Electrochimica Acta 56 (2011) 2184–2189

Contents lists available at ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

A systematic study on the changes in properties of an activated carbon cloth

upon polarization
Edip Bayram, Erol Ayranci ∗
Akdeniz University, Faculty of Science, Department of Chemistry, Dumlupinar Boulvard, 07058 Antalya, Turkey

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

Article history: The effects of polarization on properties of activated carbon cloth (ACC) have been investigated sys-
Received 11 October 2010 tematically. The polarization-treated ACC samples were prepared by polarizing them in Na2 SO4 or
Received in revised form 2 December 2010 KH2 PO4 /KOH buffer solutions at potentials from −1.5 to 5.0 V. The properties, such as surface area,
Accepted 4 December 2010
pore size distribution (PSD), total pore volume, amount and nature of the surface functional groups
Available online 15 December 2010
and surface acidity, of pristine and polarization-treated ACC samples were determined. The samples
were also characterized electrochemically by determining the properties such as specific capacitance
Keywords:
and potential at point of zero charge (EPZC ). Anodic polarization in different electrolytes was found
Activated carbon cloth
Surface characterization
to cause oxidation on ACC. Although the surface textural properties did not change considerably, the
Surface functional groups changes took place in chemical and electrochemical properties upon anodic polarization were found to
Polarization be important. The increase in surface acidity shifted the pHPZC from 7.40 to 3.21 and EPZC from 164 to
Point of zero charge 355 mV. The optimum potential range, considered to be safe for polarization of ACC, was determined
as −1.5 to +0.8 V.
© 2010 Elsevier Ltd. All rights reserved.

1. Introduction surface functional groups such as carboxylic, phenolic and lactonic

Activated carbons are widely used in electrochemistry as an
electrode material due to their high specific surface area, devel-
oped pore structure and electrical properties such as low electrical
resistance and high conductivity. Among the activated carbons,
activated carbon cloth (ACC) or fibers have the advantages of
mechanical integrity and easy handling [1]. These advantages make
ACC more favorable than the other forms of activated carbons such
as granular or powder for electrochemical processes.
Electrochemical treatment technologies, using activated car-
bons as an electrode material, include electro-oxidation/reduction
[2–4] and electrosorption [5–7]. Electro-oxidation/reduction reac-
tions are Faradaic processes containing charge transfer through
the phase boundary. However, electrosorption, which is generally
defined as potential-induced adsorption on the surface of elec-
trodes, is a non-Faradaic process. In electrosorption processes such
as waste water treatment [8], desalination [9] and electrical double
layer applications [10], the performance of the electrode depends
on the surface chemistry and pore structure of electrode. How-
ever surface chemistry, surface area and pore size distribution
of activated carbons may also be affected from electrochemical
polarization applied during electrosorption and thus weakly acidic
∗ Corresponding author. Tel.: +90 242 3102315; fax: +90 242 2278911.
E-mail address: eayranci@akdeniz.edu.tr (E. Ayranci).
may be formed [11].
Oxidation of carbon materials by nitric acid has been exten-
sively studied by some authors [12,13]. The effect of surface oxygen
groups on capacitance of samples was examined and it was found
that such groups cause an increase in the capacitance of carbon
material in strongly acidic medium. On the other hand, electro-
chemical oxidation of granular activated carbon by catholic and
anodic treatment in different electrolytic media, were studied by
several groups [14–16], but few studies focus on activated carbon
cloth or fibers [7,17–19]. The effect of polarization on the surface
morphologies of activated carbon fibers before and after polariza-
tion at the potential range between 0.2 and 0.8 V were characterized
and it was found that surface area and average pore size of fibers
increase with increasing anodic potential [11]. In another study
[17], electrochemical regeneration of activated carbon fiber previ-
ously exhausted with acid orange-7 was carried out by polarizing
the activated carbon fiber in a potential range from −10 to 10 V and
the changes in the amount of surface functional groups upon polar-
ization were determined. Gulyas et al. [18] investigated the changes
in surface chemistry of activated carbon fiber upon polarization in
different electrolytes by FTIR spectroscopy and they found that the
surface chemistry is determined by the type of electrolyte, as well
as by the other conditions of the process. Recently, Harry et al. [19]
oxidized the activated carbon cloth galvanostatically in KCl solu-
tion for different periods of time and examined the changes in the
surface properties.

