining parameters were kept constant at the best estimated values.
The result of the perturbation test at the equivalent packing heights
of 0.75m. and 0.95m. are shown in Figures 5.7 through 5.15. From
these figures, it is clear that all the parameters tested are not
sensitive with respect to the liquid and gas flow rates required for
90% absorption of S02 in flue gas.
The effect of variations in these parameters on the S02 removal
efficiency is also tested. The results are listed in Table 5.1. The
operating conditions selected are similar to those used by Borgwardt
(1974b). From this table, it is seen that the most sensitive parameter
is the overall mass transfer coefficient for CO, absorption, (Kra)rn .
^ \3 \j\J *y
A perturbation of +10% of this value results in approximately +3.8%
variation in the S0_ removal efficiency while the same amount of
perturbation in other parameter changes the removal efficiency by less
than ±2%.
60
-------
00
•
CN
to
S
0
0)
in
rsi
E
os
Oi
cd
3 •*
C ^
CM
•
CM
SO- Inlet 2430 ppm MgO Fed = 0
'Outlet 243 ppm Cl~ = n
Average Total Pressure Drop = 2370 Pa (9.5 in. of water)
Equivalent Packing Height, m.
— Variations due to +• 10% Increments of the Mass Transfer
Coefficient, k^a, for SO- Absorption.
o
0.95 m.
0.75 m.
I
27
28
29
32
33
30 31 2
Liquid Flow Rate, liter/m sec
Figure 5.8 Effect of the Variations of the Mass Transfer Coefficient, kpa, for SO
Absorption on the System Operating Lines.1 ^
-------
00
a
o
0)
m
CM
vO
(0
CO
o
0)
CM
S02 Inlet 2430 ppm MgO Fed = 0
Outlet 243 ppm Cl =0
Average Total Pressure Drop = 2370 Pa (9.5 in. of water)
Equivalent Packing Height, m.
Variations due to ±10% Increments of the Mass Transfer
Coefficient, ksa, for S02 Absorption.
(.95 m.
0.75
28
29
30
31
32
33
Liquid Flow Rate, liter/m sec
Firure 5.9 Effect of Variations of the Mass Transfer Coefficient,k a, for SO
Absorption on the System Operating Lines. ^
-------
oo
•
CM
co,
0.75 m.
27
28
29
30
31
32
Liquid Flow Rate, liter/m sec
33
Figure 5.10 Effect of Variations of the Overall Mass Transfer Coefficient, (Kf,a)(
for C02 Absorption on the System Operating Lines.
-------
o
ro
00
S02 Inlet 2430 ppm MgO Fed = 0
Outlet 243 ppm Cl~ = 0
Average Total Pressure Drop = 2370 Pa (9.5 in. of water)
Equivalent Packing Height, m.
Variations due to ±10% Increments of the pH Effect Factor
, A , in the Packing Section.
CM
U
0)
co
vO
CM
ON
-------
ON
tn
o
CO
CM
00
O •
01 CM
(0
(0
«J
-------
CM
B
01
4J
n)
O
o)
O
•
co
00
CM
Iu
01
k
CM
CM
CM
•
CM
27
S02 Inlet 2430 ppm MgO Fed - 0
Outlet 243 ppm Cl~ = 0
Average Total Pressure Drop = 2370 Pa (9.5 in. of water)
Equivalent Packing Height, m.
Variations due to ±10% Increments of the Magnesium Effect
Factor, A , in the Packing Section
0.95 m.
0.75 m.
28
29
30
31
32
Liquid Flow Rate, liter/m sec
33
Figure 5.13 Effect of Variations of the Magnesium Effect Factor, A , in the Packing
P
Section on the System Operating Lines.
-------
o
CO
00
U
0)
e
0)
01
sf
•
tM
es
CM
S02 Inlet
2430 ppm
Outlet 243 ppm
MgO Fed = 0
Cl
= 0
Average Total Pressure Drop = 2370 Pa (9.5 in. of water)
Equivalent Packing Height, m.
Variation due to +10% Increments of the Magnesium Effect
Factor, A , in the Spray Section.
0.95 m.
_L
0.75 m.
27
28
29 30
Liquid Flow Rate,
31
32
33
sec
Figure 5.14 Effect of Variations of the Magnesium Effect Factor, A , in the Spray
Section on the System Operating Lines
-------
-------
Table 5.1
Sensitivity of Parameter Accuracy on the SO,, Removal Efficiency
Perturbation of the Parameter
from the Estimated Value = -10%
S02 Removal Efficiency Based on the
Estimated Values of the Parameters = °°
Gas Flow Rate
Liquid Flow Rate
Equivalent Packing Height
Negligible Mg and Cl in the Slurry
= 2.8 m/sec
= 30 liter/n»2sec
= 0.95 m.
Parameter
kla
V
tV3co2
\
A '
s
AP
4s
a
Estimated Value of
the Parameter
.10.922
(gmole/m atm sec)
1.409
3
(gmole/m atm sec)
0.127
(gmole/m atm sec)
1.954
(Dimension less)
16.78
(Dimensionless)
1.0
(Dimensionless)
1.0
(Dimensionless)
330
(atm'1)
S02 Removal Efficiency (%)
+10% Perturbation
of the Parameter
91.7
91.0
93.6
90.6
90.1
89.2
89.5
89.3
-10% Perturbation
of the Parameter
88.1
88.9
86.1
89.3
89.9
90.5
90.4
90.6
69
-------
CHAPTER 6
CONCLUSION AND DISCUSSION
In this study, a mathematical model which can simulate the
pressure drop across a large scale TCA used for the scrubbing of S02
from flue gas has been proposed. It includes the effects of spray,
packing and grid sections. The pressure drop across a grid with
more than 50% opening is very small compared to the pressure drop
across the spray and packing sections. It accounts for about 5%
of the total pressure drop across the TCA scrubber. A correlation
for the pressure drop of a large TCA has been developed and is
compared with experimental data covering wide ranges of
packing heights, types of grids, and flow conditions up to the loading
zone. The calculated pressure drops were in agreement with experi-
mental data for a large TCA within 10%. However, the correlation can
not be used to calculate the normal pressure drop in a small scale
TCA scrubber with lime slurry. Due to the lack of the experimental
pressure drop data taken under no scaling in the small scale TCA,
further development of a better correlation could not be undertaken.
The scale formation in the small TCA scrubber may be a possible reason
for the inadequacy of the pressure drop correlation developed to
provide better agreement with the experimental data.
C02 absorption from flue gas in the presence of 802 was also
studied. The CC>2 absorption rate was reduced drastically due to
the presence of 502 i-n ^lue £as* The extremely sensitive nature of
its mass transfer coefficient to the liquid flow rate revealed that
70
-------
the operation of the EPA/RTP scrubber was near the loading zone.Thus
it is evident that the C02 absorption has been examined only in a
narrow range of gas and liquid flow rates. A more reliable mass
transfer correlation could have been obtained if data in other
operating conditions were available. The temperature dependence of
the mass transfer coefficient has not been determined since the
experimental data were available only in a narrow range of temperature.
However, the variation in the temperature of the recycled slurry was
very small, -2°C;thus,the temperature effect*on the C02 mass trans-
fer coefficient can be neglected.
