EPA-600/2-76 273
October 1976
Environmental Protection Technology Series
EQUILIBRIUM PARTIAL PRESSURE OF
SULFUR DIOXIDE IN ALKALINE
SCRUBBING PROCESSES
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped ;nto
five series. These five 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 five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed
to develop and demonstrate instrumentation, equipment and
methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the
new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U. S. Environmental Protection
Agency, and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
-------
EPA-600/2-76-279
October 1976
EQUILIBRIUM PARTIAL PRESSURE
OF SULFUR DIOXIDE
IN ALKALINE SCRUBBING PROCESSES
by
David K. Oestreich
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
ROAPNo. 21ADE
Program Element No. 1AB013
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
CONTENTS
Page
List of Figures iv
List of Tables iv
Acknowledgments v
Introduction 1
Conclusions 4
Sections
I Equilibrium Partial Pressure of S02 in
Caustic Potash Based Scrubbing Systems 6
Introduction 6
Experimental Approach 6
Discussion of Data 8
Development of Theoretical Model and
Predictive Equation for Partial Pressure 11
II Equilibrium Partial Pressure of SCL in
Sodium Based Scrubbing Systems 26
Introduction 26
Experimental Approach 26
Discussion of Data 26
III Equilibrium Partial Pressure of SCL in
Sodium Citrate Based Scrubbing Systems 28
Introduction 28
Experimental Approach 28
Discussion of Data 29
References 31
iii
-------
FIGURES
No. Page
1 Johnstone's Data for Na2S03/NaHS03 System at 90°C 5
2 Solubility of KS^ in Oxygen- Free Water 15
3 Equilibrium Data at 25°C--PSQ vs. S/C 16
4 Equilibrium Data for [S(IV)] =0.88 Molar at 50°C 17
5 Equilibrium Data for [S(IV)] =2.48 Molar at 50°C 18
6 Equilibrium Data for [S(IV)] =4.98 Molar at 50°C 19
7 Equilibrium Data for [S(IV)] =1.14 Molar at 75°C 20
8 Equilibrium Data for [S(IV)] =3.06 Molar at 75°C 21
9 Equilibrium Data for [S(IV)] =6.12 Molar at 75°C 22
10 Equilibrium Data at 25°C--PSO vs. pH 23
11 Equilibrium Data at 50°C— PSO vs. pH 24
12 Equilibrium Data at 75°C—PSO vs. pH 25
13 Citrate System Equilibrium Data 33
TABLES
No. Page
1 Equilibrium Data for KHSO-j/^SO-j System 9
2 Comparing Experimental P^Q (EP$Q ) with
Values Calculated from Equation (17) 14
3 PSQ Equilibrium Data for Sodium Based Scrubber
Liquor 27
4 Equilibrium Data for the System S02— • Citric Acid--
NaOH-H20 30
5 Pcn Calculated at 90 Percent Confidence Level --
bU2
Citrate System 32
iv
-------
ACKNOWLEDGMENTS
The help of the technicians who made many of the measure-
ments contributing to this report is gratefully acknowledged.
The technicians most directly involved in this work were Messrs.
Don Mitchell and Ron Collier. The help of Mrs. Ellen White and
Mr. Frank Briden was also very valuable in the areas of data
analysis and computer programming. Lastly, the typing skills
of Mrs. Brenda Foil are gratefully acknowledged.
-------
INTRODUCTION
A number of alkaline scrubbing processes are being developed for
S09 control in which the sorbent is regenerated. A notable example of
<- i
this is the Well man-Lord Process . Among other systems proposed is the
2
U. S. Bureau of Mines Citrate Process in which the regeneration is
accomplished by an aqueous Claus or Wackenroder reaction and elemental
sulfur is the product.
The maximum removal efficiency at any point in the scrubbing cycle
of any of these processes is related to the equilibrium partial pressure
of S02 above the solution. Thus, if the equilibrium partial pressure of
SOp above the scrubbing liquor is equal to the partial pressure of S02 in
the incoming gas stream, there will be no removal.
Conversely, during the stripping cycle, steam requirements increase
as the cycle progresses and the equilibrium partial pressure of S0? de-
creases for the scrubber liquor. Consequently, a knowledge of equi-
librium P<-n is required to operate the stripping cycle economically.
5U2
Historically, theoretical equations developed by Dr. H. F. Johnstone"
in 1935 have been used to calculate equilibrium partial pressures for
scrubber design. The equation given by Johnstone is:
SO,
= M
(2S-C)'
C- S
(equation 1)
where
= equilibrium partial pressure of S02, mm Hq
+4
S = total concentration of dissolved S /100 moles
H20
C = Total concentration of the cation associated
with the S+4/100 moles H20
M = a constant depending on T (°K) and the type
of scrubbing system (for the Na2S03/NaHS03 system,
log M = 4.619- 1987/T)
-------
Equation (1) considers only the equilibrium which exists between sulfite
and bisulfite ions and ignores the other important equilibria which exist in
this extremely complex system. Within the framework given by equation (1),
and considering the KHS03/ I^SO-j system, by definition the ratio S/C goes
from 0.5 for a completely stripped solution of KpSO, to 1.0 for a completely
saturated solution of KHSO.,. The reaction during the scrubbing cycle is
thought to be:
S03 = + S02 + H20 scrubbing 2HS03 (equation 2)
^stripping
In reality, the total picture of the complex system is given by:
2SO ~~+ 2H,0+
.