0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.12.018
 E. Bayram, E. Ayranci / Electrochimica Acta 56 (2011) 2184–2189 2185

The potential at point of zero charge, EPZC , which is believed to be was obtained from (Vtot − Vmicro). Pore size distribution (PSD) was
an important parameter in determining the surface properties of an calculated using density functional theory (DFT) method. All the
electrode, has not received much attention in the limited number calculations were done by using the software of the instrument
of works reported on electrochemical oxidation of ACC or activated which assumes slit shapes for pores. A LECO 932 elemental ana-
carbon fiber. It is also believed that there is a lack of information lyzer was used to determine the percentages of carbon, oxygen,
on the effects of variables such as the nature of electrolyte and nitrogen, hydrogen and sulfur contents of ACC samples.
both the anodic and cathodic potential on the surface chemistry of The pHPZC value of ACC, defined as the pH of the solution when
ACC. A systematic study is necessary not only for electrochemical the net charge on ACC piece dipped into it is zero, was determined
oxidation as an alternative to the traditional chemical methods, with a procedure described in our previous work [22] using batch
but also for the development of new electrochemical technologies. equilibrium method reported earlier [23].
Besides it would be helpful to solve the problems, which may be The contents of acidic and basic surface groups on ACC samples
encountered during the electrochemical processes. were determined according to the Boehm method [22,24].
The aim of this study is to investigate systematically the changes
in surface properties, chemical composition and electrochemical 2.4. Electrochemical characterization of ACC samples
properties of ACC upon polarization at different anodic and cathodic
potentials in different electrolytes. All the electrochemical experiments were performed using
a standard three electrode system connected to the potentio-
2. Experimental stat/galvanostat interfaced to a computer with its own software. In
cyclic voltammetric studies, WE was an ACC piece of about 12 mg
2.1. Materials attached to a Pt wire, CE was a Pt plate, and RE was a Ag/AgCl elec-
trode. It is known that SO4 2− ions do not specifically adsorb onto
The high specific surface area ACC was obtained from Spectra ACC [8]. Thus 0.01 M Na2 SO4 (pH ∼ 6.8) solution was used as inert
Corp., coded as Spectracarb 2225. Although the full details of its supporting electrolyte.
mode of preparation are regarded as proprietary, it originates from Potential of zero charge (EPZC ) is the potential at the surface
pyrolysis of phenolic polymer fibers followed by heat treatment in when the net charge on it is zero. Immersion potential of a material
O2 -free N2 between 800 ◦ C and 900 ◦ C for some hours [20]. in the absence of any Faradaic process should be equal to EPZC [25].
Reagents used in this study were Merck grade. High purity water Thus the immersion potential of ACC to be measured in Na2 SO4
used in all experiments was from an 18.2 M Milli-Q UV (Millipore) solution will be its EPZC . For the measurement of EPZC , ACC sample
water treatment system. of about 12 mg dipped into 0.01 M Na2 SO4 solution was degassed
first by applying vacuum for a short period of time and then by bub-
2.2. Treatment of ACC bling N2 through the solution. Then the potential was measured
against a Ag/AgCl reference electrode. The subsequent open cir-
Before it was used in the experiments, the ACC was first washed cuit potentials (EOC ) of ACC samples were determined in the same
