Amperometric detection of hydrogen peroxide at nano-nickel oxide/thionine and celestine blue nanocomposite-modified glassy carbon electrodes

Article abstract of DOI:10.1016/j.electacta.2009.05.078

A simple procedure was developed to prepare a glassy carbon (GC) electrode modified with nickel oxide (NiOx) nanoparticles and water-soluble dyes. By immersing the GC/NiOx modified electrode into thionine (TH) or celestine blue (CB) solutions for a short

Full text of DOI:10.1016/j.electacta.2009.05.078

Electrochimica Acta 54 (2009) 6312–6321
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Amperometric detection of hydrogen peroxide at nano-nickel oxide/thionine and
celestine blue nanocomposite-modified glassy carbon electrodes
Abdollah Noorbakhsha, Abdollah Salimia,b,∗
a Department of Chemistry, University of Kurdistan, P.O. Box 416, Sanandaj, Iran
b Research Center for Nanotechnology, University of Kurdistan, P.O. Box 426, Sanandaj, Iran
a r t i c l e i n f o
a b s t r a c t
Article history:
A simple procedure was developed to prepare a glassy carbon (GC) electrode modified with nickel oxide
(NiOx) nanoparticles and water-soluble dyes. By immersing the GC/NiOx modified electrode into thion-
ine (TH) or celestine blue (CB) solutions for a short period of time (5–120 s), a thin film of the proposed
molecules was immobilized onto the electrode surface. The modified electrodes showed stable and a
well-defined redox couples at a wide pH range (2–12), with surface confined characteristics. In com-
parison to usual methods for the immobilization of dye molecules, such as electropolymerization or
adsorption on the surface of preanodized electrodes, the electrochemical reversibility and stability of
these modified electrodes have been improved. The surface coverage and heterogeneous electron transfer
rate constants (ks) of thionin and celestin blue immobilized on a NiOx-GC electrode were approximately
3.5 × 10−10 mol cm−2, 6.12 s−1, 5.9 × 10−10 mol cm−2 and 6.58 s−1, respectively. The results clearly show
the high loading ability of the NiOx nanoparticles and great facilitation of the electron transfer between
the immobilized TH, CB and NiOx nanoparticles. The modified electrodes show excellent electrocatalytic
activity toward hydrogen peroxide reduction at a reduced overpotential. The catalytic rate constants
for hydrogen peroxide reduction at GC/NiOx/CB and GC/NiOx/TH were 7.96 ( 0.2) × 103 M−1 s−1 and
5.5 ( 0.2) × 103 M−1 s−1, respectively. The detection limit, sensitivity and linear concentration range for
hydrogen peroxide detection were 1.67 ␮M, 4.14 nA ␮M−1 nA ␮M−1 and 5 ␮M to 20 mM, and 0.36 ␮M,
7.62 nA ␮M−1, and 1 ␮M to 10 mM for the GC/NiOx/TH and GC/NiOx/CB modified electrodes, respectively.
Compared to other modified electrodes, these modified electrodes have many advantages, such as remark-
able catalytic activity, good reproducibility, simple preparation procedures and long-term stabilities of
signal responses during hydrogen peroxide reduction.
Received 13 January 2009
Received in revised form 24 May 2009
Accepted 24 May 2009
Available online 6 June 2009
Keywords:
Thionine
Celestine blue
Nickel oxide
Nanoparticles
Modified electrode
H2O2 detection
© 2009 Elsevier Ltd. All rights reserved.
1. Introduction
ties, low resistivity, and remarkable redox properties of metal oxide
particles and nanoparticles, they are suitable compounds for the
Metal oxide, oxyhydroxide and their relevant metal alloys
are extensively used in many different areas, such as corro-
sion protective coatings, electrochemical capacitors, the electronic
industry, magnetic nanostructures, photochemical energy conver-
sion, lithium ion batteries and display technology [1–5]. Moreover,
metal oxide films are the most interesting class of materials in
electrocatalysis and pH sensing [6,7]. Many of the metal oxides
are indirect band-gap semiconductors with electrical and optical
properties that are exploited for many different applications, such
as oxygen storage, electrochemical capacitors and super capacitors
[8], transparent conducting electrodes [9], electrochromic materi-
als [10] and semiconductor photoelectrodes [11]. In addition, due
to the low production cost, high stability, good electrical proper-
fabrication of gas sensors [12], lithium ion batteries and storage
materials [13]. Furthermore, metal oxide particles or nanoparticles
aresuitablematricesandnovelcandidatesfortheimmobilizationof
different redox molecules and biomolecules due to their high elec-
trical conductivity, high surface area, wide electrochemical working
window, excellent substrate adhesion and stable chemical, elec-
trochemical and physical properties. The electrocatalytic activity
of different electron transfer mediators, when immobilized onto
metal oxide films, is a new challenge in sensor fabrication and elec-
trode modification technologies [14,15].
Due to the high electron transfer efficiency and low cost of
water-soluble dyes, they have been used as redox mediators in elec-
trocatalytic processes [16–19]. Since the dye molecules will pollute
the reference and counter electrodes, it is not convenient to add
them as electron transfer mediators to the sample solution. There-
fore, it would be preferable to immobilize the dye molecules on the
electrode surface. Several water-soluble redox dyes, such as methy-
lene blue [20], methylene green [21], meldola blue [22], azure-I
∗
Corresponding author at: Department of Chemistry, University of Kurdistan, P.O.
Box 416, Sanandaj, Iran. Tel.: +98 871 6624001; fax: +98 871 6624008.
E-mail addresses: absalimi@uok.ac.ir, absalimi@yahoo.com (A. Salimi).
0013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2009.05.078

A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
6313
and B [23,24], orthophenylendimines, [25], neutral red [26], tolu-
idine blue [27] and other azine compounds [28], can be used as
electron transfer mediators when immobilized on the electrode
surface. Immobilization of electron transfer mediators to electrode
surfaces is a key step for the design, fabrication and performance of
the sensors and biosensors.
