Zinc porphyrin-re(I) bipyridyl-fullerene triad: Synthesis, characterization, and kinetics of the stepwise electron-transfer processes initiated by visible excitation

Article abstract of DOI:10.1021/ic502430e

A new triad system featuring one zinc porphyrin and one fullerene moieties attached to a central redox-active Re(I) connector was obtained in remarkable yield by cleverly exploiting a facile two-step synthesis. Detailed description and discussion on the characterization of this multicomponent system and of its parent free-base analogue are presented, along with a kinetic study of the stepwise electron-transfer processes occurring upon visible excitation.

Full text of DOI:10.1021/ic502430e

Article
pubs.acs.org/IC
Zinc Porphyrin−Re(I) Bipyridyl−Fullerene Triad: Synthesis,
Characterization, and Kinetics of the Stepwise Electron-Transfer
Processes Initiated by Visible Excitation
Paolo Cavigli,† Tatiana Da Ros,† Axel Kahnt,¶ Marta Gamberoni,§ Maria Teresa Indelli,*,§
and Elisabetta Iengo*,†
†Department of Chemical and Pharmaceutical Sciences, University of Trieste, Via L. Giorgieri 1, 34127 Trieste, Italy
¶Lehrstuhl fur Physikalische Chemie I, Friedrich-Alexander-Universitat Erlangen-Nurnberg, Egerlandstraße 3, 91058 Erlangen,
̈
̈
̈
Germany
§Department of Chemical and Pharmaceutical Sciences, University of Ferrara, Via Fossato di Mortara 17, 44121 Ferrara, Italy
S
* Supporting Information
ABSTRACT: A new triad system featuring one zinc porphyrin and one fullerene
moieties attached to a central redox-active Re(I) connector was obtained in
remarkable yield by cleverly exploiting a facile two-step synthesis. Detailed
description and discussion on the characterization of this multicomponent system
and of its parent free-base analogue are presented, along with a kinetic study of
the stepwise electron-transfer processes occurring upon visible excitation.
demonstrated to be very versatile as it allows the stepwise
replacement of the three dmso-O with the desired ligands,
under mild conditions.6
INTRODUCTION
■
Artificial photosynthesis, that is, the use of light energy to
produce solar fuels, is a timely research field.1 Photochemical
water splitting, as an ideal energy-storing reaction, is the focus
of worldwide research for applications in artificial photosyn-
thesis; at the same time, large efforts are being devoted
worldwide to the development of efficient molecular devices for
photocatalytic reduction of CO2.2 The chemical conversion of
CO2 has been studied intensively by several groups with
particular attention to the use of rhenium polypyridine
photocatalysts.3 The main drawback of these photocatalysts is
the lack of absorption into the visible region. One of the major
strategy to solve this problem is to build multicomponent
conjugates containing one visible-light absorber unit connected
to the rhenium catalyst. This approach has been investigated in
recent years, in particular by the group of Ishitani4a−c and of
For the dyad reported in Chart 1, a detailed photophysical
investigation has revealed the occurrence of a fast photoinduced
electron transfer from the singlet excited state of the zinc-
porphyrin chromophoric unit (ZnP) to the Re(CO)3(bpy)
fragment (Re) (bpy = 2,2′-bipyridine).7 Ultrafast time-resolved
techniques have shown that the charge-separated state [ZnP+−
Re(CO)3(bpy)•−], formed upon visible excitation, recombines
to the ground state in tens of picoseconds.7
The present work is aimed toward improving the perform-
ance of this dyad by introducing a third redox-active
component, which should promote the formation of a
charge-separated state over a longer distance. For this purpose,
a three-component system containing a fullerene as second
electron acceptor was designed (Chart 1). Among the possible
electron acceptors, fullerene has been chosen for its peculiar
properties: (i) low, and reversible, first reduction potential; (ii)
small reorganization energy, which tends to increase the rate of
charge separation and decrease that of charge recombination.8
While metalloporphyrin−fullerene conjugates have been
investigated to a larger extent,9 examples of fullerenes
coordinated to metal complexes are quite rare.10 To our
Perutz.4 4 From a structural point of view, the use of Re(I) is
d, e
particularly effective in combination with N-based ligands, as
strong and inert bonds are established. Indeed, this metal center
has been employed in the construction of robust assemblies
containing pyridyl porphyrins for multipurpose applications.5
In this context, we have previously reported on the synthesis
and characterization of Re(I) zinc-porphyrin dyads, obtained
employing, as common starting metal precursor, the complex
[fac-Re(CO)3(dmso-O)3](OTf) (dmso = dimethyl sulfoxide;
OTf = trifluoromethanesulfonate). This complex, as compared
to the more commonly employed [ReBr(CO)5], has been
Received: October 7, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
MHz) power supply. The decays were analyzed by means of
PicoQuant FluoFit Global Fluorescence Decay Analysis Software.
Transient absorption experiments based on femtosecond laser
photolysis were carried out using two different setups: (1) In Ferrara
the measurements were performed using a pump−probe setup based
on a Hurricane Ti:sapphire laser system (Spectra-Physics) equipped
with a TAPPS-Helios with a detection range between 450 and 800 nm
(Ultrafast Systems) transient absorption spectrometer. The excitation
pulses (560 nm) were generated with an OPA (Spectra Physics 800
OPA). Probe pulses were obtained by continuum generation on a
sapphire plate (useful spectral range: 450−800 nm). Effective time
resolution ca. 300 fs, temporal chirp over the white-light 450−750 nm
range ca. 200 fs, temporal window of the optical delay stage 0−1000
ps. The time-resolved spectral data were analyzed with the Ultrafast
Systems Surface Explorer Pro software. (2) In Erlangen the
measurements were carried out using an amplified Ti:sapphire laser
system CPA-2111 fs laser (Clark MXR − output: 775 nm, 1 kHz and
150 fs pulse width) using a transient absorption pump−probe
detection system equipped with a visible and NIR detector (TAPPS
Helios - Ultrafast Systems). The excitation wavelength (420 nm) was
generated with an NOPA Plus (Clark MXR); pulse widths <150 fs
with an energy of 200 nJ were selected.
Chart 1
Materials. All chemicals were purchased from common commer-
cial suppliers and used without further purification, unless otherwise
stated. All the solvents for photophysical measurements were of
spectroscopic grade. Deuterated solvents were purchased from CIL.
meso-Tetraphenylporphyrin (TPP) and its zincated derivative ZnTPP
were prepared as described in the literature.11 Complexes [fac-
Re(CO)3(dmso-O)3](PF6) (1), [fac-Re(CO)3(bpy)(dmso-O)](PF6)
(2), and [fac-Re(CO)3(bpy)(py)](PF6) (8) were prepared with the
procedures previously described for the analogues bearing the
trifluoromethanesulfonate as counterion, see also below.6,7 5-[4-(4-
Methyl-2,2′-bipyridine-4′-carboxyamidyl)phenyl]-10,15,20-(phenyl)-
porphyrin (TPP-Bpy) was prepared as previously reported by us.12
2-(Octylamino)acetic acid. A dichloromethane solution (100
mL) of benzyl bromoacetate (3 mL, 19.1 mmol) was added dropwise
to a dichloromethane (150 mL) solution of n-octylamine (9 mL, 54.5
mmol) at 0 °C and stirred overnight. The mixture was extracted with
distilled water (6 × 50 mL) and purified by column chromatography
on silica gel eluting with petroleum ether/ethyl acetate 9/1. A catalytic
amount of Pd/C (10% w/w) was added. The solution was left under
stirring overnight under hydrogen atmosphere. The mixture was
filtered on Celite, and the solvent was removed in vacuo to obtain the
product as a white solid. Yield: 1.44 g (40%). 1H NMR (δ, 500 MHz,
CD3OD): 3.49 (s, 2H, HA), 2.55 (t, J = 7.1 Hz, 2H, HB), 1.38−1.29
(m, 12H, HC-HH), 0.88 (t, J = 7.1 Hz, 3H, CH3).
knowledge three-component systems containing a metal-
loporphyrin as chromophore and, as first and second electron
acceptor, a metal complex and a fullerene, respectively, have
never been described before.
In this work, the desired triad system (Chart 1) has been
conveniently obtained by coordination, in a stepwise manner,
to the [fac-Re(CO)3(dmso-O)3]+ precursor of one zinc
porphyrin and one fullerene, appropriately functionalized. In
this system the rhenium fragment acts both as central
coordinative connector and as redox-active unit. The synthesis,
characterization, and photophystical properties of this new triad
system will be discussed in detail here.
