Syntheses and investigation of the effects of position and nature of substituent on the spectral, electrochemical and spectroelectrochemical properties of new cobalt phthalocyanine complexes

Article abstract of DOI:10.1016/j.poly.2010.01.004

The syntheses of new cobalt phthalocyanine (CoPc) complexes, tetra-substituted with diethylaminoethanethio at the peripheral (complex 3a) and non-peripheral (complex 3b) positions, and with benzylmercapto at the non-peripheral position (complex 5), are re

Full text of DOI:10.1016/j.poly.2010.01.004

Polyhedron 29 (2010) 1257–1270
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Syntheses and investigation of the effects of position and nature of substituent
on the spectral, electrochemical and spectroelectrochemical properties of new
cobalt phthalocyanine complexes
*
Isaac Adebayo Akinbulu, 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:
The syntheses of new cobalt phthalocyanine (CoPc) complexes, tetra-substituted with diethylaminoe-
thanethio at the peripheral (complex 3a) and non-peripheral (complex 3b) positions, and with benzyl-
mercapto at the non-peripheral position (complex 5), are reported. The effects of the nature and
position of substituent on the spectral, electrochemical and spectroelectrochemical properties of these
complexes are investigated. Solution electrochemistry of complex 3a showed three distinctly resolved
redox processes attributed to CoIIIPcꢀ2/CoIIPcꢀ2 (E = +0.64 V versus Ag|AgCl), CoIIPcꢀ2/CoIPcꢀ2 (E
Received 19 August 2009
Accepted 4 January 2010
Available online 11 January 2010
Keywords:
Cobalt
½
½
= ꢀ0.24 V versus Ag|AgCl) and CoIPcꢀ2/CoIPcꢀ3 (E = ꢀ1.26 V versus Ag|AgCl) species. No ring oxidation
½
Diethylaminoethanethiol phthalocyanine
Benzylmercapto phthalocyanine
Cyclic voltammetry
was observed in complex 3a. Complex 3b showed both ring-based oxidation, attributed to CoIIIPcꢀ1
/
CoIIIPcꢀ2 species (Ep = +0.86 V versus Ag|AgCl), and ring-based reduction associated with CoIPcꢀ2/CoIPcꢀ3
species (E = ꢀ1.46 V versus Ag|AgCl), with the normal metal-based redox processes in CoPc complexes:
½
Spectroelectrochemisrty
CoIIIPcꢀ2/CoIIPcꢀ2 (Ep = +0.41 V versus Ag|AgCl) and CoIIPcꢀ2/CoIPcꢀ2 (E = ꢀ0.38 V versus Ag|AgCl). Solu-
½
tion electrochemistry of complex 5 showed the same type and number of species observed in complex
3a: CoIIIPcꢀ2/CoIIPcꢀ2 (Ep = +0.59 V versus Ag|AgCl), CoIIPcꢀ2/CoIPcꢀ2 (E = ꢀ0.26 V versus Ag|AgCl) and
½
CoIPcꢀ2/CoIPcꢀ3 (E = ꢀ1.39 V versus Ag|AgCl) species. These processes were confirmed using
½
spectroelectrochemistry.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
sors; which is anchored on the interesting coordination chemistry
of cobalt. Cobalt exhibits three oxidation states: CoI, CoII and CoIII.
Metallophthalocyanine (MPc) complexes are some of the most
researched molecules in modern chemistry. The extensive research
interest generated by these macrocycles stems from their versatil-
ity, which is underscored by their wide range of applications. These
applications include their uses in electronic devices, as gas sensors,
as photosensitizers, in non-linear optics, in electrochromic devices
and in Langmuir–Blodgett films [1,2]. Metallophthalocyanine com-
plexes containing electroactive metals, such as cobalt, iron and
other transition metals have also been extensively studied because
of their electrocatalytic properties towards many analytes [3,4].
Importantly, the properties of these complexes can be tuned in or-
der to meet specific qualities, suitable for their applications, by
varying the central metal, the nature and position (peripheral
and non-peripheral) of the substituent.
CoI does not bind any axial ligand; hence exist essentially in a
square planar environment. CoII does bind axial ligands but with
low affinity, which suggests CoII may exhibits coordination number
of (i) four (square planar), (ii) five (penta-coordinate) and rarely
(iii) six (hexa-coordinate). CoIII has high affinity for axial ligands
and is usually stabilized by two axial ligands; hence exist predom-
inantly in an octahedral environment. Axial ligation has direct ef-
fect on the solution electrochemistry of CoPc complexes. CoIPc
and CoIIPc species can be observed in both coordinating and non-
coordinating solvents but CoIIIPc species is only observed in coordi-
nating solvents, such as dimethyl sulphoxide (DMSO), dimethyl-
formamide (DMF) and pyridine [8]. These solvents bind, as axial
ligands, to two of the six coordination sites on CoIII, offering some
form of stability to this oxidation state, hence the possibility of
observing CoIIIPc species in the presence of these solvents. Apart
from metal-based processes, redox activities can also be observed
on the ring in CoPc complexes. The formation of CoIIIPc species is
usually followed by the first ring oxidation (formation of CoIIIPcꢀ1
species); while the first ring reduction is preceded by the forma-
tion of CoIPc species.
Cobalt phthalocyanine (CoPc) complexes are the most inten-
sively studied phthalocyanines [4–7] because of their potential
use as electrocatalysts and for the design of electrochemical sen-
* Corresponding author. Tel.: +27 46 6038260; fax: +27 46 6225109.
E-mail address: t.nyokong@ru.ac.za (T. Nyokong).
0277-5387/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2010.01.004