Phosphate buffer solutions (PBS) (0.1 M) were prepared from
H3PO4, KH2PO4 and K2HPO4 and the pH values were regulated with
HCl and KOH solutions. Solutions were deaerated by bubbling high
purity (99.99%) argon gas through them prior to the experiments.
All electrochemical experiments were carried out at a temperature
of 25 0.1 ◦C.
The conventional methods for immobilization of electron
transfer mediators are physical adsorption, electropolymerization,
covalent attachment, cross-linking and entrapment in a carbon
paste matrix [29–34,20]. Although different modified electrodes
have been prepared with redox dye molecules, most of these
electrodes present quasi-reversible electrochemical behavior and
even ill-defined cyclic voltammograms with large background
currents. Therefore, the immobilization of different electroactive
dye molecules on the surface of various electrode materials is
receiving increasing interest in the field of chemically modified
electrodes, electrocatalysis and electroanalysis. Different nanoma-
terials, such as carbon nanotubes, gold nanoparticles and carbon
nanofibers, have been successfully used for the immobilization of
thionine on electrode surfaces [35–37]. The prepared nanocom-
posites have been used as sensitive sensors and biosensors for
the detection of fetoprotein [36], glucose [35] and ethanol [37].
Recently, we fabricated highly sensitive glucose biosensors based
on the immobilization of thionine and celestine blue onto a carbon
nanotube-modified glassy carbon electrode [38–40].
Metal oxide materials with micro- or nanostructure are promis-
ing candidate for immobilization substances because of their
significant mechanical strength, high surface area, excellent chemi-
cal stability, and high ability for immobilization of different electron
transfer mediators. Sensitive sensors and biosensors for hydrogen
peroxide detection have been fabricated with immobilizing thio-
nine, methylene blue, meldola blue, natural red and tolidine blue,
onto titanium oxide nanoparticles [41,42], manganeous oxide [43]
and niobium oxide [44,45].
Nickel oxide (NiOx) nanoparticles have received considerable
attention in recent years due to their catalytic, optical, electronic
and magnetic properties [46,47]. Because of the volume, quantum
size, surface and macroscopic quantum tunnel effects, nanocrys-
talline NiOx is expected to possess even better properties than
those of micrometer-sized NiOx particles [48–51]. Based on the
unique properties of NiOx nanoparticles, they can be used for the
immobilization of different molecules. The easy preparation, elec-
troinactivity in physiological pH solutions and high porosity are
advantages of nickel oxide nanomaterials for the entrapment of
electron transfer mediators. Recently, we successfully used NiOx
nanoparticles for the immobilization of biomolecules and their
applications as sensors and biosensors for hydrogen peroxide and
glucose detection [52–54]. In the present study, NiOx nanostruc-
tures were used for the immobilization of thionine and celestine
blue. The prepared nanocomposite has been used as an excellent
catalyst for H2O2 reduction at lower overpotentials. Cyclic voltam-
metry and amperometry have been used for the investigation of
the electrochemical properties and electrocatalytic activity of the
nanocomposite-modified electrode. The fabricated sensor was used
for the detection of micromolar or even lowers concentrations of
H2O2 at physiological pH, using hydrodynamic amperometry.
2.2. Apparatus and procedures
Electrochemical experiments were performed with a computer-
controlled ␮-Autolab modular electrochemical system (Eco Chemie
Ultecht, The Netherlands), driven with GPES software (Eco Chemie).
A conventional three-electrode cell was used with a Ag/AgCl/(sat
KCl) reference electrode, a Pt wire as the counter electrode and a
glassy carbon disk (modified and unmodified) as the working elec-
trode. Voltammetry on electrodes coated with thionine or celestine
blue nano-NiOx modified electrodes was carried out in buffers free
of dye molecules.
2.3. Modification of GC and GC/NiOx electrodes with thionine and
celestine blue
The glassy carbon electrode (2 mm diameter) was carefully pol-
ished with alumina powder (1 and 0.05 ␮m) on polishing cloth. The
electrode was placed in ethanol and sonicated to remove adsorbed
alumina particles. The electrodeposition of metallic nickel was car-
ried out using cyclic potential (20 scans between 0 and −1.0 V at a
scan rate of 50 mV s−1) in acetate buffer solution, pH 4, in a solu-
tion containing 1 mM nickel nitrate. The potential was repetitively
cycled (30 scans) from 0 to 1 V at a scan rate of 100 mV s−1 in fresh
phosphate solution for the electrodissolution and passivation of a
nickel oxide layer at a GC electrode [55]. An stable thin film of the
dye molecules adsorbed on the electrode surfaces by immersing the
GC/NiOx modified electrode in fresh buffer solution (pH 7) con-
taining 5 mM celestin blue or thionine, for 5–120 s. The effective
surface area of the electrode modified with nickel oxide nanoparti-
cles was determined to be 0.12 cm2 from the cyclic voltammogram
of 1 mM K3[Fe(CN)6] in buffer solution, pH 7. For the attachment of
thionine or celestine blue on the surface of the GC electrode with
electropolymerization, first, the glassy carbon electrode was care-
fully polished with alumina on a polishing cloth. Then, the potential
of the electrode was maintained at 1.1 V vs. the reference electrode
for 5 min in 0.1 M phosphate buffer solution. By cycling the potential
between 0.1 and −0.5 V (30 cycles) at a scan rate of 100 mV s−1, in
pH 7 phosphate buffer solution (0.1 M) containing 0.1 mM thionine
or celestine blue, a thin layer of the dye molecules was adsorbed
[56]. For adsorption of thionine or celestine blue on the surface
of the reactive GC electrode, the process was carried out in two
steps. First, the glassy carbon electrode was held under a constant
potential of 1.8 V for 20 min in 1 M sulfuric acid solution. Second,
the preanodized GC electrode was immersed in 0.1 mM thionine or
celestine blue solution for 1 h.
