Prepared for the
Office of Air Programs
ENVIRONMENTAL PROTECTION AGENCY
by the
TENNESSEE VALLEY AUTHORITY
1972
SULFUR OXIDE REMOVAL
FROM POWER PLANT
STACK GAS
study of the effect of organic acids on
limestone scrubbing process
-------
Tennessee Valley Authority
Division of Chemical Development
Fundamental Research Branch
Sulfur Dioxide Removal From Power Plant Stack Gas
STUDY OF THE EFFECT OF ORGANIC ACIDS
ON THE WET-LIMESTONE SCRUBBING PROCESS
By
J. D. Hatfield, Y. K. Kim, and R. C. Mullins
Prepared for
Office of Air Programs
Environmental Protection Agency
U. S. Department of Health, Education, and Welfare
Contract No. TV-34425A
Mqscle Shoals, Alabama
February 24, 1972
35660
-------
STUDY OF THE EFFECT OF ORGANIC ACIDS
ON THE WET-LIMESTONE SCRUBBING PROCESS
SUMMJI..R Y
The dissolution of limestone is one of the rate-limiting
steps in the process for removal of sulfur dioxide from stack gases
by scrubbing with limestone slurry. The rate of dissolution can be
increased by adding an acid that is stronger than carbonic acid
(Kl = 4.4 X 10-7) and weaker than sulfurous acid (Kl = 1.3 X 10-2).
In a search for a suitable additive, measurements were made of the
solubility of calcium and magnesium carbonates in solutions of 27
weak organic acids that were selected on the basis of their low cost
and availability from a large number of acids that meet the primary
requirements. From these results, four acids--benzoic, phthalic,
adipic, and glycolic--were chosen for further study of their physico-
chemical properties that would affect their potential usefulness in
a scrubber system. The acids suggested in the literature (U. S.
Patent 3,632,306, January 4, 1972) for this purpose--formic, acetic,
and propionic--were considered unsuitable.
A high degree of acid stability was demonstrated by tests
under conditions much more drastic than those likely to be encountered
in scrubbing operations. Gas mixtures containing as much as 6710 S02
and 33% O2 were passed through aqueous solutions of the acids at 75°C
for as long as 59 hours without producing any change in the ultra-
violet absorption patterns of the solutions. Molten acids exposed
to similar gas mixtures rich in S02 and O2 showed the same ultraviolet
peaks as solutions of the untreated melts.
The solubilities of calcium and magnesium carbonates in
0.05, 0.1, and 0 210 solutions of each of the four acids were measured
at 25°C and in 0.1% solutions at 50°C. Within experimental error the
solubilities at the two temperatures were the same. The concentration
of the acid is the limiting factor up to 50°C. Saturation solubilities
at 25° and 50°C of the calcium and magnesium salts of the four acids
were obtained also, The solubilities of the salts of the monobasic
acids increased with rising temperature and those of the dibasic
acids decreased, Arranged in the order of the calcium concentration
in the saturated solutions at 50°C they are glycolic> benzoic>
adipic> phthalic. Phthalic acid gives only about one tenth the
concentration of calcium as the other acids.
i
-------
Measurements were made of the pH of both the solutions formed
by dissolution of the carbonates and the saturated solutions of the
calcium salts. The solubility and pH data were analyzed by a computer
program to evaluate the concentration of the species in the solutions
and to determine the stability constants of the complexes. No con-
sistent values of dissociation constants for complexes of calcium
or magnesium with phthalic, adipic, or glycolic acid were obtained.
The value for the equilibrium constant of the weak complex CaBz+ was
calculated to be
log K* = -2.13 ~ 0.15 at 25°C
log K* = -1.0 ~ 0.5 at 50°C
Measurements were made of the ultraviolet absorptions of
calcium benzoate solutions, and the activities of the calcium ion
were measured with the calcium ion specific electrode. The results
were treated by Job's method of continuous variation in an attempt
to determine whether there is an aqueous complex of calcium benzoate.
If such a complex exists it is too weak to be detected by the methods
of measurement used.
The vapor pressure of a benzoic
liter) was measured at three temperatures
difficulties with vapor absorption on the
The data can be expressed by the relation
acid solution (1.6 grams/
by a dynamic method after
equipment were overcome.
7
log P = 9.20 - 478JT
where P is pressure, atmosphere, and T is temperature, oK. The relation
is consistent with the accepted value for the heat of vaporization.
Calculations were made, from the data and published values for ioniza-
tion constants and activity coefficients, of the vapor pressure over
this concentration of solution for a range of temperature and pH. The
results indicate that the loss of acid as vapor under scrubbing con-
ditions would be relatively low.
The effect of the acids on the oxidation of sulfite to
sulfate was studied also. Exploratory tests in the sodium system
showed that the acids promote oxidation. In the calcium system at
50°C and pH 4, 0.11 solutions of the acids inhibited the oxidation
of sulfite by pure oxygen in the order benzoic> glycolic> phthalic>
adipic. From plots of the rates of oxidation for the 0.11 solutions
the first-order rate constants were determined. Measurements were
made of the effect of pH on the oxidation of calcium sulfite without
an organic acid and the relationship between oxidation rate and pH
under the test conditions was shown to be linear between pH 3 and 4.5.
The data from the oxidation tests were processed by a computer program
to calculate the species concentration and activities from the con-
stants and activity coefficients suggested by the Radian Corporation.
ii
-------
The temperature dependence of benzoate equilibrium constants
in the form used by Radian Corporation (PB 193 029)
log K = -AT-l - B log T - CT + D
are shown in the tabulation
A B C D
HBz ~ H+ + Bz- 804.7 0 0.0090476 1.192
CaBz+ ~ Ca++ + Bz- 4354.5 0 0 12.475
HBz(g) ~ HBz(aq) - 4500. 0 0 -10.27
(Henry's law)
These constants, when incorporated into the Radian equilibrium program,
will describe the properties of slurries containing benzoic acid when
used in the wet limestone scrubbing process.
iii
-------
CONTENTS
Introduction. . , . . . . .
. . . . .
. . . . .
. . . . . .
Survey of Weak Organic Acids.
. . . . . .
. . . . . . . . .
Stability of Organic Acids Under Scrubbing Conditions
Solubility Studies With Selected Organic Acids
. . . .
Measurements of 25°C
. . . . . . . . .
. . . . . . .
Constants Derived From Solubility Measurements
. . . .
Solubilities of Salts of Organic Acids at 50°C
. . , .
Complexes in Solutions. .
. . . . . . . . . .
. . . . . . .
Volatility of Organic Acids Under Scrubbing Conditions. . .
Effect of Organic Acids on Oxidation of Sulfite
. . . . . .
Oxidation of Sodium Sulfite. .
. . . .
. . . . .
Oxidation of Calcium Sulfite
. . . . . . . . . .
iv
Page
1
2
9
17
17
21
28
32
41
46
46
51
-------
STUDY OF THE EFFECT OF ORGANIC ACIDS
ON THE WET-LIMESTONE SCRUBBING PROCESS
INTRODUCTION
In the wet-limestone process for the removal of sulfur
dioxide from power-plant stack gases, the gases are scrubbed with
an aqueous suspension of finely ground limestone and the dissolution
of the limestone is the rate-limiting step. It was proposed that
an organic acid be added to the system to accelerate the dissolution
of the limestone and thereby increase the scrubber efficiency, and
this report covers an investigation of the use of organic acids for
this purpose.
The work covered by this report was intended to extend over
2 years, beginning July 1, 1910. Shortly after the end of the first
year, however, financial support for the work was withdrawn and,
except for completing work in progress, the research was terminated.
This report covers work done in the period July 1, 1910, to September
30, 1911.
In the work covered by this report, several phases of the
investigation were carried out simultaneously, and each portion of
the research was reported as it was completed, so that in chronological
order the reports would be difficult to follow. The monthly reports
have been rearranged to combine all the work on each phase in a single
section, and the dates on which the results were reported are retained.
In lieu of a completion report, this report is a compilation
of the monthly progress reports on the work. Since each progress
report was essentially a complete unit, the tables and figures were
numbered separately for each report. These numbers are retained here,
and, since the tables and figures are inserted in the text in the order
in which they are referred to, the lack of unified series of numbers
for the tables and for the figures should not be confusing.
-------
-2-
SURVEY OF WEAK ORGANIC ACIDS
The rat.e-determining step in the wet limestone scrubbing
process for the removal of 802 from stack gases is the dissolution
of the limestone in the scrubbing slurry. The dissolution rate is
affected by th~ particle size and surface area of the limestone,
the source and mineralogical properties of the limestone, the tempera-
ture, and the acidity of the slurry. The most important factor that
affects the dissolution rate of a limestone is its solubility in the
liquid phase of the slurry which gives the driving force for dissolu-
tion. The calcium and magnesium salts of many organic acids are soluble
and the addition of organic aci~s to the scrubbing slurry should improve
the wet limestone scrubbing process. The Applied Research Branch has
demonstrated improvement by the use of benzoic acid, and same of the
lower aliphatic acids have likewise shown promise in improving the
process.
Measurements were made of the solubilities of calcium and
magnesium carbonates in solutions of selected organic acids and the
absorbances of the solutions in the ultraviolet region were determined.
Several of the organic acids from the list prepared by the
Applied Research Branch were chosen to test their solubilization of
CaCOs and MgCOs. The acids were selected to give a wide variation in
formula and structure and include aliphatic and aromatic acids, acids
with different chain lengths, with different substitutions of other
groups for hydrogen, with different degrees of saturation, and with
different numbers of carboxy groups in the molecule.
One gram of the organic acid (or 1 ml. if the acid is liquid)
was added to 100 ml. of water, the pH was measured, and 5 grams of
reagent grade CaCOs or MgCOs was added to the solution or mixture. The
mixtures were allowed to stand at room temperature with shaking at
irregular intervals. The pH was measured and lO-ml. aliquots of the
clear solution were taken after 3 days and after one week. The results
are shown in the table.
For a given weight of acid, the aliphatic acids were most
effective in solubilizing calcium or magnesium carbonate and the lower
molecular weight acids were superior to the acids of higher molecular
weight. The substitution of chlorine, hydroxy, or phenyl groups for
hydrogen decreased the solubility, but apparently unsaturated acids
were slightly more effective than their saturated analogues. The
polycarboxylic acids were effective solubilizers with the exceptions of
oxalic, tartaric, and citric acids for calcium. The aromatic acids
were less effective than the lower aliphatic acids, and any substitu-
tion on the benzene ring decreased the solubility of the cation.
July 1970 - FR
-------
-3-
Use of Organic Acids in the Dissolution of Calcium and Magnesium Carbonates
pH of 801., Ca, wt.~, After 7 days 801., Mg" wt. '1>, After 7 days
acid after mole rat60 after mole ratfio Price
Acid Formula solution 3 days 7 days ac1d:Ca ...E!!.. 3 days 7 days ac1d:Mg -E!!... dollars/lb.
