Synthesis of new porphyrin/fullerene supramolecular assemblies: A spectroscopic and electrochemical investigation of their coordination equilibrium in solution

Article abstract of DOI:10.1016/j.tet.2010.10.066

Two new fullerene ligands have been designed to provide relatively simple frameworks to build supramolecular systems containing both fullerene and Zn-porphyrin moieties. The coordination of the fullerene ligands to the Zn-porphyrin was supported by UV-vis

Full text of DOI:10.1016/j.tet.2010.10.066

Tetrahedron 67 (2011) 228e235
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of new porphyrin/fullerene supramolecular assemblies: a spectroscopic
and electrochemical investigation of their coordination equilibrium in solution
a
,
b
c
b
b
*
ꢀ
Leandro J. Santos , Dayse CarvalhoDa-Silva , Julio S. Rebou c¸ as , Marcos R.A. Alves , Ynara M. Idemori ,
b
b
b
Tulio Matencio , Rossimiriam P. Freitas , Rosemeire B. Alves
a
Universidade Federal de Vi c¸ osadCampus de
ˇ
Florestal, Rodovia LMG, 818-km 6, FlorestaldMG, CEP: 35690-00
iencias Exatas, Universidade Federal de Minas Gerais, Avenida Antonio Carlos 6627, Belo HorizontedMG, CEP: 31270-901, Brazil
Departamento de QuímicadCentro de Ciencias Exatas e da Natureza, Universidade Federal da Paraíba, Campus I, Jo a~ o PessoadPB, CEP: 58059-900, CP: 5093, Brazil
ˇ
0, s/n, Minas Gerais, Brazil
b
c
Departamento de QuímicadInstituto de C
ˇ
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 August 2010
Received in revised form 17 October 2010
Accepted 22 October 2010
Available online 29 October 2010
Two new fullerene ligands have been designed to provide relatively simple frameworks to build su-
pramolecular systems containing both fullerene and Zn-porphyrin moieties. The coordination of the
fullerene ligands to the Zn-porphyrin was supported by UVevis titration, nuclear magnetic resonance
and electrochemical data. The resulting spectrophotometric data were processed both graphically and
computationally to yield the stoichiometry, stability constant, and molar absorptivity of the species in
equilibrium.
Keywords:
Ó 2010 Elsevier Ltd. All rights reserved.
Supramolecular complex
Zn-porphyrin
Fullerene-C60
SQUAD
1
. Introduction
of C60 in various steps in low-to-moderate yields. The prevailing
strategy for the synthesis of fullerene ligands is via cycloaddition
11,12
Porphyrin/fullerene systems and dyads have been extensively
reactions.
For example, D’Souza and co-workers have synthe-
1,2
studied to examine donor/acceptor interactions. Donor/acceptor
interactions within fullerene/porphyrin assemblies are importantfor
the design of multi-component model systems for photoactive ma-
terials and light-harvesting devices. Thus, chemists have in-
creasingly sought to develop relatively simple donor/acceptor
sized fullerene ligands via [3þ2] cycloadditions of azomethine ylides
1
3,14
and C60
.
More recently, Deye and co-workers have designed
15
a new ligand based on the DielseAlder functionalization of C60
.
A
3
4
less frequently used method to obtain fullerene ligands employs the
Bingel’s cyclopropanation reaction, which consists of the reaction of
5
16,17
systems. Since Liddell and co-workers reported the first synthesis of
active methylene compounds with C60
.
The search for new ful-
a C60 derivative covalently linked to a porphyrin, various other ful-
lerene/porphyrin dyads have been synthesized based on covalently
lerene/porphyrin supramolecular systems prompted us to design
new functionalized fullerene ligands via the BingeleHirsch cyclo-
propanation of C60 and to evaluate the ability of these new ligands to
assemble with a simple Zn-porphyrin, 5,10,15,20-tetraphenylpor-
phyrinatozinc(II) (Zn(TPP)).
6,7
linked components. Another attractive route to prepare such su-
pramolecular assemblies has been the coordination of appropriately
functionalized fullerene derivatives to metalloporphyrins, such as
those containing zinc8 or magnesium as the metal center. In the
latter route, the fullerene has often been functionalized with an N-
heteroaromatic moiety. This metal/ligand coordination strategy
provides a simple and versatile way to construct electron donor/ac-
ceptor assemblies, which may be of interest as mimics of the efficient
,9
10
The accurate determination of the equilibrium constants (K)
associated with dyad formation is often required in various
chemical, biochemical, and pharmaceutical settings. In previous
work, such constants for fullerene/porphyrin interactions have
1
18
8,10,13,14,16,17,19
been determined by H NMR, UVevis,
cence
and fluores-
7
9,10,17,18,21
biological electron transfer systems.
spectroscopies. Whereas spectrophotometry has
With respect to the synthetic aspects, most of the fullerene li-
gands reported to date have been prepared by the functionalization
been the method of choice due to its simplicity, low cost, ease of
adaptation, and high sensitivity (which allows the use of concen-
trations as low as 10 mol L ), it has also been limited to systems
with minimal chromophore absorbance overlap.
ꢀ
6
ꢀ1
In this contribution, we utilize a multiwavelength, multiple re-
gression procedure for the determination of the equilibrium
*
Corresponding author. Tel.: þ55 313536 3402; fax: þ55 31 34095700; e-mail
address: leandroj.santos@ufv.br (L.J. Santos).
0
040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.10.066

L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
229
constant of complex absorbing mixtures in Zn-porphyrin/fullerene
systems. Our protocol employs a computational approach rather
than the usual graphical methods for treating spectrophotometric
titration data; the software of choice was SQUAD (Stability Quo-
O
O
O
O
H3C
O
O
O
O
N
N
N
Ph
N
Ph
N
2
2
N
N
tients from Absorbance Data). This program is capable of calcu-
lating simultaneously, or individually, overall stability constants for
any species formed in studied systems, as long as the species
Ph
Zn
Ph
Ph
Zn
Ph
N
Ph
N
N
Ph
N
Zn(TPP)/6
Zn(TPP)/7
2
2
contributes to the total absorbance measured. Early studies
confirmed the ability of the program to evaluate the correctness of
the equilibrium model proposed and the values of the refined
equilibrium constants.2 To the best of our knowledge, fullerene/
porphyrin assembly coordination equilibrium data have been pro-
Fig. 1. Structures of the 1:1 supramolecular complexes investigated in the present
study.
2
bridge signal was observed at 52 ppm. The molecular composition
and the purity of the fullerene ligands 6 and 7 are also confirmed by
ultra-high-resolution Fourier-transform mass spectrometry with
17,20
cessed by only one computational (multiple regression) means.
electrospray ionization. Molecular masses of all isotopologue ions
2
2
. Results and discussion
.1. Synthesis
within the [MþH]þ clusters agree well with calculated values.
2.2. Spectrophotometric titration
The strategy explored for the preparation of the fullerene
derivatives 6, 7, and 8 was based on a Bingel-type reaction
(
Zn-porphyrins are well-established, suitable spectroscopic probes
for studying Lewis acid/base-type coordination equilibria in solution.
Scheme 1).2
3,24
O
O
1
OR2
R O
O
O
i
ii
OH
1
OR2
R O
HO
or
7
N
1
2
4
1
2
3
4
R = H3C ; R =
2
1
6
7
R = H3C ; R =
N
6
2
N
4
1
2
R = R =
1
2
R = R =
N
O
6
2
N
O
1
2
5
R = R =
1
2
7
8 R = R =
7
Scheme 1. Preparation of fullerene derivatives 6, 7, and 8. Reagents and conditions: (i) methyl malonyl chloride or malonyl dichloride, pyridine, THF, room temperature, 3 (2 h, 83%),
(5 h, 57%), 5 (5 h, 74%); ii) C60, I , DBU, toluene, room temperature, 8 h, 6 (50%), 7 (40%), 8 (50%).
4
2
The commercial alcohol 1 was esterified with commercial methyl
malonyl chloride or malonyl dichloride in pyridine/THF mixtures at
room temperature to yield the malonates 3 or 4 in 83% or 57% yield,
The Zn(TPP) UVevis absorption spectrum in hydrocarbon solvents,
such as toluene, is characterized by a strong Soret band (
¼423 nm,
in toluene) and a weaker Q band (
¼550 nm, in toluene). In the
l
l
respectively. The reaction of C60 with 3 or 4, I
2
, and DBU in toluene at
presence of a nitrogen/ligand (L), a rapid Lewis acid/base equilibrium
is established to yield the Zn(TPP)/L adduct (ML) (where M¼Zn(TPP)
and L¼ligand), according to Eq. 1. The formation of the ML species is
accompanied by a red-shift of the Soret and the Q bands in a con-
centration-dependent manner, which makes UVevis spectroscopy
a suitable technique to study such systems. The spectrophotometric
data are often analyzed graphically by a classical BenesieHildebrand
(BeH) plot, given that there are only two absorbing species in equi-
librium (i.e., M/ML or L/ML). Although this is usually the case for
simple pyridine derivatives, fullerene-functionalized pyridines (such
as 6 and 7) absorb in the UVevis region, which yields three chro-
mophores (M, L, and ML) simultaneously in equilibrium and pre-
cludes the direct use of the BeH-type of analysis. To overcome these
room temperature afforded mono-adducts 6 or 7 in 50% or 40% yield,
respectively, after purification by column chromatography. The ad-
vantages of this synthetic method for the preparation of the new
compounds 6 and 7 include the overall simplicity of the procedure
and the ready availability of the starting materials; additionally, this
methodology can be easily adapted for the synthesis of other func-
tionalized fullerenes. The fullerene derivative 8, which does not
contain the pyridyl moiety, was obtained from 1-octanol via the
same method used for the synthesis of 6 and 7 (Scheme 1).
The applicability of the resulting compounds in supramolecular
chemistry is demonstrated below by their ability to engage in co-
ordination-type assemblies with metal complexes in solution, e.g.,
fullerene/Zn-porphyrin adducts. As a proof of concept, a study was
undertaken to evaluate the ability of these new ligands 6 and 7 to
assemble with Zn(TPP), to yield the corresponding Zn-porphyrin/
fullerenes (Fig. 1) via the coordination route.
8
,10,17,18,20,21,26
analytical issues, previous studies
on the coordination
equilibrium of fullerenes to metalloporphyrins in solution are based
on fluorescence data instead of UVevis spectroscopy, as the fluo-
rescence of Zn(TPP) (M) is quenched in a concentration-dependent
manner by a non-fluorescent ligand (fullerene, L) to yield a non-
fluorescent adduct (ML). We show here that the supramolecular
assembly of functionalized fullerenes and Zn porphyrins may be
conveniently measured by UVevis spectrophotometry even in the
case of strongly absorbing fullerene species, given that the data are
appropriately treated by a multiwavelength spectrophotometric ti-
tration processing program, such as SQUAD. It is worth noting that, as
The structure and purity of the compounds were confirmed by
1
13
H and C NMR spectroscopy (the spectra of the fullerene ligands
13
are available in the Supplementary data). The C NMR spectra of
2
5
3
the fullerene ligands showed the characteristic signal of the sp -
hybridized fullerene C-atoms in the cyclopropane ring at ca.
7
1 ppm and the signals between 139 and 145 ppm corresponding
2
to the sp -hybridized carbons of the fullerene sphere. The methano

230
L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
opposed to spectrofluorometers, UVevis spectrophotometers re-
quire only simple setup and are widely available in most organic/
inorganic laboratories.
in the presence of three chromophores (i.e., M, L, and ML) simul-
taneously in equilibrium. The UVevis spectra of the spectropho-
tometric titration of Zn(TPP) with 8 (Fig. 2c), which contains only
the fullerene chromophore but no Py moiety, are superimposable
on the sum of the absorbances of the two individual species. This
observation shows that (i) the interaction between the fullerene
moiety and Zn(TPP) is negligible (if present at all), (ii) the assembly
of Zn(TPP)/6, and Zn(TPP)/7 takes place through and is primarily
controlled by the coordination of the Py moiety, and (iii) fullerene-
free compounds 3 and 4 (along with Py itself) represent suitable
models for 6 and 7.
L
Zn
Zn
(K)
ð1Þ
L
A set of UVevis spectrophotometric titrations was designed to
study the ability of 6 and 7 to assemble coordinatively with Zn(TPP)
in solution. Titrations with unsubstituted pyridine (Py), with ful-
lerene-free substituted pyridines 3 and 4, and with pyridine-free
fullerene 8 were carried out as controls for the supramolecular
assemblies of Zn(TPP)/6 and Zn(TPP)/7. The spectrophotometric
titrations of Zn(TPP) with the fullerene ligands 6 and 7 (Scheme 1)
were accompanied by the expected red-shifts of the Zn(TPP) Soret
and Q bands (Fig. 2a and b). However, the titrations did not show
the characteristic well-defined isosbestic points (Fig. 2a and b)
usually associated with the two-species M/ML equilibrium of Zn
The spectrophotometric titrations of Zn(TPP) with Py, 3, and 4,
which do not absorb in the Zn(TPP) Soret and Q band regions,
showed well-defined isosbestic points along with the expected
spectral red-shift (Fig. S1e3 in the Supplementary data). Spectro-
photometric data for these titrations were evaluated both by the
classic graphical BeH method and a computational approach using
the software SQUAD (see below). These analyses reveal the stoi-
2
7
chiometry and the formation constants (K) of the Zn-porphyr-
ineligand complex in solution.