N-Octyl-2(4-pyridine)-3,4-fulleropyrrolidine (C60C8Py). A sol-
ution of C60 (250 mg, 0.350 mmol), pyridine-4-carboxaldehyde (0.21
mL, 1.73 mmol), and 2-(octylamino)acetic acid (130 mg, 0.69 mmol)
were dissolved in 200 mL of toluene and stirred at reflux for 25 min,
and then the solvent was removed in vacuo. The product was purified
by column chromatography on silica gel initially eluted with toluene to
recover unreacted C60 and then with toluene/ethyl acetate 95/5. Yield:
185 mg (55%, the yield is based on C60 conversion). 1H NMR (δ, 500
MHz, CDCl3): 8.68 (d, J = 5.3 Hz, 2H, pya), 7.76 (s, 2H, pyb), 5.13 (d,
J = 9.4 Hz, 1H, HB), 5.05 (s, 1H, HA), 4.15 (d, J = 9.4 Hz, 1H, HB′),
3.16 (dt, J = 11.9, 8.5 Hz, 1H, HC), 2.59 (ddd, J = 12.4, 8.4, 4.4 Hz,
1H, HC′), 2.06−1.82 (m, 2H, HD), 1.75−1.50 (m, 2H, HE), 1.50−1.18
(m, 10H, HF‑J), 0.92 (t, J = 6.9 Hz, 3H, CH3). 13C NMR (δ, HSQC,
CDCl3): 150.26 (CH‑pya), 124.27 (CH‑pyb), 81.25 (CH−A), 66.72
(CH−B), 53.16 (CH−C), 31.76 (CH−J), 29.30 (CH−(F−H)), 28.18
(CH−D), 27.34 (CH−E), 22.51 (CH−I), 14.01 (CH−(CH3)). Selected IR
bands (cm−1, solid state): 1054, 1032, and 1012 (s, νN−C pyrrolidine).
UV−vis (λmax, nm, CH2Cl2): 257, 308, 325, 430. ESI-MS (m/z)
(positive mode): calcd for [M + H]+ 953.2, found 953.4.
EXPERIMENTAL SECTION
■
Instrumental Methods. Mono- and bidimensional NMR experi-
ments (1H, H−H COSY, H−C COSY, H−H NOESY, ROESY) were
recorded on a Varian 500 (500 MHz) spectrometer. All spectra were
1
run at room temperature; H chemical shifts were referenced to the
peak of residual nondeuterated solvent (δ (ppm) = 7.26 (CHCl3), 3.30
(CH3OH), 2.50 (dmso), 4.33 (CH3NO2), and 8.03 (dimethylforma-
mide (DMF)). Elemental analysis was performed at the Department of
Chemical Sciences and Technologies, University of Udine (Italy).
Electrospray ionization (ESI) mass spectrometry (MS) measurements
were performed on a PerkinElmer APII spectrometer at 5600 eV by
Dr. F. Hollan, Dept. of Chemical and Pharmaceutical Sciences, Univ.
of Trieste, Italy. Solid IR spectra were recorded on a Varian 660-IR
FT-IR spectrometer in conjunction with a Pike GladiATR accessory
provided with a germanium crystal.
UV−vis absorption spectra were recorded with a Jasco V-570 UV−
vis−near-IR instrument. Emission spectra were acquired on a Spex-
Jobin Ivon Fluoromax-2 spectrofluorimeter, equipped with Hama-
matsu R3896 tubes. Nanosecond emission lifetimes were measured
using a TCSPC apparatus (PicoQuant Picoharp300) equipped with
sub-nanosecond LED sources (280−600 nm range; 500−700 ps pulse
width) powered with a PicoQuant PDL 800-B variable-pulsed (2.5−40
5-[4-(4-Methyl-2,2′-bipyridine-4′-carboxyamidyl)phenyl]-
10,15,20-triphenyl zinc porphyrin (ZnTPP-Bpy). A chloroform
solution (25 mL) of TPP-Bpy (148 mg, 0.18 mmol) was treated
overnight with 2.5 equiv of Zn(CH3COO)2·2H2O (99 mg, 0.45
mmol) dissolved in the minimum amount of methanol (3 mL). The
B
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
solution was dried in vacuo. The product was purified by column
chromatography on aluminum oxide eluted with chloroform/n-hexane
80/20, and vacuum-dried. Yield: 128.6 mg (81%). H NMR (δ, 500
11.29 (s, 1H, NH), 9.38 (d, J = 5.5 Hz, 1H, H6′), 9.30 (s, 1H, H3′),
9.03 (d, J = 5.7 Hz, 1H, H6), 8.95 (s, 1H, H3), 8.93−8.76 (m, 8H, βH),
8.35 (d, J = 5.6 Hz, 1H, H5′), 8.33−8.26 (m, 4H, o+mNH), 8.26−8.19
(m, 6H, oH), 7.90−7.81 (m, 9H, m+pH), 7.77 (d, J = 5.4 Hz, 1H, H5),
2.67 (s, 3H, CH3), 2.54 (s, 6H, CH3 dmso-O), −2.90 (s, 2H, -NH).
1H NMR (δ, 500 MHz, CDCl3): 9.44 (s, 1H, NHam), 9.16 (d, J = 5.6
Hz, 1H, H6′), 8.96−8.84 (m, 9H, βH+H3′), 8.83 (d, J = 5.6 Hz, 1H,
H6), 8.50 (s, 1H, H3), 8.30 (d, J = 6.2 Hz, 1H, H5′) 8.28 (d, J = 3.2 Hz,
4H, o+mNH), 8.26−8.20 (m, 6H, oH), 7.84−7.72 (m, 9H, m+pH),
7.47 (d, J = 6.2 Hz, 1H, H5), 2.71 (s, 3H, CH3), 2.60 (d, J = 3.9 Hz,
6H, CH3 dmso-O), −2.59 (s, 2H, -NH). 13C NMR (δ, HSQC,
CDCl3): 153.40 (CH‑6′), 151.78 (CH‑6), 135.13 (CH‑mNH), 134.51
(CH‑oH), 128.50 (CH‑5), 126.86 (CH‑(m+pH)), 126.57 (CH‑3,5′), 122.01
(CH‑βH), 118.82 (CH‑oNH), 38.35 (CH‑(CH3‑dmso)), 21.74 (CH‑(CH3)).
Selected IR bands (cm−1, solid state): 2030 (s, νCO fac), 1918 (s, br,
νCO fac),, 1677 (m, νCO), 966 (m, νSO dmso‑O), 842 (s, νP−F). UV−
vis (λmax, nm, CH2Cl2): 245, 290, 422, 552, 594. ESI-MS (m/z)
(positive mode): calcd for [M − PF6 − dmso-O]+ 1096.3, found
1096.2.
1
MHz, CDCl3): 8.98 (dt, J = 8.0, 4.0 Hz, 2H, βH), 8.95 (s, 6H, βH),
8.94 (d, J = 4.9 Hz, 1H, H6′), 8.91 (s, 1H, H3′), 8.62 (d, J = 4.8 Hz, 1H,
H6), 8.50 (s, 1H, NH), 8.37 (s, 1H, H3), 8.26 (d, J = 8.3 Hz, 2H,
oNH), 8.25−8.20 (m, 6H, oH), 8.08 (d, J = 8.1 Hz, 2H, mNH), 7.97
(d, J = 3.4 Hz, 1H, H5′), 7.80−7.71 (m, 9H, m+pH), 7.24 (s, 1H, H5),
2.51 (s, 3H, CH3). 13C NMR (δ, HSQC, CDCl3): 150.37 (CH‑6′),
149.11 (CH‑6), 134.62 (CH‑oNH), 134.09 (CH‑oH), 131.80 (CH‑βH),
126.95 (CH‑pH), 125.35 (CH‑mH), 122.30 (CH‑3), 121.61 (CH‑5′), 118.33
(CH‑mNH), 117.19 (CH‑3′), 21.24 (CH‑(CH3)). UV−vis (λmax, nm,
CH2Cl2): 245, 290, 422, 552, 594. ESI-MS (m/z) (negative mode):
calcd for [M − H]− 886.2, found 886.1.
