Electropolymerizable iron (III) and cobalt (II) dicyanophenoxy tetraphenylporphyrin complexes: Potential electrocatalysts

Article abstract of DOI:10.1016/j.inoche.2005.11.024

Solution and solid phase electrochemical features of 5-[4-(3,4- dicyanophenoxy)phenyl],10,15,20-triphenylporphyrin complexes of iron(III) (FeCNOTPP(Cl)) and cobalt(II) (CoCNOTPP) have been described. These novel asymmetric dicyanophenoxy-derivatised cobal

Full text of DOI:10.1016/j.inoche.2005.11.024

Inorganic Chemistry Communications 9 (2006) 223–227
www.elsevier.com/locate/inoche
Electropolymerizable iron (III) and cobalt (II) dicyanophenoxy
tetraphenylporphyrin complexes: Potential electrocatalysts
*
Kenneth I. Ozoemena , Zhixin Zhao, Tebello Nyokong
Chemistry Department, Rhodes University, Grahamstown 6140, South Africa
Received 30 September 2005; accepted 21 November 2005
Available online 6 January 2006
Abstract
Solution and solid phase electrochemical features of 5-[4-(3,4-dicyanophenoxy)phenyl],10,15,20-triphenylporphyrin complexes of iro-
n(III) (FeCNOTPP(Cl)) and cobalt(II) (CoCNOTPP) have been described. These novel asymmetric dicyanophenoxy-derivatised cobalt
and iron porphyrin complexes were electropolymerised onto glassy carbon electrodes, which in aqueous solutions, gave surface concen-
trations (ca. 10ꢀ10 mol cmꢀ2) typical of monolayer coverages. The films also exhibited excellent stability and electrocatalysis towards the
direct detection of important analytes as nitrite, nitric oxide, and hydrogen peroxide in aqueous solutions.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Iron and cobalt dicyanophenoxy-derivatised porphyrins; Electropolymerisation; Hydrogen peroxide; Nitrite; Nitric oxide; Chronoamperometry
Because of their utility as excellent biomimetic models
for naturally occurring hemes, iron and cobalt porphyrin
complexes have been widely used in electrochemical and
electron transfer studies [1–11]. Several authors [5–7] have
reported the potential applications of polymer films of
metalloporphyrins (MPs) and their derivatives, especially
on glassy carbon electrode (GCE), as new electrode materi-
als for chemical and biological sensors. Iron and cobalt por-
phyrin based polymer films have intensely been reported
because of their excellent electrocatalytic activities [5–11].
Electrochemical approach to film formation on elec-
trodes is more preferable to such techniques as ‘‘dip-dry/
drop-dry’’ methods, which are largely not reproducible
and the thickness of the modifier on the electrode cannot
be accurately controlled. The use of MP complexes bearing
the cyano group (for example, the dicyanophenoxy or oxo-
phthalonitrile, as reported in this paper) to form polymer
films onto electrode surfaces is not known. In this work,
we describe the first example of asymmetric dicyanophen-
oxy (or oxo-phthalonitrile) porphyrin complexes of iron
(III) and cobalt (II) (Fig. 1) electropolymerised onto
GCE. Furthermore, we report a preliminary investigation
on the feasibility of these novel film electrodes in interact-
ing with important biomolecules, nitrite, nitric oxide, and
hydrogen peroxide in aqueous solution.
The cobalt (II) 5-[4-(3,4-dicyanophenoxy)phenyl],10,15,
20-triphenylporphyrin (herein abbreviated as CoCNOTPP)
was synthesised as recently reported by our group [12] from
5-[4-(3,4-dicyanophenoxy)phenyl],10,15,20-triphenylporph-
yrin (H2CNOTPP) ligand and cobalt acetate salt. Here, the
iron (III) chloro-derivative (FeCNOTPP) was synthesised
using a procedure similar to that for CoCNOTPP complex
[12]. Briefly, a 100 mg (0.312 mmol) of H2CNOTPP ligand,
synthesised as before [12], was mixed with ferrous chloride
tetrahydrate (FeCl2 Æ 4 H2O) (53 mg, 0.264 mmol) and dis-
solved in a mixture of dimethylformamide (DMF,10 ml)
and CH2Cl2 (10 ml) and refluxed for 6 h under a N2 atmo-
sphere with stirring in a 100-ml flask equipped with a con-
denser. After cooling to room temperature, the mixture
was washed with 100 ml water three times to remove
DMF. Column chromatography on silica gel with CH2Cl2
as eluent gave three bands. The first red band was the unre-
acted ligand (H2CNOTPP). The second red band was the
*
Corresponding author. Tel.: +27 46 603 8801; fax: +27 46 622 5109.
