Chemosphere 243 (2020) 125377

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Plasma-assisted advanced oxidation process by a multi-hole dielectric

barrier discharge in water and its application to wastewater treatment
Sukhwal Ma a, Kangil Kim b, Semin Chun b, Se Youn Moon c, Yongcheol Hong b, *
a
Radwaste Management Center, Korea Atomic Energy Research Institute, 111, Daedeok-daero, 989 Beon-gil, Yuseong-gu, Daegeon, South Korea
b
Plasma Technology Research Center, National Fusion Research Institute, 814-2 Ohsikdo-dong, Gunsan, 573-540, South Korea
c
Department of Applied Plasma Engineering, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, 561-756, South Korea

h i g h l i g h t s

The plasma system can produce a large amount of active species.

The radicals can be directly injected into the water.

We analyzed the properties of plasma-treated wastewater.

We confirmed that plasma-assisted AOP process was possible.

Turbidity, BOD, and COD were improved with bacterial inactivation.

a r t i c l e i n f o a b s t r a c t

Article history: Advanced oxidation process (AOP) is a promising technology to decolorize and reduce organic con-
Received 1 August 2019 taminants in water. It is carried out using hydroxyl radicals (
OH) with an oxidizing potential of 2.80 V.
Received in revised form Non-thermal plasma can directly generate
OH while maintaining a low temperature, and O3, H2O2, and
12 November 2019
UV light are also generated; these are necessary for AOP. In this study, we developed a multi-hole
Accepted 13 November 2019
Available online 16 November 2019
dielectric barrier discharge (DBD) system capable of generating radicals and active species in water for
assisting AOP. We confirmed the optimized operating conditions based on critical parameters, including
Handling Editor: E. Brillas electrical and optical properties and O3 concentration. Furthermore, we described the plasma-based AOP
through experimental results. We performed wastewater treatment using the multi-hole DBD: turbidity,
Keywords: BOD, and COD, were reduced by 60%, 40%, and 60%, respectively, after 20 min of treatment. Finally,
Multi-hole DBD 99.99% of Escherichia coli were eliminated after plasma treatment.
Plasma-based AOP © 2019 Elsevier Ltd. All rights reserved.
Advanced oxidation process
Bacterial inactivation
Wastewater purification
Seaweed-washing wastewater

1. Introduction biochemical oxygen demand (BOD), chemical oxygen demand

Water pollution has reached a very serious level globally (Owa,
2014; Haseena et al., 2017). Release of most industrial wastewaters,
including chemical substances and heavy metal particles, are
strictly regulated, and therefore most factories are well equipped
with wastewater purification facilities. However, wastewater
generated during seaweed washing processes has a high
* Corresponding author.
E-mail addresses: shma@kaeri.re.kr (S. Ma), kikim@nfri.re.kr (K. Kim), Semin83@
nfri.re.kr (S. Chun), symoon@jbnu.ac.kr (S.Y. Moon), ychong@nfri.re.kr (Y. Hong).
(COD), and turbidity, and has relatively weak regulations, because it
does not contain unusual chemical substances or heavy metals. In
reality, most seaweed processing factories have almost no purifi-
cation facilities except simple filters. Thus, untreated seaweed-
washing wastewater containing high BOD, COD, and turbidity is
discharged into rivers, streams, and seawater (Suyasa et al., 2015).
The wastewater also contains a large number of pathogenic mi-
croorganisms and has a red coloration due to turbidity and,
consequently, this type of wastewater causes deterioration of water
quality in receiving waters. Therefore, a technology capable of
effectively reducing BOD, COD, and turbidity is required.

https://doi.org/10.1016/j.chemosphere.2019.125377
0045-6535/© 2019 Elsevier Ltd. All rights reserved.
2 S. Ma et al. / Chemosphere 243 (2020) 125377
Table SM-1 presents the oxidizing potential of the representa-
tive oxidizing agents (Hassaan et al., 2017). Conventional tech-
niques for reducing BOD and COD include injecting chemical
oxidants such as chlorine (E0 ¼ 1.36 V) and O3 (E0 ¼ 2.08 V) into
wastewater and using advanced oxidation processes (AOP) (Grimm
et al., 1998; Lalwani et al., 2013). The method used for injecting
chemical oxidants is inexpensive and effective; however, there are
disadvantages such as reactions with water or organic compounds
in water, which may generate secondary by-products. For example,
trihalomethanes (THMs) are produced by the reaction of chlorine
with contaminated organic matter in water. More than 700 halo-
genated disinfected by-products (DBPs) can be generated in chlo-
rinated water and THMs are an important group of DBPs that
forming during the chlorination of water. THMs form when free
chlorine reacts with natural organic material over an extended
contact time throughout water treatment processes. Research on
health effects showed a relationship between THMs exposure and
bladder cancer. THMs are now suspected of causing not only cancer,
but also liver and kidney damage (Clayton et al., 2019; Masoud
et al., 2019).
The advantages of the oxidation process using O3, which is a
strong oxidizing agent, are that it effectively decomposes organic
matter, O3 is rapidly decomposed into O2 and OH in water by the
following reactions, and it does not produce any secondary by-
products.
O3 þ OH / HO
2 þ O
2 (1)
O3 þ H2O / 2HO
2 (2)
HO2 4 Hþ þ O
2 (3)
O 
2 þ O3 / O2 þ O3 (4)
O 
3 / O þ O2 (5)