A procedure by which a complete scrubber-hold tank system for
FGD using lime slurries can be designed has been presented. The
procedure relies on equilibrium calculations to determine the state
oi the inlet and outlet conditions of the TCA slurry streamsjand semi-
empirical methods for determining the extent of S02 and C02 absorption
into the lime slurry in the TCA.
It is found that the simulation procedure could estimate the
S02 removal efficiency within 5% of the observed efficiency and
that the calculated inlet and outlet pH of the scrubber are within
10% -of the observed values. The procedure is believed to be good
enough for engineering estimations.
Using the simulation procedure developed in this paper, an
optimal operating regime can be identified where a maximum amount
of gas can be treated for a given lime feed rate. However, this
optimal regime is calculated based on the limited experimental values
71
-------
of the mass transfer coefficient for C02 absorption. The mass trans-
fer coefficients for C02 absorption in the regime other than for that
of the experimental conditions are obtained by extrapolation. More
experimental values are needed within the operating regime to
estimate the optimal operating conditions more accurately.
An optimal scrubber size which minimizes S02 removal costs
can be estimated. However, these costs are intimately tied to the
power requirements associated with pressure drop across the TCA
scrubber. Based on the simulation procedures given in this study,
it should be possible to design an optimal TCA scrubber-hold tank
FGD system.
72
-------
BIBLIOGRAPHY
Astarita, G.: Mass Transfer with Chemical Reaction. Elseview
publishing Co., 1967.
Balabekov, 0. S., P. G. Romankov, E. Ya. Tarat and M. F. Mikhlev:
J. of Appl. Chem. of U.S.S.R., Vol. 42, 1454 (1969).
Barile, R. G. and D. W. Meyer: Chem. Eng. Progr. Symp. Ser., Vol. 67
No. 119, 134 (1971).
Barile, R. G., Dengler, J. L. and Hertwig, T. A., A.I.Ch.E. Symposium
Series, Vol. 70, 154 (1974).
Bergelin, Kegel, Carpenter, and Gazley, Proc. Heat Transfer and
Fluid Mech. Inst., A.S.M.E., June 22-24, 19 (1949).
Berkowitz, J., EPA-R2-73-214, Environmental Protection Agency,
Washington, DC, April, 1973, "Evaluation of Problems Related to
Scaling in Limestone Wet Scrubbing".
Blyakher, L. G., L. Ya. Zhivaikin, and N. A. Yurovskaya: Int. Chem.
Eng., Vol. 7, 485 (1967).
Borgwardt, R., Limestone Scrubbing of Sulfur Dioxide at EPA Pilot
Plant, Report No. 1 (Aug. 1972 a).
Borgwardt, R., ibid, reports prepared for the EPA Since 1972 b.
Borgwardt, R., ibid, Report No. 6 (Jan. 1973).
Borgwardt, R., ibid, Report No. 14 (Jan, 1974 a).
Borgwardt, R., ibid, Report No. 15 (Feb, 1974 b).
Borgwardt, R., ibid, Report No. 16 (June, 1974 c).
Borgwardt, R., ibid, Report No. 17 (July, 1974 d).
*
Borgwardt, R., ibid, Report No. 21 (June, 1975 a).
Borgwardt, R., Sulfur Dioxide Scrubber Studies Related to Improving
Limestone Utilization, March 1975 b.
Chen, B. H., and W. J. M. Douglas: Can. J. Chem. Eng., Vol. 46,
245 (1968).
Douglas, H. R., Snider, I. W. A., Tomlinson, G. H., Chem. Eng. Progr.,
Vol. 59, 85 (1963).
73
-------
Douglas, W. J. M., Chem. Eng. Progr., Vol. 60, 66 (1964).
Epstein, M., EPA Alkali Scrubbing Test Facility at TVA Shavmee
Power Plant, Bechtel Progress Report Prepared for the EPA, 1970
to 1976.
Epstein, M., ibid, July 1, 1973 to Aug. 1, 1973 (Aug. 31, 1973).
Epstein, M., EPA Alkali Scrubbing Test Facility: Summary of Testing
Through October 1974. Prepared for EPA in June, 1975 a.
Epstein, M., EPA Alkali Scrubbing Test Facility: Advanced Program.
First Progress Report Prepared for EPA in Sep. 1975 b.
Fan, L. S., 1975, Dissertation, West Virginia University, Morgantown.
Gel'perin, N. I., Savchenko, V. I., Ksenzenko, B. I., V. Z.,
Grishko, and Dianov, E. A., Khimicheskoe Promyshlennost, 11 (1965).
Gel'perin, N. I., V. Z. Grishko, V. I. Savchenko, V. M. Shchedrov,
Khim. Neft. Masninostroenie, No. 1, 22 (1966).
Gel'perin, N. I., Yu. M. Latyshev and L. I. Blyakham: Intl. Chem.
Eng., Vol. 8, 691 (1968).
Groeneveld, K. J. W. (1967), Dissertation, Technische Hogeschool,
Delft.
Hatta, S., Technol. Repts. Tohoku Imp. Univ., Vol. 8, 1 (1928).
Khanria, R. T., Ph.D. Thesis, McGill Univ., Canada, 1971.
Kielback, A. W., Chem. Eng., Vol. 66, 106 (Dec. 14, 1959).
Kito, M., Shimada, M., Sakai, T. Sugiyama, S., and Wen, C. Y.,
paper presented at Engineering Foundation Conferences, California
(June, 1975).
Kito, M., Sawada, M. Shimada, T. Takata, T. Sakai and S. Sugiyama:
Submitted to KagakuKogaku, to be published.
Krainev, N. I., M. I. Niyazov, I. P. Levsh and S. U. Umarov J. of
Appl. Chem. of U.S.S.R., Vol. 41, 1961 (1968).
Kulbach, A. W., Chem. Eng. Progr. Symp. Ser., Vol. 57, No. 35 (1961)
LaMantia, C. R., Hunt, R. R., and Shah, I. S., Paper 25c, presented
at the 66th AIChE Meeting, New York, NY, Nov. 1973, "Dual Alkali
Process for Sulfur Dioxide Control".
74
-------
Levsh, L. P., Krainev, N. I., Niyasov, M. I., Intl. Chem. Eng.,
Vol. 8, 311 (1968).
Lowell, Philip S., A Theoretical Description of the Limestone
Injection Wet Scrubbing Process, Radian Report for EPA in June 9,
1970.
Mcllroy, R. A., Atwood, G. A., and Major, C. J., Environ. Sci. and
Technol., Vol. 7, 1022 (1973), "Absorption of Sulfur Dioxide by
Molten Carbonates".
McMichael, W. J., Fan, L. S. and Wen, C. Y., paper presented at
AIChE Meeting, Houston, Texas (March 16-20, 1975).
McMichael, W. J., Fan, L. S. and Wen, C. Y., I and EC Process Design
and Development, Vol. 15, 459 (1976).
Nelson, R. D., M. S. Thesis, West Virginia University, Morgantown,
W.Va. (1974).
Nannen, L. W., West, R. E., and Kreith, F., J. Air Poll. Control
Assoc. Vol. 24, 29 (1974).
Nijsing, R. A. T. 0., 1975, 1969, cited by Astarita, G. (ref. 1.)
O'nell, B. K., Nicklin, D. J., Morgan, N. J., and Leung, L. S.,
Canadian J. Chem. Eng., Vol. 50, 595 (1972).
