Laser-induced fluorescence of OH radicals in a dielectric barrier discharge

Article abstract of DOI:10.1063/1.1324002

We discuss the results of laser-induced fluorescence measurements of OH radicals in a dielectric-barrier discharge. The discharge is in parallel plate geometry in atmospheric pressure air and argon. Although the air discharge consists of discrete microdischarges, two-dimensional images show the spatial uniformity of the OH radical. Results show that with increasing power, the OH production decreases due to gas heating and increased ozone production. The addition of O₂ increases the OH production at low concentrations; however, at higher O₂ concentration the OH concentration decreases due to increased electron attachment.

Full text of DOI:10.1063/1.1324002

Laser-induced fluorescence of OH radicals in a dielectric barrier discharge
R. Sankaranarayanan, B. Pashaie, and S. K. Dhali
Citation: Appl. Phys. Lett. 77, 2970 (2000); doi: 10.1063/1.1324002
View online: http://dx.doi.org/10.1063/1.1324002
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v77/i19
Published by the American Institute of Physics.
Additional information on Appl. Phys. Lett.
Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors
Downloaded 05 Jun 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions

APPLIED PHYSICS LETTERS
VOLUME 77, NUMBER 19
6 NOVEMBER 2000
Laser-induced fluorescence of OH radicals in a dielectric barrier discharge
R. Sankaranarayanan, B. Pashaie,a) and S. K. Dhalib)
Department of Electrical Engineering, Southern Illinois University, Carbondale, Illinois 62901
͑
Received 22 May 2000; accepted for publication 11 September 2000͒
We discuss the results of laser-induced fluorescence measurements of OH radicals in a
dielectric-barrier discharge. The discharge is in parallel plate geometry in atmospheric pressure air
and argon. Although the air discharge consists of discrete microdischarges, two-dimensional images
show the spatial uniformity of the OH radical. Results show that with increasing power, the OH
production decreases due to gas heating and increased ozone production. The addition of O2
increases the OH production at low concentrations; however, at higher O concentration the OH
2
concentration decreases due to increased electron attachment. © 2000 American Institute of
Physics. ͓S0003-6951͑00͒04245-5͔
2
ϩ
Atmospheric pressure nonthermal plasmas are being in-
vestigated for potential application in pollution prevention.
One of the most popular and proven methods of plasma gen-
eration at atmospheric pressure is the dielectric-barrier
are laser excited to one rotational level in the A ͚ state,
4
collisions cause other rotational levels to be populated. In
this work, a single photon excitation within the ͑1,0͒ band is
used, which is followed by the detection of ͑1,1͒ manifold.
This choice of energy levels has the advantage of low back-
ground noise and reduced self-absorption. The doubled dye
laser is tuned to a wavelength of 281.9 nm with a linewidth
1
discharge. The barrier discharge has been used for oxidizing
SO2 and volatile organic compounds.2 The barrier dis-
charge in air is a rich source of O and OH radicals. In the
plasma chemical removal process, the OH radical plays a
very important role. In addition, OH is the major pathway for
,3
4
Ϫ1
of 0.45 cm . At this wavelength, the laser energy is about
0 mJ and the pulse width is 8 ns. This laser excitation
1
produces a broadband radiation in the 312–330 nm range.
An interference filter in 316–324 nm range is used to remove
scattered light from the discharge. The laser passes along the
discharge unfocused with a 15 mm spot size. The ICCD is
gated and integrates over ten laser pulses.
converting CO to CO . The efficiency of removal by oxida-
2
tion is directly related to the production of OH radicals.
In this letter we report the results of OH measurements
by laser induced fluorescence ͑LIF͒ in a dielectric-barrier
discharge. Laser induced fluorescence is a common tech-
nique used for the measurements of OH radicals, particularly
2
In air the dielectric-barrier discharge consist of numer-
ous filamentary discharges, which fill the volume over time.
In argon and other monatomic gases, the discharge is glow-
like. A typical two-dimensional image of the LIF of OH with
the discharge turned on is shown in Fig. 3. This CCD picture
was taken for a discharge in atmospheric pressure air satu-
rated with water vapor ͑2.2% by weight at 300 K͒. Clearly
over a period of 1 s, the OH radicals fill the gap uniformly,
although it is being produced in the filamentary region of a
microdischarge.
From an image as shown in Fig. 3, the LIF signal in 2
mmϫ2 mm square was integrated to give the average LIF
intensity. A typical LIF spectrum for the discharge in atmo-
spheric pressure saturated air is shown in Fig. 4. The resolu-
tion is enough to distinguish some line structures, however,
4
in flames. This work was done in atmospheric pressure satu-
rated air and argon gas. There have been reports of other
techniques for qualitative measurements of OH in barrier
5
,6
discharges. However, the LIF method is sensitive and eas-
ily lends itself to two-dimensional imaging.
Figure 1 shows the schematic of the experimental setup.
The barrier discharge consists of a circular glass dielectric
coated with a conductive coating for one of the electrodes
and the other electrode is a circular stainless steel plate.
Gases are introduced through an orifice in the middle and
flow radially outwards. The diameter of the parallel plate
discharge is 15.3 cm, the dielectric thickness is 0.32 cm, and
3
the gap distance is 2 mm. With a gas flow rate of 33.3 cm /s,
the residence time is 1.1 s. Using UV optics, the discharge is
imaged on to a Princeton Instrument ICCD ͑intensified
charge coupled device͒ camera. A continuum YAG-pumped
dye laser ͑rhodamine 590 with a tunable range of 552–584
nm͒ along with the doubler is used to provide the appropriate
wavelength for laser excitation.
Figure 2 shows the OH energy levels, which are used for
laser excitation and fluorescence. Spectroscopically, the im-
portant electronic states are X2
͟
and A ͚. As OH molecules
2
a͒Current address: Dept. of Engineering Physics, 1 University Plaza, South-
east Missouri State University, Cape Girardeau, MO 63071; electronic
mail: bpashaie@physics.semo.edu
FIG. 1. Schematic of the experimental setup for LIF measurements of OH in
a dielectric-barrier discharge.
b͒
Electronic mail: sdhali@siu.edu
0003-6951/2000/77(19)/2970/3/$17.00
2970
© 2000 American Institute of Physics
Downloaded 05 Jun 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions

Appl. Phys. Lett., Vol. 77, No. 19, 6 November 2000
Sankaranarayanan, Pashaie, and Dhali
2971
FIG. 2. Energy level diagram of OH for LIF.
instrument resolution is not high enough to estimate the tem-
perature from the linewidth.
FIG. 4. Laser induced fluorescence spectrum of OH in a dielectric-barrier
discharge in atmospheric pressure air.
In a barrier discharge the mechanism of OH generation
is through the following direct electron and molecule-
molecule processes:
eV of energy͒ from simulations. For the gas mixture the G
factors under our discharge conditions for ͑OH͒ and O D are
1
k
1
ϳ1.4/100 eV. For a discharge power of 20 W, the power
3
H Oϩe→HϩOHϩe,
͑1͒
͑2͒
͑3͒
2
density is 0.56 W/cm . Therefore, the electron impact terms
1
6
k
in Eq. ͑7͒ are k (H O)(e)Ϸk (O )(e)Ϸ4.9ϫ10
2
1 2 2 2
3
Ϫ1
3
1
cm s . The OH concentration from Eq. ͑7͒ is 2.8
O ϩe→O ͑P͒ϩO ͑D͒ϩe,
2
ϫ1014 cm . This value compares well with OH values ob-
Ϫ3
k
3
tained from absorption studies.6
1
O ͑D͒ϩH O→2OH.
2
An upper limit on the temperature rise can be obtained
by assuming all the input power goes to gas heating. Assum-
ing air temperature at 300 K, the density, ␳ϭ1.18
The rates k and k are dependent on the ratio of electric
1
2
2
field to gas pressure and also on the gas composition. The
reaction rate for k ϭ2.28ϫ10 cm /s ͑Ref. 2͒. Reactions
2͒ and ͑3͒ form the main pathways for OH production in
Ϫ10
3
Ϫ3
3
ϫ10
g/cm , the heat capacity at constant pressure, c
3
p
Ϫ1
Ϫ1
͑
ϭ1.0 Jg
K
, the rise in temperatureр25 °C/W. Since
barrier discharges. In saturated air, the important pathways
for the destruction of OH are listed below:
the dominant gas in air is nitrogen, a large fraction of the
input energy is stored in the ground-state vibrational mani-
fold of nitrogen. Typically less than 10% of the energy goes
into gas heating. For a 40 W input power, the expected rise
in temperature over ambient is equal to 100 °C. The increase
k
4
8
OHϩOH→H OϩO,
͑4͒
2
k
5
in k and k are 56% and k 110%, respectively, for 100 °C
4
5
5
OHϩO →HO ϩO ,
͑5͒
3
2
2
rise in temperature.
Ϫ11
3
where k4ϭ1.0ϫ10 exp(Ϫ500/T͒cm /s, and k5ϭ1.6
Shown in Fig. 5 is the LIF intensity of OH as a function
of the applied voltage to discharge. Also shown is the power
input to the discharge. Increasing voltage increases the
power input to the discharge as predicted by the linear rela-
tionship shown in Ref. 1. However, it does not increase the
electric field in the air gap, which remains near the critical
Ϫ12
3
7
ϫ10 exp(Ϫ1000/T͒cm /s.
If (A) represents the concentration of A, and assuming
steady-state:
2
2
k ͑OH͒ ϩk ͑OH͒͑O ͒ϭk ͑H O͒͑e͒ϩ2k ͑H O͒
4 5 3 1 2 3 2
1
1
ϫ͑O D͒.
͑6͒
field due to charging of the dielectric. As a result, the field
dependent rates do not change with increasing voltage. Ini-
tially, the increase in power goes to increase the number of
From Eqs. ͑2͒ and ͑3͒, at steady state, it can be shown
k (H O)(OD)ϭk (O )(e). At room temperature, k Ӷk
3
2
2
2
5
4
and ͑OH͒ is approximately given by
͑
OH͒ϭ ͓k ͑H O͒͑e͒ϩ2k ͑O ͒͑e͔͒/2k .
ͱ
͑7͒
1
2
2
2
4
A rough upper estimate of the ͑OH͒ concentrations can be
obtained from G factors ͑number of radicals produced/100
FIG. 5. Power input to the discharge and the relative LIF intensity of OH as
a function of the applied voltage. These data are for a discharge in atmo-
spheric pressure saturated air.
FIG. 3. Two-dimensional CCD image of the LIF intensity of OH in a
parallel-plate dielectric-barrier discharge in atmospheric pressure air.
Downloaded 05 Jun 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions

2972
Appl. Phys. Lett., Vol. 77, No. 19, 6 November 2000
Sankaranarayanan, Pashaie, and Dhali
age. The initial concentration of OH, in the absence of O , is
2
due to direct electron impact production reaction shown in
Eq. ͑1͒. With increasing oxygen concentration, reactions ͑2͒
and ͑3͒ become important and at 0.5% O concentration ex-
2
ceeds the production due to reaction ͑1͒. As predicted by
reactions ͑2͒ and ͑3͒, there is an increase in OH concentra-
tion as oxygen concentration is increased. However, it peaks
around 3% and drops off with increasing oxygen concentra-
tion. Oxygen being an attaching gas has a dramatic effect on
the nature of the discharge. With increasing oxygen concen-
tration, there is a decrease in electron density, which reduces
the radical production by direct processes ͓Eqs. ͑1͒ and ͑2͔͒.
In addition increased concentration of O and O will also
3
reduce the OH concentration due to reaction ͑5͒.
In conclusion, we have imaged the OH profile in an
atmospheric pressure air filled dielectric-barrier discharge
and found the distribution to be fairly uniform. Excited oxy-
gen atom is a major source of OH at low concentrations of
oxygen. However increasing power input to the discharge
beyond an optimum value tends to reduce the OH concentra-
tion.
FIG. 6. LIF intensity of OH, as a function of O concentration, in an argon
discharge at atmospheric pressure.
2
microdischarges in the volume of the discharge. This results
in an increase in average LIF intensity with increasing volt-
age. However, after this initial increase, there is a gradual
decrease in the LIF intensity with increase in the applied
voltage. Once the discharge volume is completely filled with
microdischarges, the increase in power results in higher heat
dissipation, which is evidenced by the increased brightness
of each microdischarge. The increase in radical production
due to increase in power is offset by the destruction of OH
This work was partially supported by a grant from the
National Science Foundation.
1
R. Sankaranarayanan, B. Pashaie, and S. K. Dhali, Appl. Phys. Lett. 74,
3
119 ͑1999͒.
2
3
due to the increase in k4 and k . In addition reaction ͑5͒
J. Li, W. Sun, B. Pashaie, and S. K. Dhali, IEEE Trans. Plasma Sci. 23,
672 ͑1955͒.
5
reduces the OH concentration with increasing voltage. Hibert
L. A. Rosocha, G. K. Anderson, L. A. Bechtold, J. J. Coogan, H. G. Heck,
M. Kang, W. H. McCulla, R. A. Tennant, and P. J. Wantuck, in Proceed-
ings of the NATO Advanced Research Workshop on Non-Thermal Plasma
Techniques for Pollution Control Part B: Electron Beam and Electrical
Discharge Processing, edited by B. M. Penetrante and S. E. Schultheis
et al. have shown increasing voltage increases the O con-
3
centration and reduces the OH lifetime in a dielectric barrier
6
discharge.
As mentioned earlier, reactions ͑2͒ and ͑3͒ are an impor-
tant pathway for the formation of OH radicals. We studied
the dependence of OH formation on oxygen in a saturated
͑
Springer, Berlin, 1993͒, pp. 281–308.
4
N. M. Laurendeau and J. E. M. Goldsmith, Combust. Sci. Technol. 63,
39 ͑1989͒.
1
5
6
argon (H O 1.6% by weight at 300 K͒ discharge. Argon is a
J. J. Coogan and A. D. Sappy, IEEE Trans. Plasma Sci. 24, 91 ͑1996͒.
C. Hibert, I. Gaurand, O. Motret, and J. M. Pouvesle, J. Appl. Phys. , 7070
2
commonly used carrier gas in pollution control application,
and it is simpler to analyze as compared to air. Figure 6
shows the LIF intensity of OH as a function of oxygen con-
centration. This experiment was done at a fixed applied volt-
͑
1999͒.
7
8
H. Matzing, Advances in Chemical Physics, edited by I. Prigogine and S.
A. Rice ͑Wiley, New York, 1991͒, Vol. LXXX, pp. 360–402.
S. K. Dhali, IEEE Trans. Plasma Sci. 17, 603 ͑1989͒.
Downloaded 05 Jun 2013 to 141.161.91.14. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions