water research 43 (2009) 895–901

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

The effect of electrode material on the generation of

oxidants and microbial inactivation in the electrochemical
disinfection processes

Joonseon Jeonga, Choonsoo Kimb, Jeyong Yoonb,*

a
Urban Water Research Center, Department of Civil and Environmental Engineering, University of California, Irvine,
Irvine, CA 92697-2175, United States
b
School of Chemical and Biological Engineering, College of Engineering, Seoul National University, Seoul 151-742, Republic of Korea

article info abstract

Article history: Electrochemical disinfection has gained increasing attention as an alternative for
Received 25 August 2008 conventional drinking water treatment due to its high effectiveness and environmental
Received in revised form compatibility. The most common method of electrochemical disinfection is the use of
12 November 2008 electro-generated oxidants, such as active chlorine and reactive oxygen species, as disin-
Accepted 20 November 2008 fectants. This study examined the role of electrode material on the generation of oxidants,
Published online 6 December 2008 and elucidated the different reaction pathways for generating individual oxidants by
employing boron-doped diamond (BDD), Ti/RuO2, Ti/IrO2, Ti/Pt–IrO2, and Pt as anode

Keywords: materials. The efficiency of OH production, as determined by para-chlorobenzoic acid
Electrochemical disinfection ( pCBA) degradation, was in the order of BDD [ Ti/RuO2 z Pt. No significant production of
 
Electrode material OH was observed at Ti/IrO2 and Ti/Pt–IrO2. The OH was found to play a key role in O3
Reactive oxygen species generation at BDD, but not at the other electrodes. The production of active chlorine was in
Active chlorine the order of Ti/IrO2 > Ti/RuO2 > Ti/Pt–IrO2 > BDD > Pt. The large difference in this order
Hydroxyl radical from that of ROS was attributed to the difference in the electrocatalytic activity of each
electrode material toward the production of active chlorine, as evidenced by linear sweep
voltammetry (LSV) measurements. In addition, the characteristics of microbial inactivation
as a function of electrode material were examined under the presence of an inert
electrolyte, using Escherichia coli as an indicator microorganism.
ª 2008 Elsevier Ltd. All rights reserved.

1. Introduction involving the formation of disinfection by-products such as

Electrochemical disinfection has gained increasing attention
as an alternative for conventional drinking water treatment
because it is environmentally friendly, amendable to auto-
mation, inexpensive, easily operated and is known to inacti-
vate a wide variety of microorganisms from bacteria to viruses
and algae (Diao et al., 2003; Gutknecht et al., 1981; Shimada
and Shimahara, 1982), even though the potential problems
ClO2, ClO 
3 , and ClO4 have to be carefully examined before any
real application (Bergmann and Rollin, 2007).
The most popular method of electrochemical disinfection
is the use of electro-generated oxidants. During electrolysis,
a variety of oxidants can be produced depending on the
reaction condition. The active chlorine species such as Cl2,
HOCl and OCl have been recognized as key oxidants
responsible for inactivating microbial cells in electrochemical

* Corresponding author. Tel.: þ82 2 880 8927; fax: þ82 2 876 8911.
E-mail address: jeyong@snu.ac.kr (J. Yoon).
0043-1354/$ – see front matter ª 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2008.11.033
896 water research 43 (2009) 895–901
disinfection with saline water due to their outstanding dis-
infecting activity (Diao et al., 2003). More recently, the role of