1258
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
The focus of the present study is to report the synthesis and
2.4. Synthesis
investigate the effects of the nature and position of substituent
on both the spectral, electrochemical and spectroelectrochemical
qualities of new CoPc complexes tetra-substituted with diethylam-
inoethanethio (peripherally: complex 3a and non-peripheral: com-
plex 3b) and benzylmercapto (non-peripherally: complex 5)
substituents. The presence of these substituents makes these com-
plexes good candidates in the formation of thin films, either by
electrochemical polymerization or self-assembly technique, thus
facilitating their use as electrocatalysts and in the development
of electrochemical sensors. The synthesis of CoPc peripherally
octa-substituted with diethylaminoethanethio has been reported
[9]. This work reports on the peripherally (3a) and non-peripher-
ally (3b) tetra-substituted CoPc derivatives. The synthesis of CoPc
peripherally tetra-substituted with benzylmercapto (6) has also
been reported [10]. This work reports on the non-peripheral sub-
stitution with benzylmercapto (5).
2.4.1. 4-(2-Diethylaminoethanethiol) phthalonitrile (2a) and 3-(2-
diethylaminoethanethiol) phthalonitrile (2b)
4-Nitrophthalonitrile (compound 1a, Scheme 1a) and 3-
nitrophthalonitrile (compound 1b, Scheme 1b), were synthesized
according to reported procedure [11]. Compound 2a was synthe-
sized using the method reported for the synthesis of 1,2-bis-(dieth-
ylaminoethanethiol)-4,5-dicyanobenzene
[12]
with
some
modifications as follows: compound 1a (3.05 g, 17.6 mmol) was
dissolved in anhydrous DMF (150 ml) under nitrogen and 2-dieth-
ylaminoethanethiol hydrochloride (9 g, 53.0 mmol) was added.
After stirring for 10 min, finely ground anhydrous K2CO3 (29.3 g,
212.0 mmol) was added in portions over 2 h with stirring. The
reaction mixture was stirred at room temperature for 48 h under
nitrogen. Then the solution was poured into ice (900 g). The precip-
itate was filtered off, washed with water, until the filtrate was neu-
tral. The product was then dried in air. Yield: 3.42 g (75%). IR (KBr)
m
max/cmꢀ1: 3084–3018 (Ar–C–H), 2968–2807 (CH2), 2229 (CN),
2. Experimental
1582, 1542, 1471, 1384, 1289, 1225, 1198, 1142, 1072, 987, 906,
825, 732, 609, 521. 1H NMR (CDCl3) d = 7.66–7.54 (m, 3H, Ar–H),
3.15–312 (t, 2H, SCH2), 2.80–2.77 (t, 2H, NCH2), 2.62–2.57 (qnt,
4H, CH2C), 1.07–1.03 (t, 6H, CH3) ppm.
2.1. Materials
Potassium carbonate, cobalt chloride, 2-(diethylaminoethane-
thiol) hydrochloride and 2-diethylaminoethanol were obtained
from Sigma–Aldrich. Benzylmercapto thiol was obtained from Flu-
ka. Tetrabutylammonium tetrafluoroborate (TBABF4) (Aldrich) was
used as the electrolyte for electrochemical experiments. Aluminum
oxide, WN-3: neutral and Silica Gel 60 (0.04–0.063 mm), for col-
umn chromatography, were purchased from Sigma–Aldrich and
Merck, respectively. Dimethylformamide (DMF), dichloromethane
(DCM) and dimethyl sulphoxide (DMSO) were obtained from
Merck. DMF, DMSO, DCM, methanol and ethanol were distilled be-
fore use.
Compound 2b, Scheme 1b, was synthesized following the meth-
od described above for compound 2a, using 3-nitrophthalonitrile
(compound 1b) in place of 4-nitrophthalonitrile (compound 1a).
Yield: 2.74 g (60%). IR (KBr) m
max/cmꢀ1: 2969–2814 (CH2), 2229
(mCN), 1565, 1520, 1445, 1372, 1290, 1231, 1193, 1133, 1065,
1029, 987, 855, 787, 730, 547, 439. 1H NMR (CDCl3) d = 7.73–7.56
(m, 3H, Ar–H), 3.22–3.19 (t, 2H, SCH2), 2.82–2.79 (t, 2H, NCH2),
2.61–2.58 (dd, 4H, CH2C), 1.06–1.02 (t, 6H, CH3) ppm.
2.4.2. 3-(Benzylmercapto) phthalonitrile (4)
Compound 4 was synthesized (Scheme 2) following the proce-
dure reported for 4-(benzylmercapto) phthalonitrile [10] with
some modifications as follows: compound 1b (6 g, 34.68 mmol)
was dissolved in anhydrous DMSO (50 ml) under nitrogen and ben-
zylmercapto thiol (5.55 g, 45 mmol) was added. After stirring
strongly for 20 min, finely ground anhydrous K2CO3 (15 g,
108.7 mmol) was added in portions over 2 h with stirring. The
reaction mixture was stirred at room temperature for 12 h under
nitrogen. Thereafter, the crude product was precipitated out from
the reaction mixture with ice (600 g). The precipitate was filtered
off, washed with water, until the filtrate was neutral. The product
was crystallized twice from ethanol and dried in air. Yield: 1.91 g
2.2. Electrochemical studies
All electrochemical experiments were performed using Autolab
potentiostat PGSTAT 302 (Eco Chemie, Utrecht, The Netherlands)
driven by the general purpose Electrochemical System data pro-
cessing software (GPES, software version 4.9). Square wave voltam-
metric analysis was carried out at a frequency of 10 Hz, amplitude:
50 mV and step potential: 5 mV. A conventional three-electrode
system was used. The working electrode was a bare glassy carbon
electrode (GCE), Ag|AgCl wire and platinum wire were used as the
pseudo reference and auxiliary electrodes, respectively. The poten-
tial response of the Ag|AgCl pseudo-reference electrode was less
than the Ag|AgCl (3M KCl) by 0.015 0.003 V. Prior to use, the elec-
trode surface was polished with alumina on a Buehler felt pad and
rinsed with excess millipore water. All electrochemical experi-
ments were performed in freshly distilled dry DMF containing
TBABF4 as supporting electrolyte.
(22%). IR (KBr)
(CH2), 2226 (
m : 3067–3031 (Ar–C–H), 2936–2852
max/cmꢀ1
m
CN), 1961, 1561, 1500, 1447, 1289, 1187, 1155,
1066, 1030, 911, 854, 785, 709, 589, 555, 483, 434. 1H NMR (CDCl3)
d = 7.62–7.55 (m, 3H, S–Ar–H), 7.36–7.30 (m, 5H, C–Ar–H), 4.31 (s,
2H, CH2) ppm.
2.4.3. Cobalt tetrakis-(2-diethylaminoethanethiol) phthalocyanines:
3a (peripheral) and 3b (non-peripheral)
Complexes 3a (Scheme 1a) and 3b (Scheme 1b) were synthe-
sized following the method reported for the octa-substituted ana-
logue [12] with some modifications.
Spectroelectrochemical data were obtained using a home-made
optically transparent thin-layer electrochemical (OTTLE) cell which
was connected to
voltammograph.
a Bioanalytical Systems (BAS) CV 27
2.4.3.1. Complex 3a, Scheme 1a. A mixture of compound 2a (0.40 g,
1.54 mmol) and cobalt(II) chloride (0.049 g, 0.38 mmol) was re-
fluxed in 2-(diethylaminoethanol) (1.2 ml) for 12 h under nitrogen.
Thereafter, the mixture was cooled to room temperature and trea-
ted with excess MeOH:H2O (1:1) to precipitate the crude deep blue
product. The product was filtered and dried in air. Purification was
achieved using column chromatography with neutral alumina as
column material and DCM/MeOH (10:1) as eluent. Yield: 1.08 g
(64%). Anal. Calc. for C56H68N12S4CoꢁCH2Cl2: C, 56.95; H, 5.76; N,
2.3. Equipment
UV–Vis spectra were recorded on Cary 50 UV–Vis/NIR spectro-
photometer. IR (KBr discs) was recorded on Bruker Vertex 70-Ram
II spectrophotometer. Elemental analysis was performed using
Vario Elementar Microcube EL111. 1H nuclear magnetic resonance
(1H NMR, 400 MHz) was obtained in CDCl3 using Bruker EMX 400
NMR spectrometer.