3. Results and discussion
3.1. Electrochemical properties of thionin and celestin blue
modified electrodes
2. Experimental
2.1. Chemicals and reagents
The formation and growth of electrodeposited nickel oxide
nanoparticles on a glassy carbon electrode were investigated by
scanning electron microscopy (SEM) and atomic force microscopy
(AFM) (Fig. 1A–C). As shown, nickel oxide particles grow by elec-
trodeposition on the amorphous glassy carbon surface. The average
size of the NiOx particle varies from under 50 nm to slightly less
than 200 nm. AFM images also show nickel oxide nanostructures
The chloride salts of celestine blue (C17 H18 ClN3O4) and thionin
acetate (3,7-diamino-5-phenothiazinium acetate) were obtained
from Aldrich and used as received. H2O2 (30%, w/w) was obtained
from Merck, and its diluted solution was prepared daily. Ni(NO3)2
and other reagents were of analytical reagent grade from Merck.

6314
A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
Fig. 1. (A) SEM image of the electrodeposited nickel oxide on glassy carbon, scale bar is 20 ␮m. (B) SEM image with higher magnification for the same sample, scale bar is
2 ␮m. (C) AFM images of the electrodeposited NiOx sample.
with average diameters ranging from 50 to 200 nm electrode-
posited on the GC electrode. Furthermore, the electrodeposition
of the nickel oxide layer on the electrode surface was verified by
recording cyclic voltammograms of the modified electrode in alka-
line solution (not shown). The cyclic voltammogram exhibits an
anodic peak at 0.4 V and a cathodic peak at 0.35 V, due to the oxi-
dation of the Ni(OH)2 phase to NiOx(OH) and the reduction of
NiOx(OH) to Ni(OH)2 [57].
Fig. 2 shows the cyclic voltammograms of NiOx, NiOx/thionine
and NiOx/celestine blue modified GC electrodes in buffer solution
(pH 2). As shown for the NiOx modified GC electrode, no redox
peaks were observed (voltammogram “a”) at a potential range of
0.4 to −0.2 V, which provides a potential window of the NiOx/GC
electrode for investigating the voltammetric behavior of celestin
blue and thionine. By immersing the GC/NiOx modified electrode in
1 mM celestin blue or thionine solution for 90 s, a stable thin layer of
dye molecules was adsorbed. As shown for GC electrodes modified
with nickel oxide nanoparticles and celestine blue, a well-defined
cyclic voltammogram with a peak potential separation of less than
15 mV and a peak current of 0.4 ␮A was observed at a formal poten-
tial of 0.130 V (Fig. 2 voltammogram “b”). For the thionin/NiOx/GC
modified electrode, a well-defined cyclic voltammogram with a
peak potential separation of 23 mV and a peak current of 1.4 ␮A
was observed at a formal potential of 0.07 V. The electrochemi-
cal properties of GC/NiOx nanoparticles modified with thionine
and celestine blue were compared with those of a GC electrode
modified with the proposed dye molecules using electropolymer-
ization or physical adsorption as the modification techniques. As
shown in Fig. 3a, by immersing the preanodized GC electrode in
10 mM thionin solution for 1 h, a cyclic voltammogram with low
peak current (0.1 ␮A) and a high peak potential separation (50 mV)
was observed. Fig. 3b shows a cyclic voltammogram of a GC elec-
trode modified by the electropolymerization of thionine. It can be
seen that, for a thionine film attached with electropolymerization,
a cyclic voltammogram with low redox response (a peak current
of 0.8 ␮A and a peak potential separation of 60 mV) was observed.

A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
6315
Fig. 2. (A) Cyclic voltamograms of: (a) NiOx nanoparticle modified GC electrode in 0.1 M phosphate buffer solution (pH 2), (b) NiOx nanoparticle/celestin blue modified GC
electrode under the conditions presented in (a), with an immersion time for electrode modification of 90 s, and a scan rate of 10 mV s−1. (B) Cyclic voltammograms of NiOx
nanoparticles/celestin blue modified GC electrode under the conditions presented in (A), with an immersion time of 100 s and a scan rate of 10 mV s−1
.
For voltammograms “a” and “b”, the peak separation between the
oxidation and reduction peaks is increased due to the increasing
resistance of the thicker film. However, for the GC/NiOx/thionine
modified electrode, a pair of well-defined redox couples, with low
peak potential separation (20 mV) and higher peak current (4 ␮A)
were observed (Fig. 3 voltammogram “c”), suggesting that the redox
reversibility of thionine is significantly improved. In addition to the
fact that the porous interfacial layer of the nano-NiOx modified
electrode with a high specific surface area increases the conduc-
tive area, thionine can penetrate through the conductive porous
channels onto the electrode more easily, leading to a higher redox
response. The same voltammetric behavior was observed for the
Fig. 3. Recorded cyclic voltammograms for the GC electrode modified with preanodization and the physical adsorption of thionin (a), GC electrode modified with the
electropolymerization of thionin (b), and glassy carbon electrode modified with thionin and nickel oxide nanoparticles, in 0.1 M phosphate buffer solution (pH 7), with an
immersion time of 100 s and a scan rate of 50 mV s−1 (c).

6316
A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
ing the scan rate from 10 to 600 mV s−1 as expected for redox
surface-controlled process. Moreover, the anodic peak currents
were almost the same as the corresponding cathodic peak currents
and the peak potential did not change with increasing scan rate.