Aliphatic monocarboxylic acids
Formic HCOOH 2.30 0.48 0.54 1.7 7.00 0.40 0.41 1.4, 7.60 0.147
Acetic CHsCOOH 2.30 0.32 0.37 1.9 7.20 0.31 0.31 1.4 7.65 0.09
Chloroacetic CICH2COOH 2.15 0.21 0.21 2.0 7.15 0.20 0.20 1.3 7.85 0.21
Glycolic HOCH2COOH 2.40 0.22 0.25 2.1 7.40 0.24 0.27 1.2 7.50 0.10
Phenylacetic C~HsCH2COOH 2.70 0.13 0.15 2.0 7.40 0.14 0.13 1.4 8.15 0.68
Propionic C2HsCOOH 2.90 0.28 0.28 1.9 7.20 0.27 0.21 1.6 7.75 0.147
Lactic CHsCHOHCOOH 2.50 0.21 0.23 1.9 6.90 0.21 0.21 1.3 7.70 0.275
Butyric C3H7COOH 2.90 0.22 0.21 2.2 6.65 0.23 0.21 1.3 7.50 0.33
Caproic CsHuCOOH 3.00 0.17 0.17 2.0 6.15 0.15 0.12 1.7 7.90
Gluconic C~OH(CHOH)4 2.80a 0.06 0.06 3.4 7.00 0.06 0.08 1.6 8.00 0.145
COOH
Acrylic CH2=CHCOOH 2.60 0.31 0.31 1.8 6.60 0.25 0.21 1.6 7.70 0.31
Oleic CH3 ( CH2 hc=c
(CH2 hCOOH 3.80a 0.01 0.01 16.7 6.60 0.03 0.03 3.4 8.45 0.23
Aliphatic polycarboxy1ic acids
Oxalic COOHCOOH 1.75 0.01 0.01 44.4 7.70 0.13 0.08 3.4 8.35 0.22
Succinic COOHCH2CH2COOH 2.30 0.27 0.32 1.1 7.35 0.29 0.30 0.69 7.80 0.62
Tartaric COOHCHOHCROH 2.25 0.01 0.01 26.7 7.70 0.24 0.26 0.62 7.85 0.415
COOK
Malic COOHCHOHCH2COOH 2.408 0.09 0.09 3.3 7.60 0.22 0.24 0.76 8.40 0.315
Fumaric COOHCH=CHCOOH 2.35a 0.25 0.28 1.2 7.30 0.28 0.27 0.78 8.00 0.225
Maleic H02CCH=CHC02H 1.95 0.23 0.31 1.1 7.25 0.31 0.30 0.70 7.70 0.48
Adipic COOH(CH2)4COOH 2.80 0.28 0.28 1.0 7.20 0.26 0.21 0.79 7.80 0.18
Citric (COOHCH2 ),eOOH 2.25 0.02 0.02 10. 7.80 0.24 0.24 0.53 8.50 0.33
COOH
Aromatic acids
Benzoic CeHsCOOH 2.75a 0.14 0.17 1.9 7.35 0.04 0.13 1.5 8.20 0.215
Salicylic HOCeI4COOH 2.608 0.11 0.14 2.1 7.35 0.21 0.13 1.4 8.1' O.lJ25
P-Amino NH2CeHeCOOH 3 . 5r:P 0.15 0.15 1.9 7.40 0.14 0.14 1.3 8.00 1.72
benzoic 2.80a
Dinitro (N02)2CeHs 0.09 0.10 1.9 7.45 0.09 0.09 1.3 8.10
benzoic COOH
Gallic (HohceH7 2.85 0.11 0.11 2.1 6.20 0;12 0.11 1.3 8.15 2.65
COOH
Phthalic CeH4(COOH)2 2.45a 0.12 6.06 4.0 7.60 0.22 0.21 0.70 8.05 0.12
a-Naphthoic CloH7COOH 3.75a 0.02 0.01 23 6.70 0.08 0.06 2.3 8.30
a 1 gram of acid did not completely dissolve in 100 ml. H20.
b
Based on solubilities after 1 week, assuming all acid dissolved.
July 1970
FR
-------
-4-
The columns in the table that show the mole ratio of acid
used to metal solubilized indicate that one calcium atom reacts with
two carboxy groups for those acids having appreciable solubilizing
power; for magnesium the number of carboxy groups required is somewhat
less--approxjmately 1.5. On the basis of solubility alone, it is
concluded that the best additives are those of low molecular weight,
since a given weight of the additive would contain more acidic groups
to react with the calcium or magnesium. Other factors such as cost
and losses due to volatility and decomposition under scrubbing condition
need to be considered, however.
Of the aliphatic monobasic acids, acrylic, glycolic, and
chloroacetic acids appear promising, provided they are stable and have
sufficiently low vapcr pressures under scrubbing conditions. It is
probable that chloroacetic acid may hydrolyze to glycolic acid under
scrubbing conditions; this type of reaction may occur with any halogen
substituted acid. The dicarboxylic acids that appear promising are
adipic, succinic, maleic, and fumaric, provided they are stable and
nonvolatile. Benzoic appears to be the most promising aromatic acid,
although phthalic acid is very effective for magnesium. Since phthalic
acid is inexpensive it should be tested for calcium at the scrubbing
temperature of 55° C.
Ultraviolet Absorbance of Benzoic Acid: Since the Applied
Research Branch has shown the effectiveness of benzoic acid in improving
the wet scrubbing process, the stability of the additive with continued
use is important to the economics of the process. A method for rapid
analysis of the additive in the solution phase of the slurry is being
tested with the Cary 17 UV spectrometer.
The absorbance spectra of benzoic acid at room temperature
showed two distinct peaks at 193 ~ and 230 ~ and a broad shoulder
at about 270 IIlIJ.. The molar absorptivity of benzoic acid was found to
be 4.015 x 104 and 9.964 x 103 at 193 ~ and 230 ~, respectively.
Plots of the absorbances at the two peaks against the concentration in
solution, Figure 1, show that Beer's law is valid for the system from
o to 5 ppm benzoic acid. The absorbances were only slightly affected
by the pH of the solution when it was varied by the addition of calcium
or sodium hydroxide. Thus by diluting the solution to this concentra-
tion range we can measure the benzoate concentration very accurately,
and can detect any loss of the acid QY decomposition or vaporization.
July 1970 - FR
-------
-5-
14
2
12
10
< 8
...
OJ
()
~
~
o
fI)
~ 6
4
o
o
1
2
3
4
5
Benzoic acid, ppm
Fip:ure 1
Calibration Curves for Analysis for
Benzoic Acid With Cary 17 UV Spectrometer
July 1970 - FR
(J. D. Hatfield, Y. K. Kim, V. L. Bulger)
-------
-6-
Other Absorption Studies: The
acid are probably due to the species Bz-
The ratio of these species is related to
the ionization constant, K,
two absorption peaks of benzoic
and HBzo that exist in solution.
the pH of the solution through
K = [Bz-J[~J/[HBzJ
(3 )
This implies that careful control of pH is needed to measure the concentra-
tion of benzoic acid in scrubber solutions as well aR the effects of cations
A series of solutions cont~ining a constant amount of benzoic
acid, 4 X 10_4M, and increasing amounts of base--NaOH or Ca(OH)2--were
prepared; the ionic strength of each solution was maintained at about
0.1M with sodium perchlorate. These simulated titration curves are shown
in Figure 4, and the absorbances at 193 and 227 ~m are shown in Figure 5.
There was very little difference between the "titration" curves with
NaOH and those with Ca(OH)2' The absorbance at 193 ~ increased steadily
as base was added until the equivalent point was reached, after whioh the
increase was slower and more erratic. There was a slight but distinct
difference between the absorbance of solutions containing sodium and of
those containing calcium. The absorbance at 227 IJlD. decreased as base was
added and then became approximately constant beyond the equivalence point.
These results will be evaluated further to assess the effect of
cation constituent on the absorbances, and the effect of solution com~
position in tne analysis of benzoate by ultraviolet absorption. The data
also will be analyzed to determine the ionization constant of benzoic
acid at 25°C and at an ionic strength of 0.1, and attempts will be made to
evaluate hypotheses concerning complexes of calcium and benzoate ions and
to obtain the magnitude of' the strength of such complexes to explain the
experimental data.
Similar studies will be made with magnesium benzoate and witn
calcium and magnesium salts of adipic, glycolic, and phthalic acids.
December 1970 - FR
-------
10.0
9.0'
8.0
~
7.0.
5.0
6.0
4.0
-7-
o NaOH
D Ca(OH)2
o 0.5
Apparent equivalents of base per equivalent of benzoic acid
Figure 4
Titration Curve of Benzoic Acid
With Sodium and Calcium Hydroxide
December 1970 - FR
-------
-8-
1.9
193 IDIJ.
1.8
1.7 .
Q)
~
.01.
f.t
o
fIJ
~
o - NaOH
D - Ca(OH)2
0.35
227 m.~
o
D
0.30
o
0.5 1 1.5 2.0 2.5
Apparent equivalents of base per equivalent of benzoic acid
Fi~e 5
Ultraviolet Absorbance of Calcium and
Sodium Benzoate Solutions
Decem.ber 1970 - FR
(J. D. Hatfield, Y. K. Kim,
R. C. Mullins)
-------
-9-
STABILITY OF ORGANIC ACIDS UNDER SCRUBBING CONDITIONS
Stability of Benzoic Acid: The stability of benzoic acid was
studied under conditions more drastic than those expected to be encountered
under scrubbing conditions to accelerate its decomposition or oxidation and
to identify any products of decomposition that may be formed.
In one test, about 700 ml of a 0.3% benzoic acid solution was
placed in a l-liter flask maintained at 75°C in a water bath. A gas mixture
(50i S02, 25% 02' and 25% N2) was passed through a coarse fritted glass
into the liquid at a rate of 200 ml/min. Samples of the solution were
withdrawn at regular intervals for determination of the ultraviolet
absorption pattern. After 51 hours of aeration with the gas mixture at
75°C, the ultraviolet absorption of the solution was compared with that of
pure benzoic acid. The absorption peaks at 193 and 227 nm were unchanged
from the treatment, and there were no new absorption peaks. (The wave
lengths were erroneously listed in the December report as ~; these should
be nm or ~).
In another test, about 20 grams of pure benzoic acid was placed
in a scrubbing tube, and the assembly was placed in a furnace at 136° = 2°C
to give a molten product (benzoic acid melts at 122°C). A gas mixture
(67% S02, 33% 02) was passed through a fritted glass into the melt at the
rate of 45 cc/minute. Some benzoic acid crystals condensed in the cool
portion of the apparatus during the run. After bubbling the gas through
the molten benzoic acid for 18 hours, the assembly was removed from the
furnace and the melt was allowed to solidify at room temperature. Samples
of the same weights of the melt, the condensed acid on the wall of the
scrubbing tube, and pure benzoic acid were dissolved in the same amounts
of water, and the ultraviolet absorption patterns and peak heights of the
three solutions were determined as shown in the figure. No new absorption
peaks were found in the melt or in the condensate, and these two materials
were identical in purity with the benzoic acid.
These tests, made under more severe conditions than would be
encountered in pilot-plant operation, indicate that benzoic acid is quite
stable and is not decomposed by S02 and oxygen.
January 1971 - FR
-------
1.8
1.7
1.6
-10-
A - Pure benzoic acid
B,C - Melt and condensed benzoic
J.cid
0.8
0.7
0.6
1. 0.5
Q) Q)
() ()
~ a
of: of:
o 0.4 g
ro1.
~ ~
L
1.
1.
1.0
January 1971 - FR
200
250
Wavelength, nm
Ultraviolet Absorption of Benzoic Acid
0.3
0.2
0.1
0.0
300
-------
-11-
Stability of Organic Acids: Accelerated measurements were made
of the stability of the 4 organic acids--benzoic) phthalic, adipic, and
glycolic--being studied as additives to the wet-limestone scrubbing process.
The results for benzoic acid were reported in January; this report describes
tests of the other three acids and summarizes all the results. All tests were
made under conditions more drastic than those expected to be encountered under
scrubbing conditions in order to reduce the time required for the tests and
to identify any products of decomposition.
In one series of tests, the organic acids were dissolved in water,
and the S02 gas mixture was bubbled through a coarse fritted glass into the
solution which was maintained at constant temperature in a water-bath.