[fac-Re(CO)3(dmso-O)3](PF6) (1). [ReBr(CO)5] (0.60 g, 1.48
mmol) was dissolved in a mixture of acetone (30 mL) and dmso (680
μL, 9.6 mmol). AgPF6 (0.38 g, 1.48 mmol) was added, and the light-
protected mixture was heated to reflux for 4 h. After removal of AgBr
by filtration, the colorless solution was concentrated in vacuo to ca. 3
mL and stored at room temperature after dropwise addition of diethyl
ether until saturation (precipitation of residual AgBr may occasionally
occur upon addition of ether). Colorless crystals formed within a few
days and were removed by filtration, washed with diethyl ether, and
vacuum-dried. Yield 0.67 g (70%). Anal. Calcd: MW = 649.60,
C9H18O6F6P1S3Re. Requires: C, 16.6%; H, 2.79%. Found: C, 16.4%;
[fac-Re(CO)3(ZnTPP-Bpy)(dmso-O)](PF6) (5). ZnTPP-Bpy (119
mg, 0.134 mmol) and 1 (79 mg, 0.121 mmol) were dissolved in
chloroform (50 mL). The mixture was stirred at reflux for 4 h under
argon. The product was collected by filtration after precipitation from
a concentrated solution with the addition of diethyl ether and vacuum-
dried. Yield: 125 mg (75%). 1H NMR (δ, 500 MHz, CDCl3): 9.30 (s,
1H, NH), 9.06 (d, J = 5.4 Hz, 1H, H6′), 8.97 (dd, J = 10.8, 4.5 Hz, 4H,
βH), 8.93 (s, 4H, βH), 8.82 (d, J = 5.6 Hz, 1H, H6), 8.71 (s, 1H, H3′),
8.44 (s, 1H, H3), 8.27−8.22 (m, 6H, oH), 8.20 (d, J = 7.9 Hz, 2H,
oNH), 7.97 (d, J = 7.7 Hz, 2H, mNH), 7.82−7.70 (m, 10H, m+pH +
H5′), 7.47 (d, J = 5.3 Hz, 1H, H5), 2.71 (s, 3H, CH3), 2.55 (s, 6H, CH3
dmso-O). 13C NMR (δ, HSQC, CDCl3): 152.92 (CH‑6′), 151.81
(CH‑6), 134.65 (CH‑oNH), 134.41 (CH‑oH), 131.96 (CH‑βH), 128.55
(CH‑5), 127.64 (CH‑5′), 127.30 (CH‑(m+pH)), 126.10 (CH‑3), 121.67
(CH‑3′), 118.26 (CH‑mNH), 38.25 (CH‑(CH3‑dmso)), 21.65 (CH‑(CH3)).
Selected IR bands (cm−1, solid state): 2030 (s, νCO fac), 1913 (s, br,
νCO fac), 1670 (m, νCO), 950 (m, νSO dmso‑O), 843 (s, br, νP−F).
UV−vis (λmax, nm, CH2Cl2): 245, 300, 423, 552, 595. ESI-MS (m/z)
(positive mode) for [M − PF6 − dmso-O]+ 1158.2, found 1158.2.
[fac-Re(CO)3(TPP-Bpy)(C60C8Py)](PF6) (6). 4 (49 mg, 0.037
mmol) and C60C8Py (36 mg, 0.037 mmol) were dissolved in 1,2-
dichloroethane (15 mL). The mixture was stirred at reflux for 24 h
under argon. The solution was concentrated in vacuo to ca. 10 mL, and
the same amount of diethyl ether was added to induce precipitation. A
brown pure product was collected by filtration and vacuum-dried.
1
H, 2.72%. H NMR (δ, 500 MHz, CD3NO2): 2.93 (s, 18H, CH3
dmso-O). Selected IR bands (cm−1, solid state): 2026 (s, νCO fac),
1902 and 1871 (s, br, νCO fac), 950 (s, νSO dmso‑O), 843 (s, νP−F), 475
(m, νRe−O).
[fac-Re(CO)3(bpy)(dmso-O)](PF6) (2). 1 (100 mg, 0.15 mmol)
and 2,2′-bipyridine (31 mg, 0.2 mmol) were dissolved in acetone (15
mL). The solution, initially almost colorless, was refluxed for 1 h under
argon atmosphere and turned yellow. Concentration in vacuo to ca.10
mL followed by addition of a few drops of diethyl ether induced the
formation of yellow microcrystals, which were removed by filtration,
washed with diethyl ether, and vacuum-dried. Yield: 84 mg (85%).
Anal. Calcd: MW = 649.52, C15H14N2O4F6P1S1Re. Requires: C, 27.7%;
H, 2.17%; N, 4.31%. Found: C, 28.1%; H, 2.12%; N, 4.24%. 1H NMR
(δ, 500 MHz, CD3NO2): 9.15 (d, J = 4.8 Hz, 2H, H6,6′), 8.55 (d, J =
8.2 Hz, 2H, H3,3′), 8.37 (t, J = 8.1 Hz, 2H, H5,5′), 7.80 (t, J = 8.1, 2H,
H
4,4′), 2.62 (s, 6H, CH3 dmso-O). Selected IR bands (cm−1, solid
state): 2032 (s, νCO fac), 1910 and 1890 (s, br, νCO fac), 951 (s,
νSO dmso‑O), 842 (s, νP−F), 485 (m, νRe−O).
[fac-Re(CO)3(bpy)(C60C8Py)](PF6) (3). C60C8Py (50.4 mg, 0.053
mmol) was added to a solution of 2 (44.9 mg, 0.069 mmol) in
anhydrous dichloromethane (15 mL). The mixture was stirred at reflux
for 24 h. A brown solid was precipitated after concentration in vacuo
followed by addition of diethyl ether. The solid was filtered and
1
Yield: 57 mg (70%). H NMR (δ, 500 MHz, dmso-d6): 10.88 (s, 1H,
NH), 10.75 (s, 1H, NH), 9.51−9.45 (m, 1H, H6′), 9.21−9.15 (m, 1H,
H6), 8.90−8.74 (m, 9H, βH+H3′), 8.71 (s, 1H, H3′), 8.40 (d, J = 6.0
Hz, 2H, pya), 8.33 (s, 1H, H3), 8.29 (d, J = 5.9 Hz, 1H, H5′), 8.26−
8.13 (m, 8H, oH+H3+H5′), 8.10 (d, J = 8.0 Hz, 2H, oNH), 8.02 (d, J =
8.3 Hz, 2H, mNH), 7.93−7.81 (m, 9H, m+pH), 7.78 (d, J = 5.9 Hz,
1H, H5), 7.70 (d, J = 5.9 Hz, 1H, H5), 7.63 (d, J = 4.7 Hz, 2H, pyb),
5.07 (s, 1H, HA), 5.02 (s, 1H, HA), 4.73 (d, J = 9.1 Hz, 1H, HB), 4.51
(d, J = 8.9 Hz, 1H, HB), 3.82−3.60 (m, 1H, HB′), 3.04−2.90 (m, 1H,
HC), 2.59−2.53 (m, 1H, HC), 2.43−2.32 (m, 3H, CH3), 1.89−1.68
(m, 1H, HD), 1.66−1.52 (m, 1H, HD), 1.49−1.26 (m, 10H, HE-HI),
1.02−0.76 (m, 3H, HJ), −2.98 (d, 2H, NH). 13C NMR (δ, HSQC,
CDCl3): 134.68 (CH‑(oH+oNH)), 127.74 (CH‑mH), 127.63 (CH‑pH),
119.34 (CH‑mNH)), 31.65 (CH−I), 29.22 (CH‑(F−H)), 27.69 (CH‑E),
22.31 (CH−I), 21.40 (CH‑(CH3)), 14.29 (CH‑J). Selected IR bands (cm−1,
solid state): 2036 (s, νCO fac), 1940 (s, br, νCO fac), 1914 (s, br,
νCO fac), 1674 (m, νCO), 842 (s, br, νP−F). UV−vis (λmax, nm,
CH2Cl2): 256, 310, 325, 419, 516, 552, 591, 647. ESI-MS (m/z)
(positive mode): calcd for [M − PF6]+ 2049.2, found 2049.0.
1
vacuum-dried. Yield: 50 mg (60%). H NMR (δ, 500 MHz, CDCl3):
9.07 (d, J = 5.5 Hz, 1H, H6), 8.99 (d, J = 5.4 Hz, 1H, H6), 8.71 (d, J =
8.2 Hz, 1H, H3), 8.67 (d, J = 8.2 Hz, 1H, H3), 8.27 (t, J = 7.1 Hz, 2H,
H5), 8.19 (d, J = 5.5 Hz, 2H, pya), 7.81 (d, J = 6.1 Hz, 2H, pyb), 7.72
(t, J = 6.8 Hz, 1H, H4), 7.65 (t, J = 6.7 Hz, 1H, H4), 5.06 (d, J = 9.7
Hz, 1H, HB), 5.04 (s, 1H, HA), 4.10 (d, J = 9.6 Hz, 1H, HB′), 3.01−
2.91 (m, 1H, HC′), 2.61−2.53 (m, 1H, HC), 1.95−1.78 (m, 2H, HD),
1.68−1.55 (m, 2H, HE), 1.49−1.28 (m, 8H, HF‑J), 0.92 (t, J = 7.0 Hz,
3H, CH3). 13C NMR (δ, HSQC, CDC13): 152.73 (CH‑6), 151.56
(CH‑pya), 141.56 (CH‑5), 128.70 (CH‑4), 127.03 (CH‑pyb), 126.13 (CH‑3),
80.03 (CH‑A), 66.49 (CH−B), 53.49 (CH−C), 31.76 (CH‑J), 29.36
(CH‑(F−H)), 28.23 (CH‑D), 27.51 (CH‑E), 22.57 (CH−I), 14.04
(CH‑(CH3)). Selected IR bands (cm−1, solid state): 2032 (s, νCO fac),
1918 (s, br, νCO fac), 842 (s, νP−F). UV−vis (λmax, nm, CH2Cl2): 254,
313. ESI-MS (m/z) (positive mode): calcd for [M − PF6]+ 1379.2,
found 1379.2.
[fac-Re(CO)3(ZnTPP-Bpy)(C60C8Py)](PF6) (7). 5 (65 mg, 0.047
mmol) and C60C8Py (45 mg, 0.047 mmol) were dissolved in 1,2-
dichloroethane (10 mL). The mixture was refluxed for 24 h under
argon. The brown solid formed during the reaction was collected by
[fac-Re(CO)3(TPP-Bpy)(dmso-O)](PF6) (4). TPP-Bpy (75 mg,
0.092 mmol) and 1 (45 mg, 0.070 mmol) were dissolved in
chloroform (15 mL). The mixture was stirred at reflux for 4 h
under argon. The purple product was collected by filtration after
concentration of the solution and addition of diethyl ether and
vacuum-dried. Yield: 60 mg (60%). 1H NMR (δ, 500 MHz, dmso-d6):
1
filtration and vacuum-dried. Yield: 85 mg (80%). H NMR (δ, 500
MHz, CDCl3 + 40 μL of CD3OD): 9.34−9.14 (m, 1H, H6′), 9.01−
8.92 (m, 1H, H6), 8.92−8.72 (m, 9H, βH+H3′), 8.57 (s, 1H, H3′), 8.35
C
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 1
a
(i) C60/pyridine-4-carboxaldehyde, toluene, reflux, 25 min, 55%; (ii) dichloromethane, reflux, 24 h, 60%. For simplicity, only the (S) enantiomers
are represented for both C60C8Py and 3.
1
Figure 1. H NMR spectra (CDCl3) of C60C8Py (above) and 3 (below), with some of the relevant cross peaks from the H−H COSY spectra
indicated with circles of the same color; see Scheme 1 for labeling.
(s, 1H, H3), 8.26−7.94 (m, 13H, oH+pya+(o+mNH)+H5′), 7.80−7.61
(m, 11H, m+pH+pyb), 7.53−7.39 (m, 2H, H5), 4.98−4.50 (m, 2H,
HA+HB), 3.79 (d, J = 8.6 Hz, 1H, HB′), 3.00−2.87 (m, 1H, HC), 2.65−
2.41 (m, 3H, CH3), 1.93−1.50 (m, 3H, HD), 1.48−1.08 (m, 10H,
HE‑I), 0.92−0.78 (m, 3H, HJ). Selected IR bands (cm−1, solid state):
2036 (s, νCO fac), 1940 (s, br, νCO fac), 1914 (s, br, νC O fac), 1673
(m, νCO), 843 (m, νP−F). UV−vis (λmax, nm, CH2Cl2): 258, 313, 326,
422, 553, 593. ESI-MS (m/z) (positive mode) for [M − PF6]+ calcd
2112.6, found 2111.5.
[fac-Re(CO)3(bpy)(py)](PF6) (8). 2 (30 mg, 0.046 mmol) was
dissolved in CH2Cl2 (15 mL). After addition of pyridine (37 μL, 0.46
mmol), the solution was stored at ambient temperature for 2 d and
then concentrated in vacuo to ca. 5 mL. Yellow microcrystals of the
product formed upon addition of few drops of diethyl ether and were
removed by filtration, washed with diethyl ether, and vacuum-dried.
Yield: 22 mg (73%). H NMR (δ, 500 MHz, CD3NO2): 9.32 (d, J =
1
4.8 Hz, 2H, H6,6′), 8.44 (d, J = 8.2 Hz, 2H, H3,3′), 8.36 (m, 4H, H4,4′
+
pya), 7.87 (m, 3H, H5,5′ + pyc), 7.35 (t, J = 7.1 Hz, 2H, pyb). Selected
IR bands (cm−1, solid state): 2026 (s, νCO fac), 1921 and 1907 (s, br,
νCO fac), 843 (s, νP−F).
RESULTS AND DISCUSSION
■
Synthesis and Characterization. The fulleropyrrolidine
monoadduct C60C8Py, equipped with both a pyridyl group for
coordination to the Re(I) metal center and a C8 alkyl chain to
foster solubility, was prepared with the well-known 1,3-dipolar
cycloaddition to C60 of azomethine ylide, formed from the
D
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Partial H−H NOESY spectrum (CDCl3) of C60C8Py with some of the relevant spatial correlations and corresponding NOE cross peaks
indicated by arrows and squares, respectively.
a
Scheme 2
a
(i) 1, CHCl3, reflux under Ar, 4 h, 70% (4) and 75% (5); (ii) 1,2-dichloroethane, reflux under Ar, 24 h, 70% (6) and 80% (7).
condensation of pyridine-4-carboxaldehyde and 2-
(octylamino)acetic acid (Scheme 1).13
To explore the reactivity of C60C8Py toward the Re(I)
center, and to get useful insights for the characterization of the
following target three-component system, the model com-
pound 3 was synthesized (Scheme 1). 3, obtained in good yield
by treatment of [fac-Re(CO)3(bpy)(dmso-O)](PF6) (2) with
C60C8Py, is well-soluble in chlorinated solvents, and has been
fully characterized in solution by ESI mass spectrometry and
NMR spectroscopy and in the solid state by IR spectroscopy
(see also Figures 6S−9S). In Figure 1 the 1H NMR (CDCl3) of
3, which shows the correct relative signal integrations, is
The nonselective generation of a chiral center (carbon CH‑A
of the pyrrolidine) is very well-revealed in the 1H NMR
(CDCl3) of C60C8Py. In fact, the geminal protons of the
pyrrolidine and of the alkyl chain are diastereotopic and give
rise to two sets of resonances, more resolved for the protons
closer to the stereogenic center, which can be correctly assigned
through the cross peak correlations found in the 2D H−H
COSY and NOESY spectra (Figures 1 and 2).
E
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
Figure 3. H NMR spectra (CDCl3) of ZnTPP-Bpy (above) and 5 (below), see Scheme 2 for labeling.
reported; all the assignments were done by means of 2D NMR
experiments (H−H COSY, H−C HSQC, and H−H NOESY).
The chemical shifts and pattern of the proton resonances of 3
are not particularly affected by its coordination to the {fac-
Re(CO)3(bpy)} fragment, with the exception of the doublet
related to the two pyridyl protons pya, which is upfield-shifted
as a consequence of the mutual perpendicular disposition of the
pyridyl and bipyridyl rings (Δδ = −0.45 ppm).6,7 On the other
hand, the presence of a chiral center on the apical
fulleropyrrolidine affects the pattern of the proton resonances
of the bpy ligand, which loses its orthogonal mirror plane, and
thus presents two nonequivalent halves. Two sets of resolved
resonances of equal intensities for the bipyridyl protons can be
detected, each set pertaining to one of the nonequivalent bpy
halves (the two doublets of the H5 protons are partly
collapsed), and presenting its own spin system as revealed in
the 2D H−H COSY spectrum (indicated with the blue/green
color code in Figure 1).
(NH, Δδ = 0.80 ppm) proton resonances, while the signals of
the zinc-porphyrin fragment are substantially unchanged. Very
similar features are also observed for 4 (Figures 12S and 13S).
Both 4 and 5 are very valuable intermediates, as they can
allow for the relatively facile preparation of a potentially infinite
variety of three-component systems. In fact, the residual apical
dmso-O is labile and may be easily replaced with any chosen
photoactive additional partner, provided that this latter is
equipped with a donor pyridyl moiety apt for coordination.