Payne, J. W., and Dodge, B. F., Ind. Eng. Chem., Vol. 24, 630 (1932).
Shale, C. C., Simpson, D. G., and Lewis, P. S., Chem. Eng. Progress,
Vol. 67, 52 (1971).
Slack, A. V., Intl. J. Sulfur Chem. Vol. 7B, 67 (1972).
Strauss, W., Industrial Gas Cleaning, Pergamon Press,
Oxford, 1966, Section 3.7.1. and Section 3.4.
Tepe', J. B., and Dodge, B. F., Trans. Am. Inst. Chem. Engrs., Vol. 39,
255 (1943).
Tichy, J., Wong A. and Douglas, W. J. M., Can. J. Chem. Eng., Vol. 39,
255 (1943).
Tichy, J., and Douglas, W. J. M., Can. J. Chem. Eng., Vol. 51, 618,
(1973).
Wen, C. Y., Wet Scrubber Study, Report prepared by West Virginia
University for the EPA, Report No. 35 (Dec. 1973).
75
-------
Woodies, T. C., Cummings, J. M., Jr., and Hunter, G. B., Environ.
Sci. and Techno!. Vol. 7, 827 (1973).
76
-------
APPENDIX A.
SIMULATION COMPUTER PROGRAM
The following program and subroutines were used to simulate the
lime scrubber-hold tank system. Flow diagram for the computations
have been given earlier in the thesis (Figure 5.3).
77
-------
C SIMULATION COMPUTER PROGRAM.
COMMON/ALIN2/PH2(12),PLT(12),Nr>TPH
COMMON/RFEF/PMn,SK.GA, PKGA,PS02,ZS,ZP,G,P
COMMON/ALINl/DLTtt,DLT9,PH4
COMMON/ENEL/AMG,CL,S03,ANK
COMMON/CC02/SOF,CAF,VV,AK,COO,S,Z,V,C02,Af;S3,Mr:M
COMMON/SOLU/MM,GPM,CAO,EFFSP
COMMON/PRI^7TITLE(lP.)/COBS(7)/PHORS/IT5/PCSR/PPCC,STn
COMMON/SYPLX/ S02/UTI , PHI N, PHOT, EFFCY
CJLJLJLJlMAJtMJLJLjlJt&JLJtJtHJlJlJlA&&A&&AAA&&£JtJlAHJlJt4tJt&JlM. JlJlJlJl M A J^ Jt M A Jf-
W if H if W T ff « ff«WHHwnn if w WlfWWWHnW"^WiTtfWiiWir iFVirjfTrfFWif if if W W TrffTTnTTrrTr
C A CROSS SECTION AREA,CM**2
C V L/SEC.CM**2,LIO FLOW
C AMG MMOL/L
C PMG. ...PPM,t'SEr IN THE CALCULATION OF S02 REM EFF
C CAO....FFED,#/HR
C CAF....LIME FED^MOLE/L
C GPM....LIQ FLOW RATE,PPM
C S02....FEED,*/HR
C SOF S02 FED,Mf'OLE/L
^ C PHIN...PH AT INLET OF TOV'ER
00 C PHOT...PH AT OUT LET OF TOWFR
C EFFCY..S02 REM EFF
C PLTI...DLT AT PH IN FROM EQLH PROGRN?
C DLTO...DI.T AT PH OUT FROM EOLM PPOCP^
C ABS1...S02 ABSORRED^MOL/L
C ****************************************************
MM»MM+1
C IF(MM.GT.l) GO TO 120
C ****************************************************
A»l*10.
Z=285.
ZP=60.
-------
S02-0.
P=9.5
MN=5
NN-6
NDTPH=12
NPDP-12
CONTINUF
SL«n.
CL=0.
S03«27.6
PPCO.12
T=51.6
READ (MM,700) TITLE
IF (T.EO.O.) STOP
OPM»A*V/.06308
CAO*CAF*nPM/35.P<)9
AL=1050.*V
CALL SUFATCSOS^L
WRITE (NN,815) TITLE
*******************************************
IF(MMM.LE.l) GO TO 150
VfRITE(6,72)
WRITE(6,£)
WPITE(6/5) AMH,CLASPS
WRITE(fi,10)P/PS02/PPrC
WRITE(6,6) rAF,SOF,T,SI
WRITE (6,11)0,V,Z,ZP
-------
72 FORMAK//,1 TEST EPA SCPDRRTR SI?17.1)
8 FORMAT(//, lvX,'Mnl,lGX,'CLl,llX,1S03l)
5 FORMAT(lX,F10.lf,5X,F10.i*,5X,F10.U
10 FORMAT(/,6X,'PI,UX,1PS02I,10X,'PCP21,/, 3X, F5. 1, IPX, F10 . 6, 6X,
C fll.2)
6 FORMAT(//6X/ICAFM2X,ISOF'/11X,ITM5X,ISLI///1X/F10.3/ 5X,F10.3,
/ 5X,F10.3,10X,F10.3)
11 FORMAT(/,6X, 'GMUXj'VMSX, 'ZMSX^ZP1 ,7,2X^7.5,10X^7. k, RX,
/ F7.U10X,F10.3)
700 FORMAT(18AI»)
701 FORMAT (OF 10. 4)
702 F0RMAT(8Fin.6)
815 FORMAT (///, 10X,18AU)
C *******************************************
CALL ENELS(T)
WR I TEC 6, 7)
7 FORMAT (//,' PH DLT')
§ DO 2 1=1,NDDD
WRITE (6,U)PH2(!),DLT(1)
k FORMAT(lX,2Fin.O
2 CONTINUE
150 CONTINUE
C ###f ######f ######## f ######*###*####### ####f ###########
DO 3 I=1,NDDD
IF(PH2(I).GT.U.O) GO TO Ul
PHU=PH2(I)
Ul IF (PH2(I).G?.9.) DLT9=DLT(I)
3 CONTINUE
#### ##################################################
AK=. 0005125
C00»2.1676
S«. 05536
-------
W=V**(-.«tl57)
ZS=Z-7P
PMG«AMG*2lu3
*********
120 CONTINUE
S02=U52.35*A*G*PS02
SOF=31.2365*SC2/GPM
SKGA=0.0013U*(r**0.8)*(AL**n.»»)
PKGA=.00132*(G**.l*7)*(AL**. 51 )*((P/6.1 )**!.!)
21 CALL PHEFF (CAF,SOF, PH I N, PHOT, EFFCY)
UTI «SOF*EFFCY/CAF
STOI=1./UTI
WRITE(6,71)V,GPM ,CAF, CAO,SOF,S02
WRITECB^OO) UTI,STOI
WRITE(6,20) ABS3/C02/G
oo WRITE (6, 9) PHIM,PHOT,EFFrY
M 500 FORMAT C/^SX/L'T I «', F5. 2, 5X/ 'STOI = ' , F6. 3)
20 FORMATC 5X,'ABS •• F5. 2, 5X, 'C02 =' t F5. 2, 5X, 'G ='^7.5)
9 FORMAT( 5X, ' PH I N=',F5. 2X 5X, ' PHOT=' , F5. 2, 5X, ' EFF^1, F6. l»
73 FORMAT(5X,'Z =', F6. 2/ 5X, 'ZP ='^7.2)
71 FORMATC/.SX/V «' ,F9. G^X, 'GPM-1 ,F7. 1,/, 5X, 'CAF= ' , F6. 27
C 5X, ICAO«'/F7.2//,5X,ISOF='/F6.2/5X/ISO?=I,F7.2)
STOP
END
SUBROUTINE PHEFF (CAP, SOF, PHI N, PHOT, FFFPY)
COMMOM/ALIN2/PH2(12),PLT(12),MPTPH
C ..... INPUT:CAF,SOF. OUTPUT: PH I M, PHOT, EFFCY.