with warm deionized water by a procedure described in our pre- manner.
vious works [8,21]. The washed and dried ACC was then cut into
desired dimensions, weighed accurately and then kept in a desic- 3. Results and discussion
cator for further use. For the purpose of polarization treatments, the
ACC piece was dipped into a 0.01 M Na2 SO4 or 0.01 M KH2 PO4 /KOH 3.1. Surface textural characteristics of ACC samples
(pH 6.7) solution to act as a working electrode (WE), a Pt plate
was used as a counter electrode (CE) and a Ag/AgCl electrode (BAS, Physical characteristics of porous materials such as surface area
MF-2030, Bioanalytical systems Inc. USA) as a reference electrode and pore structure are important in sorption processes. The surface
(RE). N2 was bubbled through the solution to remove the dis- area is the reactive zone where chemisorption and/or physisorption
solved gasses. ACC pieces polarized, either in 0.01 M Na2 SO4 or occur. Alternatively, pore size and shape will define the process
in 0.01 M KH2 PO4 /KOH solution, at constant applied potentials of performance, since kinetics of adsorption is directly related to
+0.5 V, +0.8 V, +1.5 V, +5.0 V or −1.5 V with the use of a potentio- adsorbate intraparticle diffusion. The SBET obtained from N2 adsorp-
stat/galvanostat (Gamry Instruments Inc.) for a period of 90 min, tion isotherm data and the calculated Vtot , Vmicro and Vmeso data of
were washed with water at 60 ◦ C and then dried at room temper- pristine and polarization-treated ACC samples are given in Table 1.
ature. Those polarized in 0.01 M Na2 SO4 were labeled as ACC + 0.5, All ACC samples analyzed in this study have high specific BET
ACC + 0.8, ACC + 1.5, ACC + 5.0 or ACC − 1.5 and those polarized in surface area which decreases with electrochemical polarization.
0.01 M KH2 PO4 /KOH were labeled as ACC + 1.5T or ACC − 1.5T. The Maximum decrease observed in SBET relative to pristine ACC was
sign and figures in these labels indicate the sign and magnitude of about 18% with ACC + 5.0 sample. The decrease in anodically polar-
the potential applied during polarization. ized ACC samples could be a consequence of a combination of the
direct oxidation by the anodically polarized electrode according to
2.3. Surface properties and textural characterization of ACC Eqs. (1) and (2) [8] and the indirect oxidation by electrogenerated
samples
Table 1
Prior to nitrogen adsorption experiments to determine surface Surface textural properties of pristine and polarization treated ACC samples.
properties, ACC samples were degassed at 130 ◦ C under vacuum
Sample SBET (m2 g−1 ) Vtot (cm3 g−1 ) Vmicro (cm3 g−1 ) Vmeso (cm3 g−1 )
(up to 10−6 torr) for 12 h. The N2 adsorption data were obtained
with a Quantachrome Autosorb-1-C/MS apparatus over a relative ACC 1596 0.697 0.695 0.17 × 10−2
pressure ranging from 10−6 to 1. The specific surface area was ACC + 0.5 1549 0.671 0.660 1.01 × 10−2
ACC + 0.8 1541 0.655 0.654 0.17 × 10−2
calculated from the linear part of the adsorption isotherm using ACC + 1.5 1488 0.652 0.643 0.91 × 10−2
the BET method (SBET ). The total pore volume (Vtot ) was esti- ACC + 5.0 1300 0.559 0.540 2.00 × 10−2
mated from the amount adsorbed at a relative pressure of 0.99. ACC − 11.5 1518 0.662 0.659 0.22 × 10−2
The Dubinin–Radushkevich (DR) theory was employed for esti- ACC + 1.5T 1422 0.619 0.603 1.60 × 10−2
ACC − 11.5T 1565 0.686 0.673 1.31 × 10−2
mating the micropore volume (Vmicro ). Mesopore volume (Vmeso )
2186 E. Bayram, E. Ayranci / Electrochimica Acta 56 (2011) 2184–2189