From a structural viewpoint, 4 and 5 both exist as mixtures of
enantiomers. In fact, the arrangement of the ancillary ligands
around rhenium center is, for both species, inherently
asymmetrical, and of the type fac-Re(D)3(C)(B^A), thus
identifying one pair of clockwise (C) and anticlockwise (A)
enantiomers (Figure 4), most presumably forming in equal
amounts.
The target three-component system [fac-Re(CO)3(ZnTPP-
Bpy)(C60C8Py)](PF6) (7) and its free-base analogue [fac-
Re(CO)3(TPP-Bpy)(C60C8Py)](PF6) (6) were both efficiently
synthesized via the modular approach already reported by us
for the obtainment of (zinc)porphyrin−Re(CO)3(bpy) dyads,7
and is illustrated in Scheme 2.
Initially, TPP-Bpy or ZnTPP-Bpy, prepared as previously
reported,12 was reacted with fac-[Re(CO)3(dmso-O)3](PF6)
(1), in mild conditions and short reaction times, to give the
intermediates 4 or 5, respectively, in very good isolated yields
(Scheme 2). These derivatives were fully characterized in
solution by ESI mass spectrometry, NMR, UV−vis spectros-
copies, and in the solid state by IR spectroscopy (see also
Figure 4. Two mirror images that can be schematically depicted for 4
and 5.
In the present study, replacement of the apical dmso in 5 (or
in 4) with C60C8Py, by refluxing the reactants in 1,2-
dichloroethane for ∼24 h, led to the isolation, in very good
yield, of the desired three-component system 7 (or of its free-
base analogue 6, Scheme 2). The nature of these products was
confirmed by their ESI mass spectra, both presenting a single
peak with the correct isotopic distributions for the molecular
ion (Figure 5).
The IR spectra (solid state) of both 6 and 7 are consistent
with the presence of the fac-{Re(CO)3} fragment, as they both
present three intense stretching bands: the ones at 1915 and
1940 cm−1 are assigned to the CO asymmetric stretching mode,
and the one at 2035 cm−1 is assigned to the CO symmetric
1
Figures 11S-16S). In Figure 3 a comparison between the H
NMR spectra (CDCl3) of ZnTPP-Bpy and 5 is reported. All
the proton resonances, and in particular those pertaining to the
bipyridyl and the bridging phenyl, were correctly assigned by
means of a combination of H−H COSY and 1D NOESY NMR
experiments (particularly useful are the long-range correlations
between the bpy CH3 and the H3/H5 signals, and the negative
Overhauser effects on the resonances of the bpy H3′ and the
bridging phenyl mNH, detected by saturation of the amide NH
broad singlet, Figure 16S). Coordination of the ZnTPP-Bpy to
the metal center induces a downfield shift of most of the bpy
chelating moiety (Δδ = 0.25 ÷ 0.12 ppm) and of the amide
F
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. ESI-MS spectra of 6 (above) and 7 (below), with magnifications of the molecular ion peak distributions (experimental left, and
calculatedprogram IsoProright, respectively).
stretching mode (an additional weak band at 1674 cm−1 is
assigned to the stretching of the amide CO, Figure 17S).
Given the presence of two stereogenic centers, 6 and 7 both
may form mixtures of isomers, and in particular two couples of
enantiomers can be structurally drawn, as schematically
depicted in Figure 6.
for the resonances of protons NH of the amide; H6′, H6, and H5
of the bpy fragment; HA and HB of the fulleropyrrolidine, and
the −NH of the porphyrin unit (Figure 7). An analogous
pattern can be revealed by H−H COSY for the resonances of
the H5′ bpy proton, presenting two distinct correlation peaks
with those of H6′ (Figure 19S). The two resonances arising
from H3/H3′ bpy protons were assigned thanks to their NOE
correlations with the amide NH broad singlets, and long-range
COSY correlations with the −CH3 bpy protons, respectively
(Figures 20S and 21S). Apparently, the relative intensity of the
two proton sets is not equal, which may be an indication that
the product contains different amounts of the possible
isomers.14−16 Nevertheless, by varying the NMR solvent, the
ratio of the two resonance sets is also found to be varied, and in
particular when the product is completely dissolved, an almost
By looking at the possible relative orientations of the (zinc)-
porphyrin and the fullerene moieties within the same molecule
and with respect to an orthogonal plane bisecting the bpy
ligand, the two couples of enantiomers present two distinct
situations: one in which the two moieties are located on
opposite sides (Re(C),C60(S) and Re(A),C60(R) in Figure 6, top),
and the second in which they reside on same sides (Re(C),C60(R)
and Re(A),C60(S) in Figure 6, bottom). Accordingly, two distinct
proton resonance sets are expected in the 1H NMR spectra of 6
and 7. Indeed, a high degree of complexity in terms of number
of resonances and extended overlaps, which made a correct
interpretation quite troublesome, was found in their NMR
spectra. We believe that the main spectroscopic features are
worth a brief discussion, as they may prove valuable for the
characterization of other conjugates of a similar type. A
spectrum (dmso-d6) of 6 is reported in Figure 7, with
assignments achieved by means of 1D and 2D NMR
experiments (the alkyl resonances region has been omitted,
for clarity).
1
one-to-one ratio is observed (the H NMR of 6 in DMF-d6,
with only some relevant proton resonances, as most of the
other signals overlap and/or are hidden below the residual
DMF, is reported in Figure 8).
The zincated three-component target triad 7 shows a limited
solubility in chlorinated solvents and precipitates during the
reaction progress. The solubility of the isolated product is also
poor in the other common organic solvents. If methanol, that
axially binds to the zinc ion, thus interfering with the π,π
intermolecular interactions between zinc-porphyrin macro-
cycles, and/or zinc-porphyrin and fullerene aromatic surfaces,
is added to suspensions of 7 in chloroform, a significant
increase of solubility is obtained. Still, only a partial assignment
Two sets of signals, more or less resolved, can be observed
for both the TPP-Bpy and the C60C8Py ligands attached to the
fac-Re(CO)3 fragment. This observation is particularly evident
G
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Depiction of the possible isomers of 6 and 7, constituted by two couples of enantiomers, arising from the presence of two stereocenters.
1
Figure 7. H NMR spectrum (dmso-d6) of 6, see Scheme 2 for labeling (the asterisk indicate a satellite of the residual dmso).
1
of the resonances of the H NMR spectrum was possible
(Figure 22S), with most of it derived from a combination of
H−H COSY and NOESY spectra, and also by comparison with
the free-base analogue 6 just described. In particular, at least
four sets of resolved resonances appear distinctly for some of
the protons, most propably as a consequence of the partial
coordination of methanol to the zinc, that produces a larger
distribution of isomeric species.14−16
besides some very minor shifts (∼2−3 nm), best described as
the superimposition of the absorption of the corresponding
model compounds. This fact indicates that there are only very
weak interactions between the individual components in the
ground state. The absorption spectrum of compound 7 is
dominated by the Soret band (around 430 nm) and the two Q
bands (between 500 and 700 nm) of the ZnP moiety.
Compared to ZnP, the rhenium complex and the fullerene
are predominantly absorbing light in the UV region (Figure
23S).
Photophysical Characterization of the Triad 7.17 The
absorption spectrum of 7 in dichloromethane (Figure 23S) is,
H
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
Figure 8. Partial H NMR spectrum (DMF-d6) of 6, see Scheme 2 for labeling.
Figure 9. Transient absorption spectra obtained for 7 in a DCM solution (λexc = 560 nm) in the 0−10 ps range (left) and in the 50−2000 ps range
(right).
Taking the former into account, an exclusive excitation of the
ZnP chromophore is feasible in the visible region of the
spectrum. A detailed photophysical investigation of the new
triad was carried out in dichloromethane solution, upon visible
excitation, by steady state and time-resolved absorption and
emission spectroscopy. At room temperature, a solution of the
triad shows the typical Zn-porphyrin fluorescence with
prominent vibronic bands at 599 and 648 nm (Figure 24S);
even taking a close look into the 700−750 region, no C60-
centered fluorescence was observed, neither in DCM nor in less
polar solvents (e.g., toluene). Comparative experiments, carried
out on optically matched solutions of 7 and the ZnTPP model
at λexc = 516 nm, indicate that, in the triad, the zinc-porphyrin
fluorescence, although maintaining the same emission pattern,
is strongly quenched relative to that of the model compound
(Φ0/Φ ≥ 100, where Φ0 and Φ are the fluorescence quantum
yields of the ZnTTP model and 7, respectively, Figure 24S).
Single-photon counting experiments demonstrate that the
fluorescence lifetime of the ZnP−Re−C60 triad 7 lies below
the instrumental time resolution (<250 ps). To obtain insight
into the quenching mechanism, ultrafast transient absorption
experiments were performed in different solvents (dichloro-
methane, toluene, benzonitrile, and tetrahydrofuran) and with
different experimental setups and excitation wavelengths.
The transient absorption spectra observed in dichloro-
methane upon excitation at 560 nm are shown in Figure 9.
The initial spectrum, taken immediately after the excitation
pulse (τ ≈ 1 ps), is identical with the transient absorption
spectrum typically observed for the zinc-porphyrin singlet
excited state, characterized by a broad intense positive
absorption in the 450−550 nm range, a relatively flat positive
absorption in the visible region in the 560−800 nm range with
superimposed transient bleaching of the ground-state Q bands
(580−600 nm). Additional apparent bleaches in the 650−720
nm range are caused by stimulated emission.
The transient spectral changes show a biphasic behavior,
taking place in the 1−50 ps and in the 50−2000 ps range
(Figure 9). In the first range a fast increase of the optical
density in the 620−700 nm range and a reduction of Q-band
bleach are observed. These spectral changes are relatively small,
but the signatures of the radical cation of the zinc-porphyrin
component (broad absorption bands in the 600−800 nm
range) can be easily recognized, consistent with the formation
of a charge-separated state in which this component is oxidized
(ZnP+).18 In the longer time range (50−2000 ps) a uniform
decay of ZnP+ spectrum toward the initial baseline is observed,
clearly indicating that ZnP+ converts quantitatively to the
ground state. These results are consistent with an electron-
transfer mechanism for the quenching of excited singlet state of
the ZnP component. Kinetic analysis of the spectral changes at
650 nm (Figure 25S) yields a time constant of 2 1 ps for the
first process (formation of ZnP+) and 140
10 ps for the
second one (conversion to ground state).
In these measurements, the investigated spectral window
(400−800 nm) is limited by experimental setup (see
Experimental Section). To obtain evidence on the nature of
I
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the charge separated state it is necessary to extend the study to
the infrared region, in which the C−60 shows its transient
absorptions. For this purpose, a transient absorption detection
system with a wider spectral window (400−1200 nm) was
employed. Experiments in various solvents (toluene, THF, and
benzonitrile), by using 420 nm as excitation wavelength, were
performed (see also Figures 26S and 27S).19 The spectral
changes observed in toluene are reported in Figure 10.
Kinetic analysis of the spectral changes observed in the
different solvents indicates that, while the charge-separation
process is very fast (1−2 ps) and nearly independent of the
polarity of the solvent, the charge recombination, leading to the
ground state, is by far slower and strongly solvent dependent.
The values of the time constants of the charge recombination
process reported in Table 1 are obtained from kinetics analyses
at three different wavelengths: 650, 465, 1020 nm (see also
Figures 28S−31S).
In conclusion, the interpretation of the transient spectral
changes observed in the ultrafast absorption experiments as
formation and decay of the charge-separated product ZnP+−
Re−C6−0, respectively, is straightforward. In all solvents
investigated no evidence of an intermediate species ZnP+−
Re−−C60, where the ZnP is oxidized and rhenium component
is reduced, was obtained. To discuss this point, it is useful
consider a simplified energy level diagram for 7 studied in
dichloromethane (Figure 11, the excited states of the C60 unit
are omitted for clarity).21
Figure 10. Femtosecond transient absorption spectra of 7 in argon-
saturated toluene, after excitation at 420 nm.
Immediately after excitation of the ZnP component (τ = 0.1
ps), its singlet excited state is formed. Evidences for the latter
stem from the occurrence of an intense absorption with maxima
at 462, 570, and 615 nm as well as minima at 550, 600, and 620
nm corresponding to the bleach of the ground-state Q bands
and to the stimulated emission. In the early time scale (0−30
ps), spectral changes with signatures of radical cation ZnP+
(maxima at 460, 625, and 675 nm and a minimum at 555 nm)
and a maximum at 1025 nm, which corresponds to the typical
absorption of the radical anion, C−60, are clearly discernible.
These observations indicate that a charge-separated product
(ZnP+−Re−C6−0), in which the zinc-porphyrin chromophore is
oxidized and the fullerene unit is reduced, is formed.
Importantly, the ZnP+−Re−C6−0 features grow in a time scale
that matches the decay of the ZnP singlet excited state,
indicating that a photoinduced electron-transfer process from
the ZnP excited state to C60 component occurs. In a longer
time scale, the spectroscopic signatures of the charge-separated
state, ZnP+−Re−C6−0, disappear with a decay of the whole
spectrum toward the initial baseline. The transient changes
show that, in THF and benzonitrile, ZnP+−Re−C6−0 converts by
charge recombination to the ground state (Figures 30S and
31S), whereas in toluene (Figures 28S and 29S), the
recombination process is much slower (see Table 1), the
charge-separated state ZnP+−Re−C6−0 undergoes spin inversion
and yields the Zn porphyrin triplet state upon charge
recombination.20
Figure 11. Energy level diagram for 7 and photophysical processes in
dichloromethane.
In this diagram, in addition to the energy levels of the excited
states of the ZnP chromophore, two types of intercomponent
charge transfer states, both containing the oxidized form of the
zinc-porphyrin component, must be considered: one, in which
the rhenium-based unit is reduced (ZnP+−Re−−C60) and the
other, in which the C60 unit is reduced (ZnP+−Re−C6−0). The
energy of these states can be obtained from the measured redox
potentials for the three-component system 7 (Table 2) as
Table 2. Electrochemical Data for 7 and for Its Model
a b
,
Compounds
compound
E1/2 (red) (1) (V)
E1/2 (red) (2) (V)
E1/2 (ox) (V)
c
c
ΖnTPP
8
−1.38
−1.13
−0.55
−0.58
−0.65
+0.80
c
d
Table 1. Time Constants (τ) for the Charge Recombination
Process in Different Solvents
C60
3
−1.00
−1.03
7
+0.77
a
solvent
ε
τ (ps)
a
benzonitrile
DCM
25.7
9.1
125
140
All measurements were made in N2-deaerated CH2Cl2 solutions at
298 K (0.1 tetrabutylammonium hexafluorophosphate (TBAPF6) as
supporting electrolyte); scan rate 100 mV/s; SCE as reference
THF
7.6
260
b
toluene
2.4
2800
electrode, and glassy carbon as working electrode). Half wave
c
potential in cyclic voltammetry (ΔEp = 60−80 mV). From ref 7.
a
d
ε is the dielectric constant.
From ref 9f.
J
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
difference in the potentials for oxidation of the Zn-porphyrin
unit (E1/2(ox) (ZnP/ZnP+) = +0.77 V vs SCE) and for the
reduction of the rhenium (E1/2(red) (Re/Re−) = −1.03 V vs
SCE) and the C60 (E1/2(red) (C60/C−60) = −0.65 V vs SCE)
components, respectively (see also Figure 33S), with
appropriate correction for the electrostatic work term. The
redox potential values of 7 are very close to those found for the
corresponding models ZnTPP, C60, 8, and 3 (Table 2 and
Figure 32S), in agreement with the supramolecular nature of 7.
Within the limits of this type of calculation, the ZnP+−Re−−
C60 is estimated to lie at 1.80 eV, and the ZnP+−Re−C6−0 is at
1.42 eV, both lower in energy than the singlet state of the zinc
porphyrin [E(S1) = 2.02 eV] (Figure 11).
with decreasing solvent polarity. A similar solvent dependence
was observed for the charge-recombination process occurring in
the previously studied ZnP−Re dyad (see Chart 1) and can be
qualitatively explained in terms of Marcus theory as a
consequence of the changes in the driving force and
reorganizational energy with changing the solvent in the
hypothesis that the process belongs to the Marcus inverted
region.
Of particular interest to the present work is the comparison
with the dyad studied by Perutz, constituted by the same
chromophoric (ZnP) and electron-acceptor ({Re(CO)3(bpy)})
components, covalently attached by an amide bridge (Figure
12).4d,e
1*ZnP−Re−C60 → ZnP+−Re−−C60 ΔG = −0.22eV
(eq 1)
ZnP+−Re−−C60 → ZnP+−Re−C60− ΔG = −0.38eV
(eq 2)
The charge-separated state, ZnP+−Re−C6−0, recombines to
the ground state in a longer time scale (eq 3).
