Electrochemical reductive dechlorination of 2,4-dichlorophenoxyacetic acid using a palladium/nickel foam electrode

Article abstract of DOI:10.1016/j.electacta.2012.03.038

The electrochemical reductive dechlorination of 2,4-dichlorophenoxyacetic acid (2,4-D) in an aqueous solution was investigated at ambient temperature using a palladium/nickel foam (Pd/Ni foam) electrode in batch mode experiments. The catalytic electrode prepared using the standard chemical deposition method was further characterized using X-ray diffraction and scanning electron microscopy. It was observed that the reaction followed a pseudo-first-order kinetics model, the magnetic agitator-supported system could achieve 87% removal of 2,4-D within 4 h, which is 16% higher than the efficiency obtained under a nitrogen atmosphere. No organic intermediates other than phenoxyacetic (PA), o-chlorophenoxyacetic acid (o-CPA) and p-chlorophenoxyacetic acid (p-CPA) were observed to be generated during the reaction. The dechlorination efficiency depended on several factors including the current density, the palladium loading and the initial concentrations of the supporting NaCl electrolyte and the 2,4-D. The palladium loading and the NaCl concentration had a greater effect on the dechlorination kinetics of 2,4-D. Furthermore, the efficiencies of dechlorination and PA formation could be improved by optimizing the reaction system by modifying the ventilation conditions.

Full text of DOI:10.1016/j.electacta.2012.03.038

Electrochimica Acta 69 (2012) 389–396
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Electrochemical reductive dechlorination of 2,4-dichlorophenoxyacetic acid
using a palladium/nickel foam electrode
∗
Kairan Zhu, Shams Ali Baig, Jiang Xu, Tiantian Sheng, Xinhua Xu
Department of Environmental Engineering, Zhejiang University, Hangzhou 310058, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
The electrochemical reductive dechlorination of 2,4-dichlorophenoxyacetic acid (2,4-D) in an aque-
ous solution was investigated at ambient temperature using a palladium/nickel foam (Pd/Ni foam)
electrode in batch mode experiments. The catalytic electrode prepared using the standard chemical
deposition method was further characterized using X-ray diffraction and scanning electron microscopy.
It was observed that the reaction followed a pseudo-first-order kinetics model, the magnetic agitator-
supported system could achieve 87% removal of 2,4-D within 4 h, which is 16% higher than the
efficiency obtained under a nitrogen atmosphere. No organic intermediates other than phenoxyacetic
Received 16 December 2011
Received in revised form 5 March 2012
Accepted 7 March 2012
Available online 16 March 2012
Keywords:
Dechlorination
Electrochemical
Palladium
Nickel foam
Kinetics
(PA), o-chlorophenoxyacetic acid (o-CPA) and p-chlorophenoxyacetic acid (p-CPA) were observed to be
generated during the reaction. The dechlorination efficiency depended on several factors including the
current density, the palladium loading and the initial concentrations of the supporting NaCl electrolyte
and the 2,4-D. The palladium loading and the NaCl concentration had a greater effect on the dechlorina-
tion kinetics of 2,4-D. Furthermore, the efficiencies of dechlorination and PA formation could be improved
by optimizing the reaction system by modifying the ventilation conditions.
©
2012 Elsevier Ltd. All rights reserved.
1
. Introduction
disrupting activities in both animals and humans and irreparable
damage to the central nerve system has been reported [11,12].
Due to the deleterious effects of 2,4-D on human health and
the serious risks this chemical poses to wildlife, the World Health
Organization (WHO) defined 2,4-D as moderately toxic (class II) and
recommended a maximum allowable concentration of 0.1 ppm in
drinking water in 1982 [13]. Subsequently, the United States Envi-
ronmental Protection Agency (US-EPA) also recommended that the
permissible level of 2,4-D in potable water be 0.1 ppm [14]. The
prevalence and toxicity of 2,4-D has triggered the development
of effective degradation techniques for this chlorinated aromatic
herbicide that are aimed at environmental protection. To date, a
variety of individual or integrated methods have been developed
for the removal of pesticides from water, including adsorption,
photocatalysis, photoelectrochemistry, sonoelectrochemistry, bio-
logical degradation, electrochemistry assisted by microwave and
membrane technology [15–20]. Despite these advances, there is no
universally applied technique available for the treatment of 2,4-D.
Numerous published articles regarding the electrochemical
reductive dechlorination of chlorinated organic compounds (COCs)
are widely available [21–24]. Matsunaga and Yasuhara [25] inves-
tigated the complete dechlorination of 1-chloronaphthalene using
an electrochemical reduction process. Sun et al. [26] reported the
electrochemical reductive dechlorination of 2,4-dichlorophenol.
Because there are several intrinsic advantages of using mild reac-
tion conditions, including a rapid reaction rate, good recovery of
the catalyst and the absence of recalcitrant secondary contaminants
The contamination of surface and ground waters by pesticides
is a matter of environmental concern that is recognized because
of the potential toxic, carcinogenic, mutagenic effects and mobil-
ity of the pesticides [1,2]. 2,4-D is an herbicide belonging to the
chlorophenoxyacetic acid group. Because of its low cost and good
selectivity, 2,4-D is one of the most commonly used herbicides in
the agriculture sector around the world for controlling a wide range
of broadleaf weeds and grasses from crops such as sugar cane, oil
palm, cocoa and rubber [3,4]. The application of 2,4-D for high-yield
farming is frequent in China and many other countries, and 2,4-D
can be easily transported into water bodies as a result of surface
runoff and leaching [5,6]. As a result of its widespread use and high
water solubility, 2,4-D has been frequently detected as a major con-
taminant in both surface and ground water sources [7]. Conversely,
2
,4-D is a weakly biodegradable pollutant that has existed in natural
water, soil resources and agricultural food products [8,9]. Addition-
ally, 2,4-D poses a risk for dioxin formation because it contains
chlorine [10] and introduces threats to public health. The toxico-
logical effects of 2,4-D on human health have recently been studied
extensively by different researchers [11,12]. A connection between
2
,4-D and the development of cancer in mammals, the endocrine
∗ Corresponding author. Tel.: +86 571 88982031; fax: +86 571 88982031.