oxidant species, such as hydroxyl radicals according to Eqs. (3) and Table 2
Elemental analysis (wt.%) results and pHPZC values of pristine and polarization-
(4) [26].
treated ACC samples.
C + H2 O → C–O + 2H+ + 2e− (1) Sample C H O N S pHPZC
+ −
C + H2 O → C–OH + H + e (2) ACC 85.9 1.13 13.0 – – 7.40
ACC + 0.5 85.1 1.12 13.8 – – 6.83
H2 O → HO• + H+ + e− (3) ACC + 0.8 81.4 1.15 17.4 – – 6.34
ACC + 1.5 76.8 1.45 21.8 – – 3.95
− −
OH → HO• +e (4) ACC + 5.0 68.3 0.98 30.8 – – 3.21
ACC − 1.5 84.4 1.14 13.3 – – 7.32
In the case of cathodic polarization, a competitive mechanism ACC + 1.5T 73.7 1.31 25.0 – – 3.83
involving the indirect oxidation by peroxide species which are elec- ACC − 1.5T 77.2 1.47 21.4 – – 7.59
trogenerated during electroreduction of oxygen traces dissolved
in the electrolyte in contact with air and the direct reduction (Eq.
(5)) by the electrode potential [27] may be responsible from the noted that ACC − 1.5T, ACC + 5.0 and ACC + 1.5T samples have total
decrease in SBET . micropore volumes of about 0.262, 0.159, 0.240 cm3 g−1 , respec-
tively, for pore sizes below 0.8 nm (not shown in Fig. 1 but reflected
C + H2 O + e− → C–H + OH− (5) into Vmicro values given in Table 1).
Moreover, we have found earlier [28] that in an unbuffered solu-
tion and non-divided electrochemical cell, the pH of the solution 3.2. Chemical characterization
effectively decreases with negative polarization of ACC. In acidic
solutions, the cathodic reduction of protons emerges as another Chemical characterization of ACC samples was accomplished
important competitive reaction [29]. So we can conclude that all by determining their chemical composition and surface functional
these complex processes taking place during cathodic polarization groups, and by measuring their pHPZC . The results of elemental
leading to oxidation of ACC on the one hand and to the develop- analysis of ACC samples are given in Table 2. In all polarization-
ment of the surface porosity of ACC as a result of electroreduction treated samples, the carbon content is decreasing while the oxygen
on the other hand are causing an overall decrease in SBET . content, obtained by difference, is increasing with the poten-
It is known that oxidation of activated carbon takes place at tial applied during polarization-treatment. The absence of sulfur
the edges of basal planes and at the entrance of pores due to the shows that no electrochemical reaction involving the SO4 2− ions
structural tenseness at these locations, leading to conversion of is taking place during the polarization of ACC in 0.01 M Na2 SO4
micropores to mesopores. The increasing trend in Vmeso and the solution.
decreasing trend in Vmicro values of polarization-treated ACC sam- The pHPZC value is important in determining the net charge on
ples with respect to the pristine ACC (Table 1) can be taken as ACC surface in solutions. It depends on the type and number of
evidences for the oxidation of ACC upon polarization. The conver- functional groups present on ACC. It will be shifted to lower val-
sion of micropores to mesopores also explains the decrease in SBET ues with oxidation of ACC due to the introduction of weakly acidic
values upon polarization discussed above. Further proofs for the functional groups onto its surface [30]. The curves for the change
oxidation can be seen in the pore size distribution (PSD) curves of in final pH (pHfinal ) as a function of initial pH (pHinitial ), derived
ACC samples given in Fig. 1 where micropores above 0.8 nm size for the determination of pHPZC , are given in Fig. 2a for anodically
have been displayed only. All ACC samples are microporous and polarized ACC samples and in Fig. 2b for cathodically polarized
most of the pores in all samples except ACC − 1.5 have sizes around ACC samples. The pHPZC values determined from plateau in curves
1.1 nm. The intensity of the peak at 1.1 nm decreases with increas- are given in the last column of Table 2. It is seen that ACC sur-
ing anodic polarization. In the curve of ACC − 1.5, the main peak face has been oxidized with positive polarization in both 0.01 M
is shifted to the pore size of around 0.9 nm probably due to the Na2 SO4 and 0.01 M KH2 PO4 /KOH buffer solutions for a period of
rearrangement of pores as a result of electroreduction. It should be 90 min, the extent of oxidation being increased with increasing
applied potential. pHPZC value of ACC + 1.5T is lower than that of
ACC + 1.5, showing the higher oxidation capacity in buffer solu-
ACC tion than in Na2 SO4 solution. pHPZC value of ACC did not change
0,4 ACC-1.5T significantly upon polarization-treatment at −1.5 V in buffer solu-
ACC-1.5
ACC+0.5
incrimental pore volume / cm 3 g-1