The peak-to-peak potential separation is about 20 mV for sweep
rates below 100 mV s−1, suggesting facile charge transfer kinetics
over this range of sweep rate. At higher sweep rates, the plot of
the peak current vs. scan rate deviates from linearity and the peak
current becomes proportional to the square root of the scan rate
(Fig. 4B), indicating a diffusion-controlled process, which is reflec-
tive of the relatively slow diffusion of counter ions into the electrode
surfaces. At higher sweep rates (v > 500 mV s−1), the peak separa-
tions begin to increase, indicating the limitation of charge transfer
kinetics. Based on the Laviron theory [58], the electron transfer rate
constant (ks) and charge transfer coefficient (˛) can be determined
by measuring the variation of the peak potential with scan rate.
The values of the peak potentials were proportional to the loga-
rithm of the scan rate for scan rates higher than 2.0 V s−1 (Fig. 4C).
The slope of the ꢀEpc vs. log(v) was approximately 58.4 mV. Charge
transfer coefficient, ˛ = 0.503 was obtained using the equation
Ep = K − 2.3030(RT/˛nF) log (v) and the two electrons transferred for
GC/NiOx nanoparticle electrode modified with celestine blue. The
high conductivity of the celestine blue/nano-NiOx/GC electrode
nanocomposite film increases the electrical properties of the redox
processes. Large surface area is also available for celestine blue
intercalation. Therefore, NiOx nanoparticles can also be used as new
materials for the immobilization and electron transfer reactions of
celestine blue.
Cyclic voltammograms of NiOx/celestin blue modified glassy
carbon electrode with different concentration of celestin blue
obtained by soaking the NiOx/GC electrode in 5 mM celestin blue
solution for different times in phosphate buffer solution (pH 7) was
recorded (not shown). The results indicate that, by immersing the
electrode for 10 s in dye solution, a stable thin layer of celestin
blue was adsorbed at the surface of the electrode. By increasing
the immersion time, the surface concentration of celestin blue is
increased and begins to level off after 100 s. The same behavior
for adsorption celestine blue onto GC/NiOx nanoparticles electrode
was observed.
Fig. 4A shows the cyclic voltammograms of the GC/NiOx/celestin
blue modified electrode at different scan rates. As shown in the
inset of Fig. 4C the peak currents linearly increased with increas-
Fig. 4. (A) Cyclic voltammetric responses of a nano-NiOx/celestin blue modified GC electrode in phosphate buffer (pH 2) at scan rates (inner to outer) of 10, 20, 30, 40, 50, 60,
70, 80, 90, and 100 mV s−1. (B) and (C) Plots of peak currents vs. the square root of scan rate and scan rate. (D) Variation of peak potential vs. log v.

A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
6317
Fig. 5. (A) The first (a) and 220th (b) cyclic voltammograms of the nano-NiOx/thionin modified GC electrode in buffer solutions, pH 2. (B) Cyclic voltammograms of the
nano-NiOx/thionin modified GC in phosphate buffer solutions, pH 2 (a) and after immersing the modified electrode for 30 h in phosphate buffer solution (pH 2) (b), at a scan
rate of 200 mV s−1
.
celestin blue. Using this ˛ value in the following equation [58]:
no appreciable changes (less than 3%) observed either in the peak
current or the peak potential values after 220 cycles. In addi-
tion, it was found that the current value is decreased only by 5%
when the electrode was immersed in buffer solution (pH 2) for
30 h (Fig. 5B). The stability of the modified electrodes was eval-
uated by the same method in electrolyte solutions at pH 7 and
11. The results indicated that the modified electrode was stable
in acidic and natural solution, but, in alkaline media (pH 11), a
larger decrease was observed in the peak currents (10% after 100
cycles). Therefore, the mediator is prevented from leaching. Since
the method of electrode preparation is simple and fast (less than
2 min), the current decay is not a serious disadvantageous for
this modified electrode. The high stability of adsorbed thionine
against desorption in aqueous solution is related to the chemical
and mechanical stability of the NiOx film and the strong adsorp-
tion of thionine on the surface of nickel oxide nanoparticles. The
same results were obtained for the celestin blue/NiOx/GC modified
electrode. For the celestin blue/NiOx/GC electrode, after recording
300 continuous cyclic voltammograms between 0.35 and −0.13 V
in 0.1 M phosphate solution at pH 2, the anodic and catholic cur-
rents of the modified electrode decreased only 4% (not shown). This
result indicates that celestine blue strongly adsorbed on the surface
of the NiOx nanoparticles. Due to the chemical stability, electro-
chemical reversibility and high electron transfer rate constant of
thionin and celestin blue at the nano-NiOx modified electrode,
they can be widely used in electrocatalysis as electron transfer
mediators to shuttle electrons between analytes and substrate elec-
trodes.
ꢀ
ꢁ
RT
log ks = ˛ log(1 − ˛) + (1 − ˛) log ˛ − log
nFv
ꢀ
ꢁ
nFꢀE
− ˛(1 − ˛)
(1)
2.3 RT
An apparent electron transfer rate constant, ks = 6.12 s−1, was
estimated for the celestin blue/NiOx/GC modified electrode. For
the GC/NiOx/thionine modified electrode, the ˛ and ks values were
estimated to be 0.5 and 6.58 s−1, respectively. This value for the
electron transfer rate constant is larger than that obtained for the
self-assembled monolayer of thionine (0.02 s−1 and 1.7) [59,60],
thionin immobilized by cross-linking with glutaraldehyde on the
surface of a glassy carbon electrode (0.097 s−1) [61] and celes-
tine blue immobilized onto carbon nanotubes [39]. The large value
of the electron transfer rate constant indicates a high ability of
nickel oxide nanoparticles for promoting electrons between thion-
ine, celestine blue and the electrode surface.