E-mail address: xuxinhua@zju.edu.cn (X. Xu).
0
013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2012.03.038

3
90
K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
[
27], the electrochemical reductive approach has been suggested as
acetone for 20 min, the nickel foam was rinsed with deionized
−
1
a promising method for the detoxification of chlorine-containing
aromatic hydrocarbons [28]. This technology can work without
additional chemicals and energy; thus, it is environmentally com-
patible, versatile, safe and easy to handle [19]. The electrochemical
reductive approach is the most common technology used to address
the purification of pesticides and related compounds in con-
taminated water. Tsyganok [29,30] explored the electrocatalytic
reductive dechlorination of 2,4-D on carbon materials contain-
ing supported palladium and several other noble metals. Among
the noble metals, palladium is generally utilized as the catalyst
loaded on inert supports for the electrochemical reactions. Many
publications have reported the application of palladium in the
degradation of aromatic compounds in an aqueous medium via
electrochemical dechlorination [31]. Currently, nickel foam is being
extensively applied in controlling environmental pollution due to
its high porosity and large specific surface area, which are desirable
characteristics for hydrogenation catalysts or electrode materials in
rechargeable batteries [32].
water and immersed in 80 g L sulfuric acid for 5 min to remove
the oxide layer. Finally, the nickel foam was thoroughly washed
with deionized water.
For the chemical deposition of palladium, a supporting nickel
foam electrode was placed in a conical flask (with a volume of
100 mL) consisting of 20 mL of 10 mM PdCl2 and 30 mM NaCl. This
◦
electrode-containing solution was then incubated at 32 C with
shaking until the original yellow solution became colorless.
The surface morphology of the Pd/Ni foam electrode was
observed under an XL-30 environmental scanning electron micro-
scope (ESEM; Philips, Netherlands) at 2 kV, and the X-ray diffraction
(XRD) analysis of the structural properties of the electrode was per-
formed using a Rigaku D/max 2550 PC X-ray diffractometer. The
analysis was performed at 40 kV and 300 mA, and copper metal
radiation was used to produce X-rays (ꢀ = 0.1542 nm).
2.3. Bulk electrolysis
In the current study, the electrochemical reductive dechlorina-
tion method was chosen because of its affordability and ecological
importance. The aim of this work was to evaluate the performance
of a Pd/Ni foam as the working cathode for the electrochemical
reductive dechlorination of 2,4-D in a two-compartment electro-
chemical system. Nickel foam was also employed as the electrode
substrate. Various factors that might affect the dechlorination
of 2,4-D were systematically investigated, including the initial
concentrations of the supporting electrolyte NaCl and 2,4-D, the
applied current density and the palladium loading rate. More-
over, to improve the efficiency of dechlorination under relatively
optimal conditions, the reaction system was optimized using dif-
ferent ventilation conditions in the cathode chamber. The kinetics
of the electrochemical dechlorination of 2,4-D using the Pd/Ni foam
electrode were also studied. We chose NaCl as the supporting elec-
trolyte based on the fact that the actual wastewaters contain NaCl
The electrochemical reductive dechlorination of 2,4-D was car-
ried out in a cell that consisted of two compartments separated by
a Nafion-117 cation-exchange membrane to prevent the diffusion
of the chloride ions produced in the cathode compartment into the
anode compartment to form Cl2. A Pd/Ni foam electrode was used as
the cathode, and a platinum foil was used as the anode. The distance
between the anode and the cathode was approximately 80 mm.
−
1
The cathode compartment contained 80 mL of a 2 g L
aqueous
solution of NaCl with a known amount of 2,4-D, and the anode com-
−
1
partment contained 40 mL of a 2 g L aqueous solution of NaCl.
A reductive atmosphere around the cathode was maintained by
sparging with nitrogen during the dechlorination experiments. The
electrolysis was performed at a constant current density, which
was controlled by an SK1760SL20A direct-current regulated power
supply for 4 h, and the samples (0.8 mL) were withdrawn from the
cathode compartment at pre-determined time points for further
analysis.
[
33]. Additionally, many studies have reported the use of NaCl as
the widely used supporting electrolyte [15,34].
2.4. Analytical methods
2
. Experimental
The concentrations of 2,4-D, o-CPA, p-CPA and PA that were
2
.1. Materials and methods
generated during the electrolysis were determined using a SHI-
MADZU high-performance liquid chromatograph at 283 nm with
an Agilent TC-C18 column (150 × 4.6). The mobile phase was a
60:40 (v/v) mixture of methanol:water (0.2% H PO ), and a flow
rate of 0.5 mL min was used. The volume of the injected sample
was 20 ␮L.
Palladium chloride (PdCl ; ≥99.97%) was used as the main
2
source of palladium and was purchased from the Sinopharm Chem-
ical Reagent Co., Ltd., China. The chemical was of analytical grade
and was used as received without any further purification. The
nickel foam from Shenzhen Rolinsia Power Materials, Ltd., was
used as the electrode. 2,4-D (≥97%), p-chlorophenoxyacetic acid
3
4
−
1
3. Results and discussion
(
p-CPA; ≥98%) and phenoxyacetic acid (PA; ≥98%) were obtained
from Jingchun Reagent Co., Ltd., Shanghai. o-Chlorophenoxyacetic
acid (o-CPA; ≥98%) was purchased from Alfa Aesar Chemical Co.,
Ltd., Tianjin. The original 2,4-D did not contain o-CPA or p-CPA. N2
gas (≥99.9%) stored in a steel cylinder was provided by the Jingong
Special Gas Co., Ltd., China. A stock solution of 2,4-D with a con-
3.1. Characterization of electrodes
Fig. 1 shows the XRD patterns of the nickel foam electrode, the
Pd/Ni foam electrode before the dechlorination reaction and the
Pd/Ni foam electrode after the reaction. All of the patterns exhib-
−1
◦
◦
◦
centration of 5 g L was prepared by dissolving 0.25 g of 2,4-D in
methanol and diluting this solution to 50 mL in a volumetric flask.
The stock solution was stored in a refrigerator at 4 C before use.
The reaction solutions were prepared by diluting the stock solu-
tion of 2,4-D to the desired concentration using a 2 g L aqueous
solution of NaCl.
ited peaks at 44.1 , 51.5 and 76 , corresponding to the (1, 1, 1), (2,
0, 0) and (2, 2, 0) planes of nickel metal, respectively. Additionally,
◦
◦
◦
◦
three diffraction peaks at 39.7 , 46.2 and 67.7 corresponding to
the weak (1, 1, 1), (2, 2, 0) and (3, 1, 1) planes of palladium could also
be observed in the patterns of the palladium-deposited samples,
indicating the presence of palladium on the surface of the elec-
trodes. This result was important because it directly impacted the
efficiency of the dechlorination. The difference between Fig. 1(b)
and (c) demonstrates that the presence of palladium was slightly
modified after the reaction. The strength of the palladium peak
decreased after the dechlorination reaction compared with that
before the reaction at the same degree. This result could be due
−
1
2.2. Electrode preparation and characterization
Prior to palladization, the nickel foam (20 mm × 30 mm; thick-
2
−1
ness, 1.2 mm; pore density, 110 ppi; SBET, 1.2 m g ) was cleaned
and etched using the following steps. After ultrasonic cleaning in

K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
391
Pd
Ni
(
(
(
a) Ni electrode
b) Pd/Ni electrode before reaction
c) Pd/Ni electrode after reaction
(
a)
(
b)
c)
(
3
0
40
50
60
70
80
2
Theta/degree
Fig. 1. XRD patterns of (a) the nickel foam electrode, (b) the Pd/Ni foam electrode
before the reaction and (c) the Pd/Ni foam electrode after the reaction.
to the contamination or passivation of palladium or the loss of
palladium from the nickel foam during the dechlorination process.
SEM was used to image the surface morphology of a fresh nickel
foam and the Pd/Ni foam, and the results are shown in Fig. 2. The
nickel foam had a smooth surface with clear zigzag lines (Fig. 2(a)),
and the fresh Pd/Ni foam electrode exhibited a thick uniform film
of palladium with deep grooves on the surface of the Pd/Ni foam
electrode (Fig. 2(b)), indicating that palladium was successfully
deposited on the nickel foam. This surface coating appeared to
enhance the dechlorination efficiency. In contrast, the palladium
film appeared thinner after 4 h of reaction, as shown in Fig. 2(c). The
grooves became less prominent, more granular palladium domains
were observed and the zigzag nickel lines were looming. These
observations might be due to the loss of part of the palladium
film during the reaction, which could also be verified by the XRD
results.
Fig. 2. SEM micrographs of (a) the fresh nickel foam electrode, (b) the Pd/Ni foam
electrode and (c) the Pd/Ni foam electrode after the dechlorination of 2,4-D.
According to the results of the Brunauer–Emmett–Teller (BET)
−1
−1
−1
NaCl and 1 g L NaCl, respec-
presence of 4 g L
NaCl, 1.5 g L
2
−1
−1
test, the surface area of the Pd/Ni foam was 3.5 m g , which is
tively. Meanwhile, PA and inorganic chlorine were detected as final
products of the reaction. A low NaCl concentration decreased the
PA production rate, which was expressed as the ratio of the actual
2
much larger than that of the nickel foam (SBET: 1.2 m g ). Thus, the
large surface area of the palladium particles provided adsorption
sites for [H] and pollutants.
50
45
40
3.2. Factors affecting 2,4-D dechlorination
The catalytic dechlorination of 2,4-D using a Pd/Ni foam
35
30
electrode was investigated at ambient temperature. During the cat-
alytic dechlorination process, 2,4-D was initially transformed into
o-CPA and p-CP and then converted to PA. Hence, o-CPA and p-
CPA were the chlorinated intermediates, and PA was the sole final
organic product of the reaction. No other chlorinated intermediates
or final organic products were detected.
2
2
1
1
5
0
5
0
25
g L-1NaCl
1
-
1
1.5 g L NaCl
2
0
5
-1
2
4
g L NaCl
-
1
g L NaCl
1
3
.2.1. Effects of NaCl concentration on 2,4-D dechlorination
NaCl was selected as the supporting electrolyte in the present
10
5
study. The effect of the NaCl concentration on the dechlorination
of 2,4-D using the Pd/Ni foam electrode was examined. Fig. 3 illus-
trates the dechlorination of 2,4-D using the Pd/Ni foam electrode
0
0
1
2
3
4
−
1
−1
at various NaCl concentrations ranging from 1 g L to 4 g L . As
shown in Fig. 3, the 2,4-D concentration rapidly decreased, and the
removal percentage reached 71% in 4 h using the Pd/Ni foam elec-
Time/h
Fig. 3. Effects of the supporting electrolyte NaCl concentration on the 2,4-D dechlo-
rination using the Pd/Ni foam electrode (initial concentration of 2,4-D: 50 mg L
current density: 1.667 mA cm and palladium loading: 1.78 mg cm ).