ACC+5.0
ACC+1.5 11 a 11 b
0,3
ACC+0.8
ACC+1.5T 9 9
pHPZC

0,2 7 7
pHfinal
pHfinal

5 5

0,1
3 3

1 1
1 3 5 7 9 11 1 3 5 7 9 11
0
pHinitial pHinitial
0,8 1,2 1,6 2 2,4
pore width / nm Fig. 2. pHinitial vs. pHfinal plots for the determination of pHPZC values of (a) ACC (♦);
ACC + 0.5 (); ACC + 0.8 (); ACC + 1.5 (䊉); ACC + 5.0 (*); ACC + 1.5T () and (b) ACC
Fig. 1. PSD of ACC and polarization treated ACC samples. (♦); ACC − 1.5 (); ACC − 1.5T () samples.
 E. Bayram, E. Ayranci / Electrochimica Acta 56 (2011) 2184–2189
Fig. 3. Surface functional groups of ACC and polarization treated ACC samples.
tion and increased slightly upon the same treatment in Na2 SO4
solution. This is rather surprising, considering the possible indirect
oxidation of ACC by peroxide species upon cathodic polarization
discussed above in relation to changes in SBET values. Decrease in
surface acidity, thus increase in pHPZC , of granular activated car-
bon upon cathodic polarization in 0.5 M Na2 SO4 solution was also
observed by Mehta and Flora [15]. At this point, the determina-
tion of the amount of surface functional groups gains a crucial
importance.
Functional groups and delocalized electrons of the graphitic
structure determine the apparent chemical character of an acti-
vated carbon surface. Some of the groups such as carbonyl, carboxyl,
phenolic hydroxyl and lactonic ones are acidic. The existence
of pyrone, chromene and quinone structures was postulated to
account for the basic nature of the carbon surface [31]. The sur-
face functional groups of ACC samples determined according to the
Boehm method [24] are given in Fig. 3. An increase in the amount
of total acidic and carboxylic groups and a decrease in total basic
groups was observed with increasing applied anodic potential. Lac-
tonic group content of ACC decreased with applied anodic potential
up to 0.8 V and then decreased at 1.5 V in Na2 SO4 solution. This
suggests that at sufficiently high (as high as 1.5 V) potentials CO-
type groups are further oxidized to CO2 -type groups. It is obvious
that total acidic, lactonic and phenolic group contents of ACC + 1.5T
are appreciably higher than those of ACC + 1.5. This can be corre-
lated with the pH values of the Na2 SO4 and KH2 PO4 /KOH solutions
reached at the end of polarization treatment. The pH values of
Na2 SO4 and KH2 PO4 /KOH buffer solutions were measured as 10.1
and 6.5 at the end of polarization treatment, respectively. Lower pH
of buffer solution than Na2 SO4 solution leads to CO2 -type groups
to form more favorably than CO-type groups [26]. This may fur-
ther explain the higher acidic group contents in ACC + 1.5T than
ACC + 1.5. These findings are consistent with the pHPZC values of
ACC samples given in Table 2, and SBET values given in Table 1.
On the other hand ACC − 1.5T was found to possess more acidic
and basic groups compared to the pristine ACC sample. Insignif-
icant change in pHPZC from ACC to ACC − 1.5T, discussed above,
can now be attributed to the balancing effects of the simultane-
ous increase in acidic and basic groups in ACC − 1.5T. Similarly
the slight increase in total basic groups in ACC − 1.5 compared
to ACC may be responsible for the slight increase in its pHPZC
(Table 2).
2187
1 mV/s
5 mV/s
0,06 10 mV/s
25 mV/s
0,02
I/A
-0,02
-0,06
-0,4 -0,1 0,2 0,5 0,8
E vs (Ag/AgCl)/V
Fig. 4. Cyclic voltammograms (CVs) of pristine ACC obtained in 0.01 M Na2 SO4 solu-
tion at different scan rates.
3.3. Electrochemical characteristics of ACC samples
In order to check the changes in the electrochemical proper-
ties of ACC upon polarization, cyclic voltammetric studies of ACC
samples were carried out. Cyclic voltammetric curves of pristine
ACC obtained in 0.01 M Na2 SO4 solutions at scan rates of 1, 5,
10 and 25 mV s−1 , between the potential limits of −0.3 and 0.8 V
vs. Ag/AgCl, are given in Fig. 4. Since SO4 2− ions do not specifi-
cally adsorb onto ACC [8], contribution of pseudo-capacitance to
be originated from specific adsorption of SO4 2− ions on cyclic
voltammograms (CVs) can be neglected. Therefore shape of CVs is
determined by the properties of ACC and mobility of Na+ and SO4 2−
ions in hydrated state. The quasi-rectangular shape of curves at the
scan rate of 1 mV s−1 reflects the capacitative behavior of pristine
ACC. When the scan rate is increased the shape of CVs are dis-
torted owing to increasing iR-drop which arises principally from the
complex internal distribution of resistance of electrolyte in porous
electrodes [32]. In the negative scan direction Na+ ions are stored
easily. On the other hand in the positive scan direction the Na+ ions
are desorbed while SO4 2− ions are adsorbed. The capacitance is
provided by these SO4 2− ions. However, due to the greater size and
thus lower mobility of sulfate ions in hydrated state in narrow pores
[33], voltammetric charging is slowed down in positive scan direc-
tion at relatively higher scan rates, distorting the quasi-rectangular
shape of curve.
The differences in CVs of anodically polarized ACC samples can
be seen in Fig. 5a and b recorded at 1 mV s−1 and 5 mV s−1 , respec-
tively. CVs of pristine ACC were also included in these figures
for comparison. At the slower scan rate (Fig. 5a), ACC exhibits a
weak maximum in both the anodic and cathodic current responses
around 0.1–0.5 V, probably due to a pseudo-capacitance contri-
bution from surface functional groups, e.g. of the quinonoid type
[32]. These maxima were shifted to around ∼0.3–0.7 V in anodic
scan direction and to around ∼0.0–0.3 V in cathodic scan direction
(Fig. 5a) for ACC + 1.5 and ACC + 1.5T samples. These two samples
show almost identical capacitative behavior. ACC and all anodi-
cally polarized ACC samples, except ACC + 5.0, show ideal capacitor
behavior exhibiting quasi-rectangular shapes at a scan rate of
1 mV s−1 . When the scan rate is increased to 5 mV s−1 (Fig. 5b),
shape of CVs is distorted as in pristine ACC (Fig. 4). The differences in
CVs of ACC samples become more significant. Capacitative behavior
of ACC electrode is gradually changing with the potential applied
during polarization-treatment. As a result of increased iR-drop
arising from oxidation upon polarization, the CVs show “resistive
behavior” at 5 mV s−1 . In this case, ACC exhibits very small capaci-
tative response or charge storage capacity on the anodic half-cycle,
resulting from high internal resistance and pore blockage charac-
teristics. It may be expected that the overcharge oxidation also
2188 E. Bayram, E. Ayranci / Electrochimica Acta 56 (2011) 2184–2189

