Modulating the electronic properties of asymmetric push-pull and symmetric Zn(II)-diarylporphyrinates with para substituted phenylethynyl moieties in 5,15 meso positions: A combined electrochemical and spectroscopic investigation

Article abstract of DOI:10.1016/j.electacta.2012.08.039

Push-pull Zn(II)-porphyrinates have recently shown attracting performances as light harvesting systems in dye-sensitized solar cells (DSSCs). To fully exploit their intrinsically high efficiency it is important to finely tune their HOMO and LUMO levels, which can be achieved by proper choice of the push and pull substituents. Of course such target-oriented molecular design requires the availability of reliable relationships between molecular structure and electronic properties; therefore we have carried out a combined electrochemical, spectroscopic and computational investigation on a wide, systematic range of Zn(II)-porphyrinates 5,15 meso substituted with phenylethynyl linkers, including a first symmetric series carrying on the opposite terminals the same substituent (N(CH₃)₂, OCH₃, COOCH₃, COOH, NO₂); and a second push-pull one, with the terminal positions carrying one donor and one acceptor group belonging to the series above. Moreover, two suitably modified porphyrins allowed evaluation of the effects of (i) the presence or absence of the phenyl group in the linker between the porphyrin core and the acceptor group, and (ii) the effect of perfluorination on the same phenyl group. A rationalization scheme is proposed encompassing the whole porphyrin set, affording inter alia interesting clues on the different localization of the redox centres and effective conjugation between the porphyrin core and the side chains as a function of the molecular design.

Full text of DOI:10.1016/j.electacta.2012.08.039

Electrochimica Acta 85 (2012) 509–523
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Modulating the electronic properties of asymmetric push–pull and symmetric
Zn(II)-diarylporphyrinates with para substituted phenylethynyl moieties in 5,15
meso positions: A combined electrochemical and spectroscopic investigation
Patrizia Romana Mussinia,∗,1, Alessio Orbelli Birolib,∗∗, Francesca Tessorec, Maddalena Pizzottic,b
,
Cinzia Biaggic, Gabriele Di Carloc, Maria Grazia Lobellod, Filippo De Angelisd
a Dipartimento di Chimica Fisica ed Elettrochimica, Via Golgi 19, 20133 Milano, Italy
b Istituto di Scienze e Tecnologie Molecolari del CNR (CNR-ISTM), Via C. Golgi 19, 20133 Milano, Italy
c Dipartimento di Chimica Inorganica Metallorganica e Analitica “Lamberto Malatesta” dell’Università degli Studi di Milano, Unità di Ricerca dell’INSTM, Via G. Venezian 21, 20133
Milano, Italy
d Istituto di Scienze e Tecnologie Molecolari del CNR (CNR-ISTM), c/o Dipartimento di Chimica, Via Elce di Sotto 8, 06123 Perugia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 April 2012
Received in revised form 8 August 2012
Accepted 8 August 2012
Available online 17 August 2012
Push–pull Zn(II)-porphyrinates have recently shown attracting performances as light harvesting systems
in dye-sensitized solar cells (DSSCs). To fully exploit their intrinsically high efficiency it is important to
finely tune their HOMO and LUMO levels, which can be achieved by proper choice of the push and
pull substituents. Of course such target-oriented molecular design requires the availability of reliable
relationships between molecular structure and electronic properties; therefore we have carried out a
combined electrochemical, spectroscopic and computational investigation on a wide, systematic range
of Zn(II)-porphyrinates 5,15 meso substituted with phenylethynyl linkers, including a first symmetric
series carrying on the opposite terminals the same substituent ( N(CH3)2, OCH3, COOCH3, COOH,
NO2); and a second push–pull one, with the terminal positions carrying one donor and one acceptor
group belonging to the series above. Moreover, two suitably modified porphyrins allowed evaluation of
the effects of (i) the presence or absence of the phenyl group in the linker between the porphyrin core
and the acceptor group, and (ii) the effect of perfluorination on the same phenyl group. A rationalization
scheme is proposed encompassing the whole porphyrin set, affording inter alia interesting clues on the
different localization of the redox centres and effective conjugation between the porphyrin core and the
side chains as a function of the molecular design.
Keywords:
Push–pull porphyrins
Multiple redox centres
Effective conjugation
HOMO and LUMO tuning
Dye-sensitized solar cells
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
10,20-diphenylporphyrins, carrying in the 5,15 meso positions a
para substituted phenyl ring connected by a triple bond to the
Porphyrins are extremely versatile molecular architectures with
a widespread potential in the field of organic electronics, including
emissive and optoelectronic applications [1–5] and recently also
of organic photovoltaic cells [6–11] on account of their interesting
photophysical properties, tunable through a variety of substituents
which can be linked to the meso or ␤ pyrrolic positions of the
porphyrinic ring. Moreover porphyrins are characterized by good
thermal and chemical stabilities.
porphyrin ring [12,13].
These metal porphyrins act as a push–pull system by introduc-
tion of an electron donor group on one aromatic ring and an electron
acceptor group on the other one, as first reported by Therien and co-
workers [14–16]. The large quadratic hyperpolarizability [12–16]
and the significant two-photon absorption [17,18] of this family of
push–pull metal porphyrins have been attributed mainly to a direc-
tional charge transfer process along the push–pull system, origin of
a quite strong Q absorption band corresponding to the transition
from the donor HOMO to the acceptor LUMO level, as reported for
the first time by some of us by a TDDFT investigation [19]. More-
over, also transitions between energy levels localized mainly on the
porphyrin ring, origin of the very strong B absorption band at about
450 nm, are contributing to the hyperpolarizability, although to a
less extent [19].
Some of us have investigated the synthesis and
second-order non linear optical (NLO) properties of metal
∗
Corresponding author. Tel.: +39 02 50314211; fax: +39 02 50314300.
Corresponding author. Tel.: +39 02 50314397; fax: +39 02 50314405.
E-mail addresses: patrizia.mussini@unimi.it (P.R. Mussini), a.orbelli@istm.cnr.it
∗∗
Starting from the evidence of the presence of various strong
charge transfer processes in this family of push–pull metal
(A. Orbelli Biroli).
1
ISE member.
0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2012.08.039

510
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
R3
R3
1 : R1= R2= p-C6H4NMe2; R3= t-Bu
2 : R1= R2= p-C6H4OMe; R3= t-Bu
3 : R1= R2= p-C6H4COOMe; R3= t-Bu
3': R1= R2= COOEt; R3= t-Bu
N
N
N
N
4 : R1= R2= p-C6H4COOH; R3= t-Bu
Zn
R1
R2
5 : R1= R2= p-C6H4NO2; R3= t-Bu
6 : R1= p-C6H4NMe2; R2= p-C6H4COOH; R3= t-Bu
6': R1= p-C6H4NMe2; R2= p-C6F4COOH; R3= t-Bu
7 : R1= p-C6H4NMe2; R2= p-C6H4NO2; R3= H
8 : R1= p-C6H4OMe; R2= p-C6H4COOH; R3= t-Bu
9 : R1= p-C6H4OMe; R2= p-C6H4NO2; R3= t-Bu
R3
R3
Fig. 1. A synopsis of the investigated Zn(II)-porphyrinates.
porphyrins, interest for their application as dyes in DSSCs (Dye-
Sensitized Solar Cells) has raised in the last years. Several push–pull
Zn(II)-porphyrinates of this family have been synthesized and
investigated, all of them featuring: (a) one or more carboxylic
groups as electron acceptor groups on one aromatic ring, ensur-
ing efficient anchoring on the TiO2 layer, and (b) phenylethynyl
moieties as ␲-spacers between the porphyrin core and the termi-
nal acceptor carboxylic or donor groups. Many of these push–pull
Zn(II)-porphyrinates show attractive performances when acting as
dyes in DSSCs [20–26]. To fully exploit the efficiency of these por-
phyrins as light harvesting dyes it is necessary to match specific
requirements, such as a LUMO energy level higher than that of
the TiO2 conduction band and a HOMO energy −level lower than
the energy of the electrolytic redox couple I−/I3 [7,27–31]. This
can be achieved by tuning the HOMO–LUMO levels in particular by
proper choice of the donor groups of the push–pull system. In order
to produce rational guidelines for this choice, a convenient exper-
imental approach appears to be the electrochemical systematic
characterization of an exhaustive series of Zn(II)-porphyrinates,
either asymmetrically (push–pull) or symmetrically substituted
in 5,15 meso positions affording reliable comparison not only of
HOMO–LUMO energy gaps (which can be also theoretically or
experimentally evaluated by electron absorption spectroscopy) but
also of the origin of the electronic effects, introduced by differ-
ent donor or acceptor groups, acting on the HOMO and LUMO
energy levels, critical for the design of the appropriate porphyrinic
dye.
The aryl substituent in position 10,20 is the 3,5-di-tert-
butylphenyl ring with the exception of 7, where the substituent
is a simple phenyl ring.
2. Experimental
2.1. Synthesis
All reagents and solvents were purchased from Sigma–Aldrich
and used as received except for Et3N and Et2NH (freshly distilled
over KOH) and THF (freshly distilled from Na/benzophenone under
nitrogen atmosphere). Silica gel (Geduran Si 60, 63–200 ␮m) was
purchased from Merck.
Glassware has been flame-dried under vacuum before use when
it was necessary.
Zn(II)-porphyrinates 1, 2, 3, 3ꢀ, 4, 5 and 9 have been synthesized
following the specific procedures reported in Appendix A.
[5,15-Diiodo-10,20-bis(3,5-di-tert-butylphenyl)porphyrinate]-
Zn(II) [32], [5,15-dibromo-10,20-bis(3,5-di-tert-butylphenyl)-
porphyrinate]Zn(II) [33], [5-(4ꢀ-carboxy-phenylethynyl)-15-(4ꢀꢀ-
N,N-dimethylamino-phenylethynyl)-10,20-bis(3,5-di-tert-butyl-
phenyl)porphyrinate]Zn(II) (6) [32], [5-(4ꢀ-carboxy-2ꢀ,3ꢀ,5ꢀ,6ꢀ-
tetrafluorophenylethynyl)-15-(4ꢀꢀ-N,N-dimethylamino-phenyl-
ethynyl)-10,20-bis(3,5-di-tert-butylphenyl)porphyrinate]Zn(II)
(6ꢀ) [32], [5-(4ꢀ-N,N-dimethylamino-phenylethynyl)-15-(4ꢀꢀ-nitro-
phenylethynyl)-10,20-(diphenyl)porphyrinate]Zn(II) (7) [13] and
[5-(4ꢀ-carboxy-phenylethynyl)-15-(4ꢀꢀ-methoxy-phenylethynyl)-
10,20-bis(3,5-di-tert-butylphenyl)porphyrinate]Zn(II) (8) [32]
were prepared following the syntheses reported in literature. All
the compounds have been characterized by 1H and 19F NMR spec-
tra recorded on a Bruker Avance DRX-400 in CDCl3, CDCl3 with a
drop of pyridine-d5 or in THF-d8 (Cambridge Isotope Laboratories,
Inc.) as solvent.
In order to follow this approach we have investigated, by
cyclic voltammetry and electronic absorption spectroscopy with
the support of a TDDFT theoretical study, two series of 10,20-diaryl
Zn(II)-porphyrinates carrying in 5,15 meso positions para substi-
tuted phenylethynyl linkers (Fig. 1) namely:
•
a first series of five symmetric porphyrinic architectures (1–5),
carrying in the 5,15 meso positions the same para substituted
phenylethynyl moiety, where the substituents are N(CH3)2,
OCH3, COOCH3, COOH, NO2, i.e. regularly ranging from a
strong donor to a strong acceptor group;
2.2. Electronic absorption spectra
Electronic absorption spectra were recorded in diluted THF solu-
tions at room temperature by a Jasco V-530 spectrometer. The
“onset” optical energy gaps were evaluated by the edge corre-
sponding to the intersection between the negative tangent line in
the inflection point of the lowest energy absorption band and the
tangent line to a linear portion of the absorption tail.
•
a second one of four asymmetric push–pull porphyrinic architec-
tures (6–9), carrying in the 5,15 meso positions one donor and
one acceptor group belonging to the series above;
finally, two suitably modified porphyrinic architectures in order
•
to evaluate the effect of the absence of the phenyl ring in the
ethynyl linker between the porphyrinic core and the acceptor
COOEt group (3ꢀ), and of the perfluorination of the phenyl ring
carrying the COOH group, thus producing an increased acceptor
effect (6ꢀ).
2.3. Electrochemistry
The voltammetric studies have been performed in a 4 cm3 cell,
on ∼0.0005 M solutions in dimethylformamide (Aldrich, 99.8%)

P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
511
with 0.1 M tetrabutylammonium perchlorate (TBAP, Fluka) as the
3. Results and discussion
supporting electrolyte. The solutions were deaerated by N2 bub-
bling. The ohmic drop has been compensated by the positive
feedback technique [34].
3.1. Syntheses
Deaeration is here more critical than usual, because most
reduction peaks in our series of Zn(II)-porphyrinates fall in close
proximity of oxygen reduction, a process which, moreover, por-
phyrins efficiently catalyse.
The Zn(II)-porphyrinates investigated in this work are reported
in Fig. 1, which includes:
•
the series of symmetric Zn(II)-porphyrinates, 1–5;
•
The experiments were carried out using an AUTOLAB PGSTAT
potentiostat (EcoChemie, The Netherlands) run by a PC with GPES
software. Cyclic voltammetry (CV) investigations were carried out
at scan rates typically ranging 0.05–2 V s−1, with ohmic drop com-
pensation; scans were performed both starting in positive and
negative directions, in order to check possible electrode poisoning
effects by subsequent processes, which however proved not to be
our case (see the detailed CV patterns reported for each porphyrin
in the Supporting Material (Part C), where peaks coincide irre-
spective of the starting direction and width of the potential scan).
Differential pulse voltammetry (DPV) curves (step potential: 5 mV,
modulation amplitude: 50 mV) were also recorded for each com-
pound, as a support to CV peak inspection, enhancing the porphyrin
signals with respect to the background. Moreover, in the absence of
surface poisoning as in the present case, comparison of DPV scans
in positive and negative direction (cyclic DPV, CPDV [35,36]), evi-
dences electron transfer reversibility (the more reversible a given
electron transfer process, the more similar the corresponding peaks
obtained in the positive and negative DPV scans).
The working electrode was a glassy carbon one (AMEL, diam-
eter = 1.5 mm) cleaned by synthetic diamond powder (Aldrich,
diameter = 1 ␮m) on a wet cloth (Struers DP-NAP); the counter
electrode was a platinum disk or wire; the operating reference
electrode was an aqueous saturated calomel electrode, but the
potentials were ultimately referred to the Fc+/Fc couple (the inter-
solvental redox potential reference currently recommended by
IUPAC [37,38]) by both external and internal standardization. To
prevent water and chloride leakage into the working solution a
compartment filled with the operating medium and ending with
a porous frit was interposed between the reference electrode and
the cell.
the series of asymmetric push–pull Zn(II)-porphyrinates, 6–9;
the symmetric Zn(II)-porphyrinate 3ꢀ, to be compared to 3 in
order to evaluate the effect of the absence of the phenyl ring in
the ethynyl linker;
•
the asymmetric, push–pull Zn(II)-porphyrinate 6ꢀ, to be com-
pared to 6 in order to evaluate the effect of perfluorination of
the phenyl ring carrying the COOH group.
•
Syntheses of all the Zn(II)-porphyrinates investigated were
carried out by the Sonogashira coupling [50], starting from the
5,15-diiodo or 5,15-dibromo Zn(II)-porphyrinates. For 6, 6ꢀ, 7 and
8 synthetic procedures previously reported in literature were fol-
lowed [13,32], while for the remaining Zn(II)-porphyrinates the
syntheses are reported in Appendix A. In particular, symmetric
Zn(II) porphyrinates 1, 2, 3, 3ꢀ and 5 were synthesized starting
from the 5,15-diiodo Zn(II)-porphyrinate by a one-step reaction
with the donor or the acceptor phenyl ethynyl moiety, whereas
the push–pull Zn(II)-porphyrinate 9 was synthesized by a two-
step method starting from the 5,15-dibromo Zn(II)-porphyrinate,
by introducing first the donor and then the acceptor phenyl ethynyl
moiety, following the procedure reported by Therien and co-
workers [14]. Finally 4 was obtained by hydrolysis of 3 with LiOH
in a THF/H2O mixture.
3.2. Electronic absorption spectra
The electronic absorption spectra of all the Zn(II)-porphyrinates
investigated, recorded in THF solution and normalized with respect
to the maximum absorption peak, are collected in Fig. 2; details
on the absorption maxima and onset wavelengths are reported in
Table 1.
All spectra feature the two absorption bands, typical of metal
porphyrins, one very strong centred at about 450 nm (Soret or B
band) and one quite strong centred in the 650–700 nm range (Q
band); in some cases a weaker band appears, also as a shoulder,
in this latter range (Fig. 2). The B band corresponds mainly to the
singlet S0→S2 transition, centred on the porphyrin core, while the
Q band corresponds mainly to the singlet S0→S1 charge-transfer
transition between the HOMO–LUMO levels [19].
The increased push–pull character of the asymmetric Zn(II)-
porphyrinates, resulting from increasing strengths of the electron
donor or electron acceptor groups or of both, can be expressed
in terms of Hammett parameters, which follows the sequence
8 < 9 < 6 < 7, with a regular red-shift of both B and Q bands (see
Table 1).
On the contrary, the symmetric Zn(II)-porphyrinates display
nearly constant wavelengths of both B and Q bands with the excep-
tion of 1 and 5 (actually the two opposite limiting cases in the
sequence of the Hammett parameters) (Table 1) featuring, in par-
ticular 1, a remarkable red shift of both bands which could be
tentatively attributed to quadrupolar contributes, [51] because of
the known ambivalent behaviour, both donor or acceptor, of the
porphyrinic ring [52].
2.4. DFT and TDDFT calculations
DFT ground state optimized geometries and energies of dyes
1–5 and 3ꢀ were obtained by using the B3LYP exchange correlation
functional [39] along with a 6-31G* basis set. For sake of simplicity
the tert-butyl substituents of the phenyl rings have been replaced
by hydrogen atoms.
TDDFT excited state calculations were performed on the DFT
optimized geometries at the same B3LYP/6-31G* level of theory,
both in vacuo and in water solution.
Although a TDDFT approach coupled with a GGA exchange cor-
relation functional and relatively large basis set has proven to
provide reliable agreement between predicted and experimentally
observed vertical excitation energies and other spectroscopic prop-
erties in porphyrin derivatives [40–43], some of us have shown
in previous papers [13,19,32] that B3LYP, also adopted by other
researchers [44–46], gives very good results for push–pull 5,15
meso-substituted Zn(II)-porphyrinates; in particular, B3LYP has
been found sufficient to study the charge transfer character of Q
bands by TDDFT calculations [43].
The lowest 10 excitation energies were calculated, spanning
approximately 3.5 eV. All the calculations have been performed by
the Gaussian 03 program package [47]. The nonequilibrium C-PCM
solvation model was employed for TDDFT calculations in solution
[48,49].
Moreover it can be noticed that the increase of the peak widths
of the B band appear to regularly account for the symmetric or
asymmetric character: in the symmetric series the B bands are quite
sharp, albeit widening in the two limiting cases 1 and 5 (strongest
electron donor and acceptor substituents, respectively), while in

512
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
Table 1
Onset and maximum absorption wavelengths of Zn(II)-porphyrinates 1–5, 6–9, 3ꢀ and 6ꢀ, together with the corresponding calculated values of optical energy gaps. Zn(II)-
porphyrinates are reported in increasing order of the sum of Hammett ꢀp parameters [56] for the two substituents in 5,15 meso positions, with the exception of 3ꢀ and 6ꢀ. Eg
values are highlighted in bold character.
#
Substituents
ꢁꢀp
B band
Onset
Q band
Onset
Max
Max
ꢂB/nm
Eg/eV
ꢂB/nm
Eg/eV
ꢂQ/nm
Eg/eV
ꢂQ/nm
Eg/eV
1
2
3
4
5
6
7
8
9
3ꢀ
6ꢀ
p-C6H4NMe2
p-C6H4OMe
p-C6H4COOMe
p-C6H4COOH
p-C6H4NO2
p-C6H4NMe2
p-C6H4NMe2
p-C6H4OMe
p-C6H4OMe
p-COOEt
p-C6H4NMe2
p-C6H4OMe
p-C6H4COOMe
p-C6H4COOH
p-C6H4NO2
p-C6H4COOH
p-C6H4NO2
p-C6H4COOH
p-C6H4NO2
p-COOEt
−1.66
−0.54
0.90
487
469
467
468
481
486
496
470
478
458
494
2.55
2.65
2.66
2.65
2.58
2.55
2.50
2.64
2.60
2.71
2.51
467
451
452
451
458
456
459
451
459
440
454
2.66
2.75
2.75
2.75
2.71
2.72
2.70
2.75
2.70
2.82
2.73
703
675
680
678
697
698
710
677
691
664
707
1.77
1.84
1.83
1.83
1.78
1.78
1.75
1.83
1.80
1.87
1.76
678
658
660
659
669
671
677
659
667
644
673
1.83
1.89
1.88
1.88
1.86
1.85
1.83
1.88
1.86
1.93
1.84
0.90
1.56
−0.38a
−0.12
0.49
0.83
0.90
p-C6H4NMe2
p-C6F4COOH
−0.11a
a
The difference between the two ꢁꢀp entries results from a +0.27 contribution accounting for the perfluorinated C6F5 ring [56].
the asymmetric series the B bands are significantly larger, and their
width regularly increases with increasing push–pull character. It is
worthwhile noticing also that the only spectrum of an asymmetric
Zn(II)-porphyrinate of width comparable to those of the symmetric
series is 8, characterized by a weak push–pull asymmetric charac-
ter.
This trend of the peak widths can be extended also to the Q
band widths, albeit it is less conspicuous. Interestingly the ratio
of the peak intensities of the Q band with respect to the B one,
defined as εQ/εB, shows some specific and peculiar trends namely
(Fig. 3):
0.4
0.3
0.2
0.1
symmetric porphyrins
asymmetric porphyrins
1
2
3
3'
4
5
6
6'
7
8
9
porphyrins entry
Fig. 3. Ratio of the peak intensities of the Q band with respect to the B one, defined
as εQ/εB, for Zn(II)-porphyrinates 1–5, 6–9 and 3ꢀ, 6ꢀ.
9
8
•
the symmetric series shows a nearly constant low ratio εQ/εB of
∼0.15, the only exception being clearly provided by 1 which, as
above stated, could be justified in terms of quadrupolar contri-
bution [51];
7
6’
6
3ꢀ presents the lowest ratio because in the absence of the phenyl
•
ring in the linker the electronic properties of the molecule
approach that of a non substituted Zn(II)-porphyrinates, gener-
ally presenting a very low εQ/εB value (≤0.10)2;
5
•
the asymmetric push–pull series shows a regular increase of the
εQ/εB ratio, at fixed donor group, with increasing strength of the
acceptor one; two different straight lines can be perceived for
D = OCH3 and N(CH3)2, the second one characterized by much
higher values, consistently with the higher donating ability of the
amino group and therefore with a higher push–pull character of
the porphyrinic architecture.
4
3’
3
2
1
400
450
500
550
600
650
700
750
800
2
In our experience we found the same ratio value in the case of the 5,15 diiodo
λ / nm
or dibromo Zn(II)-porphyrinates, but also in the case of the 5-bromo-15-iodo
Zn(II)-porphyrinate. Moreover it has been reported [53] for several octaalkyl metal-
porphyrinates, unsubstituted in all the meso positions, a molar extinction coefficient
10 times higher for the B band than for the Q band in vapor phase absorption spectra.
Fig. 2. A synopsis of the normalized electronic absorption spectra of the Zn(II)-
porphyrinates 1–5, 6–9, 3ꢀ and 6ꢀ. Inset: focusing on the Q bands.

P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
513
(Table B.7 in the Supplementary Material), and on the electron
density on the HOMO and LUMO orbitals (Fig. 5).
In agreement with the trends of the absorption spectra
(Figs.
2 and 3 and Table 1) the energy gap between the
HOMO–LUMO levels is lower for 1 and, although to a lesser extent,
for 5, while it is slightly higher and quite independent from the
nature of the donor or acceptor groups for 2, 3 and 4 (Fig. 4).
Such a picture is clearly in agreement with the higher values of
the ratio εQ/εB reported in Fig. 3, since the electron transfer from
the HOMO to the LUMO level involves a relevant electron density
transfer from the donor ethynyl phenyl moiety carrying the NMe2
strong donor group to the porphyrinic core for 1 (Fig. 5) and a less
significant electron density transfer from the porphyrinic core to
the ethynyl phenyl moiety carrying the nitro strong acceptor group
for 5 (Fig. 5), in accordance with the suggested ambivalent accep-
tor/donor properties of the porphyrin ring [36]. Such transfer of
electron density from or to the porphyrinic core is much less rel-
evant for 2, 3 and 4, having less strong donor or acceptor groups
respectively (see Fig. 5).
For asymmetric Zn(II)-porphyrinates 6 and 7 we have a strong
electron density transfer along the push–pull system mainly when
the donor group is N(CH3)2, due to the large electron density
centred on the ethynyl phenyl moiety carrying the donor group
(HOMO level) while the acceptor LUMO level is centred mainly on
the ethynyl phenyl moiety carrying the carboxylic group [32].
Therefore in the asymmetric architectures the HOMO or LUMO
energy levels do not significantly involve the porphyrinic core,
while for the HOMO (3, 4, 5) and for the LUMO (1, 2) levels of the
symmetric architectures the contribution of the porphyrinic core
is relevant.
Fig. 4. Schematic diagram of the HOMO–LUMO energy levels, calculated in water
solution, for Zn(II)-porphyrinates 1–5 and 3ꢀ.
From electronic absorption spectra “optical” energy gaps Eg have
been calculated by using both either the maximum or the onset
wavelengths (Table 1), according to the equations:
h (Js−1) × c (ms−1
ꢂmax(m) × e (C)
)
Eg,max (eV) =
Eg,onset (eV) =
(maxima criterion)
(onset criterion)
(1a)
(1b)
h (Js−1) × c (ms−1
ꢂonset (m) × e (C)
)
Finally for the symmetric Zn(II)-porphyrinate 3ꢀ, lacking the
phenyl ring in the ethynyl linker, the HOMO–LUMO energy gap
is higher so that the electron density transfer (Fig. 5) from the
porphyrinic ring (HOMO level) to the ethynyl moiety carrying the
COOEt group (LUMO level), is much less relevant in agreement
with the energy and intensity trend of its electronic absorption
spectrum (Table 1 and Fig. 3).
In order to confirm the above suggestions raised by the DFT
investigation we have carried out also a TDDFT computational
investigation, explicitly calculating the systems’ excited states, thus
allowing a complete definition of the electronic origin of the B and
Q bands.
The computed excitation energies and oscillator strengths of
the optical transitions calculated in vacuo and in solution taking
into consideration the lowest 10 excitation states are reported in
Table B.8 in the Supplementary Material.
In particular a quite good agreement between theoretical and
experimental values was found for the Q band, for which the
major contribution of the HOMO–LUMO transition was confirmed,
according to the trend of Table 2.
For the symmetric series 1–5 the energy gap of both the S0→S2 and
S0→S1 transitions evaluated from the maximum wavelength of the
B and in particular of the Q bands respectively, is lower for strong
donor ( NMe2) or strong acceptor ( NO2) groups, suggesting, in
accordance with the ambivalent donor and acceptor properties of
the porphyrinic core [52], a more facile electron transfer to the por-
phyrinic core ( NMe2) or vice versa ( NO2). For the other groups
(
OMe, COOMe, COOH) the energy gap is slightly higher but
quite independent from the donor or acceptor nature of the group,
consistently with a less relevant electron transfer process from or
to the porphyrinic core.
It was already reported [32] that for the asymmetric Zn(II)-
porphyrinates 6 and 8 the energy gap of the Q band, corresponding
to the HOMO–LUMO transition with a significant electron trans-
fer along the push–pull system, is lower when the donor group is
NMe2, just confirming the relevant effect of the strength of the
donor group. Moreover for the symmetric Zn(II)-porphyrinate 3ꢀ
the energy gap of the Q band is higher if compared to 1–5, suggest-
ing a less relevant electron transfer from the porphyrinic core to
the COOEt group, in the absence of an intermediate phenyl ring
in the ethynyl linker.
Finally the fluorination of the aromatic ring carrying the accep-
tor COOH group as in 6ꢀ does not produce a relevant effect on the
energy gap of both the B and Q bands which is quite similar to that
of the non fluorinated Zn(II)-porphyrinate 6 (Table 1).
In order to confirm such analysis of the properties and trends of
the electronic absorption spectra we carried out a DFT theoretical
investigation on the ground state energy level of the symmetric
Zn(II)-porphyrinates 1–5 and 3ꢀ. Such kind of investigation has been
already reported for the asymmetric Zn(II)-porphyrinates 6 and 8
[32] and 7 [19].
For both 1 and 5 the calculated values of ꢂmax and f of the absorp-
tion Q band are higher if compared to the values for 2, 3 and 4, in
full agreement with the experimental trends of Table 1 and Fig. 3.
For 3ꢀ the calculated values of ꢂmax and f of the Q band are at
higher energy and of lower intensity still in agreement with the
experimental findings (Table 1 and Fig. 3), as a consequence of a
lower HOMO–LUMO contribution to the Q band, due to the lack of
the phenyl ring in the ethynyl linker.
Finally the TDDFT calculations have shown that for
the Zn(II)-porphyrinates
close major strong transitions at about 400–410 nm char-
acterizing the band, two close transitions of minor but
1 and 5, in addition to two very
B
still relevant intensity at about 470–490 nm are found
(Figures B.1 and B.6 and Tables B.1 and B.6 in the Supple-
mentary Material), which correspond to the broad absorption
band found in the experimental spectra.
The DFT computational investigation has given informa-
tion on the energy of the HOMO and LUMO levels (Fig. 4),
together with that of other levels of higher or lower energy

514
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
Fig. 5. Electron density plots of the HOMO and LUMO levels of Zn(II)-porphyrinates 1–5 and 3ꢀ calculated in water solution (isodensity value = 0.02).
In the case of 2, 3,
very close major transitions
(Figures B.2, B.3, B.5, B.4 and Tables B.2, B.3, B.5 and B.4 in the Sup-
plementary Material) but transitions at lower energy as in 1 and
5 are lacking. All these transitions mainly involve the electronic
density of the porphyrinic core and excited states of higher energy.
4
and 3ꢀ we still have two
at about 435–445 nm
fundamental review paper [54]:
ELUMO (eV) = −e × [(Eonset,Ic/V (Fc+|Fc) + 4.8 V (Fc+|Fc vs zero)]
(onset criterion)
(2a)
ELUMO (eV) = −e × [(E◦ /V (Fc+|Fc) + 4.8 V (Fc+|Fc vs zero)]
Ic
3.3. Electrochemical characterization
(formal potential criterion)
(2b)
(2c)
(3a)
All Zn(II)-porphyrinates investigated in this work have been
electrochemically characterized by both cyclic voltammetry (CV)
and differential pulse voltammetry (DPV), the latter affording
significant enhancement of the faradaic signal vs capacitive back-
ground ratio, a useful tool considering the moderate solubility
and large size and sterical hindrance, implying low diffusion
coefficients, of the Zn(II)-porphyrinates investigated.
The investigation was carried out in DMF + 0.1 M TBAP, which
ensures sufficient solubility for all the Zn(II)-porphyrinates inves-
tigated together with a satisfactory potential window for both
reduction and oxidation processes.
ELUMO (eV) = −e × [(Ep,Ic/V (Fc+|Fc) + 4.8 V (Fc+|Fc vs zero)]
(maxima criterion)
EHOMO (eV) = −e × [(Ep,Ia/V (Fc+|Fc) + 4.8 V (Fc+|Fc vs zero)]
(maxima criterion)
ꢀ
EHOMO (eV) = −e × [(EI◦a/V (Fc+|Fc) + 4.8 V (Fc+|Fc vs zero)]
Synopses of CV and DPV patterns are provided in Fig. 6 (sym-
metric series 1–5), 7 (3 vs 3ꢀ) and 8 (asymmetric push–pull series
6–9 and 6ꢀ). Relevant oxidation and reduction peaks are accounted
for in terms of (a) onset potential, (b) formal potential, and (c) peak
(maximum) potential (Table 3), corresponding to the three cur-
rent approaches to the calculation of electrochemical HOMO and
LUMO energy levels, ultimately referred to the absolute value of
the normal hydrogen electrode (NHE) as critically assessed in a
(formal potential criterion)
(3b)
EHOMO (eV) = −e × [(Eonset,Ia/V (Fc+|Fc) + 4.8 V (Fc+|Fc vs zero)]
(onset criterion)
(3c)
where e denotes the electron charge.
Table 2
Calculated properties of the Q band of symmetric Zn(II)-porphyrinates 1–5 and 3ꢀ.
Compound
ꢂmax (nm) (calc.)
ꢂmax (nm) (exp.)
f
HOMO–LUMO composition (%)
1
2
3
4
5
707
647
644
644
680
624
678
658
658
659
669
644
1.46
0.99
1.10
1.08
1.41
0.64
82
78
79
78
82
75
3ꢀ

Table 3
A synopsis of reduction and oxidation potentials (onset, formal and peak maxima ones) for the Zn(II)-porphyrinates investigated. The data have been obtained working in DMF + 0.1 TBAP, obtained with GC electrodes and referred
to the ferricinium|ferrocene reference redox couple. Also the corresponding calculated values of HOMO and LUMO level energies and HOMO–LUMO gaps energies are reported. Data are ordered in increasing order of sum of
Hammett ꢀp parameters [56] for the two substituents in 5,15 meso positions. Eg values are highlighted in bold character.
#
Substituents
ꢁꢀp
Onset criteriona
Formal potential criterion
◦ꢀ
EIc/V
EIa/V
ELUMO/eV
−3.34
EHOMO/eV
−4.87
Eg/eV
E
/V
EI◦aꢀ /V
ELUMO/eV
−3.21
EHOMO/eV
Eg/eV
Ic
(Fc+|Fc)
(Fc+|Fc)
(Fc+|Fc)
−1.59
(Fc+|Fc)
1
p-C6H4NMe2
p-C6H4NMe2
−1.66
−1.46
0.072
1.53
0.144
0.23b
0.288
0.448
0.453
0.450
−4.94
−5.03c
−5.09
−5.25
−5.25
−5.25
1.73
1.82
1.82
1.90
1.88
1.84
2.07
2
3
4
5
p-C6H4OMe
p-C6H4COOMe
p-C6H4COOH
p-C6H4NO2
p-C6H4OMe
p-C6H4COOMe
p-C6H4COOH
p-C6H4NO2
−0.54
0.90
0.90
1.56
−1.42
−1.32
−1.34
−1.20
0.249
0.341
0.351
0.366
−3.38
−3.48
−3.46
−3.60
−5.05
−5.14
−5.15
−5.17
1.67
1.67
1.69
1.57
−1.54
−1.45
−1.42
−1.39
−1.62b
−3.26
−3.35
−3.38
−3.41
−3.18c
6
7
p-C6H4NMe2
p-C6H4NMe2
p-C6H4COOH
p-C6H4NO2
−0.38
−0.12
−1.40
−1.29
0.158
0.164
−3.41
−3.51
−4.96
−4.96
1.55
1.46
−1.51
0.230
0.36b
0.235
0.38b
0.454
0.391
−3.29
−5.03
−5.16c
−5.03
−5.18c
−5.25
−5.19
1.74
1.87
1.62
1.86
1.93
1.77
1.88
−1.39
−1.48b
−1.47
−1.38
−1.49b
−3.41
−3.32c
−3.33
−3.42
−3.31c
8
9
p-C6H4OMe
p-C6H4OMe
p-C6H4COOH
p-C6H4NO2
0.49
0.83
−1.24
−1.28
0.370
0.331
−3.56
−3.52
−5.17
−5.13
1.61
1.61
3ꢀ
6ꢀ
p-COOEt
p-C6H4NMe2
p-COOEt
p-C6F4COOH
0.90
−0.11
−1.15
−1.38
0.488
0.099
−3.65
−3.42
−5.29
−4.90
1.64
1.48
−1.25
−1.46
0.614
0.142
0.33b
−3.55
−3.34
−5.41
−4.94
−5.12c
1.86
1.60
1.76
#
Substituents
ꢁꢀp
Peak maxima criterion
Ep,Ic/V
Ep,Ia/V
ELUMO/eV
EHOMO/eV
Eg/eV
Ep,Ic − Ep,IIc/V
Ep,Ia − Ep,IIa/V
(Fc+|Fc)
(Fc+|Fc)
1
2
3
4
5
p-C6H4NMe2
p-C6H4OMe
p-C6H4COOMe
p-C6H4COOH
p-C6H4NO2
p-C6H4NMe2
p-C6H4OMe
p-C6H4COOMe
p-C6H4COOH
p-C6H4NO2
−1.66
−0.54
0.90
−1.621
−2.144b
−1.565
−2.089
−1.427
−1.917b
−1.466
−1.843b
−1.426
−1.673b
0.170
0.289b
0.359
−3.18
−3.24
−3.37
−3.33
−4.97
−5.09c
−5.16
1.79
1.91
1.92
0.52
0.52
0.49
0.38
0.25
0.12
0.464
0.470
0.507
−5.26
−5.27
−5.31
1.89
1.94
0.90
1.56
−3.37
1.93
2.18
−3.13c
6
7
8
9
p-C6H4NMe2
p-C6H4NMe2
p-C6H4OMe
p-C6H4OMe
p-C6H4COOH
p-C6H4NO2
p-C6H4COOH
p-C6H4NO2
−0.38
−0.12
0.49
−1.549
−2.090b
−1.386
−1.536b
−1.465
−2.050b
−1.400
−1.518b
0.267
0.409b
0.260
0.421b
0.490
−3.25
−5.07
−5.21c
−5.06
−5.22c
−5.29
1.82
1.96
1.65
1.96
1.96
0.54
0.15
0.59
0.12
0.14
0.16
−3.41
−3.26c
−3.34
0.83
0.426
−3.40
−5.23
1.83
1.94
0.26
0.16
0.685b
−3.28c
−5.48c
3ꢀ
6ꢀ
p-COOEt
p-COOEt
0.90
−1.270
−1.672b
−1.497
−1.953b
0.651
−3.53
−5.45
1.92
0.40
0.46
p-C6H4NMe2
p-C6F4COOH
−0.11
0.199
−3.30
−5.00
1.70
1.85
0.355b
−5.16c
a
b
c
Onset parameters are given only for first oxidation and first reduction peaks since they are not always available for the subsequent ones.
Potentials corresponding to the second reduction or oxidation process.
Energy level calculated from the potential value corresponding to the second reduction or oxidation process.

516
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
Fig. 6. CV (0.2 V s−1 scan rate) and DPV patterns for symmetric Zn(II)-porphyrinates 1–5 in DMF.
The resulting ELUMO and EHOMO levels are reported in Table 3
reduction peaks (some of which being not reversible) besides the
first ones, when necessary.
too. Concerning the choice between the three criteria, as recently
discussed in detail by one of us [55], Eonset and Ep values are
related to a complex bundle of parameters, depending on the
specific reaction mechanism and being influenced by the exper-
imental conditions particularly in the case of electrochemically
irreversible electron transfers. On the other hand, the approach
based on standard potentials E◦ (in practice, formal potentials
Looking at Fig. 6a, at least two reduction and one oxidation
peak are neatly perceivable for all symmetric Zn(II)-porphyrinates,
reversible or quasi reversible chemically (presence of a return peak
at least at high scan rates) and electrochemically (fast electron
transfer, implied by the nearly constancy of Ep at increasing scan
rates v). In the case of 1, having two N(CH3)2 strong donor groups,
two close reversible oxidation peaks are perceivable, although they
are nearly merging. The voltammetric features of 3ꢀ, when com-
pared to 3 (Fig. 7) point to a significant decrease of chemical
reversibility for both the first oxidation and the first reduction elec-
tron transfer processes. The CV and DPV of asymmetric push–pull
Zn(II)-porphyrinates 6–9 and 6ꢀ, reported in Fig. 8, show two
reversible reduction peaks and in most cases two close and nearly
merging oxidation peaks.
E
◦ꢀ, neglecting activity coefficients), being independent on both
scan rate and concentration, and directly related to the process
ꢃG◦, is the safest, most reproducible, best defined, and there-
fore most significant one; however it is applicable only when
all the peaks to be discussed are electrochemically and chem-
ically reversible or quasi reversible. In our case this holds for
the first oxidation and reduction peaks, so concerning HOMOs
and LUMOs we can consider the values calculated according to
this criterion (Table 3, middle section). They appear in excellent
consistency with the spectroscopic gaps derived by the Q band
(Table 1) both as values and trend, pointing to the same energy
levels being involved in the electrochemical and the spectroscopic
process. Moreover all HOMO and LUMO energy levels appear to
comply with the condition required for DSSC application, i.e. all
LUMOs are well above the TiO2 conduction band level (∼−3.95 eV)
The series of symmetric Zn(II)-porphyrinates 1–5 and 3ꢀ affords
a valuable benchmark in order to evaluate the extent of the elec-
tron communication with the porphyrin core along the whole
conjugated ␲ system (with concurrent localization of the redox
centres), hinging on the evaluation of the effects of the donor or
acceptor groups of the ethynyl phenyl moieties linked to 5 and
15 meso positions of the porphyrinic ring. A similar approach had
previously been followed by Therien et al. [15] although their
experiments were significantly impaired by the solvent choice,
besides lacking the carboxylic substituent in the form of an acid
or an ester, essential for applications of the metal porphyrins in
DSSCs.
[25] and all HOMOs are sufficiently below the I−/I3 couple
−
(∼−4.8 eV) [7] excepting perhaps the case of 6ꢀ falling somehow
borderline.
Further clues on the nature and localization of the electron
transfer processes can be obtained form a detailed analysis of the CV
peak position and morphology, which will be made in terms of Ep
values, since it allows taking into account second oxidation and/or
Zn(II)-porphyrinates from 1 to 5 constitute a series of sym-
metric electronic structures which, according to the following

P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
517
Fig. 7. Contrasting CV (0.2 V s−1 scan rate) and DPV patterns for symmetric Zn(II)-porphyrinates 3 and 3ꢀ in DMF.
trend from donor to acceptor substituents in para position of the
phenylethynyl linker:
potential of the first oxidation peak corresponding, for 2–5, to a
process involving the oxidation of the electronic core of the ring
while for 1 it is the second oxidation peak that correspond to the
ring oxidation process (see later and Table 3).
For symmetric Zn(II)-porphyrinates 1 or 5 the N(CH3)2 and
NO2 terminal groups can themselves undergo oxidation or reduc-
tion processes, respectively, at potentials slightly different from
N(CH3)2 > OCH3 > COOCH3 ≈ COOH > NO2
gradually decreases the electron density on the porphyrinic core,
as suggested by a regular shift towards more positive values of the
Fig. 8. CV (0.2 V s−1 scan rate) and DPV patterns for symmetric Zn(II)-porphyrinates 6–9 plus 6ꢀ in DMF.

518
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
Fig. 9. First and second oxidation EIa, EIIa) and reduction (EIc, EIIc) peak potentials, referred to the Fc+|Fc reference redox couple for the Zn(II)-porphyrinates investigated.
Symmetric 1–5: circles; asymmetric push–pull 6–9: stars; symmetric 3ꢀ without phenyl group: triangles; asymmetric 6ꢀ with a perfluorinated phenyl ring: diamonds. Straight
lines are drawn on the points reported as full symbols.
those involving only the porphyrinic core (Table 3) as it occurs for
2–4. This can result in localization of the first oxidation or reduction
process on these groups away from the porphyrinic core, as it will
be analysed and discussed more in detail later on.
The substituent effects on the peak potentials for oxidation or
reduction processes centred on the porphyrinic core can be con-
veniently rationalized in terms of the sum of the ꢀp Hammett
parameters of the electron donor or acceptor groups [56] so that the
three peaks, one oxidation and two reduction, exhibit linear rela-
tionships vs ꢁꢀp, (full circles in Fig. 9 and corresponding straight
lines).
the second one involves the oxidation of the porphyrinic core. In
accordance with such assignment, the DFT computational inves-
tigation has confirmed that the electron density of the N(CH3)2
group of the phenylethynyl linker lies always, both in symmetric
or asymmetric Zn(II)-porphyrinates, on the HOMO level, involved
in the first oxidation process. It must be noticed, though, that while
this assumption is the only possible in the asymmetric structures
6, 6ꢀ and 7 featuring only one N(CH3)2 group, in the symmetric
structure 1, featuring two N(CH3)2 groups, we could also assume
that the two nearly merging oxidation peaks may correspond to
the oxidations of both the two N(CH3)2 groups, as equivalent
redox centres slightly interacting through the ␲ conjugated system
localized on the 5,15 meso positions. Therefore we could assume
an oxidation mechanism which involves two oxidation steps as
represented in Scheme 1:
The slopes of these Hammett linear plots, which are an indica-
tion of the efficiency of the transmission of the electronic inductive
effects from the donor or the acceptor substituents to the por-
phyrinic core acting as redox site, appear to be significant and,
meaningfully, nearly equal for the three peaks, confirming that they
take place at the same redox site, that is the porphyrinic core, since
it is the only redox site common to all the Zn(II)-porphyrinates
1–5 [52]. This evidence confirms the significant efficiency of trans-
mission of the electronic inductive effects by the phenylethynyl
linkers to the porphyrinic core. It is worthwhile recalling here
a previous comparative test of ours between phenylethenyl and
phenylethynyl linkers, showing that triple bonds are more suit-
able than double ones for the task of ensuring conjugation between
the porphyrinic core and the peripheral substituted phenyl groups
[52,57].
Concerning this localization of the redox site, we must notice
that in the case of 1 carrying two N(CH3)2 groups, in which two
nearly merging reversible oxidation peaks are observed (Fig. 6),
only the second one fits in the Hammett straight line (full circle
in Fig. 9). The same feature also holds with all asymmetric Zn(II)-
porphyrinates featuring two oxidation peaks and one N(CH3)2
group since the second oxidation peak and not the first one lies
on the Hammett straight line for 6, 6ꢀ and 7 (Fig. 9), while, in the
case of 9, featuring two oxidation peaks both of them not con-
cerning N(CH3)2 groups, it is the first and not the second that
fits in the Hammett straight line. This evidence, in accordance
with what reported by some of us and by other research groups
[15,52,57], can be explained in terms of the first oxidation peak cor-
responding to the localized oxidation of the N(CH3)2 group while
•
a first oxidation step which must result in the generation of a
radical cation localized on one of the two electron rich N(CH3)2
groups; obviously, this structure is in resonance with that with
the radical cation localized on the specular N(CH3)2 group;
•
once the radical cation is formed, because of the excellent
electronic exchange via the ␲ system involving the 5,15 meso
positions [52] and of the electron donor capacity of the other
N(CH3)2 group, it is possible that it rapidly assumes an
alternative resonant structure where the unpaired electron is
delocalized on the porphyrinic core, whereas both the N(CH3)2
groups are in their quaternary form being positively charged. As a
matter of fact this electronic structure is evidently most balanced
in term of charge distribution;
•
finally since the second oxidation step must involve mainly
the porphyrinic core in accordance with the fit of second oxi-
dation peak in the Hammett linear relationship (Fig. 9), both
N(CH3)2 groups are oxidized in their quaternary form produc-
ing as expected for a direct oxidation of both the two N(CH3)2
groups.
Of course the final doubly positive charged cation is more diffi-
cult to oxidize, since in Fig. 6b the peak corresponding to the third
oxidation is at a more positive potential and well separated from
the previous oxidation peaks.

P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
519
tBu
tBu
tBu
tBu
N
N
N
N
N
N
N
N
- e
Zn
Zn
N
N
N
N
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
N
N
N
N
N
N
- e
Zn
Zn
N
C
C
C
C
N
N
C
C
C
C
N
N
N
tBu
tBu
tBu
tBu
Scheme 1. Pathway hypothesized for the first and second oxidation processes in Zn(II)-porphyrinate 1.
In the opposite limiting case of the symmetric structure 5, with a
nitro group on both the phenylethynyl moieties, featuring a couple
of nearly merging reduction peaks, the first and second reduction
process could be significantly localized on the nitro group.
This suggestion is based on the evidence found for regular
sequences of substituted aromatic systems, e.g. aryl bromides, that
the radical anion formed in the first reduction process increasingly
becomes more localized in proximity of the very strong electron
withdrawing substituents [58], with no solution of continuity, the
nitro substituent being the limiting case in the series. Moreover,
DFT calculations on the ground state energy level of the neu-
tral Zn(II)-porphyrinate 5 have shown that the reduction process
involving the LUMO level, is mainly localized on the nitro group of
the phenylethynyl linker (Fig. 5).
Significant electronic effects induced by the donor and acceptor
groups of the phenylethynyl moieties on the electronic properties
of the porphyrinic core are confirmed by the following evidence.
According to the well known Kadish relationship [59] confirmed
by a computational investigation by Zerner and Gouterman [60]
the separation between first oxidation and first reduction poten-
tials (corresponding to the electrochemical HOMO–LUMO gap of
a porphyrinic simple core) for tetraphenylporphyrin (TPP) and
diphenylporphyrin (DPP) templates must range in the interval
of a metal ion, both oxidation and reduction potentials should
shift in the same direction and to the same extent. Therefore the
above gap should remain constant, provided that both oxidation
and reduction processes are still centred on the porphyrinic core.
It is however reasonable that in our porphyrinic systems the above
rule cannot be satisfied in the following two cases:
(a) when the first oxidation or reduction centres or both do no more
coincide with the porphyrinic core as redox site, the separation
could become larger or narrower, as a consequence of a shift of
the potential of the first oxidation or reduction peak;
(b) when the effective ␲ conjugation of the porphyrinic architec-
ture is enhanced, the separation between the first oxidation
and reduction potentials should decrease, since HOMO rises and
LUMO falls implying less positive Epa (easier oxidation) and less
negative Epc (easier reduction).
Now, concerning point (b) all of the porphyrinic architectures
investigated in this work have a significantly large ␲ conjugated
system. This should increase the global effective conjugation and
therefore should contribute to decrease the HOMO–LUMO gap, as
confirmed by the DFT computational investigation (Fig. 4). Fur-
thermore, concerning point (a), in symmetric and asymmetric
Zn(II)-porphyrinates carrying one or two NMe2 or NO2 groups
the first oxidation or reduction centre does not coincide with the
porphyrinic core so that in both cases we have a decrease of the
gap. This can explain why in the porphyrinic architectures investi-
gated in this work we observed a Ep,Ia − Ep,Ic difference significantly
lower than the above value (5), i.e. about 1.85 V (obtained from the
difference of the intercepts of the Ep,Ia and Ep,Ic straight lines in
Fig. 9).
ꢀ
ꢀ
EI◦a − EI◦c = 2.25 0.15 V
(4)
for formal potentials (corresponding to a slightly larger difference,
in terms of peak potentials).
Ep,Ia − Ep,Ic ≈ 2.36 0.15 V
(5)
It is known that in the presence of purely inductive effects gener-
ated by substituents on the porphyrinic core and/or by the presence

520
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
This feature is fully consistent, inter alia, with the Therien’s
the porphyrinic core since only this peak fits the Hammett linear
investigation cited above [15].
relationship (Fig. 9) obtained taking into account the values of
the first oxidation peak involving surely the porphyrinic core, as
in the case of asymmetric Zn(II)-porphyrinates not containing a
N(CH3)2 group such as 8 and 9.
Moreover the difference between first and second reduction
peak potentials in simple tetraphenylporphyrin and diphenylpor-
phyrin templates must satisfy the Kadish relationship [61–63]:
•
•
ꢀ
ꢀ
EI◦c − EI◦Ic = 0.42 0.03 V (≈ Ep,Ic − Ep,IIc
)
(6)
A system of two close nearly merging reversible reduction peaks
EIc and EIIc is always observed in the presence of a nitro group as
in 7 and 9 (see Table 3 and Fig. 9). As already mentioned, unlike
the symmetric Zn(II)-porphyrinate 5, it is the second one that
better fits the Hammett linear relationship. Accordingly, the first
one should be associated to the formation of a radical anion local-
ized on the nitro group, while only the second one to a reduction
process centred on the porphyrinic core.
Finally the trend of CV peaks of the asymmetric Zn(II)-
porphyrinate 8 is quite consistent with that typical of the
symmetric ones (see Fig. 9); it appears also that the absence of
strong donor N(CH3)2 or acceptor NO2 groups, 8 hold as for the
symmetric ones in the Hammett plots of Fig. 9, which takes into
account only oxidation or reduction processes involving the por-
phyrinic core. It follows that in 8 the push–pull system is so weak
that both the oxidation and reduction processes involve mainly
the porphyrinic core, not too much perturbed by the push–pull
character of the ␲ conjugated system.
Taking into account the symmetric Zn(II)-porphyrinates for which
the first two reduction processes are surely localized on the por-
phyrinic core, i.e. 1 and 2, we have found a significantly higher
difference, i.e. 0.52 V (see Table 3), as a consequence of an efficient ␲
conjugation, resulting in a more efficient separation between nega-
tive charges. However it is worthwhile noticing that for increasing
strength of electron acceptor substituents the second reduction
peaks Ep,IIc gradually deviate from the Hammett linear relation-
ship, approaching Ep,Ic, that is the value of the first reduction peak,
until they nearly merge as in 5, carrying two nitro groups (Fig. 9).
This feature, pointing to two nearly equivalent and independent
redox centres in the extreme case 5, is another evidence of the
above suggested partial localization of the first negative charge not
completely on the porphyrinic core but in increasing proximity of
the NO2 electron acceptor substituent, so that the introduction of
a second negative charge becomes symmetrically localized on the
other NO2 group, far away from the first one, since this process is
more facile from the electrostatic point of view (Fig. 9).
A comparison between the trend of the CV peaks of the asym-
Similar features, namely
a
slightly larger Ep,Ic − Ep,IIc dif-
metric Zn(II)-porphyrinate 6 and its fluorinated homologous 6ꢀ
(Fig. 8) affords an evaluation of the effects of the perfluorination
of the phenyl ring carrying the carboxylic group, that is:
ference of about 0.45–0.52 V are observed for asymmetric
Zn(II)-porphyrinates 6 and 8, respectively, with the exception of
Zn(II)-porphyrinates carrying only one nitro group, e.g. 7 and 9, for
which the first reduction process is not localized on the porphyrinic
core, thus resulting in a much narrower gap of about 0.12–0.15 V
(Table 3).
•
the two reduction peaks are shifted towards more positive val-
ues (Table 3), in accordance with an easier reduction as expected
since fluorine atoms set as electron acceptors and therefore
should act synergistically to the electron withdrawing effect of
the adjacent carboxylic group in order to facilitate the reduction
of the oxidized porphyrinic system;
The symmetric Zn(II)-porphyrinate 3ꢀ (voltammetric patterns in
Fig. 7 and triangle points in Fig. 9) affords by comparison with 3 an
evaluation of the effects of the absence of the phenyl group in the
linker, which appears to be:
•
concerning the first oxidation peak EIa, one should expect a shift
•
a significant decrease in radical cation and above all radical anion
towards more positive values (more difficult oxidation), albeit
less than the reduction peaks since the oxidation site, which is
N(CH3)2 group, as in the case of 6, is away from the fluorinated
aromatic ring. However, in 6ꢀ not only the first but also the second
oxidation peak remarkably shift towards lower potential values
(easier oxidation) (Table 3). This trend could be possibly justified
in terms of a ␲ system along the push–pull system partially loos-
ing its planarity upon fluorination of the phenyl group, so that the
communication by ␲ conjugation between the electron accep-
tor moiety and the N(CH3)2 donor group is much less facile.
In accordance some of us have already shown, by a theoretical
DFT approach, that for 6ꢀ a planar and a tilted arrangement, with
respect to the porphyrin ring, of the fluorinated aromatic ring
carrying the COOH group are energetically comparable [32]. A
similar kind of relevant structural effect, has been described by
Kadish et al. about the electronic properties of tetraphenyl por-
phyrins in which the ␤ pyrrole positions are gradually substituted
by halogen atoms, resulting in loss of planarity [64]. Since in our
series of Zn(II)-porphyrinates only 1, that is a symmetric structure
with two donor N(CH3)2 groups, is oxidized more easily than 6ꢀ,
it follows that in 6ꢀ the behaviour of the N(CH3)2 group is more
similar to that of a N(CH3)2 group not perturbed as 1, by the
conjugation effect produced by an antagonist electron acceptor
group, as in the case of 6 and 7. It follows that the electron com-
munication along the push–pull system is not relevant for 6ꢀ. This
intriguing anomaly is consistent with the photoelectrochemical
behaviour of 6ꢀ when acting as dye in DSSCs [32], pointing to a
very low efficiency in comparison with 6; actually, as above men-
tioned, the experimental value of the HOMO of 6ꢀ appears too near
chemical stability: the first reduction peak, neatly observable for
3, is absent for 3ꢀ even at high scan rate and also the CV and
DPV patterns following the first reduction peak are much more
complex and less reversible for 3ꢀ. The first oxidation peak, only
partially reversible for 3 at 0.2 V s−1 scan rate, but improving at
higher scan rates, is nearly irreversible for 3ꢀ;
•
a significant shift of both the first oxidation and the first reduction
to higher values of the potential (by about 0.18–0.15 V) for 3ꢀ with
respect to 3 (Table 3), pointing to the presence of the phenyl ring
making the molecule slightly electron richer with the conjugation
efficiency remaining nearly the same.
The above approach to the analysis of the CV and DPV pat-
terns and of the values of oxidation and reduction potentials can
be conveniently extended to the asymmetric push–pull Zn(II)-
porphyrinates 6–9. Their voltammetric features, shown in Fig. 8,
are consistent with our former observations because:
•
a system of two close and nearly merging reversible oxidation
peaks is always observed in the presence of a N(CH3)2 group,
as for 6 and 7, with 0.14 V and 0.16 V difference between the
first and second oxidation peaks, smaller than the difference
of 0.26 V between these two oxidation peaks surely centred, in
the absence of a N(CH3)2 group on the porphyrinic core, as
in 8 and 9 (Table 3). The first oxidation peak in the case of 6
and 7 should correspond, as above suggested by some of us and
other research groups, [15,52,57] to an oxidation localized on the
N(CH3)2 group, while the second peak should be centred on

P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
521
to the I−/I3− couple energy level (−4.94 eV instead of −5.03 eV for
6, Table 3).
irreversible multistep processes corresponding to the break-
down of the molecule take place much earlier on both the
reduction and the oxidation side (particularly in the latter
case).
independent and not interacting by a significant electron trans-
fer process, an intriguing feature already reported by some of us
[32].
Finally the spectroscopic and TDDFT theoretical calculations on
the symmetric porphyrinic structures do confirm the ambivalent
character of the porphyrinic core, first reported by some of us [52],
which act as acceptor with donor NMe2 groups, and as donor with
acceptor NO2 groups.
•
4. Conclusions
The whole set of interpretative/predictive guidelines thus
made available will hopefully provide valuable support for target-
oriented design and optimization of the synthetic efforts in the
current development of highly efficient porphyrinic dyes for DSSC
applications.
The combined electrochemical, spectroscopic and theoretical
investigation of our series of symmetric and push–pull asymmet-
ric Zn(II)-porphyrinates 5,15-meso substituted with phenylethynyl
linkers affords a rationalization of their electronic properties as a
function of the nature of the substituents on the para position of
the phenyl ring of the linkers.
Acknowledgements
In the case of moderately electron acceptor ( COOR, COOH)
and/or donor ( OCH3) substituents the first oxidation or reduc-
tion processes are rationalized in terms of predominant inductive
effects since their potentials satisfy classical Hammett straight lines
holding for both symmetric and asymmetric structures; further-
more, such straight lines are parallel, pointing to both the electron
transfer processes being centred mainly on the porphyrinic core.
The same also applies to the second reduction process. The dif-
ference between first oxidative and reductive potentials (and,
concurrently the HOMO–LUMO gap) is quite constant, but lower
than that of the classical Kadish relationships, pointing to such
porphyrinic structures having a certain amount of global effec-
tive conjugation which introduces a perturbation of the electronic
core of the porphyrinic ring slightly different from that of TPP
or DPP templates considered by Kadish. This evidence confirms
inter alia the efficiency of the transfer of the electronic effects by
phenylethynyl linkers.
We thank the Fondazione CARIPLO (2008: Progettazione e uti-
lizzo di nuovi materiali organometallici o di coordinazione per celle
solari organiche di terza generazione) and the Italian MIUR (PRIN
2008: Progettazione e sviluppo di nuovi componenti per celle solari
a semiconduttori sensibilizzati ad alta efficienza) for financial sup-
port. AOB and FDA thank FP7-NMP-2009 project 246124 “SANS”
for financial support. MGL thanks Istituto Italiano di Tecnologia,
Platform Computation, Project SEED 2009 “HELYOS” for financial
support. We thank Dr. William Galbiati for precious collaboration
in the synthesis of some molecules and Prof. Renato Ugo for his
interest and useful discussions.
Appendix A. Syntheses of Zn(II)-porphyrinates 1, 2, 3, 3ꢀ, 4,
5 and 9
[5,15-Bis(4ꢀ-N,N-dimethylamino-phenylethynyl)-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II), 1
On the contrary, for the symmetric structure the presence of
strong electron acceptor or electron donor groups (i.e. NO2 or
N(CH3)2, respectively), not only results in a more facile promo-
tion of the first reduction or oxidation process, respectively, but
also in peripheral displacement of the first reduction/oxidation
site, possibly localized on the acceptor or donor substituent itself
(as confirmed by DFT computational investigation on the electron
density distribution of the LUMO and HOMO orbital levels), rather
than on the porphyrinic core, which appears to be involved only by
the second electron transfer. As a consequence for the asymmet-
ric push–pull structure, a higher and asymmetric shrinking of the
HOMO–LUMO gap occurs, with the HOMO or LUMO levels being
localized respectively on the porphyrinic core. The limiting case
is the one featuring both NO2 and N(CH3)2 in an asymmet-
ric push–pull structure, in which both LUMO and HOMO levels
appear to be peripherally localized on the NO2 and N(CH3)2
group, respectively, with both first reduction and first oxidation
steps highly promoted, affording in the narrowest HOMO–LUMO
gap in the series.
In a dry Schlenk tube 189.8 mg of [5,15-diiodo-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II) (0.189 mmol), 17.3 mg of
[Pd2(dba)3] (0.019 mmol), 19.8 mg PPh3 (0.075 mmol), 10.8 mg
of CuI (0.057 mmol) were deaerated three times with
a
vacuum–nitrogen cycle and then dissolved with a mixture of
20 ml of THF and 5 ml of Et2NH previously deaerated with four
freeze–pump–thaw cycles at −78 ◦C. At 70 ◦C a solution of 41.2 mg
of 4-ethynyl-N,N-dimethylaniline (1.5 equiv.) in 5 ml of THF, previ-
ously deaerated with three freeze–pump–thaw cycles at −78 ◦C.
The reaction mixture was stirred under nitrogen atmosphere for
24 h in the dark. Then the solvents were evaporated in vacuo and
the crude product was purified by column chromatography (SiO2,
n-hexane/THF 8:2) to afford 44.8 mg of pure product (yield 22.9%).
1H NMR (400.1 MHz, CDCl3 + py-d5) ı, ppm: 9.71 (4H, d), 8.87
(4H, d), 8.03 (4H, m), 7.90 (4H, d), 7.79 (2H, m), 6.87 (4H, d), 3.11
(12H, s), 1.55 (36H, s).
MS-FAB(+) m/z: calcd for C68H70N6Zn 1034, found 1035 [M+H]+.
UV–vis (THF) ꢂ, nm: 467 (ε = 220,000 M−1 cm−1); 668
(ε = 58,000 M−1 cm−1).
Interestingly, peripheral localization of the electronic density of
the HOMO and/or LUMO also modulates the intensity ratio of Q
and B absorption bands, as a consequence of an increased effective
conjugation along the push–pull system.
[5,15-Bis(4ꢀ-methoxy-phenylethynyl)-10,20-bis(3,5-di-tert-
butylphenyl)porphyrinate]Zn(II), 2
The presence of the phenyl group besides the ethynyl one
in the linker appears to be determining for the chemical sta-
bility of the incipient radical anion; it also appears to slightly
improve effective conjugation and to make the system slightly elec-
tron richer. Perfluorination of this phenyl ring, in a asymmetric
push–pull structure, intended to enhance the power of the elec-
tron attractor side chain and therefore the push–pull character of
the structure, results instead in apparent loss of conjugation, with
the electron rich and the electron poor moiety, localized on the
5,15 meso position of the porphyrinic ring, appearing reciprocally
In a dry Schlenk tube 123.7 mg of [5,15-diiodo-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II) (0.123 mmol), 14.2 mg of
Pd(PPh3)4 (0.012 mmol) and 3.5 mg of CuI (0.018 mmol) were
deaerated three times with a vacuum–nitrogen cycle and then dis-
solved in a mixture of 15 ml of THF and 2.5 ml of Et2NH, previously
deaerated with five freeze–pump–thaw cycles at −78 ◦C.
Under stirring, in nitrogen atmosphere and at 25 ◦C, a solu-
tion of 25.0 ␮l of 4-ethynylanisole (1.6 equiv.) in 5 ml of THF,

522
P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
previously deaerated with five freeze–pump–thaw cycles at −78 ◦C,
was added dropwise to the reaction. The reaction was left under
nitrogen atmosphere for 36 h in the dark. The solvents were
removed in vacuo and the crude product obtained was purified
by column chromatography (SiO2, n-hexane/THF 9:1) to afford
65.1 mg of pure product (yield 52%).
(0.0193 mmol) were dissolved in 1.7 ml of a 3:1 THF:H2O mixture;
under stirring, 2.6 mg of LiOH (0.106 mmol, 5.5 equiv.) were added.
The reaction was left at room temperature for 24 h. The solvents
were removed in vacuo and the crude product was dissolved
in 1 ml of THF and diluted with 10 ml of DCM. The dark green
solution was then treated with 10 ml of a 20% aqueous solution of
citric acid and left at room temperature overnight under vigorous
stirring. Then, the reaction mixture was transferred in a funnel, the
phases were separated and the organic phase was washed 3 × H2O
(10 ml), until pH = 6–7 was reached. The organic phase was dried
over Na2SO4, filtered and the solvent was removed in vacuo. The
title product was obtained in 45% yield (9.1 mg, dark green solid).
1H NMR (400.1 MHz, THF-d8) ı, ppm: 9.82 (4H, d), 8.97 (4H, d),
8.27 (4H, d), 8.20 (4H, d), 8.16 (4H, s), 7.96 (2H, s), 1.32 (36H, s).
1H NMR (400.1 MHz, THF-d8) ı, ppm: 9.76 (4H, d), 8.91 (4H, d),
8.15 (4H, m), 8.02 (4H, d), 7.94 (2H, m) 7.16 (4H, d), 3.94 (6H, s),
1.61 (36H, s).
MS-FAB(+) m/z: calcd for C66H64N4O2Zn 1008, found 1009
[M+H]+.
UV–vis (THF) ꢂ, nm: 451 (ε = 258,000 M−1 cm−1); 658
(ε = 37,000 M−1 cm−1).
[5,15-Bis(4ꢀ-carboxymethyl-phenylethynyl)-10,20-bis(3,5-di-
tert-butylphenyl)porphyrinate]Zn(II), 3
MS-ESI(−) m/z: 1035.3852 [M−H]−; 517.1868 [M−2H]2−
.
UV–vis (THF) ꢂ, nm: 450 (ε = 336,000 M−1 cm−1); 660
(ε = 50,000 M−1 cm−1).
In a dry Schlenk tube 96.5 mg of [5,15-diiodo-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II) (0.0963 mmol), 5.8 mg of
Pd(dba)2 (0.0101 mmol), 2.7 mg of CuI (0.0143 mmol), 1.35 ml
of Et3N and 5 ml of dry THF were deaerated with four
freeze–pump–thaw cycles at −78 ◦C.
[5,15-Bis(4ꢀꢀ-nitro-phenylethynyl)-10,20-bis(3,5-di-tert-
butylphenyl)porphyrinate]Zn(II), 5
In a dry Schlenk tube 80.1 mg of [5,15-diiodo-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II) (0.080 mmol), 29.1 mg
of 4-ethynyl-nitrobenzene (2.4 equiv.), 7.3 mg of Pd2(dba)3
(0.008 mmol), 13.1 mg of PPh3 (0.050 mmol) and 5.0 mg of CuI
(0.026 mmol) were deaerated three times with a vacuum–nitrogen
cycle and then dissolved in a mixture of 20 ml of THF and 5 ml of
Et2NH, previously deaerated with five freeze–pump–thaw cycles
at −78 ◦C were dissolved in a mixture of 8 ml of THF and 1.3 ml of
Et2NH. Under stirring, the reaction mixture was heated at 70 ◦C
for 24 h in nitrogen atmosphere. The solvents were removed in
vacuo and the crude product obtained was purified by column
chromatography (SiO2, n-hexane/THF 8:2, Rf: 0.22) to afford 8.8 mg
of product (yield 11%).
Under stirring at room temperature, a solution of 38.4 mg of
methyl 4-ethynylbenzoate (0.240 mmol, 2.5 equiv.) in 2.8 ml of
THF, previously deaerated with four freeze–pump–thaw cycles at
−78 ◦C, was added via cannula under nitrogen atmosphere to the
reaction mixture. The reaction was heated at 70 ◦C for 40 h. The sol-
vents were removed in vacuo and the crude product was purified
by column chromatography (SiO2, n-hexane/THF 85:15, Rf = 0.31)
to afford 45 mg of pure product (yield 44%).
1H NMR (400.1 MHz, CDCl3 + py-d5) ı, ppm: 9.68 (4H, d), 8.90
(4H, d), 8.13 (4H, d), 8.00 (4H, s), 7.98 (4H, d), 7.78 (2H, s), 3.91 (6H,
s), 1.51 (36H, s).
MS-FAB(+) m/z: calcd for C68H64N4O4Zn 1064, found 1065
[M+H]+.
1H NMR (400.1 MHz, THF-d8) ı, ppm: 9.82 (4H, d), 8.98 (4H, d),
8.50 (4H, d), 8.34 (4H, d), 8.16 (4H, m), 7.97 (2H, m), 1.63 (36H, s).
MS-FAB(+) m/z: calcd for C64H58N6O4Zn 1038, found 1038 [M]+.
UV–vis (THF) ꢂ, nm: 458 (ε = 98,000 M−1 cm−1); 669
(ε = 17,000 M−1 cm−1).
UV–vis (THF) ꢂ, nm: 452 (ε = 334,000 M−1 cm−1); 660
(ε = 48,000 M−1 cm−1).
[5,15-Bis(4ꢀ-carboxyethyl-ethynyl)-10,20-bis(3,5-di-tert-
butylphenyl)porphyrinate]Zn(II), 3ꢀ
[5-Bromo-15-(4ꢀ-methoxy-phenylethynyl)-10,20-bis(3,5-di-tert-
butylphenyl)porphyrinate]Zn(II)
In a dry Schlenk tube 95.8 mg of [5,15-diiodo-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II) (0.0956 mmol), 11 mg of
Pd(PPh3)4 (0.00956 mmol) and 2.9 mg of CuI (0.0152 mmol), were
deaerated three times with a vacuum–nitrogen cycle and then dis-
solved in a mixture of 17 ml of THF and 4 ml of Et3N, previously
deaerated with five freeze–pump–thaw cycles at −78 ◦C. Under
stirring, in nitrogen atmosphere and at 25 ◦C, 24 ␮l of ethyl propio-
late (2.5 equiv.) were added dropwise to the reaction. The mixture
was allowed to react at 25 ◦C for 24 h. The solvents were removed
in vacuo and the crude product obtained was purified by column
chromatography (SiO2, n-hexane/THF 9:1) to afford 55.4 mg of pure
product (yield 59.5%).
In a dry Schlenk tube 108.6 mg of [5,15-dibromo-10,20-bis(3,5-
di-tert-butylphenyl)porphyrinate]Zn(II) (0.128 mmol), 14.8 mg of
Pd(PPh3)4 (0.013 mmol) and 3.6 mg of CuI (0.019 mmol) were
deaerated three times with a vacuum–nitrogen cycle and then dis-
solved in a mixture of 12 ml of THF and 5 ml of Et2NH, previously
deaerated with five freeze–pump–thaw cycles at −78 ◦C.
Under stirring, in nitrogen atmosphere and at 25 ◦C, 20 ␮l of 4-
ethynylanisole (1.2 equiv.) was added to the reaction. The reaction
was left under nitrogen atmosphere for 24 h. The solvents were
removed in vacuo and the crude product obtained was purified by
column chromatography (SiO2, n-hexane/THF 8.9:1.1, Rf = 0.36) to
afford 58.3 mg of pure product (yield 47%).
1H NMR (400.1 MHz, CDCl3 + py-d5) ı, ppm: 9.66 (4H, d), 8.91
(4H, d), 7.96 (4H, m), 7.78 (2H, m), 4.48 (4H, q), 1.50 (36H, s), 1.48
(6H, t).
1H NMR (400.1 MHz, CDCl3 + py-d5) ı, ppm: 9.74 (2H, d), 9.64
(2H, d), 8.92 (2H, d), 8.89 (2H, d), 8.02 (4H, m), 7.97 (2H, d), 7.82
(2H, m) 7.10 (2H, d), 3.95 (3H, s), 1.57 (36H, s).
MS-FAB(+) m/z: calcd for C58H60N4O4Zn 940, found 941 [M+H]+.
UV–vis (THF) ꢂ, nm: 440 (ε = 260,000 M−1 cm−1); 644
(ε = 24,000 M−1 cm−1).
[5-(4ꢀ-Methoxy-phenylethynyl)-15-(4ꢀꢀ-nitro-phenylethynyl)-
10,20-bis(3,5-di-tert-butylphenyl)porphyrinate]Zn(II), 9
[5,15-Bis(4ꢀ-carboxy-phenylethynyl)-10,20-bis(3,5-di-tert-
butylphenyl)porphyrinate]Zn(II), 4
In
a
dry Schlenk tube 58.3 mg of [5-bromo-15-(4ꢀ-
In a one-necked round bottom flask equipped with a magnetic
stir bar 20.6 mg of [5,15-bis(4ꢀ-carboxymethyl-phenylethynyl)-
methoxy-phenylethynyl)-10,20-bis(3,5-di-tert-butylphenyl)por-
phyrinate]Zn(II) (0.061 mmol) and 21.1 mg of 4-ethynyl-
nitrobenzene (2.3 equiv.) were dissolved in a mixture of 8 ml
10,20-bis(3,5-di-tert-butylphenyl)porphyrinate]Zn(II)
(3)

P.R. Mussini et al. / Electrochimica Acta 85 (2012) 509–523
523
of THF and 1.3 ml of Et2NH. The solution was deaerated with five
freeze–pump–thaw cycles at −78 ◦C. At room temperature and
under nitrogen atmosphere 9.0 mg of Pd(PPh3)4 (0.008 mmol)
and 4.0 mg of CuI (0.021 mmol) were added. Under stirring, the
reaction mixture was heated at 75 ◦C for 24 h. The solvents were
removed in vacuo and the crude product obtained was purified
by column chromatography (SiO2, n-hexane/THF 9:1) to afford
20.4 mg of pure product (yield 33%).
[21] C.-P. Hsieh, H.-P. Lu, C.-L. Chiu, C.-W. Lee, S.-H. Chuang, C.-L. Mai, W.-N. Yen, S.-J.
Hsu, E.W.-G. Diau, C.-Y. Yeh, Journal of Materials Chemistry 20 (2010) 1127.
[22] C.-F. Lo, S.-J. Hsu, C.-L. Wang, Y.-H. Cheng, H.-P. Lu, E.W.-G. Diau, C.-Y. Lin,
Journal of Physical Chemistry C 114 (2010) 12018.
[23] T. Bessho, S.M. Zakeeruddin, C.-Y. Yeh, E.W.-G. Diau, M. Grätzel, Angewandte
Chemie International Edition 49 (2010) 6646.
[24] C.-L. Wang, Y.-C. Chang, C.-M. Lan, C.-F. Lo, E.W.-G. Diau, C.-Y. Lin, Energy &
Environmental Science 4 (2011) 1788.
[25] Y.-C. Chang, C.-L. Wang, T.-Y. Pan, S.-H. Hong, C.-M. Lan, H.-H. Kuo, C.-F. Lo, H.-Y.
Hsu, C.-Y. Lin, E.W.G. Diau, Chemical Communications 47 (2011) 8910.
[26] A. Yella, H.-W. Lei, H.N. Tsao, C. Yi, A.K. Chandiran, Md K. Nazeeruddin, E.W.-G.
Diau, C.-Y. Yeh, S.M. Zakeeruddin, M. Grätzel, Science 334 (2011) 629.
[27] B. O’Regan, M. Grätzel, Nature 353 (1991) 737.
[28] A. Hagefeldt, M. Grätzel, Chemical Reviews 95 (1995) 49.
[29] J.M. Rehm, G.L. McLendon, Y. Nagasawa, K. Yoshihara, J. Moser, M. Grätzel,
Journal of Physical Chemistry B 100 (1996) 9577.
1H NMR (400.1 MHz, THF-d8) ı, ppm: 9.79 (4H, m), 8.97 (2H, d),
8.93 (2H, d), 8.48 (2H, d), 8.31 (2H, d), 8.16 (4H, m), 8.04 (2H, d),
7.96 (2H, m) 7.17 (2H, d), 3.95 (3H, s), 1.63 (36H, s).
MS-FAB(+) m/z: calcd for C65H61N5O3Zn 1023, found 1024
[M+H]+.
[30] M. Grätzel, Nature 414 (2001) 338.
UV–vis (THF) ꢂ, nm: 459 (ε = 293,000 M−1 cm−1); 667
(ε = 78,000 M−1 cm−1).
[31] M.K. Nazeeruddin (Ed.), Michael Grätzel Festschrift, A Tribute for this 60th
Birthday, Dye Sensitized Solar Cells, vol. 248, Elsevier, Amsterdam, 2004, Spe-
cial Issue in Coordination Chemistry Reviews.
[32] A. Orbelli Biroli, F. Tessore, M. Pizzotti, C. Biaggi, R. Ugo, S. Caramori, A. Aliprandi,
C. Bignozzi, F. De Angelis, G. Giorgi, E. Licandro, E. Longhi, Journal of Physical
Chemistry C 155 (2011) 23170.
Appendix B. Supplementary data
[33] C. Monnereau, E. Blart, V. Montembault, L. Fontaine, F. Odobel, Tetrahedron 61
(2005) 10113.
[34] A.J. Bard, L.R. Faulkner, Electrochemical Methods. Fundamentals and Applica-
tions, Wiley, New York, 2002, p. 648.
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.electacta.2012.08.039.
[35] T.M. Cotton, R.P. Van Duyne, Journal of the American Chemical Society 101
(1979) 7605.
˝
˝
References
[36] Z. Odabas¸ , F. Dumludag˘, A.R. Ozkaya, S. Yamauchi, N. Kobayashi, O. Bekarog˘lu,
Dalton Transactions 39 (2010) 8143.
[37] G. Gritzner, J. Kuta, Pure and Applied Chemistry 56 (1984) 461.
[38] G. Gritzner, Pure and Applied Chemistry 62 (1990) 1858.
[39] A.D. Beck, The Journal of Chemical Physics 98 (1993) 5648.
[40] V.N. Nemykin, R.G. Hadt, Journal of Physical Chemistry A 114 (2010) 12062.
[41] V.N. Nemykin, G.T. Rohde, C.D. Barrett, R.G. Hadt, J.R. Sabin, G. Reina, P. Galloni,
B. Floris, Inorganic Chemistry 49 (2010) 7497.
[42] G.A. Peralta, M. Seth, T. Ziegler, Inorganic Chemistry 46 (2007) 9111.
[43] G. De Luca, A. Romeo, L. Monsù Scolaro, G. Ricciardi, A. Rosa, Inorganic Chem-
istry 48 (2009) 8493.
[44] J. Mack, Y. Asano, N. Kobayashi, M. Stillman, Journal of the American Chemical
Society 127 (2005) 1697.
[45] A. Muranaka, Y. Asano, A. Tsuda, A. Osuka, N. Kobayashi, ChemPhysChem 7
(2006) 1235.
[46] K. Nakai, K. Kurotobi, A. Osuka, M. Uchiyama, N. Kobayashi, Journal of Inorganic
Biochemistry 102 (2008) 466.
[47] M.J. Frish, et al., Gaussian 03, Revision B.05, Gaussian, Inc., Pittsburg, PA, 2003.
[48] M. Cossi, V. Barone, The Journal of Chemical Physics 115 (2001) 4708.
[49] M. Cossi, N. Rega, G. Scalmani, V. Barone, Journal of Computational Chemistry
24 (2003) 669.
[50] K. Sonogashira, Journal of Organometallic Chemistry 653 (2002) 46.
[51] Z.-Q. Liu, Q. Fang, D.-X. Cao, D. Wang, G.-B. Xu, Organic Letters 6 (2004) 2933.
[52] T. Morotti, M. Pizzotti, R. Ugo, S. Quici, M. Bruschi, P. Mussini, S. Righetto,
European Journal of Inorganic Chemistry (2006) 1743.
[53] M. Gouterman, Optical spectra and electronic structure of porphyrins and
related rings, in: D. Dolphin (Ed.), The Porphyrins, vol. 3, Academic Press Inc,
1978, p. 1 (Chap. 1).
[54] S. Trasatti, Pure and Applied Chemistry 58 (1986) 955.
[55] A. Tacca, R. Po, M. Caldararo, S. Chiaberge, L. Gila, L. Longo, P.R. Mussini, A.
Pellegrino, N. Perin, M. Salvalaggio, A. Savoini, S. Spera, Electrochimica Acta 56
(2011) 6638.
[56] C. Hansch, A. Leo, R.W. Taft, Chemical Reviews 91 (1991) 165.
[57] E. Annoni, M. Pizzotti, R. Ugo, S. Quici, T. Morotti, M. Bruschi, P. Mussini, Euro-
pean Journal of Inorganic Chemistry (2005) 3857.
[58] A. Pierini, D.M.A. Vera, Journal of Organic Chemistry 68 (2003) 9191.
[59] K.N. Kadish, E.V. Caemelbecke, Journal of Solid State Electrochemistry 7 (2003)
254.
[60] M. Zerner, M. Gouterman, Theoretical Chemistry Accounts 4 (1996) 44.
[61] D.W. Clack, N.S. Hush, Journal of the American Chemical Society 87 (1965)
4238.
[62] R.H. Felton, H. Linschnitz, Journal of the American Chemical Society 88 (1966)
1113.
[63] C.L. Lin, M.Y. Fang, S.H. Cheng, Journal of Electroanalytical Chemistry 531 (2002)
155.
[64] K.M. Kadish, F. D’Souza, A. Villard, M. Autret, E. Van Caemelbecke, P. Bianco, A.
Antonini, P. Tagliatesta, Inorganic Chemistry 33 (1994) 5169.
[1] B.J. Coe, Nonlinear optical properties of metal complexes, in: J.A. McCleverty,
T.J. Meyer (Eds.), Comprehensive Coordination Chemistry II: From Biology to
Nanotechnology, vol. 9, Elsevier Pergamon, Oxford, 2004, p. 621.
[2] G. de la Torre, P. Vàzquez, F. Agulló-López, T. Torres, Chemical Reviews 104
(2004) 3723.
[3] M.O. Senge, M. Fazekas, E.G.A. Notaras, W.J. Blau, M. Zawadzka, Oliver B. Locos,
E.M. Ni Mhuricheartaigh, Advanced Materials 19 (2007) 2737.
[4] X. Meng, W. Zhu, H. Tian, Major classes of organic small molecules for electron-
ics and optoelectronics, in: S.-S. Sun, L.R. Dalton (Eds.), Introduction to Organic
Electronic and Optoelectronic Materials and Devices, CRC Press, 2008, p. 129.
[5] S. Di Bella, C. Dragonetti, M. Pizzotti, D. Roberto, F. Tessore, R. Ugo, Coordination
and organometallic complexes as second-order nonlinear optical molecular
materials, in: Topics in Organometallic Chemistry, volume 28/2010 (Molecular
Organometallic Materials for Optics), Springer, Berlin, 2010, p. 1.
[6] K. Hara, Molecular catalyst for energy conversion, in: T. Okada, M. Kaneko (Eds.),
Springer Series in Materials Science 111, Springer, Berlin, 2009, p. 217 (Chap.
9) and references therein.
[7] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chemical Reviews 110
(2010) 6595, and references therein.
[8] M.V. Martinez-Diaz, G. de la Torre, T. Torres, Chemical Communications 46
(2010) 7090.
[9] M.G. Walter, A.B. Rudine, C.C. Wamser, Journal of Porphyrins and Phthalocya-
nines 14 (2010) 759.
[10] H. Imahori, Key Engineering Materials 451 (2011) 29.
[11] J.N. Clifford, E. Martínez-Ferrero, A. Viterisi, E. Palomares, Chemical Society
Reviews 40 (2011) 1635.
[12] M. Pizzotti, E. Annoni, R. Ugo, S. Bruni, S. Quici, P. Fantucci, M. Bruschi, G. Zerbi,
M. Del Zoppo, Journal of Porphyrins and Phthalocyanines 8 (2004) 1311.
[13] M. Pizzotti, F. Tessore, A. Orbelli Biroli, R. Ugo, F. De Angelis, S. Fantacci, A.
Sgamellotti, D. Zuccaccia, A. Macchioni, Journal of Physical Chemistry C 113
(2009) 11131.
[14] S.M. LeCours, H.-W. Guan, S.G. DiMagno, C.H. Wang, M.J. Therien, Journal of the
American Chemical Society 118 (1996) 1497.
[15] S.M. LeCours, S.G. DiMagno, M.J. Therien, Journal of the American Chemical
Society 118 (1996) 11854.
[16] S.M. Le Cours, C.M. Phillips, J.C. de Paula, M.J. Therien, Journal of the American
Chemical Society 119 (1997) 12578.
[17] K.S. Kim, S.B. Noh, T. Katsuda, S. Ito, A. Osuka, D. Kim, Chemical Communications
(2007) 2479.
[18] E. Collini, S. Mazzucato, M. Zerbetto, C. Ferrante, R. Bozio, M. Pizzotti, F. Tessore,
R. Ugo, Chemical Physics Letters 454 (2008) 70.
[19] F. De Angelis, S. Fantacci, A. Sgamellotti, M. Pizzotti, F. Tessore, A. Orbelli Biroli,
Chemical Physics Letters 447 (2007) 10.
[20] C.-W. Lee, H.-P. Lu, C.-M. Lan, Y.-L. Huang, Y.-R. Liang, W.-N. Yen, Y.-C. Liu, Y.-S.
Lin, E.W.-G. Diau, C.-Y. Yeh, Chemistry: A European Journal 15 (2009) 1403.