−1
trode in the presence of a 2 g L NaCl solution. In contrast, only
approximately 65%, 48% and 18% of the 2,4-D was removed in the
−1
,
−
2
−2

3
92
K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
PA amount produced to the theoretical PA amount during the total
dechlorination of 2,4-D, from 70% in 2 g L NaCl to 13% in 1 g L
3.2.3. Effects of current density on 2,4-D dechlorination
−1
−1
The 2,4-D reduction profiles at various current densities and
the amounts of the chlorinated intermediates detected in the reac-
tion are shown in Fig. 5. Increasing the apparent current density
NaCl in 4 h. The coinciding decrease of both the 2,4-D removal per-
centage and the PA production rate with the decrease in the NaCl
concentration suggested that the presence of NaCl played a sig-
nificant role in the dechlorination of 2,4-D. Concurrent with the
−
2
from 0.833 mA cm
(according to the surface area; actual cur-
−
2
−2
rent density: 0.00042 mA cm ) to 1.667 mA cm (actual current
density: 0.00083 mA cm ) elicited a distinct increase in the 2,4-
−
1
−2
removal of 2,4-D, we observed a lower PA production in 4 g L
NaCl, which was 57%. Furthermore, we observed that the initial out-
put potential differences were negatively correlated with the NaCl
concentrations in a series of experiments. When the NaCl concen-
D conversion efficiency from 35% to 71% in 4 h using the Pd/Ni
foam cathode for the electrolysis. However, a further increase of
−
2
−2
the apparent current density from 1.667 mA cm to 2.5 mA cm
−1
−1
−1
−1
−2
tration was increased from 1 g L , 1.5 g L and 2 g L to 4 g L
,
(actual current density: 0.00125 mA cm ) did not enhance the
dechlorination efficiency, but instead slightly reduced the dechlori-
nation efficiency from 71% to 68% using the Pd/Ni foam cathode. In
general, a lower current density could guarantee that a sufficient
amount of fresh adsorbed reactive hydrogen ((H)adsM) is gener-
ated by the electrolysis of water, thereby ensuring an increase
in the dechlorination rate. At a relatively higher current density,
the electric potential would be more negative, consequently lead-
ing to a faster generation of (H)adsM. However, a large portion of
adsorbed reactive hydrogen tends to generate H2 gas rather than
reacting with 2,4-D at a relatively high current density. Thus, the
polarized part of the catalysts that are thought to adsorb pollu-
tants will be covered by a H2 layer, which will hinder the further
mass transfer of the chlorinated organics to the polarized active
sites [38]. This hindrance leads to the inhibition or the reduction
the initial output potential differences were 19.3 V, 14.3 V, 11.0 V
−
1
and 9.8 V, respectively. Consequently, we chose 2 g L NaCl as the
supporting electrolyte in the subsequent studies.
The underlying phenomenon appeared to be that the use of a
NaCl solution as the supporting electrolyte provided a large number
of conductive free ions in the solution, which ensured the smooth
flow of the circuit. In addition, when the NaCl concentration was
high, more conductive free ions would be ionized to exist outside
the system, thus increasing the ionic strength and the conductivity
of the solution [35] and enhancing the dechlorination of 2,4-D. This
explanation also supported the phenomenon that explains why the
initial output potential difference increased as the NaCl concentra-
tion decreased at the beginning of the reaction process. Notably, the
efficiency of 2,4-D dechlorination would actually decrease slightly
if the NaCl concentration increased continually.
−
2
of the 2,4-D conversion. Based on these results, 1.667 mA cm
was chosen as the current density to be used for the remainder
of this study as an optimal value for the dechlorination pro-
cess.
3
.2.2. Effects of 2,4-D initial concentration on 2,4-D
dechlorination
During the catalytic degradation process, 2,4-D was initially
reduced to o- and p-CPA and then converted to PA, which was
evident from the concentration changes. The reduction of 2,4-D
resulted in an increased PA concentration. Hence, PA and inorganic
chlorine were detected as the final products of the reaction. The
generation and further catalytic degradation process of o- and p-
CPA during the same reaction period with different current density
values are presented in Fig. 5(c) and (d). The results demonstrated
that the generation of o-CPA was much faster than that of p-CPA,
and without considering the current density value, p-CPA did not
appear until the reaction was allowed to proceed for 1 h. In addition,
with the ongoing degradation process, the o-CPA production was
much higher than that of p-CPA. This difference is likely because the
chlorine atom located at the para position was more reactive than
that at the ortho position as a result of steric hindrance [39]. The
chlorine located at the para position was surrounded by a much
larger space than was the chlorine located at the ortho-position,
and it could be easily replaced by adsorbed reactive hydrogen.
Therefore, the generation of o-CPA occurred faster than did the
generation of p-CPA, and the amount of o-CPA generated was much
higher than that of p-CPA.
Four different 2,4-D initial concentrations (20, 35, 50 and
0 mg L ) were employed in batches during the reaction pro-
−
1
8
cess. The effective electrode area remained constant during the
reaction. The 2,4-D reduction profiles with different initial 2,4-D
concentrations using the Pd/Ni foam electrode and the amounts of
chlorinated intermediates detected are shown in Fig. 4. The removal
percentages of 2,4-D reached 34, 63, 60 and 71% after 2 h and then
increased to 45, 70, 71 and 80%, respectively, after 4 h of electrolysis.
It was obvious that the chlorine atoms were removed individually
from the benzene ring of 2,4-D, which is similar to other dechlori-
nation processes of chlorinated organic compounds [36]. CPA was
detected as the only intermediate generated during the electroly-
sis, and the chlorinated intermediate was observed to continually
increase throughout the electrolysis (Fig. 4). However, it should be
noted that the accumulation of CPA did not simply indicate that
there was no dechlorination reaction of CPA. With an increased
number of chlorine atoms, the electron density on the benzene
ring will be lower and vice versa. The lower electron density of
benzene is more susceptible to nucleophilic attack of the reducing
agents, so the chlorine atoms can easily be replaced by other atoms
[
37]. Therefore, it was observed that the chlorine intermediate was
Although o-CPA and p-CPA could be reduced to PA in the fur-
ther catalytic degradation process, it was observed that the o-CPA
and p-CPA concentrations were continually increasing instead of
decreasing even after 4 h of the reaction process. Hence, the con-
centration of o-CPA was much higher than that of p-CPA during the
catalytic dechlorination of 2,4-D.
continuously accumulated during the experiments. This observa-
tion could be explained by studying the kinetics of the reaction and
calculating the reaction rate constants as explained in Section 3.5.
The PA that was detected as the final product continuously
increased with the degradation of the target pollutant 2,4-
D throughout the electrolysis. The PA production percentages
observed were nearly 28, 63, 70 and 77%, respectively, when the ini-
3.2.4. Effects of palladium loading on 2,4-D dechlorination
−
1
tial 2,4-D concentration increased from 20, 35 and 50 to 80 mg L
after 4 h of the electrolysis.
The reductive dechlorination of 2,4-D using the Pd/Ni foam elec-
trode occurs on the surface of particles; hence, palladium loading
is also a significant variable. The amount of surface area available
is among the most significant experimental variables affecting the
contaminant reduction.
Although the 2,4-D removal percentages were similar when the
initial 2,4-D concentration was high, the absolute 2,4-D removal
amount increased with the increase of the initial 2,4-D concentra-
tion. The absolute 2,4-D removal amount increased from 9, 24 and
Several electrodes with different palladium loadings
−
1
3
5 to 64 mg L in 4 h as the initial 2,4-D concentration was varied
−2
−2
−2
−2
)
(
0.44 mg cm , 0.855 mg cm , 1.33 mg cm , and 1.78 mg cm
−1
from 20, 35 and 50 to 80 mg L
.
were prepared by varying the amount of PdCl2 available while

K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
393
8
6
4
2
0
0
0
0
-
1
8
5
0 mg L 2,4-D
-1
0 mg L 2,4-D
-1
35 mg L 2,4-D
-1
2
0 mg L 2,4-D
6
5
4
3
2
1
0
0
4
3
2
1
0
0
0
0
0
1
2
3
4
Time/h
Fig. 4. Effects of the initial concentration of 2,4-D on the 2,4-D dechlorination using the Pd/Ni foam electrode (supporting electrolyte NaCl concentration: 2 g L−1, current
−2
−2
density: 1.667 mA cm and palladium loading: 1.78 mg cm ).
making the electrode to explore the influence of catalyst loading
on the dechlorination of 2,4-D. Fig. 6 shows that the dechlori-
nation rate was strongly influenced by the palladium loading
during the reaction. As the palladium loading was increased from
3.3. Reaction and equilibrium analysis
A quantitative analysis of the intermediates is required to deter-
mine the conservation of mass during the dechlorination of 2,4-D.
Fig. 7(a) illustrates the intermediates and the carbon mass balance
during the dechlorination of 2,4-D using the Pd/Ni foam electrode.
Other conditions such as the supporting electrolyte concentra-
tion, the initial concentration of 2,4-D, the current density and
−2
−2
−2
−2
0.44 mg cm , 0.885 mg cm and 1.33 mg cm to 1.78 mg cm ,
the 2,4-D removal percentage using the Pd/Ni foam electrode
increased dramatically from 19%, 39% and 47% to 71%, respectively.
Consequently, the production yield of PA increased from 20%, 34%
and 50% to 70%, respectively. Moreover, both the 2,4-D removal
percentage and the PA production rate increased with the increase
in the palladium loading, suggesting that palladium played an
important role as a catalyst in the dechlorination of 2,4-D.
−
1
−1
−2
the palladium loading were 2 g L , 50 mg L , 1.667 mA cm and
1.78 mg cm , respectively.
−
2
The 2,4-D concentration decreased with the increasing elec-
trolysis time, and the reaction intermediates were released
50
40
30
20
24
(
a)
1
6
(
b)
8
0
6
4
2
0
(
c)
0
0
0
0
.15
.10
.05
.00
(
d)
0
1
2
3
4
Time/h
Fig. 5. Effects of the current density on the 2,4-D dechlorination using the Pd/Ni foam electrode (supporting electrolyte NaCl concentration: 2 g L−1, initial concentration of
−
1
−
2
2
,4-D: 50 mg L and palladium loading: 1.78 mg cm ).

3
94
K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
5
4
4
3
3
2
2
0
5
0
5
0
5
0
50
4
3
0
0
20
1
0
1
2
5
5
1
1
.78 mg cm-2
.33 mg cm-2
3
2
2
0
5
0
air
2
1
1
0
5
0
5
0
0.885 mg cm-2
no ventilation
N2
magnetic stirring
0
.44 mg cm-2
15
10
5
0
1
2
3
4
0
Time/h
0
1
2
3
4
Time/h
Fig. 6. Effects of palladium loading on the 2,4-D dechlorination using the Pd/Ni foam
−
1
electrode (supporting electrolyte NaCl concentration: 2 g L , initial concentration
of 2,4-D: 50 mg L and current density: 1.667 mA cm ).
−1
−2
Fig. 8. Variation of the concentration of 2,4-D and the final product PA using the
Pd/Ni foam electrode under different ventilation conditions (supporting electrolyte
−
1
−1
NaCl concentration: 2 g L , initial concentration of 2,4-D: 50 mg L , current den-
−
2
−2
sity: 1.667 mA cm and palladium loading: 1.78 mg cm ).
accompanied by the sequential hydrogenolysis of 2,4-D, indicating
the absence of any additional organic products apart from CPA and
PA. The initial concentration of 2,4-D was 50 mg L , and the total
carbon mass balance recorded was greater than 99.5%; thus, almost
no carbon mass was lost during the reaction. The carbon loss that
−
1
was observed could be considered within the allowable error lim-
its, such as operation error, test error, etc. The 2,4-D concentration
rapidly decreased within the first 1 h of the reaction, while the final
PA product was generated quickly and approached the maximum
attainable limit.