E-mail address: k.ozoemena@ru.ac.za (K.I. Ozoemena).
1387-7003/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2005.11.024

224
K.I. Ozoemena et al. / Inorganic Chemistry Communications 9 (2006) 223–227
GCE was first cleaned by polishing with an aqueous
slurry of alumina, followed by sonication, rinsing with dis-
tilled water and then wiping dry with a clean tissue paper.
For solution voltammetric studies, concentrations of the
porphyrin complexes were maintained at ca. 1 mM in dry
dimethylsulfoxide (DMSO) containing TBAP (tetrabuty-
lammonium perchlorate) as supporting electrolyte. Electro-
polymerisation of FeCNOTPP and CoCNOTPP films on
GCE (designated as GCE-poly-FeCNOTPP and GCE-
poly-CoCNOTPP, respectively) carried out by repetitive
scanning of a solution containing the metalloporphyrin in
DMSO and 0.1 M TBAP (as a supporting electrolyte).
Before electrochemical investigations in aqueous solutions,
the modified electrodes were thoroughly rinsed, first in
fresh DMSO and finally in pH 4.0 buffer solutions.
CN
CN
N
N
N
N
M
O
where M = Fe(Cl) or Co
Fig. 1. Molecular structure of metallo-5-[4-(3,4-dicyanophenoxy) phe-
nyl],10,15,20-triphenyl porphyrin complex (MCNOTPP), where
M = Fe(III)Cl or Co(II).
We observed that the synthetic strategy employed for
the FeCNOTPP complex (use of ferrous chloride salt)
resulted in a facile axial ligation of the iron center with
ꢀOH and Clꢀ, and axial ligation is most possible with
Fe(III) rather than the Fe(II) oxidation state. Also, elec-
tronic spectra of the axially ligated FeCNOTPP complexes
were found to resemble those of other reported high-spin
Fe(III) porphyrin complexes [13,14]. Unlike, the CoC-
NOPP that gave four couples [12], solution electrochemis-
try of FeCNOTPPCl (Fig. 2) only showed three well-
defined redox couples at 0.20 V (I), ꢀ0.85 and ꢀ1.35 V
(II) (vs. AgjAgCl). These couples did not change with
either the scan direction or starting from the rest zero
potential. It is well established in the literature [1,5,15] that
the first oxidation couple of Fe(III) porphyrin complex is
due to the ring process, thus couple (I) is due to the
[Fe(III)TPP]/[Fe(III)TPP]Å+ process while couples (II) and
(III) are attributable to the metal-centered processes
([Fe(III)TPP]/[Fe(II)TPP]) and ([Fe(II)TPP]/[Fe(I)TPP],
respectively). Electrochemistry of phthalonitrile in similar
conditions resulted in a irreversible peak at ꢁꢀ0.8 V vs.
AgjAgCl, suggesting that the phthalonitrile peak is not
observed when the molecule is bound to porphyrin. The
plot of current versus square root of the scan rate ranging
chloro derivative [(Fe(III)CNOTPP)Cl]. The final red band
was 5-[4-(3,4-dicyanophenoxy)phenyl],10,15,20-triphenyl-
porphyrin iron (III) hydroxide [(Fe(III)CNOTPP)OH].
The third band was washed with 20 ml 5% HCl solution
three times to change hydroxyl to chloride. Combination
of second and third bands, and removal of CH2Cl2 by
evaporation afforded 85 mg (76.0%) of a purple solid.
