v>EPA
United States Industrial Environmental Research EPA-600/7-79-030
Environmental Protection Laboratory January 1979
Agency Research Triangle Park NC 27711
of Sulfur Dioxide
Oxidation in Aqueous
Solution
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5 Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8 "Special" Reports
9 Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems, and integrated assessments of a wideirange of energy-related environ-
mental issues
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use
This document is available to the public through the National Technical Informa-
tion Service. Springfield, Virginia 22161
-------
EPA-600/7-79-030
January 1979
Kinetics of Sulfur Dioxide
Oxidation in Aqueous Solution
by
J.L Hudson, J. Erwin, and N.M. Catipovic
The University of Illinois
Urbana, Illinois 61801
Grant No. R800303
Program Element No. EHE624
EPA Project Officer: Norman Kaplan
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
The rate of oxidation of sulfur dioxide was studied in a one liter
semi-batch reactor. There were two aspects of the study, viz., low pH
catalyzed oxidation and high pH uncatalyzed oxidation.
The results of the low pH experiments were as follows. Both MgSO4 and
catalyzed liquid-phase oxidation of SO- at low pH are zero order in SO,
concentration. While manganese is a very effective catalyst even at
concentration as low as 3 p.p.m., magnesium shows little catalytic action
at concentrations as high as 20,000 p.p.m. Energy of activation for the
Mn-catalyzed reaction was found to be 18.7 kcal/mole. Low pH oxidation of
SO2r catalyzed by MnSO4, is independent of pH between 1 and 4, while at
pH values above 4 the reaction speeds up due to the presence of sulfite
ions. Ionic strength did not appear to affect the reaction rate significantly,
but additional experiments at higher ionic strengths should be performed to
check this effect. Sulfate ions inhibit the reaction significantly. Kinetic
results can be presented conveniently using the ratio (S04=)a(jde(j to (S$>2)
initial. The mechanism of the reaction seems to involve complexing of Mn
with SO.=, but further work is necessary to make definite conclusions.
The primary study of the high pH case was a first order rate constant
of 0.0053 sec"1. The experiments were carried out under conditions such
•that, the oxidation was kinetically controlled with a constant oxygen
concentration of 0.0013 M.
Finally the results were applied to conditions in a sulfur dioxide
scrubber. The catalyst effect even at p.p.m. concentraion is very significant.
The effect of catalysts in high pH scrubbing systems subject to oxidation
is much greater than in low pH systems.
Studies carried out with chelating agents showed that previously studied
"uncatalyzed" reactions were in fact primarily trace metal catalyzed.
ii
-------
CONTENTS
Abstract ii
Figures iv
Tables Vl
I. Introduction 1
II. Previous Kinetic Work on Liquid-Phase Oxidation of Sulfur
Dioxide 4
III. Low pH Catalyzed Oxidation of Sulfur Dioxide 10
IV. High Ph Uncatalyzed Oxidation of Sulfur Dioxide 43
V. Uncatalyzed Studies 59
VI. Application
-------
FIGURES
Number Page
1 Effect of pH on the relative concentrations of S0_ species in
solution 2
2 Experimental apparatus 11
3 SO concentration as a function of time, with 20,000 p.p.m. MgSO
catalyst 7 16
4 SO concentration as a function of time, with 1 p.p.m. MnSO
catalyst 17
5 SO concentration as a function of time, with 3 p.p.m. MnSO
catalyst 18
6 SO concentration as a function of time, with 1 p.p.m. MnSO
catalyst and lower (SO,.) 19
2 initial
7 Zero-order reaction rate constant, k , as a function of the
initial pH, with 1 p.p.m. MnSO. catalyst 21
8 Zero-order reaction rate constant kQ, as a function of the initial
pH, with 3 p.p.m. MnSO catalyst 21
9 SO concentration as a function of time, with 1 p.p.m. MnSO and
pH. ._ . of 3.4 24
initial
10 Zero-order reaction rate constant, kQ, as a function of
(SO* ) jj ,/{SO_). ... , from Johnstone s Coughanowr's data. 28
4 added 2 initial '
11 Zero-order reaction rate constant, ko, as a function of
(S°4}added/(SV initial' with X P'P'm- ""^ catalyst 29
12 Zero-order reaction rate constant, ICQ, as a function of
, with 3 p.p.m. MnSO catalyst .... 30
13 S02 concentration as_a function of time, with 1 p.p.m. MnSO.
catalyst and
-------
Number Page
14 S02 concentration as_a function of time, with 3 p.p.m. MnSO
catalyst and -/(SO = 10 ....... « . . 32
15 Dimensionless diagram of kQ(with sulfate added) Ao (without any
sulfate added) as a function of
-------
TABLES
Number Page
1 Results of kinetic experiments at various pH values 22
24
2 Results of Johnstone and Coughanowr 27
3 Zero-order reaction rate constants for different experimental
conditions 33
4 Results of kinetic experiments dealing with sulfate effects ... 34
5 Ratios of k (with sulfate added)/k (without sulfate added) for
various experimental conditions 36
6 Kinetic results for the uncatalyzed sulfite oxidation at a
constant pH of 11.1 (with average values and standard
deviations) 52
7 Kinetic results for the uncatalyzed sulfite oxidation at variable
pH i . . 57
vi
-------
I. INTRODUCTION
The emission of sulfur oxides from stationary sources has become one of
the most serious air pollution problems in the world. Sulfur dioxide is one
of the major pollutants, formed mostly as a result of combustion of coal and
oil with high sulfur content in power plants and industrial installations.
The liquid-phase oxidation of SO2 is a very important factor in the air
pollution problems caused by this gas. In aqueous solutions, sulfur dioxide
readily undergoes oxidation to sulfate, especially in the presence of metal
ions as catalysts. Formation of sulfate in SO scrubbing systems has become
one of the most important unsolved problems in the removal of SO2 from stack
gases by wet scrubbing. The oxidation of SO2 in fog droplets in heavily
polluted air has been partly responsible for some of the air pollution disasters
in the past. This oxidation can also take place in cloud and rain droplets
in relatively unpolluted atmospheres.
When sulfur dioxide comes into contact with water the product is hydrated
sulfur dioxide, which can undergo dissociation to bisulfite or sulfite:
S02'H2° *k H + HS°3 ' Kl
2 k
HSO ~ =1 H+ + SO, , K, = _2_ (2)
3
The equilibrium constants for the dissociations have been the subject of a great
deal of investigation, as reported by Powell39. In our work, we adopt the values
KI = 1.74 x 10~2 (Tartar and Garretson48; Frydaman, Torsten and Sillen15) and
K2 = 6.24 x 10~8 (Yui57).
The relative concentrations of the bisulfite and sulfite ions depend
greatly on the pH of the aqueous solution. Fig. 1 shows the effect pH on
the relative concentrations of hydrated sulfur dioxide, bisulfite and sulfite.
Although Fig. 1 may not be quantitatively exact because all activity
coefficients are assumed to be unity, it does give a good qualitative idea
of the relation between various S02 species and pH.
-------
1.0
0.8 .
0.6
o
•H
+J
U
0.4
to
0.2
0.0
PH
Fig. 1. Effect of pH on the relative concentrations of SO2 species in solution.
-------
When the oxidation of sulfur dioxide in aqueous solutions is discussed,
it must be pointed out that the various species in solution react with
oxygen at different rates. The oxidation of sulfite is the fastest process.
Therefore, it is very important to know whether high pH or low pH oxidation
is taking place. Also, various catalysts can greatly affect the oxidation
rate; hence, it is necessary to differentiate between uncatalyzed and
catalyzed SO- oxidation.
In industrial SO. scrubbers, sulfur dioxide is absorbed in the scrubbing
liquid, the pH of which depends on the particular system used. The pH
range can extend from 4 to 10. Various metal catalysts are inevitably present
in the scrubbers. Sulfates formed by SO- oxidation must be either regenerated
and removed from the system as solids or purged from the system in liquid
form, in either case, additional scrubbing solution must be added to the
system as make-up, and this significantly increases the operating cost of
the process. Also, since the sulfate form of a salt is much more difficult
to regenerate than the bisulfite or sulfite form, oxidation essentially
represents a loss of SO. product in processes which involve regeneration of
the absorbed SO-. To be economically acceptable, regenerative scrubbing processes
must minimuze this sulfate formation. SO- oxidation rates in scrubbers can
reach values as high as 20-30% (% of total SO2 absorbed which is oxidized) in
some processes (for example, in the double alkali process which uses dilute
sodium sulfite as the scrubbing solution).
In the atmosphere, SO- is absorbed by fog or rain drops and oxidized to
sulfate, which is more toxic than the sulfur dioxide gas. Atmospheric
oxidation can take place with the aid of metal catalysts which serve as
condensation nuclei for water vapor or in the presence of ammonia when even
the uncatalyzed reaction can produce significant levels of SO4 due to the
maintaining of a higher pH by NH3.
The S02 liquid-phase oxidation process itself is poorly understood in
terms of its kinetic characteristics and mechanism. Some studies that have
been done have given contradictory results. The purpose of our study,
therefore, was to perform kinetic experiments which shed more light on the
factors that influence SO2 oxidation. Since traces of metal salts that
catalyze SO- oxidation are present in both the atmosphere and the scrubbers,
a great deal of our attention was focused on the low pH catalyzed reaction.
We also considered the high pH uncatalyzed SO2 oxidation for which very few
reliable data have been obtained in the past. Better understanding of the
characterisitcs and mechanism of both the uncatalyzed and catalyzed oxidation
of SO2 in aqueous solution should lead to improved efforts of controlling the
effects of this reaction. This work is the initial part of an overall study
of SO2 oxidation kinetics in all pH ranges.
-------
II. PREVIOUS KINETIC WORK ON LIQUID-PHASE OXIDATION OF SULFUR DIOXIDE
Although many authors do not specifically mention the pH at which their
experiments were done, the kinetic work on SO_ oxidation can clearly be
divided into two groups: the low pH oxidation (pH<5) and the high pH
oxidation (pH>7) . A lot of kinetic experiments have been performed at low
pH with the purpose of extrapolating the results to the atmospheric oxidation
of SO2 in cloud, fog and rain droplets. Most of the work at high pH (sulfite
oxidation) has been done to facilitate measurements of interfacial areas in
gas-liquid contactors. If the kinetics of a reaction is known, interfacial
areas can be calculated from the theory of gas absorption accompanied by a
chemical reaction. Sulfite oxidation kinetics is also used for evaluating the
performance of aerobic fermentors.
Low pH Oxidation
Uncatalyzed SO2 oxidation at low pH has received little attention since
it occurs at very slow rates. Hoather and Goodeve22 found that, with SO2
present initially at 0.01 N, the time required for the reaction to go halfway
to completion at 406C was of the order of ten days. However, low pH oxidation
of SO_ can be very effectively catalyzed by metal salts, and the influence
of a large number of catalysts has been studied extensively. There are two
common methods for obtaining kinetic data for the catalyzed SO., oxidation at
low pH: kinetic studies in bulk solution and absorption studies, either in
water droplets containing catalyst or in packed columns.
Hoather and Goodeve22 used a dilatometric technique to follow the bulk
phase oxidation of sulfur dioxide at 35°C, with manganous sulfate as the
catalyst. SO2 concentration was in the range 0.0005 - 0.005 M, while oxygen
concentration was of the order of 10~4 M. Catalyst concentration varied from
3 x 10~6 M (0.5 p.p.m. by weight) to 8 x 10~5 M (12 p.p.m.). They found that
the reaction was zero order in both SO- and oxygen until concentrations of
either dropped below about 1.25 x 10~5 M. The rate was proportional to the
1.7 power of the catalyst. Addition of sulfuric acid reduced the rate of
reaction although it remained zero order. For unexplained reasons, the acid
formed by the reaction did not slow the reaction and cause its graph to curve
away from zero order line. By making several runs at 30°C, an activation
energy of 27,300 calories was determined.
-------
Bassett and Parker5 bubbled a mixture of SO, and O_ through solutions
of various salts for several hours at 25°C. SO2 reached high concentrations
in solution, while the catalyst concentration in various runs ranged from
10~3 M upward. Results are difficult to interpret because there is no basis
for deciding whether the reaction rate was limited by the resistance to mass
transfer. Manganese salts were the most potent catalysts, being many times
more effective than cobalt and nickel salts. Sulfates of these metals were
quite a bit more efficient than chlorides. Ferric and cupric salts were
also active as catalysts. MgSO4, ZnSO4, NH4C1 and NaCl were ineffective.
Hydrochloric acid reduced the rate only slightly for concentrations up to
0.1 M, but it retarded the reaction remarkably at higher concentrations.
This reduction in rate is attributed to decreased oxygen solubility at large
ionic strength.
Johnstone and Coughanowr2^ studied the kinetics of the liquid-phase
oxidation by mixing solutions of SO_ with solutions of oxygen containing
MnSO4 and quenching the reaction after fixed intervals of time by pipetting
samples into standard iodine to determine the amount of SO^ remaining.
Initial SO2 concentration was around 0.002 M, while oxygen concentration
varied slightly around 0.0005 M. Catalyst was present at 4-14 p.p.m.