OH 4 O þ Hþ (6)


OH þ O3 / HO
2 þ O2 (7)
HO2
þ HO2
/ H2O2 þ O2 (8)
HO
2 þ O 
2 / HO2 þ O2 (9)
O3- þ O2- / 2OH- þ 2O2 (10)
3O3 þ OH- þ Hþ / 2HO
þ4O2 (11)
The reaction involving hydroxide ions and O3 leads to the cre-
ation of the superoxide anion and hydroperoxyl radical HO
2 (Eq.
(1)). The reaction involving O3 and the superoxide anion creates the
ozonide anion O 3 (Eq. (4)). The radical instantly decays releasing


OH.
Consequently, three O3 molecules generate two
OH (Eq. (11)).


OH generated from the decomposition of O3 molecules is a strong
oxidizing agent with an oxidation potential of 2.80 V, and readily
reacts with most organic molecules. However, large and industrial
O3 generators have been limited to commercial applications owing
to drawbacks such as high power consumption and large spaces
required for installment. Recently, compact air-cooled O3 genera-
tors have been commercialized. These generators employ different
methods for O3 production; however, they primarily use a UV or
corona method. The UV method is inefficient for O3 production in
terms of power consumption and transmittance. The corona
method yields high amounts of O3, but it has disadvantages such as
relatively high power consumption and wearing of electrodes.
The advanced oxidation process (AOP) is receiving increased
attention as a method of water treatment and removal of pollutants
from water (Deng et al., 2015). The AOP utilizes the strong oxidizing
power of
OH (E0 ¼ 2.80 V) for improved O3 oxidation. The mech-
anism of OH production highly depends on the AOP technique used.
There are several well-known methods for generating hydroxyl
radicalsdO3-based AOP, UV/H2O2-based AOP, and photocatalytic
oxidation with TiO2 are generally used in the industry.
O3 þ H2O2 (12)
O3 þ UV (13)
H2O2 þ UV (14)
O3 þ H2O2 þ UV (15)
Photocatalytic (TiO2 þ UV) (16)
Photo Fenton (Feþ2 þ H2O2 þ UV) (17)
Fenton Reaction (Feþ2 þ H2O2) (18)
However, existing AOP methods have several disadvantages that
need to be addressed; O3 þ UV-based AOP have high energy con-
sumption, and installation is not cost effective when the UV source
is used for large amounts of wastewater. AOP based on O3 þ H2O2
require excessive H2O2 injection to generate OH and inhibit mi-
crobial growth. Photocatalytic AOP using photocatalysts such as
TiO2 are inactive under visible light, and exhibit low light efficiency.
Moreover, operating cost is increased when an artificial light source
is used. AOP employing the Fenton Oxidation method, Eq. (19), have
a disadvantage in that a large amount of sludge is generated due to
the injection of Fe2þ and the process efficiency is reduced due to the
scavenging effect (Collivignarelli et al., 2017).
Fe2þ þ H2O2 / Fe3þ þ
OH þ OH- (19)
Therefore, there is a need for a technology capable of purifying
wastewater simply and effectively while compensating for the
drawbacks of AOPs. Non-thermal plasma can directly generate
OH
based on the operating conditions while maintaining a low tem-
perature. H2O2, O3, and UV light are also generated from the plasma,
which are necessary for AOP (Lee et al., 2018).
In this regard, we have developed a multi-hole dielectric barrier
discharge (DBD) with a simple structure and low power con-
sumption that can produce a large amount of radical and active
species including
OH, O3, and H2O2, thereby purifying seaweed
washing wastewater. The newly developed multi-hole DBD is
characterized by the fact that the radicals can be directly injected
into the wastewater through a plurality of holes on the outer
dielectric. Most short-lived species have a very short lifetime,
including hydroxyl the radical OH (~2.7 ms); nitric oxide NO (~1.2
ms), H2O2 (1.4 ms), and superoxide O2*-, (~1.3 ms) (Gligorovski et al.,
2015).
OH has been shown to have the longest lifetime among
reactive oxygen species (Attri et al., 2015). Thus, they should be
injected directly into wastewater to enable them to contribute to
wastewater treatment before decaying. Conversely, O3 molecules
belong to a species with a long lifetime, although this varies with
water and gas temperature (Batakliev et al., 2014). However, a large
amount of O3 is lost during transportation from an O3 generator to
wastewater, thereby reducing the efficiency of the O3 oxidation
 S. Ma et al. / Chemosphere 243 (2020) 125377 3