reactive oxygen species (ROS), including OH, O3, H2O2, and
O
 
2 , which are produced by water oxidation, has been
underlined by their higher disinfecting activity of electro-
chemical disinfection compared to that of electro-chlorina-
tion in saline and even in the chlorine-free water (Jeong et al.,
2006). In addition, several other oxidants such as S2O28 , C2O6
2
4
and P2O8 can possibly be generated from the oxidation of
SO2 2 3
4 , CO3 and PO4 , respectively, when a specific electrode
material like boron-doped diamond (BDD) is used (Canizares
et al., 2005; Furuta et al., 2004).
The generation of such oxidants strongly depends on the
several key reaction parameters, including electrode material,
electrolyte composition, applied current (or voltage), pH,
temperature and type of electrolysis, of which the electrode
material has been regarded as one of the most important
parameters governing the species and yield of generated
oxidants (Martinez-Huitle and Brillas, 2008). For example, the
electrode materials for generating ROS must have a high
oxygen overpotential, because otherwise a large portion of the
applied current will be wasted to produce oxygen as a side
reaction, thereby dropping the efficiency of the electro-
chemical process. Several electrode materials such as PbO2,
SnO2-based DSA and BDD have been reported to exhibit high
oxygen overpotential. These electrode materials are able to
hinder the thermodynamically favoured oxygen evolution
reaction, thus improving the ROS generation efficiency (Mal-
pass et al., 2006; Panizza and Cerisola, 2005). In the case of
oxidants other than ROS, the anode materials are required to
have a high electrocatalytic activity toward the production of
the desired oxidants (Szpyrkowics et al., 2005).
In spite of many studies examining the influence of elec-
trode material on the degradation of chemical pollutants,
relatively few reports have investigated the disinfection of
microorganisms. Most previous studies on electrochemical
disinfection generating oxidants were limited to a single
electrode material without comparison to other electrodes and
were therefore unable to demonstrate the effect of electrode
material on the generation of the oxidants comparatively
(Martinez-Huitle and Brillas, 2008).
Therefore, the goal of this study was to investigate the
effect of electrode material on the generation of several key
oxidants, including ROS and active chlorine, and to clarify the
different reaction pathways for generating each oxidant as
a function of the following anode electrode materials: BDD,
Ti/IrO2, Ti/RuO2, Ti/Pt–IrO2 and Pt, which are widely applied
for electrochemical disinfection.
2. Materials and methods
2.1. Chemicals and analyses
All chemicals were of reagent grade and used without further
purification. Potassium dihydrogen phosphate (KH2PO4),
sodium dihydrogen phosphate (NaH2PO4), sodium sulfate
(Na2SO4), sodium bicarbonate (NaHCO3), sodium chloride
(NaCl), tert-butyl alcohol (t-BuOH), para-chlorobenzoic acid
( pCBA) and sodium thiosulfate (Na2S2O3) were purchased from
Aldrich Co. (USA). All supporting electrolytes and stock solu-
tions were prepared with distilled and deionized water from
the Millipore purification system (Barnstead NANO Pure, USA).
ROS, such as O3, H2O2, and OH, were identified by direct or
indirect methods. The concentration of O3 was measured
using the indigo method with an UV–vis spectrophotometer
(Hewlett-Packard 8453, USA) and 10 cm cuvettes. This method
is based on the quantitative decolorization of indigo trisulfo-
nate as a result of its reaction with O3, which is observed at
600 nm and whose detection limit is about 0.01 mg/L (Bader
and Hoigné, 1982). All experiments for O3 generation were
conducted at low temperature (10  C), since our previous
study revealed that the electrochemical generation of O3 is
strongly dependent upon the temperature of electrolytic
solution, such that a higher O3 concentration is achieved at
lower temperature (Jeong et al., 2006).
The formation of H2O2 was determined by a colorimetric
method using copper(II) ions and 2,9-dimethyl-1,10-phenan-
throline (DMP) at 454 nm (Ayad et al., 1998). Our preliminary
tests and results reported in the literature confirmed that
these analytical methods have no significant interference
toward each other under the experimental conditions in this
study (Ayad et al., 1998; Bader and Hoigné, 1982). For the
analysis, an appropriate amount of sample taken from the
electrolytic solution was injected into the analytical reagents
as quickly as possible to minimize the decay of oxidants.
Two different methods were employed to assess the

production of OH, as its short lifetime makes such assess-
ment difficult. The first method involved the use of t-BuOH,

a well-known OH scavenger. An excess amount of t-BuOH

(0.05 M) was introduced to verify the presence of OH. The
second method compared the degradation rate of pCBA,
which has been widely employed as a probe compound for

