Interaction between nickel hydroxy phthalocyanine derivatives with p-chlorophenol: Linking electrochemistry experiments with theory

Article abstract of DOI:10.1016/j.electacta.2010.10.007

In this work the interaction between peripherally (β) substituted nickel tetrahydroxyphthalocyanines (β-NiPc(OH)₄ and β-Ni(O)Pc(OH)₄) with p-chlorophenol is theoretically rationalised by performing calculations at B3LYP/6-31G(d) level. Density functional theory (DFT) and molecular orbital theory are used to calculate the condensed Fukui function for phthalocyanine derivatives and p-chlorophenol, in order to determine the reactive sites involved when p-chlorophenol is oxidized, and to compare theoretically predicted reactivity to experimentally determined electrocatalytic activity. Electrocatalytic activities of adsorbed NiPc derivatives: ads-α-NiPc(OH)₈-OPGE (OPGE = ordinary poly graphite electrode), ads-α-NiPc(OH)₄-OPGE and ads-β-NiPc(OH)₄-OPGE are compared with those of the polymerized counterparts: poly-α-Ni(O)Pc(OH)₈-OPGE, poly-α-NiPc(OH) ₄-OPGE and poly-β-NiPc(OH)₄-OPGE, respectively.

Full text of DOI:10.1016/j.electacta.2010.10.007

Electrochimica Acta 56 (2010) 706–716
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Interaction between nickel hydroxy phthalocyanine derivatives with
p-chlorophenol: Linking electrochemistry experiments with theory
Samson Khene, Kevin Lobb, Tebello Nyokong∗
Department of Chemistry, Rhodes University, Grahamstown 6140, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work the interaction between peripherally (␤) substituted nickel tetrahydroxyphthalocyanines
(␤-NiPc(OH)4 and ␤-Ni(O)Pc(OH)4) with p-chlorophenol is theoretically rationalised by performing cal-
culations at B3LYP/6-31G(d) level. Density functional theory (DFT) and molecular orbital theory are used
to calculate the condensed Fukui function for phthalocyanine derivatives and p-chlorophenol, in order to
determine the reactive sites involved when p-chlorophenol is oxidized, and to compare theoretically
predicted reactivity to experimentally determined electrocatalytic activity. Electrocatalytic activities
of adsorbed NiPc derivatives: ads-␣-NiPc(OH)8-OPGE (OPGE = ordinary poly graphite electrode), ads-␣-
NiPc(OH)4-OPGE and ads-␤-NiPc(OH)4-OPGE are compared with those of the polymerized counterparts:
poly-␣-Ni(O)Pc(OH)8-OPGE, poly-␣-NiPc(OH)4-OPGE and poly-␤-NiPc(OH)4-OPGE, respectively.
© 2010 Elsevier Ltd. All rights reserved.
Received 8 August 2010
Received in revised form
26 September 2010
Accepted 2 October 2010
Available online 25 October 2010
Keywords:
Nickel phthalocyanine
Chlorophenol
Hardness
Softness
Condensed Fukui function
1. Introduction
nines (Fig. 1). The Fukui function will give us information on
possible reactive sites on the metallo phthalocyanine, where elec-
Organic materials, such as phthalocyanines, have been used at
electrode interface, in order to tune or improve the electron trans-
fer properties of the electrode [1–4]. Density functional theory
(DFT) methods have proved useful in elucidating chemical proper-
ties when molecules interact [5–8]. The DFT calculations are used
in this paper to study chemical potential (ꢀ), chemical hardness
(ꢁ) and chemical softness (S) of nickel phthalocyanine tetrasubsti-
tuted with hydroxyl at the peripheral (␤) positions (␤-NiPc(OH)4
and ␤-Ni(O)Pc(OH)4, the latter was formed following transformed
␤-NiPc(OH)4 in NaOH to form O–Ni–O bridges [9–11]) on inter-
action with p-chlorophenol. We also determine the reactive site
by making use of the condensed Fukui function [6,12]. The Fukui
function derived by Parr and Yang [6] is related to the electron
density in the frontier molecular orbitals (FMO) and as a result it
plays a crucial role in chemical selectivity [13]. In the previous work
[14], the reactivity of ␤-NiPc(OH)4 and ␤-Ni(O)Pc(OH)4 towards
the electrocatalysis of p-chlorophenol was determined using DFT
methods.
trocatalytic oxidation of analytes such as 4-chlorophenol may
occur. The Fukui function plots of three differently substituted
metallo phthalocyanines {non-pheripherally-(␣-NiPc(OH)4) and
peripherally tetrasubstituted (␤-NiPc(OH)4) and non-peripherally
octasubstituted (␣-NiPc(OH)8), Fig. 1} are comparatively studied
and the information was used to explain their catalytic activity
towards the oxidation of 4-chlorophenol. The electrocatalytic stud-
ies of these molecules towards the oxidation of 4-chlorophenol are
also presented in this work. For the electrochemical studies the
NiPc derivatives are adsorbed on ordinary poly graphite electrode
(OPGE) to form ads-␤-NiPc(OH)4-OPGE, ads-␣-NiPc(OH)4-OPGE
and ads-␣-NiPc(OH)8-OPGE or polymerized in NaOH to form,
poly-␤-Ni(O)Pc(OH)4-OPGE, poly-␣-Ni(O)Pc(OH)4-OPGE and poly-
␣-Ni(O)Pc(OH)8-OPGE, respectively. The latter contain O–Ni–O
bridges as discussed above.
2. Theoretical background: interaction and reactivity
descriptors
The aim of this paper is to make use of the Fukui function
in order to theoretically investigate if there will be any effect
on selectivity and reactivity of the ring, due to the hydroxy
group substituted at different positions on metallo phthalocya-
Chemical potential (ꢀ), chemical hardness (ꢁ) and chemical
softness (S) are three useful molecular indices in DFT framework.
According to Koopman’s theorem [15], the three quantities are
defined by Eq. (1)–(3):
ꢀ
ꢁ
∂E
1
2
ꢀ =
≈
( ∈ HOMO + ∈ LUMO
)
(1)
∗
Corresponding author. Tel.: +27 46 6038260; fax: +27 46 6225109.
E-mail address: t.nyokong@ru.ac.za (T. Nyokong).