(
a) 0.25
The current efficiency (CE) of the 2,4-D dechlorination in the
period of t = 0–t using the Pd/Ni foam electrode can be calculated
as follows:
0.20
0.15
0.10
0.05
0.00
(
n C + n (C
+ Cp-CPA))FV
1
PA
2
o-CPA
CE =
Q
PA
(n1CPA + n2(Co-CPA + Cp-CPA))FV
o-CPA
p-CPA
2,4-D
=
ꢀ
t=t
I dt
t=0
Total C/8
In these equations, n1 and n2 are the electron transfer numbers
per one molecule of PA (n = 2) and CPA (n = 4), respectively, CPA,
1
2
Co-CPA and Cp-CPA are the concentrations (M) of PA, o-CPA and p-
−
1
CPA, respectively, F is the Faraday constant (96,500 As mol ), V is
the volume of the total cathode electrolyte (L), Q is the total electric
charge (As) and I is the electrolysis current (A). As calculated using
the above equation, the CE of the 2,4-D dechlorination using the
Pd/Ni foam electrode is shown in Fig. 7(b). Although the current
efficiency of the dechlorination reaction obtained in the study did
not exceed 10% and the current efficiency diminished as the elec-
trolysis time increased beyond 1 h, the 2,4-D dechlorination to PA
was readily achieved. The results [30] presented in A.I. Tsyganok’s
study also demonstrated a low current efficiency, which did not
exceed 10%. Li et al. [40] described a decay in the current efficiency
with thane increase in the reaction time. The current efficiency can
be influenced by many factors, such as the catalytic electrode, the
current density and the pollutant type. We could not determine
which factor has a greater influence on the current efficiency in this
study. Future work will focus on improving the current efficiency.
0
1
2
3
4
Time/h
(b)
10
8
6
4
2
0
3.4. Improved dechlorination efficiency
0
.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
To improve the dechlorination efficiency, many researchers
Time/h
have already conducted studies on various applicable options [41].
However, this part of the investigation introduces another applied
perspective in improving the dechlorination efficiency.
Fig. 8 demonstrates the results of the laboratory-scale dechlo-
rination of 2,4-D and the PA production rate under different
Fig. 7. Dechlorination of 2,4-D (a) and the current efficiency (b) during the elec-
trochemical dechlorination of 2,4-D using the Pd/Ni foam electrode (supporting
electrolyte NaCl concentration: 2 g L , initial concentration of 2,4-D: 50 mg L
current density: 1.667 mA cm and palladium loading: 1.78 mg cm ).
−
1
−1
,
−
2
−2

K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
395
ventilation conditions. The best dechlorination efficiency was
obtained when the reaction was stirred using a magnetic agitator
during the reaction process. The removal percentage of 2,4-D was
100
80
60
40
20
0
first run
second run
third run
8
7%, and the PA production rate reached 83% when the reaction
underwent magnetic stirring. Furthermore, these reaction con-
ditions resulted in an efficiency that was 16% higher than the
efficiency obtained in a recent study conducted in the presence
of nitrogen, which was aimed to guarantee the reductive atmo-
sphere of the cathode chamber throughout the degradation process
[
26]. Additionally, it was estimated from the trend charts of 2,4-
D and PA that the reaction rate remained relatively stable during
the entire degradation process under the magnetic stirring condi-
tions, whereas the reaction rates during the entire experimental
procedure under the nitrogen atmosphere were changed. These
phenomena can be explained by the mechanism of the electro-
chemical reductive dechlorination reaction. It is well known that
the adsorbed reactive hydrogen that is produced on the surface
of the cathode is used to break the C Cl bond and replace the
chlorine atom on the benzene ring of chlorinated organic com-
pounds [36,42]. This process is the most important step of the 2,4-D
dechlorination reaction, which results in the reduction or even the
disappearance of the toxicity of 2,4-D. Thus, it can be observed that
the generated reactive hydrogen plays an important role in the
dechlorination of 2,4-D. However, this reactive hydrogen may be
stripped off by nitrogen gas throughout the reaction and thus may
impede the reaction. The ventilation conditions are thus critical for
this type of reaction.
Two sets of experiments were conducted to investigate the
effect of the presence of air on the reaction. The removal percent-
ages of 2,4-D reached up to 43% and 26%, and the PA productions
were 45% and 14%. Magnetic stirring was used to achieve an
even distribution of the reaction solution in the cathode cham-
ber during the entire reaction process, and the reactive hydrogen
generated could stably be adsorbed onto the electrode surface,
thus facilitating the dechlorination reaction. However, although
the presence of nitrogen in the cathode chamber could ensure
the reductive atmosphere, it affected the dechlorination effi-
ciency. The presence of gases reduced the efficiency of 2,4-D
dechlorination by causing the reactive hydrogen to be stripped
from the surface of the catalytic electrode or, more often, to be
completely removed from the reaction system. Significantly, the
reductive dechlorination process was negatively affected in the
presence of the air because the air contains more than 20% oxy-
gen. These batch-mode experiments suggested that the reaction
system could be optimized by modifying the ventilation, and the
electrochemical reductive 2,4-D dechlorination using the Pd/Ni
foam catalytic electrode performed better under magnetic stirring
conditions.
0
1
2
3
4
Time/h
Fig. 9. Removal percentage of 2,4-D using the same Pd/Ni foam electrode under
magnetic stirring conditions for three cycles of dechlorination (supporting elec-
trolyte NaCl concentration: 2 g L , initial concentration of 2,4-D: 50 mg L , current
density: 1.667 mA cm and palladium loading: 1.78 mg cm ).