C f#####f#############*####i»#####a#*# ##*####»####*#***##
NTT-0
PHI1=7.5
-------
C #* ##M##ff#f*#*#t ###############*###
CALL PRniT 00 TO 63
IF(CAFD1.GT.O.) GO TO 62
PW2=«5.5
GO CALL PRDIT(PHI2,CAF1,CAF2,CAFP2,EFF2,PH02)
IF(ABSCCAFD2/CAF).LE..0005> GO TO GU
61 CONTINUE
IF(ABS(CAFD2-CAFDl).CT..no001) GO TO 50
PH13=(PHI2+PHI1)*.5
NTT»NTT+1
WRITE(6,53) PHI1,PHI2,CAFD1,CAFP2
53 FORMAT(//^F10.6//,5X«*IN SUB PHEFF ')
GO TO 51
50 PHI3=PHIl-CAFDl*(PHI2-PHIl)/(CAFP2-CAFni)
51 A=PHI3
B.»CAF2
CALL PRHITCA^B^C^D^E.F)
CAF3»C
CAF03=D
EFF3»E
PH03=F
IF(NTT.GT.O) GO TO 65
!F(ABS(CAFD3/CAF).LE..0005) r,0 TO 65
PHI1«PH!2
CAFD1*CAFD2
CAF1=CAF2
PH!2=PHI3
CAFD2=CAFD3
CAF2=CAF3
GO TO 61
C
62 PHI2»10.7
-------
GO TO 60
63 PHIN«PHI1
PMOT=PHP1
EFFCY=EFF1
RETURN
BU PHIN=PHI2
PHOT*PH02
EFFCY=EFF2
RETURN
65 PHIN=PHI3
PHOT=PH03
EFFCY=EFF3
70 RETURN
END
SUBROUTINE PRO IT (PHI ,ABS1,A*S2, CAFP, EFF, PHD)
COMMON/ALIN2/PH2(12),nLT(12),NnTPH
COMMON/AUNl/DLTU,nLT9,PHlv
COMMON/CC02/SOF/CAF/VV/AK/COO/S/Z/V/Cn2,AP?.3/M
C ..... ABS1... GUESS Sf^ABS t MK'OL/L
C ..... ABS2...S02/AES CALC'D
C ..... CAFD... DIFFERENCE BETWEEN CAP TATA A Pt>EMCTf^ VAIA'E
C ..... PHI PHO...PH AT IN & OUT.
SF*0.
1=0
IF(N.LF..l») GO TO 90
IF(PHI .LE.10.2) GO TO 90
WPITE(6/62)
WR I TE ( 6^1 )SOF,EFF,PH I ^PPO^A^Sl,
62 FORMAT(///I PHirOlO.2,1^ SUP- PPO|T')
STOP
90 IF(PHI.GE.9.) GO TO 5P.
IFCPV'I .GT.PHlf) GO TO G3
-------
WRITF.(6,B1)
WRITE(6,81)SOF,EFF,PHI,PHO,ABSl,C02,nLTI,DLTO
61 FORMAT (' PHIN IS TOO LOW. IN SUP PPPIT')
STOP
******************************************************
63 CALL AL!NE(MDTPH,PH2,DLT,PHI,nLTI)
GO TO 51
#########*###########«## ########**#
56 rLTI«DLT9
51 OLTO*=nLTI+ABSl
1=1+1
IF (I.GT.30) GO TO 50
IFCDLTO.r-T.DLTU ) GO TO 58
CALL ALIME (NPTPH,DLT, PH2,DLTO, PHO)
GO TO 5k
58 V'RITE(6,80)
WRITE(6,81)SOF/EFF,PHI/PHO/AESl/C02,nLTI/DlTO
80 F'ORMATC//,1 CALCULATED TLT OUT IS TOO Hi™.',/,1 PROPABLY l.OLT IS
/ NEARLY CONSTANT. 2.DLTOUT IS HEIHHEP THAN THE PLJ AT THE LOV/FST P
/H2.1)
81 FORMATC//,' SOF EFF PH I M PHOT ' , /, kF7 . T>, II, ' ABS1
/C02 DLTI DLTO *,/,kF7.1)
STOP
C ######################################*######/*### ##f##
5k PHLM=.362+ALOfUO((PHI-PHO)/(10.**(-PHO)-10.**(-PHI )))
C ######################?#############*###*#############
CALL REMEF (PHLM,EFF)
ABS2=EFF*SOF
SPER-A8S1-A8S2
IF(A8S(SPER) .LE..001) GO TO 50
ABS1=SF*SPER+ABS2
C ..... SF ..... SCALE FACTOR
-------
GO TO 51
C #**#*#######,#########*#######*#########*#######$######
C C02 PORTION
50 PHM=(PHI+PHO)/2.
H«10.**(3.-PHM)
CKGA=EXP(1.108*PHM-1.50U*VV)
AKH-l.+AK/H
C02=COO*AKH*(1.-EXP(-CKGA*S*Z/A!OYV))
C ******************************************************
ABS3»ABS2
CAFD«CAF-C02-ABS2
IF(N.LE.2) RETURN
WRITE(6,81)SOF,EFF,PHI,PHO,ARS1,C02,DITI,DLTO
60 RETURN
END
SUBROUTINE REMEF(PHLM,EFF)
COMMON/REEF/PMG,SKGA^PKRA/PS02/ZS,ZP/n/P
COMMON/RRR/AS^S^P.AP, SPRAY, PACK
C AL GM/SEC.CM**2
C G GMOLE/SEC.CM**2
C P IN OF WATEP
C PMG.... PPM
C PS02...ATM
C ZS ZP..CM
IF(PMG.GT.350.)GC TO 3
SnELT«l.
PDELT»1.
GO TO 6
3 SDELT»5n.l/(PMG**n.6682)
IF(PMG.GT.3600.)GO TO 5
PDELT-1.
GO TO 6
5 PPELT=22./(0.0012«t29*P^)**2.0(:,5
-------
6 CONTINUE
PHS»PHLM
IF(PHLM.GE.7.19)PHS»7.19
AS»l./(EXP(-1.35*PHS+7.82)-0.15)
IFCPHLM.GT.G.) 00 TO 1
AP«1./0.308
GO TO 2
1 CONTINUE
C IF PH VALUE IS GREATER THAN 6.59 , AP VALUE WILL RE NEGATIVE
PHP-PHLM
IF(PHLM.GE.6.59)PHP=6.59
AP=l./(-.517*PHP+3.M)
2 E«EXPC-33n.*PS02)
RS=(AS/SOELT)*E
RP=(AP/PDELT)*E
PACK»PKGA*ZP*RP/G/(1.+RP)
SPRAY«SKGA*ZS*RS/G/(1.+RS)
EFF -l.-l./EXP(SPRAY+PACK)
RETURN
END
SUBROUTI-NE ALINE (N^,Y^XX,YY)
DIMENSION Y(12),X(12)
C ##########*ffffMff#**#**#f##**#f**##*################
MaN-1
no it 1=1^'
X1=X(I)-XX
X2»X(I+1)-XX
X3=X1*X2
IF(ABS(X1).LE..0002) GO TO 5
IF(ABS(X2).LE..OOQ2) GO TO 6
IFCX3.LT.O.) RO TO 7
I* CONTINUE
5 YY=Y(I)
-------
GO TO 8
YY=Y(I+1)
PO TO P.