process. The multi-hole DBD can directly affect the wastewater out under optimized conditions, and it was confirmed that 99.99%
treatment with almost no loss of radicals and active species. of Escherichia coli (E. coli), a representative microorganism, was
Moreover, it can be installed in the wastewater, thus improving the eliminated.
cooling efficiency of the electrodes.
In the multi-hole DBD, O3 is dominantly produced, which con-
2. Material and methods
tributes greatly to the oxidation and decomposition of organic
contaminants in seaweed washing wastewater. Depending on the
2.1. Configuration of the multi-hole DBD
pH, water temperature and concentration, O3 rapidly decomposes
into O2 and
OH in water. Depending on the plasma conditions, a
Hong et al. (2019) first designed a multi-hole DBD system and
small amount of H2O2 is produced by water dissociation, which is
described the detailed characteristics and effects of the electrodes.
the initiator of hydroperoxyl radical formation. Generated H2O2
Schematic drawings of the multi-hole DBD electrode and the sys-
also contributes to the formation of
OH through the following
tem are shown in Fig. 1. The multi-hole DBD system consists of a
reaction (Eq. (20)) when reacted with UV light. Generated
OH from
high-voltage electrode, a ground electrode, an inner dielectric tube,
O3 and H2O2 decomposition can induce AOP to purify wastewater
an outer dielectric tube, a high-voltage power supplier and a
with higher efficiency (Andreozzi et al., 1999; Chen et al., 2015).
reactor containing wastewater. The power supply used in this
experiment was a commercially available 25 kHz transformer
H2O2 þ UV/ 2 HO
(20)
(NTO-500, NT electronics, South Korea), which was connected to
the high-voltage electrode and the ground electrode. The high-
The DBD plasma is generated by a breakdown of the injected air
voltage electrode shown in Fig. 1A is made of ∅ 2.4 stainless steel
gas in the strong E-field generated between the inner electrode and
rod with asymmetric electrode configuration. This electrode
the outer electrode; the generated DBD is discharged into the water
arrangement can enhance the emission of plasma plumes toward
through the multiple holes according to the gas flow rate. In this
water, thereby increasing the contact area between water and
study, we analyzed the electrical and optical properties of the
plasma and decreasing unnecessary discharge of plasma plumes
newly developed multi-hole DBD and measured the concentration
into water. It is enclosed within a quartz tube with an inner
of O3 after passing through the water. We also measured dissolved
diameter of 3 mm as the dielectric material. The high voltage
O3 and H2O2 concentrations using a dissolved O3 meter and a
electrode is located at a 2 mm eccentricity in the direction of the
photometer, as described in section 2.3. Furthermore, we
water. The outer quartz tube is located at the discharge gap of 1 mm
confirmed the optimal conditions based on the gas flow rate, the O3
from the inner quartz tube and is surrounded by wastewater as
and H2O2 concentrations, the wastewater temperature during the
ground electrode. On the water side of the outer quartz tube,
treatment, and the electrical and optical characteristics. Based on
multiple holes (1 mm diameter) are located at 5 mm intervals in the
the optimized condition, we analyzed the properties of wastewater
axial direction. The reactive gas is injected into the discharge gap
treated by the multi-hole DBD. As a result, the turbidity, BOD, and
between the outer quartz tube and the inner quartz tube. The active
COD were improved. Furthermore, the sterilization test was carried
species generated in the discharge gap are injected directly into the