Conditions of the tests for each organic acid are summarized in Table I. In
another series of tests, the pure acid (solid) was placed in the bottom of a
scrubbing tube, the assembly was heated in a furnace at a fixed temperature
above the melting point, and the gas mixture was passed through a fritted
glass into the melt. The conditions of these tests a130 are summarized in
Table I. Glycolic acid was not treated in this manner because it was available
only as a 70% aqueous solution.
In the first series of tests, the solutions after scrubbing had a
strong odor of S02 and exhibited strong peaks of ultraviolet light absorption
at 188 and 207 nm which interfered with the organic acid absorbance measurements.
The odors and the absorbance peaks due to 802 were completely eliminated by
boiling for about 10 minutes on a hot plate. The ultraviolet light absorbances
of the acid solutions were measured after this treatment and after appropriate
dilution.
Phthalic, adipic, and glycolic acids, after treatment to expel the
S02, showed essentially the same spectrum as the pure acids at the same concen-
tration (Figures 1, 2, 3). The slight differences between the absorption
curves for the treated solutions of phthalic and glycolic acid and the reference
curves probably are due to the loss by evaporation of the acid during heating
to expel the S02 fram the solutions; there appeared to be little or no loss
of adipic acid in the treatment. No new absorbance peaks were observed in
any of the acids.
March 1971 - FR
-------
-12-
TABLE I
Reaction Conditions for the Organic Acid Stability Measurement
Gas compn.,
Time, Temp., parts by volume now rate,
Acid hours °c Sample SO,2 ~ !& cc/min
Tests with solutions
Benzoic 51 75 1ooml, 0.3~ soln. 2 1 1 200
Phthalic 53 75 400ml, O.lj soln. 2 1 0 120
Adipic 59 75 500 ml, 2j soln. 1 1 0 40
Glycolic 28 55 200 ml, 7(1fD soln. 2 1 0 45
Tests with melts
Benzoic 18 136 20 grams 2 1 0 45
Phthalic 0.5 212 60 grams 2 1 0 4'5
Adipic 10 162 30 grams 2 1 0 45
March 1971 - FR
-------
2.5
..
I
1
~
1.0
0.5
o
March 1971 - FR
-13-
---- Original solution
- - - After treatment
,
,
,
,
,
~
,
,
,
,
,
,
,
,
~
\
~-.&'....
o 2
Wavelength, nil
300
Figure 1
Ultraviolet Absorbance ot Phthalic Acid
(0.1' lolution in a l-mm cell)
-------
<
..
cu
o
~ 0.8
~
o
rI2
~
1..4
1.2
1.0
0.6
0.4
0.2
200
-14-
----- Original solution
-A- After treatment
220
Figure 2
Ultraviolet Absorbance 01' Adipic Acid
March 1971 - FR
(2~ solution in a l-mm cell)
2
-------
2.0
1.8 .
1.6
-<
..
81.4
~
o
G
~
1.2
1.0
0.2
o
March 1971 - FR
200
-15-
---- Original solution
-lr After treatment
If
220
2 0
Wavelength, nm
Figure 3
Ultraviolet Absorbance of Glycolic Acid
~.4~ solution in a l-mm cell)
-------
-16-
In the second series of tests, phthalic acid did not melt until the
temperature reached 212°C, and the vapor pressure of the phthalic acid was so
high that the condensing vapor completely blocked the exit tube and the test
was terminated after 30 minutes. Adipic acid became grayish in color after
10 hours at 162°C but its ultraviolet absorption pattern was identical with
that of pure adipic acid. No weight change was observed during the tests and
no organic compound that absorbed ultraviolet light was detected in the exit-
gas scrubbing tube.
The results of this study indicate that benzoic, phthalic, adipic,
and glycolic acid are stable compoun~s in the presence of 802 and O2 under
conditions more severe in temperature and concentration than would be encountered
in the wet-limestone scrubbing process. The principal losses of the organic
acids would be from vaporization and these will be calculated from vapor pressure
measurements in progress. However, there is a possibility of organic acid
degeneration by the accumulation of highly oxidizing ions and radicals in a
closed-loop system; no plans are made to study this possibility at the present
time. (J. D. Hatfield, Y. K. Kim)
March 1971 - FR
-------
-17-
SOLUBILITY STUDIES WITH SELECTED ORGANIC ACIDS
Measurements at 25°C
The solubility at 25°C of calcium and magnesium carbonates in
solutions of the four organic acids being considered as additives for the
wet limestone scrubbing of S02 are shown in Table I. Available acids used
for the tests were B & A reagent-grade benzoic, Fisher pure flake phthalic
anhydride, Fisher certified adipic, and Eastman technical-grade glycolic
(70% aqueous solution). The benzoic acid and phthalic anhydride were assayed
by comparing the ultraviolet absorption of their solutions with those of
Bureau of Standards benzoic acid and potassium acid phthalate. The acid
equivalent of adipic acid was found by potentiometric titration with a
standard base, and the acid content of the glycolic acid solution was
determined by oxidation with standard ceric sulfate. Weighed amounts to
give the desired concentrations were dissolved in water with the aid of
a mechanical shaker before an excess of either reagent-grade calcium carbonate
(Baker's) or basic magnesium carbonate (B & A) was added, and they were allowed
to stand at room temperature with intermittent shaking until equilibrated.
Calcium and magnesium in the liquid phases were determined by EDTA titration.
Benzoic, phthalic, and adipic acid were determined by ultraviolet spectro-
photometry after removing the interfering calcium and magnesium from the
adipate solutions with an ion-exchange column. Glycolic acid was determined
volumetrically by the method of Williard and Young [J. Am. Chem. Soc., ~,
132 (1930)J. Solutions were considered to be equilibrated when the calcium
or magnesium analysis of succeeding samples taken at least 3 days apart
agreed within the limits of the method used. The values shown are averages
of the final two determinations. The concentrations of the organic acids
found were within analytical variation of the weighed amounts added. The
mole ratios M:(-OCOH) listed in Table I were obtained by taking the average
of the organic acid determined by analysis of the saturated solution and
that added to the original solution.
The ratios of metal to -COOH groups of the organic acid, shown
in Table I, increased with decreasing amounts of the acid added. This may
be caused by the slight solubility of the metal carbonates in water. These
ratios are shown in the figure as a function of the weight percent of the
acid added to the solution. Magnesium carbonate is dissolved quite efficiently
by the organic acid solutions, and calcium carbonata also is dissolved in
ratios slightly greater than the stoichiometric value of 0.5 for most of
the solutions in this concentration range. The higher pH values of solutions
saturated with MgC03 may be caused by the basicity of the reagent carbonate
used as well as the greater solubility of the magnesium compound.
Work done under OAP-TVA Contract No. TV-34425A
~il~n-rn
-------
Acid
Benzoic
Phthalic
Glycolic
Adipic
April 1971 - FR
-18-
Ti\BLF; I
Solubil1~f CRlciulD ~tndMlJ.gneBium Ca.rbonl.1tee
in Solutions of Organic Acids ~t 25°C
S~td. CaC03 Bolna. Satd. MgC03 801ns.
J.lolad ty X 103 Mole Molarity X 103 Mole
Wt. ." Organic ra.tio Organic ratio
of acid J2.!!.. CEl Held Ca:(-COOH) E!L ~ acid Mg: (-COOH)
0.05 8.0 2.42 3.95 0.60 9.4 5.88 4.05 1.45
0.1 7.9 4.42 8.00 0.55 9.1 8.51 8.20 1.04
0.2 8.2 8.58 16.4 0.52 8.9 13.7 16.0 0.85
0.05 8.1 3.19 2.88 0.51~ 9.4 6.53 2.88 1.11
0.1 8.0 6.08 5.85 0.51 9.1 10.2 5.90 0.86
0.2 7.9 11.9 11.6 0.50 8.8 17.1 11.8 0.72
0.05 8.1 3.23 5.75 0.53 9.3 7.12 6.60 1.08
0.1 8.0 6.64 12.9 0.51 9.2 11.3 13.7 0.84
0.2 7.8 13.8 26.9 0.52 9.0 18.8 28.2 0.69
0.05 8.0 ,.69 3.13 0.52 9.' 7.17 3.13 1.00
0.1 7.9 7.15 7.00 0.52 9.1 11. 5 7.00 0.8,
0.2 7.8 14.1 16.0 0.48 9.0 19.3 14.0 0.70
TABLE II
--
Solubility of Calcium and Magnesium Salts
of Organic Acids at 25°C
Salt
Saturated solution
Molar! ty X 103
Ca or Mg Organic acid
£!!...
7.4
6.4
7.1
7.6
9.2
196.
15.5
158.
171.
700.
99.8
16.1
80.4
173.
356.
Calcium benzoate
Calcium phthalate
Calcium glycolate
Calcium adipate
Magnesium benzoate
-------
Ai
~O.55
~
~
k
1
....01:
~ ..;'
W)
....
't1
~
W)
u
~
0.4
o.
April 1971 - FR
-19-
o Benzoic
[] Phthalic
6 Glycolic
. Adipic
~
\
\
\
o ,
,
1,~ 1
" []
" 1
'4
D
o
o.
1.6
[ 1.4
~
k 1.2
't1
~
M
o
ro
ro
:a
I 1.
ro
~
:i
o.
o.
.2
Weight ~ acid
~
\
\
\
\
D \
/1
0'
" \
\ "
~ ~
" i
''i
Dissolution ot CaCO~ and MgCO~ by
Solutions ot Organic Acids at 2SoC
-------
-20-
The saturation solubility at 25°C of the calcium salts of the
four acids and magnesium benzoate 1.s shown in Table II. The purity of all
the salts (February report) wag checked by wet analysts for calcium or
magnesium and all were practically theoretical in composition. Saturation
was obtained by placing an excess of the salt in water and allowing it to
stand at roam temperature with occasional shaking until analysis indicated
that the concentration of the liquid phases was constant.
The analytical methods were those used for the dilute solutions
in Table I even though ultraviolet spectrophotometry is not as suitable
for concentrated solutions because of dilution errors. This could have
contributed to the lack of stoichiometry in the mole ratios of metal to
acid for the saturated solutions of Table II.
Similar studies at 50°C are in progress. A computer program is
being written to calculate the species concentrations of the solutions and
to evaluate from the data the association of the metals with the organic
acids in solution. (J. D. Hatfield, R. C. Mullins)
April 1971 - FR
-------
-21-
Constants Derived From Solubility Measurements
The data reported in April on the solubility of CaC03 in solutions
of the four organic acids being studied (benzoic, glycolic, phthalic, and
adipic) were analyzed by a computer program to evaluate the concentration
of species in the solutions and to determine the stability constant of
the complex, CaA (A = organic anion)
CaA +t Ca + A, Kcpx = acaaA/acaA
The activity coefficients and equilibrium constants for other equilibria
used by Radian Corporation (PB193029) were used in this program.
(1)
The densities of the solutions were measured and the original
molar concentrations were converted to molalities for the computations.
In addition, the pH of each solution was redetermined more accurately to
obtain smaller variance of the stability constant, Kcpx' The ionization
constants of the organic acids with the exception of benzoic were taken
from Special Publication No. 17 of The Chemical Society (1964). The
ionization constant of benzoic acid was taken from Smo1yakov and Primanchuk,
[Russ. J. Phys. Chem. ~, No.3, 331-3 (1966)]; their data agree well with
those of other investigators and cover the temperature range 25° to 90°C.
The thermodynamic ionization constants (infinite dilution) are
summarized in the tabulation.
