Wet plasma reactor for remidiation of SO2

Article abstract of DOI:10.1063/1.1426692

In pollution control applications, the presence of water in the electrical discharge enhances oxidation of pollutants. The results of an electrical discharge in gas when it flows through a heterogeneous mixture of water and dielectric pellets are reported. The discharge in the wet plasma reactor is more uniform compared to dry dielectric-barrier reactors. The electrical characteristics of such a discharge are discussed. Also the results of removal of SO₂ with the wet reactor are reported. The wet reactor was found to be 5-10 times more energy efficient in removing SO₂ compared to conventional dry plasma reactors.

Full text of DOI:10.1063/1.1426692

Wet plasma reactor for remidiation of SO 2
S. Seethamsetty, S. K. Dhali, and Bakul Dave
Citation: Applied Physics Letters 79, 4298 (2001); doi: 10.1063/1.1426692
View online: http://dx.doi.org/10.1063/1.1426692
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/79/26?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Effect of water on sulfur dioxide ( SO 2 ) and nitrogen oxides ( NO x ) removal from flue gas in a direct current
corona discharge reactor
J. Appl. Phys. 101, 103304 (2007); 10.1063/1.2733762
Diesel Participate Matter Removal Using DBD Pulsed Plasmas
AIP Conf. Proc. 799, 209 (2005); 10.1063/1.2134602
Abatement of perfluorinated compounds using microwave plasmas at atmospheric pressure
J. Appl. Phys. 93, 9483 (2003); 10.1063/1.1574595
Temperature resolved modeling of plasma abatement of perfluorinated compounds
J. Appl. Phys. 89, 2047 (2001); 10.1063/1.1337088
Low-voltage operation of a plasma reactor for exhaust gas treatment by dielectric barrier discharge
Rev. Sci. Instrum. 69, 1847 (1998); 10.1063/1.1148854
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7
On: Wed, 03 Dec 2014 16:39:50

APPLIED PHYSICS LETTERS
VOLUME 79, NUMBER 26
24 DECEMBER 2001
Wet plasma reactor for remidiation of SO2
S. Seethamsetty, S. K. Dhali,a) and Bakul Dave
Southern Illinois University, Carbondale, Illinois 62901
͑Received 8 June 2001; accepted for publication 8 October 2001͒
In pollution control applications, the presence of water in the electrical discharge enhances oxidation
of pollutants. The results of an electrical discharge in gas when it flows through a heterogeneous
mixture of water and dielectric pellets are reported. The discharge in the wet plasma reactor is more
uniform compared to dry dielectric-barrier reactors. The electrical characteristics of such a discharge
are discussed. Also the results of removal of SO2 with the wet reactor are reported. The wet reactor
was found to be 5–10 times more energy efficient in removing SO2 compared to conventional dry
plasma reactors. © 2001 American Institute of Physics. ͓DOI: 10.1063/1.1426692͔
The use of nonthermal plasma for pollution control is an
active research area. To name only a few, nonthermal plas-
mas have been used for flue gas treatment,1,2 volatile organic
compound ͑VOC͒ removal,3 and vehicle exhaust pollution