SLOP=(Y(I)-Y(I+1))/(X(I )-X(l+D)
YY=SLOP*(XX-X(I))+Y(I)
RETURN
ENO
SUBROUTINE ENELS (T)
DIMENSION C(35),EK(25),CONS<10)
COMMON/AI.IN2/PH2(12),ni.T(l?),NnTPH
COMMON/ PP I NM/T I TIE ( 18 ) , CO BS ( 7 ) , PHOPS t I T5 , PCSP , PPCC , STO I
COf-'MOM/nFRJ/nELTB
COMMON/ ENEL/AMG,CL,S03,ANK
COMMON/TESTS/TEST/TFST2,TFST3/TESTIf 00003000
MN=5 00005000
MN=6
COBS(1)=0.
COBS(6)=0.
COBS (2)= AMG
COBS (3)= ANK
COBS(5)= S03
COBS(7)= Cl.
TT = T + 773.16
CALL EOCON(FK,TT) 00017000
C f####f ######
K«0
302 CONTINUE
K«K+1
PH»PH2(K)
CONS(l) = 0.6 * CORSC5) / 1000. 00037000
COMS(2) « 0.9 * CCBS(?) / 1POO. 0003ROOO
-------
CONS(3) -
CONS(U) •
CCNS(5) '
CONS(6) •
•CO-NSC 7) •
CO'NS(g) •
COMS(9) <
CONS(10)
COBS(7) /
COBS(3) /
PPCC
10.**(-PH)
0.0
CONS(1)
CONS(2)
« CONSC5)
1000.
1000.
ON
oo
oo
If f #f#f*#f##fl*f*f *##*####*###*########## ##f»####
CALL EQUIC(C,EK,CONS,TT,PH)
QSMG •« CONS(l) + CONSC2)
IFCQSMG .LT. 0.001) GO TO 1*15
CHECK FOR SPECIFIED SULFATE REYONP SATURATI
IF (CONS(l) .LT. 0.001) GO TO 1*13
IF (TEST .EQ. 0.) GO TO 201
IF (CONS(l) .GT. C(33)) GO TO 1*13
CONS(l) = 0.001 * COBS(5) * CCHO /
CHECK FOR SPECIFIED MAGNESIUM ABOVE
IF (CONS(2) .LT. 0.001) CO TO UH
IF (TEST2 .EQ. 0.)GO TO 203
IF (CONS(2) .GT. C(35)) GO TO klk
CONS(2) = 0.001 * COBS(2) * C(19) /
#################*######£######## ##
CALL EQUIC(C,EK/CONS,TT/PH)
CONTINUE
201
»*13
203
klk
l»15
2) + C(33)+C(3U-C(31))*inoO.
IF(K.LT.12)GO TO 302
RETURN
END
SUBROUTINE SUFAT(S03,CL, AMG)
C ..... INPUT:CL,AMG . OUTPUT:S03
C S03 CONC PRFPICTION AT CL ANT Mr
C(33)
SATURAT
C(35)
OM
00039000
OOOitOOOO
00041000
000^2000
OOOU3000
0004UOOO
OOOU5000
0001+6000
0001*7000
OOOU8000
00050000
00051000
00052000
00053000
00051*000
00055000
00056000
00057000
00058000
00059000
00060000
00061000
00063000
OOOFOOOO
-------
C MG...AMG ,MMOL/L
C CL...CL,MMOh/L
0 S03..AMC+X,Mr
-------
c
c
S03..MMOLE/L
100 RETURN
END
SUBROUTINE EQCON (FK,TT)
DIMENSION EK(25)
C GIVEN THE TEMPERATURE,
P = 1.9872
C EK(2), EK(U), AMD EK.U7) ARE
C DEPENDING ON SATURATION IN
EK(2) - 0.
EK(1U) = 0.
EK(17) •= 0.
C FIRST GROUP IN STANDARD FORM FOR
C KSP FOR CAS03
EK(1) = EXP(U933.5/TT-U6.27)/R)
C KB FOR H2S03
EK(3) = EXP((2900.8/TT-U2.71)/R)
C KA FOR H2S03
EKC5) = EXP((3861
KDISS FOR ION PAIR
= EXP((2310
FOR ION PAIR
= EXP((2176
FOR ION PAIR
= EXP((1381
C KDISS FOR ION PAIR
EXP((1250.l*/TT-10.^5)/R)
C 1/H FflR S02 (H=P/C)
EK(12) = EXP((6269.8/TT-20.6U)/R)
C 1/H POP 002 (H=P/C)
EK(13) = EXP(O»6lt5.2/TT-22.2?)/R)
C KDISS FOR ION PAIR MGOH+
THIS RETURNS THE EOIJI LI PR I Mr
FILLF.P
CAC03,
If
THE
SIPPOUTINE
EK(7)
KDISS
EK(8)
KDISS
EK(9)
KDISS
EK(10)
.l/TT-21
CAS 03
.2/TT-23
CAC03
l/TT-21
CAHC03+
2/TT-10
CAOH+
FNTHALPY AND RNTPCPY
5lt)/R)
30)/R)
00203000
0020^000
00206000
00207000
00208000
00209000
00210000
00211000
00212000
00213000
00211*000
00215000
00216000
00217000
00218000
00219000
00220000
00221000
00222000
00223000
0022l»000
00225000
00226000
00227000
00228000
00229000
00230000
00231000
-------
C
C
C
C
n
C
c
c
c
c
c
c
c
EK(1E) = EXP((2370.6/TT-13.7*)/R)
KSP FOR MRSQ3
EK.C25) = EXP((-3fi71.3/TT-7.2IO/P)
KPISS FOR ION PAIR MCHC03+
EK(18) = EXP((1075.