OH production. The concentration of pCBA was analyzed by
HPLC with an XTerra Rp-18 reverse-phase column (5 mm,
150 mm  2.1 mm) and Gilson 151 UV/vis detector at 230 nm.
A solvent mixture of 35% acetonitrile/65% water containing
40 mM phosphate buffer was used as the mobile phase.
The concentration of active chlorine and the other
oxidants generated was determined as mg/L (as total Cl2) by
N,N-diethyl-p-phenylenediamine (DPD) colorimetric method.
DPD is oxidized to form a red-violet product, the concentra-
tion of which was measured by spectrophotometer (DR/2010,
HACH Co., Loveland, USA) at 530 nm (APHA, 2005).
2.2. Preparation and analysis of microorganisms
Escherichia coli (ATCC 8739) was employed as an indicator
bacterium in all disinfection experiments. The stock suspen-
sion of E. coli was prepared following the procedures described
elsewhere (Cho et al., 2004). The initial populations of E. coli for
each disinfection experiment ranged approximately from 105
to 106 CFU/mL. During the experiments, 1 mL of suspension
was withdrawn at each sampling time and was immediately
quenched with excess Na2S2O3 (10 mM) to eliminate the
residual disinfectants in the sample solution. Then, the
sample was diluted accordingly, depending on the initial
number of viable cells. Three replicates of the diluted and
undiluted 0.1 mL suspensions were used for counting by the
spread plate method with nutrient agar grown at 37  C for 24 h
 water research 43 (2009) 895–901
and showed good reproducibility within a standard deviation
of 10%.
2.3. Electrochemical cells and apparatus
The experimental measurement of the oxidant formation and
microbial inactivation by electrolysis was carried out in
a divided cell consisting of a 100-mL anodic compartment and
a 50-mL cathodic compartment with a Nafion 117 cation
exchange membrane (Aldrich Co., USA) as a separator. Five
different electrode materials, a BDD film on a niobium
substrate (Nb/BDD; CONDIAS GmbH, Germany), Ti/IrO2
(Samsung DSA Co., Korea), Ti/RuO2 (Samsung DSA Co., Korea),
Ti/Pt–IrO2 (Magneto Co., The Netherlands), and Pt (Aldrich Co.,
USA), were employed as the anode with a working geometric
area of 6 cm2 (2 cm  3 cm). The stainless steel (SUS316)
cathode was the same size. The electrodes, prepared by
connecting pieces of the tested materials with a copper wire
and sealing them in an epoxy resin, were installed in parallel
in each compartment and kept 20 mm apart by a Teflon cell
top. A laboratory DC power supply (UDP-150I, Unicorn Tech
Co., Korea) was employed to provide electric current.
In order to investigate the electrocatalytic activities of
the electrode materials, linear sweep voltammetry (LSV)
measurements were performed in a conventional, three-
electrode cell using a computer-controlled potentiostat/
galvanostat (PARSTAT 2273A, Princeton Applied Research,
USA). The five selected materials were used as the working
electrode, Pt wire as the counter electrode, and Ag/AgCl
(saturated NaCl) as the reference electrode. All potential values
quoted in this study were with reference to the Ag/AgCl
electrode. The apparent exposed area of the working electrode
was 0.177 cm2. The measurements were conducted in a base
electrolyte containing 0.1 M KH2PO4, to which 0.1 M NaCl was
added with different electrode materials.
2.4. Experimental procedures
Prior to the experiments, the electrodes were sonicated for
10 min (60 kHz and 500 W, Powersonic 410, Whasin Tech Co.,
Korea) to remove any contaminants, and then washed several
times with sterile water. For the disinfection study, any
residual microbial cells adsorbed on the electrodes and
reactor surface were eliminated by irradiation with a suffi-
cient UV intensity for 5 min (IT ¼ 225 mJ/cm2, 6 W  2 low
pressure mercury lamp, Philips Co., The Netherlands; USEPA,
2003). The electrodes were then immersed in the electrolytic
solution and the resultant electrochemical cell was placed in
a thermostatic water bath (Model RW-2025G, JEIO TECH,
Korea). An inert electrolyte, KH2PO4, was used to investigate
the effect of the electrode material on ROS generation in order
to minimize the formation of oxidants other than ROS. In the
experiment for generating oxidants other than ROS, the
sodium salts of the anions of interest, including NaCl,
NaH2PO4, Na2SO4 and NaHCO3, were employed as the elec-
trolyte to eliminate any influence from the cation.
Unless otherwise stated, the pH and temperature were 7.1
and 25  C, respectively. For disinfection experiments, the
appropriate amount of E. coli stock was suspended in the
electrolytic solution of the anode compartment and stirred for
897
at least 10 min prior to electrolysis. The electrolysis was begun
by applying a constant current density ranging from 17 to
167 mA/cm2. The samples were withdrawn at timed intervals
to analyze the concentrations of oxidants or to count the
number of viable cells.
3. Results and discussion
3.1. Production of reactive oxygen species (ROS)
The most common oxidants in the electrochemical disinfec-
tion processes under the presence of inert electrolyte are the
ROS produced directly from the oxidation of water molecules

involving OH, O3, and H2O2. First of all, Fig. 1 shows the effect

of electrode material on the OH production, as accessed by

the degradation of the OH probe compound ([pCBA]0 ¼
1.9  106 M) in 0.2 M KH2PO4 that was employed as an inert
electrolyte. Even though the possible formation of HPO