2S02(q) + excess H20/1xi=±2S02*XH20^±2H2S03^=±2HS03 + 2H30 (equation 3)
%
S205"+ H20
It can be seen that, in regions of high bisulfite ion concentration, meta-
bisulfite ions will be formed. Since this equilibrium is not considered in
the Johnstone equation, the equation is invalid in these regions. It might
also be pointed out that the experimental work from which the parameters for
the Johnstone equation were derived was done using a transpiration technique.
In a transpiration experiment, the P-Q of a scrubbing liquor is determined
by bubbling an inert gas such as nitrogen through the scrubber solution and
measuring the mole fractions of the components of the exit gas stream. Thus:
p = S02 p (equation 4)
S02 XS02 + XN2 + XH20 T
where X. = mole fraction i and
PT = total pressure
The mole fraction of S02 is determined iodometrically and,in order to
make a reasonably accurate determination of XSQ , it is necessary to
-------
transpire a considerable amount of SCL from the solution thus causing a
considerable uncertainty in the S/C of the liquor. Another large uncertainty
is present in X,, Q which is obtained from the Literature values of water
vapor pressure, but corrected for the Raoults Law vapor pressure lowering
introduced by the ionic strength of the scrubbing liquor. There is much
uncertainty as to the actual ionic strength of the liquor, because in a sense
this is what is being determined by the total experiment.
Figure 1 shows Johnstone's experimental transpiration data taken at 90°C
for the sodium sulfite/bisulfite system at various total sulfur IV concen-
trations. A semi-log plot of Pcn versus S/C is shown in this plot. The
^2
line through the cluster of points, drawn by the author of this paper to
indicate the apparent insensitivity of Pcn to total sulfur IV concentration,
oUo
appears to fit the points derived from different S IV concentrations equally
well: this contradicts the implications of Henry's Law. Johnstone's equation
may be rewritten:
so,
= MS
[2 -
(equation 5)
'2
Because of the above mentioned inadequacies of the Johnstone equation,
it was decided to do a more definitive equilibrium partial pressure experi-
ment which would hopefully yield a more accurate predictive equation for Pen
as a function of easily measured scurbber liquor parameters. "
-------
CONCLUSIONS
An experimental technique has been developed for determining equi-
librium partial pressures of sulfur dioxide (Pcn ) over various scrubber
9 o /i
liquors. It has been shown that equations developed by H. F. Johnstone '
for equilibrium partial pressure are incorrect.
Experimentally verified theoretical expressions have been developed for
PSQ for strong base (potassium hydroxide and sodium hydroxide) scrubbing
systems and also for the buffered sodium citrate scrubbing system. The
equations developed are capable of predicting PSQ from the input para-
meters--pH, temperature, and concentration of sulfur (IV) in the scrubber
liquor. These equations are presented in a form which includes the expected
uncertainty of the equations at the 90 percent confidence level.
It is recommended that the equations developed in this paper be
used to define maximum thermodynamic control efficiencies achievable for the
scrubbing systems studied. Definition of maximum theoretical control effi-
ciency is essential to scrubber design.
It is suggested that the methodologies described in this paper be
utilized to develop essential relationships and understandings for other
scrubbing systems.
-------
or
i
tu
£
CM
8
100
90
80
70
60
50
40
30
20
10
9
8
7
6
5
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
OS =-7.1 moles SULFUR(IV)/100 moles H20
D S=~5.4 moles SULFUR(IV)/100 moles H20
A S=~3.6 moles SULFUR(IV)/100 moles H20
• S=~4.4 moles SULFUR(IV)/100 moles H20.
0.70
0.80 0.90
S/C, moles S(IV)/ moles K+
Figure 1. Johnstone's data for I^SOa/NaHSOs system at 90 °C.
5
-------
Section I
EQUILIBRIUM PARTIAL PRESSURE OF S02 IN CAUSTIC POTASH BASED
SCRUBBING SYSTEMS
INTRODUCTION
The We11 man-Lord Process as originally conceived involved the scrubbing
of S02 with a concentrated solution of ICSO.,. The reaction product of the
scrubbing cycle is a concentrated solution of KHSO- from which KpSpOj- crys-
tallizes because of its lower solubility. The process then called for
the separation of solid K2S205 from the liquor followed by the dissolution
of the K2$205 in water and then the steam-stripping of S02 from this second
solution. The limited solubility of KoS9^5 Defines tne total possible sul-
fur IV concentration of a liquor which is saturated with S02.
