Electrosynthesis and electrochemical properties of porphyrin dimers with pyridinium as bridging spacer

Article abstract of DOI:10.1039/c1nj20177h

Two porphyrin dimers (more precisely two isomers) with pyridinium as bridging spacer have been obtained by controlled potential electrolysis. This method is based on a nucleophilic substitution of a porphyrin substituted by a pyridyl group (namely 5,10,15

Full text of DOI:10.1039/c1nj20177h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 2534–2543
www.rsc.org/njc
PAPER
Electrosynthesis and electrochemical properties of porphyrin dimers with
pyridinium as bridging spacer
Delphine Schaming,a Sonia Marggi-Poullain,a Iftikhar Ahmed,a Rana Farha,bc
Michel Goldmann,bd Jean-Paul Gisselbrechte and Laurent Ruhlmann*ae
Received (in Montpellier, France) 28th February 2011, Accepted 22nd July 2011
DOI: 10.1039/c1nj20177h
Two porphyrin dimers (more precisely two isomers) with pyridinium as bridging spacer have been
obtained by controlled potential electrolysis. This method is based on a nucleophilic substitution
of a porphyrin substituted by a pyridyl group (namely 5,10,15-tritolyl-20-(4-pyridyl)porphyrin
(H2T3P-4-Py) or 5,10,15-tritolyl-20-(3-pyridyl)porphyrin (H2T3P-3-Py)) on an electrogenerated
radical cation of a zinc b-octaethylporphyrin (ZnOEP). The new compounds have been
1
characterized by HR-MS, H NMR, UV-vis and fluorescence spectroscopies. Difference of
electrochemical behaviour of the pyridinium spacers between the two dimers has been also
discussed.
porphyrin with a Lewis base (like a pyridyl group for example)
substituted on the second porphyrin.8 The second approach
Introduction
Porphyrins constitute a major class of chemical compounds
due to their optical and redox properties,1,2 and because of
metals of two porphyrins.9
corresponds to the formation of an oxo-bond between the
their biological importance.3 Indeed, naturally occurring
Strategies allowing the formation of covalently linked
porphyrins play essential roles in photosynthesis, cellular
porphyrin dimers are more numerous, and a large variety of
respiration and biological electron transfer reactions.3 Thus,
linkages between the two macrocycles are permitted. For
considerable effort is devoted to design new materials that
instance, Osuka and Shimidzu have proposed a strategy based
incorporate porphyrins as molecular building blocks in
on the chemical oxidation of porphines allowing a direct
order to mimic natural systems.3–6 In particular, polynuclear
meso–meso coupling of the macrocycles.10 Another method
porphyrins attract considerable interest owing to the possibility
consists in using an elementary brick containing the bridging
of energy or electron transfer between the macrocycles.
spacer which is linked to four pyrroles. Afterwards, these
Therefore, various strategies have been developed to synthe-
pyrroles can be condensed to other compounds via aldehyde
functions in order to form two macrocycles on both sides.11–13
size porphyrin dimers (or oligomers).
Such dimers can be promoted for instance by electrostatic
But a more current strategy consists in using two porphyrins
interactions, coordination interactions or covalent bonds.
substituted by judicious groups allowing a coupling. For
For example, electrostatic cofacial dimers have been obtained
instance, the synthesis can be performed via metal-catalyzed
reactions (Pd-catalyzed Sonogashira14 or Stille15 coupling,
Cu-catalyzed Ullmann16 coupling. . .).
in aqueous solution between b-tetracationic and meso-
tetraanionic manganese(III) porphyrins.7 In the case of coor-
dination bonds, two approaches can be envisaged. The first
one consists in the axial coordination of the metal of one
In our precedent works, we have shown an electrochemical
method based on a nucleophilic attack of a Lewis base
containing two distinct nucleophilic sites which can react with
electrogenerated radical cations of two porphyrins. This
strategy has allowed us to synthesis porphyrin dimers with
diphosphonium bridges17,18 or viologen spacers.19
a
´
Laboratoire de Chimie Physique, UMR 8000 CNRS/Universite
´
ˆ
Paris-Sud 11, Faculte des Sciences d’Orsay, Batiment 349,
91405 Orsay Cedex, France. E-mail: laurent.ruhlmann@u-psud.fr;
Fax: +33 1 69 15 61 88; Tel: +33 1 69 15 44 38
b
´
Institut des NanoSciences de Paris, UMR 7588 CNRS/Universite
Using a similar electrochemical process, we have taken
porphyrins substituted by a pyridyl group which can also act
as a Lewis base towards the electrogenerated radical cation of
a second porphyrin. In this way, porphyrin dimers with a
pyridinium spacer can be obtained. Many porphyrins bearing
pyridinium units as substituents have already been studied in
the literature and have shown interesting physical chemistry
properties.20–23 Nevertheless, to our knowledge, it is the first
Paris 6, 4 place Jussieu, boıˆte courrier 840, 75252 Paris Cedex 05,
France
c Laboratoire d’Analyse et Controˆle des Syste`mes Complexes, ECE
´
Paris Ecole d’Ingenieurs, 37 quai de Grenelle, 75015 Paris, France
Universite Paris Descartes, 45 rue des Saint Peres, 75006 Paris,
d
´
`
France
e Laboratoire d’Electrochimie et de Chimie-Physique du Corps Solide,
´
UMR 7177 CNRS/Universite de Strasbourg, 4 rue Blaise Pascal,
CS 90032, 67081 Strasbourg Cedex, France
c
2534 New J. Chem., 2011, 35, 2534–2543
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Scheme 1 Electrosynthesis scheme of the two dimers studied, 4-H2–Zn and 3-H2–Zn.
example of porphyrin dimers with such a spacer. A pyridinium-
type spacer has the advantage of allowing a short distance
between the two macrocycles. Moreover, this spacer could also
permit interesting electronic coupling between the two
porphyrins.
Table 1 Electrochemical data for ZnOEP, H2T3P-4-Py, H2T3P-3-Py,
4-H2–Zn and 3-H2–Zn
Oxidation
Ring
Reduction
Py+
Ring
Compounds
ZnOEP
Thus, the present work reports on the synthesis of the first
two dimers with a pyridinium spacer (abbreviated 4-H2–Zn
and 3-H2–Zn, Scheme 1). These dimers have been obtained
from the zinc b-octaethylporphyrin (ZnOEP) and two isomers
of the 5,10,15-tritolyl-20-pyridylporphyrin, where the pyridyl
group is either in the para position (H2T3P-4-Py) or in the meta
position (H2T3P-3-Py), respectively. Like this, the angle
between the two macrocycles can be modulated.
1.04 0.65
H2T3P-4-Py 1.30 1.06
H2T3P-3-Py 1.33 1.04
À1.59
À1.14 À1.43
À1.16 À1.49
À1.17 À1.49 À1.69
À1.26 À1.59
4-H2–Zn
3-H2–Zn
1.59 1.24 0.95 À0:75
À0:97irr
1.47 1.22 0.90
All potentials in V vs. SCE were obtained from cyclic voltammetry in
1,2-C2H4Cl2 containing 0.1 mol LÀ1 NBu4PF6 at 100 mV sÀ1 (working
electrode: glassy carbon). Bold characters are used for potential values
observed for ZnOEP; italic characters are used for potential values
observed for H2T3P-x-Py (x = 3 or 4) and underlined characters are
used for pyridinium signals. irr: irreversible.
Results and discussion
Electrosynthesis
It is well known that the oxidation of the p-ring of the
porphyrins proceeds via two one-electron steps generating
radical cations and dications.2,24,25
in the absence of electroactive compounds) after ca. 24 h for
both the synthesized dimers. At the end of the electrolysis, the
number of electrons transferred was 2.1 and 2.0 per molecule
of ZnOEP, for 4-H2–Zn and 3-H2–Zn respectively.
Prior to the electrosynthesis of the two dimers, the electro-
chemical behavior of the three used porphyrins has been
investigated by cyclic voltammetry in 1,2-C2H4Cl2. The first
oxidation step of H2T3P-4-Py and H2T3P-3-Py is observed at
1.06 and 1.04 V/SCE, respectively, while for ZnOEP, this first
oxidation step is observed at a lower potential (0.65 V/SCE,
Table 1). Consequently, an exhaustive electrooxidation at
0.75 V/SCE of a mixture of ZnOEP and one of the two pyridyl
porphyrins (H2T3P-4-Py or H2T3P-3-Py) generates the radical
cation of ZnOEP (ZnOEPꢀ+) without oxidation of the other
macrocycle. Thus, the pyridyl group of this porphyrin can
react as a nucleophile with an electrochemically generated
ZnOEPꢀ+, leading to the formation of a dimer.
According to our precedent works,19 an ECNECB mechanism
can be proposed to explain the formation of these dimers.
+
In the first step, the radical cation ZnOEPꢀ is electro-
chemically obtained by oxidation of the ZnOEP macrocycle
(Scheme 2, step E1). Then, the pyridyl group of H2T3P-x-Py
(x = 3 or 4) can carry out a nucleophilic attack on this radical
cation (Scheme 2, step CN), leading to the formation of a
dimer for which the ZnOEP macrocycle leads to an
isoporphyrin. This isoporphyrin corresponds to a porphyrin
for which the aromaticity is broken, because of the simul-
taneous presence of a pyridinium group and a proton on the
same meso carbon. Afterwards, this isoporphyrin can be
oxidized again (Scheme 2, step E2). Finally, this spare proton
is lost, leading to the rearomaticity of the ZnOEP macrocycle
During exhaustive electrolysis, the current decreases expo-
nentially with time, and reaches the residual current (measured
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2534–2543 2535

View Article Online
Scheme 2 Mechanistic scheme for the electrosynthesis of the 4-H2–Zn dimer. A similar mechanism can also be proposed for the formation of
3-H2–Zn.
(Scheme 2, step CB). Thus the global reaction can be sum-
marised like this:
HR-MS spectra
The dimers 4-H2–Zn and 3-H2–Zn have been characterized by
HR-MS mass spectroscopy. The mass spectrum of the mole-
cular ion region corresponds to the theoretical isotopic pattern
for [M]+ as shown in Fig. 1 in the case of 3-H2–Zn. The mass
spectrum gives [M]+ at m/z 1252.5666. This supports the
ZnOEP + H2T3P-x-Py
- H2T3P-x-Py+–ZnOEP + 2eÀ + H+ (x = 3 or 4)
After electrolysis for 24 h at 0.75 V/SCE, the initial red
solution turned brown. The purified H2T3P-x-Py+–ZnOEP
(abbreviated x-H2–Zn, x = 3 and 4) are obtained with a yield
of 87 and 85%, respectively.
It is to be noted that the applied potential (0.75 V/SCE) is
too low to permit multi-substitutions of H2T3P-x-Py (x = 3 or
4) on the ZnOEP macrocycle. Indeed, the first oxidation
potential of the mono-substituted ZnOEP is higher than that
for the non-substituted ZnOEP, because of the electron-with-
drawing effect of the pyridinium group on the ZnOEP
macrocycle (see below).
Of course, we have also tried to synthesize the dimer H2T3P-
2-Py+–ZnOEP from the H2T3P-2-Py macrocycle bearing the
pyridyl group in the ortho position. Nevertheless, a mixture
of several products has been obtained, with a small yield
besides. The poor reactivity can be explained by an important
steric hindrance. Moreover, this steric hindrance can also
lead to a drastic decrease of the rate of the nucleophilic attack
of the pyridyl group onto the electrogenerated radical
cation of the ZnOEP. Nevertheless, the radical cation
+
ZnOEPꢀ is too reactive to remain stable for a long time,
and consequently, can react further with small traces of water
present in the organic solvent,26 leading to degradation of
products.
Fig. 1 Obtained (top) and predicted (bottom) HR-MS data for
3-H2–Zn.
c
2536 New J. Chem., 2011, 35, 2534–2543
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
notion that compound 3-H2–Zn is indeed prepared as m/z
calculated for the dimer, [ZnC82H78N9+] (1252.5688). Similar
data are obtained for 4-H2–Zn.
1H NMR spectra
1H NMR spectroscopy of the two dimers has been carried out
in (CD3)2CO. Complete NMR data are reported in the Experi-
mental section. For each dimer, the observation of two singlets
in the range of 10.30–10.55 ppm, which are integrated with a ratio
equal to 1 : 2 and which are attributed to the meso-H protons,
is consistent with the proposed mono-substitution in a meso
position of the ZnOEP. In the range of 9.50–11.35 ppm, supple-
mentary signals are attributed to the protons of the pyridinium
groups. Indeed, two doublets (with integrations of two protons
for each of them) are present in the case of the compound
4-H2–Zn, whereas four signals (with integrations of one
proton for each) are observed for the dimer 3-H2–Zn. In the range
of 8.90–9.45 ppm are present three signals (two doublets and
one singlet) attributed to the b-H protons of the H2T3P-4-Py or
H2T3P-3-Py porphyrin unit. The singlet is assigned to the four
b-H protons farthest from the pyridinium group and the two
doublets, more deshielded, are assigned to the four b-H
protons closest to the pyridinium group. Moreover, in the aromatic
part, the signals assigned to the protons corresponding to the
tolyl groups of the H2T3P-4-Py or H2T3P-3-Py porphyrin are
observed in the range of 7.60–8.30 ppm. Concerning the ethyl
protons of ZnOEP units and the methyl protons of the tolyl
groups of the other porphyrin, they are observed in the shielded
part, between 1 and 4.5 ppm. The methyl protons of the tolyl
are present at ca. 2.66–2.67 ppm. Finally, the inner N–H
protons of the H2T3P-4-Py or H2T3P-3-Py porphyrin appear
at ca. À2.5 ppm as a singlet. This singlet suggests that N–H
tautomerization is rapid in comparison to the NMR time scale.27
Fig. 2 UV-vis absorption spectra of H2T3P-4-Py and H2T3P-3-Py
(c = 5 Â 10À6 mol LÀ1, dashed line), ZnOEP (c = 5 Â 10À6 mol LÀ1
,
dash-dotted line) and (A) 4-H2–Zn (c = 5 Â 10À6 mol LÀ1, full line) and
(B) 3-H2–Zn (c = 5 Â 10À6 mol LÀ1, full line) in 1,2-C2H4Cl2
(1 cm path length cell). Inset: zoom of the Q bands.
UV-vis absorption and fluorescence spectra
The UV-vis absorption spectra of the dimers (Fig. 2A and B)
do not correspond to the mere addition of the spectra of the
corresponding individual porphyrins.
Indeed, the Soret band of each dimer appears split, due to
the presence of the two different porphyrins, but a red-shift of
the Soret and Q bands (ca. 5 nm) is also observed compared to
that for the porphyrins alone, in particular for the ZnOEP.
This red-shift can be explained by an electron-withdrawing
effect of the pyridinium spacer.22,23 This splitting is accompanied
by a broadening of the absorption bands, which suggests also
excitonic interactions between the two porphyrin rings.19 In
the case of 4-H2–Zn, a broad band around 440–480 nm is
observed. This band may characterize a partial delocalization
of the positive charge of the pyridinium onto the ZnOEP
macrocycle leading to the red-shift. Such an additional band is
not due to the partial protonation of the free base macrocycle
of the dimer (possible presence of an acid trace in 1,2-C2H4Cl2
due to the degradation of the solvent under light illumination).
Indeed, no change is observed after the addition of an excess
of base such as 2,6-lutidine. Such a delocalisation is less or not
observed in the case of 3-H2–Zn.
Fig. 3 Fluorescence spectra of 4-H2–Zn (full line), H2T3P-4-Py (dashed
line) and ZnOEP (dash-dotted line) in 1,2-C2H4Cl2 (excitation wavelength:
530 nm). A similar behaviour is observed by comparison of 3-H2–Zn with
H2T3P-3-Py and ZnOEP.
the dimers. This result can be ascribed to rapid electron
transfer from the excited porphyrins to the pyridinium
acceptor.23,28
Concerning the fluorescence spectra (Fig. 3), a total quenching
of the luminescence is observed for both porphyrins within
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2534–2543 2537

View Article Online
Electrochemistry
The electrochemical behaviour of free base porphyrins and
metalloporphyrins involving nonelectroactive metals is well
documented.2 In general, in the anodic domain, the oxidation
of the p-ring proceeds via two one-electron steps generating
radical cations and dications. Similarly, in the cathodic domain,
two one-electron steps yield radical anions and dianions.
The potential values of these redox processes attributed to the
three porphyrins and the two dimers are gathered in Table 1.
Fig. 4 illustrates the CV curves recorded for the 3-H2–Zn
dimer. In the anodic part, three successive quasi reversible
processes can be observed at 0.90, 1.22 and 1.47 V/SCE
(Table 1 and Fig. 4). The first one could be attributed to the
first oxidation of the ZnOEP ring, the second one both to the
first oxidation of the H2T3P-3-Py ring and the second
oxidation of the ZnOEP ring, and the last one to the second
oxidation of the H2T3P-3-Py ring. Compared to the potential
values recorded for the corresponding unsubstituted monomers
of porphyrins, these oxidation potentials are shifted about
150–250 mV. These shifts may be ascribed to the electron-
withdrawing effect of the positively charged pyridinium on the
p systems of the porphyrins.19,22,29
Fig. 5 Cathodic wave corresponding to the reduction of the Py+ spacer
for 4-H2–Zn (full line) and 3-H2–Zn (dashed line) in 1,2-C2H4Cl2
containing 0.1 mol LÀ1 NBu4PF6 at 100 mV sÀ1 (working electrode:
glassy carbon).
and Fig. 4). The first wave is quasi irreversible while the two
last waves are reversible. The first quasi irreversible process at
À0.97 V/SCE could be assigned to the reduction of the
pyridinium spacer whereas the second and the third processes
could be attributed to the reduction of the macrocycles
(reduction of the H2T3P-3-Py ring at À1.26 V and simul-
taneous reduction of the H2T3P-3-Py and ZnOEP rings at
À1.59 V). In contrast to the oxidation waves, the reduction
processes are not significantly shifted compared to the unsub-
stituted porphyrins. This can be explained by the fact that the
pyridinium spacer is reduced in a first step, leading to a neutral
dimeric compound.
Concerning the cathodic part, three successive processes can
be also observed at À0.97, À1.26 and À1.59 V/SCE (Table 1
Similar behaviours are observed for the 4-H2–Zn dimer
(Table 1), except concerning the reduction process of the
pyridinium spacer (Fig. 5). Indeed, while this electronic trans-
fer was quasi irreversible in the case of 3-H2–Zn, a reversible
wave is observed at À0.75 V/SCE for 4-H2–Zn. Consequently,
the pyridyl radical generated from the reduction of the
pyridinium group appears more stable for the 4-H2–Zn dimer
than for the other dimer.
We can also note an important change in the potential
values for the pyridinium reduction for these two dimers.
Indeed, the pyridinium of the 4-H2–Zn compound is more
easily reduced (À0.75 V/SCE) than that of the 3-H2–Zn
compound (À0.97 V/SCE). This observation might be also
tentatively explained by a most important increase in stability
of the radical for the 4-H2–Zn dimer, this increase of stability
being in favour of an easier reduction.
Spectroelectrochemistry
Spectroelectrochemical studies have been carried out on the
first reduction and the first and second oxidations under the
same experimental conditions as the electrochemical measure-
ments (i.e. in 1,2-C2H4Cl2 containing 0.1 mol LÀ1 NBu4PF6).
The optical absorption spectra of the systems studied can be
used as an additional support for the assignments of the
observed redox steps.
Fig. 4 CVs for 3-H2–Zn (c = 2.5 Â 10À4 mol LÀ1) in 1,2-C2H4Cl2
containing 0.1 mol LÀ1 NBu4PF6 at 100 mV sÀ1 (working electrode:
glassy carbon). (A) First cathodic scan. (B) First anodic scan.
For 4-H2–Zn, the spectral changes recorded during electro-
lysis at the first reduction wave at À0.75 V are illustrated
c
2538 New J. Chem., 2011, 35, 2534–2543
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
onto the ZnOEP macrocycle does not still occur. Examination
of the UV–vis spectrum recorded after the first reduction step
involving the pyridinium spacer in the case of 3-H2–Zn
shows also a red-shift for the Soret band (ca. 11 nm). However,
a blue-shift attributed to the loss of the pyridinium (loss of the
electron-withdrawing effect during the reduction of the spacer)
is usually observed. Nevertheless, such an unusual red-shift was
already noticed for the pyridinium reduction of H2TPP(Py)+,
ZnTPP(Py)+, ZnOEP(Py)+ and ZnOEP(Py)44+ 22,32
This red-
.
shift during electrolysis at the pyridinium reduction can be
tentatively explained by a fast protonation of the pyridyl
radical (obtained after the reduction of the pyridinium) which
would maintain the overall positive charge of the molecule.
Such a protonation is a common reaction with basic radicals
even in nonaqueous media, and has been already described in
the case of pyridinium reductions.30,31,33–35
For 4-H2–Zn, the spectral changes recorded during electro-
lysis at the first and second oxidation waves at 0.95 V and
1.24 V are illustrated in Fig. 6B and C. The first oxidation
reveals a slight decrease in the intensity of the Soret band,
which appears less important than that usually observed in the
generation of porphyrin cation radicals. This can be tentatively
explained by the decrease of the excitonic interactions between
the two macrocycles during the oxidation of the metallo-
porphyrin subunit. Thus, the usual decrease of the intensity
of the Soret band of ZnOEPꢀ+ can be compensated by an increase
of the intensity of the free base. For 3-H2–Zn similar behavior
has been obtained.
Examination of the UV–vis spectrum recorded after the second
oxidation step shows a new decrease in the intensity of the Soret
band accompanied by an increase in the intensity of a new band
at 470 nm and 451 nm for 4-H2–Zn and 3-H2–Zn, respectively.
This new Soret band is in agreement with the formation of
diprotonated H4T3P2+-Py macrocycles, respectively, 4-H24+–Zn
and 3-H24+–Zn. Similarly, the intensity of the initial Q bands
decreases with an increase in new Q band(s) at 673 nm and
712 nm for 4-H24+–Zn and 670 nm for 3-H24+–Zn, typical also
of the formation of diprotonated species. Apparently, this
band has already emerged during the first oxidation because of
the applied potential used (close to the beginning of the second
wave). These results indicate that the second oxidation wave of
the dimer corresponds to the oxidation of the free base. This is
in agreement with our previous assignments, since from the
voltammetric measurements, we have attributed this second
wave simultaneously to the second oxidation of ZnOEP and
the first oxidation of the free base. Indeed, in contrast to
metalloporphyrins with non-electroactive metal centres, the p
radical cations of porphyrin free bases are very reactive and can
further react with the solvent or the supporting electrolyte via
proton abstraction to give the diprotonated porphyrin, as
already reported by Guilard et al.36 According to these authors,
a radical reaction occurs, which can be written as follows:
Fig. 6 Evolution of the UV-vis absorption spectrum of 4-H2–Zn in
1,2-C2H4Cl2 containing 0.1 mol LÀ1 NBu4PF6 recorded during
(A) reduction at À0.74 V, (B) oxidation at 0.95 V and (C) oxidation
at 1.24 V vs. Ag|AgCl.
in Fig. 6A. From our voltammetric measurements, this potential
has been assigned to the reduction step of the pyridinium
spacer. The reduction product exhibits a typical porphyrin
spectrum profile with a red-shift of the Soret band by ca. 14 nm
when compared to the parent compound. This result indicates
that the electrons are not transferred to the porphyrin macro-
cycle but to the pyridinium cation and confirms our primary
assignment. The initial broad band around 440–480 nm disappears
during the reduction of the spacer showing that the partial
delocalization of the positive charge of the pyridinium
H2T3P-x-Py - H2T3Pꢀ+-x-Py + eÀ
2H2T3Pꢀ+-x-Py + 2S–H
- H4T3P2+-x-Py + H2T3P-x-Py + 2Sꢀ
where the substrate S–H can represent the solvent or the
supporting electrolyte.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2534–2543 2539

View Article Online
Formation of a ‘‘tetramer’’ resulting from the coupling between
two reduced 3-H2–Zn dimers
whose color becomes brown. When the deposition time (tD)
increases, the intensity of this peak becomes also more
important, showing a greater amount of the ‘‘tetramer’’ of
porphyrins formed. Moreover, this peak shifts anodically
when the deposition time increases. This can be tentatively
explained by the formation of aggregates on the electrodes
with strong p–p interactions between either the macrocycles or
the dipyridyl spacers. In such a case, the anodic peak shift for
the oxidation of the dipyridyl spacer, with the deposition time,
can be explained by a greater stabilization of these dipyridyl
spacers due to their aggregation by p-stacking. Indeed, the
more the aggregates are important, the more the dipyridyl
spacers are stabilized by p-stacking, and consequently the
more they must be harder to oxidize. Nevertheless, this first
oxidation peak assigned to the oxidation of the dipyridyl
spacers leads to their breakage. Consequently, the initial
3-H2–Zn dimers with pyridinium spacers are again obtained.
Thus, this oxidation leads to the disorganisation and
disappearance of the aggregates, and consequently, further
peaks are not affected (no shift observed).
As already reported below, the pyridyl radical appears less stable
in the case of 3-H2–Zn, so it must be more reactive. Thus we can
suggest a coupling in the 4-position of two pyridyl radicals
yielding a ‘‘tetramer’’ of porphyrins (Scheme 3A). Indeed, the
dimerization of pyridyl radicals when the 2-, 4- or 6-positions are
free has been often discussed in the literature.22,23,30–32 This
dimerization can be explained by the delocalisation possibilities
of the radical onto the pyridyl group, which can take place only
in positions 2, 4 and 6. Regarding the 4-H2–Zn compound, the
4-position of the pyridyl radical is not free. Moreover the 2- and
6-positions seem to be too encumbered to allow a radical
coupling dimerization. Furthermore, the presence of the tritolyl-
porphyrin in the 4-position allows supplementary delocali-
sations of the radical onto this macrocycle (Scheme 3B),
increasing even more the stability of this radical. Concerning
the 3-H2–Zn dimer, the 4-position is free, and consequently,
the pyridyl radical is able to dimerize allowing the formation
of a ‘‘tetramer’’ of porphyrins.
The morphology of the deposits has also been scrutinized by
atomic force microscopy (Fig. 7C): whatever the deposition
time, the deposits appear in the form of tightly packed coils
(diameter around 50–80 nm) which seem agglomerated in a
binder. This result is also in favour of the formation of
aggregates where p–p interactions lead to the formation
of coils.
To support this hypothesis of reactivity of the pyridyl radical
in the case of the 3-H2–Zn dimer, we have carried out a steady
state polarization at a potential equal to À1.1 V/SCE in
CH3CN, in order to reduce the pyridinium spacer. After that,
further sweep reveals a new well defined oxidation peak
(Fig. 7A) characteristic of the oxidation of the product result-
ing from the coupling (i.e. oxidation of the dipyridyl spacer22
of the ‘‘tetramer’’ of porphyrins) which coats the working
electrode (the ‘‘tetramer’’ being not very soluble in CH3CN)
A similar experiment has been performed with 4-H2–Zn. But
in this case, no new anodic peak appears in the CV curve after
Scheme 3 (A) Electrochemical coupling mechanism proposed upon reduction of pyridinium for the 3-H2–Zn compound. The 3D structure of the
tetramer has been obtained by the MM2 force field method. Tentative representation of the formation of aggregates of the tetramer on the
electrodes via strong p–p interactions. (B) Delocalization possibilities of the radical onto the pyridyl group in the case of the 4-H2–Zn compound.
c
2540 New J. Chem., 2011, 35, 2534–2543
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 7 (A) CVs of the films obtained from 3-H2–Zn after different deposition times (tD) at an applied potential equal to À1.1 V/SCE (dashed line:
CV of 3-H2–Zn in CH3CN containing 0.1 mol LÀ1 NBu4PF6; working electrode: glassy carbon); (*) a cathodic redissolution process which ensues
from the reduction of the insoluble ‘‘tetramers’’ at the surface of the working electrode. Inset: Epaapp = f(tD) for the wave corresponding to the
oxidation of the ‘‘tetramer’’. (B) UV-vis absorption spectra of the films deposited onto an ITO electrode after different tD (dashed line: UV-vis
absorption spectrum of 3-H2–Zn in 1,2-C2H4Cl2). Inset: Abs = f(tD) at 426 nm. (C) Atomic force micrograph of a coated ITO electrode obtained
after a deposition time equal to 60 s.
the electrolysis allowing the reduction of the pyridinium spacer.
This is in good agreement with the expected absence of the
coupling process in the case of this dimer.
electron-withdrawing character of the pyridinium spacer
induces an anodic shift of the oxidation potentials of the
two porphyrins. Moreover, pyridinium spacers are also
electroactive. Indeed, a new wave is observed in the cathodic
part for these two dimers. Regarding the 4-H2–Zn dimer, this
wave is reversible while in the case of the 3-H2–Zn dimer, it is
quasi irreversible. In addition, this pyridinium spacer is more
easily reduced in the case of the 4-H2–Zn compound. To
explain these observations, we have invoked the different
delocalisation possibilities of the radical coming from the
pyridinium reduction, leading to a better stabilisation of this
radical in the case of the 4-H2–Zn dimer. On the contrary, in
the case of the 3-H2–Zn dimer, this radical appears more reactive.
Formation of a tetramer has been mentioned, which has been
studied in more detail after obtainment by electrolysis.
In order to try to characterize the tetramer of porphyrins
resulting from the coupling of two 3-H2–Zn dimers, similar
polarizations at À1.1 V/SCE have been performed on trans-
parent ITO electrodes where the UV-visible absorption spectra
(Fig. 7B) can be obtained. A red-shift of ca. 20 nm is observed
for the Soret band of the tetramer of porphyrins compared to
that of the dimer 3-H2–Zn. This red-shift is rather surprising
since the tetramer of porphyrins results from a coupling of the
reduced electron-withdrawing pyridinium groups. Consequently,
a blue-shift was expected. The red-shift can be tentatively
explained by the existence of important interactions between
the macrocycles within the deposits on the electrode, in
agreement with the formation of aggregates. Indeed, J-type
aggregates between macrocycles could be present when a
red-shift is observed, according to the report of Kasha.37
Experimental section
Chemicals
All solvents were of reagent grade quality and used without
further purification.
Conclusion
The zinc-b-octaethylporphyrin (ZnOEP) was purchased
from Sigma-Aldrich. The 5,10,15-tritolyl-20-(4-pyridyl)por-
phyrin (H2T3P-4-Py) and the 5,10,15-tritolyl-20-(3-pyridyl)-
Two porphyrin dimers with pyridinium as bridging spacer
have been obtained by controlled potential electrolysis. This
method is based on a nucleophilic substitution of a porphyrin
substituted by a pyridyl group (H2T3P-x-Py, x = 3 or 4) on an
electrogenerated radical cation of another porphyrin
(ZnOEP). In order to study the influence of the angle between
the two macrocycles, two isomers of the H2T3P-x-Py have
been used, where the pyridyl group was either in the para
position (H2T3P-4-Py) or in the meta position (H2T3P-3-Py).
To our knowledge, they are the two first examples of
porphyrin dimers with pyridinium spacers.
porphyrin (H2T3P-3-Py) were prepared according to
modified literature method.38
a
Electrochemistry
All electrochemical measurements were carried out under argon
with a standard three-electrode system using a PARSTAT
2273 potentiostat. The working electrode was a glassy carbon
disk electrode (d = 3 mm). A platinum wire was used as an
auxiliary electrode. The reference electrode was a saturated calomel
electrode (SCE) that was electrically connected to the studied
solution by a junction bridge filled with the corresponding
solvent-supporting electrolyte solution.
UV-vis absorption spectra of these two synthesized dimers
have shown the presence of excitonic interactions between the
porphyrin subunits. Concerning the fluorescence spectra, a
total quenching of the luminescence has been observed.
The electrochemical study of these two dimers has been
carried out by cyclic voltammetry. The dimers exhibit
an electrochemical behaviour that is close to that of the
individual porphyrins in the cathodic region. Nevertheless, the
Electrosynthesis was performed using a large platinum wire
as a working electrode. The anodic and cathodic compartments
were separated by a fritted glass disk to prevent diffusion of
the electrogenerated species.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2534–2543 2541

View Article Online
A one sided indium-tin-oxide (ITO, Aldrich, 8–12 O per
square) electrode with a surface of about 1 cm2 was used to
prepare tetramer porphyrin films with various deposition
times (tD).
8.26 (4H, d, tolyl), 8.18 (2H, d, tolyl), 7.77 (4H, d, tolyl), 7.69
(2H, d, tolyl), 4.24 (4H, q, 3J = 7.5 Hz, 2CH2 of -CH2-CH3),
4.14 (8H, q, 3J = 7.5 Hz, 4CH2 of -CH2-CH3), 3.22 (4H, q, 3J =
7.5 Hz, 2CH2 of -CH2-CH3), 2.66 (9H, s, CH3 of tolyl), 1.87
3
3
(12H, t, J = 7.5 Hz, 4CH3 of -CH2-CH3), 1.81 (6H, t, J =
7.5 Hz, 2CH3 of -CH2-CH3), 1.24 (6H, t, 3J = 7.4 Hz, 2CH3 of
-CH2-CH3 adjacent to Py+), and À2.48 (2H, s, NH). HR-MS
m/z: 1252.5677; calculated for [ZnC82H78N9+]: 1252.5688.
Molecular mass: [ZnC82H78N9+] (M = 1254.96 g molÀ1).
3-H2–Zn: this second dimer was obtained by the same
procedure using H2T3P-3-Py in the place of H2T3P-4-Py. Then
Spectroscopic measurements
UV-visible spectra were recorded with a Hewlett-Packard
HP8453 spectrophotometer. Steady-state luminescence emission
spectra were obtained using
spectrofluorimeter.
a Spex fluorolog 1681
3-H2–Zn was obtained with
a yield of 85% (19.7 mg,
Spectroelectrochemistry
1.41 Â 10À5 mol).
Spectroelectrochemical measurements were carried out using a
potentiostat/galvanostat (EI30 M Brucker) and a diode array
spectrophotometer (Hewlett-Packard 8453A). A homemade
borosilicate glass cell with an optical pathway of 0.1 mm was
used. The working electrode was a platinum grid (1000 mesh)
placed in the optical pathway, the auxiliary electrode was a
platinum wire, and the reference electrode was an aqueous
Ag|AgCl electrode.
UV-vis lmax(1,2-C2H4Cl2)/nm 407 (e/dm3 molÀ1 cmÀ1
137 600), 535 (13 200), 576 (13 000) and 657 (3000). NMR
dH(300 MHz; (CD3)2CO; Me4Si; 25 1C) 11.35 (1H, d, Py+),
10.99 (1H, s, Py+), 10.44 (1H, d, Py+), 10.42 (2H, s, meso-H),
10.31 (1H, s, meso-H), 9.52 (1H, dd, Py+), 9.44 (2H, d, b-H),
9.16 (2H, d, b-H), 8.90 (4H, s, b-H), 8.17 (4H, m, tolyl), 8.04
(2H, m, tolyl), 7.74 (4H, m, tolyl), 7.65 (2H, m, tolyl), 4.17
3
3
(4H, q, J = 7.5 Hz, 2CH2 of -CH2-CH3), 4.11 (8H, q, J =
3
7.5 Hz, 4CH2 of -CH2-CH3), 3.23 (4H, q, J = 7.5 Hz, 2CH2
Atomic force microscopy
3
of -CH2-CH3), 2.67 (9H, s, CH3 of tolyl), 1.89 (12H, t, J =
7.5 Hz, 4CH3 of -CH2-CH3), 1.83 (6H, t, 3J = 7.5 Hz, 2CH3 of
-CH2-CH3), 1.19 (6H, t, 3J = 7.4 Hz, 2CH3 of -CH2-CH3
adjacent to Py+), and À2.74 (2H, s, NH). HR-MS m/z:
1252.5666; calculated for [ZnC82H78N9+]: 1252.5688. Mole-
cular mass: [ZnC82H78N9+] (M = 1254.96 g molÀ1).
AFM was performed directly on the surface of the ITO
electrode using a Dimension 3100 (Veeco) in the tapping mode
under ambient conditions. Silicon cantilevers (Veeco probes)
with a spring constant of 300 N mÀ1 and a resonance
frequency in the range of 120–139 kHz were used. The
scanning rate was 1.0 Hz.
Acknowledgements
Electrosynthesis
4-H2–Zn: 10 mg of ZnOEP (1.67 Â 10À5 mol) and 11 mg of
H2T3P-4-Py (1.67 Â 10À5 mol) were dissolved in 100 mL of a
1,2-C2H4Cl2/CH3CN (3 : 1) solution containing 0.1 mol LÀ1
tetraethylammonium hexafluorophosphate. 10 mL of 2,6-lutidine
were added to the solution. This compound acts as a base with
the protons released during the synthesis. Then an electrolysis
was carried out for 24 h at 0.75 V vs. SCE. Indeed, the
electrolysis was stopped when the current value had reached
the residual current value. During this anodic oxidation, the
electrolyzed solution was continuously stirred and maintained
under argon. After electrolysis, the initial red solution had
turned brown, and the number of electrons per molecule of
ZnOEP transferred was found to be equal to 2.1 electrons.
Then the solvents were removed, and the residue was dissolved
in 100 mL of 1,2-C2H4Cl2. This organic solution was washed
ten times with water in order to remove the supporting
electrolyte. The organic layer was concentrated (B5 mL)
and chromatographed on silica. The first and the second
fractions (eluted with CH2Cl2) were the unreacted ZnOEP
and H2T3P-4-Py, respectively. The third fraction (eluted with a
mixture of CH2Cl2/CH3OH (99.75 : 0.25)) was the dimer
4-H2–Zn. The solvent was removed and 4-H2–Zn was
obtained (20.2 mg, 87%).
This work was supported by the CNRS, the Universite
Paris-Sud (Paris 11, Orsay), the Universite de Strasbourg,
the Universite Paris Descartes (Paris 5), the Universite
et Marie Curie (Paris 6) and the ECE Paris Ecole d’Inge
´
´
´
´
Pierre
´
nieurs.
The authors thank also R. Thouvenot and B. Hasenknopf,
Institut Parisien de Chimie Moleculaire, UMR 7201 CNRS-
Universite Paris 6, for the NMR and HR-MS mass
´
´
spectroscopies.
References
1 M. Gouterman, The porphyrins, vol. 3, Academic Press, 1978.
2 K. M. Kadish, K. M. Smith and R. Guilard, The porphyrin
handbook, vol. 8, Academic Press, 2000.
3 L. R. Milgrom, The colours of life: an introduction to the chemistry
of porphyrins and related compounds, Oxford University Press,
1997.
4 S. G. Boxer, Biochim. Biophys. Acta, 1983, 726, 265–292.
5 M. R. Wasielewski, Photochem. Photobiol., 1988, 47, 923–929.
6 D. Gust and T. A. Moore, Science, 1989, 244, 35–41.
7 L. Ruhlmann, A. Nakamura, J. G. Vos and J.-H. Fuhrhop, Inorg.
Chem., 1998, 37, 6052–6059.
8 S. Punidha and M. Ravikanth, Tetrahedron, 2004, 60, 8437–8444.
9 E. Vanover, Y. Huang, L. Xu, M. Newcomb and R. Zhang, Org.
Lett., 2010, 12, 2246–2249.
10 A. Osuka and H. Shimidzu, Angew. Chem., Int. Ed. Engl., 1997, 36,
135–137.
11 R. G. Khoury, L. Jaquinod and K. M. Smith, Chem. Commun.,
1997, 1057–1058.
12 M. G. H. Vicente, L. Jaquinod and K. M. Smith, Chem. Commun.,
1999, 1771–1782.
13 K. F. Cheng, C. M. Drain and K. Grohmann, Inorg. Chem., 2003,
42, 2075–2083.
UV-vis lmax(1,2-C2H4Cl2)/nm 407 (e/dm3 molÀ1 cmÀ1
108 500), 537 (7400), 579 (11 200) and 671 (3900). NMR
dH(300 MHz; (CD3)2CO; Me4Si; 25 1C) 10.87 (2H, d, Py+),
10.52 (2H, s, meso-H), 10.41 (1H, s, meso-H), 9.70 (2H, d,
Py+), 9.38 (2H, d, b-H), 9.30 (2H, d, b-H), 9.00 (4H, s, b-H),
c
2542 New J. Chem., 2011, 35, 2534–2543
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
14 T.-H. Huang, Y.-J. Chen, S.-S. Lo, W.-N. Yen, C.-L. Mai,
M.-C. Kuo and C.-Y. Yeh, Dalton Trans., 2006, 2207–2213.
15 N. N. Sergeeva, A. Scala, M. A. Bakar, G. O’Riordan, J. O’Brien,
G. Grassi and M. O. Senge, J. Org. Chem., 2009, 74, 7140–7147.
16 X.-L. Jiang, H.-L. Zhang and J. Wu, Heterocycles, 2006, 68,
2153–2160.
17 L. Ruhlmann and A. Giraudeau, Eur. J. Inorg. Chem., 2001, 659–668.
18 L. Ruhlmann, M. Gross and A. Giraudeau, Chem.–Eur. J., 2003,
9, 5085–5096.
19 A. Giraudeau, L. Ruhlmann, L. El-Kahef and M. Gross, J. Am.
Chem. Soc., 1996, 118, 2969–2979.
20 K. Kalyanasundaram and M. Neumann-Spallart, J. Phys. Chem.,
1982, 86, 5163–5169.
26 D. Dolphin and R. H. Felton, Acc. Chem. Res., 1974, 7, 26–32.
27 H. Garcıa-Ortega, J. Crusats, M. Feliz and J. M. Ribo, J. Org.
Chem., 2002, 67, 4170–4176.
28 J. Hirota and I. Okura, J. Phys. Chem., 1993, 97, 6867–6870.
29 A. Giraudeau, S. Lobstein, L. Ruhlmann, D. Melamed,
K. M. Barkigia and J. Fajer, J. Porphyrins Phthalocyanines,
2001, 5, 793–797.
30 V. Carelli, F. Liberatore, A. Casini, S. Tortorella, L. Scipione
and B. D. Renzo, New J. Chem., 1998, 22, 999–1004.
31 V. Carelli, F. Liberatore, S. Tortorella, B. D. Rienzo and
L. Scipione, J. Chem. Soc., Perkin Trans. 1, 2002, 542–547.
32 A. Brisach-Wittmeyer, S. Lobstein, M. Gross and A. Giraudeau,
J. Electroanal. Chem., 2005, 576, 129–137.
21 K. Kalyanasundaram, Inorg. Chem., 1984, 23, 2453–2459.
22 D. Schaming, A. Giraudeau, S. Lobstein, R. Farha,
M. Goldmann, J.-P. Gisselbrecht and L. Ruhlmann,
J. Electroanal. Chem., 2010, 635, 20–28.
33 J. Volke, J. Urban and V. Volkeova, Electrochim. Acta, 1992, 37,
2481–2490.
34 J. Volke, L. Dunsch, V. Volkeova, A. Petr and J. Urban, Electro-
chim. Acta, 1997, 42, 1771–1780.
23 N. Karakostas, D. Schaming, S. Sorgues, S. Lobstein,
J.-P. Gisselbrecht, A. Giraudeau, I. Lampre and L. Ruhlmann,
J. Photochem. Photobiol., A, 2010, 213, 52–60.
24 J. Fajer, D. C. Borg, A. Forman, D. Dolphin and R. H. Felton,
J. Am. Chem. Soc., 1970, 92, 3451–3459.
35 J. E. H. Buston, F. Marken and H. L. Anderson, Chem. Commun.,
2001, 1046–1047.
36 C. Inisan, J.-Y. Saillard, R. Guilard, A. Tabard and Y. L. Mest,
New J. Chem., 1998, 22, 823–830.
37 M. Kasha, Radiat. Res., 1963, 20, 55–71.
25 J.-H. Fuhrhop, K. M. Kadish and D. G. Davis, J. Am. Chem. Soc.,
1973, 95, 5140–5147.
38 K. Funatsu, T. Imamura, A. Ichimura and Y. Sasaki, Inorg.
Chem., 1998, 37, 1798–1804.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2534–2543 2543