4 or
PO2

4 by the oxidation of the KH2PO4 that was used as elec-
trolyte cannot be totally excluded, the pCBA degradation
caused by such radicals was assumed to be negligible due to

their weaker oxidizing activity than OH (Neta et al., 1988).
The rate of pCBA degradation was in the order of BDD [ Ti/
RuO2 z Pt, while no significant pCBA degradation was
observed by Ti/IrO2 and Ti/Pt–IrO2 during the 30 min elec-
trolysis. The pseudo-first order kinetics of pCBA degradation
(inset of Fig. 1) gave apparent rate constants of 1.33  101,
1.15  102 and 7.4  103 s1 for BDD, Ti/RuO2 and Pt,

respectively. As a result, the production of OH by BDD was
about 10 times higher than that by Ti/RuO2 and Pt.

The large difference in the OH production shown in Fig. 1
was attributed to the different nature of the electrode
materials employed. It has been suggested that the reaction

pathway of OH after being produced at the electrode surface
(M) (reaction (1)) strongly depends on the chemical interaction

between the electrode surface and OH. Two classes of

Fig. 1 – Effect of electrode material on the production of OH
as accessed by the degradation of pCBA (i [ 100 mA/cm2,
[pCBA]0 [ 1.9 3 10L6 M, [KH2PO4]0 [ 0.2 M, pH [ 7.1,
T [ 25 8C).
898 water research 43 (2009) 895–901

electrode materials can be defined as non-active and active

electrode (Marselli et al., 2003).
In a non-active electrode, the electrode surface acts only as

an electron sink and has no interaction with OH. Thus, the

OH on a non-active electrode preferentially diffuses to the
bulk solution to react with any oxidizable species (R) adsorbed
on or in the vicinity of electrode surface (reaction (2)). The
BDD, PbO2, and SnO2-based DSAs have been reported as the
non-active electrode (Canizares et al., 2002). The finding that
the pCBA degradation was much faster at BDD than that at the
other electrodes, as shown in Fig. 1, can be explained by the

non-active characteristic of BDD, which allows OH to partic-
ipate in degrading pCBA in the bulk solution.

However, in an active electrode the OH favorably interacts
with the electrode surface, resulting in the transformation into
a higher oxide surface (MO) (reaction (3)). The MO is also able to
participate in the oxidation reaction (reaction (4)), but its
oxidizing capacity has been reported to be weaker than that of

OH (Comninellis, 1994). The materials like Pt-, IrO2- and RuO2- Fig. 2 – Effect of electrode material on the generation of O3
based DSAs and stainless steel belong to the active electrode (i [ 167 mA/cm2, [KH2PO4]0 [ 0.2 M, [t-BuOH]0 [ 0 or
(Canizares et al., 2002). The slower pCBA degradation by the 0.05 M, pH [ 7.1, T [ 10 8C).
three metal oxide electrodes tested in this study was attrib-

uted to the weaker oxidation capacity of MO than that of OH.
The difference in the pCBA degradation observed between the
active electrodes tested in this study suggests that the MO However, the preference in the generation of O3 for the
reactivity varies according to the electrode material. electrode materials other than BDD was not in agreement
Oxygen evolution is accomplished either by the further 
with that of OH observed in Fig. 1, suggesting that alternative

oxidation of OH (reaction (5)) or by the release of oxygen from 
reaction pathways other than by OH may be involved in the
MO (reaction (6)). generation of O3.
The effect of electrode material on the formation of H2O2
M þ H2O / M( OH) þ Hþ þ e

(1)
was also investigated by measuring the concentration of H2O2
 after 5 min of electrolysis at 100 mA/cm2. The H2O2 formation
M( OH) þ R / M þ Products (2)
was 7  105 M and 4  105 M by BDD and Ti/RuO2, respec-
M( OH) / MO þ Hþ þ e