I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
1259
NC
NC
N
NC
K2CO3
+
HS
NO2
DMF
S
NC
1a
2a
N
CoCl 2
Diethylaminoethanol
N
S
N
N
N
N
N
N
N
S
N
N
Co
N
S
S
N
3a
Scheme 1a. Synthetic pathway for the formation of complex 3a.
14.23; S, 10.85. Found: C, 56.87; H, 5.78; N, 13.89; S, 11.25%. UV–
Vis (DMF): kmax (nm) (log ): 675 (5.6), 615 (5.1), 406 (4.8), 352
(5.1); IR (KBr)
max/cmꢀ1; 2965–2800 (CH2), 1601, 1519, 1449,
0.49 mmol) was refluxed in 2-(diethylaminoethanol) (3 ml) for
6 h under nitrogen. The deep green crude product was precipitated,
after cooling, in excess methanol. The precipitate was filtered and
dried in air. Purification was performed using column chromatog-
raphy with Silica Gel as column material and CHCl3/MeOH (10:1)
as eluent. Yield: 1.06 g, (42%). Anal. Calc. for C60H40N8S4CoꢁCH2Cl2:
C, 62.93; H, 3.50; N, 9.79; S, 11.19. Found: C, 62.61; H, 3.91; N,
e
m
1393, 1312, 1143, 1091, 930, 815, 771, 748, 689.
2.4.3.2. Complex 3b, Scheme 1b. Complex 3b was synthesized
(Scheme 1b) and purified following the same method described
for complex 3a, using compound 2b (0.40 g, 1.54 mmol) as the
starting material in place of 2a. Yield: 1.01 g (60%). Anal. Calc. for
C56H68N12S4Co: C, 61.37; H, 6.21; N, 15.34; S, 11.69. Found: C,
60.96; H, 6.31; N, 14.80; S, 12.11%;. UV–Vis (DMF): kmax (nm)
9.42; S, 10.68%. UV–Vis (DMF): kmax (nm) (log
e): 692 (5.5), 355
(5.3); IR (KBr)
m
max/cmꢀ1; 3059 (w, Ar–C–H), 2920 (w, CH2) 1584,
1448, 1318, 1241, 1114, 746, 702.
(log e): 693 (5.3), 628 (4.8), 478 (4.2); IR (KBr) m
max/cmꢀ1; 2966–
2799 (CH2), 1619, 1572, 1514, 1462, 1384, 1318, 1239, 1113,
3. Results and discussion
919, 739, 595.
3.1. Synthesis and spectral characterization
2.4.4. Cobalt tetrakis-(benzylmercapto) phthalocyanine 5 (non-
peripheral), Scheme 2
Complex 5 was synthesized following the method reported for
the peripheral derivative [10] with some modifications. A mixture
of compound 4 (0.6 g, 1.85 mmol) and cobalt(II) chloride (0.063 g,
The synthetic routes for complexes 3a, 3b and 5 are shown in
Schemes 1 and 2. The phthalonitriles, compounds 2a, 2b and 4
were obtained via a based-catalyzed (K2CO3) nucleophilic aromatic
substitution reaction. Cyclotetramerization of 2a, 2b and 4 oc-
curred in the presence of CoCl2 to form the desired complexes:

1260
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
NC
NC
NC
K2CO3
N
HS
+
NC
DMF
S
NO2
2b
N
1b
CoCl2
Diethylaminoethanol
N
N
S
S
N
N
N
Co
N
N
N
N
N
S
S
N
N
3b
Scheme 1b. Synthetic pathway for the formation of complex 3b.
3a, 3b and 5, respectively. Purifications of 3a and 3b were accom-
plished using column chromatography on alumina while that of 5
was performed on Silica Gel. The complexes were soluble in sol-
vents such as DMF, DCM and DMSO.
Characterizations of the complexes were carried out using IR
and UV–Vis spectroscopies as well as elemental analysis. The for-
mation of complexes 3a, 3b and 5 was confirmed by the disappear-
ance of the sharp C„N vibration at 2229, 2229 and 2226 cmꢀ1 of
2a, 2b and 4, respectively.
band. The degree of steric limitation is closely related to the com-
plexity of the substituents. In the present study, a red-shift in Q-
band, characteristic of non-peripherally substituted MPcs com-
plexes, was observed for complex 3b compared to complex 3a
(Fig. 1). The Q-band of complex 5 (692 nm) was also red-shifted
(bathochromic effect) compared to that of the recently reported
peripheral analogue (674 nm) [10]. However, the position of Q-
band in complex 3b (693 nm), Fig. 1, was similar to that of complex
5 (692 nm).
Fig. 1 shows the UV–Vis spectra of complexes 3a, 3b and 5
(3.82 ꢂ 10ꢀ6 M) in DMF. The effects of the nature and position of
substituent on the spectral qualities of these complexes are clearly
observable. The position and nature of substituent impact both
hypochromic and bathochromic effects on the spectral properties
of these complexes. The shift in spectra and changes in absorption
behavior have been explained in terms of conformational distor-
tion caused by the stress of the substituents in MPcs complexes
[13] and have also been attributed to coordination of certain li-
The spectra in Fig. 1 are relatively broad suggesting aggregation.
Fig. 2a (1.28–8.96 ꢂ 10ꢀ6 M), 2b (1.27–8.89 ꢂ 10ꢀ6 M) and 2c
(1.27–5.72 ꢂ 10ꢀ6 M) shows the effect of changing concentration
on the spectra of complexes 3a, 3b and 5, respectively. Beer0s law
was obeyed within the range of concentration investigated for all
the complexes.
3.2. Cyclic and square wave voltammetry studies
gands to the central metal and enlargement of the
p-system of
3.2.1. Complex 3a
the Pc [14–16]. They can also be attributed to aggregation in MPcs.
Complex 3a, (peripheral), shows Q-band absorption at 675 nm
in DMF, while that of complex 3b, (non-peripheral) was observed
at 693 nm, hence the latter was red-shifted compared to that of
complex 3a. Usually, non-peripheral substitution in MPc com-
plexes imposes some measure of steric limitation on these macro-
cycles, leading to conformational distortion of the Pc skeleton,
resulting in change in absorbance and shift in the position of Q-
Fig. 3a and b shows, respectively, the square wave and cyclic
voltammetry profiles of 1 ꢂ 10ꢀ3 M of complex 3a in freshly dis-
tilled dry DMF containing 0.1 M TBABF4 as supporting electrolyte.
Three distinctly defined redox processes (labeled I, II and III) can
be observed. Process III (Fig. 3b) is assigned to metal oxidation,
CoIIIPcꢀ2/CoIIPcꢀ2 (E = +0.64 V versus Ag|AgCl), Table 1 in accor-
½
dance with literature [8]. Metal oxidation is expected, since this
species is usually observed in coordinating solvents [8]. This pro-

I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
1261
NC
NC
NC
DMSO
+
HS
CH2
NC
K2CO3
NO2
SR
4
1b
CH2
R =
CoCl2
Diethylaminoethanol
SR
SR
SR
N
N
N
N
N
N
Co
N
N
N
N
Co
N
N
N
N
SR
N
RS
N
SR
RS
5
SR
6
(Published before [10])
Scheme 2. Synthetic pathway for the formation of complex 5.
2.5
2
1.5
1
3a
3b
5
0.5
0
350
400
450
500
550
600
650
700
750
800
Wavelength (nm)
Fig. 1. UV–Vis spectral of 3.82 ꢂ 10ꢀ6 M of complexes 3a, 3b and 5 in DMF.
cess is virtually irreversible. The irreversibility of oxidation pro-
cesses of sulfur containing MPc processes is well documented
[17,18] hence the irreversibility of process III is not surprising. Pro-
cess
II
is
assigned
to
metal
reduction,
CoIIPcꢀ2
/
CoIPcꢀ2(E = ꢀ0.24 V versus Ag|AgCl). It has cathodic to anodic
½
peak current ratio of near unity, but quasi-reversible in that peak

1262
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
4
3.5
3
a
4
3
2
1
0
y = 0.371x + 0.2109
R² = 0.9964
0
2
4
6
8
10
Concentration (M) (10 -6
)
2.5
2
1.5
1
0.5
0
350
400
450
500
550
600
650
700
750
800
850
Wavelength (nm)
2
1.8
1.6
1.4
1.2
1
2
1.6
1.2
0.8
0.4
y = 0.2039x + 0.086
R² = 0.9971
b
0
0
2
4
6
8
10
Concentration (M) (10 -6
)
0.8
0.6
0.4
0.2
0
350
400
450
500
550
600
650
700
750
800
850
Wavelength (nm)
3
2.5
2
c
2
1.5
1
y = 0.2845x + 0.1465
R² = 0.9974
0.5
0
0
2
4
6
8
Concentration (M) (10 -6
)
1.5
1
0.5
0
350
450
550
650
750
850
Wavelength (nm)
Fig. 2. Effect of changing concentration on the UV–Vis spectra of (a) complex 3a (1.28–8.6 ꢂ 10ꢀ6 M), (b) complex 3b (1.27–8.89 ꢂ 10ꢀ6 M) and (c) complex 5 (1.27–
5.72 ꢂ 10ꢀ6 M). Inset: Plot of absorbance vs. concentration. Solvent = DMF.
to peak separation (
D
E) of 161 mV was obtained (
D
E for ferrocene
cathodic to anodic current ratio of near unity. Spectroelectrochem-
istry is used below to confirm these processes.
was 90 mV within the same potential window with the same scan
rate (100 mVsꢀ1 versus Ag|AgCl)). Reversibility did not improve at
shorter potential window. Process I is assigned to ring reduction,
3.2.2. Complex 3b
Electrochemical properties of complex 3b were slightly differ-
ent from that of complex 3a. Effect of the position of substituent
CoIPcꢀ2/CoIPcꢀ3 (E = ꢀ1.26 V versus Ag|AgCl). This process was
½
also quasi-reversible with
DE value of 151 mV, although with