Dissociation Constants at 25°C
Acid K1 log K1 K:;::> log K:;::>
Benzoic 6.24.10-5 -4.205
Glycolic 1.48.10-4 -3.831
Phthalic 7.25.10-4 -3.14 4.0.10-6 -5.40
Adipic 3.8.10-5 -4.42 3.9.10-6 -5.41
June 1971 - FR
OAP-TVA Contract No. TV-34425A
-------
-22-
The reported data for the ionization constants of benzoic and
glycolic acids with temperature were put in the form used by Radian Corporation.
log K = -A/T -B log T - CT + D
(2)
The coefficients were determined by least squares and their significance was
determined by statistical tests. The coefficients are shown in the tabula-
tion.
Acid
A
B
C
D
Benzoic
Glycolic
804.7
1327.1
o
o
0.0090476
0.0144969
1.192
4.942
The ionization constants of dibasic phthalic and adipic acids at
temperatures other than 25°C have not been reported.
The calculations were made for the system CaO-C02-HnA-H20 where
A is the acid anion, n = 1 for benzoic and glycolic acid, and n = 2 for
phthalic and adipic acids. The solutions were saturated with CaCOs, and
the total concentrations of calcium and acid were determined, the acid
both by analysis and by weighing the amount of acid added to the total
solution. The pH of each solution ~s determined and equations corresponding
to the following equilibria were solved.
H20 ~ H + OH
[H~ i! HA + A]
HA4!H+A
CaOH 4! Ca + OH
CaHCOs +! Ca + HCOs
CaCOs ~ Ca + COs
H2COS it H + HCOs
RCOs ~ H + COs
CaA 4! Ca + A
ECA = Ca + CaOH + CaHCOs + caCos + CaA
EA = [H~] + HA + A + CaA
EC02 = CaHCOs + CaCOs + H2COS + HCOs + COs
EZi mi = 0
Activity coefficients, Vi, which relate the activity, ai' and molality,
mi, of each aqueous species, ~ere calculated from the expression
log Vi = A*Zi2 [-I 1/2/(1 + B*~i I 1/2) + biI] + UiI
(3 )
June 1971 - FR
OAP-TVA Contract No. TV-34425A
-------
-23-
where A* and B* are the temperature-dependent Debye-Hilckel terms, a.nd Zi,
~i' bi' and Ui are the valence, ionic size, and two experimentally determined
parameters for each i species, respectively. The parameters for the organic
species, for which there were no available values, were taken such that
equation 3 reduced to the Davies equation. These parameters were ~ = 3.0,
b = 0.3, u = 0.0 for all charged species, and u = 0.076 for uncharged species.
The ionic strength, I, was calculated from I = ~Zi2mi' and the activity of
the water was calculated from the Gibbs-Duhem relation that has been adapted
by Radian Corporation for computer evaluation with the above activity co-
efficient form (Technical Note 200-004-02, 27 May 1970).
Since the solutions were saturated with CaCOs and the pH was
measured, a direct measurement was obtained of the activities of the aqueous
species CaCOs and H+; this leaves 12 unknowns and 12 equations for the
monobasic acids (13 for the dibasic acids), one of which is the complexity
constant, Kcpx, of equation 1. Table I summarizes the values of solubility,
density, and pH of the solutions and the calculated values of the logarithm
of the complexity constant, Kcpx' Table II is an example of the computer
printout for the solution containing 0.2i benzoic acid when the acid was
determined by ultraviolet spectrophotometry.
The data in Table I include values for the organic acid content
of the solutions determined by analysis, as described in April, and determined
from the weighed amounts added to the solutions. The value of Kcpx was
calculated for each of these acid concentrations as well as for the average
of the two. For benzoic acid, the equations were solvable for all solutions,
but for the other acids no values of Kcpx were obtained for many solutions
because some of the species concentrations were calculated to be negative.
This indicates the need for more accurate analyses to describe the systems
adequately and to evaluate the complexity constants from solubility data.
,For example, a li error in the analysis of banzoic acid caused a change in
the value of Kcpx of 32i, 47i, and 8Li when the acid concentrations were
approximately 0.05, 0.1 and 0.2i, respectively. Similarly, an error of
0.03 pH unit caused an error of about 30% in the value of Kcpx'
In Table II the symbol M stands for metal, here calcium, TC02
is the total carbonate present in solution in moles per kg H20, PC02 is
the partial pressure of CO2 in atmospheres over the solution, EN is the
electroneutrality balance from solving the equations (the sum of positive
charges over negative charges), and I is the ionic strength.
June 1971 - FR
OAP-TVA Contract No. TV-34425A
-------
TABLE I
Comp~exity Cons'tants at 25°C in Solutions Saturated With Calcium
Carbonate Co:ataiD1ng Organic Acids
Wt ., i 103 molal1 ty
Density, Acid log Kcpx when acid is
Acid ot acid ...E!L fl./ml. Ca Ana.l. Added analyzed added averaged
Benzoic 0.05 8.04 0.9913 2.428 3.963 4.096 -2.27 -1.85 -2.~0
o.~ 7.95 0.9915 4.437 8.030 8.197 -2.3~ -1.96 -2.16
0.2 7.81 0.9984 8.615 16.47 ~6.41 -2.13 -2.22 -2.J.8
Glycolic 0.05 8.05 0.9978 3.239 5. 766 6.578 -2.14 ns ns I
!'\)
0.1 7.98 0.9991 6.655 12.93 13.16 ns DB ns .j::""
I
0.2 7.82 1. 000 ~3.84 26.98 26~35 -2.71 DB -2.53
Phthalic 0.05 8.14 0.9979 3.199 2.888 3.0il -2.82 DB -2.14
0.1 8.00 0.998~ 6.100 5.869 6.025 -2.26 I1S DB
0.2 7.87 0.9990 U.94 u.64 12.063 -2.95 ns ns
Adipic 0.05 8.02 0.9987 3 .698 3.138 3.424 ns -~.59 DB
0.1 7.92 0.9990 7.167 7.017 6.849 ns -2.80 -2.00
0.2 7.81 0.9988 14.16 16.07 13.71 ns -3.44 us
ns No solution obtained because some ot the species conce:atration were calculated to be
negati ve.
June 1971 - FR
OAP-TVA Contract No. TV-34425A
-------
-25-
TABLE Il
Distribution of Species and Other Properties of Saturated Solutions
of CaC03 Containing O.~ Benzoic Acid at 25.0
SPEOIES
H10
H
04
HA
A
"'A
H
MOI-I
HHC03
HC03
~2C03
HC03
C03
TCOP:
I(CP)(:
t14
H
,",OH
M~C03
t1C03
ACTIVITY
0.1549E-07
O.653JE-06
(I.2653E-05
o .1070E-(I1
O.3752E-02
O.2584E-02
O.3998e-07
O.2931E-04
O.7732F-05
O.2170E-04
O.6237E-OJ
O.1685E-05
O.7756e-03
O.736ee-02
METAL
49.flJ
49.90
".00
0.39
0.09
MOL All TY
0.1737E-0?
O.7444E"06
O.2645F.-05
O.12'9E-Oi
0.4275E-02
O.4299F.-02
0.45':>5E-07
0.33391:-04
0.7709E-05
0.2164F.-04
O.7097E-03
O. ;HfllE-05
PC02:
LOOKCPX:r
DISTRIRUTION
ACID
HA 0.02
A 74,02
MA 25.96
June 1971 - FR
OAP-TVA Contract No. TV-34425A
ACT. COeF"F"
0.?9964
O,A91AE+OO
0, A776E+OO
0.100JE+Ol
O.A776E+OO
O.8776E+OO
O.6012E+OO
O,8776E+OO
O.A776E+OO
O,100jE+Ol
0.100JE+Ol
O.8788E+OO
O.5964E+OO
O.6342E-O~
-2.13263
CARBONATE
MHC03 o4.:H
MC03 0.99
H2COJ 2.19
HCOJ 91,50
COJ 0,41
EN- O.132,e-09
I- O.1120ee-01
-------
-26-
Table III is a summary of the calculated distributions of calcium
and acid between ionic and complex forms in the solutions. Small amounts
of other species are present, but 99% or more of both calcium and acid
at these pH's exists as the complex, CaA, and the free ion, Ca2+ or A-
(A2- for dibasic acids). The data for benzoic acid are quite reliable
because of the high precision of the analysis and the relative constancy
of Kcpx at different concentrations of acid. The fraction of calcium or
acid in the complex form increases as the acid concentration is increased.
The data for glycolic, phthalic, and particularly adipic acids are less
reliable and only indicate a trend; little significance should be assigned
to the magnitude of the distribution.
A similar study was made for the solutions saturated with MgC03
(April report), but the solution of the equations resulted in some negative
concentrations of aqueous species. This indicates that either the analyses of
the solutions are not sufficiently accurate, the model is incorrect, or the
constants for the magnesium system equilibria are in error (as has been
suspected).
June 1971 - FR
OAP-TVA Contract No. TV-34425A
-------
-27-
TABLE III
Calculated Distributions of Calcium and Acid Between Ionic
and Complex Forms in Saturated CaC03 Solutions
Containing Organic Acids
Distribution of
Wt. ~ Calc! urn Acid
Acid of acid M MA A MA
Benzoic 0.05 75 24 86 14
0.1 61 38 79 20
0.2 50 50 74 26
Glycolic 0.05 31 68 83 17
0.1
0.2 24 76 61 39
Fhthalic 0.05 86 13 86 14
0.1 78 22 77 23
0.2 41 59 40 60
Adipic 0.05 94 5 95 5
0.1 54 46 52 48
0.2 24 76 21 79
(J. D. Hatfield, R. C. Mullins, R. L. Dunn)
June 1971 - FR
oAP-TVA Contract No. TV-34425A
-------
-28-
Solubilities of Salts of Organic Acids at 50°C: The results of
measurements of the solubilities at 50°C of the calcium and magnesium salts
of the four organic acids--benzoic, phthalic, adipic, and glycolic--are
shown in Table I, together with the solubilities at 25°C that were reported
in April. Stoichiometric mixtures of each acid with calcium or magnesium
carbonate were shaken intermittently in water at 50.0° ~ O.loC until the
compositions of the liquid phases became constant. Saturated solutions were
prepared similarly from the calcium salts of the acids also. Except for the
glycolate, the results obtained by the two methods for the calcium salts
were the same within experimental error.
Microscopic examination showed that the solid phases in the benzoate,
phthalate, or adipate mixtures prepared from the carbonate were the same as
those prepared from the reagent salts. The solid phase in the solution made
from calcium carbonate and glycolic acid was predominately 2(CH20HCO)2Ca.3H20,
whereas that in the solution prepared from anhydrous calcium glycolate was
anhydrous. The difference in composition of the glycolate solutions therefore
represents the difference in solubility of the trihydrate and the anhydrous salt.
The pH of the solutions prepared from the carbonates and the acids, however,
were consistently lower than those prepared from the salts, presumably because
they were saturated with CO2,
The solubility of the
glycolic--increased with rising
phthalic and adipic--decreased.
is noteworthy.
salts of the monobasic acids--benzoic and
temperature and those of the dibasic acids--
The relative insolubility of calcium phthalate
As shown in Table II, the solubilities at 50°C of calcium and
magnesium carbonates in O.l~ solutions of the organic acids are the same
within experimental error as those at 25°C (April report). The concentration
of acid is the limiting factor, and temperature, up to 50°C, has no effect
on the solubility. The pH's of the solutions at 50°C were significantly
lower than those at 25°C; the reason for this effect is not apparent.