The purple product was recrystallised from CH2Cl2 with
hexane. IR [(KBr) Vmax/cmꢀ1]: 2223 (CN), 1248 (C–O–
C); UV–Vis [CHCl3, kmax/nm (loge)]: 375 (4.23) , 413
(5.32), 508 (4.07), 572 (3.57), 660 (3.28), 690 (3.35);
MALDI-TOF-MS m/z: (FeCNOTPP)+, Calcd. = 810.698;
Found = 810.886. CoCNOTPP [12] and FeCNOTPPCl
were characterised spectroscopically using a Varian 500
UV–Vis/NIR spectrophotometer, a Perkin–Elmer 200
FTIR spectrometer, and MALDI-TOF (Perseptive Biosys-
tems Voyager DE-PRO Biospectrometry Workstation and
possessing Delayed Extraction Technology at the Univer-
sity of Cape Town).
Ultrapure water of resistivity 18.2 MX from a Milli-Q
Water System (Millipore Corp., Bedford, MA, USA) was
used for all aqueous studies. Phosphate buffer solutions
(PBS) at various pHs were prepared with appropriate
amounts of Na2HPO4 and NaH2PO4, and the pH adjusted
with 0.1 M H3PO4 or NaOH. All solutions were de-aerated
by bubbling nitrogen prior to each electrochemical studies.
H2O2 solution (30%) was obtained from SAARCHEM. All
other reagents were of analytical grade and were used as
received from the suppliers without further purification.
All electrochemical experiments were carried out with
BioAnalytical System (BAS) 100 B/W Electrochemical
Workstation using a conventional three-electrode set-up
with a glassy carbon electrode (GCE, 3.00 mm diameter,
modified with FeCNOTPP or CoCNOTPP polymer films)
as a working electrode, platinum wire as a counter elec-
trode and AgjAgCl wire as a pseudo-reference electrode.
The modified GCE is herein referred to as GCE-poly-
MCNOTPP (where M = Co(II), Fe(III)(Cl)). All experi-
ments were performed at 25 1 °C and under purified
nitrogen gas.
20 A
µ
I
II
III
-1.8
-1.2
-0.6
0
0.6
1.2
E / V (vs Ag AgCl)
|
Fig. 2. CV responses of a 1 mM solution of FeCNOTPP on a GCE in dry
DMSO solution containing 0.1 M TBAP at 50 mV sꢀ1
.

K.I. Ozoemena et al. / Inorganic Chemistry Communications 9 (2006) 223–227
225
from 5 to 800 mV sꢀ1 (not shown) was linear as expected
for diffusion-controlled processes.
to mass transport through the polymer film as the film
thickness increases [16]. After about 60 repetitive scans,
there was no detectable increase in current, indicating the
end of the electropolymerisation process. Attempts to
obtain similar electropolymers in narrower windows or
anodic windows alone were not successful, which may sug-
gest a possible involvement of some reduction products in
the electropolymerisation process. Several steps have been
recognised to be involved in the electrochemical-induced
formation of polymeric films on electrode surfaces and
these include the transport of monomeric species to the
electrode by diffusion, monomer oxidation at the appropri-
ate potential to produce radical cations, radical–radical
coupling, electrochemical oxidation of the formed oligo-
mers and the precipitation of the polymer on the electrode
surface [17]. Although there has been no report yet on the
cyano-terminated porphyrin and related complexes, the
same mechanisms may apply.