EXPERIMENTAL APPROACH
It was decided to do static experiments rather than dynamic ones, to
determine equilibrium partial pressures. With static experiments, it is
only necessary to measure the mole fraction of S02 present and the total
pressure. Dynamic experiments would require the accurate determination
of the mole fractions of three components as well as total pressure. It
was obvious that the simplest approach would be to keep the total pressure
in the equilibrium vessel equal to atmospheric pressure which is easily
and accurately measured. This was accomplished by using Tedlar plastic
bags fitted with septums as equilibrium vessels.
The Tedlar bags were loaded with simulated scrubber liquors in such
a fashion as to avoid the introduction of oxygen or carbon dioxide into
the vessels. Water used to make the liquors was boiled and sparged with
nitrogen to remove any dissolved gases. This was done to avoid the pos-
sibility of any side reactions such as the oxidation of sulfite to sulfate
and thus assure that the effects measured were caused by shifts in the
equilibrium under consideration and not by random factors.
The loaded Tedlar bags were placed in a constant temperature bath
which was capable of controlling the temperature to ± 0.5°C. It was
-------
determined experimentally that the mole fraction of SC^ existing in the gas
phase in the vessel reached a steady state within 2 hours of the time that -
the vessel was placed in the bath. The bath itself contained ethylene glycol
as the heat exchange medium and was agitated with an air-driven stirrer. The
total capacity of each plastic bag was 200 milliliters with each bag con-
taining 150 milliliters of liquor and approximately 50 milliliters of nitrogen.
The gas phase from each equilibrium vessel was analyzed using a Model 990
Perkin-Elmer Gas Chromatograph (G.C.) equipped with a thermal conductivity
detector. The G.C. column used was 60 cm x 3.175 mm 316 stainless steel tub-
ing filled with Davidson #12 Silica Gel 60/80 mesh. The flow rate was set at
60 milliliters/minute, the column temperature at 130°C, the detector at 200°C,
and the detector current at 275 mi Hi amperes for the determination. Water was
removed from the column after the SOp peak eluted by raising the column temper-
ature to 195°C for 10 minutes.
Prior to the analysis of the gas phase of the bag for mole fraction of
S02, the bags were shaken by hand periodically to assure adequate contacting
of the gas and liquid phases. The gas samples for G.C. analysis were taken
in a 1 cc Hamilton syringe which was heated to the operating temperature of /
the bath.
The concentration of SOg in the gas samples was quantified by comparing
the peak height of SOp in the sample to the peak height of SOp in a 2000 ppm
SOp in nitrogen span gas standard which was purchased from Matheson. The
G.C. analysis procedure was shown experimentally to have a coefficient of
variation of 6.9 percent.
Total pressure was measured using a mercury barometer. The S02 partial
pressure was calculated by the product of total pressure and mole fraction
so2.
pH readings were taken on samples of liquor after the gas phase was
analyzed. These readings were taken on a temperature-compensated, expanded-
scale pH meter using a combination Ag-AgCl/glass electrode.
Samples of the liquor were also titrated iodometrically for total
sulfur IV to confirm that the weighing arid transfer of solid reagents was
quantitative.
-------
Levels of Parameters Utilized
It was decided that the sulfur IV concentration of the liquors which
were to be studied by this experiment would be defined by the solubility
of KoS2^5 at tne vari°us temperature levels. Accordingly, sulfur IV concen-
trations were held at as many as three levels for each temperature level.
These levels were arbitrarily chosen as fully saturated, half saturated,
and one-fifth saturated with respect to ICS^Or.
Supplemental measurements were made to determine experimentally the
solubility of K^SpOg in de-oxygenated water as a function of temperature.
The results of these measurements are shown in Figure 2. These measure-
ments were made by saturating solutions of KpSp^B at various temperatures
and determining sulfur IV iodometrically.
The experimentally determined solubilities are lower than those quoted
by Linke . The solubilities quoted by Linke are shown as the dashed line
on Figure 2. The difference might possibly be due to the de-gasing of the
water used in this experiment. The presence of dissolved oxygen could cause
oxidation of l^SpO,- to K^SCL, a more soluble compound.
The simulated scrubber liquors were made for a given experimental run
by using a mixture of ^2^5 and I^SOg to adjust the S/C's of the liquors
over the S/C range 0.5 to 1.0 but holding S, the total sulfur IV concentra-
tion, constant for all solutions of a given run. PSQ and pH were measured
for each liquor as described previously.
Data was taken for liquors at three different temperatures, chosen
both to bracket the typical operating temperature of a scrubber, and also
to approach the temperature at which steam stripping would take place. The
typical temperature of an operating scrubber is 50°C and the typical temper-
ature of the steam stripper is approximately 100°C. The raw data for the
KHS03/K2S03 system is presented in Table 1.