September 1971 - FR
-------
-29-
TABLE I
Solubility ot Calcium and Magnesium Salts
ot Organic Acids
Satd. soln. at 50.C Satd. soln. at 2'.ca
Conen.. .It. )( 1()2 Conen.. M. )( lOR
Ca or Acid Ca or Acid
Acid R!L Mg anion !L Mg anion
Calc1UID salts
Benzoic
Acid + CaCOe ,.8 1,.6 21.4
Reagent salt 6.8 1,.6 26.8 1.4 10.0 19.6
A1thaUc
Acid + CaCOe 1.1 1.21 1.1' -
Reagent ual t 1.4 1.22 1.18 6.4 1.61 1."
Adipic
Acid + CaCG" 6.2 11.0 11.' -
Reagent .al t 1.6 10.1 10.9 7.6 17.' 17.1
01yco11c
Acid + CaCOs ,., 21." "2.8 . -
Reagent salt 6.7 11.1 ,..., 7.1 8..0 1,.8
ItI,gneetwa salts
Benzoic ,.2 89.4 179. 9.2 ".6 70.0
A1thalic 6.6 119. 18,. .
Adipic ~.9 1"2. 1..8.
Olyco1ic .9 71.' 146.
. Data reported in April 1971.
September 1971 - Fa
-------
-30-
TABLE II
SolUbility of Calcium and MagneBitUII Carbonates
in 0.1~ Solutions of Organic Acids
Satd. soln. at 50.C 8atd. soln. at 25.CR
Mole Mole
Conen. J H, X 103 ratio Concn. J .H, X 103 ratio
Ca or Acid metal: Ca or Acid m~ta.l:
Acid P!L Mg anion anion R!L Mg anion anion
Calcium salts
Benzolc 7.2 4.44 8.08 0.55 7.9 4.42 8.00 0.55
Fbthal1c 6.8 6.19 5.94 1.04 8.0 6.08 5.85 1.02
Adipic 6.8 7.80 7.19 1.~ 7.9 7.1' 7.00 1.02
Glycolic 6.7 8.71 16.5 0.53 8.0 6.64 12.9 0.51
tBgnesiwn salts
Benzoic 8.; 8.05 8.31 0.97 9.1 8.51 8.20 1.04
Fbthal1c 8.4 10.7 6.09 1.75 9.1 10.2 5.90 1.72
Adipic 8.3 11.9 6.95 1.71 9.1 11.5 7.00 1.66
Glycolic 8.2 12.2 15.1 0.81 9.2 11.3 13.7 0.8~
a
Data reported in April 1971.
September 1971 - FR
-------
-31-
The calcium and magnesium contents of the 50°C solutions were
determined by EDTA titration, and the organic acid contents were determined
by potentiometric titration of eluates from passage of the solutions through
a cation resin column. The eluates were heated to boiling to remove C02 and
cooled before the titrations.
Attempts to calculate stability constants for assumed aqueous
complexes from the solubility data at 50°C were unsuccessful for the glycolates
and phthalates because the calculations gave negative concentrations of some
species. For the benzoate and adipate solutions containing calcium carbonate
the data are consistent with the following equilibria at zero ionic strength.
CaBz+ ~ Ca++ + Bz-
CaAd ~ Ca++ + Ad--
Kcpx = 0.1
Kcpx = 0.003
Bunting and Thong [Can.J. Chern. 48, 1654 (1970)] reported a value of 0.63
for the ionization constant of C8Bz+ at 30° and an ionic strength, I, of
0.4. From the Davies expressions for activity coefficients, the thermo-
dynamic ionization constant (I = 0) of CaBz+ at 30°C is calculated from
Bunting's data to be 0.16 as compared with our measured value of 0.1 at
50°C. No comparable data were found for the adipate.
September 1971 - FR
-------
-32-
COMPLEXES IN SOLUTION
This report describes studies of the chemical composition of
solutions containing calcium and benzoic acid. The literature contains no
info~mation on the formation of complexes of calcium benzoate, such as
CaBz or CaBz~ as aqueous species; this information is needed to help
define the chemistry of reactions occurring in wet-limestone scrubbing in
the presence of organic additives. The method of continuous variation, .
Job's method [Ann. Chim. [lO]~, 113 (l928)],was used to treat data obtained
with the calcium ion selective electrode (Beckman) and the Cary 17 ultra-
violet spectrometer.
Calcium Ion Electrode Studies: Calcium perchlorate tetrahydrate,
Ca(C104)204H20, was prepared from reagent grade CaCOs and 7010 HC104. A
slight excess of freshly calcined CaCOs was added to the acid, the mixture
was filtered, and the filtrate was concentrated by evaporation. The
crystals that separated on cooling were filtered off and dried in a desiccator.
A stock solution of the crystals was analyzed for calcium, and portions of
the solution were diluted for electrode calibration and testing.
Sodium benzoate (U. S. P.) was purified by recrystallization. The
crystals, dried at 105°C, analyzed l5.9l~ Na (theory 15.96) and were used
for preparing a stock solution for dilution.
.----- -~. -
The calcium ion electrode was calibrated immediately prior to
making the measurements for complexation. The calibration from 10-1 to
10-5 molar calcium is shown in Figure 1 in terms of both concentration
and activity of the calcium ion. The activity coefficients, YCa2+, were
calculated from the extended Debye-Hucke1 equation used by Radian Corporation
[Final report for NAPCA Contract No. CPA-22-69-138, Vol. 1, June 9, 1970]
by the expression
log YCa2+ = 4A [_11/2/{1 + 4.5 B 11/2) + 0.1 I]
(1)
December 1970 - FR
-------
-33-
where A and B are the theoretical Debye-Huckel constants (0.5116 and
0.3292, respectively, at 25°C), and I is the ionic strength (I = 1/2 t Zr Ci'
with Zi the valence and Ci the concentration of the ith ion). The calibra-
tion measurements that spanned the ra.nge to be used in the complexation
measurements, 10-3 to 10-4 N, were run in duplicate. The least-squares
equation between lO-~ and 10-4 M,
log ACa = -3.0636 + 0.03328E
(2)
corresponds to a slope of 30.0 mv/decade (theoretical 29.6) and represents
the activity of the calcium ion with a standard deviation of 29'10. This
variability agrees with the duplicate measurements and is perhaps due to
the slow response time of the electrode, and the influence of the immediate
past history of the electrode (memory effect). The procedure in determining
the "equilibrium" emf of the cell
calcium electrode/solution containing Ca2+/standard calamel electrode
was to record the readings at l-minute intervals until two successive
readings were identical to tenths of a millivolt; this sometimes required
more than 10 minutes and a slow drift may have continued much longer--
depending on the solution in which the electrode had been immersed in the
previous measurement.
A series of samples was prepared by mixing 10-3M Ca(C104)2 with
10-3M CeHsCOONa in the volume ratios indicated in Table I: The volume of
each sample was 100 ml, and the concentrations of calcium and benzoate
varied in regular order as shown in columns 3 and 4 of Table I. Concentra-
tions between 10-3 and 10-4M were chosen to minimize the effect of ionic
strength on the activity coefficients and of Na+ on the electrode. The
millivolt readings in column 5 of Table I were converted to activities of
the calcium ion by use of equation 2, and the results are given in column 6;
the pH values of the mixtures are shown in column 7.
December 1970 - FR
-------
..
GJ
£+30
1)
GJ
M
GJ
M
~+15
~
u
~
1 0
of.>
fI)
.
fI)
>
rz1 -15
-34;.
+60
o - Concentration
D - Activity
-30
-45
10-1
10-2
10-3
Concentration of Ca2+, ~
Figure 1
10-4
10-5
Standardization of Calcium Ion Selective Electrode
[Calcium supplied by standard Ca(C104)2 solutions]
December 1970 - FR
-------
-35-
TABLE I
Study ot Calcium Benzoate Complex at 2'.C
~ Total Total &at, Ca electr~e Ca activity,
1 )( 101 1)( 10"3 calcium, benzoate, vs. standard M )( 10'
~ CeH!lCo6Na If)( lot B)( 10" electrode I mv J trCIII eQ.. 2 t ..R!..
o 100 0 10.0 6.'0
10 90 1.0 9.0 -25.6 1.2 6.'2
20 80 2.0 8.0 -20.' 1.8 6.29
'0 70 '.0 1.0 -17.1 2.' 6.'7
"0 60 4.0 6.0 -15., 2.6 6."
,0 ,0 ,.0 '.0 -12.1 ,.4 6.29
60 40 6.0 4.0 -10.6 ,.8 6.,..
70 '0 7.0 ,.0 -7.0 '.1 6.26
80 20 8.0 2.0 -5.2 ,.8 6.'9
90 10 9.0 1.0 -4.2 6., 6.29
100 0 10.0 0 -2 . rI' 7.4 6.17
a Calibration reading.
December 1970 - FR
-------
-36-
Figure 2 shows the relation between the activity of the calcium
ion, Aca2+, and the total calcium in solution. Some of the variation in
the data can be attributed to the performance of the electrode, and some
of the variation may be attributed to the effect of atmospheric CO2 on the
system over which there was no control. There were no sharp breaks in the
relation to indicate a definite complex as CaBz+ or CaBz20 (Bz = benzoate
ion, CeHsCOO-). While the measured activity of calcium ion was only 70 to
80~ of that expected if completely ionized (dotted line, Figure 2), the
difference is more than likely due to complexes such as CaOIf'" and CaCOso
that are known to exist. The variability of pH would indicate differences
in the amount of CO2 absorbed by the solutions and would preclude the use
of these measurements in calculating the strength of any possible calcium
benzoate complex.
Ultraviolet Absorption Studies: The characteristic absorption
peaks for solutions of benzoic acid or sodium benzoate are located at
193 ~ and 227 J,Lm, and the absorbances at these wave lengths obey Beer's
law (July report). Freshly prepared solutions with the same compositions
as those used for the calcium electrode measurements were used to study
the effect on the absorbances in a l-mm celL The results are shown in
Figure 3. There is a linear increase in the absorbances with increase in
total benzoate, and there are no breaks in the absorbanct:!s. There were
also no new absor'J:>ance bands as would be expected if a new species were
formed by adding calcium. It is possible that a new calcium benzoate
species forms that absorbs at the same wave lengths as does benzoic acid,
but this is very unlikely.
The results of the ultraviolet absorption and the calcium ion
electrode both indicate that no complex between calcium ion and benzoate
ion that can be measured by these methods is formed in solution. If such
a complex is formed it is very weak, and calcium benzoate solutions can be
considered completely ionized--other than the effects of hydrolysis of the
ions and reactions with other components of the solution.
December 1970 - FR
-------
b 6
....
)(
it.,
c?