Temperature was 25°C. The rate of reaction always remained constant until
the oxygen content was essentially depleted. The apparent zero-order
reaction constant was found. When sulfuric acid was not present in the
solution initially, the rate constant was proportional to the square of
MnS04 concentration. When sulfuric acid was added to the catalyst solution
before mixing, the initial rate decreased as the acid concentration increased.
The initial apparent zero-order reaction rate constants were found for
several concentrations of externally added acid at a MnSO. concentration of
12.5 p.p.m. As in the case of Hoather and Goodeve22, the sulfuric acid
formed by SO2 oxidation itself did not appear to slow the reaction down.
Iron sulfate was found to be several orders of magnitude less effective
than MnSO., while NiSO. at 1000 p.p.m. did not show any catalytic action.
Junge and Ryan28 bubbled dilute mixtures of SO2 (5-200 p.p.m. by volume)
in air through different 1 p.p.m. catalyst solutions to investigate the
formation of sulfate. Their results showed that oxidation did not take place
significantly without a catalyst; there was also no noticeable increase in the
reaction rate with MaCl present. Of the salts examined McCl2 was the most
effective catalyst, followed by CuCl2» FeC12 anc* CoC12* T*le authors found
that formation of sulfate is highly dependent on the pH of the solution in
the pH range 1-5. As the pH decreases by the formation of sulfuric acid
from SO- oxidation, the production of sulfuric acid slows down and finally
stops. A decrease in pH by addition of hydrochloric or citric acid causes
similar effects—lower final concentrations of sulfuric acid are produced.
Junge and Ryan concluded that the major factor which restricts sulfate
formation is the hydrogen ion concentration.
-------
Johnstone and Moll2^ used fog chambers to study the production of
H2SO4 when high humidity SO2 fogs were exposed to nuclei of oxidation
catalysts. The effect of manganese and iron salts was investigated.
Manganese was found to be the better catalyst. Sodium chloride did not
catalyze the reaction at all.
Pritchett4^ mixed a solution of dissolved oxygen with a solution of
sulfur dioxide containing manganese sulfate catalyst in a continous flow
reactor. The completion of reaction was measured by the rise in temperature
using a thermistor detector. SO- concentration varied from 0.01 to 0.15 M,
oxygen from 10~3 to 10~4 M, while MnSO4 concentration was unusally high —
0.01 to 0.25 M. Oxidation of SO2 occurred at an unusual accelerating rate.
This was attributed to the formation of a catalytic complex by an autocatalytic
reaction. Oxidation proceeded at a rate proportional to the concentration
of the active catalyst and the square root of the oxygen concentration. This
behavior, not observed by anyone else was probably due to extremely high
catalyst concentrations. Sulfuric acid did not have any significant effect
on the rate of the reaction, even when it was added at a concentration which
was 15 times greater than that which would be produced by the reaction itself.
Conghanowr and Krause10 examined the rate of oxidation of sulfur dioxide
in a solution of manganous sulfate over a broad range of catalyst concentration.
Initial sulfur dioxide concentration was around 0.002 M, while the oxygen
concentration in solution varied from 0.3 to 0.7 x 10"^ M. Up to 15 p.p.m.
catalyst concentration a batch method was used; from 100 to 10,000 p.p.m.
a flow method was applied. The reaction was zero order with respect to both
sulfur dioxide and oxygen at all concentrations of MnSO4 used. The
concentration of manganous sulfate had a large effect on the rate of reaction;
from 0 to 100 p.p.m., the reaction rate constant, ko, is proportional to
the square of the catalyst concentration, but above 100 p.p.m., kQ increases
less rapidly up to about 500 p.p.m., after which it increases very slowly with
catalyst concentration. The reaction was easily inhibited by very small
amounts of contaminant.
Matteson, Stober and Luther32 studied the kinetics of the oxidation
of SO2 in aqueous aerosols of MnS04. They observed that the rate of reaction
decreased with increasing amounts of sulfuric acid formed. A decrease in pH
slowed and eventually halted the reaction. A four-step mechanism, involving
intermediate complexes with the catalyst, was proposed.
Commenting on iron and manganese catalyzed reactions, Foster14
concludes that it is the pH conditions existing at the beginning of the
experiment which determine the subsequent rate of oxidation. Discussing
the retarding effect of H2SO4 on the rate of reaction catalyzed by manganese,
he mentions the work of Grozdovskii1^ and Kashtanov and Ruizthov2', which
indicated that the reaction rate is reduced with time as a result of part of
the manganese forming a stable complex with the sulfuric acid. Foster
suggests that at the low concentrations of MnSO* and H-SO. used by Hoather &
-------
Goodeve22 and Johnstone & Coughanowr24, this complexing reaction takes places
so slowly that very little complexing occurs in the time of the experiment.
Gunn and Saleem20 studied the kinetics of the absorption of sulfur
dioxide and oxygen and the catalyzed liquid phase oxidation of SO in a
co-current gas-liquid contactor packed with glass fibers. Using the
steady-state theory of absorption accompanied by a chemical reaction, the
reaction kinetics are deduced from SOg absorption rates. Partial pressures
of SO2 and O2 in the gas varied from 0.003 - 0.02 atm and 0.09 - 1 atm
respectively. The concentration of manganous sulfate catalyst in the liquid
phase was maintained at 0.01 M (around 1500 p.p.m.). Order of reaction with
respect to oxygen was found to be zero, while the order in SO2 was zero if
pSOp is hi?h enough (if not, there is a small positive dependence on SO2
concentration). An increase in sulfuric acid concentration, obtained by
recirculating the liquid in the contactor and building up H2SO4, decreased
the reaction rate. The authors presented a graph showing the zero order
reaction rate constant as a function of acid concentration.
Cheng, Corn and Frohliger8 studied the reaction kinetics of water
soluble aerosols and SO- in air at p.p.m. concentrations. A steady-state
in the aerosol droplet absorption and oxidation of SO2 was attained in which,
according to the authors, SO2-H2O (unionized form) combines with metal ions
to form an intermediate complex. MnS04 was about four times more effective
in catalyzing the reaction than MnCl2 or CuSO4.
Walter-*2 was the only one who studied catalyzed SO2 oxidation kinetics
in bulk solutions by employing a continous supply of oxygen. However, for
a very effective catalyst like MnSO4, he could not get oxygen into the solution
fast enough, so that his reactions were limited to the oxygen initially
present in to SO, solution. The catalysts used were MnSO4 (1.6 - 6 p.p.m.),
MgSO. (1000 - 18,000 p.p.m.), ZnSO4 (4000 - 20,000 p.p.m.), CaSO4 (20 - 120 p.p.m.).
In all runs the initial SO. concentration was around 0.002 M (pH^2.7) and the
oxygen about 0.006 M. Temperature was kept constant at 25°C. A zero order
reaction was obtained in all cases. Relatively good agreement with the results
of Johnstone and Coughanowr24 for MnSO4 catalyst was obtained. The reaction
rate constant depended on the 1.9 power of MnSO4 concentration. For MgSO4,
ZnS04 and CaSO4, the power dependence of k0 on the catalyst concentration was
0.71, 0.73 and 0.39 respectively, but not enough points were obtained to
confirm these values more definitely. MnSO4 appeared to be at least three
orders of magnitude more effective than the other catalysts.
In recent work, Powell ' and Ness" focused their attention on the
uncatalyzed oxidation of aqueous solutions of bisulfite, with the hope that
the information obtained by their approach would yield a firmer basis for
understanding the chemistry, kinetics and mechanism of more complicated SO2
systems. The reaction was studied in the pH range 1.2 - 1.7. The rate of
oxidation was found to be first order in bisulfite ion concentration. Powell's
first order reaction rate constant was 0.061 hr"1, while Ness, using more realistic
-------
expressions for the activity coefficients of the sulfur dioxide-bisulfite
equilibria, obtained a value of 0.0065 hr"1.
In most of the work presented, the pH values of the solution in tlie course
of experiments were not given, but since in most cases the initial concentration
of sulfurous acid was of the order of 10~2 - 10~ M, we can assume that the
pH was below 4. It is obvious that manganese salts are by far the best
catalysts for SO2 oxidation at these low pH values. The reaction appears to
be zero order with respect to SO2 concentration, no matter what catalyst
is used. The zero order dependence in oxygen, observed by many authors,
might me questionable because in most experiments the initial oxygen
concentration did not vary much, and in all cases oxygen was present in very
small amounts and was the limiting reactant.
High pH Oxidation
High pH SO2 oxidation occurs at higher rates than low pH oxidation due
to the presence of sulfite which reacts very rapidly with oxygen, especially
in the presence of catalysts. Methods employed in studying this process
include bulk kinetic experiments, usually rapid-mixing techniques in flow
reactors, and absorption studies, either in wetted-wall columns or in
mechanically agitated sulfite solutions.
The experimental work on the uncatalyzed sulfite oxidation and the numerous
problems that various workers had with it be will be described in Section IV.
The only two catalysts for which high pH kinetic data are available are
cobaltous (CO"*"1") and cupric (Cu++) ions. Absorption of oxygen in Na2S03
solutions in the presence of CoSO^ as catalyst is particularly useful in
evaluating interfacial areas of gas-liquid contactors. The reaction kinetics
have to be well known in order that the theory of absorption with chemical
reaction could be used for this purpose. Interfacial areas have to be measured
at high chemical enhancement factors of the absorption process in order to
eliminate the influence of hydrodynamic conditions at the gas-liquid interface
on the absorption rate. Cobalt-catalyzed sulfite oxidation proceeds at
extremely fast rates, satisfied the above requirements, and has therefore been
the subject of a great deal of research. Cobalt is a more powerful catalyst
than copper (at least two times according to Westerterp et^al^4) and data
for cobalt are more reproducible than those for copper.
The reaction has very complex kinetics; for instance, according to
Astarita et al^2, it is zero order in O2 when the sulfite concentration
is 0.06 M; first-order when it is 0.25 M and second-order when it is 0.25 - 1 M.
Reith and Seek43 reviewed most of the data on cobalt catalysis obtained by
absorbing oxygen in Na2SO3 solutions. For 0.213=)<0.8 M, most investigators
agree that the reaction is second order in oxygen, first order in cobalt and
zero order in sulfite. Using the absorption method at low Na2SO3 concentrations
8
-------
(0.002 - 0.07 M), Yagi and Inoue56 found the reaction to be first order in
oxygen, first order in sulfite and first order in cobalt (at cobalt concentrations
of 5 x 10~8 - 6 x 10~7 M).
Fuller and Crist16 found that copper catalyst greatly affects the sulfite
oxidation rate even at concentrations as low as 10 M. This reaction, as well
as the cobalt-catalyzed one, has been studied by several researchers who
used the rapid-mixing flow method of Hartridge and Roughton21. Aqueous
solutions of oxygen and sodium sulfite were passed through a rapid mixer,
and the extent of reaction was followed by the temperature rise of the reacting
stream in an observation tube following the mixer. Barron and O'Hern4
used sulfite concentrations of 0.05 - 0.5 M and cupric ion concentration of
10~9 - 10~4 M. The reaction order was zero in oxygen, 3/2 in sulfite and
1/2 in catalyst. Chen and Barron?, working at a pH of 9, obtained the same
orders when cobaltous ion was used at concentrations of 10~7 - 10~ M.
Srivastava, McMillan and Harris47 worked with cobaltous ion at 10~-> - 10~ M,
the pH of the solution being 8.3. They obtained an oxygen order of one, cobalt
order of one, and sulfite order of zero for (S03=)>0.04 M. For (SO3=)<0.02 M,
there was a marked deviation from zero order. A couple of mechanisms were
proposed by all these researchers, one involving a free-radical chain and one
involving the formation of a cobalt complex with sulfite.
It is obvious from the above review that the order of reaction with respect
to both oxygen and sulfite changes with sulfite concentrations. It has been
established that cobalt is a more potent catalyst than copper, but nothing
is known about other metal catalysts. Considering the results of low pH
catalyzed oxidation of SC>2, experiments with manganese catalyst at high pH
should be very valuable.
-------
III. LOW pH CATALYZED OXIDATION OF SULFUR DIOXIDE
In this part of our work, we focused our attention on two catalysts:
magnesium and manganese. Magnesium is present in large quantities in many
SOj scrubbing systems, but any conclusions regarding its catalytic action
have yet to be made. Of all the catalysts studied, manganese has been shown
to be the most effective one by far, and is therefore very suitable for
studies of the influence of various factors on the catalyzed S(>> oxidation
rate.
All of our experiments were conducted in the pH range 1-4, where
bisulfite is the predominant SO2 species in solution. Therefore, we actually
studied the catalyzed oxidation of bisulfite. Of the factors which could
influence the rate of oxidation, we examined the effects of varying the pH,
increasing the sulfate concentration, and slightly altering the ionic strength.
From these effects and the observed order of reaction, certain conclusions
could be made about the manganese-catalyzed oxidation of SO2.
Experimental Apparatus
The experimental apparatus consisted of a reaction flask and associated
connections, oxygen supply, thermostat (see Fig. 2) and t'itrametic analysis
equipment.
The reaction flask was a three-neck, 1-liter Pyrex flask, modified by
four indentation baffles and a flattened bottom to accommodate a 5-cm.
teflon-coated magnetic stirring bar. The same flask was utilized for all
experiments dealing with catalyzed liquid-phase oxidation of SO-.