0,01 a
ACC a 0,007 ACC
ACC+0.5 ACC-1.5
ACC+0.8 ACC-1.5T
0,006 ACC+1.5
ACC+5.0 0,003
ACC+1.5T
0,002

I/A
-0,001
I/A

-0,002

-0,005
-0,006

-0,009
-0,01 -0,4 -0,2 0 0,2 0,4 0,6 0,8
-0,4 -0,2 0 0,2 0,4 0,6 0,8

ACC b
0,026
ACC b 0,02 ACC-1.5
ACC+0.5 ACC-1.5T
ACC+0.8
0,016 ACC+1.5 0,01
ACC+5.0
ACC+1.5T
0,006
I/A
0
I/A

-0,004
-0,01

-0,014
-0,02
-0,024 -0,4 -0,2 0 0,2 0,4 0,6 0,8

-0,4 -0,2 0 0,2 0,4 0,6 0,8 E vs (Ag/AgCl)/V

E vs (Ag/AgCl)/V Fig. 6. Cyclic voltammograms (CVs) of pristine ACC and cathodically polarized ACC
samples obtained in 0.01 M Na2 SO4 solution at (a) 1 mV s−1 and (b) 5 mV s−1 .
Fig. 5. Cyclic voltammograms (CVs) of pristine ACC and anodically polarized ACC
samples obtained in 0.01 M Na2 SO4 solution at (a) 1 mV s−1 and (b) 5 mV s−1 .