ZnP+−Re−C−60 → ZnP−Re−C60 ΔG = −1.42 eV
Figure 12. Dyad studied by Perutz.4d,e
(eq 3)
The kinetics of the processes (eq 1, eq 2, eq 3) can be
analyzed in the light of standard electron-transfer theory. In the
limit of weak interaction (non-adiabatic regime), the rate
constant for an electron-transfer process from a donor to an
acceptor is given by the following equation:
The rate of photoinduced charge-separation process reported
by Perutz is practically the same (a few picoseconds) as we
measured in the present study. This fact confirms the
hypothesis of a stepwise mechanism for the formation of the
charge-separated product ZnP+−Re−C6−0 where the electron
transfer from the ZnP first singlet excited state to the rhenium
component is the rate-determining step. On the other hand, the
result that the charge-recombination process in Perutz system is
faster (53 ps vs 260 ps measured in the same solvent, THF) can
be explained by considering that, in our system, the charge-
recombination process involves two components (ZnP and
fullerene unit) that are not directly bound.
2π
ℏ
k =
HDA2FCWD
(eq 4)
where HDA is the electronic coupling between donor and
acceptor and FCWD is the nuclear term (Franck−Condon
weighted density of states) that accounts for the combined
effects of the reorganizational energies and driving force. In
principle, the fact that the electron-transfer processes (eq 1, eq
2, and eq 3) have different electronic factors as well as different
nuclear factors does not allow making a quantitative
comparison between the rate constant values experimentally
measured. However, some remarkable comments can be useful
to rationalize the kinetic data. As far as the charge-separation
and charge-shift processes (eq 1 and eq 2) are concerned, since
the driving forces of the processes place both in the normal
Marcus region (assuming approximately similar reorganiza-
tional energies), the charge-shift process (eq 2) is expected to
be faster being less activated. Moreover, this process is
kinetically favored by the electronic factor: it involves the C60
component directly connected to the rhenium unit via the
pyridyl group, while, in the case of the photoinduced charge
separation process (eq 1), an amido bridge is interposed
A related dyad system constituted by a zinc porphyrin
covalently linked to a C60 unit has been investigated by Imahori
et al. (Figure 13).9d
1
between *ZnP and Re components. Regarding the kinetics of
processes 1 and 3, it is worth also pointing out that the rate of
the charge recombination (eq 3) to ground state is significantly
slower than the charge separation (eq 1) in all investigated
solvents. This result can be easily explained by considering that,
for process 3, both nuclear (process 3 presumably falls in the
Marcus inverted region) and electronic (the electron transfer
occurs between two sites not directly connected) factors are
less favorable.
Figure 13. Zinc porphyrin−C60 dyad studied by Imahori.9d
They found that, in benzonitrile, a photoinduced charge
separation from ZnP to fullerene was followed by a
recombination process occurring with time constants of 100
and 760 ps, respectively. In our system, the photoinduced
process is by far faster (ca. 1 ps), probably because there is a
stronger electronic interaction between the components
involved in the electron-transfer reaction (the singlet excited
As reported above, the data of Table 1 clearly indicate that
the kinetics of the charge-recombination process was found to
be strongly solvent-dependent, with rate constants decreasing
K
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Notes
state of ZnP and the bpy ligand of the {Re(CO)3(bpy)}
fragment). On the other hand, the rate of 125 ps observed in
our case for the charge recombination of the ZnP+−Re−C6−0
state, and faster with respect to that reported by Imahori, may
be ascribed to a superexchange interaction involving the state
ZnP+−Re−−C60 (Figure 11),22 relatively close in energy.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. C. Chiorboli (Univ. of Ferrara, Italy) for the
time-resolved ultrafast measurements (560 nm excitation) in
the visible region and Dr. M. Natali (Univ. of Ferrara, Italy) for
the cyclic voltammetry measurements. Prof. D. M. Guldi
CONCLUSIONS
■
(Friedrich-Alexander-Universitat Erlangen-Nurnberg, Ger-
many) is acknowledged for his support during the photo-
physical experiments in Erlangen.
̈
̈
In summary, we have here shown how the facial tris-acceptor
metal precursor [Re(CO)3(dmso-O)3]+ can be conveniently
employed to obtain systems containing an increased number of
photoactive partners, that is, triads versus dyads, while
maintaining an established synthetic strategy. In the present
study, the preparation of a new triad featuring one
chromophore and two different electron-acceptor moieties
has been presented. Careful attention to the characterization
studies of this new adduct, featuring two stereogenic centers, of
its model and precursor compounds, with special regard to the
NMR features, has been given. We believe that this aspect,
often neglected or poorly analyzed, will give a valuable
contribution to the structural comprehension of similarly
inspired derivatives.
REFERENCES
■
(1) (a) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 15729. (b) Barber, J. Chem. Soc. Rev. 2009, 38, 185−196.
(c) Aratani, N.; Kim, D.; Osuka, A. Acc. Chem. Res. 2009, 42, 1922−
1934. (d) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2009,
42, 1890−1898. (e) Hambourger, M.; Moore, G. F.; Kramer, D. M.;
Gust, D.; Moore, A. L.; Moore, T. A. Chem. Soc. Rev. 2009, 38, 25−35.
(f) Frischmann, P. D.; Mahata, K.; Wurthner, F. Chem. Soc. Rev. 2013,
̈
42, 1847−1870. (g) Sartorel, A.; Bonchio, M.; Campagna, S.;
Scandola, F. Chem. Soc. Rev. 2013, 42, 2262−2280. (h) Fukuzumi,
S.; Ohkubo, K.; Suenobu, T. Acc. Chem. Res. 2014, 47, 1455−1464.
(i) Karkas, M. D.; Jhonston, E. V.; Verho, O.; Akermark, B. Acc. Chem.
Res. 2014, 47, 100−111.
(2) (a) Morris, A. J.; Meyer, G. J.; Fujita, E. Acc. Chem. Res. 2009, 42,
1983−1994. (b) Takeda, H.; Ishitani, O. Coord. Chem. Rev. 2010, 254,
346−354. (c) Windle, C. D.; Perutz, R. N. Coord. Chem. Rev. 2012,
256, 2562−2570.
(3) (a) Hawecker, J.; Lehn, J.-M.; Ziessel, R. J. Chem. Soc., Chem.
Comm. 1983, 536. (b) Hori, H.; Ishihara, J.; Koike, K.; Takeuchi, K.;
Ibusuki, T.; Ishitani, O. J. Photochem. Photobiol., A 1999, 120, 119−
124. (c) Takeda, H.; Koike, K.; Inoue, H.; Ishitani, O. J. Am. Chem.
Soc. 2008, 130, 2023−2031. (d) Agarwal, J.; Fujita, E.; Schaefer, H. F.;
Muckerman, J. T. J. Am. Chem. Soc. 2012, 134, 5180−5186.
(e) Marimoto, T.; Nishiura, C.; Tanaka, M.; Rohacova, J.;
Nakagawa, Y.; Funada, Y.; Koike, K.; Yamamoto, Y.; Shishido, S.;
Kojima, T.; Saeki, T.; Ozeki, T.; Ishitani, O. J. Am. Chem. Soc. 2013,
135, 13266−13269. (f) Tsuyoshi, A.; Nakagawa, Y.; Funada, Y.; Sawa,
S.; Takeda, H.; Morimoto, T.; Koike, K.; Ishitani, O. Inorg. Chem.
2014, 53, 7170−7180.
(4) (a) Sato, S.; Koike, K.; Inoue, H.; Ishitani, O. Photochem.
Photobiol. Sci. 2007, 6, 454−461. (b) Kiyosawa, K.; Shiraishi, N.;
Shimada, T.; Masui, D.; Tachibana, H.; Takagi, S.; Ishitani, O.; Tryk,
D. A.; Inoue, H. J. Phys. Chem. C 2009, 113, 11667−11673.
(c) Tamaki, Y.; Koike, K.; Morimoto, T.; Ishitani, O. J. Catal. 2013,
304, 22−28. (d) Gabrielsson, A.; Hartl, F.; Zhang, H.; Lindsay Smith,
J. R.; Towrie, M.; Vlcek, A.; Perutz, R. N. J. Am. Chem. Soc. 2006, 128,
The photophysical behavior of the system, upon visible light
excitation, has been thoroughly investigated. Electron-transfer
processes were detected and kinetically characterized by using a
combination of emission spectroscopy and femtosecond time-
resolved transient absorption techniques. An ultrafast (ca. 1 ps)
photoinduced electron transfer from the zinc-porphyrin
chromophore to the {Re(CO)3(bpy)} fragment occurs,
followed by a faster charge shift process to the fullerene unit,
not allowing for the accumulation of the ZnP+−Re−−C60
intermediate species. Ultrafast transient absorption techniques
in the NIR region show clear evidence for the formation of the
charge-separated species ZnP+−Re−C6−0. While the rate of
charge separation is practically independent of the polarity of
the solvent, the kinetics of recombination process depends on
the solvent, in agreement with a Marcus inverted behavior.