4
-37-
10
8
~
2
1
Mole ratio Bz:Ca
o
o
10
2
8
4 6
6 4
Total concentration, M x 104
Figure 2
Activity of Calcium in Calcium
PercPlorate-Sodium Benzoate Solutions
December 1970 - FR
8
2
D
10 Ca2+
o Bz
-------
-38-
1.0
3.0
2.5
0.8 2.0
! ~
~
~ r-I
.to) ~
as
II) 0.6 1.5cv
t) tJ
~ ~
'E 'E
o 0
fIJ fIJ
~ ~
0.4 1.0
2
8
4 6
6 4
Total concentration, ~ X 104
0.5
0.2
o
o
10
o
8
2
Ca2+
Bz
Figure 3
Ultraviolet Absorbance of Calcium
Perclorate-Sodium Benzoate Solutions
December 1970 ~ FR
-------
-39-
Spectroscopic Study of Calcium Benzoate Complexes: Information
on the formation of complexes is needed in understanding the behavior of
organic additives in the wet-limestone scrubbing of S02' Benzoic acid
solutions absorb at 275 and 192 nm in the ultraviolet region (July 1970
report); the peaks are sharp and Beer's law is obeyed in the concentration
range 10-3 to 10-4 molar- A calcium benzoate complex formed in aqueous solu-
tion probably would have different absorption characteristics, and infrared
spectroscopy should be useful in identifying this complex. In exploratory
tests (December 1970 report) no complex was detected, but it was found that
the pH of the benzoate solutions affects the absorbance. In further studies
the pH was adjusted to either 5 or 8 by addition of tetramethylammonium
hydroxide to solutions of calcium perchlorate and benzoic acid before measuring
the absorbance. The results are given in Table IV. Neither perchlorate nor
tetramethylammonium ions absorb appreciably under the experimental conditions
and do not form complexes. Tests at pH 5 and 8 showed that the absorbance is
proportional to the benzoic acid concentration when the calcium concentration
is constant and that the absorbance is constant when the calcium concentra-
tion is varied. These results indicate that any aqueous calcium benzoate
complex formed is so weak that it does not affect the absorbance of the
normal benzoate species in solution. This is in agreement with results of
previous studies with the ion-selective electrode (December 1970 report) that
indicated that calcium benzoate solutions could be considered almost completely
ionized. The ionization constant calculated from the solubility data at 50°C
(Kcpx = 0.1) indicates that less than 1% of the benzoate in solution is com-
plexed as CaBz+ when the calcium ion concentration is in the range shown in
Table IV. This complex is too weak to affect ultraviolet absorbance measure-
ments, and the ion-selective electrode does not detect such small changes in
concentration.
September 1971 - FR
-------
-40-
TABLE IV
Ultraviolet Absorbance of Calcium Benzoate
Solution at Roam ~mperature (250 -27°C l
Absorbance at
indicated
Composition, M )( 10", of solution wavelength, DID
Ca( C10... 1.& CeH!5COOH If ( CHs ):;Q! 225 ~
Measurements at pH 5.0
2 1 0.96 0.074 0.40
2 1.9 0.159 0.85
3 2.7 0.243 1.30
4 3.6 0.325 1. 72
5 4.5 0.405 2.16
1.6 4 3.6 0.324 1.71
1.2 4 3.6 0.323 1.70
0.8 4 3.6 0.324 1.71
0.4 4 3.6 o. 324 1.71
0.0 4 3.6 0.323 1.70
Measurements at pH 8.0
2 1 1.7 0.086 0.46
2 2.8 0.173 0.92
3 3.2 0.255 1.37
4 4.2 0.342 1.84
5 5.2 0.4~2 2.30
1.6 4 4.3 0.431 1.83
1.2 4 4.3 0.431 1.82
0.8 4 4.3 0.432 1.82
0.4 4 4.3 0.431 1.83
0.0 4 4., 0.431 1.82
Septemuer 1971 - FR
-------
-41-
VOLATILITY OF ORGANIC ACIDS UNDER SCRUBBING CONDITIONS
Vapor Pressure of Benzoic Acid in Water: The partial pressure
of benzoic acid over its aqueous solution was measured by a dynamic method
similar to that described by Johnstone [Ind. Eng. Chem. g]} 587 (1935)].
Nitrogen gas was saturated by passage through a solution of 1.6 grams of
benzoic acid per liter, and the exit gas was scrubbed with a Vanier-type
bulb on the exit tube. The benzoic acid in the scrubbing bulb was
determined by diluting the sample with deionized water and measuring the
ultraviolet absorbance of the solution.
It was found that the vapors of benzoic acid were absorbed by
both the neoprene stopper and the stopcock grease in the joints of the
apparatus. The stopcock grease was eliminated from the apparatus, and
after many determinations the neoprene stopper in the top of the satura-
tion vessel apparently became equilibrated or perhaps saturated with the
benzoic acid. Reproducible results then were obtained in duplicate runs.
The partial pressures of benzoic acid over a solution containing
1.6 grams/liter were measured at three temperatures with the following
results:
Temp.,
°c
Partial pressure
of benzoic acid
atm, X 105
40
55
65
1.336
2.07
2.79
From the temperature dependence of the ionization constant, K1, of benzoic
acid [Russian J. of Phys. Chem. ~, 331 (1966)]
log Kl = -804.7/T - 0.OO90476T + 1.192
(1)
and the activity coefficients used by Radian Corporation [Final report,
NAPCA Contract No. CPA-22-69-l38, Vol. 1, June 9, 1970], the degree of
dissociation at each temperature was calculated, and from this the
activity of undissociated benzoic acid, ~z' and the Henry's law
constant, h, were calculated.
h = PHBz/AHBz
(2)
January 1971 - FR
-------
-42-
The results are shown in the tabulation.
Tenw . ,
°c
40
55
65
h, atm/molality
1.088 X 10-3
1.684 X 10-3
2.266 X 10-3
Both the partial pressure of benzoic acid, PHBz, and the
Henry's law constant, h, were related to the absolute temperature, T,
by equations determined by the method of least squares.
log PHBz. -0.5662 - l349.4/T
(3 )
and
log h = 1.3285 - l344.6/T
(4)
Equation 3 applies only to the concentration of benzoic acid used (1.6
grams/liter) and implies a heat of vaporization of 6,200 cal/mole from
solution as compared with a sublimation energy of 21,900 cal/mole for the
pure acid. Equation. 4 and equation 2 may be used to calculate the vapor
pressure at any temperature and concentration at which AHBz can be
determined or calculated.
The vapor pressure of benzoic acid over scrubbing solutions
in the pH range 4 to 1 were estimated from the relation
log(~z/~Z) = pK1 - pH
( 5)
where AB is the activity of the benzoate ion and the symbol p indicates
the negative logarithm. If it is assumed that the activity coefficients for
the benzoate species are unity, the following vapor pressures at 55°C are
calculated for a tot~ benzoate concentration of 0..01 m (about 1.2 grams/
li ter ) . -
E!!
4
5
6
7
PHBzJ ppm (volume basis)
11.0
2.4
0.28
0.028
January 1971 - FR
-------
-43-
The vapor pressures of benzoic acid solutions in the pH range
4 to 1 is too low to measure in our laboratory equipn1ent, but values
estimated from the consistent data obtained thus far probably are correct
within an order of magnitude. This range of vapor pressures is a serious
consideration in scrubbing operations, since a vapor pressure of 1 ppm
over solutions used to scrub the gas from a 500-megawatt boiler would
produce more than 40 liters per minute (>200 grams) of benzoic acid vapor
in the exit flue gas.
The above results will be checked with an apparatus fabricated
with Teflon seals. The vapor pressm.'es of solutions of phthalic, glycolic,
and adipic acids will be determined also. (J. D. Hatfield, Y. K. Kim,
R. C. MUllins, M. E. Deming)
January 1971 - FR
Vapor Pressure of Benzoic Acid Solutions: The partial pressure of
benzoic acid at 3 temperatures was reported in January for a concentration
of 1.6 grams benzoic acid per liter. The data reported were consistent with
the integrated Clausius-Clapeyron equation, but they implied a heat of
vaporization of 6.2 kcal/mole as compared with a value of 21.9 kcal/mole
for sublimation. Experimental difficulties of absorption of benzoic acid
vapors on the apparatus and on the stopcock grease used at the joints cast
doubt on the validity of the results.
September 1971 - FR
-------
-44-
The study was repeated in a 4-stage saturator with the sidearm'
heated about 10°C above that of the run temperature. The grease-free joints
were flamed briefly at the end of a run to remove absorbed vapors. After
conditioning the apparatus by many runs to a quasi-steady state, consistent
results were obtained, as shown in the fi~re. The data are expressed by
the relation
log Patm = 9.20 - 4,787/T
which indicates an energy of vaporization of 22 kcal/mole. In January the
Henry's law constant for evaporation was used, and its reciprocal is used
here to conform to the convention used by the Radian Corporation in their
computer program described in PB193029. The Henry's law constant, h, for
absorption then becomes
log h = -10.27 + 4500/T
and the vapor pressures of 0.01 molar benzoate solution at different
temperatures and pH are shown in the tabulation.
E!!
2
3
4
5
6
7
Vapor pressure of b~nzoic acid, ppb,
at indicated temperatures
4Soc (113°F) SO°C (122°F) 5SoC (131°F)
136 217 355
129 206 337
85 137 225
19 31 52
2 4 6
0.2 0.4 0.6
These data indicate that the losses of benzoic acid under scrubbing
conditions will be relatively low. A vapor pressure of 100 ppb would result
in the loss of about 2.5 pounds of benzoic acid per hour if the acid were
used in the limestone slurry to scrub the gas from a 500-megawatt boiler. It
is probable, however, that greater losses would result from absorption on
scrubber solids, on surfaces of the equipment, and in the mist from the mist
eliminator at the top of the scrubber.
September 1971 - FR
-------
20
ClO
,...
...
~
...
GJ
~ 4
CIJ
~
,..
~
r> 2
1
2.9
September 1971 - FR
-45-
65
Temperature, 00
55
,.0
,.1
1000 IT
VapOr Pressure of Benzoic Acid
Over Its 0.013 U Aqueous Solution
(1.6 grams OeHsCOOH per liter)
,.2
(J. D. Hatfield, Y. K. Kim,
R. C. Mullins, M. E. Deming,
R. L. Dunn)
-------
-46-
EFFECT OF ORGANIC ACIDS ON OXIDATION OF SULFITE
Oxidation of Sodium Sulfite
The addition of an organic acid to limestone slurries to promote
S02 absorption may also affect the rate of oxidation of sulfite to sulfate
in the scrubber. Initial results in the Applied Research pilot unit indicated
a considerable increase in the oxidation to sulfate (May 1970). Since the
oxidation rate is probably dependent on pH, and the pH changes with the
degree of oxidation, a study was made of the effects of the four organic
acids (benzoic, phthalic, glycolic, and adipic) in the sodium system in which
considerable sulfite can be held in solution to offset rapid pH changes and
lengthen the time before oxidation is complete. Similar studies will be
made in the calcium system.
The reaction solution was prepared by dissolving reagent-grade
Na2S03 and Na2S20S in the mole ratio 2:3 (mole ratio S:Na-== 0.8), and the
final concentrations of the Na and S were 2.12 and 1.69 molal, respectively.
About 8 to 11% of the total sulfur was in the sulfate form before the run
was started because of the oxidation of the chemical in storage and during
the preparation of the sample. In each run, 500 ml of the solution was
placed in a l-liter gas-absorption tube equipped with a coarse fritted-
glass gas disperser, and the vessel was placed in a water bath at 55°C.
Pure oxygen was passed through the glass frit at the rate of 400 cc per minute,
and a 25-ml sample was withdrawn every hour for analysis. Immediately after
sampling, the total sulfite was determined by iodine titration and the pH
was measured; the total sulfur and sodium contents then were determined. The
losses of water and S02 by evaporation in the large flow of oxygen were
determined by the increases in sodium concentration and by the ratio of
total sulfur to sodium (assuming no losses of sodium or sulfate); these
losses were checked by a computer calculation of the vapor pressures of H20
and S02 and numerically integrating between samples. Results of a typical
run are shown in Figure 1; the S02 evaporation losses determined by chemical
analysis usually agreed fairly well with those determined by vapor pressures
but they differed most in the range in which the pH of the solution changed
most rapidly. Since the oxidation of sulfite to sulfate lowered the pH, the
vapor pressure of S02 increased. Therefore, the evaporative loss of sulfite
as S02 started after about 1 hour and gradually increased until. oxidation was
essentially complete.