The specific capacitance values calculated for pristine and

polarization-treated ACC samples are sufficiently high as expected
occurs at the internal micro-pores, which causes change in their
from high SBET values and decrease upon polarization-treatment of
interfaces or micro-porosity [32]. It is well known that the oxygen
ACC (Table 3). The loss in capacitance values upon polarization of
functionalities are mainly located at the edges of the basal planes
ACC samples is attributable to the decrease in the electrochemically
and the entrance of the pores in activated carbons [7]. Besides, the
accessible surface area of ACC and to the unavoidable blocking of
oxygen groups are ideal sites for water adsorption, favoring the
nano- or micropores, accompanied by increases in internal resis-
creation of water clusters at the entrance of the pores and block-
tance, as discussed above in relation to CVs. The huge decrease in
ing of access of the electrolyte ions down into their double-layer
specific capacitance of ACC upon polarization-treatment at 5.0 V,
interfaces. This observed decrease in voltammetric charges in both
in spite of its still huge surface area (1300 m2 g−1 ), is attributable
positive and negative scan direction proves the oxidation of ACC
to both blockage of pores and disruption of RC-network of ACC.
surface by polarization.
Since capacitance is related to the electrosorptive desalting capac-
The current responses in both anodic and cathodic scan direc-
ity, it can be concluded from the capacitance values in Table 3
tions for ACC decrease upon cathodic polarization-treatment at
that the desalting capacity of ACC decreases with the potential
−1.5 V in buffer solution (Fig. 6a). This further confirms the bal-
applied in polarization-treatment. The result that specific capac-
ancing effects of the increased acidic and basic surface functional
groups. CV of ACC − 1.5 seems to be almost identical with that of
ACC at the scan rate of 1 mV s−1 . However, at a scan rate of 5 mV s−1
Table 3
the differences in CVs of ACC samples become distinctive, sup-
Specific capacitance, EPZC and EOC valuesa of pristine and polarization-treated ACC.
porting the finding of different textural and chemical properties
of cathodically polarized ACC samples. Sample Specific capacitance EPZC (mV) EOC (mV)
(F g−1 )
The effect of anodic and cathodic treatments on CVs of ACC
can be quantified by calculating specific capacitance values of ACC ACC 266 164 187
samples. The specific capacitance in F g−1 of each ACC sample ACC + 0.5 254 218 240
ACC + 0.8 231 275 276
has been estimated from the voltammetric charge surrounded by
ACC + 1.5 189 298 272
the cyclic voltammetric curves at 5 mV s−1 scan rate according to ACC + 5.0 20 355 308
C = (qa + |qc |)/mV and are given in Table 3. Here, qa , qc , m and V ACC − 1.5 234 35 107
are the anodic and cathodic voltammetric charges on positive and ACC + 1.5T 202 310 268
ACC − 1.5T 206 40 133
negative sweeps, the weight of ACC and the potential range of cyclic
a
voltammetry, respectively. Potentials vs. Ag/AgCl.
 E. Bayram, E. Ayranci / Electrochimica Acta 56 (2011) 2184–2189 2189

itance decreases with blockage of pores upon oxidation is rather Acknowledgement

specific to the ACC material of this study. It is known that for some
other activated carbon materials an increase in capacitance occurs The support of this work by the Scientific Research Projects
upon oxidation [12,13]. Coordination Unit of Akdeniz University is acknowledged.
EPZC and EOC , the two important parameters closely related to
the surface chemistry of ACC, were measured and given in the last References
two columns of Table 3. Both EPZC and EOC are seen to increase [1] E. Ayranci, B.E. Conway, J. Electroanal. Chem. 513 (2001) 100.
with increasing anodic potential applied in polarization-treatment [2] M. Panizza, G. Cerisola, Electrochim. Acta 48 (2003) 1515.
of ACC and to decrease upon cathodic polarization-treatment. Ini- [3] A. Matsunaga, A. Yasuhara, Environ. Sci. Technol. 37 (2003) 3435.
[4] H. Cheng, K. Scott, P.A. Christensen, J. Electrochem. Soc. 150 (2003) 17.
tial potentials (EPZC ) are decreasing for anodically polarized ACC
[5] E. Bayram, E. Ayranci, Environ. Sci. Technol. 44 (2010) 6331.
samples and increasing for pristine and cathodically polarized ACC [6] E. Bayram, E. Ayranci, Carbon 48 (2010) 1718.
samples toward equilibrium (EOC ). Most of the surface groups are [7] C.O. Ania, F. Beguin, Water Res. 41 (2007) 3372.
of electron-acceptor character in ACC. Such groups increase in [8] E. Ayranci, B.E. Conway, J. Appl. Electrochem. 31 (2001) 257.
[9] L. Zou, L. Li, H. Song, G. Morris, Water Res. 42 (2008) 2340.
number upon polarization and withdraw more p-electrons from [10] W.G. Pell, B.E. Conway, J. Power Sources 96 (2001) 57.
graphene layers of ACC leading to an increase in both EPZC and [11] Y. Han, X. Quan, S. Chen, S. Wang, Y. Zhang, Electrochim. Acta 52 (2007) 3075.
EOC . Conversely, reduction of ACC increases the electron density [12] M.J. Bleda-Martinez, J.A. Marsia-Agullo, D. Lozano-Castello, E. Morallon, D.
Cazorla-Amoros, A. Linares-Solano, Carbon 43 (2005) 2677.
of graphene layers causing a decrease in EPZC and EOC values of [13] M.J. Bleda-Martinez, D. Lozano-Castello, E. Morallon, D. Cazorla-Amoros, A.
ACC, as observed in both ACC − 1.5 and ACC − 1.5T. Since anodi- Linares-Solano, Carbon 44 (2006) 2642.
cally polarized samples have more acidic groups, the dissociation [14] S.S. Barton, M.J.B. Evans, E. Halliop, J.A.F. MacDonald, Langmuir 13 (1997) 1332.
[15] M.P. Mehta, J.R.V. Flora, Water Res. 31 (1997) 2171.
of these groups with time causes the decrease in going from EPZC [16] M. Garcia-Oton, F. Montilla, M.A. Lillo-Rodenas, E. Morallon, J.L. Vazquez, J. Appl.
to EOC . On the other hand the adsorption of protons by graphene Electrochem. 35 (2005) 319.
layers in pristine and cathodically polarized ACC samples over- [17] Y. Han, X. Quan, X. Ruan, W. Zhang, Sep. Purif. Technol. 59 (2008) 43.
[18] J. Gulyas, E. Földes, A. Lazar, B. Pukanszky, Compos. Part A 32 (2001) 353.
comes the dissociation effect. This causes an increase from EPZC [19] I.D. Harry, B. Saha, I.W. Cumming, Carbon 45 (2007) 766.
to EOC in both samples proving the reduction of ACC upon cathodic [20] E. Ayranci, E. Bayram, J. Hazard. Mater. B 122 (2005) 147.
polarization. [21] N. Hoda, E. Bayram, E. Ayranci, J. Hazard. Mater. B 137 (2006) 344.
[22] E. Ayranci, N. Hoda, E. Bayram, J. Colloid Interf. Sci. 284 (2005) 83.
[23] B.M. Babic, S.K. Milonjic, M.J. Polovina, B.V. Kaludierovic, Carbon 37 (1999) 477.
4. Conclusions [24] H.P. Boehm, Advances in Catalysis, Academic Press, New York, 1966.
[25] H. Tobias, A. Soffer, J. Electroanal. Chem. 148 (1983) 221.
[26] R. Berenguer, J.P. Marco-Lozar, C. Quijada, D. Cazorla-Amorós, E. Morallón, Car-
ACC samples were found to be oxidized upon anodic polariza- bon 47 (2009) 1018.
tion, especially above the applied potential of 1.5 V. Specific surface [27] A. Soffer, J. Electroanal. Chem. 38 (1972) 25.
[28] E. Bayram, N. Hoda, E. Ayranci, J. Hazard. Mater. 168 (2009) 1459.
area, the amount of basic functional groups on the surface, pHPZC
[29] P.Z. Cheng, H.S. Teng, Carbon 41 (2003) 2057.
and specific capacitance of ACC decreased while the amount of [30] J.B. Donnet, R.C. Bansal, Carbon Fibres, Marcel Dekker, New York, 1984.
acidic functional groups on the surface, EPZC and EOC increased [31] L.R. Radovic, C. Moreno-Castilla, J. Rivera-Utrilla, Chemistry and physics of car-
upon anodic polarization treatment. Cathodic polarization caused bon, in: L.R. Radovic (Ed.), A Series of Advances, vol. 27, Marcel Dekker, New
York, 2001.
a decrease in specific surface area, capacitance, EPZC and EOC of ACC [32] J. Niu, W.G. Pell, B.E. Conway, J. Power Sources 156 (2006) 725.
which was taken as an indication of reduction took place on ACC [33] G. Salitra, A. Soffer, L. Eliat, Y. Cohen, D. Aurbach, J. Electrochem. Soc. 147 (2000)
surface. 2486.