DISCUSSION OF DATA
In order to compare the experimental data with the Johnstone equation,
Pcn in mm of mercury was plotted against S/C on semi-log plots in Figures
^2
3 through 9. A computer program, written for the Johnstone equation, was
8
-------
Table 1. EQUILIBRIUM DATA FOR KHS03/K2$03 SYSTEM
Molarity
Sulfur IV
1.82
1.82
1.82
1.82
1.82
3.66
3.66
3.66
3.66
3.66
0.882
0.882
0.882
0.882
2.48
2.48
2.48
2.48
4.98
4.98
4.98
4.98
1.14
1.14
1.14
3.06
3.06
3.06
3.06
3.06
6.12
6.12
6.12
6.12
6.12
S/C
0.99
0.98
0.97
0.96
0.94
0.99
0.98
0.97
0.94
0.90
0.99
0.98
0.96
0.92
0.99
0.98
0.94
0.80
0.99
0.98
0.94
0.88
0.98
0.94
0.88
0.99
0.98
0.94
0.88
0.80
0.99
0.98
0.94
0.88
0.80
'«,
irniHg
1.25
0.381
0.194
0.138
0.066
1.32
0.45
0.23
0.065
0.068
2.19
1.09
0.30
0.14
5.62
2.21
0.634
0.070
6.27
2.57
0.701
0.351
5.49
1.19
0.27
17.9
8.74
2.04
0.656
0.210
21.1
9.61
2.16
0.625
0.155
pH
4.6
5.0
5.2
5.35
5.5
4.8
5.15
5.3
5.7
6.0
4.7
5.1
5.4
5.75
4.7
5.0
5.55
6.3
4.95
5.3
5.9
6.36
5.05
5.6
6.0
4.8
5.1
5.65
6.0
6.4
4.9
5.3
6.0
6.35
6.9
Temperature,°K
298
298
298
298
298
298
298
298
298
298
323
323
323
323
323
323
323
323
323
323
323
323
348
348
348
348
348
348
348
348
348
348
348
348
348
-------
used to define the dashed lines on these plots. Johnstone did not give
an equation for "M" for the potassium system; therefore, his equation
developed for the sodium system was used:
log M = 4.619-1987/T (equation 6)
The assumption that the equation for the sodium system may be used for the
potassium system can be rationalized on the basis that sodium and potassium
hydroxides are both very strong bases and are completely ionized in aqueous
solutions.
An examination of Figures 3 through 9 reveals very little similarity
between the data of this work and the line calculated from the Johnstone
equation. It can be seen that on any given plot, the ratio between the
data and the line calculated from the Johnstone equation is not constant.
The differences between the data and the Johnstone equation cannot there-
fore be resolved by an adjustment of the "M" term of the Johnstone equation.
In all cases except for that shown on Figure 7, the Johnstone equation pre-
dicts partial pressures which are high when compared to the data. Although,
the Johnstone equation is actually a fairly good fit of the data shown on
Figure 7, the fit is probably just coincidence.
The practical implications of the differences between the data and
the Johnstone model follow.
First, the actual SOp removal in a KgSOg scrubber which operates at
equilibrium should be better at any given S/C than that predicted by the
Johnstone equation. Secondly, the data predict that the steam require-
ments, for the stripping of SC^ from the pregnant liquor and the resultant
regeneration of the liquor, will be higher than those predicted using the
Johnstone model. This would have a significant effect upon the economics
of the total scrubbing process.
In an attempt to relate partial pressures to the solution parameters
in a simple straightforward manner, the experimental P-0 's were plotted
against pH on semi-log paper. These plots are shown in Figures 10 through
12. As would be expected, these plots show a straight line relationship
which can be expressed mathematically by the use of a least squares fit of
the data.
10
-------
The least squares straight line fits of the data shown in Figures 10
through 12 appear generally to fit the data quite well. An exception to
this is the data point occurring at the lowest pH for each run. As pH's
drop from about 4.25 to 4.00, the line inexplicably turns over. Conse-
quently, the initial data point for each run was dropped for the least
squares analysis.
The anomaly at low pH's is very possibly due to equilibria other
than the first ionization of sulfurous acid coming into prominence. The
formation of metabisulfite is known to be enhanced at low pH's.
DEVELOPMENT OF THEORETICAL MODEL AND PREDICTIVE EQUATION FOR PARTIAL
PRESSURE
As a first approximation, if one considers the chemical equilibria
of interest as given in equation (3), within the pH range of interest,
the only equilibrium of great importance is the first ionization constant
of sulfurous acid:
H2S03 ^ H*
[HSO,]
[H2S03]
K = - =5_ = 1.7 x 10"^ (equation 7)
For all practical purposes, the total S(IV) concentration in solution is
equal to the sum of the sulfurous acid concentration and the bisulfite
ion concentration. Thus with substitution:
- [H2S03]}
i/ _ *••• j ii.-\--/j u-'v—jj; (equation 8)
n
[H2S03]
Solving this expression for [H9SOo]»one obtains:
(elation 9)
In the denominator of equation 9, it is seen that within the pH range
of interest, [H ] is inconsequentially small compared to K-, and may be
11
-------
dropped. (At a pH of 3, [H+] = 1 x 10"3 which is 6 percent of K, and as pH
j_ I
rises [H ] rapidly becomes smaller.)