Interestingly, repetitive scanning of the two asymmetric
metalloporphyrin complexes in DMSO resulted in a grad-
ual increase in the amplitude of the cyclic voltammetric
peaks (exemplified for CoCNOTPP in Fig. 3a). The opti-
mised potential window for reproducible electrosynthesis
was between ꢀ1.10 and + 1.10 V vs. AgjAgCl. As evident
in Fig. 3, the first scan (i) is different from the subsequent
scans that increase steadily in the anodic directions, indi-
cating electrosynthesis of the dicyanophenoxy-derivatised
metalloporphyrin polymer on the GCE surface. The
increase in peak separations with scan number is an indica-
tion that the polymer film became increasingly less conduc-
tive as the film thickness increased. This behaviour is the
result of increased electrical resistance and the resistance
The half-wave potentials (E1/2) of the GCE-poly-FeC-
NOTPP and GCE-poly-CoCNOTPP in acidic pH were
observed at ꢀ0.34 and ꢀ0.07 V vs. AgjAgCl, respectively,
attributable to their metal-centered, one-electron process
[M(III)TPP(-2)/M(II)TPP(-2)] couples [18,19]. Fig. 3b pre-
sents typical cyclic voltammograms obtained for the GCE-
poly-CoCNOTPP, after 100 continuous scans (at
50 mV sꢀ1) in pH 4.0 solution, indicating excellent electro-
chemical stability. The plots of anodic and cathodic peak
(i)
75 nA
-1.2
-0.8
-0.4
0
0.4
0.8
1.2
current (not shown) vs. scan rate (ranging 5–500 mV sꢀ1
)
a
E / V (vs Ag AgCl)
|
were linear as is expected for surface adsorbed species.
The effect of solution pH (1.0–11.0) on the observed poten-
tial (E1/2) of M(III)TPP(-2)/M(II)TPP(-2) redox couples
for the immobilised complexes was investigated with
SWV (Fig. 3c). The redox couples showed shifts to negative
potential with increasing pH as a result of proton exchange
occurring during the electron transfer, typical of
M(III)TPP(-2)/M(II)TPP(-2) couples [5]. For a one-proton
one-electron reaction, one would expect a ꢀ60 mV/pH unit
dependence. The slopes of the plots (ca. ꢀ45 mV/pH) may
be due to the protonation of the tethered porphyrin ligand.
It may thus be suspected that there is a microenvironment
effect around the porphyrin molecules in the surface. The
potential of the redox couple M(II)/M(I) observed
ꢁ0.80 V (vs. AgjAgCl) remained essentially the same
throughout the pH ranges studied in this work. The immo-
bilized CoCNOTPP and FeCNOTPP exhibited quasi-
reversible behaviour by showing anodic-to-cathodic peak
currents ratio (Ipa/Ipc) of approximately unity expected
for an ideal reversible process but with peak-peak separa-
tion (DEp) that is much higher than the expected 60 mV
for an ideal reversible process or surface-adsorbed species
of zero volt. Deviation from the ideal zero DEp values [20]
for surface-confined species, suggests possible kinetic limi-
tations or electrostatic interactions of the porphyrin mole-
75 nA
-1.2 -0.8 -0.4
0
0.4
0.8
1.2
b E / V (vs Ag AgCl)
|
200
100
0
y = -44.603x + 198.91
R2 = 0.9901
CoCNOTPP
-100
-200
-300
-400
FeCNOTPP
y = -45.419x + 164.95
R
2 = 0.9823
0
2
4
6
pH
8
10
12
c
Fig. 3. CV responses of a 1 mM solution of: (a) CoCNOTPP on a GCE in
DMSO solution containing 0.1 M TBAP during 60 continuous scans at
50 mV sꢀ1. Curve (i) is the 1st scan; 2nd–4th scans were removed for
clarity, the broken arrow shows the direction of the growth of the
polymer; (b) 60 repetitive scans of GCE-poly-CoCNOTPP in pH 4.0 PBS
at 50 mV sꢀ1 and (c) plot of E1/2 vs. pH.
cules in the polymer films. The surface concentration (CMP
/
)
mol cmꢀ2) of the immobilised films (ca. 10ꢀ10 mol cmꢀ2
was in the range reported for monolayer [20–22] rather
than the thick porphyrin polymer films (10ꢀ9 mol cmꢀ2

226
K.I. Ozoemena et al. / Inorganic Chemistry Communications 9 (2006) 223–227
range) [5]. This observation indicates that upon rinsing in
DMSO and/or in aqueous conditions, nearly all the immo-
bilised electropolymer is lost thus retaining an electrochem-
ically stable monolayer. The formation of a stable
monolayer is an advantage since thin solid films or mono-
layer can exhibit some important properties (e.g., fast elec-
trode kinetics due to short electron tunnelling distances)
over the thick polymer films.