One fitting was a teflon-stoppered 25-ml. burette for the addition of
catalyst solution, with the tip sealed in a 24/40, tapered, ground-glass
joint. A pressure equalizing side-arm was attached to the burette for operating
at a slight gauge pressure, which served to increase the available oxygen
for the reaction, to eliminate the possibility of any leaks into the flask,
and to facilitate sample-taking. A three-way teflon stopcock was located
between the side-arm and the burette.
The second Bitting was a capillary glass tubing through which samples
of solution from the flask were taken. It was sealed in a 24/40, tapered,
ground-glass joint such that the bottom of the tubing cleared the top of
the magnetic stirring bar by approximately one centimeter. A male glass
10
-------
Magnetic
stirring bar
Stirrer
Sample
needle
Cooling
coil
Temperature
controller
Heater
Sample tube
Fig. 2. Experimental apparatus.
11
-------
Luer joint was connected to the top of the caplillary tubing. A stainless
steel valve (B-D MSO1) was fastened to this joint, with a curved syringe
needle connected to the valve.
The third fitting was a teflon stopcock which was sealed into a 24/40,
tapered, ground-glass joint. Oxygen was supplied to the flask through
this fitting.
The oxygen supply was a cylinder of oxygen, Linde Division, Union
Carbide Corp. After the high pressure cylinder regulator, a Foxboro low
pressure regulator (1.7 x 10" Pa. maximum inlet pressure, 4.1 x 10^ Pa. maximum
outlet pressure) was connected in series. Tygon tubing, 0.64 cm inside
diameter, led from the low pressure regulator to the third fitting. A second
section of Tygon tubing led from the outlet side of the low pressure regulator
to a mercury manometer.
The thermostat was a water-filled Plexiglas tank with stirrer,
heating element and auxiliary coil for cooling-water. Temperature was controlled
to within +_ 0.01°C by a Thermotrol controller (model 1053A). A magnetic
stirrer capable of 1850 r.p.m. was mounted close under the tank in order
to continuously stir the reactants by turning the teflon stirring bar in the
reaction flask. The stirrer was usually operated at 900 r.p.m.
To measure the pH of the reacting solution a Corning Model 10 expanding-
scale pH meter, equipped with a combination AgCl-KCl electrode, was used.
The accuracy is estimated to be ±0.02 pH units.
The main part of the titrametric analysis equipment was a special 5-ml
burette, on which the titration end-point could be read to within 0.002 ml.
The analysis used for determining SO2 concentrations will be described later.
The chemicals used for titrations, the catalyst salts (MgSO. and MnSO.),
sodium hydroxide and hydrochloric acid (used for changing the pH), and sodium
sulfate (used for increasing the sulfate concentration) were all reagent
grade. Magnesium sulfate was from Mallinchrodt Chemical Works, while manganese
sulfate, MnSO4'H2O, and sodium sulfate, Na2SO4*10H2), were from the
J. T. Baker Chemical Co.
All necessary weighing of chemicals was done on a Mettler or an Ainsowrth
balance of +_ 0.0001 gm sensitivity.
Deionized water used in the experiments was obtained from Roger Adams
Laboratory supply; its specific resistance was around 400,000 ohm-cm.
Experimental Procedure
Prior to each experiment, the reaction flask and associated fittings were
cleaned with a concentrated potassium dichromate cleaning solution (H_SO4 + K.Mn.O..)
12
-------
and rinsed many times with deionized water.
Before the start of a run 375 ml. of sulfur dioxide solution and
the stirring bar were placed in the flask. The SO2 solution of desired
concentration was obtained by diluting a 1 M SO2 solution with deionized
water. The 1 M solution was prepared by bubbling sulfur dioxide gas in
deionized water. SO2 gas was standard grade supplied from cylinders
obtained from the Linde Division of Union Carbide Corp.
The fittings were attached to the flask necks, the joints being sealed
with teflon sleeves to insure purity, and the Tygon tube leading from the
oxygen supply was connected to the oxygen inlet. The flask was then
placed in the constant temperature bath.
The temperature of the water bath was adjusted to 25.00°C and was measured
to +_ 0.01°C accuracy by a thermometer calibrated by the National Bureau
of Standards
Oxygen was blown through the reaction flask for several minutes to
purge the air. The pressure was then set a 800 mm Hg, the magnetic stirrer
was turned on, and a couple of hours were allowed for oxygen absorption.
To actually start a run, the catalyst solution was placed in the burette,
the oxygen inlet stopcock was closed, while the stopcock on the pressure
equalizing arm of the burette was opened. Then the burette stopcock was
opened to let 25 ml. of catalyst solution into the flask (the concentration
of the catalyst solution was so adjusted that additional of 25 ml. to the
375 ml. of SO_ solution in the flask resulted in the desired p.p.m. of
catalyst). Both stopcocks on the burette were then shut and the oxygen pressure
again adjusted to 800 mm Hg by opening the oxygen inlet stopcock. Stirrer
speed was set at 900 r.p.m.
The first sample was taken a few minutes after the catalyst, was added,
while other samples (usually 10 - 15 ml.) followed after desired time
intervals. The change in SO2 concentration in the flask was thus followed
as a function of time. Prior to each sample-taking, the sampling lines
were purged by letting several milliliters of S02 solution out. The solution
flowed through the lines readily when the valve was opened since the system
in the flask was under slight gauge pressure. pH measurements were made
several times in the course of the reaction by drawing a number of 5-ml. samples
and measuring acidity with a pH meter which was previously calibrated against
a known standard. These samples were used only for pH measurements and were
not titrated.
SO2 concentration of a sample (or more precisely, the total sulfur
concentration in the +4 oxidation state) was measured by quenching a
known volume of reactant solution in an iodine solution of known volume
and concentration. The iodine reacts with the various SO2 species in solution
to form sulfate ions according to the following reactions:
13
-------
SO -HO + I + HO N SO ~ + 2I~ + 4H+ (3)
^ ^ £ £ Q
HSO3~ + I2 + H2O ^ SO4= + 2I~ + 3H+ (4)
SO= + I + H O *• SO = + 2I~ + 2H+ (5)
322 •* 4
The excess iodine was then determined by titration with arsenious acid.
2H+ (6)
Since reactions (3) , (4) and (5) are pH-dependent, sodium bicarbonate
was used as a buffer to increase the pH and force the reactions to completion.
All reagent solutions were approximately 0.1 N (or diluted further) and
were prepared and standardized by standard methods46.
A sample calculation of the total SO2 concentration in the flask at a
particular time is shown in Appendix A. The time necessary to take a sample
and quench it was less than one minute.
A special advantage of our experimental method is the continuous oxygen
supply during the course of the reaction. All other investigators either
worked with only the oxygen that was initially dissolved in the SO2 solution
(0.0012 M at 25°C and 1 atm. oxygen pressure) 37 or could not get oxygen
into the solution fast enough (i.e., the process was mass-transfer controlled).
Due to good stirring, the process in our system was not mass-transfer controlled,
which was shown by calculating the activation energy from runs at 25 and 38°C.
This fact enabled us to feed the oxygen to the solution and sustain the reaction
for several hours, sometimes even days. The time scale in most of the previous
experimental works on catalyzed oxidation of 862 at low pH values was of
the order of second or minutes, depending on the catalyst concentration.
In most of our experimental runs, the total SO. (or the total sulfur
concentration in the +4 state) was between 0.01 and 0.02 M, giving an initial
pH of the solution of about 2.2 - 2.3. For runs at higher pH, we added
drops of 1 N NaOH to the solution prior to the start of the reaction. The
pH was raised to around 4 with 2.7 ml. of 1 N NaOH added to 375 ml. of SO2 solution.
To obtain pH values below 2.2, we added certain amounts of 1 N HC1. Twenty-five
ml. of IN HC1, added to 375 ml. of reactant solution, have a pH of approximately
1. In experiments in which the initial sulfate concentration in the reacting
solution was increased, sodium sulfate was added to the initial 375-ml. volume
in desired amounts, and ample time was allowed for it to dissolve completely.
NaOH and Na_SO, were used in the experiments since sodium was shown to have
14
no catalytic effect on SO- oxidation.
-------
Results and Discussion
We first studied the catalytic action of magnesium sulfate. In several
S02 scrubbing processes magnesium ion is present in large quantities in the scrubbing
solution; therefore, we worked with a very high MgSO4 concentration —
20,000 p.p.m. (0.17 M). We observed a zero-order reaction with respect to
SO2 concentration, with the reaction rate constant, ko» having a value of
1.11 x 10-4 moles/lit, hr. (see Fig. 3).
We also made a run for the uncatalyzed reaction at the same initial SO2
concentration. This run showed that 0.001 moles/lit, of SO- reacted in 15
hours. Comparing these numbers with the reaction shown in Fig. 3, we can see
that 20,000 p.p.m. MgSC>4 in solution speeds up the reaction only about 60%
over the uncatalyzed case. Since this is a rather small enhancement of the
reaction rate at a very high catalyst concentration, we conclude that MgSO4
is a poor catalyst for the bisulfite oxidation.
Bassett and Parker^, using solutions saturated in sulfur dioxide,
found that magnesium retarded the reaction at high concentration (above 0.05 M) .
Our result does not agree with their findings. However, in the MgSO4 that we
used, there was 0.001% manganese which could account for part of the catalytic
action since Mn is a very good catalyst.
Walter52 found that the catalytic action of calcium sulfate is roughly
the same as that of MgSC>4, while zinc sulfate was only 50% as effective.
In light of our result, we can state that Ca, Mg and Zn, all of which can be
present in scrubbing solutions in large quantities, are not very effective
from the standpoint of catlyzing" the undesirable SO2 oxidation to sulfate.
As already noted, many investigators found manganese sulfate to be a
very effective catalyst for bisulfite oxidation. Having this in mind, in
our studies of the manganese-catalyzed reaction, we used low catalyst
concentrations — 1 and 3 p.p.m. MnSO4 (6.62 x 10~6 M and 1.99 x 10~5 M
respectively). The reaction was found to be zero order with respect to the
SO2 concentration (i.e., total sulfur concentration in the +4 state).
The reaction rate constants were 2.5 x 10~4 moles/lit, hr. for 1 p.p.m.
catalyst (Fig. 4), and 10.6 x 10~4 moles/lit, hr. for 3 p.p.m. catalyst
(Fig. 5). Initial SO2 concentrations in both of these cases were around
0.012 M. However, the same reaction rate constants were obtained when the
initial SO2 concentration was almost an order of magnitude smaller
around 0.002 M (Fig. 6 shows the case for 1 p.p.m. MnSO4). We could not find
a value for ko at 1 or 3 p.p.m. MnSO4 in any of the articles described in
Section II because none of the authors used such low manganese concentrations.
However, if Johnstone & Coughanowr's24 or Coughanowr & Krause's^O results,
obtained for MnSO4 concentrations above 4 p.p.m., were extrapolated to 3 p.p.m.
by assuming a second order dependence of k0 on the catalyst concentration, a
value of 9.6 x 10~4 moles/lit, hr. would be obtained. Our ko of 10.6 x 10~4
moles/lit, hr. for 3 p.p.m. MnS04 is then in very good agreement with this value .
15
-------
17
°-02
pH. = 2.35
init.
PH(t=28) = 2'25
Slope = kQ = 1.11 x 10~4 moles/lit, hr.
8
12
20
24
28
Fig. 3.
catalyst.
16
Time, hours
SO- concentration as a function of time, with 20,000 p.p.m. MgSO4
-------
m
o
X
J
in
0)
M
i
CM
O
0)
12
11
10
" 9
(SO2)init^ 0.012 M
PHinit. = 2-24
PH(t=14)= 2-16
Slope = k_ = 2.50 x 10~4 moles/lit, hr
8
Time, hours
12
16
Fig. 4. 862 concentration as a function of time,
with 1 p.p.m. MnSO^ catalyst.
17
-------
12
2 11 .
-P
-rH
•H
(0
0)
fN
O
°-012 M
PHinit. = 2-18
2'15
Slope = k0 = 10.5 x 10~4 moles/lit, hr
30
60 90
Time, minutes
120
150
Fig. 5. S02 concentration as a function of time, with 3 p.p.m. MnSO^ catalyst.
-------
2 .
1.5 .
o
•H
X
4J
(0
(1)
1 .
CN
0.5
PHinit. ' 2'7
PH(t=6) = 2'65
Slope = k0 = 2.50 x 10~4 moles/lit, hr.
) 1234567
Time, hours
Fig. 6. S02 concentration as a function of time, with 1 p.p.m.
catalyst and lower (S^^ initial'
-------
In order to make sure that we were obtaining purely kinetic results
and that the reaction was not controlled by the rate of oxygen transfer to
the solution, the activation energy for the MnSC^-catalyzed reaction was
calculated from the kinetic data at 25 and 38.2°C with 1 p.p.m. catalyst.
The value for the activation energy obtained from k0 = 2.5 x 10~4 moles/lit, hr.
at 25°C, and kQ + 9.25 x 10~4 moles/lit, hr. at 38.2°C, was 18.7 kcal/mole.
This value is far too high for a mass-transfer limited process.