(3) tively, but was insignificant in the electrolysis employing the
other electrodes (data not shown). This result was explained
MO þ R / M þ Products (4) in terms of the different activities of electrode materials

toward OH production, as shown in Fig. 1, since the mecha-
þ  

M( OH) / 1/2 O2 þ M þ H þ e (5) nisms of H2O2 formation is the combination of two OHs
(reaction (9)) (Jeong et al., 2006).
MO / 1/2 O2 þ M (6)
 
OH þ OH / H2O2 (9)
Fig. 2 shows the influence of electrode material on the O3
generation as determined by measuring the concentration of O3
accumulated during the electrolysis at 10  C. In the case of BDD,
the O3 concentration gradually increased to a maximum of 3.2. Production of active chlorine
1.1 mg/L for 20 min of electrolysis, compared to below 0.1 mg/L
in the other electrodes. The slight reduction in the rate of O3 Fig. 3 shows the generation of active chlorine with different
generation with time was attributed to the growing decompo- electrode materials for 10 min of electrolysis in 1.7  102 M
sition of O3 with increasing O3 concentration. BDD exhibited NaCl solution. The concentration of active chlorine increased
a much higher efficiency in O3 generation than the other elec- linearly with elapsed time, and its rate was in the order of
trodes, and the addition of excess t-BuOH (0.05 M) remarkably Ti/IrO2 > Ti/RuO2 > Ti/Pt–IrO2 > BDD > Pt, which agreed with
inhibited the formation of O3 at BDD, which confirmed the key a previous study (Kraft et al., 1999). This finding clearly

role played by OH in the generation of O3 at BDD. According to demonstrated that the electrode materials employed signifi-
one proposed mechanism, O3 can be generated by the reaction cantly affected the efficiency of chlorine evolution.

of O2 with O, which is the one-electron oxidation form of OH

To better understand the different efficiencies toward the
(reactions (7) and (8)) (Michaud et al., 2003). evolution of active chlorine with the five electrode materials
shown in Fig. 3, the LSV measurement of the five electrode
OH / O þ Hþ þ e
 
(7) materials employed in this study was carried out in the base
electrolyte (0.2 M KH2PO4) alone and those including 0.1 M

O þ O2 / O3 (8) NaCl, and the difference of current densities at each anode
 water research 43 (2009) 895–901
Fig. 3 – Effect of electrode material on the evolution of
chlorine (i [ 17 mA/cm2, [NaCl]0 [ 0.017 M, T [ 25 8C).
potential is presented in Fig. 4. At a given anode potential, the
total current densities obtained after adding NaCl (IT) are
always higher than those in the absence of NaCl (IB) with the
five electrode materials examined, since the current for
chlorine evolution is added to those for oxygen evolution in
the presence of NaCl. The order of the current difference
(IT  IB) magnitude in Fig. 4 agrees well with that of the chlo-
rine evolution measured in Fig. 3, indicating that the differ-
ence in the evolution of active chlorine with electrode
materials can be attributed to the different electrocatalytic
activity toward the reaction of chlorine evolution.
Fig. 4 – Comparison of voltammetric currents associated
with the evolution of chlorine for five electrode materials
(IT: current densities measured after adding NaCl, IB:
current densities measured in the absence of NaCl,
[KH2PO4]0 [ 0.2 M as base electrolyte, [NaCl]0 [ 0 or 0.1 M,
scan rate [ 50 mV/s).
899
According to the previous research (Trasatti, 1987), the
evolution of active chlorine is accomplished by direct oxida-
tion of Cl ion at the surface of metal oxide electrode (M)
(reactions (10) and (11)).
M þ Cl / M( Clad) þ e

(10)

2M( Clad) / 2M þ Cl2 (11)
Pt exhibited the lowest electrocatalytic activity toward the
evolution of active chlorine, as observed by LSV in Fig. 4, which
is consistent with the result of Fig. 3 in that the lowest
concentration of active chlorine measured among the tested
electrodes was for Pt. On the other hand, in the case of BDD,
a certain level of active chlorine was generated even though
the current measured at BDD was comparable to that at Pt and
much lower than that at the three metal oxide electrodes,
implying that reaction pathways other than direct oxidation of
Cl may be involved in the evolution of active chlorine at BDD.
The most plausible mechanism would be the indirect oxida-
tion of Cl mediated by OH because BDD exhibited an excel-


lent efficiency in the production of OH, as shown in Fig. 1.