∂N
v(r)
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.10.007

S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
707
OH
function f(r) which is defined by Eqs. (5) and (6):
ꢀ
ꢁ
ꢀ
ꢁ
HO
ıꢀ
∂ꢂ(r)
∂N
f (r) =
ꢂ(r) =
=
(5)
(6)
OH
ıv(r)
N
N
N
v(r)
N
N
N
N
OH
ꢀ
ꢁ
N
Ni
N
N
∂E
O
Ni
N
∂v(r)
N
N
N
N
N
OH
N
N
where ꢂ(r) is the electron density. In the DFT framework, the Fukui
function is a local reactivity descriptor and provides the reactive
site where the reaction occurs. The Fukui function gives the site of
higher reactivity because the function gives the largest variation in
the electronic density for a molecular system when it accepts or
donates electrons.
OH
OH
(A)
(B)
The Fukui function (Eq. (5)) is a quantity involving the electron
density of the atom or molecule in its frontier orbitals and has a
local quantity which has different values at different points [18].
This function is indicative of how the electronic density varies at
a constant external potential when there is a change in the num-
ber of electrons in the system. When electrons are added to the
LUMO or removed from the HOMO, the density as a function of N
has slope discontinuities, hence three different reaction indices can
be defined, f+(r), f−(r) and f0(r) which are the Fukui functions for
nucleophilic attack, electrophilic attack and radical attack, respec-
tively. They are defined by Eqs. (7)–(9):
HO
OH
OH
OH
OH
Cl
N
N
N
N
Ni
N
N
N
N
OH
OH
HO
OH
ꢀ
ꢁ
∂ꢂ(r)
(C)
(D)
f
f
+(r) =
≈ ꢂN+1(r) − ꢂN(r)
(7)
(8)
∂N+
v(r)
Fig. 1. Molecular structure of (A) ␤-NiPc(OH)4, (B) ␣-NiPc(OH)4, (C) ␣-NiPc(OH)8,
and (D) p-chlorophenol.
ꢀ
ꢁ
∂ꢂ(r)
∂N−
−(r) = −
≈ ꢂN(r) − ꢂN−1(r)
v(r)
ꢀ
ꢁ
ꢀ
ꢁ
1
2
∂ꢀ
∂N
1
2
∂2E
∂N2
1
2
ꢁ =
=
≈
( ∈ LUMO − ∈ HOMO
)
(2)
ꢀ
ꢁ
∂ꢂ(r)
∂N0
1
2
1
2
v(r)
v(r)
f 0(r) =
≈
[ꢂN+1(r) − ꢂN−1(r)] = [f +(r) + f −(r)]
v(r)
and
(9)
ꢀ
ꢁ
where f−(r), measures the reactivity towards an electrophile, f+(r),
measures reactivity towards a nucleophile and f0(r) measures reac-
tivity towards radical attack. The quantities ꢂN+1(r), ꢂN(r) and
ꢂN−1(r) are the electronic densities for the system with N + 1, N and
N − 1 electrons, respectively.
∂N
∂ꢀ
1
2ꢁ
S =
=
(3)
v(r)
where E is energy of interacting molecules, v(r) is the external elec-
trostatic potential an electron at position (r) experiences due to
the nuclei, N is the number of electrons, ∈HOMO is the energy of
the highest occupied molecular orbital (HOMO) and ∈LUMO is the
energy of the lowest unoccupied molecular orbital (LUMO). Chem-
ical hardness (ꢁ) governs reactivity (Eq. (2)) [15,16]. When studying
the chemical reactivity of two interacting molecules the chemical
potential (ꢀ) has an important role to play. In order for a reaction
to proceed via a low-energy path, an electron should be transferred
from a molecule with higher chemical potential to a molecule with
a lower chemical potential [17]. This means that the larger the dif-
ference in the chemical potential between the two molecules, the
easier the reaction will be.
If a reaction is to take place, the driving force in a reaction is the
equalization of the chemical potential between the two reacting
species. Changes in the number of electrons and the external poten-
tial will cause the chemical potential of a system to change. If only
infinitesimal changes in ꢀ are considered, the following relation is
established, Eq. (4):
When charges of two interacting molecules are small or may
be zero, for example in a reaction between two neutral molecules,
the maximal Fukui function (f(r)) site is preferred for the reaction
[18]. When the charges of the molecules are relatively large, then
the Fukui function (f(r)) needs to be minimised in order for the
reaction to proceed smoothly [18,19].
3. Computational aspects and calculation procedure
In this work, Gaussian 03 program [20] running on an Intel/Linux
cluster was used to perform all DFT calculations. The calculations
were done at the B3LYP level with basis set 6-31G(d) for both
optimization and excited energy calculations (using time depen-
dent density function theory, TDDFT). Gausview 4.1 program was
used for all visualization [20]. B3LYP employs Becke’s method of
using Lee–Yang–Parr’s gradient-correction, exchange-correlation
density functional, which includes a hybrid of the Hatree–Fock
exchange and the DFT exchange.
Geometry optimization for the ground state of a given molecule
A (e.g. ␤-NiPc(OH)4) was obtained. From this calculation, the value
of the energy for molecule A with N electrons (EA(N)) and the
Mulliken charges on the kth atom of the molecule (qAk(N)) were
obtained. Single point energy calculations were performed on
molecule A with N − 1 and N + 1 electrons. Accordingly EA(N − 1),
ꢀ
ꢁ
ꢀ
ꢁ
ꢂ
∂2E
∂N2
ıꢀ
dꢀ =
dN +
ıv(r)dr
(4)
ıv(r)
v(r)
N
The first term gives the chemical hardness shown in Eq. (2),
which is an important factor governing the reactivity of two chem-
ical species. The second term in Eq. (4) is related to the Fukui

708
S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
HO
OH
p-Toluenesulfonyl
chloride
OH
OTos
1-Pentanol
(reflux for 7 hr)
OH
OH
OH
OH
N
N
CN
CN
CN
CN
N
Ni
N
N
N
K2CO3, Acetone,
reflux(2hr)
DBU, Ni acetate
NaOH
N
N
OH
(1)
OTos
(2)
HO
OH
Scheme 1. Synthesis pathway for non-peripherally substituted ␣-NiPc(OH)8.
qAk(N − 1) and EA(N + 1), qAk(N + 1) are obtained from the calcula-
tions.