From our results it is clear that manganese is a very effective catalyst
even when present at only several parts per million in solution. Three p.p.m.
manganese sulfate in a scrubbing solution will cause 10 times higher SC>2
oxidation rate than 20,000 p.p.m. MgS04. It is very likely that manganese is
present in the scrubbing liquid in p.p.m. concentrations.
We saw in Section II that several authors observed a zero order in oxygen
for the catalyzed SO- oxidation at low pH values. In the same section (p. 8) ,
we questioned the validity of such observations. In the case of our experimental
results, the oxygen order cannot be determined since the oxygen concentration
in solution was kept constant by the continuous supply of the gas and by
efficient stirring of the reacting solution.
Neither manganese-catalyzed nor even magnesium-catalyzed S(>2 oxidation
at high pH values could be followed in our experimental system since they
were too rapid; all of SO2 in solution reacted within several minutes.
In studying various factors which could influence the catalyzed S02
oxidation, we worked with MnSO4 as catalyst since it is effective, and since
even small changes in reaction rates can then be observed. The influence
of pH was investigated first. As before, all experiments were done at 25°C
and 800 mm Hg oxygen pressure.
As mentioned already, Junge and Ryan28 observed that the production of
sulfate by SO2 oxidation was dependent on the pH of the solution in the
range 1-5, and concluded that hydrogen ion concentration is the major factor
which restricts sulfate formation. Fosterl4 made similar conclusions.
We varied the pH of our reacting solutions from 1 to 4. Working with 1
and 3 p.p.m. MnSO4, we found that the rate of bisulfite oxidation was
independent of the pH. The results obtained from 1 p.p.m. manganese sulfate,
with the initial SO2 concentration of around 0.013 M, are shown in Fig. 7,
while the reaction rate constants for 3 p.p.m. MnS04 and the same initital
SO2 concentration are plotted in Fig. 8. In both of these graphs, the pH
values shown are the ones observed at the start of the reaction. All of these
results, as well as those when the initial SO2 concentration was almost in
order of magnitude smaller, are summarized in Table 1. A representative
run is shown in Fig. 9.
20
-------
•sr
o
M
.e
(0
<0
i
.0
(S02)inlt.<1'
M
PH
Fig. 7. Zero-order reaction rate constant, krt,
as a
function of the initial pH, with 1 p.p.m. MnSO4 catalyst.
sr
o
rH
0)
14
12
10
8
2
0
init.'x' °-013 M
PH
Fig. 8. Zero-order reaction rate constant k , as <
function of the initial pH, with 3 p. p.m. MnSO4 catalyst,
21
-------
TABLE 1. Results of kinetic experiments at various pH values
(S02) initial' M
pK *• 2
0.013
0.010
0.010
0.011
0.009
0.012
0.011
0.008
0.002
0.002
0.002
0.002
pH * 3-4
0.012
0.011
0.011
0.010
0.009
0.009
Length of
run (hr)
15
14
14
12
14
11
11
7
4
4
6
6
20
20
15
12
12
11
pHinitial
1 p. p.m.
2.19
2.30
2.26
2.18
2.24
2.22
2.27
2.33
2.7
2.7
2.7
2.7
3.15
3.22
3.09
3.22
3.75
3.37
(continued)
22
pHfinal ko'
MnS04
2.12
2.15
2.13
2.13
2.16
2.15
2.20
2.26
2.65
2.65
2.65
2.65
k0/ = 2.5 x
°(ave)
2.70
2.77
2.70
2.82
2.98
2.90
moles/lit. hr. x 104
2.84
2.55
2.90
2.36
2.50
2.20
2.40
2.32
2.60
2.20
2.50
2.60
10~4 moles/lit. hr.
2.20
2.25
2.45
2.90
2.45
2.70
-------
TABLE 1. (Continued)
(so 1 M Length of
-------
n
o
10
rvi
O
ui
(SO). ...
init.
0.009 M
•
P«init. =
PH(t=ll)= 2-90
lope = kn = 2.70 x 10~4 moles/lit.hr
Time, hours
12
Fig. 9. S02 concentration as a function of time, with 1 p.p.m. MnS04 and
PHinitial of 3-4-
-------
When the pH was raised above 4, the reaction rate was greatly increased
in the first few moments, indicating that sulfite ion was present in solution.
We can therefore conclude that the hydrogen ion concentration has an important
influence on the rate of SO2 oxidation when the pH of the solution is above
4 because it determines the relative amounts of bisulfite and sulfite present.
However, when the pH is below 4, the hydrogen ion concentration does not affect
the reaction in any way.
It is interesting to see how the ionic strength of our reacting solution
was altered as we changed the initial pH. The ionic strength of our original
solution (initial (SO2) - 0.013 M, pH = 2.2) was 0.0065 moles/lit, while
upon additon of NaOH to obtain a pH of 4, the ionic strength increased to
about 0.01 moles/lit. According to the Debye-Huckel theory, the dependence
of our zero-order reaction rate constant on the ionic strength of the solution
of 25°C can be expressed as
log —il = 1.02 ZZ y, -MTT (7)
where kQ = zero-order reaction rate constant for ionic strength yj
kQ. . = zero-order reaction rate constant for ionic strength y__
Z , Z = valences of species involved in the reaction (see Laidler3",
A B
pp. 220-221)
For the change of pH from 2.2 to 4 with our method, the change in ionic
strength is too small to cause any significant change in the zero-order
reaction rate constant, even if the absolute value of the product ZAZB were
as large as 3.
However, when the pH of our solution was changed from 2.2 to 1 (by adding
25 ml. of 1 N HC1) , the ionic strength changed from 0.0065 to 0.04 moles/lit.
Theoretically, this change would cause a 25% change in ko if the absolute
value of the product Z&ZB were 1, or a 63% change if the absolute value
of ZAZB were 2. Large changes of kQ were not observed in our experiments,
indicating a rather small influence of ionic strength on the reaction. However,
it must be mentioned that the reliability of our results of pH values close
to 1 is less than the reliability at pH values above 2 (there was a large scatter
in data before the values for pH ^ 1, shown in Table 1, were obtained). To get
a clearer idea of the ionic strength influence on the rate of reaction, a larger
change in ionic strength than in our experiments would have to be made. At any
rate, the primary aim of this part of our investigation was not to study the
ionic strength effects, but to find out whether the hydrogen ion is in any
way involved in the reaction.
25
-------
In the past, several researchers (Hoather & Goodeve2^ , Johnstone &
Coughanowr2^ , Matteson e_t al^. 2, Foster*^ , Gunn and Saleem20} have observed
that increasing acidity slows down the SO2 oxidation rate. However, in all
these cases the pH of the solution was changed either by external addition
of sulfuric acid or by the formation and accumulation of H2SO4 in the course
of the oxidation reaction itself. Judging from our results, which show
independence of the bisulfite oxidation rate on the hydrogen ion concentration, it
it appears that it is the sulfate ion which affects the reaction rate in some
manner. Our next series of experiments, therefore, dealt with the influence
of externally added sulfate ions on the S02 oxidation rate at low pH.
Johnstone and Coughanowr2^ were the only authors who published detailed
data on the retarding effect of H2SO4 addition on the reaction. The reaction
rates went down as increasing amounts of sulfuric acid were added to the
solution prior to the start of the oxidation, but the reaction remained zero
order with respect to SO2 concentration. The authors analyzed the initial
apparent zero-order reaction rate as a function of H2SO4 concentration.
Table 2 shows the results of their experiments, which were done with 12. 5 p.p.m.
MnSO4 and an initial SO2 concentration of 0.002 M. Similar behavior was noticed
by Hoather & Goodeve22 (for 1-10 p.p.m. MnSO4) and Gunn & Saleem20 (for
1500 p.p.m. MnSO4) , but their results were not given in a clear form.
One of the most noteworthy conclusions which emerged from our experimental
analysis dealing with sulfate effects is the fact that kinetic results can be
conveniently presented using the ratio (S04=) added/
-------
TABLE 2. Results of Johnstone and Coughanowr
(S02}initial - °-°02 M
(02}initial - °-0003 - 0-007 M
12.5 p.p.m. MnSO.
Initial H2SO4 (added), M
kQ, moles/1it.hr. x 103
0
0.002
0.02
0.035
0.05
0.1
12.10
4.20
1.65
1.08
0.42
0.30
27
-------
NJ
00
•r4
H
(fl
0)
l-t
12
10
(S02}init. = °'002 M
12.5 p.p.m. MnSO.
0 1
10
20 30
(S04=)added/(S02>initial
40
50
(SO
Fig. 10. Zero-order reaction rate constant, k0, as a function of
~) ,, ,/(SO0). ... , from Johnstone & Coughanowr's data.
4 'added
-------
tsj
o
rH
X
.c
•
4J
U>
init. = 0-013 M
1 p.p.m. MnSO4
0
10
adde
initial
Fig. 11. Zero-order reaction rate constant, ko, as a function of
(S04 >added/(SO2}initial' with 1 P-P-m- M"5^ catalyst.
-------
12
10
(S02)init. = 0.013 M
3 p.p.m. MnS04
U)
o
o
iH
X
CO
(U
10 15
added/added/(S02>initial' with 3 P-P-m- MnS04 catalyst.
-------
13
°-013 M
P"init. = 2-34
PH(t=18)= 2'32
ro
O
4J
•rl
H
in
Q)
12
Slope = k0 = 0.85 x 10~4 moles/lit.hr.
CM
O
w
11
8 10 12
Time, hours
14
16
18
Fig. 13. S02 concentration as a function of time, with 1 p.p.m. MnS04 catalyst
and (S04=)added/(S02) initial = 1-
-------
w
to
CN
O
w
12
(S02)init.'1'0-0013M
pHinit. ' 2'75
PH(t=24)= 2.42
o
rH
X
•
•u
to
0)
11
10
(Slope = k = 1.65 x 10~4 moles/lit.hr.
8
12 16
Time, hours
20
24
28
Fig. 14. SO2 concentration as a function of time, with 3 p.p.m. MnS04 catalyst
and (SO =)
4 adde
initial
10.
-------
Table 3. Zero-order reaction rate constants for different experimental conditions.
Units of k : moles/lit.hr.
o
12.5 p.p.m. MnSO
1 p.p.m. MnSO
3 p.p.m. MnSO
3 p.p.m. MnSO
(S°4 'added
(S°2) initial
0
0.5
1
(S02)lnit.
(Johnstons
k
o
k
o
k
o
= 12.1
= 5.8
= 4.2
^
= 0.002 M
24
& Coughanowr )
x 10"3
x 10"3
x 10"3
(S02}iiilt. =
k =
o
k =
o
(pH. ..
f init
k =
o
2.5 x
1.24
- 2
•
0.84
(PHinit. -
10
20
k
o
k
0
= 1.65
= 0.75
x 10"3
x 10"3
k =
o
0.34
(pHinit. -
0.013 M
io-4
x 10~4
.30)
X 10~4
2.32)
x IO"4
2.75)
(so2)
(pH
k
o
(pH
k
o
init. = °
init.
= 3.63 x
init. = 2
= 1.55 x
-------
TABLE 4. RESULTS OF KINETIC EXPERIMENTS DEALING WITH SULFATE EFFECTS
.. . Length of
(S°4 } added' tS°2J initial run (hr)
1 p. p.m. MnSO , (SO_)
0
0.51
0.54
0.54
0.54
0.99
0.95
0.95
1.00
10.00
10.25
3 p. p.m. MnSO , (SO )
0
0
0
0
0
0
1.04
• .,_.,= 0.013 M
initial
SEE T
23
23
15
24
10
13
14
20
24
24
. . . , = 0.013 M
initial
2
2
2
2
2
2
7
initial
ABLE
2.28
2.30
2.32
2.30
2.31
2.31
2.29
2.34
2.75
2.75
2.17
2.17
2.16
2.16
2.18
2.15
2.26
4
k , moles/lit. hr. x 10
o
1
1.20
1.32
1.30
1.20
0.82
0.83
0.81
0.85
0.33
0.37
11.1
10.2
9.7
11.0
10.5
10.3
3.7
34
-------
TABLE 4 (continued)
(S°4=) added/ (S°2) initial
3 p. p.m. MnS04, (SO2>initi
1.06
1.05
1.06
1.10
1.00
10.10
10.05
10.00
10.10
20.00
21.00
3 p. p.m. MnS04, (S02)initi
0
0
1.00
1.05
Length of
run (hr)
= 0.013 M
7
8
8
6
7
24
24
24
24
20
20
= 0.080 M
.al
2
2
7
7
pHinitial
(cont'd)
2.26
2.21
2.23
2.25
2.20
2.80
2.80
2.75
2.78
2.83
2.90
1.60
1.60
1.80
1.80
k , moles/lit. hr.x li
o
3.7
3.5
3.5
3.65
3.7
1.5
1.6
1.65
1.55
0.74
0.75
11.1
10.7
3.7
3.8
35
-------
TABLE 5. RATIOS OF k (WITH SULFATE ADDED)/k (WITHOUT SULPATE ADDED) FOR VARIOUS EXPERIMENTAL
o o
CONDITIONS
k (sulfate added) 12.5 p.p.m. MnSO. 1 p.p.m. MnSO.