To confirm this hypothesis, the possible role of OH on the
evolution of active chlorine taking place in the electrolysis
was further investigated separately at two electrodes (BDD
and Ti/IrO2) which had been determined in this study as the

most efficient electrode materials for producing OH and
active chlorine, respectively (Figs. 1 and 3). Fig. 5 shows the
different behaviors for chlorine evolution by BDD and Ti/IrO2

with and without adding excess OH scavenger ([t-BuOH]0 ¼
0.05 M). At BDD, the evolution of chlorine was completely
inhibited by t-BuOH addition, while no significant effect was

observed at Ti/IrO2, indicating that OH plays an important
role at BDD, whereas at Ti/IrO2 most of the chlorine is gener-
ated by the direct oxidation of Cl ions rather than by the

mediation by OH.

Fig. 5 – Effect of OH scavenger on the evolution of chlorine
with Ti/IrO2 and BDD (i [ 17 mA/cm2, [NaCl]0 [ 0.01 M,
[t-BuOH]0 [ 0 or 0.05 M, T [ 25 8C).
900 water research 43 (2009) 895–901

3.3. Production of other oxidants

The production of oxidants other than ROS and active chlo-

rine, such as S2O2 2 4
8 , C2O6 , and P2O8 , was also investigated
because the importance of these oxidants in the chlorine-free
disinfection process has been frequently reported (Furuta
et al., 2004). Table 1 summarizes the formation of such
oxidants (in mg/L as total Cl2) accumulated for 5 min of elec-
trolysis at constant current (100 mA/cm2) with Na2SO4,
NaHCO3 and NaH2PO4 as the electrolyte, which is compared in
parallel with that of active chlorine under the same experi-
mental condition. The electrolyte concentrations were set as
high as 0.1 M in order to obtain a high concentration of
oxidants based on the analytical problems at low concentra-
tions of these oxidants (Bergmann and Rollin, 2007), even
though the concentration of such electrolytes in natural water
is much lower than that used in this study.
As a result, the formation of active chlorine is always
higher than that of other oxidants in all electrode materials
Fig. 6 – Effect of electrode material on the inactivation of
examined, which may be attributed to both the higher elec-
E. coli without and with t-BuOH in the electrolyte of KH2PO4
trocatalytic activity of electrode materials toward Cl ions
(i [ 100 mA/cm2, electrolysis time [ 3 min,
than to the other anions and the longer lifetime of active
[KH2PO4]0 [ 0.2 M, [t-BuOH]0 [ 0 or 0.05 M, pH [ 7.1,
chlorine than that of the other oxidants produced. In the
T [ 25 8C).
presence of electrolytes other than NaCl, the concentrations of
oxidants formed at BDD were considerably higher than those
at the other electrode materials, which is consistent with the
 E. coli at BDD, which agrees well with the result of our previous
tendency for OH production. This implies that the oxidants
study (Jeong et al., 2006). Similar behavior in the influence of
S2O2 2
8 , C2O6 , and P2O8
4
are likely to be formed by indirect
 t-BuOH was observed at Ti/RuO2, indicating the main effect
oxidation mediated by OH rather than by direct oxidation. 
of OH in the inactivation at Ti/RuO2, even though the level of
inactivation at Ti/RuO2 was much lower than that at BDD. The
large difference in the level of inactivation between at BDD and
3.4. Inactivation of microorganisms 
Ti/RuO2 was ascribed to the different production of OH, which
is supported by the result of Fig. 1.
Fig. 6 shows the effect of electrode materials on the inactiva-
In the case of Pt, an approximately 1.3 log inactivation of E.
tion of E. coli during the electrolysis for 3 min in 0.2 M of
coli was achieved for 3 min without t-BuOH, which was much
KH2PO4 electrolyte in the absence and presence of t-BuOH 
 higher than that observed at Ti/RuO2 as the production of OH
(0.05 M), in order to assess the role of OH on the inactivation of
at Pt was comparable with that at Ti/RuO2. In the presence of t-
microorganisms with five different electrode materials. The
BuOH, the inactivation of E. coli at Pt was partially reduced but
preliminary test confirmed that electrolyzed water resulting
not completely inhibited. These observations indicated the
from the electrolysis of KH2PO4 for 3 min had no residual 
involvement of OH, but not as the solo species responsible for
disinfecting activity toward E. coli.
the inactivation of E. coli. The most probable cause for the
The highest reduction in the inactivation of E. coli by adding
inactivation of E. coli at the Pt anode shown in Fig. 6 is the
t-BuOH was found at BDD, as an approximately 2.4 log inacti-
direct electron transfer reaction of E. coli adsorbed onto the Pt
vation of E. coli, accomplished after 3 min without t-BuOH, was
anode surface. The possibility of this reaction has been
mostly inhibited by the addition of 0.05 M t-BuOH. This result
 already reported in our previous study (Jeong et al., 2007), and
indicated that OH is mainly responsible for the inactivation of
presumably induces the inactivation of E. coli without being
influenced by the t-BuOH addition.
The inactivation of E. coli at other metal oxide electrodes
Table 1 – Comparison of oxidant formation (in mg/L as (Ti/Pt–IrO2, Ti/IrO2) was not significant, regardless of t-BuOH
total Cl2) with different electrode materials and addition. This was attributed to the extremely low production
electrolytes ([Electrolyte]0 [ 0.1 M, i [ 100 mA/cm2, 
of OH, which agrees with the result of Fig. 1.
electrolysis time [ 5 min).
Electrode Electrolyte