of NOꢀ2 and NO was studied at fixed potential of
+750 mV at pH 7.4 (Fig. 4b(i)) and pH 2.0 (Fig. 4b(ii)),
respectively. Following the literature precedents [26], we
used sodium nitrite (NaNO2) as a precursor for NO. In
acid solution (pH < 4), NaNO2 can generate free NO by
disproportionation. In pH 7.4, however, the disproportion-
ation is minimal, thus NOꢀ2 is mainly present. All analytical
parameters, linear concentration range (LCR), limit of
detection (LoD) and sensitivity, were obtained after six
repeated measurements performed during a two-week per-
iod (RSD < 2%). Hydrogen peroxide was detected up to
10 mM with a LoD of ꢁ0.5 mM (S/N = 3); its reductive
detection being approximately 28 times more sensitive
(0.890 lA/mM) than for the oxidative one (0.032 lA/
mM). Both NOꢀ2 and NO were detected in the lM range
and with LCR up to 173 lM (r2 = 0.9963 and 0.9955 for
NOꢀ2 and NO, respectively). Larger sensitivity was
observed for NOꢀ2 (1.694 nA/lM) than for the NO
(0.643 nA/lM). The reductive detection of H2O2 occurred
close to the formal potential of the metal-centered
(Fe(III)TPP/Fe(II)TPP) process, suggesting the involve-
ment of this redox couple in the catalysis. The oxidative
detections, on the other hand, suggest the involvement of
the ligand ([Fe(III)TPP]/[Fe(III)TPP]Å+) process or even
the metal-centered Fe(IV)/Fe(III) process as recently pro-
posed by Aguirre and co-workers [7].
To probe the potential analytical applications of these
GCE modified porphyrin films, preliminary investigations
with some biologically significant analytes: hydrogen per-
oxide (H2O2), nitrite ðNOꢀÞ and nitric oxide (NO) were
2
performed. The need for a reliable, rapid and simple
method for the detection and monitoring of H2O2, NOꢀ,
2
and NO in many biomedical and environmental samples
has continued to be the focus of electroanalytical tech-
niques [5,23–25]. The two porphyrin based electrodes stud-
ied in this work showed comparable electrocatalytic
activities in aqueous solution spiked with hydrogen perox-
ide, nitrite and nitric oxide. However, since electron trans-
fer rates are faster with the Fe(III)/Fe(II) in porphyrins
than the Co(III)/Co(II) in porphyrins analogous [3], the
GCE-poly-FeCNOTPP was preferred in this work for
chronoamperometric analyses. From CV and SWV (not
shown) obtained for the electrocatalysis of H2O2, NO
and NOꢀ, chronoamperometric detection of H2O2 was car-
2
ried out at pH 7.4 (at fixed potential of ꢀ650 mV (reduc-
In conclusion, this work has presented the first example
of asymmetric dicyanophenoxy-derivatised iron and cobalt
porphyrin polymer films on GCEs as novel electrode mate-
rials. This type of electrode promises to be potentially use-
ful for numerous sensing applications, such as in
amperometric sensors for molecules of environmental and
biomedical importance. Some of these potential applica-
tions are being investigated in our laboratories.
tion) and +650 mV (oxidation)), (Fig. 4a). The detection
200nA
0
100
200
300
400
time / s
300 nA
Acknowledgements
We thank Rhodes University, the National Research
Foundation (NRF, GUN 2069275 and GUN 2053657)
and the Andrew Mellon Foundation for Accelerated
Development Programme for their support.
0
100
200
time / s
300
400
a
(i)
References
150 nA
[1] K.M. Kadish, E. Van Caemelbecke, G. Royal, in: K.M. Kadish,
K.M. Smith, R. Guilard (Eds.), Electrochemistry of Metalloporphy-
rins in Nonaqueous Media ‘‘The Porphyrin Handbook’’, Electron
Transfer, vol. 8, Academic Press, New York, 2000, pp. 38–55.
[2] F.C. Anson, C. Shi, B. Steiger, Accounts Chem. Res. 30 (1997)
437.