−
1
−1
−2
−2
3
.6. Kinetics for the dechlorination of 2,4-D by a freshly prepared
Pd/Ni electrode
Fig. 7 illustrates the transformations of 2,4-D to PA using the
Pd/Ni foam electrode, and the values in the figure are expressed
in molar concentrations. The organic products consisted of PA and
o- and p-CPA. Another two sets of experiments were performed
in which the initial waste waters contained 50 mg L o-CPA and
0 mg L p-CPA. The other conditions were the same as described
in Section 3.3. PA was detected as the only organic product. Thus,
it was deduced that the hydrodechlorination step of 2,4-D can be
described in the following sequence:
−
1
−
1
5
ꢁ
− k21
−
−
k
o-CPA + Cl −→ PA + 2Cl
p-CPA + Cl −→ PA + 2Cl
1
2, 4-D−→ CPA
(1)
k
−
22
In this sequence, CPA represents the total combination of o-CPA
and p-CPA in the reaction process. Previously established literature
results have shown that the degradation reaction of chlorinated sol-
vents in an aqueous solution followed a pseudo-first-order kinetics
mechanism [44]. The corresponding reaction rate equation for the
disappearance of 2,4-D, o-CPA and p-CPA can be represented as
follows:
−
−
−
dC2,4-D
=
=
k C
1
(2)
(3)
(4)
2,4-D
dt
dCo-CPA
k21Co-CPA
3.5. Electrode recovery
dt
dCp-CPA
Good reproducibility and stability are generally considered to
= k22Cp-CPA
dt
be important factors in choosing an ideal electrode. The elec-
trode stability was evaluated in terms of the change of the 2,4-D
removal percentage using the same Pd/Ni foam electrode after
three consecutive dechlorination cycles. All of the reactions were
conducted under magnetic stirring conditions. Fig. 9 shows the
removal percentage of 2,4-D in each 4 h experiment. The 2,4-D
removal percentage decreased from 87% or 84% to 78% at the Pd/Ni
foam electrode. This result suggested that the stability of the elec-
trode was not very good. This phenomenon might be due to the
desquamation of palladium particles from the catalysts [43]. This
result was consistent with the characterization of the electrodes
described in Section 3.1, which caused a reduction of the catalyst
lifetime.
After solving the above simultaneous rate equations, the reaction
leads to the following molar fractions:
−k1
t
˛2,4-D
= e
(5)
(6)
(7)
˛o-CPA = e−
k
t
t
2
1
˛p-CPA = e−
k
2
2
In the above equations, ˛ represents the molar fraction of the
subscript compound to the initial concentration of the parent com-
pound (i.e., 2,4-D). The k values were derived from fitting the
experimental data into Eqs. (5)–(7) according to the non-linear
least-square regression. The results showed that the k1 value for

3
96
K. Zhu et al. / Electrochimica Acta 69 (2012) 389–396
Table 1
both the NaCl concentration and palladium loading had a substan-
k1 values under different reaction conditions.
tial impact on the reaction kinetics. In conclusion, electrochemical
reduction using the Pd/Ni foam cathode is a promising technology
for the degradation of 2,4-D in the environment.
k1 (min 1
−
)
Reaction conditions
NaCl concentration (g L−1)
1
0.0008
0.0027
0.005
1
2
.5
Acknowledgments
2
,4-D initial concentration (mg L−1)
This research was supported by the Fundamental Research
Funds for the Central Universities, the National Water Pollution
Control and Management Project of China (2011ZX07101-012-
20
35
50
80
0.0043
0.0047
0.005
0.0061
0
2
08) and the National Natural Science Foundation of China (No.
0977086).
Current density (mA cm−2)
0.833
1.667
2.5
0.0018
0.005
0.005
References
Palladium loading (mg cm−2)
0
0
1
1
.44
.885
.33
0.0007
0.0018
0.0018
0.005
[
1] Y. Sunohara, S. Shirai, N. Wongkantrakorn, H. Matsumoto, Environ. Exp. Bot. 68
2010) 157.
2] I. Vergili, H. Barlas, Desalination 249 (2009) 1107.
(
[
.78
[3] Y.F. Xi, M. Mallavarapu, R. Naidu, Appl. Clay Sci. 49 (2010) 255.
[4] N. Maleki, A. Safavi, H.R. Shahbaazi, Anal. Chim. Acta 530 (2005) 69.
[5] R.A. Rebich, R.H. Coupe, E.M. Thurman, Sci. Total Environ. 321 (2004) 189.
[6] K. Shareef, G. Shaw, Chemosphere 72 (2008) 8.
Ventilation condition
Magnetic stirring
N2
No ventilation
Air
0.0088
0.005
0.0025
0.0012
[7] S. Deng, R. Ma, Q. Yu, J. Huang, G. Yu, J. Hazard. Mater. 165 (2009) 408.
[
[
8] J.R. Lipthay, S.R. Sorensen, J. Aamand, Environ. Pollut. 148 (2007) 83.
9] Y. Yasman, V. Bulatov, V.V. Gridin, S. Agur, N. Galil, R. Armon, I. Schechter,
Ultrason. Sonochem. 11 (2004) 365.
[
10] H. Aydin, N. Ozdemir, N. Uzunoren, Forensic Sci. Int. 153 (2005) 53.
2
,4-D dechlorination was 0.005 min 1, and the k21 and k val-
−
22
[11] S. Cenkci, M. Yildiz, I.H. Cigerci, A. Bozdag, H. Terzi, E.S.A. Terzi, Ecotoxicol.
Environ. Saf. 73 (2010) 1558.
[12] B. Bukowska, K. Hutnik, Pestic. Biochem. Physiol. 85 (2006) 174.
−
1
ues for o-CPA and p-CPA dechlorination were 0.0033 min
and
−
1
0.0037 min , respectively. From the calculated data, we could infer
[
[
13] B.H. Hameed, J.M. Salman, A.L. Ahman, J. Hazard. Mater. 163 (2009) 121.
14] L.J. Merini, V. Cuadrado, C.G. Flocco, A.M. Giulietti, Chemosphere 68 (2007)
259.
that the k21 and k22 values were smaller than the k value for 2,4-D
1
dechlorination. This result is consistent with the common phe-
nomenon that CPA accumulated continually as the intermediates
during the 2,4-D dechlorination reaction.
Table 1 lists the k1 values under different reaction conditions.
These data showed that k1 value for the 2,4-D dechlorination
[15] S.A. Neto, A.R. Andrade, Electrochim. Acta 54 (2009) 2039.
[16] V.K. Gupta, I. Ali, Suhas, V.K. Saini, J. Colloid Interface Sci. 299 (2006) 556.
[17] A.M.D. Freitas, C. Sirtori, P. Peralta-Zamora, Environ. Chem. Lett. 9 (2011) 97.