NaCl Na2SO4 NaHCO3 NaH2PO4

4. Conclusions
BDD 219 26.0 1.46 1.24
Ti/RuO2 288 1.02 0.01 0.34 This study examined the effect of electrode material on the
Ti/IrO2 333 0.12 0.33 0.24
generation of several key oxidants, including ROS and active
Ti/Pt-IrO2 240 0.05 0.22 0.31
chlorine, which are mostly available in electrochemical
Pt 163 0.03 0.32 0.26
disinfection, and clarified the different reaction pathways via
 water research 43 (2009) 895–901
which each oxidant is generated depending on the electrode
material used. The following conclusions were made based on
the study results.

1. The production of OH at BDD was much higher than that at

Ti/RuO2 and Pt, and no significant production of OH was
found at Ti/IrO2 and Ti/Pt–IrO2.
2. In the generation of O3, BDD was much more efficient than

the other electrodes, and OH played a key role in O3
generation at BDD. However, the preference in O3 genera-
tion for the electrode materials other than BDD was not in

agreement with that of OH, indicating the operation of

alternative reaction pathways other than by OH.
3. The production of active chlorine as a function of electrode
material followed the order of Ti/IrO2 > Ti/RuO2 > Ti/Pt–
IrO2 > BDD > Pt, which was markedly different from that of
ROS. This difference was attributed to the difference in the
electrocatalytic activity of each electrode material toward
the production of active chlorine, as evidenced by LSV
measurements.
4. In the disinfection study conducted under the presence of
inert electrolyte (KH2PO4), the inactivation of E. coli at BDD