Ionization potentials (IPA) and electron affinities (EAA) for
molecule A, are calculated using Eqs. (10) and (11):
2,3-Dicyanohydroquinone (1) was converted to 3,6-bis(4-
methylphenylsulfonyloxy) phthalonitrile (2) following the litera-
ture methods [23]. For the synthesis of ␣-NiPc(OH)8, compound 2
(1 g, 2.13 mmol) and dry 1-pentanol (3 ml) were placed in a stan-
dard round bottom flask and refluxed for 30 min. Anhydrous nickel
acetate tetrahydrate (0.13 g, 0.53 mmol) was added to the hot solu-
tion followed by a drop of DBU. The mixture was further refluxed for
7 h to afford ␣-NiPc(OTos)8 (OTos = Tosyl). ␣-NiPc(OTos)8 (0.77 g,
0.53 mmol) was stirred in NaOH (1 mol dm−3) solution for 24 h
to afford ␣-NiPc(OH)8. The product was separated and purified
using 1 mol dm−3 sodium hydroxide, 1 mol dm−3 hydrochloric acid
and finally with water. ␣-NiPc(OH)8 was further stirred in hot
methanol:water mixture (1:1) and finally dried at 80 ◦C overnight.
Yield 0.22 g (60%). 1H NMR: ıH (DMSO), 7.3–7.8 (8H, m, Ar–CH),
5.8 (Ar–OH): IR spectrum (cm−1): 3314 (OH), 3057 (Ar–CH), 1588
(C C), 1554 (C N), 1491, 1474, 1361, 1286, 1172, 1142, 1081, 928,
812, 747, 704, 747. Calc. MS (ESI-MS) m/z: (NiPc(OH)8) Calculated.
699; Found: 699 [M+].
IPA = EA(N − 1) − EA(N)
EAA = EA(N + 1) − EA(N)
(10)
(11)
From the values obtained for A, absolute electronegativity,
absolute hardness and absolute softness values, ꢃA, ꢁA and SA,
respectively can be calculated using Eqs. (12)–(14):
1
2
ꢃA
ꢁA
SA
=
=
=
(IPA + EAA)
(IPA − EAA)
(12)
(13)
(14)
1
2
1
2ꢁA
A broad 1H NMR singlet due to hydroxy substituent at 5.8 ppm,
disappeared upon addition of D2O.
Mulliken charges for each of the kth atoms in A with N, N − 1
and N + 1 electrons, allow for the calculation of the condensed Fukui
function according to Eqs. (15)–(17):
4.3. Equipment
fA+k = qAk(NA + 1) − qAk(NA)
fA−k = qAk(NA) − qAk(N − 1)
(15)
(16)
1H NMR spectra were obtained, in deuterated solvents, using a
Bruker EMX 400 MHz or Bruker Avance II+ 600 MHz NMR spectrom-
eters. FT-IR spectra (KBr pellets) were recorded on a Perkin-Elmer
spectrum 100 ATR spectrometer. MALDI-TOF mass spectra were
recorded ABI Voyager DE-STR Maldi TOF instrument at University
of Stellenbosch.
1
2
1
2
fA0k
=
[qAk(NAk + 1) − qAk(NA − 1)] = [fA+k + fA−k
]
(17)
Condensed local softness is represented by Eqs. (18)–(20):
SA+k = SAf +
(18)
(19)
Ak
SA−k = SAf −
4.4. Electrochemical methods
Ak
SA0k = SAf 0
(20)
Cyclic voltammograms (CV) were obtained using Autolab poten-
tiostat PGSTAT 30 (Eco Chemie, Utretch, The Netherlands) driven
by the General Purpose Electrochemical Systems data processing
software. A conventional three-electrode set up with ordinary poly
graphite electrode (OPGE) as a working electrode, platinum wire
as counter electrode and Ag|AgCl (3 M KCl) as a reference electrode
was employed. The working electrode (OPGE) was constructed in
house and the conduction between the copper wire and OPG mate-
rial was facilitated by making use of high purity silver paint and
Teflon was used for the outer casing. The OPGE (area = 0.44 cm2)
was cleaned by polishing on 800 and 1200 grit emery paper, fol-
lowed by rinsing with ultra-pure Millipore water.
Ak
4. Experimental
4.1. Materials
Nickel acetate tetrahydrate, diazabicyclo{5.4.0}-undec-7-ene
(DBU), 1-pentanol, dimethylsulphoxide (DMSO), dicyano hydro-
quinone, sodium hydroxide, hydrochloric acid, sodium sulphide
nonahydrate (Aldrich), sodium nitrite and p-toluenesulfonyl chlo-
ride were from Aldrich.
4.2. Synthesis of NiPc derivatives
The NiPc derivatives (␤-NiPc(OH)4, ␣-NiPc(OH)4 and ␣-
NiPc(OH)8) were adsorbed onto the OPGE by dipping the electrode
in dry dimethysulfoxide (DMSO) containing NiPc derivative
and allowing the solvent to dry to give adsorbed deriva-
tives: ads-␤-NiPc(OH)4, ads-␣-NiPc(OH)4 and ads-␣-NiPc(OH)8,
respectively. The electrochemical transformation of the NiPc
derivatives in alkaline aqueous solution to form the interconnected
O–Ni–O oxo bridges was achieved by cycling ads-␤-NiPc(OH)4,
ads-␣-NiPc(OH)4 and ads-␣-NiPc(OH)8, in 0.1 mol dm−3 NaOH,
The synthesis of peripherally substituted nickel hydroxyph-
thalocyanine (␤-NiPc(OH)4) has been reported in the literature
[21]. The non-peripherally substituted derivative (␣-NiPc(OH)4)
was synthesized and characterized as reported for the ZnPc deriva-
tive [22]. The synthesis of non-peripherally octahydroxy nickel
phthalocyanine (␣-NiPc(OH)8) is briefly described as follows (see
Scheme 1):

S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
709
i
20
ii
iii
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5
E vs (Ag|AgCl)/V
Fig. 3. Cyclic voltammograms of (i) poly-␤-Ni(O)Pc(OH)4, (ii) poly-␣-Ni(O)Pc(OH)4
and (iii) poly-␣-Ni(O)Pc(OH)8 on OPG electrode in 0.1 mol dm−3 NaOH. Scan
rate = 100 mV s−1
.