Fig. 1. A) A multi-hole dielectric barrier discharge, and B) configuration of the treatment system.
4 S. Ma et al. / Chemosphere 243 (2020) 125377
wastewater through multiple holes located in the water direction,
according to the gas flow. We designed a simple rectangular reactor
with the size of 160 mm  700 mm x 115 mm (H) for wastewater
treatment (Fig. 1B). The multi-hole DBD was installed at the bottom
of the reactor providing the active species with a longer reaction
time. In addition, the O3 concentration was measured after passing
water through the gas outlet in the upper part of the reactor.
2.2. Preparation of seaweed washing wastewater
The seaweed washing wastewater used in this study was
collected from a seaweed processing plant (SH tech) in Suwon,
South Korea. Wastewater was collected just prior to its discharge
into a stream, and was filtered through a 20 mm filter and had
almost no suspended particles. The sample had a mild red color,
and the average turbidity and pH were 14 NTU and 6.98, respec-
tively (n ¼ 10). The initial average BOD and COD of wastewater were
46.3 mg L1 and 39.7 mg L1, respectively (n ¼ 5).
2.3. Analytical equipment
A digital phosphor oscilloscope (DPO4054B, Tektronix, USA) and
an emission spectroscope (Spectra view 2100, K-MAC, South Korea)
were used to analyze the electrical and optical characteristics,
respectively, of the multi-hole DBD. After passing through the
wastewater, the amount of O3 generated from the multi-hole DBD
was measured at the gas outlet point via an O3 monitor (106-M, 2B
Technologies, USA) with a measurement error of 0.01 mg L1. The
dissolved O3 concentration was measured by portable dissolved O3
meter (OZ-20, TKK-TOA, Japan) with measurement error of ±2%.
The amount of H2O2 in the water was measured via a photometer
(V-2000 and K-5543, CHEMetrics, USA). We also measured pH,
conductivity, and turbidity using a pH meter (LAQUA, Horiba,
Japan), and a portable turbidity meter (TB-31, TKK-TOA, Japan) with
measurement errors of ±1.01 and ±0.5 NTU, respectively. We also
measured the water temperature using a thermocoupler
(A1.T9304C, Daihan, South Korea). Images of the multi-hole DBD (as
shown in Fig. SM-1) were taken using a camera (ILCE-5100, Sony,
Japan) and its properties of F-stop, exposure time, ISO sensitivity
and Pixel were F/22, 5 s, ISO-3000 and 4000  6000, respectively.
The BOD and COD were analyzed by an authorized analytical
institution (Water Environment Research Center, Jeonju, South
Korea); BOD was measured by the dilution method, which is the
most common method of measuring BOD, and COD was analyzed
by CODMn measurement using potassium permanganate (K2MnO4).
2.4. Set-up for sterilization test against Escherichia coli
We prepared the sterilization experiment using Escherichia coli
(E. coli) as a representative pathogen. E. coli was incubated for 2
days at 37  C, after which 1.5  104 CFU mL1 (colony-forming unit
per milliliter) of E. coli was injected into the simple rectangular
reactor (Fig. 1B) containing 500 mL of water and sterilized by the
multi-hole DBD. Bacterial inactivation experiments were repeated
five times.
3. Results and discussion
3.1. Generation of the multi-hole DBD in water
The main advantage of the multi-hole DBD is that active species
and radicals can be directly injected into wastewater to oxidize or
decompose organic contaminants with high efficiency. In addition,
a higher efficiency of purification can be expected because
OH
generated by the decomposition of O3 and H2O2 species can
activate the AOP reaction.
The discharge mechanism is that DBD plasma (Ma et al., 2018) is
generated by a strong electric field at a discharge gap of 1 mm
between the inner electrode and the outer electrode, and a large
amount of active species are generated from the DBD and are
injected into the water through 23  1 mm holes drilled in the
external insulator according to the gas flow (Fig. SM-1). The gas
velocities were 1.84, 2.3, 2.76, 3.23 and 3.69 m s1 at flow rates of
20, 25, 30, 35 and 40 L min1, respectively. Fig. SM-1 shows the
stable plasma generation at an air flow rate of 25 L min1. More-
over, the multi-hole DBD plasma is capable of generating a high O3
concentration with a simple structure and low power consumption
by using a 15 kV, 25 kHz AC transformer, which is a commercially
available power source generally used for neon signs.
3.2. Analysis of voltage and current waveform
The recorded voltage and current waveforms presented in
Fig. 2A show that discharge operated in a filamentary mode
(Rajasekaran et al., 2009). The voltage of the multi-hole DBD in
water with a flow rates of 20, 25, 30, 35 and 40 L min1 were
Vrms ¼ 7.106, 7.133, 7.426, 7.07 and 7.371 kV, respectively, i.e., there
was almost no change in the voltage with changes in flow rate. In
addition, each half cycle was characterized by a few tens of sub-
microsecond current peaks. These are expected features of such