The surface coverage concentration (ꢁ ) of celestin blue and thio-
nine was estimated by integrating the area under the cathodic
peak. According to the equation ꢁ = Q/nFA, and assuming a
number of two electrons transferred for thionine and celes-
tine blue, the surface concentrations of the proposed molecules
were about 3.5 × 10−10 mol cm−2 and 5.9 × 10−10 mol cm−2 for the
GC/NiOx/celestine blue and GC/NiOx/thionine modified electrodes,
respectively. These results indicate high loading ability of the NiOx
nanoparticles for immobilization of thionine and celestine blue.
For investigating the effect of pH on the redox response of mod-
ified electrodes, cyclic voltammograms of the NiOx/thionine/GC
electrode at various pH values (2–12) were recorded (Fig. 6). As
3.2. Stability and pH dependence of the NiOx based modified
electrodes
can be seen in the inset of Fig. 6, the formal potential of the surface
ꢀ
redox couple was pH dependent. A plot of Eo vs. pH values gives
The stability of the GC/NiOx/thionine modified electrode was
examined by continuous potential cycling between −0.15 and
0.25 V in 0.1 M phosphate buffer solution (pH 2) (Fig. 5A). As shown,
a straight line from pH 2–7 with a slope of 66.4 mV/pH, which is
very close to the anticipated Nernstian value of 59 mV for a two

6318
A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
Fig. 6. Cyclic voltammetric response of the thionine/nano-NiOx modified GC carbon electrode at a pH of 2, 4, 6, 8 and 10 (from right to left) at a scan rate of 50 mV s−1. The
inset shows the variation of standard potentials vs. pH values.
electron–two proton process.
(2)
Another straight line with a slope of 31.5 mV/pH is obtained from
thionine/nano-NiOx/GC electrode, the reduction of hydrogen per-
pH 8–12. This behavior is in accordance with a two electrons–one
proton process. The same observation was reported for thionine
when adsorbed onto carbon nanotubes [40]. As can be seen, a dra-
oxide starts at −0.1 V and an obvious catalytic reduction peak
appears at a potential of −0.27 V. As is shown, the cathodic peak
current was greatly enhanced in the presence of hydrogen per-
matic decrease was observed in the surface coverage for pH values
oxide and the oxidation peak current disappeared, suggesting
above 8. There is also a change in the slope for pH values above 8,
a typical excellent electrocatalytic reduction process. The same
which can be related to the deprotonation of the deposited thionine.
electrocatalytic behavior was observed for the celestine blue/nano-
NiOx modified glassy carbon electrode (Fig. 7B). By increasing the
The pKa of the surface deposited thionine (7.7) is very close to the
value reported in the literature for azure I with a similar structure
concentration of H2O2, the cathodic peak current of the modi-
(7.5) [23]. For the CB/nano-NiOx/GC electrode, the formal poten-
fied electrode increased, while its anodic peak current decreased,
tial of the aꢀdsorbed redox couple was pH dependent (not shown).
A plot of Eo vs. pH gives a straight line with a slope of 62 mV/pH,
which is very close to the anticipated Nernstian value of 59 mV for
a two electron–two proton process. Furthermore, with increasing
pH values, the peak current decreased, due to the desorption and
dissolution of anionic forms of celestine blue.
indicating a typical electrocatalytic reduction process (ECꢀ) (not
shown). The catalytic currents increased linearly with H2O2 con-
centration in the range of 1.0–10 mM, and were fitted to the
equation Ip (␮A) = 0.3677[H2O2] (mM) − 0.0316 ␮A and R2 = 0.9988.
The same cyclic voltammograms were observed for different
concentrations of hydrogen peroxide for the thionine/nano-NiOx
modified glassy carbon electrode. The variation of peak current vs.
H2O2 over the entire concentration range of 20–100 ␮M fitted the
equation Ip (␮A) = 0.21322 ␮A m M−1 + 0.063 ␮A, R2 = 0.9981.
Cyclic voltammograms of 5 mM H2O2 at different scan rates
were recorded (not shown). The peak current for the cathodic
reduction of hydrogen peroxide is proportional to the square root
of the scan rate, suggesting that the process is controlled by diffu-
sion of the analyte, as expected for a catalytic system. It can also
be noted that, by increasing the sweep rate, the peak potential for
the catalytic reduction of H2O2 shifts to more negative values and
the plot of the peak current vs. square rate of the scan rate deviates
from linearity (at v > 200 mV s−1), suggesting a kinetic limitation
in the reaction between the redox sites of adsorbed celestine blue
3.3. Electrocatalytic reduction of H2O2 at thionine and celestin
blue modified nano-NiOx/GC electrodes
The electrocatalytic reduction of H2O2 by celestine blue and
thionine/nano-NiOx modified GC electrodes was investigated by
cyclic voltammetry. Fig. 7A shows the cyclic voltammograms of
nano-NiOx/GC and thionine/nano-NiOx modified GC electrodes
in buffer solution (pH 7) in the absence and presence of hydro-
gen peroxide. There was no reduction peak observed at the
nano-NiOx/GC electrode in the absence or presence of H2O2 in
the potential range of 0.0 to −0.5 V, suggesting that the nano-
NiOx/GC electrode was inactive to the direct reduction of hydrogen
peroxide (Fig. 7A and B, voltammogram “b”). However, for the
and H2O2. Also, a plot of the scan rate-normalized current (Ip/v1/2
)

A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
6319
Fig. 7. (A) Recorded cyclic voltammograms of nano-NiOx modified GC electrode in phosphate buffer solution (pH 7) at scan rate 20 mV s−1 (a); (b) as (a) in the presence
12 mM H2O2, (c & d) as (a & b) for thionine/nano-NiOx modified GC electrode. (B) as (A) for nano-NiOx modified GC electrode and celestine blue/nano-NiOx modified GC
electrode in tha absence and presence 6 mM of H2O2.
vs. scan rate exhibited the characteristic shape of a typical ECꢀ cat-
alytic process. The same results were observed for the reduction
of hydrogen peroxide at the thionine/nano-NiOx modified GC elec-
trode. Based on the observed results, the following catalytic scheme
describes the reaction sequence in the reduction of hydrogen per-
oxide by a celestine blue redox center.
in 1 mM hydrogen peroxide at pH 7. According to the approach
of Andriex and Saveant, and using Fig. 1 in Ref. [62], the aver-
age values of the calculated Kcat are 7.96 ( 0.2) × 103 M−1 s−1 and
5.5 ( 0.2) × 103 M−1 s−1 for celestine blue and thionine, respec-
tively.