ple began to form near 0.35 V. The cyclic voltammograms in Fig. 2A
can be explained by the formation of the O–Ni–O oxo bridges in
alkaline aqueous solution and are indication of the transformation
of the NiPc(OH)4 into the O–Ni–O oxo bridged to form poly-␣-
Ni(O)Pc(OH)4 [9–11]. Fig. 2B shows the cyclic voltammograms of
ads-␣-NiPc(OH)8 in 0.1 mol dm−3 NaOH. Unlike the case with poly-
␣-Ni(O)Pc(OH)4 (Fig. 2A) there was an immediate formation of
a pair of peaks associated with NiIII/NiII for poly-␣-Ni(O)Pc(OH)8
(Fig. 2B). This was followed by a gradual increase of the current with
scanning in 0.1 mol dm−3 NaOH. In the case of the reported ads-␤-
NiPc(OH)4 [14] the formation of NiIII/NiII complexes was immediate
compared to ads-␣-NiPc(OH)4 where cycling was needed for for-
mation of the couple as observed in Fig. 2A, showing that the ease
of formation of O–Ni–O bridges depends on the point of substitu-
tion, with formation being easier for the peripherally substituted
derivatives. Comparing the formation of poly-␣-Ni(O)Pc(OH)4 with
poly-␣-Ni(O)Pc(OH)8, the latter forms O–Ni–O bridges more readily
than the former in that the peaks were formed instantly in the latter.
Thus the ease of formation of the O–Ni–O bridges also depends on
the number of substituents. The scan rate dependence of the peak
current was linear, suggesting surface bound redox processes.
Fig. 2. Cyclic voltammograms of (A) ads-␣-NiPc(OH)4, and (B) ads-␣-NiPc(OH)8
adsorbed on the OPG electrode in 0.1 mol dm−3 NaOH, forming poly-␣-Ni(O)Pc(OH)4
and poly-␣-Ni(O)Pc(OH)8, respectively. Scan rate = 100 mV s−1
.
giving the polymerized derivatives: poly-␤-Ni(O)Pc(OH)4, poly-␣-
Ni(O)Pc(OH)4 and poly-␣-Ni(O)Pc(OH)8, respectively.
The type of substituent also plays
a role. For example
5. Results and discussion
nickel tetrasulfophthalocyanine (NiTSPc) readily forms O–Ni–O
oxo bridges without the need of activation, whereas for nickel
tetraamino phthalocyanine (NiTAPc), activation is needed [9–11].
5.1. Analysis of derivatives on OPGE
Fig.
3
overlays the cyclic voltammograms of (i)
OPGE is made of ordered sheets of hexagonally bonded carbon
atoms arranged in the same direction and has both basal and the
edge planes. Pyrolytic graphite is anisotropic, very reproducible and
slightly denser than natural graphite. It is more porous than glassy
carbon, thus allows easy adsorption of electrode modifiers. Hence
it was employed in this work.
Depositions of NiPc derivatives to form ads-␤-NiPc(OH)4, ads-
␣-NiPc(OH)4 and ads-␣-NiPc(OH)8 on ordinary poly graphite
electrode were performed by adsorption from 1 × 10−3 mol dm−3
solution of the complex in DMSO. The adsorption was performed
by the dip dry method where the electrode was dipped in a solution
of DMSO containing the NiPc derivative and then dried, resulting
in adsorbed species. This was followed by recording of voltammo-
grams in 0.1 mol dm−3 NaOH (Fig. 2). Fig. 2A shows only a weak
feature at ∼−0.2 V for ads-␣-NiPc(OH)4, this is due to the ring based
reduction: (NiPc2−/NiPc3−) since adsorbed NiPc complexes do not
show metal based (e.g. NiIIIPc/NiIIPc or NiIIPc/NiIPc) processes. On
cycling the adsorbed ␣-NiPc(OH)4 in NaOH solution, NiIII/NiII cou-
poly-␤-Ni(O)Pc(OH)4, (ii) poly-␣-Ni(O)Pc(OH)4 and (iii) poly-
␣-Ni(O)Pc(OH)8 in 0.1 mol dm−3 NaOH. The peaks due to NiIII/NiII
are observed at E = 0.39 V for poly-␣ Ni(O)Pc(OH)8, 0.40 V for
poly-␣-Ni(O)Pc(OH)4 and at E = 0.28 V for poly-␤-Ni(O)Pc(OH)4
½
½
in 0.1 mol dm−3 NaOH (Table 1). The NiIII/NiII peaks for poly-␣-
Ni(O)Pc(OH)4 and poly-␣-Ni(O)Pc(OH)8 occur at more positive
potentials compared to poly-␤-Ni(O)Pc(OH)4, suggesting difficulty
of oxidation for the former complexes which are non-peripherally
substituted.
The surface coverage (ꢄ , mol cm−2) for poly-␤-Ni(O)Pc(OH)4,
poly-␣-Ni(O)Pc(OH)4 and poly-␣-Ni(O)Pc(OH)8 was estimated
from the charge under their relative oxidation peak using Eq. (21)
[24]:
Q = nFAꢄ
(21)
where Q is the total charge (C), A is the electrode surface area
(0.07 cm2) and ꢄ is the surface coverage of the redox species.

710
S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
Table 1
Electrochemical parameters for NiPc derivatives and 4-chlorophenol (7 × 10−4 mol dm−3) electrocatalysis in 0.1 mol dm−3 NaOH.
a
NiPc derivative
E
vs (Ag/AgCl)/V (NiIII/NiII)a
1010 × ꢄ (mol cm−2
)
Number of NiPc derivative moleculesa,b
EP vs (Ag/AgCl)/V (4-chorophenol)c
0.26
–
0.41
0.39
0.42, 0.45
0.37
½
ads-␤-NiPc(OH)4,
ads-␣-NiPc(OH)4,
ads-␣-NiPc(OH)8
poly-␤-Ni(O)Pc(OH)4
poly-␣-Ni(O)Pc(OH)4
poly-␣-Ni(O)Pc(OH)8,
–
–
–
0.28
0.40
0.39
–
–
–
1.49
0.37
5.10
–
–
–
d
6.0 × 1012 (2.9 × 1013
)
)
1.5 × 1012 (2.9 × 1013
2.10 × 1013 (3.7 × 1010
)
a
Values for adsorbed species were not determined due to NiIII/NiII (or any clearly defined oxidation peak).