an air DBD discharge (Han et al., 2015). The measured current by
flow rates of 20, 25, 30, 35 and 40 L min1 were Arms ¼ 4.694, 4.817,
5.204, 5.75 and 5.924 mA, respectively. As shown in Fig. 2B, power
consumption was 33.35, 34.359, 38.644, 40.65 and 43.66 W for flow
rates of 20, 25, 30, 35 and 40 L min1, respectively. Thus, it was
confirmed that the current and power consumption increased ac-
cording to the flow rate.
3.3. Generated O3 and H2O2 concentrations and optimized
conditions
In the case of air DBD, large amounts of active species (e.g., O3,
NO, NO2) can occur depending on the operating conditions (Ma
et al., 2017). In the filamentary mode of DBD plasma, O3 is the
dominant species and plays a major role as a strong oxidizing agent
in organic contaminants oxidation and bacterial inactivation of
wastewater (Ma et al., 2018). Fig. 3A shows the time-dependent
concentration of O3 generated from the multi-hole DBD passes
through water. It shows real-time measurements of O3 concentra-
tions, and the averaged concentrations of 180, 240, 225, 210 and
214 mg L1 with flow rates of 20, 25, 30, 35 and 40 L min1,
respectively. It can be seen from these results that the ozone con-
centration was measured at a very low level at 20 L min1. This is
because ozone molecules are strongly influenced by the gas tem-
perature. The half-life of ozone changes rapidly with gas temper-
ature, and the half-life of ozone above 200  C are only a few seconds
(Batakliev et al., 2014). In case of DBD system, the ozone formation
reaction decreases rapidly as the temperature rises. (Pekarek.,
2003; Kogelschatz., 2003). Therefore, cooling of electrodes and
dielectrics is important in ozone production using DBD. The plasma
apparatus used in this study cooled the temperature down using air
injection as the reaction gas. However, the flow rate of 20 L min1 is
too low to sufficiently cool the electrode and the dielectric. A pre-
vious study (Ma et al., 2018) confirmed that the gas temperature
was measured at about 200  C under low flow conditions, resulting
in low ozone concentrations. In this paper, it is difficult to measure
the gas temperature because the plasma device is submerged in
water. However, referring to the water temperature rise as shown
in Fig. SM-2, the results indicate that high temperature gas is
injected into the water with the flow rate of 20 L min1. From this
Fig. 2. A) voltage and current waveform of multi-hole air dielectric barrier discharge (DBD) under different flow rates, and B) power consumption at different flow rates.
6 S. Ma et al. / Chemosphere 243 (2020) 125377
Fig. 3. Time-dependent changes under different air flow rates, of; A) concentration of
O3 after passing through the water, and B) H2O2 concentration generated by water
dissociation. Data are shown as the mean ± SEM (n ¼ 5), where SEM represents the
standard error of mean and n the measurement times.
point of view, the condition of 20 L min1 is not favorable for ozone
generation and can lead to deterioration of the durability of the
device. Therefore, it was confirmed that the flow rate of 25 L min1
is the optimized condition for ozone generation in this study.
Dissolved O3 concentration increased linearly by 0.02 mg L1
per min at 20 L min1, and 0.03 mg L1 per min at 25, 30, 35 and
40 L min1. Fig. 3B shows the time-dependent H2O2 concentration
in water; H2O2 is mainly produced either by the combination of an
OH OH pair or by combining with oxygen in the dissociation pro-
cess of water (Dodet et al., 2005); the generated plasma jet con-
tacting water (Fig. SM-1) contributes to the dissociation process of
water. After 20 min, the highest concentration of 6.2 mg L1 H2O2
was produced at a gas flow rate of 20 L min1, and 5.6 mg L1 was
produced at a gas flow rate of 25 L min1, i.e., the H2O2 concen-
tration decreases when the flow rate increases. As shown in Fig. 3B,
the concentration of hydrogen peroxide changes linearly with flow
rate, unlike ozone. This is because hydrogen peroxide produced in
the water is decomposed at a water temperature of about
100e280  C, thus it is hardly affected by temperature (Lin et al.,
1991). The decomposition of water in this study was achieved
mainly by the plasma plume in contact with the water surface as
shown in Fig. SM-1. Plasma plume is affected by electric field and
gas density. According to E n1, the plasma plume is strongly
generated under relatively low gas density conditions, which may
be considered to have an effect on water decomposition. E is electric
field and n is gas density. As a result, the amount of hydrogen
peroxide measured as the flow rate increases is as shown in Fig. 3B.
According to the results, we determined that the optimal flow rate
was 25 L min1 by comparing the O3 and H2O2 concentrations and
the power consumption.
Fig. SM-2 shows the real-time water temperature during the
plasma treatment (according to the flow rate), which increased by