O 44
3 p.p.m. MnSO. 3 p.p.m. MnSO.
k (without sulfate (SO.) . .. =0.002 M (S0_) . €j_ =0.013 M (SO.). .. =0.013 M (SO,). .. =0.08 M
o ,, ,v 2 init. 2 init. 2 init. 2 init.
added)
kQ(S04~/S02 = 0)
k (SO "/SO. = 0)
O 4 2.
1.00
1.00
1.00
1.00
W
k (SO. /S0_ = 0.5)
O 4 2
ko(so4=/so2 = o)
0.48
0.50
ko(so4 /so2 = i)
ko(S04~/S02 = 0)
0.35
0.34
0.34
0.34
= 10)
k (SO/SO, = 0)
O 4 2
0.14
0.14
0.15
20)
k (S0./S00 = 0)
O 4 £
0.6
0.07
-------
to infer that the reaction is decreased proportionally in each column. This
is the basis for combining the results into a dimensionless diagram (Fig. 15),
showing the dependence of the ratio ko(with sulfate added)Ao(without any
sulfate added) on the ratio (SO4=)added/(SO2)initial is an extremely important
kinetic index of the SO2 oxidation rate. No matter what the initial SO2
concentration and the catalyst concentration are, at a certain value of this
ratio, the reaction rate is always slowed down to the same extent. This
particular characteristic of SO2 oxidation kinetics at low pH has not been
reported before.
It is interesting to note that Johnstone & Coughanowr's data fit in very
well with ours, although their source of sulfate ions was sulfuric acid. This
fact seems to confirm the independence of the SO2 oxidation rate on the
hydrogen ion concentration at low pH values, which we observed in earlier
experiments. Therefore, in the present SO2 system, the sulfate ion is the
only species responsible for retarding the reaction.
The experiments described above were performed with sulfate that was
externally added to the SO2 solution. It must be remembered, though, that
sulfate is also produced in the course of the SO2 oxidation itself. Hoather &
Goodeve^S and Johnstone & Coughanowr24 observed that, for unexplained reasons,
the acid formed by the reaction did not slow the oxidation down. However,
they used a limited amount of oxygen and could not follow the reaction to any
great extent. With our stystem of continuous oxygen supply, we could follow
the reaction until practically all of SO2 had reacted. Nevertheless, the
"internally produced" sulfate does not appear to affect the reaction rate
significantly. In some of our runs, a slight retarding effect could be
noticed after the reaction was allowed to proceed for more than 25 hours
(see Fig. 16). Even when observed, the effect was much smaller than would
be expected from the generalized diagram of Fig. 15. This relative lack of
influence of the "internally produced" acid is rather puzzling. Foster's^-4
theory that, at the low concentrations of MnS04 and H2SO4 used by Hoather &
Goodeve and Johnstone & Coughanowr, the complexing reaction between them
takes place so slowly that very little complexing occurs in the time of the
experiment, does not appear fully adequate since sulfate concentrations did
reach significant levels in the course of our reaction (up to 0.01 M) .
It is true that in all cases examined, the manganese concentration was very
low — it did not exceed 12.5 p.p.m. (8.2 x 10~5 M).
An attempt at explaining the shape of graphs in Fig. 11 and Fig. 12
could include the effect of the decreasing oxygen solubility with increasing
ionic strength on the reaction rate constant. It could be assumed that the
reaction rate constant, ko, can be written as ko = k (O2)n, where k is some
constant and n is the unknown order of reaction with respect to oxygen.
As more sulfate is added, the oxygen concentration might be decreasing due to
increased ionic strength. Bassett and Parker^ observed an effect of this kind
as they added large amounts of HC1 to their reacting solution; the reaction
rate was affected for HC1 concentrations above 0.1 N. Also, the changing ionic
strength, due to the addition of sulfate, might be reducing the reaction rate
constant according to the Debye-Huckel theory (see equation (7) ).
37
-------
U)
CO
•a
0)
•a
•a
(0
o
in
CN
(0
-P
3
4J
•H
-a
0)
T3
•a
(0
o"
en
(N
•P
-rH
1.0
0.8
0.6
0.4
0.2
0 1
10
(S02)init.
init.
= 0.002 M, 12.5 p.p.m. MnS04
= 0.013 M, 1 p.p.m. MnS04
= 0.013 M, 3 p.p.m. MnS04
= 0.08 M, 3 p.p.m. MnSO.
20 30
(S04-)added/(S02>initial
40
50
Fig. 15. Dimensionless diagram of kQ(with sulfate added)/kQ(without any sulfate
added) as a function of (SO4=)added/(S02)initial.
-------
10
10
(0
0)
-* 5
1 p.p.m. MnSO4
PHinit. - 2'30
PH(t=37)= 2'15
4 8 12 16 20 24 28 32
Time, hours
Fig. 16. Effect of the "internally produced" sulfate.
36 38
-------
However, the dimensionless graph on Fig. 15, which is independent of
the initial SO_ concentration and the concentration of SO4= added, seems to
indicate that oxygen solubility effects cannot account for the observed behavior.
The retardation of the reaction rate depends only on the ratio (S04 )a(jded/
^SO2^initial' and not on the absolute concentrations of the added sulfate
(which varied from 0.001 M to 0.26 M) , and, therefore, not on the ionic strength.
We can then conclude that oxygen solubility and ionic strength effects do
not appear to be significant in our system.
The fact that SO2 oxidation rate is zero order in SO2 concentration and
second order in manganese (the latter observed by Johnstone and Coughanowr24
Coughanowr & Krause10, and Walter52) r while inhibited by sulfate, seems to
indicate that an important factor in the kinetic mechanism is the formation
of a complex between manganese and sulfate. Since the order with respect to
manaanese is two, the catalyst must be involved in the rate-de terrain ing step.
It is possible that manganese in the +3 state is formed and that it complexes
with sulfate. It appears from dimensionless diagram in Fig. 15 that the complexing
reaction, either with sulfate or bisulfite, or both, involves an equilibrium
since the same portion of catalyst is always tied up at a certain (SO4=) a(jded^
2 oxidation at low pH,
the oxygen order of the reaction should be determined and a closer study of the
influence of the ionic strength on the rate of reaction should be made. A
better idea of the species involved in the rate-determining step of the reaction
would then be gained.
The results of our experiments can be summarized as follows:
1. Both MgSO4 and MnSO4 catalyzed liquid-phase oxidation of SC>2 at low
pH are zero order in SO2 concentration.
2. While manganese is a very effective catalyst even at concentrations
as low as 3 p.p.m. , magnesium shows little catalytic action at concentrations
as high as 20,000 p.p.m.
3. Energy of activation for the Mn-catalyzed reaction was found to be
18.7 kcal/mole.
4. Low pH oxidation of SO2, catalyzed by MnSO4, is independent of pH
between 1 and 4, while at pH values above 4 the reaction speeds up due to
the presence of sulfite ions.
5. Ionic strength did not appear to affect the reaction rate significantly,
but additional experiments at higher ionic strengths should be performed to
check this effect.
40
-------
6. Sulfate ions inhibit the reaction significantly.
7. Kinetic results can be presented conveniently using the ratio
initial-
8. Dimensionless diagram of ratio ICQ (with sulfate added) /l^ (without
any sulfate added) vs. ratio (SO4=) added/(s°2) initial was obtained for a
range of SO2 concentrations from 0.002 to 0.08 M and MnSO4 catalyst concentrations
from 1 to 12.5 p. p.m.
9. The mechanism of the reaction seems to involve complexing of Mn
with SO4=, but further work is necessary to make definite conclusions.
41
-------
IV. HIGH pH UNCATALYZED OXIDATION OF SULFUR DIOXIDE
The uncatalyzed SO2 oxidation at high pH values is extremely important
in the atmospheric chemistry of sulful dioxide, particularly for its
oxidation in drops with NH3 present. Also, since most industrial SO2
scrubbers work at higher pH, the oxidation rate is significant in demonstrating
how rapid the SO2 oxidation at high pH is, even without the presence of
catalyst.
As the pH of a solution containing SO2 rises, 803" becomes the pre-
dominant species; therefore, by high pH SO2 oxidation we primarily mean
sulfite oxidation. If the pH is above 9, all of SO2 is in SOo~ form (the
equilibrium constants being K^ = 1.74 x 10~2, K2 = 6.24 x 10-°) (Fig. 1).
Literature
The most extensive work on uncatalyzed sulfite oxidation was done by
Fuller and Crist16 in 1941. They were the first to show that the oxidation
of sulfite could occur in the absence of metals. The essential parts of
their apparatus were a quartz reaction vessel in which a weighed amount of
Na2SO3 and 40 ml. of water were placed, a gas buret for the direct measure-
ment of the rate of oxygen absorption, and a thermostat (the temperature
was 25°C). The amount of oxygen absorbed during a run, together with the
iodine titer at the end, was used to calculate the sulfite concentration
present at t0. Fuller and Crist went through extreme care in purifying
all reactants involved, distilling the water in a special way to make
certain that no heavy metals came over. Their results show a first order
reaction in sulfite, with the reaction rate constant, k», having a value of
0.013 sec"1. They worked with the oxygen pressure of 1 atm. Initial sulfite
concentrations ranged from 0.01 to 0.05 M, and the pH of the solution was
in the range 8.2 - 8.8. However, the authors indicate that at sulfite
concentrations above 0.015 M, the reaction was affected by the rate of
oxygen transfer to the solution at their maximum practical stirrer speed
of 1600 r.p.m. Considering that they used a value of 5 x 10~6 for the second
dissociation constant of SO2, which is two orders of magnitude off the
correct value, some question arises over their results. Taking the correct
value for K2, there is about 15% less sulfite at pH = 8.2 than Fuller and
Crist assumed (see Fig. 1). Only when pH is above about 8.9 is all of S02
converted to the 803" form.
In spite of some deficiencies, the work of Fuller and Crist is far more
reliable than that of other workers before them (BackstrOm3, Reinders et al.41,
Tittoff49) due to the purity of the substances used. The reaction is
43
-------
extremely sensitive to the presence of metal ions; some catalysts (primarily
copper) affect the rate even at concentrations as low as 10~8 M1^.
It must be pointed out that Fuller and Grist's16 constant kx includes
in it the oxygen concentration since the amount of dissolved oxygen remained
more or less constant in the course of the reaction. At PQ2 = 1 atm. ,
solubilities of oxygen in sulfite solutions of various concentrations are
given by Linek and Mayrhoferova31. For (SO3=) = 0.02 M, this solubility is
around 0.0013 M, which is close to the oxygen solubility in water given in
Perry's Handbook37 as 0.00123 M. The order of reaction with respect to
oxygen was not determined by Fuller and Crist.
Recent kinetic investigations do little to clarify the reaction
kinetics or mechanism. Barren and O'Hern4 used a rapid-mixing method to
eliminate the mass transfer of oxygen to the solution. Their tube-reactor
was made of plastic. Sulfite concentrations ranged from 0.05 M to 0.05 M,
while the oxygen pressure was 3.44 x 105 pa. The temperature was 25°C.
They concluded that the oxidation rate is proportional to three-halves
power in sulfite and independent of the oxygen concentration:
r = ke (S03=)3/2 (8)
with ke = 0.065 (1) /(gmole) (sec). Barren and O'Hern used deionized
water, but had copper impurities present on the order of 10~7 M.
Yagi and Inoue studied sulfite oxidation using a polarographic method.
They used distilled water, but did not de-ionize it. Sulfite concentrations
ranged from 0.001 to 0.08 M. Their rate equation at 20°C is
r = 1.4 (SO3~) • (02) gmol/lit.sec., (9)
giving a first-order dependence on oxygen concentration.
47
Srivastava, McMillan and Harris also found that the reaction rate is
first order with respect to the oxygen concentration. Their sulfite
concentration was above 0.04 M, while the oxygen concentration in solution
ranged from 0.0006 to 0.0024 M. Temperature was set at 20°C. They used
Barron and O1Hern's4 technique. However, the authors themselves admit that
they had copper impurities on the order of 3 x 10~6 M, a concentration which
would have a marked effect on the reaction rate.
Winkelmann found the reaction rate independent of oxygen concentration.
Obviously, there are uncertainties regarding the order of oxygen for
the uncatalyzed reaction rate. As far as the actual kinetic results are
concerned, Fuller and Grist's16 findings seem to be more reliable than
others due to high purity of their reactants; however, the controversy of
kinetic expressions requires a check of their results.
44
-------
Experimental Apparatus and Procedure
In a work closely associated with ours. Ness tried to check the results
of Fuller and Crist16, primarily having in mind the fact that they did not
measure the activation energy of the uncatalyzed reaction and therefore could
have been mass-transfer controlled in their experiments. However, Ness'
results indicated that the oxygen transfer to the sulfite solution was the
controlling factor in his experiments. The sulfite oxidation rate was
constant with time and the energy of activation was virtually zero.
Our apparatus was similar to that used by Ness; however, certain
modifications were made to insure operation which would yield purely kinetic
data instead of being controlled by mass transfer of oxygen to the solution.
The apparatus resembled the one used in our work with the catalyzed SC>2
oxidation at low pH, except that now the reactor was a three-neck, 1-liter
quartz flask (see Fig. 17).
Ness used an initial solution volume of 500 ml. and continuously took
samples through the sample tube to follow the change in sulfite concentration
with time. His stirring speed was around 500 r.p.m.
