22nd International Symposium on Plasma Chemistry
July 5-10, 2015; Antwerp, Belgium
Decomposition of micro-pollutants in water by combination of non-thermal
electrical discharge and adsorption on nano-fiber materials
A. Nikiforov1, P. Vanraes1, S.W.H. Van Hulle2, J. Van Durme3 and C. Leys1
1
Department of Applied Physics, Ghent University, Sint-Pietersnieuwstraat 41 B4, BE-9000 Ghent, Belgium
Department of Industrial Biological Sciences, Ghent University, Graaf Karel de Goedelaan 5, BE-8500 Kortrijk, Belgium
3
Research Group Molecular Odor Chemistry, Department of Microbial and Molecular Systems (M2S), KU Leuven,
Technology Campus, Gebroeders De Smetstraat 1, BE-9000 Ghent, Belgium
4
Separation Science Group, Department of Organic Chemistry, Universiteit Gent,
Krijgslaan 281 S4-bis, BE-9000 Gent, Belgium
2
Abstract: A new type of plasma/liquid reactor for decomposition of micro-pollutants in
water is tested. System based on atmospheric direct barrier discharge (DBD) combined
with micro-pollutant adsorption on nano-fibers. Tests of the reactor performance are
carried out with 3 types on pesticides. Investigation of reaction kinetics and by-product
analysis shows high reactor efficiency with high energy yield.
Keywords: water treatment, barrier discharge, micro-pollutants, atrazine
1. General
With increase of human’s impact on environment
different types of organic pollutants are detected more
frequently all over the world in surface water, ground
water and drinking water [1]. Some of these pollutants,
including pesticides, antibiotics and hormones, have
endocrine disruptive, carcinogenic or bioaccumulative
properties and are therefore considered hazardous even in
relatively small concentrations [2]. Amongst the different
proposed advanced oxidation processes (AOP), the use of
low temperature plasma may prove to be a sustainable and
efficient approach. Up to now, studies have focused
mostly on plasma initiated decomposition of a few very
simple compounds, e.g., phenolic compounds, sulfonol
and textile dyes [3, 4]. Here, the pesticides: atrazine,
pentachlorbenzene, and α-HCH are used as model micropollutants, since they are good representative of persistent
and hazardous micropollution in water.
Amongst various plasma reactor configurations that
have been investigated, AC powered direct barrier
discharge over moving water film has been found to be
both energy efficient and sustainable [5, 6]. This reactor
concept is further optimized in our work by adding an
adsoptive material underneath the water surface and
recycling the generated plasma gas by additionally
bubbling it through the solution under treatment.
Adsorption allows reaching higher local micropollutant
concentration close to the active plasma region. As a
result, more collisions will take place between plasma
species and micropollutants, increasing overall energy
efficiency of the treatment process. Next to that, the
plasma gas contains reactive species such as ozone and
H 2 O 2 that can further degrade organic compounds. In
this contribution we have tested our new reactor based on
these two principles. Two voltage waveforms were used
for discharge excitation: 50 kHz AC and 1 kHz 500 ns
P-I-3-21
pulsed voltage. A scheme of the setup and a picture of
the plasma are shown in Fig. 1(a)-(b).
Fig. 1. (a) Scheme of the setup with DBD reactor
chamber (right) and ozonation chamber (left) used for
atrazine desctruction, (b) photograph of the setup during
plasma generation.
1
The emission spectrum of the discharge is dominated by
radiative relaxation of excited vibrational states of the N2
molecule, indicating the presence of metastable N2
molecules. Metastable states of N2 formed in air plasma
are involved in formation of O and O3. Despite the
presence of atomic oxygen in the plasma (measurements
not shown here), emission lines of excited oxygen atoms
are very small, such as the OI triplet at 777.5 nm for
example. The OH(A-X) (0,0) Q band head around 309
nm and the OH(A-X) (1,1) band around 315 nm are
overlapped with the N2* bands and are therefore not
observed in the spectrum. Nevertheless, the small
OH(AX) (0,0) R-band head around 307 nm indicates the
presence of excited OH radicals. The total emission from
the discharge almost directly related to discharge power.
As spectrum indicates, increase in power leads to increase
in number of the micro- discharges, which is also
observed by eye whereas appearance of single microdischarges is not affected by increase of the discharge
power in used in our work range.