control.4,5
At atmospheric pressures, there are two popular methods
of generating plasmas: ͑1͒ Corona in dc or pulsed form, and
͑2͒ barrier discharge with ac sources. Corona discharges are
due to inhomogeneous electrode geometries such as a point
electrode and a plane. Although corona discharges have
found numerous applications,6,7 they are limited due to the
low active volume available for processing the gas. The bar-
rier discharge, also known as a silent discharge or a partial
discharge, is widely used in commercial ozone synthesis and
excimer-UV generation.8,9
the filamentary behavior typical of dielectric-barrier dis-
charge.
Shown in Fig. 1͑b͒ is a simplified model of the reactor in
the presence of water. The glass beads are lumped as uniform
glass dielectric, the water drops are lumped as a layer of
water dielectric, and the air gaps between the beads and wa-
ter drops are lumped as the gas gap. The addition of water
increases the dielectric constant, which increases the overall
capacitance of the reactor. A large fraction of the discharge
current is primarily due to the capacitance ͑displacement cur-
The conventional barrier discharges are in air gaps.
However in a wet-barrier discharge a considerable portion of
the air gap is filled with water and dielectric beads to dis-
perse the water. The electric discharge takes place in the gas
bubbles formed in this mixture. Having water in the dis-
charge has several advantages: ͑1͒ It improves the removal
efficiency by chemically and physically enhancing the re-
moval mechanism, and ͑2͒ it prevents the fouling of the di-
electric surface due to contaminants. The discharge in the
wet plasma reactor behaves like a partial discharge in voids.
In this letter, we present some basic characteristics of the wet
plasma scrubber followed by results of removal of SO2.
The schematic of the wet-plasma reactor is shown in Fig.
1͑a͒. The configuration is a cross flow reactor with the water
coming down and the gas moving up. It consists of a cylin-
drical glass dielectric ͑1.6 cm internal diameter, and 1.5 mm
thickness͒ with the inside filled with glass pellets. The size of
the glass pellets is in the range of ͑4–5 mm͒. The inner
electrode is a 6 mm diameter stainless steel rod and the out-
erelectrode is a conducting film on the outersurface of the
dielectric. The plasma is created in the gas, liquid, and solid
mixture between the beads. This produces a heterogeneous
medium with the plasma created in gas bubbles as it flows
through the volume of the reactor. Due to the small dimen-
sion of the gas bubbles, the microdischarges in the bubble do
not reach the arc phase. Hence the discharge does not show
FIG. 1. ͑a͒ Schematic of the wet plasma reactor; ͑b͒ the equivalent
dielectric-barrier discharge representation of a wet plasma reactor.
a͒
Electronic mail: sdhali@siu.edu
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7
0003-6951/2001/79(26)/4298/3/$18.00 4298 © 2001 American Institute of Physics
On: Wed, 03 Dec 2014 16:39:50

Appl. Phys. Lett., Vol. 79, No. 26, 24 December 2001
Seethamsetty, Dhali, and Dave
4299
reactions are at the critical field ͑110 Townsend͒ for an air
discharge.4 The radicals produced by electron impact un-
dergo fast reactions with other radicals and molecules in the
discharge to form products with longer lifetimes. Shown be-
low are the reactions that determine the concentration of OH,
which is the most important radical for oxidation of
pollutants:1,11
k
4
3
O͑1D͒ϩO → O P͒ϩO :k ϭ4.0ϫ10Ϫ11 cm3sϪ1
,
͑5͒
͑
2
2
4
k
5
O͑1D͒ϩH2O→OHϩOH:k5ϭ2.28ϫ10Ϫ10 cm3 sϪ1, ͑6͒
k
6
OHϩOH→H2OϩO:k6ϭ1.0ϫ10Ϫ11
FIG. 2. Discharge power as a function of applied voltage for with and
without water flow in the reactor.
ϫ exp Ϫ500/T͒cm3 sϪ1
,
͑7͒
͑
The final product in the removal of SO2 is H2SO4. The
removal of SO2 takes place primarily through the following
reactions:1,11
rent͒ of the reactor as is generally the case with dielectric-
barrier type discharges. When no water is passed through the
system, the capacitive current is slightly lower.
The amount of power deposited to the discharge changes
significantly with the addition of water. The discharge power,
P, in a dielectric-barrier type discharge is given by10
k
7
SO2ϩOHϩM→HSO3ϩM:k7ϭ5.0
ϫ10Ϫ13 300/T͒3.3cm6 sϪ1
,
͑8͒
͑9͒
͑
c ϩc ͒
͑
g
d
k
8
Pϭ4fVЈc V Ϫ
V
,
͑1͒
ͩ
ͪ
d
p
c
c
HSO3ϩOH→H2SO4:k8ϭ9.8ϫ10Ϫ12cm3 sϪ1
,
cd
k
where f is the power frequency, cg and cd are the gap and
dielectric capacitance, respectively,
9
HSO3ϩO2→SO3ϩHO2:k9ϭ4.0ϫ10Ϫ13cm3 sϪ1
,
͑10͒
,
Ј , and
are the
vc
vp vc
peak to peak applied voltage, the critical gap voltage re-
quired to sustain a discharge, and the critical voltage at
k
10
SO2ϩOϩM→SO3ϩM;k10ϭ4.0ϫ10Ϫ32
which the gap breaks down, respectively. The voltages
vc
and Ј are nearly equal, however we have observed that Ј
vc
vc
ϫ exp Ϫ1000/T͒,
͑11͒
͑12͒
͑13͒
͑
10
tends to be lower than
.
When water is added to the
vc
k
11
discharge, cg and cd change significantly: As shown in Fig.
1͑b͒, the dielectric capacitance which consists of the glass
and water in series decreases due to addition of water in the
reactor. However the addition of water reduces the gas gap
volume, thus increasing the gas gap capacitance, cg . Figure
2 shows the discharge power as a function of the applied
voltage with and without water flow. Water flow decreases
the capacitance, cd and increases the term (1ϩcg /cd) caus-
ing power to decrease according to Eq. ͑1͒.
The discharge takes place in air bubbles at atmospheric
pressures saturated with water ͑ϳ3.2% at 21 °C͒. In a barrier
type discharge, the field in the air gap is equal to the critical
field ͑the field at which ionization is equal to loss by attach-
ment͒ required to maintain a discharge. Since water is a
highly attaching gas, the addition of moisture increases the
critical field. The energetic electrons in the discharge disso-
ciate and ionize the gas components. In an air–water mixture
the important electron impact processes are shown below:4
SO3ϩH2O→H2SO4;k11ϭ6.0ϫ10Ϫ13cm3/s,
k
12
SO3ϩOϩM→SO2ϩO2ϩM;k12ϭ8.1ϫ10Ϫ30cm6/s.
The reactions shown in Eqs. ͑8͒ and ͑9͒ are the primary
removal pathways. The reactions shown in Eqs. ͑10͒, ͑11͒,
and ͑12͒, form an additional removal pathway. Through the
reaction shown in Eq. ͑13͒, the SO3 is converted back into
SO2. Therefore, in order for reactions ͑9͒ and ͑10͒ to be
effective, water has to be present so that reaction ͑12͒ can
remove SO3.
The results of the removal of SO2 were studied with the
wet plasma reactor. A typical removal study on water flow is
shown in Fig. 3. The two sets of plots for a starting concen-
tration of 3000 ppm of SO2 in air corresponds to ͑1͒ concen-
tration of SO2 in the exhaust as a function of discharge en-
ergy, and ͑2͒ removal of SO2 per kWh of energy. Removal
studies were done for no water flow and with different water
flows through the reactor. There is an initial removal of SO2
when water is passed through the discharge without the
plasma, which depends on the water flow and is significant.
The SO2 gas is partially absorbed by water, and its solubility
depends on the temperature. When the discharge is turned
on, there is further removal due to the plasma. With increas-
ing flow of water, the fall in the concentration of SO2 is
faster as a function of discharge power. The wet plasma re-
k
1
3
eϩO → 2O P͒ϩe;k ϭ1.04ϫ10Ϫ9 cm3/s,
͑2͒
͑3͒
͑4͒
͑
2
1
k
2
eϩO →O P͒ϩO D͒ϩe;k ϭ4.8ϫ10Ϫ9cm3/s,
3
1
͑
͑
2
2
k
3
eϩH2O→OHϩHϩe;k3ϭ6.8ϫ10Ϫ12 cm3/s.
The reaction rates k1–3 are strongly dependent on the
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7
gas mixture and electric field. The rates shown for the above
actor was found to be more energy efficient compared to dry
On: Wed, 03 Dec 2014 16:39:50