Celestineblue(oxdidized form) + 2H+ + 2e−
3.4. Amperometric detection of H2O2 at celestine blue and
thionine/NiOx nanoparticle modified electrodes
→ celestineblue(reduced form)
(3)
As shown in the previous section, the thionine and celestine
blue modified nano-NiOx/GC electrodes have excellent and strong
mediation properties and facilitate the low potential amperomet-
ric measurement of hydrogen peroxide. Fig. 8 shows a typical
current–time plot of the rotated nano-NiOx/thionine modified GC
electrode (rotation speed 1000 rpm) on successive additions of
5 ␮M and 1 mM of hydrogen peroxide at an applied potential of
−0.3 V (vs. Ag/AgCl). As shown, the modified electrode responded
rapidly and approached 95% of the steady-state current within 3 s.
The plot of current response vs. H2O2 concentration is shown
in the inset of Fig. 8. The calibration plot is linear over a wide
concentration range (5 ␮M to 20 mM), while for a high concentra-
tion of H2O2 the plot of current vs. analyte concentration deviates
from linearity (insets of Fig. 8). The linear least squares calibra-
tion curve over the range of 5–70 ␮M (14 points) is I (nA) = 4.8428
[H2O2] ␮M + 1.2195 nA with a correlation coefficient of 0.9994,
indicating that the regression line is fitted very well with the
experimental data and the regression equation can be applied in
the unknown sample determination. The detection limit (when
signal to noise ratio was 3) and sensitivity were 1.67 ␮M and
4.14 nA ␮M−1, respectively. Highly stable amperometric response
toward hydrogen peroxide is extremely attractive feature of the
Celestine blue (reduced form) + H2O2
K
cat
−→ celestine blue (oxdidized form) + 2H2O
(4)
For an ECꢀ mechanism, the Andriex and Saveant theoretical
model[62]canbeusedtocalculatethecatalyticrateconstant. Based
on this theory, a relation between the peak current and the con-
centration of substrate for slow scan rates and large catalytic rate
constant exists:
ꢀ
ꢁ
1/2
vF
RT
Ip = 0.44nFAD1/2
CS
(5)
where D and Cs are the diffusion coefficient (cm2 s−1) and the
bulk concentration (mol cm−3) of substrate (hydrogen peroxide),
respectively, and other symbols have their usual meanings. Low
values of Kcat result in values of the coefficient lower than 0.496.
For low scan rates (5–20 mV s−1), the average value of this coef-
ficient was found to be 0.309 and 0.314 for a celestine blue and
thionine NiOx/GC modified electrode, with a coverage of 3.5 × 10−10
and 5.9 × 10−10 mol cm−2 and a geometric area (A) of 0.12 cm2

6320
A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
Fig. 8. Amperometric response at the rotating thionine/nano-NiOx modified GC electrode (rotation speed 1000 rpm) held at −0.3 V in buffer solution (pH 7) for successive
additions of (A) 1 mM and (C) 5.0 ␮M H2O2. (B) and (D) Plots of chronoamperometric currents vs. H2O2 concentrations. (E) Recorded chronoamperogram for 100 mM H2O2
during 15 min.
Table 1
Analytical characteristics of dye derivatives-based hydrogen peroxide sensors and biosensors.
Electrode
Method
ks (s−1
)
Dynamic range
Limit of detection (␮M)
Sensitivity (␮A ␮M−1
)
TH1/HRP2/SAM3 [60]
HRP/TH/GC4 [61]
Amperometry
Amperometry
Choronoamperometry
Amperometry
Amperometry
Choronoamperometry
Amperometry
Amperometry
Cyclic voltammetry
Cyclic voltammetry
Amperometry
–
Up to 40
3.4–2000
9.9–649
Up to 1 mM
–
0.1
3
0.06
0.9
0.1
1
85
1.53
–
1.38
0.38
1.3
0.05
0.36
1.67
2.50 cm−2
–
0.097
0.12
–
–
–
HRP/MB5/WSG6 [32]
PTH7/Nafion/HRP/GC [56]
HRP/MDB8/FIOHM9 [33]
MB/Nafion/HRP/GC [21]
MB-HRP/NME10 [63]
–
–
Up to 0.6 mM
0.5 ␮M to 1.6 mM
4 ␮M to 2 mM
250–5000 ␮M
3.8–4070 ␮M
1.37–344 ␮M
3.44–3070 ␮M
10 ␮M to 6.0 mM
1.5–8100 ␮M
50–22,300 ␮M
1–10,000 ␮M
5–20,000 ␮M
0.075 cm−2
0.0135
–
Poly (MB) [22]
–
–
1.05
0.1
–
–
–
6.58
6.12
0.000494
–
0.1089
–
0.0115
0.00855
6.07
0.00414
0.00762
TH sol–gel electrode [64]
TH/MWCNTs/PIGE11 [34]
TH/CCE12 [65]
TH/MWCNTs/GC [38]
MB/HRP/MnO2/GC [43]
TH/DNA/nano-TiO2/GC [41]
TH/nano-NiOx/GC [this work]
CB13 /nano-NiOx/GC [this work]
Amperometry
Amperometry
Amperometry
Amperometry
Amperometry
1: Thionine, 2: Hourseradish peroxidase, 3: Self assembled monolayer electrode, 4: Glassy carbon electrode, 5: Methylen blue, 6: Wax spectrographic graphite, 7: Poly thionine,
8, Meldola blue, 9: Functionalized inorganic–organic hybrid material, 10: Carbon nanotube modified glassy carbon electrode, 11: Paraffin impregnated graphite electrode, 12:
Ceramic composite electrode, 13: Celestine blue.