Our tests have shown that plasma formation in gas
phase following by reaction of active species in the
liquid/plasma interface results in generation of H 2 O 2 in
water and dissolved O 3 . In pulsed discharge reactor, the
measured energy yield of H 2 O 2 production in liquid
phase is about 0.23 g/kWh and concentration of H 2 O 2 is
linearly dependent on treatment time. The energy yield of
O 3 production is measured to be 5.07 g/kWh. Change of
the applied voltage from sub-ns pulses to AC leads to
increase of average H 2 O 2 production energy yield to
0.51 g/kWh at high discharge power but lower O 3
production which is halves from 0.189 g/kWh when input
power is increased from 24 to 64 W due to gas heating in
the reactor. While the H 2 O 2 and O 3 production in the
reactors is not influenced by the presence of the
nano-fiber materials, there is a significant increase in
micro-pollutant decomposition when the membrane is
added to the setup.
Thus an atrazine removal yield of 85% in case of puled
DBD can be obtained with nano-fiber material at 45 min
of treatment where only about 61% removal is reached
with plasma alone. The observed effect is caused by
atrazine adsorption on the nano-fibers close to the plasma
active region, leading to a higher local atrazine
concentration near the plasma/liquid interface. The
higher local concentration increases the frequency of
direct and indirect oxidizing interactions of the
micropollutant with reactive species from the active
plasma region.
To reinforce this explanation, the
contribution of the dominant atrazine degradation
processes in the water bulk has been estimated.
According to literature, direct oxidation by ozone and
oxidation by peroxone are the dominant processes in the
water bulk in absence of UV light. According the kinetic
model [7], the contribution of these bulk processes to the
overall atrazine degradation is calculated to be only about
20% as shown on Fig. 2. Therefore, the determining
oxidation reactions are occurring in the thin water layer
2
near the plasma active region.
Fig. 2. Atrazine and OH radical concentration in function
of treatment time for direct oxidation with ozone (O3), for
dark peroxone process and for combination of both, as
calculated with the kinetic model of [7]. The theoretical
limit is calculated taking into account the peroxone
process with direct ozonation in the bulk solution.
The same positive effect of adsorbing nano-fibre
material has been observed also in AC plasma system
with active carbon textile but overall energy efficiency of
the system was lower. In AC DBD system detected byproducts with HPLC-MS analysis are identified as the
first or second generation intermediates simazine amide,
atrazine amide and deethylatrazine and the deeper
oxidation product didealkylatrazine. In case of pulsed
discharge excitation the deeper oxidation with formation
of ammelide has been observed for atrazine desctruction.
Comparison of our approach for water treatment with
reactors described in literature indicates that combination
of plasma discharge with pollutant adsorption and ozone
recycling is an energy efficient new water treatment
technology. We have found that system efficiency is
almost doubled when the nanofiber material or active
carbon textile is placed in the plasma reactor with deeper
degradation of micro-pollutant to the by-products. These
results show the benefits of combining non-thermal
plasma with pollutant adsorption for degradation of
micropollution, a synergetic effect that yet has to receive
more attention.
2. Acknowledgements
This research was partly funded by COST action
TD1208 through STSM action. Authors thanks Carbon
Cloth Division for Zorflex samples and personally Jack
Taylor for fruitful discussion of active carbon water
treatment processes.
3. References
[1] R.P. Schwarzenbach et al.
1072-1077 (2006)
Science, 313(5790),
P-I-3-21
[2]
[3]
[4]
[5]
[6]
[7]
N. Bolong, A. Ismail, M. Salim and T. Matsuura.
Desalination, 239(1-3), 229-246 (2009)
L.R.
Grabowski,
E.M.
van
Veldhuizen,
A.J.M. Pemen and W.R. Rutgers. Plasma Chem.
Plasma Process., 26(1), 3-17 (2006)
B.P. Dojcinovic et al. J. Hazard. Mater., 192(2),
7630771 (2011)
M. Magureanu, D. Piroi, N.B. Mandache and
V. Parvulescu. J. Appl. Phys., 104(10), 103306
(2008)
M.A. Malik. Plasma Chem. Plasma Process., 30,
21-31 (2010)
A. Hong, M. Zappi, C. Kuo and D. Hill. J. Environ.
Engng., 122(1), 58-62 (1996)
P-I-3-21
3