= [H+] [s"v)] [S(IV)] e'2-303
Thus:
[H SO 1 = LH -< 1SUVJJ = LSdVJJ e -• (equation 10)
2 3 Kl Kl
Henry's Law states that
pcn = Kh [H-jSOo] (equation 11)
jup n £. o
where K. is the Henry's Law constant.
Substitution of equation (10) into equation (11) yields:
Pso = K [S(IV)] e"2'303 pH (equation 12)
where K = Kh/K].
The integrated form of the van't Hoff Equation may be used to relate equilib-
rium constant to absolute temperature and has the form:
In K = a(1P.00) + b. (equation 13)
Taking the data for the potassium sulfite/potassium bisulfite system given in
Table 1, K may be calculated using equation (12) for each data point. A
least squares fit may then be done for the coefficients a and b from equation
(13) using calculated K and measured T for each data point. The resulting
equation for K using the regression analysis coefficients is:
In K = -5.77(1000) + 29-24 (equation 14)
Equation (14) may be manipulated and substituted into equation (12) to
yield:
PSO = [S(IV)] exp |-2.303pH - 5-77(1000) + 2g>24| (equation 15)
12
-------
In the approach taken above, since K is determined empirically from a
theoretical expression which employs measured data, all the uncertainty
existing in the data is translated to an uncertainty in K. The uncertainty
in K is further translated to an uncertainty in the coefficient -5.77 of
equation (14). This uncertainty was calculated at the 90 percent confidence
level using Student's t distribution. Thus equation (14) becomes at the
90 percent confidence level:
in K . -(5.77 ± 0.17 (1000) + 2g_24 (equatl.on „,
and equation (15) becomes:
exp 1-2.303PH~- (5-77 ± 0.17)(1000)+ 29.24 (equation 17)
/ T J
A comparison of experimental PSQ values with PSQ values calculated using
equation (17) is given in Table 2. The lower and upper calculated values
are given for the 90 percent confidence level. Out of the 34 data points
shown in the table, five experimental PSO 's are out of limits whereas
the confidence level predicts statistically that 3.4 data points will be
out of limits.
Equation (17) is demonstrated to be quite usable for engineering
purposes, and far superior to the Johnstone Equation. It should, however,
be remembered that the assumptions made in developing equation (17)
limit the pH range in which the equation is useful. At pH's lower than
3.5 and higher than 6.0 the equation becomes less exact.
13
-------
Table 2. COMPARING EXPERIMENTAL P$() (EP$0 ) WITH VALUES
CALCULATED FROM EQUATION (17)
[S(IV)], M
1.82
1.82
1.82
1.82
1.82
3.66
3.66
3.66
3.66
0.88
0.88
0.88
0.88
2.48
2.48
2.48
2.48
4.98
4.98
4.98
4.98
1.14
1.14
1.14
3.06
3.06
3.06
3.06
3.06
6.12
6.12
6.12
6.12
6.12
pH
4.65
5.00
5.20
5.35
5.50
4.80
5.15
5.30
5.70
4.70
5.10
5.40
5.75
4.70
5.00
5.55
6.30
4.95
5.30
5.90
6.36
5.05
5.60
6.00
4.80
5.10
5.65
6.00
6.40
4.90
5.30
6.00
6.35
6.90
T °K
1 , IV
298.00
298.00
298.00
298.00
298.00
298.00
298.00
298.00
298.00
323.00
323.00
323.00
323.00
323.00
323.00
323.00
323.00
323.00
323.00
323.00
323.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
Exper.
H
mmHg
1.25
0.38
0.19
0.14
0.07
1.32
0.45
0.23
0.07
2.19
1.09
0.30
0.14
5.62
2.21
0.63
0.07
6.27
2.57
0.70
0;35
5.49
1.19
0.27
17.90
8.74
2.04
0.66
0.21
21.10
9.61
2.16
0.63
0.16
Pcn Calculated for 90%
5U2
Confidence Level , mmHg
Lower
0.45
0.20
0.13
0.09
0.06
0.64
0.28
0.20
0.08a
0.91
0.36a
0.18
0.08
2.54
1.28
0.36
0.06
2.87
1.28
0.32
o.na
1.96a
0.55
0.22
9.35
4.68
1.32
0.59
0.23a
14.85
5.91
1.18
0.53
0.15
Expected
0.79
0.35
0.22
0.16
0.11
1.13
0.50
0.36
•0.14a
1.53
0.61a
0.31
0.14
4.31
2.16
0.61
0.11
4.86
2.17
0.55
0.19a
3.19a
0.90
0.36
15.23
7.63
2.15
0.96
0.38a
24.20
9.63
1.92
0.86
0.24
Upper
1.40
0.63
0.40
0.28
0.20
2.00
0.89
0.63
0.25
2.59
1.03a
0.52
0.23
7.29
3.65
1.03
0.18
8.23
3.68
0.92
0.32a
5.20a
1.47
0.58
24.83
12.44
3.51
1.57
0.62a
39.45
15.70
3.13
1.40
0.39
Equation failure
14
-------
H.U
LU
O
O
O
CM
•a
40 50 60
TEMPERATURE, °C
70
90
100
Figure 2. Solubility of
15
in oxygen-free water.
-------
UJ
ac.
o.
CM
O
10
9
8
7
6
5
4
I
[S (IV]]=1.82 moles
USING
JOHNSTONE
EQUATION
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
THIS WORK
0.1
0.90
[S (IV)) = 3.68 moles
USING JOHNSTONE
EQUATION
1.00 0.90
S/C, moles i(IVJ] /moles K .+
Figures. Equilibrium data at 25 °C.
1.00
16
-------
100
%
80
70
60
50
40
30
20
LU
oc
a.
CM
10
9
8
7
6
5
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1"
USING JOHNSTONE
EQUATION
0.70
0.80
S/C, moles [S(IVj] /moles K
0.90
1.00
Figure 4. Equilibrium data for [S(iv2 a 0.88 molar at 50
17
-------
100
90
80
70
60
50
40
30
20
-------
100
90
80
70
GO
50
40
30
20
OO
ce
Q_
Q.
CM
10
9
8
7
6
5
4
3
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
USING JOHNSTONE
EQUATION
V
0.1
0.70
0.80 0.90
S/C, moles [S(IVJ1 /moles K+
Figures. Equilibrium data for §>(IV3 = 4.98 molar at 50 °C.
19
1.00
-------
e
LU
cc
LU
Q£
O.
o.
CM
40
30
20
10
9
8
7
6
5
4
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.70
USINGJOHNSTONE
EQUATION /
0.80 0.90
S/C, moles [S(IVO /moles K +
Figure 7. Equilibrium data for [S(IV|j s1.14 molar at 75 °C.
20
1.00
-------
00
e
III
or
LLJ
Q-
_l
«t
OL
CM
60
50
40
30
20
10
9
8
7
6
4
3
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
USING JOHNSTONE
EQUATION
0.70
0.80
S/C, moles [SdV]] /moles K
0.90
1.00
Figures. Equilibrium data for (S(IV)] = 3.06 molar at 75 °C.
21
-------
CO
in
UJ
CC
CL.
CM
o
t/0
100
90
80
70
60
50
40
30
20
10
9
8
7
6
4
3
SE 2
£
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
USING JOHNSTONE
EQUATION
0.70
0.80
S/C, moles |S(IVj /moles K
0.90
Figures. Equilibrium data for [S(IVJ] r 6.12 molar at 75
22
-------
pH
Figure 10. Equilibrium data at 25 °C.
23
-------
cc
I
UJ
oc
a.
BC.
-------
00
OS
Q_
5
O_
CNI
pH
Figure 12. Equilibrium data at 75 °C.
25
-------
Section II
EQUILIBRIUM PARTIAL PRESSURE OF S02 IN SODIUM
BASED SCRUBBING SYSTEMS
INTRODUCTION
The present versions of the Wellman-Lord Process, the Stone and Webster
Process, and the Double-Alkali Process all utilize the scrubbing of SOp with
a sodium-based scrubber liquor. It was considered to be worthwhile to take
a small amount of equilibrium partial pressure data in the sodium system to
compare the experimental Pcn 's to the theoretical values generated using
OUn
equation (17). This would provide a test of the hypothesis that the identity
of the cationic portion of the alkali contained in the scrubber liquor is of
no consequence as long as we are talking about a strong base.
EXPERIMENTAL APPROACH
The procedures and apparatus used for the acquisition of data for the
sodium system are the same as those described in Section I for the caustic
potash system. For this work, the S(IV) concentration was fixed at two
levels, 2.961^ and 6.012N|. Temperature was held at 50°C and pH was varied
using different mixtures of sodium sulfite and sodium metabisulfite to make
up the liquor. Pcn was again measured with the gas chromatograph.
bU2
DISCUSSION OF DATA
The experimental data is given in Table 3 along with the Pcn 's calcu-
oUp
lated using equation (17). In all but two cases, experimental values fall
within the boundaries defined by the 90 percent confidence level for equation
(17) which was developed for the caustic potash system. This proves the
hypothesis that equation (17) can be used to describe PSQ for any scrubbing
system which employs a strong base.
26
-------
Table 3.
EQUILIBRIUM DATA FOR SODIUM BASED SCRUBBER LIQUOR
[S(IV)], M
6.012
6.012
6.012
6.012
6.012
6.012
2.96
2.96
2.96
2.96
2.96
2.96
pH
4.05
4.50
4.75
5.33
5.5
5.63
4.18
4.52
4.8
5.3
5.67
5.98
Exper.
mmHg
73.1
23.7
11.2
3.58
1.63
b
31.5
12.1
5.52
2.43
0.44
0.20
Pso2
Lower
27.6
9.78
5.50
1.45
0.98
0.72
10.1
4.60
2.41
0.76
0.33
0.16
Calculated
Confidence
Expected
46.6
16.5
9.30
2.45
1.65
1.23
17.0
7.78
4.08
1.29
0.55
0.27
for 90%
Level
Upper
79.0
28.0
15.7
4.14
2.80
2.07
28.8
13.2
6.91
2.18
0.93
0.46
a Temperature = 323°K
Sample lost
27
-------
SECTION III
EQUILIBRIUM PARTIAL PRESSURE OF S02 IN
SODIUM CITRATE BASED SCRUBBING SYSTEMS
INTRODUCTION
It has long been considered that a scrubbing system which is based on a
buffer system other than the sulfite/bisulfite system would offer consider-
able advantage by minimizing the troublesome oxidation of S(IV) to S(VI).
The U. S. Bureau of Mines (USBM) has developed a process based on the
2
sodium citrate buffer system through the pilot plant stage. At a pH of
3.7 the scrubbing reaction probably is:
S02 + H20 + H2Cit~ ^==^ HSOg t H3Cit (equation 18)
(In reference (2), the USBM postulated the formation and existence of an
intermediate complex between bisulfite and citric acid; however, the data
of this work does not support the contention.)
The bulk of the USBM data, showing the equilibrium S02 concentration
of the gas phase as a function of the S(IV) concentration in the scrubber
liquor and temperature, was taken for PSQ 's which are much higher than
the inlet concentrations would be for power plant flue gas.
It is of interest to consider the possible application of the citrate
process to power plant flue gases having S02 concentrations of approximately
3000 ppm. The maximum removal efficiency is defined by the equilibrium
partial pressure of S02 for a given set of liquor parameters.
EXPERIMENTAL APPROACH
The basic mechanics of conducting a static equilibrium partial pressure
experiment were discussed in Section I with reference to the caustic potash
based scrubbing system. The same apparatus was used for the citrate scrub-
bing system work.
28
-------
For this work, the liquor concentrations for [S(IV)J and citric acid
were set at approximately the same levels corresponding to inlet gas S02
concentrations of ^2500 ppm which are indicated in the USBM work. Thus, the
liquor sulfur (IV) loading was fixed at two levels —0.0264M and 0.0557M--
which are equivalent to 1.69 and 3.57 g SC^/liter, respectively. The
citrate concentration was fixed at 0.834NL The pH was varied over the range
of interest by the addition of different amounts of solid sodium hydroxide.
Three levels of temperature were chosen— 303, 323, and 348°K-- to bracket
the range of interest.
DISCUSSION OF DATA
The raw data obtained from this effort are given in Table 4.
Individual runs are presented as semi-log plots of P_n versus pH for
oUo
various temperatures in Figure 13. The lines drawn through the data points
are defined by least squares analysis of the data and the equations given
utilize the coefficients of the regression analysis.
The effect of total citrate concentration was not studied experimentally;
however, it can be seen from the USBM work that at a given mole ratio of
H3 Cit to NaHp Cit, the effect of raising the total citrate concentration
only increases the buffer's capacity to resist a change in pH. As the cap-
acity of the buffer is increased, the capacity of the liquor for S(IV)
increases.
Thus, holding the total citrate concentration constant, it is found that
PSO is a function of pH, [S(IV)], and temperature. This relationship has
the same form as equation (12) developed for the caustic potash system:
Pen = K CS(IV)] e~2'303 pH (equation 12)
bU2
where:
In K = aHOOO) + b (equation 13)
The coefficients a and b were again calculated by doing a least squares fit
of the K's calculated for each data point using equation (12). The re-
sulting expression for the equation at the 90 percent confidence level is:
In K = -(8.20 ± 0.23) (100°) + 37.57 (equation 19)
29
-------
Table 4. EQUILIBRIUM DATA FOR THE SYSTEM S09—CITRIC ACID—NaOH-H?0
[S(IV)], M
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
PH
3.57
3.65
3.83
4.02
4.11
4.25
3.50
3.69
3.80
4.00
4.10
4.18
4.26
3.65
3.78
3.84
3.92
4.10
4.24
4.33
3.30
3.52
3.65
3.72
3.85
4.00
4.10
T,°K
323.00
323.00
323.00
323.00
323.00
323.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
348.00
303.00
303.00
303.00
303.00
303.00
303.00
303.00
Exper.