Small amounts of each of the four organic acids were added to
similar solutions to study their effects on the oxidation rate; equivalent
amounts of NaOH to neutralize the acid were added so that all the starting
solutions had the same pH (5.9 % 0.05) and the same total sulfur concentration.
The results are shown in the table.
Work done under OAP-TVA Contract No. TV-34425A
May 1971 - FR
-------
-47-
100
002 volatilize,,-
Determined from
0- vapor pressure
6 - chemical analysis
f4
~60
..
,,~
. Sulf'ate
ft
..
!
I
i40
a
Sulfite and
bisulfite
2
~
20
o
o
Timt!, hours
F~ure 1
Oxidation of Sulfite to Sulfate
in the Sodium System at 55.C
Work done under OAP-TVA Contract No. TV-34425A
May 1971 - FR
-------
-48-
Effect of.Organic Acids on the Rate of Oxidation
at 550C in the Sodium Sulfite-Bisulfite System
Distribution, 'to, of S
Concn" Time, Sulfi te + S02 vola-
Acid '10 hr. ~ bisulfite Sulfate tUlzed
--
None 0 5.88 88.8 11.2 0.0
1 3.78 68.7 30.3 1.0
2 3.10 43.8 41.3 14.9
3 2.90 21.6 52.6 25.8
4 2.85 9.6 57.8 32.6
5 3.03 4.0 62.3 33.7
6 3.10 2.6 61.6 35.8
Benzoic 0.3 0 5.85 91.7 8.3 0.0
1 3.12, 59.4 33.6 7.0
2 2.85 22.3 53.0 24.7
3 2.93 7.7 61.3 31.0
4 3.32 4.5 62.5 33.0
5 3.45 2.9 63.5 33.6
6 3.48 2~3 63.5 3!L2
Phthalic 0.3 0 5.87 90.2 9.8 0.0
1 4.00 67.9 29.6 2.5
2 3.08 40.1 45.1 14.8
3 2.77 13.3 58.9 27.8
4 3.00 5.9 63.1 31.0
5 3.15 3.6 64.4 32.0
6 3.35 2.5 65.5 32.0
Glycolic 1.0 0 5.92 91.2 8.8 0.0
1 4.28 69.3 29.5 1.2
2 3.45 51.0 40.3 8.7
3 3.23 30.8 51. 7 17.5
4 3.10 14.5 59.8 25.7
5 3.05 ' 6.5 64..5 29.0
6 3.28 3.7 66.1 30.2
7 3.32 2.5 66.5 31.0
Adipic 1.0 0 5.90 90.2 9.8 0.0
1 4.50 64.5 32.5 3.0
2 2.88 31.0 52.0 17.0
3 2.75 9.0 62.8 28.2
4 3.10 4.5 66.1 29.4
5 3.35 3.0 66.4 30.6
6 3.45 2.1 67.3 30.6
Work done under OAP-TVA Contract No. TV-34425A
May 1971 - FR
-------
-49-
With each organic acid the oxidation was almost complete in 6 hours,
and the pH change and S02 evaporative 1088 pattern were similar for all runs.
The effects of the organic acids on the rate of oxidation of sulfite are
shown in Figure 2; the data indicate that all the organic acids promoted
oxidation in the sodium sulfite-bisulfite system in the order benzoic>
adipic> phthalic> glycolic> no acid.
Similar measurements are being made of the rate of oxidation of
sulfite in the calcium sulfite-bisulfite system.
Work done under OAP-TVA Contract No. TV-34425A
May 1971 - FR
-------
-50-
60
o
~
N
oM
'CI40
'R
o
G)
.p
~
i
\.4
o
..
'#.
Acid
o - None
o - Benzoic
V - Adipic
IJ - Phthalic
6 - Glycolic
..
8
oM
~20
.t:
2
6
Time, hours
Figure 2
Effect of Organic Acid on the OXidation of
Sulfite to SUlfate in the Sodium System at 55°C
(J. D. Hatfield, Y. K. Kim,
M. E. Deming, R. L. ~~nn)
Work done under OAP-TVA Contract No. TV-34425A
May 1971 - FR
-------
-51-
Oxidation of Calcium Sulfite
The addition of organic acid to sodium sulfate solutions resulted
in an increase in the rate of oxidation of sulfite to sulfate by oxygen
gas at 55°C (May report), and the following order of increase was established:
benzoic> adipic> phthalic> glycolic> no acid. In these tests, however,
the concentrations of Na and S were 2.12 and 1.69 molal, respectively, and
the presence of organic acid gave a bUffering effect to the solution as
the oxidation proceeded.
This report concerns oxidation studies in the calcium system at
50°C in which the concentration of S is much smaller than in the sodium
system, because of solubility, and in which the buffering effect of the
organic additive would be expected to be more pronounced. The effects of
changing pH, constant pH, addition of organic additives, and the organic
acid con~entration (benzoic only) were studied.
The first tests were made under conditions similar to those used
in the sodium system. The .solutions were prepared by mixing 0.3 gram of
powdered reagent CaC03 with 500 ml of water and adjusting the pH of the
solution to about 4.0 by slowly absorbing gaseous S02 into the solution;
the CaC03 dissolved completely. After purging the solution at 50°C with
nitrogen gas at the rate of 20 ml/min) oxygen was added at a rate of 21
ml/min through a coarse fritted glass to a gas absorber that contained
400 ml of the solution. Samples (10 ml) were removed at intervals, and
the total sulfite and pH were determined as quickly as possible to prevent
further oxidation. The values for pH, total sulfite, and total calcium
were used in a computer program to calculate the species concentrations
and activities, using the constants and activity coefficients suggested
by the Radian Corporation (PB 193029). The partial pressure of CO2 was
varied in the calculations from 0.0 to 0.0003 atmosphere, but this did
not affect the distribution of calcium and sulfur species because most
of the CO2 is present in these solutions as dissolved CO2 (or H2C03).
An example of the computer printout is given in Table I for the sample
taken at 1 hour in run I. The water lost due to the flowing oxygen was
calculated from its vapor pressure and the rate and time of gas flow;
this was used to correct the calcium and total sulfur concentrations for the
various times. The sulfur volatilized was obtained by a mass balance be-
tween the initial total sulfur content and the sum of the sulfite and
sulfate; this was superior to integrating the partial pressure of 802 curve
with time because of the rapid rise in 802 pressure as the pH decreased.
July 1971 - FR
-------
-52-
TABLE ,I
Distribution of Aqueous Species and the Properties
of the Solution During the Oxidation of 8ulfite
to 8ulfate With Oxygen at 50DC
.-..-....---
.... . --""< "".'''''''' ------_. -"-'"
.._-~...__. .---...
----- _.. .. - -.
SPECIES
ACTIVITY
MOLALITY
ACT.COeF"r
1 H20 0.99970
~ . 'W--O.8511E-03-0;9'616E-1J3-0';'S851E+OO
3 OH D.6438E-10 O.1405E-10 O.8695E+OO
~"-"---4---H2S03 0',8 065E- 03 '--0 ~ 8059Eoio OJ --0.10 03E+01
. , ~S03 ~OJ 1436&-02 ._O.S540E:02 O.8707E+OO
& SOJ O.372.2E-06 'J.6475E-06-0~574ae:+OO
1 HS04 O.1403E-03 O.i614E-03 O,8695E+OO
- ---.- g---'--' . 50<4 . tJ ~ 7978 E - 03 .. O. 1434 E - 0 2 -. b . 5562 E + 0 0 "
9 H2COJ O.OOOOE+OO O.OOOOE+OO O.1003E+01
---'iO"~toJ -(f~ OOOOE+OO -'0; 000 8E+OO--O", 870 7e.i:OO'-
11 C03 0.0000£+00 O.OOOOE+OO O.5748E+OO
---'-12 . ..-.- CA . "0'.30 71E- 02 ....\,. 5305S- 02 ... 0.580 iE'" 0 0-'
13 OAOH O.5522E-11 O.6351E-11 O.8695E+OO
-'--ir C4503' 0.387 oe"05 '-'-:0. 3e5SE-05 - -If.1003~+of-
. l' CAS04 0.695JE-03 O.6931E-03 0.1003(+01
----H. .'CAco3 O. OOOOE+OO --'o'.'oOOOE+OO.~"O~100JE,.;01----
17 CAHCD3 o.ooooe+oo O.ODOOE+OO O.6695~+OO
,~ . . ................ .. .. ~ ~.......-.,.. '--' .-, _..~-"'-'......._....-._.......
-----.-....
_. .. - ...
_.--
-- _.a .._-_.._- -..
....----_.. ~. ..
-.
.----"-T- ---...-- "1. o----'--PJ.fii----:--:r.01-------TCO--
T502: 0.00935 TS03a 0.00229 TC02.
pS02" --0.-150 ?'f- 02------PC0211 -.--tr.o 0 0 DE .00 ----- -- PH20.-
sv= 0.00019 HV. 0.1004
"_.'-_._'---'----':~. .--. o.2j.e"10'---"RAa '-----0-;254&75 ..----..--
.* 0.1831E"01
....~ ......-., ...-' ._.. ..-..-.., "'--- ....--------...........-----....-.---
TCA, '1'802' '1'8<>3, TC02 = total concentrations of components, molalities.
PB02' FC02' PH20 = partial pressure of component, atmospheres.
SV = Sulfur volatilized, moles/KgH20.
BV . Water volatilized, grams.
EN = Electroneutrality check on solution of equations.
RR = ~<>3/(aH)1/2.
I a IoDIc strength.
0.-00600 - ,.
o.OOOnE+OO
O..1211e...OO"
--..-.-....... .
---.-....
July 1971 - FR
-------
-53-
~ee ~uns were made under conditions as cpnstant as possible,
and the results are s~own in Figure 1 (runs I, II~ and ,III). The initial
pH of the solution ranged from 4.00 to 4.13, and the oxidation that occ\~red
duriQg preparation of the solutions ranged from 1.9 to 5.1i of the total
su).t'ur. The total calcium ~olality was 0.006 and the initial tota~ sulfur
molality varied from 0.0116 to 0.01185. The pH of the solutions decreased
rapidly at first and then decreased more slowly as the s~lfate content
increaaeQ. The i~itial oxidation rate decre~Bed with time perhaps due to
the decreasing sulfite content and to the effect of pH. The rate of
oxidation of the three solutions (I, II, and III) appeared to be different,
but the pH patterns of the solu~ions appeared to be identical.
A similar run (IV) was IIlade with a solut;ion containing 0.16%
benzoic acid. The added benzoic acid was neutralized by ad~ition of
0.0064 mole of calcium carbonate per kg of H20 to keep the pH constant,
with the sulfite concentration similar to that in runs I, II, and III.
The pH change and the rate of oxidation for run IV are shown in Figure 1
by the dashed curves. The oxidation was much slower in tb~ solution con-
taining benzoic acid, reaching abo~t 15~ of the sulfite oxidized in 5
hours as compared to about 55% oxidation in the absence of ben~oic acid.
The pH did not fall as rapidly in the presence of benzoic acid beca~se of
the lowe~ oxidation rate and the buffering effect of the additive.
A series of tests was made to determine the effect of pH on the
rate of oxidation. The starting concentrations of sulfite ranged between
0.0114 and 0.0130 molal but with different amounts of CaCOs to give different
pH values. After purging with nitrogen as before and starting the flow
of oxygen (2l cc/min) into the solution maintained at 50°C, the pH was
maintained constant throughout the run by the addition of ~ll amounts of
powdered OaCOs. The results of runs made in the absence of Qrganic additive
are shown in Table II. The pH of the last run listed ~n Table II could not
be held ~onstant because the oxidation was so rapid that excess CaCOs was
added 1 causing the pH to ~ncreaRe gradually during the run f~om 6.1 to 1.45.