4.5  C at the optimized condition of 25 L min1 and increased by
0.7  C at 40 L min1. Although this was a very low temperature rise
compared to the exposure time by the plasma in water, future
studies should enable the prevention of this increase in the water
temperature by changing the conditions or structure of the plasma
device.
3.4. Analysis of the optical emission spectra
Fig. 4 presents the emission spectra in the range of 200e900 nm
for the multi-hole DBD in water, operating at 7.133 kVrms and
25 L min1. The emission measurement was taken within the first
minute of the initiation of discharge. It is well known that the
spectra of light emitted by DBD in air are dominated by emission
bands originating from excited states of N2 molecules and Nþ 2 ions.
The majority of the distinct peaks obtained in the UV region cor-
responded to strong emissions from N2 (C3Pu -B3Pg) for the second
positive system, and Nþ 2 2
2 (B S -X S) for the first negative system
(Xiong et al., 2013; Han et al., 2017; Han et al., 2017). The absence of
atomic oxygen peaks can be explained based on the quenching of O
(3P) and O (5P) in air plasma (Walsh et al., 2010). Dave et al. (2012)
reported that quenching of atomic oxygen peaks by N2 and O2
during O3 generation is two orders of magnitude faster than the
radiative process of atomic oxygen 5S0e5P and 3S0e3P. Thus, no
atomic oxygen was observed in the emission spectrum.
3.5. Analysis of water properties and sterilization of E. coli in water
Fig. 5 shows the kill curves of E.coli treated by the multi-hole
DBD in water. The cultured E.coli was injected into the reactor
containing 500 mL of water, and treated for 10 min by O3 and
radicals generated from the multi-hole DBD. The vertical axis rep-
resents the logarithmic value of N N1 0 , where N0 denotes the con-
trol colony’s number of surviving bacteria (in units of CFU mL1).
The average control number was 1.5  104 CFU mL1 and the results
showed a Log (N N1 0 ) value of 7.204 after 10 min treatment. This
log scale means that >99.9% of bacteria were sterilized. As shown in
Fig. 5, we counted the E.coli colonies in the treated water for 5 days.
We found that E.coli slightly proliferated (to a Log (N N1 0 ) value
of 6.505) after 3 days of treatment, and the number increased (to
a Log (N N10 ) value of 4.727) after 5 days of treatment. Almost all
E.coli were eliminated by the multi-hole DBD; however, they
proliferated gradually over time due to their growth rate (Brock,
1971; Marr, 1991).
Fig. SM-3 illustrates the pH and conductivity changes with
increasing treatment time. As previously described, O3-based AOP
can be induced in wastewater treatment using DBD plasma to
achieve higher treatment efficiency. As a result of the O3 treatment
for 20 min, the pH increased from 6.98 to 8.51; Fig. SM-3 shows an
exponential increase in pH within 5 min of treatment followed by a
gradual stabilization. Gardoni et al. (2012) reported that the higher
the pH, the faster the decomposition of O3. They also defined the
 S. Ma et al. / Chemosphere 243 (2020) 125377
Fig. 4. Optical emission spectrum of multi-hole air dielectric barrier discharge (DBD).
Fig. 5. Disinfection curve of E.coli treated with O3 produced from the multi-hole air
dielectric barrier discharge (DBD). Data are shown as the mean ± SEM (n ¼ 5).
half-lives of O3 vs. pH according to when the O3 decomposition rate
increases exponentially versus pH increase, and directly as when
the half-life of O3 decreases versus pH. In the present study, it was
assumed that OH produced by the decomposition of O3 in water
and generated from the multi-hole DBD (Ma et al., 2018) contrib-
uted to the pH increase, which is expected to increase the waste-
water treatment efficiency.
On the other hand, there was minimal change in the conduc-
tivity during the O3 treatment, demonstrating that the conductive
ion or material was mostly unchanged in water, and hence did not
contribute considerably to the wastewater treatment as shown in
Fig. SM-3.
7
3.6. Analysis of seaweed washing wastewater by O3 treatment
As previously described, it was confirmed that a large amount of
active species (including O3) was generated from the multi-hole
DBD (Fig. 3A), and that they readily dissolved in water. Either
directly or indirectly, the generated active species affected organic
waste oxidation and bacterial inactivation (Malik et al., 2001). In
this regard, we measured turbidity, BOD, and COD as the most
important factors for the analysis of the wastewater purification
results following O3 treatment. Suspended solids (SS) were not
measured as samples were collected after filtering and almost no
suspended particles were evident, as previously noted. Fig. 6A
shows the turbidity changes of the seaweed washing wastewater
with O3 treatment time; the color of the wastewater was light red
and the turbidity was 14.0 NTU. The turbidity gradually decreased