-------
Kl
COMMON/PRINT/A(22),XF(22),L,PCC,PCS,PCM2
COMMON/TESTS/TEST,TEST2,TEST3,TESTU
DIMENSION F(70,22),ZI(22)
DIMENSION C(35),EK(25),CONS(10)
PATA ZI/.U,.U,.8,.8,1., 1.,!.,!.,.8,.8, . 8,.R,,U,.U,. 8,
* .8,1.., .8, .U.,1., !.,!./
C THIS SUBROUTINE CONTROLS THE AOL SET OF EOlMLIBRIUM SUBROUTINES
C INITIALIZE VALUES FOR THIS CALCULATION
PO 501 J=l,22
A(J)«0.0
DO SOU J=l,35
C(J)=0.0
TEST =0.0
TEST2 = 0.0
TESTU=1.0
15 = 0
16 = 0
17 = 0
CONS(6)
CONS(8)
CONSO)
C(23) =
C(2U) «
CONS(10)
501
SOU
326
= 10.**(-PH)
= CONS(l)
= CONS(2)
CONS(3)
CONS(U)
= CONS (5)
A(ll) = CONS(R)
SET SATURATION SITUATION
EK(2)=EK(23)
IF(TESTU .LT. 1.)EK(2)=EK(8)
IF (TEST .LT. 1.)EK(1U)=EK(21)
EK(17)=EK(25)
IF (TEST2 .LT. 1. )EK(17)»EK(1Q)
002B6000
00267000
00268000
00269000
00270000
00271000
00272000
00273000
0027UOOO
00275000
00276000
00277000
00278000
00279000
00280000
00281000
00282000
00283000
00284000
00285000
00286000
00287000
00288000
00289000
00290000
00291000
002H2000
00293000
H029UOOO
00295000
00296000
00297000
-------
to
17 = 17+1
IF (17 .LE.' 10) GO TO 401
WRITE (C,805)
805 FORMAT (10X,'TOP MANY CHANCES IM SPLIP? PDFSEMT1 )
STOP
T REPEAT THE CALCULATION UNTIL ALL TESTS ARE MET OP >70 L^PPS
l»01 DO 510 L = 1,70
C FIRST LOOP PASS USES ESTIMATED ACTIVITY COEFFICIENTS
IF (L .EO. 1) OP TO U12
C AFTER 15 LOOPS, USE ALTERNATE FPRf OF C.ORREOTIMC ACTIVITY COFFS
IF (L .LE. 15) GP TO U02
HO 513 JN = 1,22
PELTA « (FCL^JfO-Fd-l/JN
F(L,JM) » F » COMS(7)
IF(TEST3 .OT. 0.) GO TO 32F
C
#####*###### ##
F(l.,m)
F(L,19)
FL1U
FL19
CALL
no 507 J=l,22
507 C(J) « A(J) / F(L,J)
IF (L ,EO. 1) GO TO 510
00 505 J^ =1,2?
00298000
00299000
00300000
00301000
00302000
0030300"
0030l»000
00305000
0030FOOO
0030700"
OP308000
00300000
00310000
00311000
00312000
00313000
0031UOOO
00315000
00316000
00317000
0031ROOO
00319000
00320000
00321000
00322000
00323000
0032UOOO
00325000
0032POOO
00327000
00328000
003?9000
00330000
-------
IF (ARS((F(L-1,JM)-F(L,JM))/F(L,JM)> .GT. 0.001) PO TO 510
505 CONTINUE
TEST3=0.
#!#####f#####*#f#################*###############
CALL TTEST(C,EK,A,CONS,PCC,PCS,PCM2)
IF (TF.ST3 .NE. 0.) GO TO 326
FOLLOWING ARE ION SWS, IONIC IMBALANCE, I^MIC STPENPTH
C(?.5) = C(9) «• C(10) + C(ll) + C(16) + C(1R) + CC21<)
C(26) = C(3) + C(l|) + C(l?) +C(15) + C(23)
C(27) » C(13) + C(19)
C(28) « C(l) + C(2) * C(1U)
C(29) = C(25)-C(2fi) + 2. * (C(27)-C(28))
C<30) = (C(?.5)+C(26))/2. + 2. * (C(27)+C(2R))
FOLLOWING ARE TOTAL CA, SULFITE, SULFATE, CAPROMATE, MG
C(31) «= C(7) + C(8) + C(9) + C(10) + C(13) •«• C(21)
C(32) = C(l) + C(3) + C(5) + C(7) + C(17)
C(33) = C(ll*) + C(15) + C(20) + C(21)
C(3I») = C(2) + C(l*) + C(6) + C(8) + C(9) + C(1R) + C(?2)
0(35) » C(16) + C(17) + C(18) + C(10) + C(22) + C(20)
RETURN
510 -CONTINUE
WRITE (6,806)
806 FORMAT (10X,'MORE THAN 70 ITERATIONS IN EOl'ir')
STOP
ENP
SUBROUTINE CFUGU(C,EF,TT,IG )
DIMENSION C(35),EF(22)
DIMENSION CA(2?),CR(22),U(22)
REAL IZ(22)
DATA IZ/l*.,lt.,l.,l.,0., O.,0.,0.,!.,!., I.,!.,!*.,!*.,!.,
DATA u/o.,o.,o.,o.,.o76/.07fi,.07R,.07P/n.,n./n.,n./n.,n.,n./
:,.076/
00331000
nn332onn
00333000
00335000
0033FOOO
00337000
00338000
00339000
003HOOOO
0031*1000
0.03U2000
0031*3000
0031*1*000
003U5000
003U6000
0031*7000
0031*8000
0031*9000
00350000
00351000
00352000
00353000
00351*000
00355000
00356000
00357000
00359000
00359000
00360000
003^1000
00362000
"0363000
-------
<£>
tn
TATA CAA.5,l*.5,lt.5,l».5,3.,3.,3.,3.,3.,3.,F.,3.,lu5,3.,3., 003fil»000
13.,3.,3.,3..,3.,3.,3./ 00365000
PATA CB/0./0./n.,0.,.3,.3,.3,.3/.3/.3/.ti/.3/.l/0./.3/ 003FFOOO
1.3,.3,.3,.3,.3,.3,.3/ 00367000
TC = TT-273.16 00369000
OS = 87.7U - .l»0008*TC + 9. 38E-«t*(TC**2. ) - l.UF.-f *(TC**3. ) 00370000
PA = (1.8248E6)/«nS*TT)**1.5) 00371000
DB = 50.292/<(ns*TT)**.5) 00372000
SI IS THE IONIC STRENGTH TIMES TWO 00373000
SI = C(23) + C(2U) 00371(000
DO 501 I = 1,22 00375000
SI = SI + IZ(I) * C(l) 00376000
501 EF(I) = 0.0001 00377000
DO 503 I = 1,22 00378000
IF (C(l) .LE. 0.) GO TO 503 00370000
EF(I) = EXP(2.-303*(l!(l)*SI + PA* I Z( !)*(-(S I ** . 5 )/( 1. + PB*r/Kl)* 00380000
KSI**.5)) + CB(I)*SI))) 00781000
IF (EF(I) .LT. 0.001) GO TO t»01 003P200Q
IF (EF(I) .LE. 10.) GO TO 503 00383000
Ml EF(I) = -1. 003PI»000
'/.'RITE (6,801) I,SI,C(I) no3850no
801 FORMAT (10X,'POSSI RLE ERROR IM SPFCIFS ',I 2,' I S= '^FlP.fi, 0038*000
I1 CONG = ',F12.6) 00387^00
1C »IB + 1 00388000
IF (16 .r-E. 8) STOP 003P°000
503 CONTINUE 003^0000
PETURN
END
SUBROUTINE CABAL(F,EK,COMS,\., 15, IH)
PIMENSIOM E(35,3),A(13) 003TUOOO
PIMENSION F(70,22),EK(25),CPMf;(10) 00395000
COHMON/TESTS/TEST,TEST2,TEST3,TFSTli 003«>F^no
MM=G 00399000
-------
220
A(ll) = COMS(B)
S = CONS*F(L,10)*EKdO))
EK(19)*EKdl)/(2.*EK(2)*FKdr>)*F(L,lG)*Mll))
EK(19)/(EK(2)*F(L,19))
E(1,1)+E(2,1)
n*EK(6)*EK(13)/(2.*EK(9)*F(L,9)*A(ll))
£(5,1)^(7,1)
F.K(17)/(F(L,1P)*F:K(1))
E(10,l) = EK(17)*EK(11)/(A(11)*F(L,16)*EK(1)*EK(16)*?.)
E(ll,l) = EK(17)*EK(F)*EK(13)*EK(1»)*^/(EK(1)*PK(1P)*
1F(L,18)*EKU)*A(11)*2.)
E(8,l)
£(9,1)
E(2,
APP
E(l,
ADP
E(2,
E(3,
F(U,
E(5,
E(6,
E(7,
E(8,
2)
IN
2)
IN
2)
2)
2)
2)
2)
2)
2)
= A(11)/(2.*F(L,1D)
ANY OTHER NOMCOMPLEX INfi SI HOLY CHARCEP - SPECIFS PPMC
= E(l, 2) + CONS(3)/2.
ANY OTHER MONCOMPLEX I NO SINGLY CMARrF.P + SPECIFS POMC
= E(2,2) + HOMS(it)/2.
= EK(2)*Adl)/(2.*EK(t*)*F.K(9)*r(L,o))
= S
= A(ll)*S*F(l,Ui)/(?.*EK(15)*F(L,15))
= TM*F(L,lP)*EK(ll)/(2.*EK(16)*F(t.,16)*A(ll))
= TM
npn
oounoooo
oomiooo
OOU12000
nniiuooo
oouisnoo
nnuiRnnn
noi4i7nno
nomsnno
nnmgooo
nnu2nono
onii2ionn
001*22000
OOU2UOOO
OOU26000
00427000
OOI<31000
001*33000
0043UOOO
-------
to
0(9,2) = E(8,2)-E(3,2)
E(10,2) = E09,2) + EU,2) + E(5,2)
£(11,2) = E(10,2)-E(6,2)-E(7,2)
E(12,2) = E(9,2)-E(6,2)-E(7,2)
E(13,2) = EK(19)*A(11)/(2.*FK(18)*F(L,18)*EKU))
E(l«»#2) = E(10/2)-E(13/2)
E(15,2) = E(<5/2)-E(13,2)
E(16,2) = P*EK(6)*EK(13)*EK(U)/(A(11)**2.*F(L,2))
E(17,2) = n*EK(6)*EK(13)/(2.*A(ll)*F(L^))
= E(8/2)+E(16/2)+E(17/2)
=EK(19)*A(ll)**2.0/(D*EK(B)*EK(13)*EK(U)*F(t./19))
=EK(ll)*EK(19)*A(ll)/(2.0*n*F(L/16)*EK(16)*EK(6)*EK(13)
E(19,2)
E(20,2)
1*EK(U))
E(21,2)
E(22,2)
E(23,2)
E(2U,2)
E(25,2)
E(26,2)
E(27,2)
E(?8,2)
E(29,2)
E(30,2)
E(31,2)
CR TERMS
E(l,3)
E(2,3)
E(3,3)
FU,3)
E(5,3)
IN
E(18/2)-E(19/2)-E(2n/2)-E(13/2)
E(21,2)+E(U/2)+E(5/2)
(lP/2)+E(U/2)+E(5/2)-E(P/2)-E(7,2)
E(18/2)-E(6/2)-E(7/2)
TM*F(L,in)*EK(6)*EK(13)*P/(EK(18)*F(L/18)*A(ll)*?.)
E(23,2) - E(25,2)
£(2^, 2) - E(25,2)
EK(17)*EK(2)*A(11)/(F(L,18)*EK(1)*FK(18)*EK(U)*2. )
E(18,2)+E(U,2)+F(5,2)
(10,2)-E(28,2)
E(9,2)-E(28,2)
CONCENTRATION TIMES ACTIVITY OF CAI.flW
A(11)*EK(1)/(2.*F(L,3)*FK(3))
EK(2)/F(L,2)
EK(2)*A(ll)/(2.*EK(t»)*F(L/IO)
EK(HO/F(L,1U)
A(ll)*EK(]l»)/(2.*EK(l!»)*F(L,15))
TM*r(l.,19)*A(ll)*FK(2)/(2.*r(L/18)*FK(l?)*FK(l;))
00«i36POO
001*37000
OOU38000
00^39000
00^1*2000
00450000
OOU51000
00^53000
001*56000
00^59000
OO^ROOOO
OHU62000
OO^fiSOOO
nnijBSnnn
npl»R7000
-------
», rr " c \. J. J. , J ;
00 TO 128
206 *P = F(5,l)
BR = E(12,2)
CP. = E(l?, 3)
OOU69000
f*0l*70000
OCU71000
OOU72000
001473000
001*71*000
OOU76000
IT = TEST + 2.*TEST2 + U.*TESTU + 1. 001*78000
CO TO (201, 202^03, 201i,205, 206,2*7,208), IT OOU70000
AR - E(R,1) POU80POO
BR - E(26,2) OOUP1000
CR = E(8,3) OOHP2000
HO TO 128 001*83000
202 AR = F(P,1) OOI«8I*000
RR • F(27,2) OOM5000
CR « E(13,3) OOU86000
H'O TO 128 OOU87000
203 AR = E(12,l) nO«*8POPO
PR « E(?f)/2) OOliSQOOO
CR = E(S,3)
PO TO 128
20U AR = E(12,l) 001»o?000
RR = E(18,2) OOU93000
CR - E(13,3) OOI*gi»000
^0 TO 128 001*95000
205 AR « E(5,l) noi»9FOOO
BR = E(ll,2) 001*97000
= E(ll,3)
CR = E(ll,3) OOU18000
00500000
00501000
005020no
-------
207
208
128
to
to
1*51
198
1*21
C
l»22
199
76
701
»».*AR*CR
) CO TO U22
ARG**.5)/(2.*AP)
RETURN
GO TO 421
00 TO 128
AR - E(13,U
BR = E(30,2)
CR = E(9,3)
GO TO 128
AR « E(13,l)
BR = E(31,2)
CR * E(10,3)
CR = -CP
BR = -BR
ARC - BR**2 -
IF (ARC .LT. 0
A(13) = (-BR +
CONSm = A(13)
IF (A(13) .GT. 0.)
IF (A(13) .EQ. 0.)
WRITE(NN,198)
FORMAT(10X,'THE ROOT
15 = 15*1
IF (15 .GT. 3) GO TO 76
A(13) •= (-BR-ARG**.5)/(2.*AR)
CONS(7) » A(13)
PO TO 1*51
WRITE (NN,197)
FORMAT(10X,'THE ROOT IS ZERO1,/)
GO TO 76
WRITE(NN,199)
FORMATC10X,'THERE IS NO PEAL ROOT',/)
RPN DOESN'T UMPEPSTAMD WPY API. SETS TEST2
TEST? • 1.