(
TPP)/simple Py ligand systems because 6 and 7 themselves absorb
ꢀ
ꢁ
strongly in the titration region. As previously mentioned, fullerene-
functionalized pyridines absorb in the UVevis region, which results
A ꢀ A
0
log
¼ nlog½Lꢁ þ logðKÞ
(2)
A
N
ꢀ A
Fig. 2. Spectral shifts for the change of absorbance in the spectrophotometric titration titration in toluene. (a1ꢀa2) Zn(TPP)/6; (b1ꢀb2) Zn(TPP)/7; and (c1ꢀc2) Zn(TPP)/8. Con-
ditions: 7.93ꢂ10ꢀ mol L <[ligands]<2.73ꢂ10 mol L ; 1.23ꢂ10 mol L <[Zn(TPP)]<1.83ꢂ10 mol L
6
ꢀ1
ꢀ3
ꢀ1
ꢀ6
ꢀ1
ꢀ6
ꢀ1
.

L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
Table 2
231
The BeH method consists of the analysis of the changes in absor-
Parameters supplied by the SQUAD program for the spectrophotometric titration of
Zn(TPP) with Py, 3, 4, 6, and 7
bance (at a single wavelength) that accompany the change in
concentration of L at a fixed Zn(TPP) concentration, according to
System
log K
K (M 1
ꢀ
)
s
data
S (obsdꢀcalcd)
2
Eq. 2, where A
wavelength in the absence of L; A
the titration once A becomes independent of the concentration of L
and corresponds to the species Zn(TPP)/L ; A is the absorbance of
the titrated solution at any intermediate concentration between
those associated with A and A ; K is the formation constant; and n
0
is the initial absorbance of Zn(TPP) at a given
3
is the absorbance in the end of
Zn(TPP)/Py
3.686ꢃ0.003
4.853ꢂ10
0.004
0.01
N
a
(
3.62ꢃ0.01)
3
3
3
Zn(TPP)/3
Zn(TPP)/4
Zn(TPP)/6
3.646ꢃ0.002
3.930ꢃ0.001
3.683ꢃ0.007
4.426ꢂ10
8.511ꢂ10
4.819ꢂ10
0.003
0.003
0.006
0.009
0.008
0.03
n
a
0
N
(3.53ꢃ0.02)
3
is the number of ligands (L) in the complex.
Zn(TPP)/7
3.794ꢃ0.008
6.223ꢂ10
0.002
0.004
a
(3.71ꢃ0.02)
Table 1 shows the equilibrium parameters obtained for the
spectrophotometric titration of Zn(TPP) with Py, 3, and 4 by the
a
Analyses carried out independently in the visible region of the spectrum
(516e618 nm) of Zn(TPP) to yield Zn(TPP)/L species.
BeH method at two different wavelengths (
l
¼423 nm or 429 nm).
The experimental data relative to the slope (n) in Eq. 2 (Table 1)
indicate that the species M, L, and ML in the Zn(TPP)/Py, Zn(TPP)/3,
and Zn(TPP)/4 systems coexist in solution in an equilibrium state
governed by the formation constant K. However, the BeH analysis
resulting Zn(TPP)/L species. When the analyses of the Zn(TPP)/Py,
Zn(TPP)/6, and Zn(TPP)/7 were carried out in the visible region of
the spectrum (516e618 nm), the formation constants (K) calculated
were comparable to those obtained in the Soret region (Table 2).
This result indicates that the absorption band of fullerenes near the
Soret band of Zn(TPP) does not interfere with the analysis of this
spectral region by SQUAD. Convergence was not achieved for any
equilibrium model in the Zn(TPP) titration with 8, which is con-
sistent with the lack of any significant interaction between the
pyridine-free fullerene and Zn(TPP).
Good reproducibility in the equilibrium parameters that govern
the solution assembly of Zn(TPP) with Py, 3, 4, 6, and 7 was obtained
by the computational approach; the formation constants varied by
w5% with a second titration (replicate). This result demonstrates the
accuracy of both the experimental methodology and the data pro-
cessing and calculations by SQUAD. The formation constants found
at different wavelengths (
l
¼423 nm or 429 nm) returns K values of
up to one order of magnitude apart in the same titration. This
wavelength-dependence on the quantitative description of the
equilibrium constant, associated with the fact that the BeH analysis
is limited to ligands whose absorbance do not overlap considerably
with Zn(TPP), precluded the appropriate analysis of the Zn(TPP)
titration with 6 and 7 by the BeH method and prompted us to in-
vestigate the alternate multiwavelength computational approach
provided by SQUAD.
Table 1
Parameters supplied by the BeH method in analyzing the spectrophotometric
titration of Zn(TPP) with Py, 3, and 4
(
Table 2) are in an excellent agreement with those previously
System
l
¼423 nm
l
¼429 nm
reported for the interaction of Zn(TPP) with other related fullerenes
functionalized with N-containing moieties, for which different
methods of analysisof the experimental datawere used.
For the titration of Zn(TPP) by 7 an attempt to fit the experi-
mental data to a model including a 2:1 porphyrin/fullerene com-
plex was unsuccessful: no binding constant could be calculated
from the UV titration data as the chemical model including this
log K K (Mꢀ1)
N
R2
log K K (M
ꢀ1
)
n
R2
3
3
4
4
4
4
Zn(TPP)/Py 3.38 2.40ꢂ10
0.9 0.9931 4.46 2.88ꢂ10
1.0 0.9931 4.14 1.38ꢂ10
1.1 0.9983 4.37 2.34ꢂ10
1.2 0.9995
1.1 0.9992
1.1 0.9991
13,14,16,18,19,20
Zn(TPP)/3
Zn(TPP)/4
3.65 4.47ꢂ10
4.16 1.46ꢂ10
2
R ¼coefficient of linearity.