with the plasma exposure time and was 5.8 NTU after 20 min
treatment.
Fig. 6Bshows the wastewater prior to treatment for comparison;
it can be seen that after 20 min of exposure by the multi-hole DBD,
the red color disappeared due to oxidation of contaminated organic
compounds in the wastewater.
Fig. 7 shows the BOD and COD concentrations with increasing
plasma exposure time. BOD is a measure of the amount of oxygen
that is required for bacteria to degrade the organic components
present in a sample under test conditions and COD is the total
oxygen consumption by a waste liquid sample under a standard test
condition. All oxidizable organic matter is reacted by a strong re-
agent. BOD and COD are both expressed as mg L1 O2. The BOD and
COD are the most common indicators of contaminated organic
compounds in water. In the wastewater treatment using plasma-
based AOP, organic substances are reduced due to oxidation by
generated oxidants, including O3 and OH. Therefore, COD and BOD
decrease as shown in Fig. 7. The initial BOD and COD of wastewater
were 49.3 mg L1 and 39.7 mg L1, respectively. After 20 min
treatment, the BOD and COD gradually decreased to 18.6 mg L1 (a
60% decrease) and 23.9 mg L1 (a 40% decrease), respectively.
8 S. Ma et al. / Chemosphere 243 (2020) 125377
Fig. 6. A) time-dependent turbidity of wastewater during O3 treatment. Data are
shown as the mean ± SEM (n ¼ 5). B) wastewater before and after treatment.
4. Conclusions
In this study, we proposed the multi-hole DBD system to purify
seaweed washing wastewater, which presents an environmental
issue. The multi-hole DBD was installed in the wastewater and
operated smoothly. Moreover, the results show that large amounts
of active species including O3 and H2O2 can be produced and be
injected directly into the wastewater, which enables highly efficient
wastewater purification. We determined the optimal operating
conditions based on O3 concentration, water temperature and po-
wer consumption. According to the optimized condition, 99.99% of
E. coli was eliminated following treatment for 20 min. We also
analyzed the change of water properties by plasma treatment and
found that the pH gradually increased with treatment time from
6.98 to 8.51.
This was a result of
OH generated by the decomposition of
water and O3. Thus, there was an advantage in that an O3-based
Fig. 7. Time-dependent biochemical oxygen demand (BOD) and chemical oxygen de-
mand (COD) of wastewater. Data are shown as the mean ± SEM (n ¼ 5).
AOP reaction could be additionally obtained. Consequently, we
confirmed that the turbidity, BOD and COD of seaweed washing
wastewater were improved. We expect that the developed multi-
hole DBD system could be effective to purify large volumes of
wastewater with higher concentrations. In the future, further
studies will be carried out to improve the efficiency of the treat-
ment by modifying the structure of the plasma system.
Authors contributions section
Sukhwal Ma: Term, Methodology, validation, investigation, re-
sources, writing-original draft, data curation. Kangil Kim: Concep-
tualization, Writing- Review & editing, supervision, project
administration, funding acquisition. Semin Chun: Investigation,
Resources. Se Youn Moon: Writing- Review & editing, supervision.
Yongcheol Hong: Conceptualization, Writing- Review & editing,
supervision, project administration, funding acquisition.
Acknowledgment
This work was supported by the R&D Program of Plasma
Convergence & Fundamental Research through the National Fusion
Research Institute of South Korea (NFRI) funded by the Government
as well as, in part, by the National Research Council of Science &
Technology (NST) grant from the Korea government (MSIP)
(No.CAP-18-08-KIMM).
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.chemosphere.2019.125377.
References
Andreozzi, R., Caprio, V., Insola, A., Marotta, R., 1999. Advanced oxidation processes
(AOP) for water purification and recovery. Catal. Today 53, 51e59.
Attri, P., Kim, Y.H., Park, D.H., Park, J.H., Hong, Y.J., Uhm, H.S., Kim, K.N., Fridman, A.,
Choi, E.H., 2015. Generation mechanism of hydroxyl radical species and its
lifetime prediction during the plasma-initiated ultraviolet(UV) photolysis. Sci.
Rep. 5, 9332.
Batakliev, T., Georgiev, V., Anachkov, M., Rakovsky, S., Zaikov, G.E., 2014. Ozone.
Decomposition. Interdiscip. Toxicol. 7, 47e59.
Brock, T.D., 1971. Microbial growth rates in nature. Bacteriol. Rev. 35, 39e58.
Chen, Y., Peng, R., Shen, T., Tong, S., Ma, C.A., 2015. Promising ozone-based advanced
 S. Ma et al. / Chemosphere 243 (2020) 125377
oxidation process for effective generation of hydroxyl radical in acidic solution.
Separ. Purif. Technol. 151, 269e275.
Clayton, G.E., Thorn, R.M.S., Reynolds, D.M., 2019. Comparison of trihalomethane
formation using chlorine-based disinfectants within a model system; applica-
tions within point-of use drinking water treatment. Front. Environ. Sci. 22,