Pso2
mmHg
1.35
0.96
0.70
0.45
0.28
0.17
8.84
7.59
4.79
3.89
2.22
1.43
1.01
20.15
16.60
11.50
8.99
7.12
4.92
3.52
1.39
0.98
0.65
0.38
0.27
0.20
0.12
30
-------
The equation for partial pressure becomes:
PSQ - [S(IV)] exp |-2.303pH -(8.20 ± 0.23) C1000U 37.57^ (equation 20)
Please note that equation (20) is valid only for a total citrate concentration
of 0.834 FL The coefficients for equation (19) may be different for different
total citrate concentrations.
It can be observed that the coefficients a and b for the caustic potash
system are quite different from the coefficients derived for the citrate
system. This can be expected not only for the reasons given above, but also
because the two experiments represent large differences in the ionic strengths
of the scrubber liquors. These differences in ionic strengths can have ex-
treme effects upon the activity coefficients for the equilibria under considera-
tion.
A comparison of experimental PSQ values to PSQ values calculated at
the 90 percent confidence level using equation (20) is given in Table 5.
REFERENCES
1. Miller, L.A. and Terrance, J.D. Process for Recovering Sulfur
Dioxide from Gases Containing Same. (Wellman-Lord, Inc., Lakeland,
Fla.) U.S. Pat. 3,485,581. 5p., Dec. 23, 1969. 4 refs. (Appl.
Nov. 15, 1966, 20 claims).
2. Rosenbaum, J.B. et al. The Citrate Process for Removing S02 and
Recovering Sulfur from Waste Gases, Proceedings AIME Environmental
Quality Conference, Washington, D.C., June 7-9, 1971.
3. Johnstone, H.F. et al. Recovery of Sulfur Dioxide from Waste
Gases. Industrial and Engineering Chemistry, 30. 101 - 109,
January 1938.
4. Perry, J.H. Chemical Engineers' Handbook (Fourth Edition),
Chapter 14, page 11. New York, McGraw-Hill, 1963.
5. Linke, W.F. Solubilities, Inorganic and Metal Organic Compounds.
4th ed. Volume II p. 295, Washington, American Chemical Society,
1958.
31
-------
Table 5. P$() CALCULATED AT 90 PERCENT
CONFIDENCE LEVEL-CITRATE SYSTEM
[S(IV)L M
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0264
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
0.0557
PH
3.57
3.65
3.83
4.02
4.11
4.25
3.50
3.69
3.80
4.00
4.10
4.18
4.26
3.65
3.78
3.84
3.92
4.10
4.24
4.33
3.30
3.52
3.65
3.72
3.85
4.00
4.10
T, °K
323.
323.
323.
323.
323.
323.
348.
348.
348.
348.
348.
348.
348.
348.
348.
348.
348.
348.
348.
348.
303.
303.
303.
303.
303.
303.
303.
Exper.
Pso2
1.35
0.96
0.70
0.45
0.28
0.17
8.84
7.59
4.79
3.89
2.22
1.43
1.01
20.15
16.60
11.50
8.99
7.12
4.92
3.52
1.39
0.98
0.65
0.38
0.27
0.20
0.12
P
Lower
0.68
0.57
0.37
0.24
0.20
0.14
5.21
3.36
2.61
1.65
1.31
1.09
0.91
7.78
5.77
5.03
4.18
2.76
2.00
1.63
0.48
0.29
0.21
0.18
0.13
0.10
0.08
SQ Calculated at 90$
Confidence Level, mmH
-------
e -2.36 pH +11.57
+13.40
Psoe-3.18pH+11.00
ULFUR(iyj=0.0557 molar AT75°C
|ULFUR(IVJ =
0.0264 molar AT 50 °C
iULFUR(IV))=
0.0264 molar
A AT 75 °C
0.0557 molar AT30°C
4.5
PH
Figure 13. Citrate system equilibrium data.
33
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-76-279
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Equilibrium Partial Pressure of Sulfur Dioxide in
Alkaline Scrubbing Processes
5. REPORT DATE
October 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
David K. Oestreich
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Chemical Processes Branch
Industrial Processes Division, IERL-RTP
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
1AB013: ROAP 21ADE
11. CONTRACT/GRANT NO.
NA
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
In-house; 1970-71 (Lab Work)
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The report gives results of IERL-RTP in4iouse studies in which equilibrium
partial pressure of SO2 was measured as a function of pH, temperature, and concen-
tration of sulfur (3Y) on various scrubber liquors. These studies were done for
potassium-, sodium-, and citrate-based scrubbing systems. It is shown that
equations developed by earlier workers for predicting SO2 partial pressures are
incorrect. Theoretical expressions are developed to relate the equilibrium partial
pressure of SO2 to the important scrubber parameters. These expressions are
experimentally validated at the 90 percent confidence level.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Flue Gases
Scrubbers
Sulfur Dioxide
Equilibrium
Partial Pressure
Potassium
Sodium
Citrates
Air Pollution Control
Stationary Sources
Alkaline Scrubbing
Equilibrium Partial
Pressure
13B
2 IB
07A
07B
07D
07C
3. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
34
------- |