In our experiments, we used a reacting volume of 50 ml., and a stirrer
speed of over 1500 r.p.m., which brought about a large change in the surface
area for oxygen absorption; as a result, mass transfer rate was increased
manyfold over the one obtained by Ness.
The reaction rate was so rapid that it was impossible to continuously
take samples without introducing a large error. Also, the volume of the
solution in the reactor was so small that sample taking would have been
difficult. Therefore, the reaction was quenched with iodine directly in
the reactor after desired period of time. One run in the reactor gave only
one data point. For different reaction times the procedure had to be
repeated.
For the above reason, the neck used for the sample tube and valve in
Ness* procedure (and in our low pH work) was now fitted with a 25-ml.
teflon-stoppered burette with a pressure equalizing side-arm, which was used
to supply the necessary iodine for quenching. The other burette, just as in
Ness' work, was used to supply the sulfite necessary for the reaction.
Before each run, the reactor and attachments were cleaned with a con-
centrated potassium dichromate (^804 + ^Mr^Oy) cleaning solution and
rinsed many times with deionized water.
The reactor was filled with 40 ml. of deionized water to which an
appropriate number of drops of HaOH were added in order to raise the pH
above 10, the stirring bar was placed in it, and the fittings were attached
to the flask necks. A Tygon tube leading from the oxygen supply was
attached to the oxygen inlet. The joints were sealed with teflon sleeves.
The assembly was placed in the constant temperature bath; all experiments
45
-------
Sulfite
Iodine
Oxygen
Fig. 17. Reactor for high pH uncatalyzed oxidation of
sulfur dioxide.
46
-------
were conducted at 25°C. Oxygen was blown through the reactor for several
minutes to purge the air. The pressure was then set at 800 mm mercury, the
magnetic stirrer was set on, and at least two hours were allowed for oxygen
absorption.
The reaction started when 10 ml. of sodium sulfite solution were added
to the volume in the flask through one of the burettes. The initial sulfite
concentration in the reactor was close to 0.01 M. This was attained by
weighing out 0.0650 g. of Na2S<>3 and dissolving it in 10 ml. of deionized
water. Na2SO3 and NaOH were used in the experiments since it was shown by
several workers that sodium does not catalyze SO2 oxidation5'25'28.
To start the reaction, the oxygen supply was temporarily shut off, the
stopcock on the pressure equalizing arm of the burette was opened, followed
by the burette stopcock, thus adding the sulfite solution to the 40 ml. of
solution already in the reactor. Both stopcocks were then shut and the
oxygen pressure again adjusted to 800 mm Hg. Stirrer speed was increased
to 1500 r.p.m. before the start of the reaction.
After the desired reacting time had elapsed, the reaction was quenched
by introducing 15 ml. of 0.1 N iodine through the other burette. At the
moment that quenching began, stirring speed was increased to a maximum of
1850 r.p.m., causing extreme violent stirring and splattering, so that
iodine could quench the reaction even in drops on walls of the reaction flask.
Iodine remaining on burette walls was washed down with 5 ml. of deionized
water.
After the addition of iodine, 10 ml. of the resulting solution in the
flask was withdrawn with a pipette and backtitration with 0.1 N arsenious
acid was performed to determine the sulfite concentration at the time the
reaction was quenched. To obtain the sulfite concentration at some other
time, the apparatus was cleaned carefully in the manner already described
and the procedure repeated. A sample calculation of the sulfite concentration
is shown in Appendix B.
To determine the pH of the reacting solution, a separate series of runs
was performed. The burette for iodine addition was replaced by the sample-
taking tube and valve. At the desired time, a smaple was withdrawn and its
pH measured as rapidly as possible.
Before the actual kinetic results for the uncatalyzed reaction are
presented, the methods used in preparing the various solutions and the care
taken in insuring their greatest possible purity under the given conditions
should be described.
To obtain water that contained no traces of metal ions, regular
deionized water from Roger Adams Laboratory supply was passed at rates of
under 1.0 lit./hr through a quartz ion exchange column which held 0.45 kg of
a cation-anion mixed bed exchange resin (Ag 501 - x 80), obtained from
47
-------
BioRad, Inc. This column was capable of producing water whose specific
resistance was in excess of 1.5 million ohm-cm. At the end of all our
experiments, we found the specific resistance to be around 800,000 ohm-cm,
which indicates excellent purity of the water used.
The pH of the deionized water was raised by adding drops of 1 N NaOH
to it. IN NaOH was prepared from sodium hydroxide electrolytic pellets
obtained from Fisher Scientific Company. Their certificate of actual lot
analysis shows that there is a maximum 0.0002% of both iron and copper in
the pellets. The calculation of iron and copper impurities in our reacting
solution is done in Appendix C; there will be more comments about impurities
later on in the text.
To prepare a reacting solution of pH 11.1, 0.84 ml. of 1 N NaOH was
added to 300 ml. of deionized water. Forty ml. of this solution were placed
in the quartz flask prior to the start of the reaction. Upon addition of
Na2SO3, the pH was 11.1.
In order to keep the 1 N NaOH solution free of metal ions, it was
placed in a special quartz container.
The Na2SO3 used was anhydrous sodium sulfite made by Fisher Scientific
Company. The certificate of actual lot analysis shows a maximum of 0.0003%
ion present. Other impurities are either present in much smaller quantities
or are not significant from the standpoint of sulfite oxidation catalysis.
Extreme care was taken during the experimental procedure to keep the
chemicals as pure as possible. Weighing of chemicals, done on a Mettler
or an Ainsworth balance with +_ 0.0001 gm sensitivity, was performed in
clean pyrex flasks with the aid of a teflon spatula. A31 glassware was
kept extremely clean at all times with the aid of the cleaning solution and
constant rinsing with deionized water. Preparation of Na2SO3 solution was
done immediately prior to the start of the reaction, so that the solution
would not be in contact with either the pyrex preparation flask or the
burette for more than a few minutes.
As can be seen in Appendix C, the maximum amount of impurities from
the chemicals used gives concentrations of around 3 x 10~^ M copper and
7 x 10~8 M iron. Since it appears that copper is a much better catalyst
than iron, it seems that we achieved a satisfactory level of purity for
the uncatalyzed reaction.
Results
In order to determine how rapidly oxygen could be transferred to the
solution, a series of runs with cobalt catalyst was performed. Cobaltous
sulfate is known to be an extremely powerful catalyst for sulfite oxidation
and it is obvious that for a solution containing 10~4 M CoS04, the rate of
reaction would be so rapid that the process had to be mass-transfer con-
trolled in our system. Our 50-ml. reacting solution contained 1.2 x 10~4 M
48
-------
CoSC>4 and runs were made at an initial pH of 10.5. The rate of mass
transfer of oxygen is shown in Fig. 18. When the solution volume was
increased to 500 ml., the stirring speed decreased to 500 r.p.m., and the
method of continuous sampling used, a rate of oxygen transfer very similar
to Ness'35 was obtained. The improvement in mass transfer in our system is
significant (our scale is in second. Ness1 is in minutes). The rate of
reaction measured by Fuller and Crist1** is compared with our rate of oxygen
transfer to the solution (Fig. 18), and it is clear that, if the actual rate
is not too much faster than that of Fuller and Crist, we would not be mass-
transfer controlled and would be capable of obtaining pure kinetic data in
our system.
It must be mentioned that Fuller and Grist's experiments were done at a
pH of 8.2 - 8.8, while we performed two series of experiments - one at a
constant pH of 11.1 (the pH did not change noticeably during the 5 minutes
in which the reaction was followed) and one at a variable pH (from 10.25
at t = 0 to 8.7 at t = 5 min.). The initial sulfite concentration was
0.0093 M in all runs.
There was a scatter in our data due to the method used. All of the
recorded kinetic data for the uncatalyzed reaction at a constant pH of 11.1
are shown in Fig. 19. Table 6 shows the exact concentrations measured at
various times from the start of the reaction. When the average values for
each reaction time are calculated. Fig. 20 is obtained. The standard
deviation for each data point is also shown. Our results can be compared
with the curve that would result from Fuller and Grist's reaction rate
constant. Our data fit the first-order kinetics in sulfite quite well,
with the rate constant being k^ = 0.0053 +_ 0.0005 sec"1. This represents
an error of about 10%.
The concentrations measured in the runs for which the pH varied from
10.25 at t = 0 to 8.7 at t = 5 min. are shown in Fig. 22. Table 7 summarizes
these values.
The results show that the rate was independent of the pH. This is not
surprising since sulfite is the only SO2 species existing in solutions that
are as basic as ours.
The deficiencies of our method could be described as follows: there
are variations in the exact duration of the reaction due to the method of
adding sulfite and the splattering of the reacting solution on the walls of
the vessel (although the walls are extremely smooth and there was constant
and rapid flow back to the bulk solution); as a result, certain portions of
the liquid started reacting before others and some portions were not
quenched at exactly the same time as others. Furthermore, impurities often
interferred with our experiments. If extreme care was not taken, the results
would indicate zero sulfite concentration within 2 or 3 minutes. This
happened on quite a few occasions and extreme caution had to be exercised
in cleaning and preparing the equipment after the incidents. The whole
49
-------
ro
O
W
0)
C
-S
0)
r-l
3
U)
Rate of reaction
easured by
uller & Crist
120
Time, seconds
180
240
Fig. 18. Rate of mass transfer of oxygen to solution containing
1.2 x 10-4 M CoS04 catalyst.
50
-------
10
n
H
P
w
o
o
o
o
o
o
o
o
0
8
8
0
o
o
60
120 180
Time, seconds
240
300
Fig. 19. Complete kinetic data for the uncatalyzed sulfite oxidation
at a constant pH of 11.1.
51
-------
TABLE 6. KINETIC RESULTS FOR THE UNCATALYZED SULFITE OXIDATION AT A CONSTANT
pH OF 11.1 (WITH AVERAGE VALUES AND STANDARD DEVIATIONS).
t = 0 0.00884
0.00887
0.0923
0.00926
0.00959
0.00908
0.00954
0.00947
0.00922
0.00928
0.00925
Cs03=Ht_0 = 0.00924+0.00024 M
t = 60 sec. 0.00714
0.00589
0.00630
0.0062
0.00613
0.00694
0.00686
0.00655
0.00656
0.00640
0.00631
0.00620
0.00620
Cso3 D, M
t = 30 sec. 0.00721
0.00735
0.00758
0.00777
0.00735
0.00714
0.00700
0.00742
LSO,=T o« = 0.00735+0.00025 M
3 t=30 —
t = 90 sec. 0.00506
0.00613
0.00504
0.00500
0.00463
0.00463
0.00428
0.00426
0.00645
0.00506
Ls03=Jt_go = 0.00506+0.0073 M
80 ~
0.00640+0.0036 M
52
-------
TABLE 6 (continued)
1
t = 120 sec. 0.00340
0.00366
0.00623
0.00492
0.00479
0.00467
0.00485
0.00708
0.00652
0.00603
0.00595
0.00408
0.00436
0.00588
0.00546
0.00486
0.00428
0.00463
0.00470
0.00422
0.00463
0.00366
ES03=\-120 = 0.00498+0.00102 M
t = 240 sec. 0.00316
0.00264
0.00233
0.00268
0.00282
0.00274
[sop, M
t = 150 sec. 0.00436
0.00367
0.00413
0.00409
0.00429
0.00340
0.00425
0.00436
Lso3~Jt_150 = 0.00404+0.00035 M
t = 180 sec. 0.00232
0.00276
0.00289
0.00324
0.00338
0.00262
0.00317
0.00373
ES03=Jt_180 = 0.00300+0.00045 M
t = 120 sec. 0.00221
0.00189
t = 300 sec. 0.00152
0.00190
0.00199
0.00205
0.00209
CSO3=Jt 3QO = 0.00191+0.00023 M
LSO =J. = 0.00273+0.00027 M
3 t=240 —
53
-------
60
120
180
240
300
Time, seconds
Fig. 20.
Average kinetic data and standard deviations for the
uncatalyzed reaction at a constant pH of 11.1.
54
-------
k = 0.0053 + 0.0005 sec
(k = 0.013 sec >
0
Fig. 21.
120 180
Time, seconds
240
300
Natural logarithm of sulfite concentration as a function
of time for the uncatalyzed reaction at a constant pH
of 11.1.
55
-------
10
I I
4J
•H
to
0)
H
i
o
•H
•P
(0
c
(U
o
o
u
0)
•U
•H
O
8
o
8
o
8
8
o
8
o
o
8
60
1 1 1__
120 180
Time, seconds
240
300
Fig. 22. Complete kinetic data for the uncatalyzed sulfite oxidation
at variable pH.
56
-------
TABLE 7. KINETIC RESULTS FOR THE UNCATALYZED SULFITE OXIDATION AT VARIABLE pH
t = 0
(pH=10.25)
t = 60 sec.
t = 1?0 sec.
t = 240 sec.
0.00927
0.00922
0.00952
0.00940
0.00918
0.00652
0.00630
0.00666
0.00644
0.00479
0.00470
0.00472
0.00463
0.00436
0.00433
0.00253
0.00247
0.00267
Cso3 D, M
t = 30 sec. 0.00723
0.00758
0.00707
0.00700
t = 90 sec. 0.00515
0.00518
0.00498
0.00493
0.00477
t = 180 sec. 0.00373
0.00379
0.00324
0.00316
0.00282
t = 300 sec. 0.00191
(pH = 8.7)
0.00198
57
-------
procedure is very time consuming and that is the primary reason that only
experiments with the initial SO3= concentration of 0.0043 M were performed
and that an attempt at finding the activation energy by running the
reaction at some lower temperature was not made. A rapid-mixing flow
system would probably be more suitable for this rapid reaction.
We have already seen that the oxygen order of the uncatalyzed reaction
is practically not known. It if were zero, we would have, from our results:
d(SO =)
r = zi = k! <10>
at l 3
with kx = 0.0053 sec"1
However, if it were one, we would obtain
r = k2 (S03=) (02) (11)
with k = 4.08 (gmole/lit)^sec"1
(This was calculated assuming an oxygen solubility of 0.0013 M at
P = 1 atm.31)
58
-------
V. UNCATALYZED STUDIES
A comparison of the kinetic results clearly indicates that the low pH
Fe(II) and Mn(ZI) catalyzed reactions proceed by significantly different
reaction mechanisms. For the Mn(II) catalyzed reaction no induction period
was observed with the initial reaction rate being independent of pH
(0.75-4.0), zero order in oxygen concentration (0.00145-0.01 M), zero order
in S(IV) concentration (0.002-0.073 M), and second order in catalyst con-
centration (1.33 x 10~5 -2.98 x 10~4 M). In contrast, induction periods of
up to 100 minutes were observed for the Fe(II) catalyzed reaction. Follow-
ing the induction period the steady-state reaction rate was dependent on the
inverse hydrogen ion concentration (pH 0.89-2.2), zero order in oxygen
concentration for [02] < 0.01 M, and first order in oxygen for [02] £. 0.01 M,
approximately first order in Fe(II) concentration (6 x 10~5 - 1 x 10~2 M) ,
and dependent upon the 0.60 power of the bisulfite ion concentration. Both
the Mn(II) and Fe(II) catalyzed reactions exhibited negative salt effects
(ZAZB = -3.75 and -1.26) and positive activation volumes (+7.2 and +9.9 cc/mole)
with apparent activation energies of 19.8 and 31.2 kcal/mole, respectively.
Since the oxidation of aqueous S(IV) has deleterious effects on many
flue gas desulfurization systems, there is a need for techniques to effectively
inhibit the oxidation reaction. Therefore, the effects of chelating agents
(EDTA and 1,10-phenanthroline) and phenolic antioxidants (phenol, phloroglucinol,
resorcinol, hydroquinone, pyrogallol, and pyrocatechol) were investigated.
The results of the chelation studies led to the somewhat astounding conclusion
that the widely studied, previously-accepted "uncatalyzed" reactions were,
in fact, primarily trace metal catalyzed. These results indicate that it is
only the catalyzed reaction which is of importance to industrial wet scrubbers.
While both the low pH Mn(II) and Fe(II) catalyzed reactions are inhibited
in the presence of phenolic antioxidants the effects are substantially
greater on the Mn(II) catalyzed reaction. However, the activity of the Fe(Il)
catalyst is markedly more inhibited by chelation with EDTA. The extreme
susceptibility of the Mn(II) catalyzed reaction to phenolic antioxidants
(present in trace quantities in flue gases) has, in fact, severely limited
its usefulness as a catalyst in systems requiring complete oxidation to gypsum.
However, at the level of phenolic impurities present, the Fe(II) catalyzed
reaction would be relatively unaffected.
The high pH Cu(II) catalyzed reaction was markedly inhibited in the
presence of chelating agents (ETDA) of phenolic antioxidants. The most
effective inhibitors were those compounds capable of complexing the catalyst
in addition to terminating free-radical chains. The tremendous inhibiting
59
-------
effect of pyrocatechol can be attributed to this dual chelation/antioxidant
action. A UV-visible spectrophotometric study of the antioxidant inhibited
reactions provided further insight into the inhibitor effects.
60
-------
VI. APPLICATION OF KINETIC DATA ON SULFUR DIOXIDE OXIDATION TO SCRUBBERS
In industrial SO2 scrubbers, the pH of the scrubbing solution can vary
from 4 to 10, depending on the particular SO2 absorption method used. We
have already described the negative effects of the sulfur dioxide oxidation
in Section I. It is interesting to see how some of the kinetic data
obtained in our experiments, as well as those reported by other researchers
can be applied to scrubbers in order to evaluate SO2 oxidation effects.
We shall restrict ourselves to cases where the scrubbing solution
contains large amounts of bisulfite or sulfite ions (a good example is the
concentrated mode of the double alkali process where the concentration of
bisulfite and/or sulfite is greater than 0.5 M). To study the kinetic
effects in a flow system of this kind, the theory of gas absorption with
simultaneous chemical reaction has to be applied. The book Gas-Liquid
Reactions by P. v. Danckwerts11 represents the most complete work in this
area, and the criteria which will be used in our analysis of SO2 oxidation
can be found in Chapter 5 of this book, which is entitled "Absorption into
Agitated Liquids".
We shall assume that the scrubber is a packed tower in which SO2 and O,
are absorbed by a solution of sodium bisulfite or sodium sulfite (however,
the analysis which follows could be applied to any kind of absorption
tower). Since SO2 species are present in the liquid in large excess, oxygen
is the limiting reagent, and it is of interest to see how fast it can be
absorbed into the solution for the oxidation to take place. The film model
for absorption will be used, with the gas-side resistance to mass transfer
considered negligible (this is a reasonably good assumption for oxygen
absorption since the oxygen solubility in water is rather low. The overall
resistance to mass transfer consists of the gas-side resistance and the
liquid-phase resistance : I/KG = lkG + ^e/\, where H& = P /c . H for
oxygen is of the order of 106 atm cm3/mole and the term H /£ il the dominant
resistance term since kQ and kL usually do not differ by mroe than two
or three orders of magnitude)II. To see how strong the effect of SO,
liquid-phase oxidation on the process is, we shall calculate the absorption
enhancement factor, which is the ratio of the absorption rate into a
reacting liquid to the absorption rate if there were no reaction (physical
absorption rate). The enchancement factor is, therefore, the factor by which
the rate of absorption is increased by the reaction.
For an irreversible reaction
A(gas) + B(liquid) >- products (12)
which is m-th order in A and n-th order in B, and whose rate is therefore
r = k (A)m(B)n
mn
(13)
61
-------
the following dimensionless quantity was found to be of extreme importance
in determining the enhancement factor and the absorption rate
-2— k DJA*)"1'1 (B")" | (14)
» + l mn A
where k = mass-transfer coefficient of the liquid film, cm/sec
L
m = order of reaction with respect to reactant A
k = reaction rate constant, whose units depend on orders m and n
mn
2
D = diffusion coefficient of gas A in solution, cm /sec
A
A* = concentration of dissolved gas corresponding to equilibrium
with the partial pressure of the gas at the interface between
gas and liquid, gmole/cm3
B° = bulk concentration of liquid reactant B, gmole/cm
n = order of reaction with respect to reactant B
In the case of SO2 liquid-phase oxidation, oxygen is the gas A, while
bisulfite or sulfite (depending on the pH considered) is the reactant B.
Let us first consider the case where the pH of the scrubbing solution is
around 4. Bisulfite is the species present in solution, and the data
obtained in our experiments can be used.
We shall assume that the temperature in the scrubber is 25°C because
that is the only temperature for which we have kinetic data. We shall also
assume that manganese is present in solution at 3 r.p.m. The partial
pressure of oxygen in the gas phase is taken as 0.04 atm. (this value taken
from Johnstone and Singh26 who give typical composition of flue gases from
which SO2 is scrubbed). The order of reaction with respect to both
reactants is taken as zero (m = 0, n = 0). Sodium bisulfite concentration
in the scrubbing liquid is 0.5 M. kL is takenas 10~2 cm/sec (a typical
value for packed towers taken from Danckwerts , p. 214).
Under these conditions, we have the following:
83 42
A* = (o ) = 2.44 x 10 gmole/cm (obtained from Reith )
—5 2 42
D =D =1.85x10 cm /sec (Reith )
A °2
— T 3
B° = (NaHSO ) = 0.5 x 10~ gmole/cm
With no sulfate present in the scrubber.
62
-------
^mn = ko = 10'6 x 10~ moles/lit.hr. = 2.94 x 10~10 moles/cm3sec
(from our experimental data).
The dimensionless quantity M then becomes
M =
1
-1
0.067 (15)
According to Danckwerts1 theory of gas absorption with chemical reaction,
when M 1, as in this case, the enhancement factor is close to one, which means
that the bisulfite oxidation is relatively slow and does not enhance the
absorption of oxygen significantly. In scrubbers whose operating conditions
are close to the ones described here, the S(>2 liquid-phase oxidation does not
appear to represent a problem.
However, most scrubbers operate at higher pH values. As discussed
before, the kinetic data for high pH catalyzed SO2 oxidation are very scarce,
especially for good catalysts like manganese and iron, which are most likely
to be present in industrial scrubbers. In the following illustrative
calculation, we shall use Keith's42 data for CoSO4 catalyst. The catalyst
concentration in the scrubbing solution is taken as 7 x 10~5 M (around 11
p.p.m.). If the relative catalyst efficiency at high pH is anywhere close
to that at low pH, then, according to Junge and Ryan28, 11 p.p.m. 00804
would be roughly as efficient as 2 p.p.m. MnSO4- This, however, is only
hypothetical; actual kinetic data at high pH are definitely needed.
For the scrubber, we shall assume the following:
pH = 8
temperature = 50°C (temperature typical of many SO scrubbing
systems, according to Schmidt44)
P
o_ = 0.04 atm.
A* = (02) = 2.2 x 10~8 g mole/cm3 (Reith42)
DQ = 3.26 x 10~5 cm2/sec (Reith42)
k =10 cm.sec (Danckwerts )
B° = (Na2S03) =0.8M=0.8x 10~3 gmole/cm3
42
Under these conditions, Reith has found that the reaction is second order
in oxygen and zero order in sulfite. For the catalyst concentration of
63
-------
5 93
7 x 10 M, he obtained k = k = 4.9 x 10 cm /gmole sec.
mn 2
The dimensionless quantity M becomes
= 23.62 (16)
- - r* [f **
/M~ =4.86
For /M>3, the reaction is rapid and most of the dissolved oxygen reacts in the
liquid film. The enhancement factor is equal to «/M , which means that the
absorption rate is enhanced almost five times compared to the case of no
reaction.
We can conclude that even at p.p.m. concentration, the catalyst effect
on SO2 oxidation at high pH values in scrubbers is very significant. The
oxidation to sulfate is rapid and could amount to a significant percent of
the total SO, absorbed. Obviously, scrubbing at lower pH values is more
desirable from the standpoint of SO2 liquid-phase oxidation, but the S02
absorption effectiveness is decreased in more acidic solutions.
In order for the scrubber calculations at high pH values to have more
meaning, kinetic experiments of sulfite oxidation with manganese and other
catalysts should be performed. It would also be interesting to see what the
effect of the sulfate ions on high pH SO2 oxidation would be. If a retarding
effect, similar to the one observed in our experiments at low pH, should
result, addition of sulfate and its recirculation with the scrubbing solution
would prevent SO2 oxidation to a certain extent.
64
-------
VII. FURTHER STUDIES
Our experimental results and their application to scrubber calculations
have only touched upon some of the many questions in the area of SC>2 liquid-
phase oxidation. As mentioned in the introduction, our work is only an
initial part of an overall kinetic and mechanistic study of the oxidation
of sulfur dioxide inall pH ranges.
Most of our experimental work involved the low pH catalyzed oxidation of
SC>2. We observed certain characteristics of the reaction which were not
reported before. However, a great deal of work still remains to be done in
this area. The knowledge of the detailed reaction mechanism should lead to
better understanding of the SO2 oxidation kinetics in general and should
facilitate the search for ways of reducing the adverse effects of this
reaction. We only touched upon some of the possible steps that could be
involved in the overall mechanism. Complexing of the manganese catalyst with
sulfate and possibly bisulfite might be involved, as well as the formation
of Mn species in the +3 oxidation state. Further work should examine the
effects of ionic strength (by changing the ionic strength more drastically
than we did, an idea of the species involved in the rate determining step of
the reaction could be formed, i.e., whether the species are positive ions,
negative ions, or molecules). Determination of the order of reaction with
respect to oxygen could bring important clues in investigating the details
of the reaction mechanism (in order to vary the oxygen concentration by several
orders of magnitude, high pressure kinetic equipment should be used).
Spectroscopic studies of possible complexing between various manganese species
and sulfate could shed more light on some of the kinetic behavior observed in
our experiments. Once a better understanding of the mechanism is achieved,
the reasons behind the dimensionless kinetic diagram presented in our work
might be found. Since all of our work regarding the various factors
affecting the SC>2 catalyzed oxidation rate was done with manganese catalyst,
further work should examine other catalysts as well.
In the high pH region, we investigated only the uncatalyzed SO2 oxidation.