I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
1263
a
46µA
II
I
III
-2
-1.5
-1
-0.5
0
0.5
1
1.5
Potential/V vs. Ag|AgCl
b
III
25µA
II
I
-2
-1.5
-1
-0.5
0
0.5
1
1.5
Potential/V vs. Ag|AgCl
Fig. 3. (a) Square wave and (b) cyclic voltammetry profiles of 1 ꢂ 10ꢀ3 M of complex 3a in freshly distilled DMF containing 0.1 M TBABF4 supporting electrolyte. Step
potential: 5 mV, amplitude: 50 mV vs. Ag|AgCl, frequency: 10 Hz. Scan rate: 100 m Vsꢀ1 vs. Ag|AgCl.
Table 1
Summary of peak and half wave potentials, in volt vs. Ag|AgCl, of the new complexes in comparison with that of previously reported thio derivatised CoPc complexes. Values were
recorded in DMF containing TBABF4 unless otherwise stated.
Complexa
CoIPcꢀ3/CoIPcꢀ2
CoIIPcꢀ2/CoIPcꢀ2
CoIIIPcꢀ2/CoIIPcꢀ2
CoIIIPcꢀ1/CoIIIPcꢀ2
Referenceb
3a
3b
5
6
ꢀ1.26
ꢀ1.46
ꢀ1.39
ꢀ1.41
ꢀ0.58
ꢀ0.24
ꢀ0.38
ꢀ0.26
ꢀ0.38
ꢀ0.096
ꢀ0.46
+0.64
+0.41
+0.59
+ 0.42
+0.72
+0.44
TW
TW
TW
10
22
10
+0.86
+0.89
+1.16
+0.66
CoOBMPcc
CoTDMPcd
a
b
c
TDMPc = tetra dodecylmercapto phthalocyanine (peripheral) and OBMPc = octa benzylmercapto phthalocyanine (peripheral).
TW = this work.
Values recorded in DMF using tetrabutylammonium perchlorate instead of TBABF4.
Values recorded in dichloromethane (instead of DMF) using TBABF4.
d
on the electrochemical properties of complex 3b was obvious
when compared to that of complex 3a. As discussed previously,
non-peripheral positions are sterically crowded positions that
may impose some conformational stress on MPc complexes.
Fig. 4a and b shows, respectively, the square wave and cyclic vol-
tammetry profiles of 1 ꢂ 10ꢀ3 M of complex 3b in freshly distilled
dry DMF containing 0.1 M TBABF4. Two irreversible oxidations,
(process III and IV) are observed. Process III at Ep = +0.41 V versus
Ag|AgCl is assigned to a metal based oxidation: CoIIIPcꢀ2/CoIIPcꢀ2
and process IV at Ep = +0.86 V versus Ag|AgCl to ring based oxida-
tion: CoIIIPcꢀ1/CoIIIPcꢀ2. The larger currents for IV are due to the
ring based Pc ring oxidation and the oxidation of the substituents

1264
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
a
IV
14µA
I
II
III
-2
-1.5
-1
-0.5
0
0.5
1
1.5
Potential/V vs. Ag|AgCl
b
IV
III
II
40µA
I
-2
-1.5
-1
-0.5
0
0.5
1
1.5
Potential/V vs. Ag|AgCl
Fig. 4. (a) Square wave and (b) cyclic voltammetry profiles of 1 ꢂ 10ꢀ3 M of complex 3b in freshly distilled DMF containing 0.1 M TBABF4 supporting electrolyte. Step
potential: 5 mV, amplitude: 50 mV vs. Ag|AgCl, frequency: 10 Hz. Scan rate: 100 m Vs-1 vs. Ag|AgCl.
as is typical of sulfur containing MPc complexes [17,18]. Two
reduction processes (I and II) were observed. The metal reduction
The absence of ring oxidation in complex 3a (Fig. 3b) underscores
the limitation imposed on the complex with respect to undergoing
oxidative processes. These observations were expected, since non-
peripheral substitution makes the electron-donating substituent
closer to the ring relative to peripheral substitution, thus increas-
ing the electron-donating tendency of the substituent, making
complex 3b more susceptible to oxidation than complex 3a. This
suggestion was corroborated by results obtained from spectro-
electrochemical studies, discussed later in this work.
process II (at E = ꢀ0.38 V versus Ag|AgCl) is reversible with
DE va-
½
lue of 81 mV and cathodic to anodic peak current ratio of near
unity, and is assigned to CoIIPcꢀ2/CoIPcꢀ2. Process I at E = ꢀ1.46 V
½
versus Ag|AgCl is then assigned to ring reduction (CoIPcꢀ2/CoIPcꢀ3
)
as is typical of CoPc complexes in DMF solutions. This couple was
quasi-reversible with E value of 110 mV.
D
The more sensitive square wave voltammetry technique showed
other peaks whose origin may be due to aggregated species at the
concentrations used for cyclic voltammtery [8]. The redox pro-
cesses in 3b were confirmed using spectroelectrochemistry.
Ring-based reduction (Pcꢀ2/Pcꢀ3) occurred at ꢀ1.26 V and
ꢀ1.46 V (versus Ag|AgCl) in complexes 3a and 3b, respectively,
while metal-based reductions (CoII/CoI) was observed at ꢀ0.24 V
and ꢀ0.38 V in complexes 3a and 3b, respectively, thus showing
that peripheral substitution (3a) results in easier reduction than
non-peripheral substitution (3b), while oxidation is easier for
non-peripheral derivative (3b) as Table 1 shows. The ease of oxida-
tion in 3b could be a result of conformational distortion, induced
by non-peripheral substitution.
3.2.3. Complex 5
Fig. 5a and b, respectively, shows the square wave and cyclic
voltammetry profiles of 1 ꢂ 10ꢀ3 M of complex 5 in freshly dis-
tilled dry DMF containing 0.1 M TBABF4. Electrochemical proper-
ties of complex 5 were not substantially different from that of
complex 3a. The major difference was the extent of reversibility
of the redox processes. Three distinct redox processes were ob-
served in complex 5 as was the case for 3a. Like complex 3a, no
ring-based oxidation was observed in complex 5. Although, com-
plex 5 is non-peripherally substituted, like complex 3b, the elec-
tron-withdrawing nature of the phenyl group of the
benzylmercapto substituent favored reduction processes at the ex-
pense of oxidation, Table 1. Process III was an irreversible (and
The metal-based oxidation (CoIII/CoII), the only oxidative pro-
cess observed in complex 3a, was observed at +0.64 V and
+0.41 V (versus Ag|AgCl) in complexes 3a and 3b, respectively.
weak) process, and is assigned to metal oxidation, CoIIIPcꢀ2
/