and Pt was mainly achieved by the reaction of OH and the
direct electron transfer reaction, respectively.
Acknowledgements
This work was supported by the Korea Research Foundation
Grant funded by the Korean Government (MOEHRD)
(KRF-2006-311-D00406).
references
APHA, 2005. Standard Methods for the Examination of Water and
Wastewater, 21st ed. American Public Health Association,
Washington, DC.
Ayad, N.B., Johnson, G.R.A., Nazhat, N.B., Saadalla-Nazhat, R.A.,
1998. A simple spectrophotometric determination of hydrogen
peroxide at low concentration in aqueous solution. Anal.
Chim. Acta 204, 349–353.
Bader, H., Hoigné, J., 1982. Determination of ozone in water by the
indigo method: a submitted standard method. Ozone: Sci. Eng.
4, 169–176.
Bergmann, M.E.H., Rollin, J., 2007. Product and by-product
formation in laboratory studies on disinfection electrolysis of
water using boron-doped diamond anodes. Catal. Today 124
(3–4), 198–203.
Canizares, P., Larrondo, F., Lobato, J., Rodrigo, M.A., Saez, C., 2005.
Electrochemical synthesis of peroxodiphosphate using boron-
doped diamond anodes. J. Electrochem. Soc. 152 (11), D191–D196.
Canizares, P., Martinez, F., Diaz, M., Garcia-Gomez, J., Rodrigo, M.A.,
2002. Electrochemical oxidation of aqueous phenol wastes
901
using active and nonactive electrodes. J. Electrochem. Soc. 149
(8), D118–D124.
Cho, M., Lee, Y., Chung, H., Yoon, J., 2004. Inactivation of
Escherichia coli by photochemical reaction of ferrioxalate at
slightly acidic and near-neutral pHs. Appl. Environ. Microbiol.
70 (2), 1129–1134.
Comninellis, C., 1994. Electrocatalysis in the electrochemical
conversion/combustion of organic pollutants for waste water
treatment. Electrochim. Acta 39, 1857–1862.
Diao, H.F., Li, X.Y., Gu, J.D., Shi, H.C., Xie, Z.M., 2003. Electron
microscopic investigation of the bactericidal action of
electrochemical disinfection in comparison with
chlorination, ozonation and Fenton reaction. Process
Biochem. 39, 1421–1426.
Furuta, T., Tanaka, H., Nishiki, Y., Pupunat, L., Haenni, W.,
Rychen, P., 2004. Legionella inactivation with diamond
electrodes. Diamond Relat. Mater. 13, 2016–2019.
Gutknecht, J., Hartmann, F., Kirmair, N., Reis, A., Schoeberl, M.,
1981. Anodic oxidation as water disinfection process in food
plants and breweries. GIT Fachz. Lab. 25, 427–481.
Jeong, J., Kim, J.Y., Cho, M., Choi, W., Yoon, J., 2007. Inactivation of
Escherichia coli in the electrochemical disinfection process
using a Pt anode. Chemosphere 67, 652–659.
Jeong, J., Kim, J.Y., Yoon, J., 2006. The role of reactive oxygen
species in the electrochemical inactivation of
microorganisms. Environ. Sci. Technol. 40, 6117–6122.
Kraft, A., Stadelmann, M., Blaschke, M., Kreysig, D., Sandt, B.,
Schroder, F., Rennau, J., 1999. Electrochemical water
disinfection. Part I: hypochlorite production from very dilute
chloride solution. J. Appl. Electrochem. 29, 859–866.
Malpass, G.R.P., Miwa, D.W., Machado, S.A.S., Olivi, P., Motheo, A.J.,
2006. Oxidation of the pesticide atrazine at DSA electrodes. J.
Hazard. Mater. B 75, 99–113.
Marselli, B., Garcia-Gomez, J., Michaud, P.-A., Rodrigo, M.A.,
Comninellis, C., 2003. Electrogeneration of hydroxyl radicals
on boron-doped diamond electrodes. J. Electrochem. Soc. 150
(3), D79–D83.
Martinez-Huitle, C.A., Brillas, E., 2008. Electrochemical
alternatives for drinking water disinfection. Angew. Chem. 47
(11), 1998–2005.
Michaud, P.-A., Panizza, M., Ouattara, L., Diaco, T., Foti, G.,
Comninellis, C., 2003. Electrochemical oxidation of water on
synthetic boron-doped diamond thin film anodes. J. Appl.
Electrochem. 33, 151–154.
Neta, P., Huie, R.E., Ross, A.B., 1988. Rate constants for reactions
of inorganic radicals in aqueous solution. J. Phys. Chem. Ref.
Data 17 (3), 1027–1284.
Panizza, M., Cerisola, G., 2005. Application of diamond electrodes
to electrochemical processes. Electrochim. Acta 51, 191–199.
Shimada, K., Shimahara, K., 1982. Responsibility of hydrogen
peroxide for the lethality of resting Escherichia coli B cells
exposed to alternating current in phosphate buffer solution.
Agric. Biol. Chem. 46, 1329–1337.
Szpyrkowics, L., Radaelli, M., Daniele, S., 2005. Electrocatalysis of
chlorine evolution on different materials and its influence on
the performance of an electrochemical reactor for indirect
oxidation of pollutants. Catal. Today 100, 425–429.
Trasatti, S., 1987. Progress in the understanding of the
mechanism of chlorine evolution at oxide electrodes.
Electrochim. Acta 32 (3), 369–382.
USEPA, 2003. Ultraviolet Disinfection Guidance Manual. EPA
815-D-03-007. Office of Water, Washington, DC.