[18] N. Philippidis, S. Sotiropoulos, A. Efstathiou, I. Poulios, J. Photochem. Photobiol.
A 204 (2009) 129.
[19] J.X. Gao, G.H. Zhao, W. Shi, D.M. Li, Chemosphere 75 (2009) 519.
−1
increased from 0.0008 or 0.0027 to 0.005 min with increasing
NaCl concentrations from 1 or 1.5 to 2 g L . However, the k1 did
[
20] M.D. Esclapez, V. Saez, D. Milan-Yanez, I. Tudela, O. Louisnard, J. Gonzalez-
Garcia, Ultrason. Sonochem. 17 (2010) 1010.
−1
not change significantly when the initial concentration of 2,4-D was
[21] V. Saez, M.D. Esclapez, A.J. Frias-Ferrer, P. Bonete, J. Gonzalez-Garcia, Water Res.
43 (2009) 2169.
−1
increased from 20, 35 or 50 to 80 mg L . In short, the k1 value for
[
22] O. Scialdone, A. Galia, L. Gurreri, S. Randazzo, Electrochim. Acta 55 (2010)
01.
−
1
the 2,4-D dechlorination changed from 0.0018 to 0.005 min when
7
−
2
the current density was increased from 0.833 to 1.667 mA cm
,
[23] S.M. Kulikov, V.P. Plekhanov, A.I. Tsyganok, C. Schlimm, E. Heitz, Electrochim.
Acta 41 (1996) 527.
but this value did not increase with a further increase in the cur-
[
24] V.A. Sauro, D.C. Magri, J.L. Pitters, M.S. Workentin, Electrochim. Acta 55 (2010)
5584.
−
2
rent density to 2.5 mA cm . The k value also varied from 0.0007,
1
−1
0
.0018 and 0.0018 to 0.005 min with increases in the palladium
loading from 0.44, 0.885 and 1.33 to 1.78 mg cm , respectively.
[25] A. Matsunaga, A. Yasuhara, Environ. Sci. Technol. 37 (2003) 3435.
[26] Z.R. Sun, H. Ge, X. Hu, Y.Z. Peng, Sep. Purif. Technol. 72 (2010) 133.
−
2
[
27] Z.Q. He, L.Y. Zhan, Q. Wang, S. Song, J.M. Chen, K.R. Zhu, X.H. Xu, W.P. Liu, Sep.
Purif. Technol. 80 (2011) 526.
Finally, this k1 value obviously increased from 0.0012, 0.0025 and
−1
0
.005 to 0.0088 min if the reaction condition changed from air or
[28] A.I. Tsyganok, K. Otsuka, Appl. Catal. B: Environ. 22 (1999) 15.
[
[
[
29] A.I. Tsyganok, I. Yamanaka, K. Otsuka, J. Electrochem. Soc. 145 (1998) 3844.
30] A.I. Tsyganok, K. Otsuka, Electrochim. Acta 43 (1998) 2589.
31] A.I. Tsyganok, I. Yamanaka, K. Otsuka, Chemosphere 39 (1999) 1819.
nitrogen atmosphere to a magnetic stirring environment. The influ-
ence of palladium loading and the supporting NaCl concentration
on the k1 value was quite evident.
[32] B. Yang, G. Yu, D.M. Shuai, Chemosphere 67 (2007) 1361.
[
33] C. Sirtori, A. Zapata, I. Oller, W. Gernjak, A. Aguera, S. Malato, Water Res. 43
2009) 661.
(
4
. Conclusion
[
34] Z. Zainal, C.Y. Lee, M.Z. Hussein, A. Kassim, N.A. Yusof, J. Photochem. Photobiol.
A 172 (2005) 316.
Our experimental results indicated that the electrochemical
process using the Pd/Ni foam cathode electrode could efficiently
degrade 2,4-D in an aqueous solution. In these experiments under
[35] M. Boroski, A.C. Rodrigues, J.C. Garcia, L.C. Sampaio, J. Nozaki, N. Hioka, J. Hazard.
Mater. 162 (2009) 448.
[
36] Z. Zhang, Q.H. Shen, N. Cissoko, J.J. Wo, X.H. Xu, J. Hazard. Mater. 182 (2010)
52.
2
−
1
the optimal circumstances—namely 2 g L
NaCl, an initial 2,4-D
[37] J. Wiegel, Q.Z. Wu, FEMS Microbiol. Ecol. 32 (2000) 1.
[38] H. Cheng, K. Scott, P.A. Christensen, J. Electrochem. Soc. 150 (2003) 17.
[39] K. Miyoshi, T. Nishio, A. Yasuhara, M. Morita, Chemosphere 41 (2000) 819.
−1
−2
concentration of 50 mg L , 1.78 mg cm palladium loading and a
−2
current density of 1.667 mA cm —71% removal of 2,4-D and 70%
PA production could be achieved during a 4-h reaction period.
Moreover, the efficiencies would be enhanced by improving venti-
lation of the system, and a dechlorination efficiency of 87% could be
achieved under magnetic stirring conditions. For 2,4-D, the concen-
tration decay was fitted to a pseudo-first-order kinetics model, and
[
[
40] Y.P. Li, H.B. Cao, Y. Zhang, Chemosphere 63 (2006) 359.
41] M.A. Prasad, M.V. Sangaranarayanan, Electrochim. Acta 51 (2005) 242.
[42] Y. Han, W. Li, M.H. Zhang, K.Y. Tao, Chemosphere 72 (2008) 53.
[
[
43] E.N. Alvar, M.R. Golmohammadi, M. Rezeai, H.N. Alvar, A. Mardanloo, S.H.
Nouhian, M. Didari, J. Nat. Gas. Chem. 17 (2008) 321.
44] M.A. Oturan, I. Sires, N. Oturan, S. Perocheau, J.L. Laborde, S. Trevin, J. Elec-
troanal. Chem. 624 (2008) 329.