I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
1265
a
I
23µA
III
II
-2
-1.5
-1
-0.5
0
0.5
1
1.5
Potential/V vs. Ag|AgCl
b
III
II
30µA
I
-2
-1.5
-1
-0.5
0
0.5
1
1.5
Potential/V vs. Ag|AgCl
Fig. 5. (a) Square wave and (b) cyclic voltammetry profiles of 1 ꢂ 10ꢀ3 M of complex 5 in freshly distilled DMF containing 0.1 M TBABF4 supporting electrolyte. Step potential:
5 mV, amplitude: 50 mV vs. Ag|AgCl, frequency: 10 Hz. Scan rate: 100 m Vsꢀ1 vs. Ag|AgCl.
CoIIPcꢀ2(Ep = +0.59 V versus Ag|AgCl). This process was also ener-
getically less feasible, as it was with complex 3a, when compared
710 nm), with increased intensity, was observed. Considerable
disaggregation of the complex was also noticed upon reduction,
as evident from the narrowing of the absorption bands. Also there
was emergence of a new intense peak at 474 nm during reduc-
tion. New peaks with intense signal between 400 and 500 nm
are typical of the formation of CoIPc species [19]. The shift in
Q-band without decrease in intensity is a signature of metal-
based reduction in MPc complexes. Hence, spectral changes
shown in Fig. 6a are a confirmation of the reduction of CoII to
CoI, corroborating the assignment of process II, in Fig. 3b, to
CoIIPcꢀ2/CoIPcꢀ2 species. Also clear isobestic points, observed at
556 and 691 nm, indicated that the process is a clean reduction
involving two species. However, the first scan does not fall within
the isosbestic point due to disaggregation during reduction. About
85% of the original spectrum was regenerated on the application
of zero potential, showing some measure of reversibility of the
process. The number of electron transferred was calculated to
be approximately 1, using Eq. (1):
to complex 3b (Ep = +0.41 V versus Ag|AgC. Process II, CoIIPcꢀ2
/
CoIPcꢀ2 (E = ꢀ0.26 V versus Ag|AgCl), was quasi-reversible,
½
although with a cathodic to anodic peak current ratio of near unity,
but a peak separation larger than 90 mV for ferrocene. Process I,
CoIPcꢀ2/CoIPcꢀ3 (E = ꢀ1.39 V versus Ag|AgCl), was also quasi-
½
reversible. The quasi-reversibility of the processes suggests slug-
gish electron transfer in all cases.
3.3. Spectroelectrochemical studies
3.3.1. Complex 3a
The spectral changes observed on the application of potential
slightly more negative of that of process II (E = ꢀ0.24 V versus
½
Ag|AgCl, Fig. 3b) are shown in Fig. 6a. The presence of split Q-
band in the starting spectrum was an indication of aggregation,
as result of the concentration used for controlled electrolysis.
The peak at about 628 nm is due to the aggregate while that at
672 nm is the monomeric peak. The wavelength of the Q-band
of the peak associated with the monomer is slightly lower than
that observed for the complex in Fig. 1 (675 nm) due to the ef-
fects of the electrolyte. Upon reduction at potentials of process
II, a red-shift in Q-band of the monomeric peak (from 672 to
Q ¼ nFVC
ð1Þ
where n, F, V and C are the number of electrons transferred, Fara-
day’s constant, volume and concentration of the electroactive spe-
cies, respectively. Further reduction of the species formed in
Fig. 6a, at potential more negative of process I (E = ꢀ1.26 V versus
½

1266
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
a
350
450
550
650
750
850
Wavelength (nm)
b
350
450
550
650
750
850
Wavelength (nm)
c
350
450
550
650
750
850
Wavelength(nm)
Fig. 6. UV–Vis spectral changes observed for complex 3a during controlled potential electrolysis at (a) ꢀ0.3 V (process II), (b) ꢀ1.4 V (process I) and (c) +0.7 V (process III) vs.
Ag|AgCl. Dotted line in (a) and (c) (spectrum before electrolysis), dashed line (spectrum after electrolysis). Electrolyte = DMF containing 0.1 M TBABF4. The final spectrum in
(a) is the starting spectrum in (b).
Ag|AgCl, Fig. 3b) resulted in the spectral changes shown in Fig. 6b.
Decrease in intensity of the new Q-band with shift in position of
the band at 474–478 nm was noticed. Also there was formation of
new bands at 555 and 585 nm. Emergence of new bands in the
500–600 nm region is characteristic of ring-based reduction and
the formation of Pcꢀ3 species [20]. Hence, the spectral changes ob-

I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
1267
served in Fig. 6b confirmed the assignment of process I (in Fig. 3b)
to CoIPcꢀ2/CoIPcꢀ3 species.
Ag|AgCl, Fig. 4b). The Q-band of the starting spectrum (694 nm)
is similar to that of the spectrum of complex 3b (693 nm) in
Fig. 1, showing minimal effects of the electrolyte which was ob-
served for 3a. Reduction resulted in shift of the Q-band from 694
to 712 nm with constant intensity. There was also the emergence
of new intense band at 473 nm. Presence of new intense peaks be-
tween 400 and 500 nm region typifies the formation of CoIPc spe-
cies, as discussed previously [19]. The shift in Q-band without
change in intensity is characteristic of metal-based reduction in
MPc complexes. Thus, the spectral changes noticed in Fig. 7a con-
firms the assignment of process II (Fig. 4b) to CoIIPcꢀ2/CoIPcꢀ2 spe-
cies. Clear isobestic points, noticed at 570 and 705 nm, showed
that process II (Fig. 4b) is a clean reduction involving two species.
Regeneration of over 90% of the starting spectrum, on the applica-
tion of zero potential, confirmed the reversibility of the process.
The number of electron transferred was calculated to be approxi-
mately 1 using Eq. (1). The species obtained in Fig. 7a was further
reduced (Fig. 7b) on the application of potential more negative of
The spectral changes in Fig. 6c were obtained on the application
of potential more positive of process III (E = +0.64 V versus
½
Ag|AgCl, Fig. 3b). The starting spectrum has a split Q-band as ob-
served in Fig. 6a. This can be attributed to aggregation as discussed
earlier. The higher energy peak (628 nm) is due to the aggregate
while the lower energy peak (672 nm) is associated with the
monomer. Upon oxidation, there was increase in intensity of the
Q-band due to the monomer and shift in position from 672 to
677 nm. These spectral changes are consistent with the oxidation
of CoII to CoIII species. Thus, confirming the assignment of process
III (in Fig. 3b) to CoIIIPcꢀ2/CoIIPcꢀ2 couple. However, the minimal
shift in position and slight change in intensity of the Q-band in
Fig. 6c underline the difficulty, highlighted previously, with which
oxidative processes occurred in complex 3a.
3.3.2. Complex 3b
that of couple I (E = ꢀ1.46 V versus Ag|AgCl, Fig. 4b). There was
Fig. 7a shows the spectral changes observed on the application
½
a decrease in intensity of the new Q-band and emergence of new
of potential more negative of process II (E = ꢀ0.38 V versus
½
a
350
450
550
650
750
850
Wavelength (nm)
b
350
450
550
650
750
850
Wavelength (nm)
Fig. 7. UV–Vis spectral changes observed for complex 3b during controlled potential electrolysis at (a) ꢀ0.45 V (process II), (b) ꢀ1.5 V (process I), (c) +0.5 V (process III) and
(d) +0.9 V (process IV) vs. Ag|AgCl. Dotted line in (a) (spectrum before electrolysis), dashed line (spectrum after electrolysis). Electrolyte = DMF containing 0.1 M TBABF4. The
final spectra in (a) and (c) are the starting spectra in (b) and (d) respectively.

1268
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
c
350
450
550
650
750
850
Wavelength (nm)
d
350
450
550
650
750
850
Wavelength (nm)
Fig. 7 (continued)
band (570 nm) between 500 and 600 nm region. These spectral
changes are consistent with the formation of Pcꢀ3 species, as dis-
cussed previously [20]. This confirms the assignment of couple I
(Fig. 4b) to CoIPcꢀ2/CoIPcꢀ3 species.
the complex as reported previously for other thiol derivatised
MPc complexes [21,22]. Nonetheless, the assignment of process
IV to CoIIIPcꢀ1/CoIIIPcꢀ2 was consistent with the nature of CoPc
complexes (Table 1)
Application of potential more positive of process III
(Ep = +0.41 V versus Ag|AgCl, Fig. 4b) resulted in the spectral
changes shown in Fig. 7c. There was red-shifting of the Q-band
of the starting spectrum (from 694 to 702 nm), with increased
intensity, upon oxidation. These spectral changes are consistent
with the oxidation of CoII to CoIII. Hence, confirming the assign-
ment of process III (Fig. 4b) to CoIIIPcꢀ2/CoIIPcꢀ2 species. The well
defined spectral changes observed for this process, relative to that
for the corresponding process in complex 3a (Fig. 6c), corroborate
the previous suggestion that complex 3b was more susceptible to
oxidative process than complex 3a. Further oxidations are ex-
pected to occur on the ring by virtue of the nature of CoPcs [18].
However, application of potential more positive of process IV
(Ep = +0.86 V versus Ag|AgCl, Fig. 4b) resulted in the spectral
changes shown in Fig. 7d. A decrease in Q-band intensity, without
any appreciable increase in intensity around the 500 nm region,
was observed. Increased intensity around the 500 nm region is
characteristic of ring oxidation in MPc [18]. Hence, spectral
changes observed in Fig. 7d may have suggested degradation of
3.3.3. Complex 5
The spectral changes observed on the application of potential
more negative of process II (E = ꢀ0.26 V versus Ag|AgCl, Fig. 5b)
½
is shown in Fig. 8a. The starting spectrum is the same as that
shown for complex 5 in Fig. 1. The band around 628 nm is associ-
ated with vibronic transition but complicated by slight aggregation
tendency. The main Q-band at 692 nm was red-shifted upon reduc-
tion (692–707 nm) without decrease in intensity. Also, there was
the appearance of new peak, with increased intensity, at 473 nm.
Presence of new peak with increased intensity in the 400–
500 nm region is a signature of CoIPc [19], as observed for Figs.
6a and 7a. The shift in Q-band, with the same intensity, is in agree-
ment with metal-based reduction in MPc. Therefore, the spectral
changes observed in Fig. 8a confirmed the assignment of process
II (Fig. 5b) to CoIIPcꢀ2/CoIPcꢀ2 species. Clear isobestic point at
572 and 700 nm showed that the process was a clean reduction,
involving two species. The number of electron transfer was
obtained to be approximately 1 using Eq. (1). Less than 70% of

I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
1269
a
350
450
550
650
750
850
Wavelength (nm)
b
350
450
550
650
750
850
Wavelength/nm
c
350
450
550
650
750
850
Wavelength/nm
Fig. 8. UV–Vis spectral changes observed for complex 5 during controlled potential electrolysis at (a) ꢀ0.35 V (process II), (b) ꢀ1.45 V (process I) and (c) +0.65 V (process III)
vs. Ag|AgCl. Dotted line (spectrum before electrolysis), dashed line (spectrum after electrolysis). Electrolyte = DMF containing 0.1 M TBABF4. The final spectrum in (a) is the
starting spectrum in (b).
the original spectrum was regenerated on the application of zero
potential. Although, this shows some measure of reversibility, it
does testifies to the sluggish nature of electron transfer process
in this complex, relative to that of the corresponding process in
complexes 3a and 3b. The spectral changes are also not as resolved
as that observed for the corresponding process in complex 3a

1270
I.A. Akinbulu, T. Nyokong / Polyhedron 29 (2010) 1257–1270
(Fig. 6a) and complex 3b (Fig. 7a), which shows that electron trans-
fer processes are better in these complexes than in complex 5. Fur-
ther reduction of the species obtained in Fig. 8a occurred on the
confirmed the assigned processes in these complexes. The spectral
changes noticed were in agreement with that of CoPc complexes
and they corroborated the observed electrochemical identities of
the new complexes.
application of potential more negative of process I (E = ꢀ1.39 V
½
versus Ag|AgCl, Fig. 5b) resulting in spectral changes shown in
Fig. 8b. There was a decrease in the absorbance of the new Q-band,
emergence of new band between 500 and 600 nm region and shift
in position of the peak at 473 to 477 nm. These spectral changes
are characteristic of Pcꢀ3 species [20]. This confirms the assign-
ment of process I (Fig. 5b) to CoIPcꢀ2/CoIPcꢀ3 species.
Acknowledgements
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 Initia-
tive for Professor of Medicinal Chemistry and Nanotechnology and
Rhodes University.
Application of
a potential more positive of process III
(Ep = +0.59 V versus Ag|AgCl, Fig. 5b) gave the spectral changes
shown in Fig. 8c. An increase in absorbance and shift in position
of the Q-band (692 to 697 nm) of the starting spectrum were ob-
served. Although, these changes are minimal, they are characteris-
tic of the oxidation of CoII to CoIII. This confirms the assignment of
process III (Fig. 5b) to CoIIIPcꢀ2/CoIIPcꢀ2 species. The minimal spec-
tral changes noticed in Fig. 8c is a testimony to the reluctance of
the complex, like complex 3a, to undergo oxidative process, as ex-
plained previously.
References
[1] C.C. Leznoff, A.B.P. Lever, Phthalocyanines: Properties and Applications, vols.
1–4, Weinheim, VCH, 1989, 1993, 1996.
[2] N.B. Mckeown, Phthalocyanine Materials, Cambridge University Press, 1998.
[3] C.A. Caro, F. Bedioui, J.H. Zagal, Electrochim. Acta 47 (2002) 1489.
[4] J. Zagal, F. Bedioui, J.P. Dodelet (Eds.), N4-macrocyclic Metal Complexes,
Springer, New York, 2006.
[5] K.I. Ozoemena, T. Nyokong, in: C.A. Grimes, E.C. Dickey, M.V. Pishko (Eds.),
Encyclopedia of Sensors, vol. 3, American Scientific Publishers, California, 2006,
p. 157 (Chapter E).
[6] T. Nyokong, F. Bedioui, J. Porphyrins Phthalocyanines 10 (2006) 1101.
[7] S.-I. Mho, B. Ortiz, N. Doddapaneni, S.-M. Park, J. Electrochem. Soc. 142 (1995)
1047.
4. Conclusion
The syntheses of new cobalt phthalocyanine complexes are re-
ported. Their spectral, electrochemical and spectroelectrochemical
properties have been extensively investigated. The Q-band of com-
plex 3b was red-shifted, with lower intensity, relative to that of
complex 3a. This was attributed to conformational changes arising
from non-peripheral substitution. The Q-band of complex 5, like
that of complex 3b, was also red-shifted and of lower intensity
with respect to that of complex 3a. Change in the nature of substi-
tuent (diethylaminoethanethio versus benzylmercapto) did not
significantly affect the position of Q-band. All the complexes
showed interesting electrochemical properties consistent with that
of CoPc complexes. However, oxidative processes were energeti-
cally more feasible in complex 3b while complexes 3a and 5 were
more receptive to reductive processes. Complex 3b has the best
electron transfer processes by virtue of relatively smallest cathodic
to anodic peak differences of the observed redox processes, while
complexes 3a and 5 showed a rather sluggish electron transfer pro-
cesses, evident from relatively larger cathodic to anodic peak dif-
ferences. Electrochemical properties were markedly affected by
the nature of substituent. Spectroelectrochemical investigation
[8] A.B.P. Lever, E.R. Milaeva, G. Speier, in: C.C. Leznoff, A.B.P. Lever (Eds.),
Phthalocyanines: Properties and Applications, vol. 3, VCH Publishers, New
York, 1993, p. 1.
[9] I.A. Akinbulu, T. Nyokong, Polyhedron 28 (2009) 2831.
[10] B.O. Agboola, K.I. Ozoemena, T. Nyokong, Electrochim. Acta 51 (2006) 4379.
[11] B.N. Acher, G.M. Fohlen, J.A. Parker, J. Keshavayya, Polyhedron 6 (1987) 1463.
[12] Z.A. Bayur, Dyes Pigm. 65 (2005) 235.
[13] I. Chambrier, M.J. Cook, P.T. Wood, Chem. Commun. (2000) 2133.
[14] T. Muto, T. Temma, M. Kimura, K. Hanabusa, H. Shirai, J. Org. Chem. 66 (2001)
6102.
[15] N. Kobayashi, O. Toshie, M. Sato, N. Shin-ichiro, Inorg. Chem. 32 (1993) 1803.
[16] N. Kobayashi, T. Ishizaki, K. Ishii, H. Konami, J. Am. Chem. Soc. 121 (1999)
9096.
[17] B. Agboola, K. Ozoemena, P. Westbroek, T. Nyokong, Electrochim. Acta 52
(2007) 2520.
[18] A.R. Ozkaya, A.G. Gurek, A. Gul, A.O. Bekaroglu, Polyhedron 16 (1997) 1877.
[19] M.J. Stillman, T. Nyokong, in: C.C. Leznoff, A.B.P. Lever (Eds.), Phthalocyanines:
Properties and Applications, vol. 1, VCH, New York, 1989 (Chapter 3).
[20] M.J. Stillman, in: C.C. Leznoff, A.B.P. Lever (Eds.), Phthalocyanines: Properties
and Applications, vol. 3, VCH Publishers, New York, 1993 (Chapter 5).
[21] K. Takahashi, M. Kawashima, Y. Tomita, M. Itoh, Inorg. Chim. Acta. 232 (1995)
69.
[22] F. Matemadomboa, M.D. Maree, K.I. Ozoemena, P. Westbroek, T. Nyokong, J.
Porphyrins Phthalocyanines 9 (2005) 484.