nano-NiOx/thionine modified GC electrode. Fig. 8C shows the
amperometric response of 100 ␮M H2O2 during a prolonged 800 s
experiment. The response remains stable throughout the experi-
ment (only a 3% decrease in current was observed), indicating that
thionine is effectively stabilized with immobilization onto nickel
oxide nanomaterials. The same amperometric response for H2O2
reduction was obtained for nano-NiOx/celestine blue modified GC
electrode. The calibration plot is linear over a wide concentration
range (1.0 ␮M to 10 mM). Limit of detection is calculated based on
the blank signal plus three standard deviation of the blank signal
(3sb). The sensitivity and detection limit of this nanocomposite-
modified electrode were 0.36 ␮M and 7.62 nA ␮M−1, respectively.
These analytical parameters are comparable or even better than the
results reported for the electroanalytical determination of hydro-
gen peroxide with different sensors and biosensors using other
redox dye compounds as electron transfer mediators (Table 1).
4. Conclusion
Thionine and celestine blue immobilized onto glassy carbon
electrode modified with electrodeposited nickel oxide nanopar-
ticles. The nanocomposite-modified electrode showed stable and
reproducible electrochemical behavior at a wide pH range (2–12).
The thionine and celestine blue immobilized onto nickel oxide

A. Noorbakhsh, A. Salimi / Electrochimica Acta 54 (2009) 6312–6321
6321
nanoparticles can be used as excellent electron transfer media-
tors to shuttle electrons between hydrogen peroxide and substrate
electrodes. The proposed modified electrodes were used as sen-
sitive amperometric hydrogen peroxide sensors. These sensors
presented good stability as the current response remained almost
constant even after a long period of use, confirming the strong
adherence of the proposed dye molecules on the nickel oxide
nanoparticles. The fabricated hydrogen peroxide sensor possesses
excellent characteristics and performance, such as fast response
time, good stability, low detection limit and high sensitivity, mani-
festing that the immobilization of electroactive dye molecules onto
NiOx nanocomposites is a suitable method for sensor construction.
The detection limit, sensitivity and linear concentration range of the
GC/NiOx/TH and GC/NiOx/CB modified electrodes were 1.67 ␮M,
4.14 nA ␮M−1 and 5 ␮M to 20 mM, and 0.36 ␮M, 7.62 nA ␮M−1, and
1 ␮M to 10 mM, respectively. These analytical parameters are com-
parable or better than the values reported in the literature using
other dye redox-based modified electrodes.
[19] Y.L. Chen, J.H. Yuan, C.X. Tian, X.Z. Wang, Anal. Sci. 20 (2004) 507.
[20] J. Liu, S.L. Mu, Synth. Met. 107 (1999) 159.
[21] B. Wang, S. Dong, Talanta 51 (2000) 565.
[22] A. Vasilescu, S. Andreescu, C. Bala, S.C. Litescu, T. Naguer, J.L. Martr, Biosens.
Bioelectron. 18 (2003) 781.
[23] C.X. Cai, K.H. Xue, J. Electroanal. Chem. 427 (1997) 147.
[24] D. Shan, S. Mu, B. Mao, Electroanalysis 13 (2001) 493.
[25] J. Wang, L.H. Wu, L. Augnes, Anal. Chem. 63 (1991) 2993.
[26] X. Zhong, J. Chen, B. Liu, Y. Xu, Y. Kuang, J. Solid State Electrochem. 11 (2007)
463.
[27] A.S. Santos, A.C. Pereira, L.T. Kubota, J. Braz. Chem. Soc. 4 (2002) 495.
[28] A.A. Karrakin, E.E. Karrakina, H.L. Schmidt, Electroanalysis 11 (1999) 149.
[29] J. Ye, R.P. Baldwin, Anal. Chem. 60 (1988) 2263.
[30] Q. Chi, S. Dong, Anal. Chim. Acta 285 (1994) 125.
[31] A.R. de Lucca, A.S. Santos, A.C. Pereira, L.T. Kubota, J. Colloid Interf. Sci. 25 (2002)
113.
[32] X.Q. Lin, J. Chen, Z.H. Chen, Electroanalysis 12 (2000) 306.
[33] J. Zhang, B. Li, Z. Wang, G. Chen, S. Dong, Anal. Chim. Acta 388 (1999) 71.
[34] D.R.S. Jeykumari, S. Ramaprabhu, S.S. Narayanan, Carbon 45 (2007) 1340.
[35] L. Deng, Y. Wang, L. Shang, D. Wen, F. Wang, S. Dong, Biosens. Bioelectron. 24
(2008) 951.
[36] Y. Zhao, R. Yuan, Y. Chai, Y. Zhang, X. Li, N. Wang, Q. Zhu, Sens. Actuators B 114
(2006) 631.
[37] L. Wu, M. McIntosh, X. Zhang, H. Ju, Talanta 74 (2007) 387–392.
[38] A. Salimi, A. Noorbaksh, H. Mamkhezri, R. Ghavami, Electroanalysis 19 (2007)
1100.
Acknowledgements
[39] A. Noorbakhsh, A. Salimi, E. Sharifi, Electroanalysis 20 (2008) 1788.
[40] A. Salimi, A. Noorbaksh, S. Soltanian, Electroanalysis 18 (2006) 703.
[41] P.H. Lo, S.A. Kumar, S.M. Chen, Colloids Surf. B: Biointerf. 66 (2000) 266.
[42] A.A. Hoffman, S.L.P. Dias, J.R. Rodrigues, F.A. Pavan, E.V. Benvenutti, E.C. Lima, J.