2
M L species precluded the minimization and convergence process
A systematic analysis of the spectrophotometric titration data of
of the regression spectral analysis. This is likely ascribed to either
the 2:1 complex not being formed at all or because the conditions
for its formation are not accessible within the concentration limits
of the spectrophotometric titrations. It is unreasonable to assume
that only one pyridine unit of 7 coordinates to the zinc porphyrin,
but considering that Zn(TPP) forms exclusively pentacoordinate
Zn(TPP) with the various ligands Py, 3, 4, 6, 7, or 8 was undertaken
using the software SQUAD, a well-established Fortran-based data
processing software22 capable of executing calculations of the best
values of the formation constants for a proposed equilibrium
model. This software uses a nonlinear least squares approach to
simultaneously fit absorption data at multiple wavelengths for
solutions of varying ligand concentration.
The formation constants and the statistical parameters gener-
ated by SQUAD for the proposed M/L/ML equilibrium model for the
titrations of Zn(TPP) with the various ligands are summarized in
Table 2. In all calculations using SQUAD, the input file contained
data of 25 spectra per titration; a total of 50 wavelengths per
spectrum (i.e., 1250 data points per titration) were simultaneously
examined in the region of the Soret band of Zn(TPP) and the
2
8
stable complexes at room temperature, the two pyridine-moiety
nitrogen atoms of 7 cannot bind simultaneously to the Zn por-
phyrin, as this would result in a hexacoordinated complex.
Therefore, a dynamic, intramolecular equilibrium is likely to be
established for the ‘two-point’-bonded 1:1 Zn(TPP)/7 complex as
shown in Scheme 2. It is worth noting that a ‘two-point’ binding
motif has been suggested for systems as such by Li and co-
2
9
workers.
O
O
O
N
N
O
O
O
N
N
N
N
Zn
Zn
N
N
N
N
O
N
N
O
Scheme 2. Dynamic process for the proposed ‘two-point’ binding mode for the complexes of 7 with Zn(TPP) guest.

232
L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
SQUAD calculates both the formation constants and, if reques-
2.3. NMR titration
ted, the molar absorptivity of the species in equilibrium. A signifi-
cant improvement of 26e65% in the statistical parameters was
achieved by fixing the molar absorptivity of L to the experimental
values and letting SQUAD calculate the molar absortivities of both
M and ML (as opposed to fixing both L and M). The agreement
between the molar absorptivity of Zn(TPP) determined experi-
mentally and that calculated by SQUAD in each titration model
directly from the titration data (Fig. 3) represents another means to
validate the calculations performed by SQUAD. Once the constants
for the modeled chemical equilibrium have been determined, the
emulation of the experimental absorption spectra, including the
molar absorptivity coefficients at each wavelength, becomes pos-
sible via data processing using SQUAD (Fig. 3). It is also worth
noting that the access to the molar absorptivity of all of the species
involved in the solution equilibrium is not granted by the BeH
method. Fig. 3 shows the molar absorptivity spectra of all of the
The complexation between Zn(TPP) and the fullerene ligands 6
and 7 was also demonstrated by NMR titration. Fig. 4 shows the H
1
NMR spectral data for Zn(TPP) at various concentrations of 6. A
significant shielding of the protons of the ligand was observed upon
coordination to Zn(TPP). The 2,6-pyridine protons, 4-pyridine
proton and 5-pyridine proton showed up as broad singlets shifted
upfield as compared to that of unbound ligand 6. Upon addition of
1 equiv of ligand 6 (Fig. 4c), the 2,6-pyridine protons experienced
a shielding of nearly 4.5 ppm due to the complexation; the 4-pyr-
idine proton and 5-pyridine proton located at 7.27 and 7.56 ppm in
the unbound ligand were shifted to 6.54 and 5.92 ppm, re-
2
spectively; and the CH O protons experienced a shielding of
0.5 ppm. The greater shift observed for the pyridine protons rela-
tive to other protons of 6 can be rationalized by the greater prox-
imity of the axially coordinated Py moieties to the porphyrin plane,
which results in a greater shielding by the porphyrin ring
p-cur-
17
rents. No peaks for unbound ligand 6 were observed under these
conditions. With increasing concentration of ligand in the mixture,
the spectral pattern approximated that of the free ligand 6, as the
Fig. 3. Molar absorptivity of the species calculated by SQUAD (M, ML and L) or ex-
perimentally determined (M). Titration in toluene of Zn(TPP) with either 6 (top) or 7
(
bottom).
chromophore species present in the Zn(TPP) titration with 6 and 7,
including those of the Zn(TPP)/6 and Zn(TPP)/7 assemblies. The
spectra calculated by SQUAD for all Zn(TPP)/L species (L¼Py, 3, 4, 6,
and 7) in equilibrium were very similar to each other, which is
consistent with Zn(TPP) being the chromophore of higher molar
absorptivity in theses species; the molar absorptivity of the ful-
lerenes 6 and 7 calculated by SQUAD agrees with that measured
experimentally and is one order of magnitude lower than that of Zn
(
TPP). The similarity of the molar absorptivity of the Zn(TPP)/L
species (L¼Py, 3, 4, 6, and 7) is also consistent with the Zn(TPP)/L
binding motif in all ML species being essentially the same (i.e.,
coordination via an unencumbered pyridine moiety).
The formation of the supramolecular systems Zn(TPP)/6 and Zn
TPP)/7 is also supported by H nuclear magnetic resonance (NMR)
1
(
Fig. 4. 1H NMR spectrum of (a) Zn(TPP); (b) 6; (c) Zn(TPP)/6 (1:0.5); (d) Zn(TPP)/6
and electrochemical data.
3
(1:0.75); (e) Zn(TPP)/6 (1:1); (f) Zn(TPP)/6 (1:1.5); (g) Zn(TPP)/6 (1:2.5) in CDCl .

L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
233
M/L/ML rapid equilibrium in Zn-porphyrin/nitrogen-base systems
Table 3
3
0
E1/2 versus Ag/AgNO3 for C60 and fullerene ligands 6 and 7
is much faster than the NMR time-scale.
1
The H NMR titration of Zn(TPP) by ligand 7 presented the same
pattern as of that of 6, up to the 1:1 M ratio of Zn(TPP) and 7 (Fig. S5
in the Supplementary data). With increasing concentration of 7 in
the mixture, there happens an unfolding of pyridine signs, which
may originate from the stabilization of the non-symmetric system
suggested by the dynamic equilibrium illustrated in Scheme 2.
Compounds
E1/2 versus Ag/AgNO3
/V
E
1
E
2
/V
3
E /V
C
6
7
60
ꢀ1.54
ꢀ1.77
ꢀ1.65
ꢀ1.00
ꢀ1.27
ꢀ1.14
ꢀ0.55
ꢀ0.87
ꢀ0.75
Fig. 6 shows the cyclic voltammograms of Zn(TPP) and the Zn
ꢀ
1
2
.4. Electrochemical studies
(TPP)/6 and Zn(TPP)/7 assemblies at 50 mV s . For Zn(TPP)
Fig. 6a), the presence of two oxido-reduction processes can be
ascribed to the formation of a -cation radical (0.57 V vs Ag/AgNO )
(
The investigation of the electrochemical properties of the ful-
p
3
15
lerene ligands and their corresponding assemblies with Zn(TPP)
was done by cyclic voltammetry. The study was aimed at gathering
information on the redox potentials of the compounds and the
possible electronic interaction between the donor (Zn(TPP)) and
acceptor (fullerene in ligands 6 and 7) moieties of the Zn(TPP)/6
and Zn(TPP)/7 assemblies in solution.
Fig. 5 shows the cyclic voltammograms of C60, 6, and 7 recorded
under the same conditions for comparison purposes, while Table 3
lists the first three E1/2 values associated with these species. The
cyclic voltammogram of 6 and 7 is characterized by a cathodic shift
and a dication radical (0.88 V vs Ag/AgNO3). Zn(TPP)/6 (Fig. 6a)
and Zn(TPP)/7 (Fig. 6c) have an oxido-reduction process at 0.58 V
and 0.62 V, respectively, in which the potential values of the
peak-to-peak (anodic and cathodic) separation and the cathodic-
to-anodic peak current ratio indicated an electrochemically quasi-
reversible behavior. The coordination of 6 with Zn(TPP) (Fig. 6b) is
accompanied by a significant positive shift of the redox potentials,
indicating the presence of interactions between the porphyrin
system (donor) and the fullerene core (acceptor).
p-
of these potentials relative to the corresponding processes in C60
.
An irreversible anodic peak at ꢀ0.34 V and ꢀ0.19 V for 6 and 7,
respectively, was also observed.
5
0µA
(a)
Zn(TPP)
(
a)
50 A
6
(b)
Zn(TPP)/6
5
0µA
(
b)
50 A
(c)
7
Zn(TPP)/7
5
0µA
-2,0
-1,5
-1,0
-0,5
E/V
0,0
0,5
1,0
(c)
Fig. 6. Cyclic voltammogram of (a) Zn(TPP) and (b and c) fullerene-free ligands and
ꢀ
1
complexes. Scan rate: 50 mV s
.
50 A
Conversely, the coordination of 7 with Zn(TPP) did not result in
any significant change in the redox potential of the C60 moiety. This
result implies an absence of
pep charge transfer interactions be-
-2.0
-1.5
-1.0
E/V
-0.5
0.0
tween theporphyrin -systemand the fullerenecore intheZn(TPP)/7
p
assembly (Fig. 6c), which suggests a rapid coordination equilibrium
ꢀ
1
between the two pyridyl moieties of the fullerene derivative 7 and Zn
(TPP) (Scheme 2). This results in low-to-nonexistent pep charge
4 4
Fig. 5. Cyclic voltammogram of (a) C60, (b) 6 and (c) 7 in 0.1 mol L of Bu NBF o-
dichlorobenzene/N,N-dimethylformamide (1:1). Scan rate: 50 mV s
ꢀ
1
.

234
L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
transfer interactions between the porphyrin
p-system and 7 within
spectroscopy (FT-MS) was performed with an LTQ FT ULTRA (The-
the supramolecular complex Zn(TPP)/7 (Scheme 2).
moScientific-7 T-Germany) instrument with the TriVersa NanoMate
system (Advion, USA) operating in the chip-based infusion mode
and in the positive ion mode using a silicon-based integrated
nanoelectrospray microchip. Column chromatography was carried
out using silica gel 60, 70e230 mesh (Merck). UVevis spectra
(190e1100 nm) were recorded in a HewlettePackard 8453 diode-
array spectrophotometer. Multiwavelength spectrophotometric
data analyses were carried out on a Pentium IV computer using
3
. Conclusions
We have shown here that new fullerene ligands can be easily
prepared following a Bingel-type cyclopropanation reaction, and
we have demonstrated the applicability of these ligands in supra-
molecular chemistry by their coordinative assembly with a Zn-
porphyrin. The coordination equilibrium was investigated by
UVevis and NMR titrations and by electrochemical means.
The formation constants relative to the axial coordination of Zn
2
2
SQUAD software.
4.2. General procedures for the synthesis of fullerene
derivatives
(
TPP) with the new fullerene ligands were determined spectropho-
tometrically and conveniently calculated by using the program
SQUAD. The analysis of multiwavelength data by SQUAD is often
significantly more robust than examination of a single wavelength
To a solution of C60 (0.35 g, 0.48 mmol) in toluene (350 mL)
were added iodine (0.18 g, 0.72 mmol), malonate 3, 4, or 5
(
BꢀH method) and was used here for the analysis of porphyrin/ful-
(0.48 mmol), and diaza(1,3)bicyclo[5.4.0]undecane (DBU) (180 mL,
lerene dyads for the first time. Additionally, SQUAD analysis allows
the determination of the pure spectra of all of the species and in-
termediates in the equilibrium system. These results show that
analyses of multiwavelength data are particularly promising for
studying the nature, stoichiometry, and spectral properties of com-
plex assemblies of fullerenes and metalloporphyrins in solution.
The formationconstants forthe interaction between Py, 3, 4, 6, or 7
withZn(TPP)are closetoeachother(Table 2), whichindicatesthatthe
assembly of the adducts in solution is controlled primarily by the
pyridine moietyand is little influenced by the fullerene in 6 and 7. It is
worth noting that in the absence of any synergistic effect between the
porphyrin and the fullerene moiety, the relatively low values for the
formation constants of Zn-porphyrin/N-heteroaromatic-functional-
ized fullerene dyads imply that the isolated supramolecular assem-
1.2 mmol). The reaction mixture was stirred for 8 h at room tem-
perature, under nitrogen atmosphere, and then filtered. The solvent
was removed under reduced pressure, and the residue was purified
by column chromatography, eluting first with toluene (to remove
unreacted C60). The pure mono-adduct was isolated by eluting with
30% MeOH in EtOAc (for compounds 6 and 7) or with 5% AcOEt in
toluene (for compound 8) to provide the fullerene derivatives 6, 7
or 8 in 50%, 40% and 50% yield, respectively.
4.2.1. Compound 6. IR (compression cell):
1540, 1428, 1740, 1231, 1186. H NMR (CDCl
n
3022, 3002, 2951, 2853,
): 2.20 (pseudo qn, 2H,
), 4.55
O), 7.27 (br s, 1H, H-Py), 7.58 (d, 1H, J¼7.8 Hz, H-
1
3
d
CH
(t, 2H, J¼6.2 Hz, CH
Py), 8.50e8.52 (m, 2H, H-Py). C NMR (50 MHz, CDCl
(2CH ), 52.0 (methano bridge), 54.2 (OCH ), 66.3 (CH
2
CH
2
O), 2.85 (t, 2H, J¼7.2 Hz, CH
2 3
Py), 4.13 (s, 3H, OCH
2
13
3
):
d 29.4, 30.0
31
blies (ML) may not be of particular relevance as cellular sensitizers
2
3
2
O), 71.5 (2C60-
3
given that once in solution, the equilibrium is shifted to the individual
constituents M and L under biologically relevant conditions (i.e., at
sp ),123.7,136.0,136.1,148.0,150.0 (5C-Py),139.0,139.3,141.1,141.3,
142.0, 142.5, 143.5, 144.4, 144.8, 145.0, 145.1, 145.3, 145.4 (58C-ful-
ꢀ
5
ꢀ1
total stoichiometric concentrations lower than 10 mol L , ML
lerene), 163.6 (C]O), 164.1 (C]O). C72
4
H13NO . UVevis in toluene,
constitutes <10% of the M/L/ML mixture). Additionally, whereas the
l
max, nm (log
3
): 428 (3.38), 493 (3.16). Exact mass calculated for
ꢀ
3
ꢀ1
þ
use of more concentrated solutions (>10 mol L ) may warrant the
formation of the Zn(TPP)/fullerene adduct in considerable amount, at
concentrated solutions other processes such as association between
ground-state molecules, which gives rise to dimers and higher ag-
gregates, and interactions between excited and ground-state of the
molecules may take place and limit, thus, the use of these assemblies
as photoactive materials.32
Additionally, the lack of significant steric hindrance between the
two very bulky and sterically demanding moieties (i.e., the por-
phyrin and the fullerene moieties) in the dyads is granted by the
flexibility of the malonate chains, which function as an appropriate
pyridine-fullerene linker for the design of unencumbered porphy-
rin/fullerene assemblies. The easy access to malonate-functional-
ized fullerene derivatives via the Bingel-type chemistry as
described here provides a convenient method for the preparation of
other more elaborate fullerene-based ligands for the synthesis of
tailored porphyrin-based donor/acceptor dyads.
[MþH] (M¼956.08446). Found FT-MS ESI(þ): m/z 956.09276.
4.2.2. Compound 7. IR (compression cell):
n
3027, 2957, 2921, 2860,
2.18 (pseudo qn, 4H, CH CH O),
Py), 4.55 (t, 4H, J¼6.2 Hz, CH O), 7.22e7.28
(m, 2H, H-Py), 7.54 (d, 2H, J¼7.6 Hz, H-Py), 8.50 (br s, 4H, H-Py).
NMR (50 MHz, CDCl ): 29.5, 30.0 (4CH ), 52.2 (methano bridge),
66.4 (2CH O), 71.6 (2C60-sp ), 123.8, 136.8, 147.9, 149.8 (10C-Py),
139.1, 141.2, 142.0, 142.3, 143.2, 144.0, 144.7, 144.8, 145.1, 145.2, 145.4
(58C-fullerene), 163.7 (2C]O). C79 . UVevis in toluene,
): 428 (3.30), 493 (3.07). Exact mass calculated for
1
1726, 1148, 1022. H NMR (CDCl
2.82 (t, 4H, J¼7.2 Hz, CH
3
):
d
2
2
2
2
13
C
3
d
2
3
2
20 2 4
H N O
lmax, nm (log
3
þ
[MþH] (M¼1061.14231). Found FT-MS ESI(þ): m/z 1061.15122.
4.2.3. Compound 8. IR (compression cell):
n
2951, 2920, 2850, 1742,
0.79e0.82 (m, 6H, CH ),
), 1.53 (br s, 4H, CH CH CH O),
O), 4.42 (t, 4H, J¼6.6 Hz, CH O). C NMR
14.3 (2CH ), 22.8, 26.1, 28.7 (8CH (CH CH ),
O), 32.0 (2CH CH O), 52.5 (methano bridge), 67.6
O), 71.8 (2C60-sp ),139.1,141.0,142.0,142.3,143.1,144.0,144.7,
144.8, 145.0, 145.3, 145.4, 145.5 (58C-fullerene), 163.8 (2C]O).
1
1539, 1427, 1226, 1185. H NMR (CDCl
1.22 (br s, 16H, CH (CH CH
1.70e1.81 (m, 4H, CH CH
50 MHz, CDCl ):
29.4 (2CH CH CH
2CH
3
):
d
3
3
2
)
4
2
2
2
2
1
3
2
2
2
(
3
d
3
3
2
)
4
2
4
4
. Experimental
2
2
2
2
2
3
(
2
.1. Materials and methods
ꢀ
1
C
79
34
H O
4
(1046.00 g mol ). UVevis in toluene,
lmax, nm (log 3):
Reagents and solvents were of reagent grade and used without
428 (4.39), 476 (4.20).
further purification unless stated otherwise. Tetrahydrofuran (THF)
was distilled over sodium benzophenone. C60 (99.5%) was obtained
from M.E.R. Corporation. Preparation and characterization data of
the malonates 3, 4, and 5 are described in the Supplementary data.
4.2.4. Zn(TPP). Zn(TPP) was prepared by metalation of H
maeAldrich) with Zn(OAc) $2H O using the standard chloroform/
Briefly, to a refluxing solution of H TPP
(83.27 mg, 0.135 mmol) in CH Cl (20 mL) wasaddedZn(OAc) $2H
(147.76 mg, 0.67 mmol) in MeOH (5 mL). The resulting solution was
kept under reflux for 1 h. After this time, the solvent was evaporated,
2
TPP (Sig-
2
2
3
3
methanol method.
2
NMR spectra were recorded in CDCl
3
on a Bruker Avance DPX-200
200 MHz) spectrometer and referenced to either the residual CDCl
7.26) or tetramethyl silane (TMS) protons. Fourier-transform mass
2
2
2
2
O
(
(
3
d

L.J. Santos et al. / Tetrahedron 67 (2011) 228e235
235
and the residue was purified by column chromatography on neutral
alumina. The column was eluted with CH Cl . The purple fraction
corresponding to Zn(TPP) was collected and evaporated to dryness.
Supplementary data
2
2
Synthetic details of the malonates 3, 4, and 5. Spectral shifts,
molar absorptivity of species calculated by SQUAD (M and ML) or
experimentally determined (M) and BeH plot for change of absor-
The resulting purple solid was kept over P
2
O
5
under vacuum in
3): 423 (5.68), 550
a desiccator. UVevis in toluene, max, nm (log
l
1
1
(
8
4.40). H NMR (CDCl
3
, TMS):
d
7.77e7.79 (m; 12H; m- and p-HPh),
-pyrrole).
bance in the systems Zn(TPP)/Py, Zn(TPP)/3, and Zn(TPP)/4. H NMR
1
13
.22e8.24 (m; 8H; o-H Ph), 8.95e8.96 (m; 8H;
b
spectra of the titration of ZnTPP by ligand 7. H and C NMR spectra
of the new fullerene derivatives 6, 7, and 8. Supplementary data as-
sociated with this article can be found in online version at
doi:10.1016/j.tet.2010.10.066. These data include MOL files and
InChIKeys of themost important compounds describedin this article.
4
.3. Spectrophotometric titration
Spectrophotometric titrations were performed in a borosilicate
glass cuvette tightly capped with a Teflon-coated silicon septum us-
ꢄ
ing a 25.0ꢃ0.1 C thermostatized cell-holder. All stock and working
References and notes
solutions were kept in the dark fully wrapped with aluminum foil.
Additionally, thetitrationwascarried out withminimalambientlight
exposure, and the spectrophotometer shutter was kept closed in
between measurements. The initial concentration of the toluene
solutions of Zn(TPP) in the cuvette was determined spectrophoto-
1. Clifford, J. N.; Accorsi, G.; Cardinali, F.; Nierengarten, J.-F.; Armaroli, N. C. R.
Chim. 2006, 9, 1005.
2
3
4
. (a) Imahori, H. Org. Biomol. Chem. 2004, 2, 1425; (b) Ito, O.; Yamanaka, K.-I. Bull.
Chem. Soc. Jpn. 2009, 82, 316.
. Marczak, R.; Sgobba, V.; Kutner, W.; Gadde, S.; D’Souza, F.; Guldi, D. M. Lang-
muir 2007, 23, 1917.
. Guldi, D. M. Chem. Soc. Rev. 2002, 31, 22.
ꢀ
6
ꢀ1
metrically; typical values were from 1.23e1.83ꢂ10 mol L . Ali-
quots of the toluene stock solutions of the titrating ligands (Py, 3, 4, 6,
5. Liddel, P. A.; Sumida, J. P.; Macpherson, A. N.; Noss, L.; Seely, G. R.; Clark, K. N.;
Moore, A. L.; Moore, T. A.; Gust, D. Photochem. Photobiol. 1994, 60, 537.
7
, or 8) were added consecutively to the cuvette through the cuvette
silicon septum with Hamilton microsyringes; immediately prior to
recording the UVevis spectrum, the system was magnetically stirred
6
. (a) Fazio, M. A.; Lee, O. P.; Schuster, D. I. Org. Lett. 2008, 10, 4979; (b) Iehl, J.;
Osinska, I.; Louis, R.; Holler, M.; Nierengarten, J.-F. Tetrahedron Lett. 2009, 50,
2245; (c) Urbani, M.; Nierengarten, J.-F. Tetrahedron Lett. 2007, 48, 8111; (d)
Schuster, D. I.; Li, K.; Guldi, D. M.; Palkar, A.; Echegoyen, L.; Stanisky, C.; Cross, R. J.;
Niemi, M.; Tkachenko, N. V.; Lemmetyinen, H. J. Am. Chem. Soc. 2007, 129, 15973;
ꢄ
for 1 min at 25.0ꢃ0.1 C to allow for thermal and chemical equili-
bration after the addition of each ligand aliquot. The total concen-
(
e) Fukuzumi, S.; Ohkubo, K.; Imahori, H.; Shao, J.; Ou, Z.; Zheng, G.; Chen, Y.;
Pandey, R. K.; Fujitsuka, M.; Ito, O.; Kadish, K. M. J. Am. Chem. Soc. 2001,123,10676.
7. Fukuzumi, S.; Kojima, T. J. Mater. Chem. 2008, 18, 1427.
ꢀ
6
ꢀ1
trationof the ligandsinthe cuvette rangedfrom7.93ꢂ10 mol L to
ꢀ3
ꢀ1
2
.73ꢂ10 mol L . The end of the titration was determined when
8. D’Souza, F.; Smith, P. M.; Rogers, L.; Zandler, M. E.; Islam, D.-M. S.; Araki, Y.; Ito,
UVevis spectral variations ceased. Dilution effects along the titration
O. Inorg. Chem. 2006, 45, 5057.
were accounted for when calculating the total concentration of Zn
TPP) and ligand at each titration point. The mathematical treatment
of the absorbance versus concentration data was carried out either
graphically (via BenesieHildebrand plots) or computationally (us-
9. Tat, F. T.; Zhou, Z.; MacMahon, S.; Song, F.; Rheingold, A. L.; Echegoyen, L.;
Schuster, D. I.; Wilson, S. R. J. Org. Chem. 2004, 69, 4602.
(
10. D’Souza, F.; Zandler, M. E.; Gadde, S.; McCarty, A. L.; Karr, P. A.; El-Khouly, M. E.;
Araki, Y.; Ito, O. J. Phys. Chem. B 2005, 109, 10107.
2
2
11. Yin, G.; Xu, D.; Xu, Z. Chem. Phys. Lett. 2002, 365, 232.
ing the software SQUAD).
12. Uyar, Z.; Satake, A.; Kobuke, Y.; Hirota, S. Tetrahedron Lett. 2008, 49, 5484.
1
1
3. D’Souza, F.; Deviprasad, G. R.; Rahman, M. S.; Choi, J.-P. Inorg. Chem.1999, 38, 2157.
4. D’Souza, F.; Deviprasad, G. R.; Zandler, M. E.; El-Khouly, M. E.; Fujitsuka, M.; Ito,
O. J. Phys. Chem. A 2003, 107, 4801.
4
.4. Electrochemical experiments
15. Deye, J. R.; Shiveley, A. N.; Goins, S. M.; Rizzo, L.; Oehrle, S. A.; Walters, K. A.
Inorg. Chem. 2008, 47, 23.
6. Wu, Z.-Q.; Li, C.-Z.; Feng, D.-J.; Jiang, X.-K.; Li, Z.-T. Tetrahedron 2006, 62, 11054.
17. Trabolsi, A.; Urbani, M.; Delgado, J. L.; Ajamaa, F.; Elhabiri, M.; Solladi eꢀ , N.;
The electrochemical experiments were carried out using an
AUTOLAB (PGSTAT 302) ECOChemie potentiostategalvanostat. The
1
Nierengarten, J.-F.; Albrecht-Gary, A.-M. New J. Chem. 2008, 32, 159.
18. Armaroli, N.; Diederich, F.; Echegoyen, L.; Habicher, T.; Flamigni, L.; Marconi, G.;
Nierengarten, J. F. New J. Chem. 1999, 23, 77.
19. (a) D’Souza, F.; Smith, P. M.; Gadde, S.; McCarty, A. L.; Kullman, M. J.; Zandler, M. E.;
Itou, M.; Araki, Y.; Ito, O. J. Phys. Chem. B 2004, 108, 11333; (b) D’Souza, F.; Devi-
pradad, G. R.; El-Khouly, M.E.; Fujitsuka, M.; Ito, O. J. Am. Chem.Soc. 2001,123, 5277;
electrochemical experiments were carried out in a typical three-
ꢀ2
2
electrode cell; a vitreous carbon disk electrode (A¼2.8ꢂ10 cm )
was used as the working electrode, a platinum wire was used as the
counter electrode, and a Ag/0.1 M AgNO
electrode in acetonitrile. The electrochemical profile was recorded in
:1o-dichlorobenzene/N,N-dimethylformamide solutionsofZn(TPP)
3
served as the reference
(c) D’Souza, F.; Devipradad, G. R.; El-Khouly, M. E.; Fujitsuka, M.; Ito, O.; Klykov, A.;
1
VanStipdonk, M.; Perera, A. J. Phys. Chem. A 2002, 106, 3243; (d) Wilson, S. R.;
MacMahon, S.; Tat, F. T.; Jarowski, P. D.; Schuster, D. I. Chem. Commun. 2003, 226.
20. Trabolsi, A.; Elhabiri, M.; Urbani, M.; de la Cruz, J. L. D.; Ajamaa, F.; Solladi eꢀ , N.;
ꢀ1
ꢀ1
(
0.002 mol L ), fullerene or fullerene-free ligand (0.002 mol L ),
ꢀ
1
andtetrabutylammoniumtetrafluoroborate(Bu
4
NBF
4
, 0.1 molL )at
Albrecht-Gary, A.-M.; Nierengarte, J.-F. Chem. Commun. 2005, 5736.
room temperature under nitrogen atmosphere.
.5. NMR titration
The coordination of ligand 6 and 7 to Zn(TPP) was monitored
2
1. Hauke, F.; Swartz, A.; Guldi, D. M.; Hirsch, A. J. Mater. Chem. 2002, 2088.
22. Leggett, D. J.; Kelly, S. L.; Shiue, L. R.; Wu, Y. T.; Chang, D.; Kadish, K. M. Talanta
1
983, 30, 579.
4
23. Bingel, C. Chem. Ber. 1993, 126, 1957.
24. Nierengarten, J.-F.; Habicher, T.; Kessinger, R.; Cardullo, F.; Diederich, F. Helv.
Chim. Acta 1997, 80, 2238.
1
25. Diederich, F.; Isaacs, L.; Philp, D. Chem. Soc. Rev. 1994, 243.
spectroscopically by H NMR. To a CDCl
(
3
solution of Zn(TPP)
6.50 mmol L ) were added appropriate amounts of 6 directly into
the NMR tube to yield total concentrations of 6 within the
26. Xiang, X.; Wei, X.-W.; Zhang, X.-M.; Wang, H.-L.; Wei, X.-L.; Hu, J.-P.; Yin, G.; Xu,
ꢀ
1
Z. Inorg. Chem. Commun. 2006, 9, 452.
27. Benesi, H. A.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71, 2703.
ꢀ
1
28. Smith, K. M. Porphyrins and Metalloporphyrins; Elsevier: Amsterdam, 1975.
2
.95e10.80 mmol L range; upon each addition the corresponding
spectrum was recorded. A analogous titration was carried out using
instead of 6 (data are given in the Supplementary data).
29. Li, C. Z.; Zhu, J.; Wu, Z. Q.; Hou, J. L.; Li, C.; Shao, X. B.; Jiang, X. K.; Li, Z. T.; Gao,
X.; Wang, Q. R. Tetrahedron 2006, 62, 6973.
7
30. Kirskey, C. H.; Hambright, P.; Storm, C. B. Inorg. Chem. 1969, 8, 2141.
3
1. (a) dos Santos, L. J.; Alves, R. B.; de Freitas, R. P.; Nierengarten, J.-F.; Magalh ~a es,
L. E. F.; Krambrock, K.; Pinheiro, M. V. B. J. Photochem. Photobiol., A 2008, 200,
277; (b) Yamakoshi, Y.; Umezawa, N.; Ryu, A.; Arakane, K.; Miyata, N.; Goda, Y.;
Masumizu, T.; Nagano, T. J. Am. Chem. Soc. 2003, 125, 12803; (c) Wang, S.; Gao,
R.; Zhou, F.; Selke, M. J. Mater. Chem. 2004, 14, 487.
2. (a) Zinimermanc, A. A.; Orlando, J. R.; Gianni, M. H. J. Org. Chem. 1969, 34, 73; (b)
Mochalov, I. A.; Lapshin, A. N.; Nadtochenki, V. A.; Smirnov, V. A.; Goldshleger,
N. F. Russ. Chem. Bull. 2006, 55, 1598.
3. (a) Wijesekera, T. P.; Dolphin, D. A. In Metalloporphyrins in Catalytic Oxidations;
Sheldon, R. A., Ed.; MarcelDekker: New York, NY,1994; pp 193e239; (b) Rebou c¸ as, J.
S.; de Carvalho, M. E. M. D.; Idemori, Y. M. J. Porphyrins Phthalocyanines 2002, 6, 50.
Acknowledgements
Financial support from The Brazilian Research Council (CNPq)
and Funda c¸ ~a o de Amparo aꢁ Pesquisa do Estado de Minas Gerais
3
(
FAPEMIG) is gratefully acknowledged. We also thank Laborat oꢀ rio
3
ThoMSon de Espectrometria de Massas (Universidade Estadual de
Campinas, SP-Brazil) for the mass spectrometric analyses.