1e13.
Collivignarelli, M.C., Pedrazzani, R., Sorlini, S., Abba, A., Bertanza, G., 2017. H2O2
based oxidation processes for the treatment of real high strength aqueous
wastes. Sustainability 9, 1e14.
Dave, H., Ledwani, L., Chandwani, N., Kikani, P., Desai, B., Chowdhuri, M.B.,
Nema, S.K., 2012. Use of dielectric barrier discharge in air for surface modifi-
cation of polyester substrate to confer durable wettability and enhance dye
uptake with natural dye eco-alizarin. Compos. Interfac. 19, 1e11.
Deng, Y., Zhao, R., 2015. Advanced oxidation processes (AOPs) in wastewater
treatment. Curr. Pollut. Rep. 1, 167e176.
Dodet, B., Odic, E., Goldman, A., Goldman, M., Renard, D., 2005. Hydrogen peroxide
formation by discharge in argon/water vapor mixtures at atmospheric pressure.
J. Adv. Oxid. Technol. 8, 91e97.
Gardoni, D., Vailti, A., Canziani, R., 2012. Decay of ozone in water: a review. Ozone
Sci. Eng. 34, 233e242.
Gligorovski, S., Strekowski, R., Barbati, S., Vione, D., 2015. Environmental implica-
tions of hydroxyl radical (∙OH). Chem. Rev. 115, 13051e13092.
Grimm, J., Bessarabov, D., Sanderson, R., 1998. Review of electro-assisted methods
for water purification. Desalination 115, 285e294.
Han, D.S., Cho, J.H., Lee, R.H., Bang, W., Park, K.H., Kim, M.S., Shim, J.H., Chae, J.I.,
Moon, S.Y., 2017. Antitumorigenic effect of atmospheric-pressure dielectric
barrier discharge on human colorectal cancer cells via regulation of sp1 tran-
scription factor. Sci. Rep. 7, 43081.
Han, D.S., Moon, S.Y., 2015. Development of superhydrophobic surface on glass
substrate by multi-step atmospheric pressure plasma treatment. Thin Solid
Films 587, 34e38.
Han, D.S., Moon, S.Y., 2017. Development of vacuum ultraviolet emission source
using a cone-shaped O2/He hollow cathode discharge. J. Nanosci. Nanotechnol.
17, 8480e8484.
Haseena, M., Malik, M.F., Javed, A., Arshad, S., Asif, N., Zulfiqar, S., Hanif, J., 2017.
Water pollution and human health. Environ. Risk. Assess. Remediat. 1, 16e19.
Hassaan, M.A., Nemr, A.E., 2017. Advanced oxidation processes for textile waste-
water treatment. Int. J. Photoch. Photobio. 1, 27e35.
Hong, Y.C., Ma, S.H., Kim, K.I., Shin, Y.W., 2019. Multihole dielectric barrier discharge
with asymmetric electrode arrangement in water and application to
9
sterilization of aqua pathogens. Chem. Eng. J. 374, 133e143.
Kogelschatz, U., 2003. Dielectric-barrier discharges: their history, discharge physics,
and industrial applications. Plasma Chem. Plasma Process. 23, 1e46.
Lalwani, P.K., Devadasan, M.D., 2013. Reduction of cod and bod by oxidation: a cetp
case study. Int. J. Eng. Res. Afr. 3, 108e112.
Lee, S.J., Ma, S.H., Hong, Y.C., Choi, M.C., 2018. Effects of Pulsed and Continuous
Wave Discharges of Underwater Plasma on Escherichia coli, vol. 193,
pp. 351e357.
Lin, C.C., Smith, F.R., Ichikawa, N., Bada, T., Itow, M., 1991. Decomposition of
hydrogen peroxide in aqueous solutions at elevated temperatures. Int. J. Chem.
Kinet. 23, 971e987.
Ma, S.H., Kim, K.I., Huh, J.Y., Hong, Y.C., 2017. Characteristics of microdischarge
plasma jet in water and its application to water purification by bacterial inac-
tivation. Separ. Purif. Technol. 188, 147e154.
Ma, S.H., Kim, K.I., Huh, J.Y., Kim, D.E., Lee, S.J., Hong, Y.C., 2018. Regeneration and
purification of water-soluble cutting fluid through ozone treatment using an air
dielectric barrier discharge. Separ. Purif. Technol. 199, 289e297.
Ma, S.H., Kim, K.I., Lee, S.J., Moon, S.Y., Hong, Y.C., 2018. Effects of a porous dielectric
in atmospheric-pressure plasma jets submerged in water, 083519 Phys. Plasmas
25, 1e7.
Malik, M.A., Ghaffar, A., Malik, S.A., 2001. Water purification by electrical discharges.
Plasma Sources Sci. Technol. 10, 82e91.
Marr, A.G., 1991. Growth rate of escherichia coli. Microbiol. Rev. 55, 316e333.
Masoud, M.S., Ismall, A.M., El-Hoshy, M.M., 2019. Kinetics and thermodynamics of
the formation of trihalomethanes. Appl. Water. Sci. 9, 1e6.
Owa, F.W., 2014. Water pollution: sources, effects, control and management. Int.
Lett. Nat. Sci. 3, 1e6.
Pekarek, S., 2003. Non-thermal plasma ozone generation. Acta. Polytech. 43, 47e51.
Rajasekaran, P., Mertmann, P., Bibinov, N., Wandke, D., Viol, W., Awakowicz, P., 2009.
DBD Plasma source operated in single-filamentary mode for therapeutic use in
dermatology, 225201 J. Phys. D Appl. Phys. 42, 1e8.
Suyasa, B., Dwijani, W., 2015. Biosystem treatment approach for seaweed processing
wastewater. J. Environ. Waste. Manag. 2, 59e62.
Walsh, J.L., Liu, D.X., Iza, F., Rong, M.Z., Kong, M.G., 2010. Contrasting characteristics
of sub-microsecond pulsed atmospheric air and atmospheric pressure
heliumeoxygen glow discharges. J. Phys. D Appl. Phys. 43, 032001.
Xiong, Q., Nikiforov, A.Y., Gonzalez, M.A., Leys, C., Lu, X.P., 2013. Characterization of
an atmospheric helium plasma jet by relative and absolute optical emission
spectroscopy. Plasma Sources Sci. Technol. 22, 015011.