Braz. Chem. Soc. 19 (2008) 943.
The financial supports of the Iranian nanotechnology Initiative
and the Research Office of the University of Kurdistan are gratefully
acknowledged.
[43] X. Yang, X. Chen, X. Zhang, W. Yang, D.G. Evans, Sens. Actuators B 129 (2008)
784.
References
[44] A.S. Santos, R.S. Freire, L.T. Kubota, J. Braz. Chem. Soc. 13 (2002) 495.
[45] A.S. Santos, R.S. Freire, L.T. Kubota, J. Electroanal. Chem. 547 (2003) 135.
[46] C.L. Carnes, K.J. Klabunde, J. Mol. Catal. A, Chem. 194 (2003) 227.
[47] Y. Ichiyanagi, N. Wakabayashi, J. Yamazaki, S. Yamada, Y. Kimishima, E. Komatsu,
H. Tajima, Physica B 329–333 (2003) 862.
[48] L. Wu, Y. Wu, H. Wei, Y. Shi, C. Hu, Mater. Lett. 58 (2004) 2700.
[49] X. Yi, D. Zhong, Mater. Lett. 58 (2004) 276.
[50] D.L. Tao, F. Wei, Mater. Lett. 58 (2004) 3226.
[51] V. Biju, M.A. Khadar, Spectrochim. Acta Part A 59 (2003) 121.
[52] A. Salimi, E. Sharifi, A. Noorbakhash, S. Soltanian, Electrochem. Commun. 8
(2006) 1499.
[53] A. Salimi, E. Sharifi, A. Noorbakhash, S. Soltanian, Biophys. Chem. 125 (2007)
540.
[54] A. Salimi, E. Sharifi, A. Noorbakhash, S. Soltanian, Biosens. Bioelectron. 22(2007)
3146.
[55] D. Giovanelli, N.S. Lawrence, S.J. Wilkins, L. Jiang, T.G.L. Jones, R.G. Compton,
Sens. Actuators B 88 (2003) 320.
[56] R. Yang, C. Ruan, W. Dai, J. Deng, J. Kong, Electrochim. Acta 44 (1999) 1585.
[57] C. Zhang, S.M. Park, J. Electrochim. Soc. 136 (1989) 3333.
[58] E. Lavion, J. Electroanal. Chem. 101 (1979) 19.
[59] H.Y. Chen, D.M. Zhou, J.J. Xu, H.Q. Fang, J. Electroanal. Chem. 421 (1997) 21.
[60] C. Ruan, F. Yang, C. Lei, J. Deng, Anal. Chem. 70 (1998) 1721.
[61] J.J. Xu, D.M. Zhou, H.Y. Chen, Electroanalysis 10 (1998) 713.
[62] C.P. Andriex, J.M. Saveant, J. Electroanal. Chem. 93 (1978) 163.
[63] J.Z. Xu, J.J. Zhu, Q. Wu, Z. Hu, H.Y. Chen, Electroanalysis 15 (2003) 219.
[64] K. Thenmozhi, S.S. Narayanan, Sens. Actuators B125 (2007) 195.
[65] K. Thenmozhi, S.S. Narayanan, Biosens. Bioelectron. 23 (2007) 606.
[1] H. Zaman, A. Yamada, H. Fukuda, Y. Ueda, J. Electrochem. Soc. 145 (1998) 565.
[2] Y. Jyoko, S. Kashiwabara, Y. Hayashi, J. Electrochem. Soc. 144 (1997) L5–L8.
[3] I.G. Cassella, M. Gatta, J. Electroanal. Chem. 534 (2002) 31.
[4] G.X. Wang, Y. Chen, K. Konstantinov, J. Yao, J.H. Ahn, H.K. Liu, S.X. Dou, J. Alloys
Compd. 340 (2002) L5–L10.
[5] Y. Liu, M. Wang, F. Zhao, Z. Xu, S. Dong, Biosens. Bioelectron. 21 (2005) 984.
[6] G.M. da Silva, S.G. Lemos, L.A. Pocrifka, P.D. Marreto, A.V. Rosario, E.C. Pereira,
Anal. Chim. Acta 616 (2008) 36.
[7] E. Gill, K. Arshak, A. Arshak, O. Korostynska, Microsyst. Technol. 14 (2008) 499.
[8] S.L. Chou, J.Z. Wang, S.Y. Chew, H.K. Liu, S.X. Dou, Electrochem. Commun. 10
(2008) 1724.
[9] E. Budianu, M. Purica, F. Iacomi, C. Baban, P. Prepelita, E. Manea, Thin Solid Films
51 (2008) 1629.
[10] H. Kamal, E.K. Elmaghraby, S.A. Ali, K. Abdel Hady, Thin Solid Films 483 (2005)
330.
[11] K. Sayama, T. Oi, R. Abe, M. Yanagida, H. Sugihara, Y. Iwasaki, Solar. Energy Mater.
Solar Cells 90 (2006) 2429.
[12] R.J. Wu, J.G. Wu, T.K. Tsai, C.T. Yeh, Sens. Actuators B 120 (2006) 104.
[13] H. Kim, M.G. Kim, T.J. Shin, H.J. Shin, J. Cho, Electrochem. Commun. 10 (2008)
1669.
[14] Y. Yavuz, A.S. Koparal, J. Hazard. Mater. B 136 (2006) 296.
[15] F. Manea, C. Radovan, J. Schoonman, J. Appl. Electrochem. 36 (2006) 1075.
[16] C. Cai, J. Chen, Anal. Biochem. 332 (2004) 75.
[17] F.D. Munteanu, Y. Okamoto, L. Gorton, Anal. Chim. Acta 476 (2003) 43.
[18] A.M. Lazarin, C. Airoldi, Sens. Actuators B 107 (2005) 446.