The oxygen order and the activation energy of this reaction still remain
unknown. The catalyzed reaction, which is of much greater interest, has not
been a subject of careful investigation, except for the cobalt-catalyzed
case. As a matter of fact, there are almost no data on catalysts which are
significant in industrial SC>2 scrubbers and polluted atmospheres. Neither
the calculation of oxidation effects in scrubbers nor the modeling of SO2
oxidation in atmospheric drops to predict sulfate levels in polluted air can
be complete without detailed data for the high pH catalyzed reaction. Since
65
-------
manganese was shown to be the most effective of all catalysts studied at low
pH, it seems logical that the first step in further work in the high pH area
should include kinetic studies of the manganese-catalyzed reaction. A flow
system or an absorption technique would have to be used since the reaction is
extremely rapid. Studies of catalysts like iron, copper, magnesium and
calcium might also be necessary. With regard to S(>2 oxidation in drops, it
would be germane to confirm whether ammonia actually acts as a catalyst or
just maintains a pH high enough for sulfite oxidation to take place. It
would also be interesting to see whether the addition of sulfate inhibits the
high pH oxidation as it did the low pH reaction. As in the low pH case, the
prediction or knowledge of the reaction mechanism, which would surely differ
from the one at low pH, would be of great practical importance (for example,
for the effective inhibition of undesired oxidation effects in scrubbers).
When the overall kinetic study of SO2 catalyzed reaction in both the low
and the high pH regions is complete, optimal operating conditions for
scrubbers and various ways of eliminating the adverse effects of SO2 oxida-
tion in both the scrubbers and the drops should be within reach of realization.
66
-------
REFERENCES
1. Amdur, M. O., J. Air Pollut. Control Ass., 19,. 638 (1969).
2. Astarita, G., Marucci, G. , and Coleti, L., Chim. e L'Ind., 46, 1021 (1964)
3. BSckstrom, J. Amer. Chem. Soc., 49, 1460 (1927).
4. Barron, C. H. and O'Hern, H. A., Chem. Eng. Sci., 21, 397 (1966).
5. Bassett, H., and Parker, W. G., J. Chem. Soc., 1540 (1951).
6. Beilke, S., and Georgii, H. W., Tellus, 20, 435 (1968).
7. Chen, T. I., and Barron, C. H., Ind. Eng. Chem. Fund., 11, 466 (1972).
8. Cheng, R. T. , Corn, M. , and Frohliger, J. O., Atmospheric Environment,
5_, 987 (1971).
9. Coughanowr, D. R., Ph. D. Thesis, University of Illinois, 1956.
10. Coughanowr, D. R., and Krause, F. E., Ind. Eng. Chem. Fundam. , 4_, 61
(1965).
11. Danckwerts, P. V., Gas-Liquid Reactions, McGraw-Hill, New York, 1970.
12. Davidson, B., J. Air Pollut. Control Ass., 17, 154 (1967).
13. Eriksson, E. , Tellus, 4_, 215 (1952).
14. Foster, P. M., Atmospheric Environment, 3_, 157 (1969).
15. Frydaman, G. N., Torsten, R., and Sillen, L. G., Acta Chem. Scand., 12,
878 (1958).
16. Fuller, B. C. and Crist, R. H., J. Amer. Chem. Soc., 63, 1644 (1941).
17. Gartrell, F. E., Thomas. F. W., and Carpenter, S. B., Am. Ind. Hyg. Ass.
J. , 2£, 113 (1963).
18. Gerhard, E. R., and Johnstone, H. F., Ind. Eng. Chem., 47, 972 (1955).
67
-------
19. Grozdovskii, Chem. Abst.. 2090 (1936).
20. Gunn, D. J., and Saleem, A., Trans. Instn. Chem. Engrs., 48, T46 (1970).
21. Hartridge, H. , and Roughton, F. J. , Proc. R. Soc. , A104, 376 (1923).
22. Hoather, R. C. , and Goodeve, C. F. , Trans. Farad. Soc., 30, 1149 (1934).
23. Houghton, H., J. Met., 12, 355 (1955).
24. Johnstone, H. F. , and Coughanowr, D. R. , Ind. Eng. Chem. , 50, 1169 (1958).
25. Johnstone, H. G. , and Moll, A. J. , Ind. Eng. Chem. , 52, 861 (1960.
26. Johnstone, H. F. , and Singh, A. D. , "The Recovery of Sulfur Dioxide from
Dilute Waste Gases by Chemical Regeneration of the Absorbent," University
of Illinois Bulletin, Vol. XXXVIII, No. 19 (1940).
27. Junge, C. E., Tellus, 8_, 127 (1956).
28. Junge, C. E. , and Ryan, T. G. , Quart. J. Roy. Meteorol. Soc., 84, 46
(1958).
29. Kashtanov andRuizthov, Chem. Abst., 3596 (1956).
30. Laidler, K. J. , Chemical Kinetics, Second Edition, McGraw-Hill, New York,
1965.
31. Linek, V. and Mayrhoferova, J. , Chem. Eng. Sci. , 25, 787 (1970).
32. Matteson, M. J. , StBber, W., and Luther, H. , Ind. Eng. Chem. Fund. , 8_,
677 (1969).
33. McKay, H. A., Atmospheric Environment, 5^, 7 (1971).
34. Miller, J. M. and DePena, R. G. , J. Geophys. Res., 77, 5905 (1972).
35. Ness, A. I., M. S. Thesis, University of Illinois, 1974.
36. Oddie, B. C. , Quart. J. Roy. Meteorol. Soc., 31, 525 (1962).
37. Perry, J. H. , Chemical Engineers' Handbook, Fourth Edition, McGraw-Hill,
New York, 1963.
38. Petrenchuk, O. P., and Drozdova, V. M. , Tellus, 18, 280 (1966).
39. Powell, R. H. , M. S. Thesis, University of Illinois, 1973.
40. Pritchett, P. W. , Ph. D. Thesis, University of Delaware, 1961.
68
-------
REFERENCES (continued)
41. Reinders, W., and Vies, S. T. , Rec. Trav. Chim.. 44_, 244 (1925).
42. Reith, T., "Physical Aspects of Bubble Dispersions in Liquids," Thesis,
Delft Technical University, Delftsche Uitgevers Maatschappij N.V. , 1968.
43. Reith, T., and Beek, W. J., Chem. Eng. Sci. , 28, 1331 (1973).
44. Scmidt, M., World Health Organization Project No. U. S. 3100, 1970.
45. Scott, W. D., and Hobbs, P. V., J. Atmos. Sci., 24, 54 (1967).
46. Skoog, A. D., and West, D. M., Fundamentals of Analytical Chemistry,
Second Edition, Chapter 20, Holt, Rinehart and Winston, New York, 1969.
47. Srivastava, R. D., McMillan, A. F., and Harris, I. J., Canad. J. Chem.
Eng., 46, 181 (1968).
48. Tartar, H. V., and Garretson, H. H., J. Amer. Chem. Soc., 48_, 1543 (1926)
49. Titoff, A., Z. Phys. Chem., 45, 641 (1903).
50. Van Den Heuvel, A. P., and Mason, B. J., Quart. J. Roy. Meteorol. Soc.,
89, 271 (1963).
51. Vasilev, S. S,, Kastanov, L. I., and Kastorkaja, T. L., Acta Physicochim.
URSS, 3_, 413 (1935).
52. Walter, P. E., M. S. Thesis, University of Illinois, 1972.
53. Wang, J. C. and Himmelblau, D. M., J. Amer. Inst. Chem. Eng., 10, 574
(1964). ~ ~
54. Westerterp, K. R., Van Dierendonk, L. L., and De Kraa, J. A., Chem. Eng.
Sci., 18_, 157 (1963).
55. Winkelmann, V. D., 2. Elektrochem.. 59, 891 (1955).
56. Yagi. S., and Inoue, H., Chem. Eng. Sci., 17, 411 (1962).
57. Yui, T. , Tokyo Inst. Phys. Chem. Res. Bull.. 19_, 1229 (1940).
69
-------
APPENDIX A
Sample Calculation of Total Sulfur Dioxide^oncentration
Iodine concentration: 0.01103 N
- Arsenious acid concentration: 0.1012 N
- 5-ml. sample of reacting solution placed in flask containing 10 ml. of
iodine and some sodium bicarbonate buffer
- 0.570 ml. of arsenious acid used up in titration
(S02) = (10X0.01103) - (0.570) (0.1012) = ^^ R
= 0.01092 M
70
-------
APPENDIX B
Sample Calculation of Sulfite Concentration
- Iodine concentration: 0.0966 N
- Arsenious acid concentration: 0.0996 N
- 0.0650 g of Na SO dissolved in 10 ml. of HO and added at t = 0 through
burette to the oxygenated volume of 40 ml. in the reactor
- After 90 sec., reaction quenched with 15 ml. of iodine; 5 ml. of HO used
for washing iodine remaining in burette
- 10-ml. sample of quenched solution used for titration
- 1.36 ml. of arsenious acid used up in titration
40 ml. HO + NaOH
+ 0.065 g Ha SO in 10 ml HO
Reacting volume = 50 ml.
+ 15 ml. iodine
+ 5 ml. H20
Quenched volume = 70 ml.
Iodine concentration in quenched volume = -— ' -
= 0.02077 N
Sulfite concentration in quenched volume
(0.02077)(10) - (1.36)(0.0996)
51 •-- • ^^^^— I • —••- — !• •• i ^-^^—i^—^•^^••^—^»—.
10
71
-------
APPENDIX B (continued)
Sulfite concentration in reacting volume just before quenching
_ 2£ (0.02077X10) - (1. 36) (0.0996) = Q 01012 N
50 10
= 0.00506 M
72
-------
APPENDIX C
Estimated Maximum Concentration of Important Metal Ions Stemming from Chemicals
Used
From Na SO
0.0003% =
1 g
0.0650 g Na^SO, , , -6
,„ . ^23 3 x 10 g Fe
(Fe) = i .. ——..•— •• «
max 0.05 lit. reacting solution 1 g Na SO
1 mole Fe _ , _-8 „
x -=^ = 7 x 10 M
56 g Fe
From NaOH
2 x 10~6 q
0.0002% iron and copper =
1 N NaOH = 40 g NaOH/lit.
To prepare the initial solution of 40 ml., 0.842 ml. of 1 N NaOH was
added to 300 ml. of deionized water.
0.842 ml. x — r- *= 0.03368 g
1000 ml. *
. 0.1123 g HaOH/lit.
0.04 lit. x ' = 4-49 x 1°~3 9 NaOH in initial
40 ml. solution
In reacting 50 ml. solution:
4.49 x 10"3 g NaOH = g NaOH _
0.05 lit. ' lit. reacting solution
73
-------
APPENDIX C (continued)
Total
(Cu) = 0.0898 g NaOH x 2 x 10"6 9 ^" x * m°le
^
.
max ^ g NaOH 63,5 g Cu
(Cu) = 2.83 x 10~9 M
max
(Pe) = 0.0898 x 2 x 10~6 x ^ = 3.21 x 10~9 M
max 56
(Pe) = 7.32 x 10~8 M
max
(Cu) = 2.83 x 10~9 M
max
74
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1 REPORT NO.
EPA-600/7-79-030
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Kinetics of Sulfur Dioxide Oxidation in
Aqueous Solution
5. REPORT DATE
January 1979
5. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
J.L.Hudson, J.Erwin, and N.M.Catipovic
. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
The University of Illinois
Urbana, Illinois 61801
1O. PROGRAM ELEMENT NO.
E HE 62 4
11. CONTRACT/GRANT NO.
Grant R800303
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
Final: 4/73 - 12/74
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES IERL-RTP project officer is Norman Kaplan, MD-61, 919/541-
2556.
. ABSTRACT Tne r6port gives results of a study of the rate of oxidation (low pH cata-
lyzed oxidation and high pH uncatalyzed oxidation) of SO2 in a 1 liter semi-batch
reactor. Low pH experiment results included: (1) both MgSO4- and MnSO4-catalyzed
liquid-phase oxidation of SO2 at low pH are zero order in SO2 concentration; (2)
while Mn is a very effective catalyst even at concentrations as low as 3 ppm, Mg
shows little catalytic action even at concentrations as high as 20,000 ppm; (3) energy
of activation for the Mn-catalyzed reaction was 18.7 kcal/mole; (4) low pH oxidation
of SO2, catalyzed by MnSO4, is independent of pH between 1 and 4, while at pH values
above 4, the reaction speeds up due to the presence of sulfite ions; (5) ionic strength
did not appear to affect the reaction rate significantly (additional tests at higher ionic
strengths should be performed to check this effect); and (6) sulfate ions inhibit the
reaction significantly. The primary result of the high pH study was a first order rate
constant of 0.0053 per second; during the experiments, oxidation was kinetically con-
trolled with a constant oxidation concentration of 0.0013 M. Results applied to condi-
tions in a SO2 scrubber Indicated that the catalyst effect, even at ppm concentration,
is very significant. The effect of catalysts in high pH scrubbing systems subject to
oxidation is much greater than in low pH systems.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Sulfur Dioxide
Oxidation
Kinetics
Catalysis
Manganese
Maernesium
Pollution Control
Stationary Sources
13 B
07B
07C
20K
07D
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report}
Unclassified
21. NO. OF PAGES
81
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
75
------- |