Numbers in brackets are theoretically calculated values.
The peak for oxidation on bare OPGE (ordinary poly graphite electrode) is 0.38 V.
No clearly defined peak.
b
c
d
Using Eq. (21), the total surface coverages of the electrodes
one on bare OPGE. It is possible that the 4-chlorophenol peak
occurs beyond the potential range of the OPGE. The peaks observed
near −0.2 for ads-␣-NiPc(OH)4-OPGE and near −0.3 for poly-␣-
Ni(O)Pc(OH)4-OPGE due to ring reduction (NiPc2−/NiPc3−) process,
of 5.10, 0.37, 1.49 × 10−10 mol cm−2 were obtained for poly-
␣-Ni(O)Pc(OH)8, poly-␣-Ni(O)Pc(OH)4 and poly-␤-Ni(O)Pc(OH)4
(Table 1).
The surface coverage values are within the order of magnitude
of surface coverage expected for phthalocyanine molecule lying flat
on the surface (10−10 mol cm−2 [25]). Poly-␣-Ni(O)Pc(OH)8 gave
a surface coverage larger than the monolayer value, while poly-
␣-Ni(O)Pc(OH)4 gave a lower value. Only poly-␤-Ni(O)Pc(OH)4
gave a value near the monolayer. The differences reflect the ease
of film formation for the NiPc derivatives. Number of molecules
covering the surface area of the electrode was estimated exper-
imentally using the surface coverage above and the geometric
area of the OPGE, whereas the theoretical number of molecules
(in brackets in Table 1) covering the surface area of the electrode
were determined from optimized molecular structures (hence
area of one molecule) using Gaussian at B3LYP/6-31G(d) level
(Table 1). The calculated (theoretical) number of molecules for
poly-␣-Ni(O)Pc(OH)4 and poly-␤-Ni(O)Pc(OH)4, are larger than the
experimentally determined values, which suggest that the surface
is less than a monolayer. For poly-␣-Ni(O)Pc(OH)8 the calculated
number of molecules is much lower than the experimental val-
ues, which suggests that the surface coverage exceeds that of a
monolayer.
5.2. Analysis of 4-chlorophenol
Please note that in this section the modified electrodes will
be shown with OPGE at the end (as NiPc-OPGE) since the
cyclic voltammetry on OPGE alone is also studied. Fig. 4A shows
cyclic voltammograms of ads-␤-NiPc(OH)4-OPGE (ii), and poly-␤-
Ni(O)Pc(OH)4-OPGE (iii) in the presence of p-chlorophenol. There is
an increase in current for poly-␤-Ni(O)Pc(OH)4-OPGE as compared
to bare OPGE [14], showing catalytic activity. Catalytic activity is
evidenced by enhancement of currents and lowering of poten-
tial compared to analysis on the unmodified electrode. The peak
potential did not change for poly-␤-Ni(O)Pc(OH)4 but shifted to
less positive values for ads-␤-NiPc(OH)4-OPGE (Table 1), showing
catalysis in terms of lowering of potential. The reduction of Ni(III)
back to Ni(II) in Ni(O)Pc complexes is generally barely affected by
the chlorophenols [11], but an enhancement in the NiIII peak is
observed when there is catalysis. In Fig. 4A, there is an increase
in the anodic currents relative to the cathodic currents for poly-␤-
Ni(O)Pc(OH)4 in the presence of 4-chlorophenol, hence confirming
electrocatalysis. Plots of current (for p-chlorophenol) versus square
root of scan rate (not shown) were linear for all the modified
electrodes, confirming diffusion controlled mode of transport for
p-chlorophenol.
Fig. 4. Cyclic voltammograms of 7 × 10−4 mol dm−3 4-chlorophenol in 0.1 M NaOH
on OPGE modified with (A) ␤-NiPc(OH)4, (B) ␣-NiPc(OH)4 and (C) ␣-NiPc(OH)8.
(i) Bare OPGE, (ii) adsorbed complexes and (iii) polymerized complexes contain-
ing O–Ni–O bridges. (iv) in (B) is for poly-␣-Ni(O)Pc(OH)4 on OPG electrode in the
Fig. 4B shows the cyclic voltammograms of 4-chlorophenol
on bare (i), ads-␣-NiPc(OH)4-OPGE (ii) and poly-␣-Ni(O)Pc(OH)4-
OPGE (iii). There is no peak observed for the oxidation
4-chlorophenol on ads-␣-NiPc(OH)4-OPGE even though there is
absence of p-chlorophenol (in 0.1 mol dm−3 NaOH). Scan rate = 100 mV s−1
.

S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
711
as discussed above (Fig. 2A). This process is not involved in the
catalysis of 4-chlorophenol since it is at a much more negative
potential than expected for the catalytic oxidation of this species. It
has been reported that for the oxidation of 2-chlorophenol on poly
NiTSPc (tetrasulfo phthalocyanine) two anodic peaks are observed
[26]. The first appears at the same potential as for the bare elec-
trode, followed by a second peak located at about 30 mV more
positive than that of the potential of NiIII/NiII couple in the absence
of 2-chlorophenol but with higher currents. The two peaks are
observed in Fig. 4B for the oxidation of 4-chlorophenol on (iii) poly-
␣-Ni(O)Pc(OH)4-OPGE (Table 1). As stated above, the first peak for
oxidation of 4-chlorophenol [26] appears at the same potential as
for the bare electrode, followed by a second peak located at about
30 mV more positive than that of the potential of NiIII/NiII couple in
the absence of 2-chlorophenol but with higher currents. We sug-
gest that first peak observed in Fig. 4B(iii) is due to oxidation of
4-chlorophenol on the part of the bare electrode not covered with
and the second peak is due its oxidation on poly-␣-Ni(O)Pc(OH)8-
OPGE.
Oxidation of 4-chlorophenol on poly-␣-Ni(O)Pc(OH)4-OPGE
occurs with currents which are lower than those on bare OPGE,
suggesting no catalytic activity compared to bare OPGE. However,
there is an enhancement in the anodic currents of the NiIII/NiII
couple in the presence of 4-chlorophenol (Fig. 4B(iii)), compared
with the cyclic voltammogram of poly-␣-Ni(O)Pc(OH)4-OPGE in the
absence of 4-chlorophenol, suggesting the involvement of the for-
mer species in the oxidation process of the latter. However, since
the currents for oxidation of 4-chlorophenol on bare OPGE are more
than on poly-␣-Ni(O)Pc(OH)4-OPGE, catalytic activity cannot be
proved convincingly.