4300
Appl. Phys. Lett., Vol. 79, No. 26, 24 December 2001
Seethamsetty, Dhali, and Dave
FIG. 4. Water pH at the exit of the reactor.
the measured values, thus indicating that the conversion of
SO2 to H2SO4 is the major pathway for the removal of SO2.
Typically in a dielectric-barrier discharge, the power fre-
quency range is from 60 Hz–10 kHz. Since the discharge is
highly capacitive, the reactive power increases with fre-
quency. Due to power supply limitation, the highest dis-
charge power achieved decreases as the power frequency is
increased. We have observed that the removal efficiency de-
pends on the power input to the discharge, and the power
frequency has little influence on the removal.
In conclusion, the characteristics of a wet plasma reactor
were discussed. Due to the relative changes in gas and di-
electric capacitance, it was found that the power coupled to
the discharge decreases due to water flow. In addition, with
water flow in the reactor, the removal efficiency of pollutants
is enhanced due to lower temperatures and the minimal re-
actor fouling due to acidic aerosols.
FIG. 3. SO2 removal as a function of power. The dashed lines show the SO2
concentration for various water flows as a function of discharge power. The
solid lines show the energy cost for removal for various water flows as a
function of discharge power.
plasma techniques, and the power requirements were less by
a factor of 5–10 compared to dry reactors.12 Li et al. re-
ported SO2 removal in dielectric-barrier type discharges in
the order of 10 g/kWh.12 Also the energy efficiency of the
wet reactor was an order of magnitude better compared com-
bined plasma photolysis, which produced removals of the
order of few g/kWh.13
The addition of water in the reactor causes enhanced
production of OH radicals, which improves the removal ef-
ficiency. In addition, the SO3 formed is quickly dissolved and
removed before the plasma can dissociate it back to SO2.
The change in temperature of the water from the inlet to the
outlet is only a few °C. However without water flow, the
reactor temperature rises by 50–100 °C. Therefore, the wet
plasma discharge operates much cooler than dry reactors.
From the temperature dependent rates shown for some of the
reactions, it can be shown that lower temperatures favor the
removal of SO2. The rate k6 , which is lower at lower tem-
peratures, reduces the loss of OH with decreasing tempera-
ture. The rate coefficient, k7 , which determines the removal
of SO2, increases at lower temperatures. In dry reactors, acid
aerosols on the dielectric foul the reactor, which can cause
intense arc-like microdischarges with reduced efficiency for
radical production.
The energy efficiency tends to be high at low discharge
power as shown in Fig. 2. At such low powers, the concen-
tration of OH is low. Due to the low destruction of OH by
reaction ͑7͒, the OH utilization for removal of SO2 is very
high. However, the percentage removal is low at such low
energies and is of little practical importance. Similar results
have been reported for SO2 removal at low discharge
energies.11,12
The solution pH at the exit of the reactor was measured
and is shown in Fig. 4. Also shown is the calculated value of
pH under the assumption that each SO2 molecule removed
produces one H2SO4. The calculated values agree well with
This research was partially supported by a grant from the
National Science Foundation.
1 W. Sun, B. Pashaie, S. K. Dhali, and F. I. Honea, J. Appl. Phys. 79, 3438
͑1996͒.
2 B. Pashaie, J. Li, W. Sun, S. K. Dhali, and F. I. Honea, Non-Thermal
Plasma for Cleanup of Flue Gas, Emerging Technologies in Hazardous
Waste Management vol. VI, edited by D. William Tedder and Frederick G.
Pohland ͑American Academy of Environmental Engineers, Atlanta, GA,
1996͒, p. 209.
3 L. A. Rosocha, G. K. Anderson, L. A. Behold, J. L. Cogan, H. G. Heck,
M. Kang, W. H. Macula, R. A. Tenant, and P. J. Wanton, in Proceedings of
the NATO Advanced Research Workshop on Non-thermal Plasma Tech-
niques for Pollution Control Part B: Electron Beam and Electrical Dis-
charge Processing, edited by B. M. Penetrante and S. E. Schultheis
Springer, Berlin, 1993, pp. 281–308.
4 A. C. Gentile and M. J. Kusher, J. Appl. Phys. 78, 2074 ͑1995͒.
5 B. M. Penetrante, in Book of Abstract for the Symposium on Emerging
Technologies in Hazardous Waste managements VI, edited by W. Tedder
͑American Chemical Society, Atlanta, 1994͒, pp. 195–198.
6 B. Eliason and U. Kogelschatz, J. Phys. B 19, 1241 ͑1986͒.
7 S. Masuda, Pure Appl. Chem. 60, 727 ͑1988͒.
8 B. Eliason and U. Kogelschatz, IEEE Trans. Plasma Sci. 19, 1063 ͑1991͒.
9 B. Eliason and U. Kogelschatz, IEEE Trans. Plasma Sci. 19, 309 ͑1991͒.
10 B. Sankaranarayanan, B. Pashaie, and S. K. Dhali, Appl. Phys. Lett. 74,
3119 ͑1999͒.
11 H. Matzing, Advances in Chemical Physics, edited by I. Prigogine and S.
A. Rice ͑Wiley, New York, 1991͒, Vol. LXXX, pp. 360–402.
12 J. Li, W. Sun, B. Pashaie, and S. K. Dhali, IEEE Trans. Plasma Sci. 23,
672 ͑1995͒.
13 M. B. Chang, J. H. Balbach, M. J. Rood, and M. J. Kosher, J. Appl. Phys.
69, 4409 ͑1991͒.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 129.22.67.7
On: Wed, 03 Dec 2014 16:39:50