(ii)
[3] H. Sun, V.V. Smimov, S.G. DiMagno, Inorg. Chem. 42 (2003) 6032.
[4] J.P. Collman, L. Fu, P.C. Hermann, X. Zhang, Science 275 (1997)
949.
[5] F. Bedioui, S. Trevin, V. Albin, M. Guadalupe, G. Villegas, J.
Devynck, Anal. Chim. Acta 341 (1997) 177.
[6] M. Biesaga, K. Pyrzynska, M. Trojanowicz, Talanta 51 (2000) 209.
[7] M. Lucero, G. Ramirex, A. Riquelme, I. Azocar, M. Isaacs, F.
Armijo, J.E. Forster, E. Trollund, M.J. Aguirre, D. Lexa, J. Mol.
Catal. C 221 (2004) 71.
0
100
200
time / s
300
400
b
Fig. 4. Chronoamperograms recorded at GCE-poly-FeCNOTPP on
successive additions of: (a) 1 mM H2O2 to a phosphate buffer pH 7.0
solution poised at +650 and at ꢀ650 mV (inset) (vs. AgjAgCl) and (b)
13 lM NaNO2 aliquots to a pH 7.4 (i) and pH 2.0 (ii) solutions poised at
+750 mV (vs. AgjAgCl). All solutions stirred at 250 rpm.

K.I. Ozoemena et al. / Inorganic Chemistry Communications 9 (2006) 223–227
227
[8] S.-K. Yoo, J.H. Han, S.J. Lee, J.Y. Ryu, C. Kim, S.W. Jin, Y. Kim,
W. Nam, Inorg. Chem. Commun. 6 (2003) 1148.
[16] H.S. Sharma, S.-M. Park, J. Electrochem. Soc. 151 (2004) 61.
[17] W. Schuhmann, C. Kranz, H. Wohlschla¨ger, J. Strohmeier, Biosens.
Bioelectron. 12 (1997) 1157.
[9] I. Zilbermann, E. Maimon, R. Ydgar, A.I. Shames, E. Korin, L.
Soifer, A. Bettelheim, Inorg. Chem. Commun. 7 (2004) 1238.
[10] S. Yamazaki, Y. Yamada, T. Ioroi, N. Fujiwara, Z. Siroma, K.
Yasuda, Y. Miyazaki, J. Electroanal. Chem. 576 (2005) 253.
[11] S.Lj. Gojkovic, S. Gupta, R.F. Savinell, Electrochim. Acta 45 (1999)
889.
[12] Z.X. Zhao, K.I. Ozoemena, D.M. Maree, T. Nyokong, Dalton Trans.
(2005) 1241.
[13] J. Wojaczynski, M. Stephen, L. Latos-Grazynski, Eur. J. Inorg.
Chem. (2002) 1806.
[18] W.S. Cardoso, Y. Gushikem, J. Electroanal. Chem. 583 (2005) 300.
[19] E. Song, C. Shi, F.C. Anson, Langmuir 14 (1998) 4315.
[20] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals
and Applications, second ed., Wiley, New York, 2001.
[21] K. Ozoemena, T. Nyokong, Electrochim. Acta 47 (2002) 4035.
[22] K.I. Ozoemena, T. Nyokong, Inorg. Chem. Commun. 6 (2003) 1192.
[23] Y. Lin, X. Cui, L. Li, Electrochem. Commun. 7 (2005) 166.
[24] K.I. Ozoemena, Z. Zhao, T. Nyokong, Electrochem. Commun. 7
(2005) 679.
[14] M. Kasuno, A. Uehara, N. Ichieda, T. Kitano, K. Banu, S. Kihara, J.
Electroanal. Chem. 579 (2005) 223.
[25] J. Oni, N. Diab, I. Radtke, W. Schuhmann, Electrochim. Acta 48
(2003) 3349.
[15] D. Lanc¸on, K.M. Kadish, J. Am. Chem. Soc. 105 (1983) 5610.
[26] A. Yu, Z. Liang, J. Cho, F. Caruso, Nanoletters 3 (2003) 1203.