Scheme 2. Proposed mechanism for the catalytic oxidation of 4-chlorophenols by
poly-␤-Ni(O)Pc(OH)4 derivatives.
The mechanism for the oxidation of 4-chlorophenol in basic
media may be proposed as shown by Scheme 2, Eqs. (22) and
(23) [27] on poly-␤-Ni(O)Pc(OH)4 where clear catalytic activity was
involved:
In highly basic media (such as 0.1 mol dm−3 NaOH), 4-
chlorophenol is deprotonated. The first step is the electrochemical
oxidation of NiII to NiIII for poly-␤-Ni(O)Pc (Eq. (22)), followed by
the reaction of NiIII with deprotonated 4-chlorophenol in the pres-
ence of water molecules from electrolyte to form benzoquinone
(Eq. (23)). Quinones are well known as products of electrocatalytic
oxidation of chlorophenols [10]. For the adsorbed species, it is pos-
sible that ring based processes of the NiPc derivatives are involved
since NiIII/NiII are not known when NiPc derivatives are not poly-
merized.
5.3. Theoretical results
Molecular orbital calculations together with Fukui func-
tions, hardness and softness indices, were used to analyse the
chemical interaction of ␤-NiPc(OH)4 and ␤-Ni(O)Pc(OH)4 with p-
chlorophenol.
In this paper the interaction between ␤-NiPc(OH)4 and p-
chlorophenol and the effect of the central metal is studied further
than reported in Ref. [14], by making use of the condensed Fukui
function (fA+k), in order to increase our understanding of the electron
transfer process. The theoretical calculations are compared with
experimental results.
Fig. 4C shows that the anodic current on bare OPGE
(i) is higher than for ads-␣-NiPc(OH)8-OPGE(ii) and poly-␣-
Ni(O)Pc(OH)8-OPGE (iii) modified electrodes for 7 × 10−4 mol dm−3
4-chlorophenol. For the ads-␣-NiPc(OH)8-OPGE (ii), the currents
for 4-chlorophenol are lower than for poly-␣-Ni(O)Pc(OH)8-OPGE
(iii). Figs. 4C(iii) shows that the anodic currents for poly-␣-
Ni(O)Pc(OH)8-OPGE in the presence of p-chlorophenol are higher
than the cathodic currents confirming the involvement of the
NiIII/NiII couple in the oxidation of 4-chlorophenol.
Fig. 5A shows optimized structure of ␤-NiPc(OH)4 and
p-chlorophenol adduct. The energy levels of ␤-H2Pc(OH)4, ␤-
NiPc(OH)4 and ␤-Ni(O)Pc(OH)4 and the interaction of the latter two
with 4-chlorophenol were calculated (Table 2). The HOMO–LUMO
energy gaps of ␤-H2Pc(OH)4, ␤-NiPc(OH)4 and ␤-Ni(O)Pc(OH)4
HOMO–LUMO gap are 2.11 eV, 2.19 eV and 2.06 eV, respectively
(Table 2).
We previously used the donor acceptor molecular hardness
(ꢁDA) factor to study the interaction of p-chlorophenol and the
NiPc complexes [14], the result obtained suggested that the ␤-
Ni(O)Pc(OH)4 (ꢁDA = 1.87 eV) interacted better with p-chlorophenol
compared to ␤-NiPc(OH)4 (ꢁDA = 1.98 eV). In this work, we used a
more explicit and quantitative description of the interaction energy
between the NiPc complexes and p-chlorophenol which agreed
much better with experimental data. The intermolecular interac-
tion energy calculations were obtained using the supramolecular
In summary, for the oxidation of 4-chlorophenol on poly-
␣-Ni(O)Pc(OH)4-OPGE and poly-␣-Ni(O)Pc(OH)8-OPGE, the
observed currents are less than those of the bare OPGE,
with no improvement in potential. However for poly-␤-
Ni(O)Pc(OH)4-OPGE higher currents for 4-chlorophenol oxidation
are observed than on bare OPGE showing good electro-
catalytic activity compared to poly-␣-Ni(O)Pc(OH)4-OPGE
and poly-␣-Ni(O)Pc(OH)8-OPGE. In terms of surface cover-
age, it would be expected that poly-␣-Ni(O)Pc(OH)4-OPGE
(ꢄ = 0.37 × 10−10 mol cm−2) would have lower catalytic currents
than poly-␤-Ni(O)Pc(OH)4-OPGE (ꢄ = 1.45 × 10−10 mol cm−2), as
observed. poly-␣-Ni(O)Pc(OH)8-OPGE with the largest surface
coverage (ꢄ = 5.10 × 10−10 mol cm−2) would be expected to show
better catalytic activity than the other two. This is not observed
as discussed above since less currents are observed on poly-␣-
Ni(O)Pc(OH)4-OPGE and poly-␣-Ni(O)Pc(OH)8-OPGE than on bare
OPGE.
It can thus be concluded that non-peripheral substitution (␣)
does not favour electrocatalytic oxidation of 4-chlorophenols.
The behaviour could be related to the ease of formation and
nature of the O–Ni–O bridges, which form more readily for
poly-␤-Ni(O)Pc(OH)4 compared to either poly-␣-Ni(O)Pc(OH)4 or
poly-␣-Ni(O)Pc(OH)8. The better catalysis observed for poly-␤-
Ni(O)Pc(OH)4 could also be due to the fact the NiIII/NiII peaks are
observed at a much lower potential (0.28 V) when compared to
poly-␣-Ni(O)Pc(OH)4-OPGE and poly-␣-Ni(O)Pc(OH)8-OPGE (0.40
and 0.39 V, respectively) (Table 1).
Table 2
Calculated HOMO–LUMO energy gaps (eV) in gas phase. References in parentheses.