The results in Table II are shown in Figure 2 as a first~order
mechantsm, This implies, at constant pH,
-dSt/dt = kSt
(1)
where
k ~ specific reaction rate, sec-l
t = time, see
St ~ total sulfite concentr~tion, ~
July 1911 - FR
-------
-54-
90
100
&> " ~.O
'-....--- SULFITE
"
10 ,IV ,.8
"
"
"
~ 60 "'"
'-t tg.
o
1i 50 ,.~
f:
£
18.0 ,.2
Initial Solutions
--2tl- S as s04.i
'0 I 0 4.05 2.8 '.0
II C 4.00 5.1
III I:J. 4.13 1.9
IV - - 4.05
Soln. IV contained 0.16i
20 benzoic acid 2.8
10
o
-t 2.6
o
o
1
2
, 4
Time, hours
Figure 1
5
6
The Oxidation of Bulfi te to Sulfate in the
System CaO-SO~-SO~-H20 at 50°C With Op Gas
July 1971 - FR
-------
-55-
TABLE II
Effect of pH on the Rate of OxidatiQn in Calcium-
Bisulfite-Sulfite System With O2 at ~O°C
Time, traction -1
-R!L hours oxidizedJ ~ kJ sec
, 0 0 ,.66 X 10-e
0.' 2.'
1 5.4
2 13.0
3 22;8
4 22.7
, 37.9
6 48.3
,., 0 0 1. 72 X 10-9
0.' 6.1
1 20.4
2 43.2
3 59.0
4 70.4
4.0 0 0 4.31 X 10-9
0.5 18.6
1 44.0
2 79.5
3 89.5
4.5 0 0 1. 03 X 10-4
0.5 34.8
1 76.0
1.5 99.0
6.10 0 0 1.16 )( 10-40
6.45 0.5 40.3
7.10 1 87.0
7.45 2 -100
July 1971 - FR
-------
100
'#.
..
.,
.p~
~
f
20
10
o
July 1971 - FR
-56-
1
,
Time, hours
2
1!tl
o ,.0
A '.5
D 4.0
. 4.5
A 6.0 to 1.0
Figure 2
Effect of pH on the Oxidation of Sulfite to
SUlfate in Calcium System at SO~C
4
5
-------
-57-
The integral form
log (St/StO) = -kt/2.303
(2)
indicates that the slopes of the lines (Figure 2) are equal to -k/2.303
for each experimental condition. The values of k are listed in the last
column of Table II. The magnitude of k is directly proportional to the
oxidation rate.
The value of the total sulfite, St, can be expressed by
St = S032- + HS03- + H2S03 + CaS03
(3 )
Omitting the negligible amount of CaS03, and using the ionization constants
of the H2S03' K, and K2,
St = (HS03-)([K2/(~)J + 1 + [(~)/K1J)
Therefore dSt/dt 0: d(HS03 -) /dt at constant pH.
(4)
then
If the oxidation rate is related to the pH in a simple relation,
dSt/dt = k'St(~)n
(5 )
where
k = k' (:tr+")n
(6)
or
log k ~ log k' - npH
(7)
The values of k in Table II are plotted in Figure 3 according to equation
7. The relation is a~ost linear betweer. pH 3 and 4.5 with a slope of
about -1; at the highest pH the relation js obscured perhaps by the difficulty
in maintaining constant pH. The oxidation rate in the high pH range may be
influenced also by the presence of precipitated CaS03 and other solids
which make the system heterogeneous instead of homogeneous. Consequently
the oxidation rate at the high pH should be considered differently. For
the system in the pH range 3 to 4.5, the oxidation rate may be expressed
by equation 5 with n = -1.
July 1971 - FR
-------
~)(10"4
10"4
....
,
(J
Q)
fI)
..
~
..
$ 5)(10..5
QI
J.4
~
.p
~
f:
(J
5
(J
~
10..5
,
July 19~(1 - FR
-58-
n "* -1
.----------0
4
6
5
pH
Figure 3
Effect of pH on the Oxidation Rate
of CaSO~..Ca(HS~)~ at 50°C
-------
-59-
A series of tests was made to compare the four organic acids
(benzoic, phthalic, glycolic, and adipi.c) at constant pH. The test solution
was prepared by suspending 0.3 gram CaCOs in 500 ml H20 and absorbing 802
gas so that the final solution had a pH of 4.0 and a total sulfite concentra-
tion of 0.0123~. Then 0.5 gram of each organic acid was added to portions
of the above solution and the pH adjusted t.o 4.0 by 9.dding appropriate
amounts of CaCOs. After purging with nitrogen for about 40 minutes at 50°C,
pure oxygen was introduced at the rate of 21 ml/min through the same apparatus.
The pH of the solution was maintained constant at 4.0 during the run by
adding powdered GaCOs as needed; 10-ml samples were taken at intervals for
determination of the total sulfite concentration by iodine titration.
The same straight-line relationship was found between log total
sulfite in the solution and time (Figure 4).' The specific reaction velocity
values, k, are listed in Table III. All the organic acids tested retarded
the oxidation of sulfite to sulfate by oxygen gas at pH 4.0 and 50°C in the
order benzoic> glycolic> phthalic> adipic> no acid.
The effect of the concentration of benzoic acid on the specific
rate constant for oxidation at 50°C and pH 4 is shown in Figure 5. The
lower the concentration of benzoic acid, the smaller the effect of oxidation
retardation, and finally at about O.oooy~ benzoic acid the effect was not
detectable. The concentration effect may be expressed empirically by
k = kOA/(A + Bm)
where A = 4 X 10-4, B = 0.5, m = benzoic acid concentration, and kO =
reaction rate constant when m = O.
Our results definitely indicate that all four organic acids
behave as oxidation inhibitors to the oxidation of sulfite by pure O2; this
is contrary to the pilot-plant results using flue gases (May 1970 Applied
Research Branch report). Further study will be made to clarify this
discrepancy and to determine the mechanism of the oxidation.
July .1971 - FR
-------
100
.,p..
..
., ~
~
j
20
1
-60-
~
o Benzoic
A Phthalic
D Adipic
. Glycolic
. None
Time, hours
F1~e~
Effect of Orp;anic Acids, on the OXidation of Sulfite to
Sulfate in Calcium SYstem at sooa and pH ~.O
JulY 1971 - FR
-------
-61-
TABLE III
Effect of Organic Acids on the Rate of Oxidation
in the Calcium Sulfite-Bisulfite
System at pH 4 and 50°C
Concn., Time, Fraction
Acid --L- hours oxidizedJ~ k, sec-1
Benzoic 3.2 X 10-3 0 0 1.89 X 10-5
0.5 7.0
1 21.4
1.5 35.3
2 44.5
3 53.5
Benzoic 0.1 0 0 2.36 X 10-6
0.5 1.7
1 4.3
2 8.5
3 12.6
4 16.2
Phthalic 0.1 0 0 4.25 X 10-6
0.5 2.6
1 7.6
2 14.2
3 21.1
4 27.3
5 32.4
Glycolic 0.1 0 0 2.93 X 10-6
0.5 2.5
1 5.0
2 10.0
3 15.0
4 19.0
5 24.0
6 27.2
Adipic 0.1 0 0 3.20 X 10-5
0.5 5.4
1 23.2
1.5 43.4
2 57.9
2.5 68.6
3 77.1
4 88.0
July 1971 - FR
-------
,
""
~ 2
..
....
..
"
4>
r: 10.!!!
EI
o
'"
~
II
...
tI
E
tI
i
-62-
5
1
5
,
2
10.e
0,000'
0.00
0.01
Benzoic acid conoentration, i
0.10
F1~~ 5
Effect of Benzoic Ac!4 Concentration on the
Oxidation of SUlfite to SUlfate ill Calo4.ul\I System at 50Ge and pH '" 4..0
(J. D. Hatfield, Y. K. Kim,
M. E. Deming, R. L. Dunn)
July 1971 - FR
-------
-63-
Effect of Benzoic Acid on the Oxidation of Sulfite: In pilot-
plant studies (Applied Research Branch report, May 1970) in which simulated
flue gas containing sulfur dioxide and oxygen was passed through an aqueous
suspension of limestone, addition of benzoic acid to the suspension increased
the oxidation of sulfite to sulfate. In our laboratory studies, in which
oxygen was bubbled through solutions of sulfites or bisulfites, benzoic acid
increased the oxidation of sodium sulfite (May report) but decreased the
oxidation of calcium sulfite (July report). In further laboratory studies
of the effect of benzoic acid on the oxidation of sulfite, mixtures of sulfur
dioxide and oxygen were bubbled through aqueous suspensions of powdered reagent
calcium carbonate.
A mixture of sulfur dioxide (1.4 ml/min), oxygen (6.5 ml/min),
and nitrogen (63 ml/min) was bubbled through 500 ml of water at 50°C in a
l-liter reaction vessel until the pH dropped to 4.0. The gas flow was
maintained for 3 hours and the pH was held at 4.0 % 0.1 by addition of small
amounts of powdered calciUm carbonate. The sulfite sulfur in solution was
determined by iodine titration of 10-ml samples that were withdrawn periodically.
The exit gas from the reaction flask con~ained no detectable sulfur dioxide,
and the sulfate sulfur in the suspension was obtained by difference. The test
was repeated under the same conditions except that the initial water contained
O.l~ benzoic acid. As shown in Table III, with no benzoic acid 35 to 40% of
the sulfite was oxidized, whereas only 7 to 14% was oxidized in the presence
of benzoic acid. If all the oxidation is assumed to be that of the S02
added between samplings, the rate of oxidation is lowered by benzoic acid,
as shown in the last collillm of Table III.
In a similar test made for 5 hours with a mixture of the same
flows of .sulfur dioxide and nitrogen but with 17 ml/min of oxygen, o.li
benzoic acid was added after the gas had been passed for 3.5 hours. The
amounts of sulfate found, run 2 in Table III, indicate that benzoic acid
retards the oxidation of sulfite at the higher oxygen flow rate also.
Although benzoic acid retards the oxidation of sulfite under these conditions,
increasing the amount of oxygen in the gas increases the oxidation, but not
enough to overcome the effect of the benzoic acid.
September 1971 - FR
-------
-64-
TABLE III
Effect of Benzoic Acid on the Oxida.tion of
8ulfi te in the Presence of Calcium at 50:2
(Gas mixture of 1.4 ml/min 802, '63 ml/min N2, andOa
bubbled through suspension; CaCOs added a8
required to maintain pH 4)
Fraction, ~,
of 802 added
Conen., M, x 102 80. 8, during sampling
Run Benzoic acid Time, of 8 ;, of period that was
1!2:. concn. J ~ hr Total As 80.2- total 8 oxidized to 80.2-
O2 flow, 6.5 ml/min
1A 0.0 1 0.15 0.48 36.0 36.0
2 1. 515 0.98 35.4 34.6
3 2.295 1.395 39.2 46.8
IB 0.1 1 0.15 0.695 7.3 7.3
2 1. 515 1.37 9.5 11.1
3 2.295 1.978 13.8 22.0
O2 flow, 11.0 ml/min
2 0.0 1 0.75 0.62 17.3 11.3
2 1. 515 0.99 34.6 52.
3 2.295 1.03 55.2 95.
0.1 3.5 2.705 1.07 60.5 90.
4 3.11 1.345 56.8 34.0
4.5 3.54 1.635 53.8 32.5
5 3.98 1.96 50.8 25.0
september 1971 - FR
------- |