WRITE (MN,701)
FORMAT (' PROBLEM WITH POOT IM TABAL')
16 = 16+1
IS NEGATIVE1,/)
00503000
0050l»000
00505000
00506000
00507000
00508000
oosoonoo
00510000
00511000
00512000
00513000
0051l»000
00515000
00516000
00517000
00518000
oosinooo
00520000
00521000
00522000
00523000
0052UOOO
00525000
00526000
00527000
00528000
00529000
00530000
00531000
00532000
00533000
O053'l000
00535000
-------
TEST3 * 1. 0053POOO
IF (16 .LT. 8.) RETl'RM 00537000
STOP
END
SUBROUTINE CONCE(EK,A,CONS,1,FL1U,FL19)
DIMENSION EK(25),A(22),CONS(10) "OSMOOO
COMMON/TESTS/TEST,TEST2,TEST3,TE?TU 005U2000
A(ll) - COMS(6) OOSU5000
A(13) » CONS(7) 0051*6000
S » CONS(P) no5'»7000
TM - CONS(9) 0051*8000
P * CONS(IO) 005U9000
A(l) = EK(1)/A(13) 00550000
A(2) - FK(6)*EK(13)*EK(l»)*n/A(ll)**2. 00551000
IF (TESTU .EQ. 1.) A(2) = EK(2)/A(13) 005570PO
A(3) = A(11)*A(1)/FK(3) 00553000
o A(U) = A(11)*A(2)/EK(I») 00551*000
0 A(5) = A(11)*A(3)/EK(5) 00555000
A(6) «= A(ll)*A(l»)/EK(6) 00556000
A(7) «= A(13)*A(1)/EX(7) 00557000
A(8) - A(13)*A(2)/EK(8) 00558000
A(9) = A(13)*A(U)/EK(9) 00559000
A(12) = EK(11)/A(11) 00560000
A(10) = A(13)*A(12)/EK(10) 005R1000
A(ll») - FLU* * S 00562000
IF (TEST .EQ. 1.) A(l«*) = EK(1U)/A(13) 005P3000
A(15) = A(11)*A(1U)/EK(15) n-05Ri»000
A(19) = TM * FL19 00565000
IF (TEST2 .EO. 1.) A(l<)) » EK(17)/A(1) 00566000
A(1G) «= A(19)*A(12)/HK(1P) 00567000
A(18) = A(19)*A(I»)/EK(18) 005RPOOO
A(17) = A(19)*A(1)/EK(19) 005P9000
A(20) = A(1
-------
uni
U02
A(13)*A(1I»)/EK(21)
A(19,)*A(2)/EK(22)
A(21) •
A(22) •
RETURN
END
SUBROUT!NE TTEST(C,EK,A,CONS,PCC,PCS,PCM2)
DIMENSION C(35),EK(25),A(22),COMS(10)
COMHON/TESTS/TEST,TEST2,TEST3,TESTU
Tl = TEST
T2 = TEST2
TU = TESTU
ES » CONS(l)
TMG = COMS(2)
C03 = CONS(5)
S » CONS(R)
TM = COMS(9)
P = CONS(in)
PCC = 100.*A(13)*A(2)/FK(23)
IF (TESTU .EQ. 1.) HO TO U01
(PCC .CT. 100.) TEST-U = 1.
TO U02
n =A(11)**2.*EK(2)/(EK(«O*EK(6)*EK(13)*A(13))
IF (P .GT. C03) TESTU =• 0.
IF (TESTI* .EQ. 0.) P = 003
PCS»100.*A(13)*A(1I*)/E.KC2U)
IF(TEST .EQ. 1.) GO TO Un3
IF (PCS .GT. 100.) TEST = 1.
GO TO Unii
S = CUU)
(C(1U) .GT. ES) TEST = 0.
(TEST .FQ. 0.) S = ES
100.*A(19)*A(1)/EK(25)
IF (TEST2 .EO. 1.) GO T^ l»05
IF (PCM? .HT. inn.) TEST? = 1.
no57iono
IF
GO
IF
IF
0057finon
nns77non
nossooon
nnssmnn
nns9nonn
nn5
-------
GO TO UOR 00608000
U05 TM • C(19) ooeogoon
IF (TM .HT. TMG) TEST2 = 0. 00630000
IF (TEST2 .£0. 0.) TM « TMG OOF11000
IF (Tl .ME. TEST) TEST3 = 1. OOP12000
IF (T2 .NE. TEST2) TEST3 = 1. 00613000
IF (TU .ME. TESTI») TEST3 = 1. 0061UOOO
CONS(8) = S 00615000
CONS(9) = TM Onf.16000
CONS(IO) »= P OOF17000
RETURN
END
-------
APPENDIX B.
Plots of the Mass Transfer Coefficient for CCU Absorption vs.
Various Average pH Values.
103
-------
•P
a
«H
-------
o
t/1
C
0>
•rl
O
•H
CO
-1
CO
10
-2
Borgwardt's Data (1974b,d)
6
Liquid Flow Rate, liter/m sec
• 30.0
A 36.2
3.5
Figure B (Con't)
_L
J_
4.0
4.5 5.0 5.5
Average Slurry pH = pH
6.0
7.0
1m
Effect of Slurry pH and Flow Rate on the Overall Mass
Transfer Coefficient for C02 Absorption into Recycled
Lime Slurries in a TCA Scrubber.
-------
•p
a
0)
-------
•H
a
0)
•H
o
•H
«H
«H
0)
o
o
SH
0)
2 Absorption into Recycled
Lime Slurries in a TCA Scrubber .
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-77-026
2.
3. RECIPIENT'S ACCESSION-NO.
4 TI7LEANDSUBTITLEAnalysis and Simulation of Recycle
SO2-Lime Slurry in TCA Scrubber System
5. REPORT DATE"
March 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOH(S)
C. Y. Wen and Fred K. Fong
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
West Virginia University
Department of Chemical Engineering
Morgantown, West Virginia 26506
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
Grant R800781-03-0
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
13. TYPE OF REPORT AND P
Final; 6/74-8/76
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES EPA project officer for this report is R.H. Borgwardt, Mail Drop
65, 919/549-8411 Ext 2234.
i6. ABSTRACT
rep()rt gwQS results of an analysis of flue gas desulfurization by a tur-
bulent contact absorber (TCA) employing lime slurry, including the development of
performance equations for the scrubber-hold tank recycle system. Performance
characteristics investigated include pressure drop of the scrubber, CO2 and SO2
absorptions, and lime utilization. Experimental data obtained from EPA/Research
Triangle Park and TVA/Shawnee Power Station are used for the analysis and cor-
relation. The analysis of CO2 absorption indicates that the overall mass transfer
coefficient is a function of the pH of inlet and outlet scrubber liquor and is very
sensitive to the liquor flow rate. (The rate of SO2 absorption in a TCA has been devel-
oped previously by McMichael et al. , 1976. ) The correlations developed are used to
formulate a simulation procedure for predicting SO2 scrubbing efficiency as a function
of pH of slurry and gas and liquor flow rates. The result of simulation indicates that,
for a given lime feed rate and a fixed inlet and outlet SO2 concentration, a maximum
flue gas flow rate exists which the scrubber can treat by the recycling slurry. An
example is shown for the design of a TCA capable of desulfurizing flue gas from a
50-MW power station.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Flue Gases
Desulfurization
Calcium Oxides
Slurries
Sulfur Dioxide
Absorption
Scrubbers
Circulation
Analyzing
Simulation
Carbon Dioxide
Air Pollution Control
Stationary Sources
TCA
Turbulent Contact Absor
ber
Lime Slurry
13B
21B
07A,07D
07B 14B
11G
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
118
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
108
------- |