Molecules
|HOMO–LUMO| (eV)
␤-H2Pc(OH)4
␤-NiPc(OH)4
␤-Ni(O)Pc(OH)4
␤-NiPc(OH)4/p-chlorophenol
Ni(O)Pc(OH)4/p-chlorophenol
p-Chlorophenol
2.11
2.19 [14]
2.06 [14]
2.17
1.64
6.01

712
S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
Fig. 5. (A) Optimized structure of ␤-NiPc(OH)4 and p-chlorophenol using B3LYP/6-3g(d). (B) Electron density distribution of ␤-NiPc(OH)4 in the absence and presence of and
p-chlorophenol ␤-Ni(O)Pc(OH)4 in the absence and presence of p-chlorophenol.
Fig. 6. ␤-NiPc(OH)4 (A), ␤-NiPc(OH)4/p-chlorophenol (B), ␤-Ni(O)Pc(OH)4/p-chlorophenol (C) and ␤-Ni(O)Pc(OH)4 (D) frontier molecular orbitals (LUMO).

S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
713
approach, using Eq. (24):
is the sum total energy of isolated NiPc and p-chlorophenol.
Interaction energy of p-chlorophenol and NiPc complexes was
calculated using DFT calculations to be −205.9 kcal/mol (␤-
NiPc(OH)4) and 19.1 kcal/mol (␤-Ni(O)Pc(OH)4). Fig. 5B shows
the electron density distribution before and after the forma-
tion of the adduct. The interaction energy suggests that there
ꢃ
Eint = Ecomplex
−
Emolecules
(24)
where Eint is the interaction energy of the molecules, Ecomplex is
ꢄ
the energy of the complex (NiPc/p-chlorophenol) and
Emolecules
(A)
(B)
(C)
Fig. 7. Fukui 2D plots of condensed softness for maximum softness, H2Pc(OH)4 (A), ␤-NiPc(OH)4 (B), ␣-NiPc(OH)8 (C), NiPc (D), and ␣-NiPc(OH)4 (E). (F) Atomic numbering
in the phthalocyanine molecule which is used for describing peripheral and non-peripheral positions.

714
S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
(D)
(E)
(F)
Fig. 7. (Continued ).
is greater interaction of p-chlorophenol with ␤-NiPc(OH)4 than
with ␤-Ni(O)Pc(OH)4. This is evidenced by shifting of the oxida-
tive peak towards less positive potentials for ␤-NiPc(OH)4 (Fig. 4A)
compared to ␤-Ni(O)Pc(OH)4 in the presence of chlorophenol.
The opposite was observed when we employed ꢁDA values [14].
The results suggest that one has to be very careful when using
molecular indices to study interaction between supramolecular
complexes.

S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
715
Fig. 6A shows that for ␤-NiPc(OH)4 the frontier MO (LUMO)
is localized on the ligand and partly on the Ni atom. Also Fig. 6
shows that the LUMO of ␤-NiPc(OH)4 appear as an antibonding
pi orbital (␲*) and the LUMO of ␤-Ni(O)Pc(OH)4 appear as an
antibonding sigma orbital (␴*). When ␤-NiPc(OH)4 interacts with
p-chlorophenol (Fig. 6B) the LUMO shows extensive localization
of MO mainly on the Pc ligand with some on the Ni atom. For
␤-Ni(O)Pc(OH)4 in the presence p-chlorophenol (Fig. 6D), there
is less extensive delocalization on the Pc ring compared to when
␤-NiPc(OH)4 interacts with p-chlorophenol. This suggests that for
␤-NiPc(OH)4 the ring is more extensively involved in the oxida-
tion of 4-chlorophenol when compared to ␤-Ni(O)Pc(OH)4. When
adsorbed specie (without cycling or conditioning in 0.1 mol dm−3
NaOH) are employed for the catalytic oxidation of 4-chlorophenol,
the oxidation potentials are observed at different values com-
4-chlorophenol compared to ␤-NiPc(OH)4 in terms of catalytic cur-
rents (Fig. 4).
6. Conclusions
NiPc
derivatives:
poly-␣-Ni(O)Pc(OH)4-OPGE,
poly-␣-
Ni(O)Pc(OH)8-OPGE and poly-␤-Ni(O)Pc(OH)4-OPGE and their
adsorbed counterparts (ads-␣-NiPc(OH)4-OPGE, ads-␣-NiPc(OH)8-
OPGE and ads-␤-NiPc(OH)4-OPGE) were studied for their catalytic
activity towards the oxidation of p-chlorophenol. In terms of
electrocatalytic activity, poly-␣-Ni(O)Pc(OH)4-OPGE and poly-
␣-Ni(O)Pc(OH)8-OPGE, showed evidence for the involvement
of the NiIII/NiII couple in the oxidation of 4-chlorophenol, but
the catalytic currents were less than those of the bare OPGE.
Poly-␤-Ni(O)Pc(OH)4-OPGE showed higher currents than the bare
OPGE showing good electrocatalytic activity compared to for poly-
␣-Ni(O)Pc(OH)4-OPGE and poly-␣-Ni(O)Pc(OH)8-OPGE. We have
shown, using interaction energy values that when p-chlorophenol
interacts better with ␤-NiPc(OH)4 than ␤-Ni(O)Pc(OH)4, hence the
former is a better catalyst in terms of lowering of overpotential.
We have used the condensed Fukui function to determine the
reactive sites where electron transfer would take place between
p-chlorophenol and the nickel phthalocyanine complexes. We
have theoretically rationalised the interaction of p-chlorophenol
with nickel phthalocyanine complexes.
pared to when the electrode has been cycled in 0.1 mol dm−3
,
suggesting that there are different catalytic sites involved. Fol-
lowing cycling in 0.1 mol dm−3, there will be sites containing
NiIII/NiII, hence the central metal will be involved in the catalytic
reaction.
The oxidation process of p-chlorophenol is viewed as a nucle-
ophilic attack of ␤-NiPc(OH)4 by p-chlorophenol. According to Eqs.
(7)–(9) and (15)–(17), in order to calculate the Fukui function and
condensed Fukui function (f(r) and fA+k), one needs to determine the
electron density of the molecule and the charges of the atoms in the
molecule. In this work we have used the Mulliken gross population
analysisfor qAk, Eqs. (15)and (18) to determine the condensedFukui
function and the condensed local softness.
Acknowledgements
Tetra substituted phthalocyanine derivatives contain a mix-
ture of four possible structural isomers, with the position of the
substituents being at 2(3), 9(10), 16(17) and 23(24) for the periph-
erally substituted derivatives and at 1(4), 6(11), 15(18) and 22(25)
for non-peripherally substituted derivatives (see numbering in
Fig. 7). However for the purposes of the present calculations, the
substituents have been fixed to one position on each benzene
group. Fig. 7A–E gives us a clearer picture of what is actually hap-
pening to the carbons attached to nitrogens when H2Pc(OH)4 is
metallated. Comparing Fig. 7A and B, we see that the condensed
local softness (less reactive) values for carbons at 5, 12, 19, 26
(red region) lie below the ones at 7, 14, 21, 28 (blue regions),
for unmetallated H2Pc(OH)4 (Fig. 7A). Upon metallation (Fig. 7B)
the condensed local softness values for carbons at 5, 12, 19, 26
(red region) have increased (more reactive) to a point where
they have surpassed the values for carbons at 7, 14, 21, 28 (blue
regions).
This work was supported by the Department of Science and
Technology (DST) and National Research Foundation (NRF) of South
Africa through DST/NRF South African Research Chairs initiative for
Professor of Medicinal Chemistry and Nanotechnology and Rhodes
University.
References
[1] T. Nyokong, in: J.H. Zagal, F. Bedioui, J.-P. Dodelet (Eds.), N4-Macrocyclic Metal
Complexes, Springer, United States of America, 2006.
[2] D.R. Trackley, G. Dent, W.E. Smith, Phys. Chem. Chem. Phys. 2 (2001) 1419.
[3] M.S. Ureta-Zanartu, P. Bustos, C. Berrios, M.C. Diez, M.L. Mora, C. Guitierrez,
Electrochim. Acta 47 (2002) 2399.
[4] S. Griveau, F. Bedioui, Electroanalysis 13 (2001) 253.
[5] G. Heimel, L. Romaner, J.-L. Bredas, E. Zojer, Phys. Rev. Lett. 96 (2006) 196806.
[6] R.G. Parr, W. Yang, J. Am. Chem. Soc. 106 (1984) 4049.
[7] R.G. Parr, R.A. Donnelly, M. Levy, W.E. Palke, J. Chem. Phys. 68 (1978) 3801.
[8] R.G. Parr, R.G. Pearson, J. Am. Chem. Soc. 105 (1983) 7512.
[9] M.S. Ureta-Zanartu, A. Alarcon, C. Berrios, G.I. Cardenas-Jiron, J. Zagal, C. Gutier-
rez, J. Electroanal. Chem. 580 (2005) 94.
Fig. 7A and B suggests that when Ni is inserted into the
ring, the carbon atoms at 5, 12, 19, 26 (red region) are acti-
vated differently compared to unmetallated. The 5, 12, 19, 26
(red region) carbons are situated on the side where the OH is
substituted and show greater local softness (hence greater reac-
tivity). The carbons at 7, 14, 21, 28 (blue regions) are situated
on the side where there are no OH substituents and have lower
local softness values (less reactive compared to 5, 12, 19, 26 car-
bons). This suggests that even though the OH substituents have
no effect on the bond length between the Ni atom and the N
[14], it has some distabilization effect on the 5, 12, 19, 26 car-
bons for ␤-NiPc(OH)4. Fig. 7D shows that for unsubstitured NiPc
all the carbons are activated equally compared to ␤-NiPc(OH)4.
Fig. 7C suggests that, for ␣-NiPc(OH)8, the Ni atom is made softer
or more reactive relative to all other atoms. The Fukui plots of
␣-NiPc(OH)8 suggests that if this complex was to be used for elec-
tro catalytic oxidation of p-chlorophenol, catalysis through the
nickel atom would be favoured over the ring. Fig. 7E shows the
Fukui plots of ␣-NiPc(OH)4, showing that only the carbons are
activated. It is observed experimentally that ␣-NiPc(OH)4 and ␣-
NiPc(OH)8 shows bad catalytic activity towards the oxidation of
[10] M.S. Ureta-Zan˜artu, C. Berríos, J. Pavez, J. Zagal, C. Gutíerrez, J.F. Marco, J. Elec-
troanal. Chem. 553 (2003) 147.
[11] C. Berríos, M.S. Ureta-Zan˜artu, C. Gutíerrez, Electrochim. Acta 53 (2007) 792.
[12] P.W. Ayres, R.G. Parr, J. Am. Chem. Soc. 122 (2000) 2010.
[13] M. Berkowitz, J. Am. Chem. Soc. 109 (1987) 4823.
[14] S. Khene, K. Lobb, T. Nyokong, Inorg. Chim. Acta 362 (2009) 5055.
[15] R.G. Pearson, Acc. Chem. Res. 26 (1993) 250.
[16] B.E. Douglas, D.H. McDaniels, J.J. Alexander, Concept and Models of Inorganic
Chemistry, John Wiley and Sons, Inc., New York, 1994, pp. 193–195.
[17] Y. Li, J.N.S. Evans, J. Am. Chem. Soc. 117 (1995) 7756.
[18] F. Mendez, J.L. Gazquez, J. Am. Chem. Soc. 116 (1994) 9298.
[19] K. Flurchick, L. Bartolotti, J. Mol. Graph. 13 (1995) 10.
[20] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheese-
man, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P.
Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Strat-
mann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari,
J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Ste-
fanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. John-
son, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision E.01,
Gaussian, Inc., Wallingford, CT, 2004.
[21] B.N. Achar, P.K. Jayasree, Synth. Met. 104 (1999) 101.

716
S. Khene et al. / Electrochimica Acta 56 (2010) 706–716
[22] C.C. Leznoff, M. Hu, C.R. McArthur, Y. Qin, J.E. van Lier, Can. J. Chem. 72 (1994)
1990.
[23] S. Khene, D.A. Geralo, C.A. Togo, J. Limson, T. Nyokong, Electrochim. Acta 54
(2008) 183.
[25] Z. Li, M. Lieberman, W. Hill, Langmuir 17 (2001) 4887.
[26] C. Berrius, J.F. Marco, C. Gutierrez, M.S. Ureta-Zan˜artu, Electrochim. Acta 54
(2009) 6425.
[27] B. Agboola, T. Nyokong, Electrochim. Acta 52 (2007) 5039.
[24] H.O. Finklea, in: A.J. Bard, I. Rubinstein (Eds.), Electroanalytical Chemistry, vol.
19, Marcel Dekker, New York, 1996, pp. 109–335.