Efforts are currently focused on the rational design of
alternative three-component photoactive systems, by prepara-
tion of suitable electron acceptors capable of coordinating to
the ZnP−Re(I)(CO)3(bpy) intermediate via an appended
pyridyl group.
ASSOCIATED CONTENT
■
̂
4253−4266. (e) Windle, C. D.; Campian, M. V.; Duhme-Klair, A.-K.;
Gibson, E. A.; Perutz, R. N.; Schneider, J. Chem. Commun. 2012, 48,
8189−8191.
(5) (a) Hupp, J. T. Struct. Bonding (Berlin, Ger.) 2006, 121, 145−165.
(b) Boccalon, M.; Iengo, E.; Tecilla, P. J. Am. Chem. Soc. 2012, 134,
20310−20313. (c) Iengo, E.; Cavigli, P.; Milano, D.; Tecilla, P. Inorg.
Chim. Acta 2014, 417, 59−78.
S
* Supporting Information
Additional compound characterization (ESI-MS, NMR, IR,
UV−vis, fluorescence spectra, transient absorption spectra, time
profiles, and CV experiments). This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
(6) (a) Casanova, M.; Zangrando, E.; Munini, F.; Iengo, E.; Alessio,
E. Dalton Trans. 2006, 5033−5045. (b) Ghirotti, M.; Chiorboli, C.;
Indelli, M. T.; Scandola, F.; Casanova, M.; Iengo, E.; Alessio, E. Inorg.
Chim. Acta 2007, 360, 1121−1130. (c) Casanova, M.; Zangrando, E.;
Iengo, E.; Alessio, E.; Indelli, M. T.; Scandola, F.; Orlandi, M. Inorg.
Chem. 2008, 47, 10407−10418.
■
Corresponding Authors
*E-mail: eiengo@units.it. (E.I.)
*E-mail: idm@unife.it. (M.T.I.)
Funding
(7) Gatti, T.; Cavigli, P.; Zangrando, E.; Iengo, E.; Chiorboli, C.;
Indelli, M. T. Inorg. Chem. 2013, 52, 3190−3197.
Financial support from the Italian MIUR (PRIN Project No.
2010N3T9M4 and FIRB Project No. RBAP11C58Y), from the
University of Trieste (FRA2009 and FRA2012), from
Fondazione Beneficentia Stiftung, and from Deutsche For-
schungsgemeinschaft (Grant No. KA 3491/2−1) is gratefully
acknowledged.
́ ́
(8) (a) Martín, N.; Sanchez, L.; Illescas, B.; Perez, I. Chem. Rev. 1998,
98, 2527−2548. (b) Guldi, D. M.; Prato, M. Acc. Chem. Res. 2000, 33,
695−703.
(9) To cite some of them: (a) Da Ros, T.; Prato, M.; Guldi, D. M.;
Ruzzi, M.; Pasimeni, L. Chem.Eur. J. 2001, 7, 816−827. (b) Guldi,
L
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
D. M. Chem. Soc. Rev. 2002, 31, 22−36. (c) Armaroli, N.; Accorsi, G.;
Felder, D.; Nierengarten, J.-F. Chem.Eur. J. 2002, 8, 2314−2323.
(d) Imahori, H. Org. Biomol. Chem. 2004, 2, 1425−1433. (e) D’Souza,
F.; Ito, O. Coord. Chem. Rev. 2005, 249, 1410−1422. (f) 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, 15973−15982. (g) D’Souza, F.; Ito, O. Chem. Commun.
2009, 4913−4928. (h) Poddutoori, P. K.; Sandanayaka, A. S. D.;
Zarrabi, N.; Hasobe, T.; Ito, O.; van der Est, A. J. Phys. Chem. A 2011,
115, 709−717. (i) Garg, V.; Kodis, G.; Chachisvilis, M.; Hambourger,
M.; Moore, A. L.; Moore, T. A.; Gust, D. J. Am. Chem. Soc. 2011, 133,
2944−2954. (j) Bottari, G.; Trukhina, O.; Ince, M.; Torres, T. Coord.
Chem. Rev. 2012, 256, 2453−2477. (k) Wijesinghe, C. A.; El-Khouly,
M. E.; Zandler, M. E.; Fukuzumi, S.; D’Souza, F. Chem.Eur. J. 2013,
19, 9629−9638. (l) Fukuzumi, S.; Ohkubo, K. Dalton Trans. 2013, 42,
15845−15858.
(10) (a) Guldi, D. M.; Maggini, M.; Menna, E.; Scorrano, G.; Ceroni,
P.; Marcaccio, M.; Paolucci, F.; Roffia, S. Chem.Eur. J. 2001, 7,
1597−1605. (b) Armaroli, N.; Accorsi, G.; Felder, D.; Nierengarten, J.-
F. Chem.Eur. J. 2002, 8, 2314−2323. (c) Chaignon, F.; Torroba, J.;
Blart, E.; Borgstrom, M.; Hammarstrom, L.; Odobel, F. New J. Chem.
̈
̈
2005, 29, 1272−1284. (d) Possamai, G.; Menna, E.; Maggini, M.;
Carano, M.; Marcaccio, M.; Paolucci, F.; Guldi, D. M.; Swartz, A.
Photochem. Photobiol. Sci. 2006, 5, 1154−1164. (e) Allen, B. D.;
Benniston, A. C.; Harriman, A.; Mallon, L. J.; Pariani, C. Phys. Chem.
Chem. Phys. 2006, 8, 4112−4118. (f) Karlsson, S.; Modin, J.; Becker,
H.-C.; Hammarstrom, L.; Grennberg, H. Inorg. Chem. 2008, 47, 7286−
̈
7294. (g) Iehl, J.; Vartanian, M.; Holler, M.; Nierengarten, J.-F.;
Delavaux-Nicot, B.; Strub, J.-M.; Van Dorsselaer, A.; Wu, Y.;
Mohanraj, J.; Yoosaf, K.; Armaroli, N. J. Mater. Chem. 2011, 21,
1562−1573. (h) Walsh, E. A.; Deye, J. R.; Baas, W.; Sullivan, K.;
Lancaster, A.; Walters, K. A. J. Photochem. Photobiol., A 2013, 260, 24−
36.
(11) Fleischer, E. B.; Shachter, A. M. Inorg. Chem. 1991, 30, 3763−
3769.
(12) Gianferrara, T.; Bratsos, I.; Iengo, E.; Milani, B.; Ostric, A.;
́
Spagnul, C.; Zangrando, E.; Alessio, E. Dalton Trans. 2009, 10742−
10756.
(13) Maggini, M.; Scorrano, G.; Prato, M. J. Am. Chem. Soc. 1993,
115, 9798−9799.
(14) VT 1H NMR experiments have been performed for 6 and 7, but
no appreciable variations of the line width and/or resolution were
detected.
(15) 13C NMR spectra have been recorded for 6 and 7, but no useful
information on the carbonyl resonances, which requires long
acquisition times or higher sensitivity instruments, could be derived.
(16) The 2D ROESY spectra registered for 6 and 7 (room
temperature) did not present any exchange cross peaks, so the
precence of (slow) exchange processes between different species can
be excluded.
(17) The photophysical characterization of 6 was not carried out,
since the photoinduced electron-transfer process from the singlet
excited state of the free base porphyrin to the {Re(CO)3(bpy)} is
thermodynamically forbidden.6c
(18) Fajer, J.; Borg, D. C.; Forman, A.; Dolphin, D.; Felton, R. H. J.
Am. Chem. Soc. 1970, 3, 3451−3459.
(19) DCM as solvent for the laser photolysis experiments exciting at
420 nm was not used since photodegradation was observed under our
experimental conditions.
(20) Ghirotti, M.; Chiorboli, C.; You, C. C.; Wurthner, F.; Scandola,
̈
F. J. Phys. Chem. A 2008, 112, 3376−3385.
(21) The singlet (E0−0 = 1.76 eV) and the triplet (E0−0 = 1.50 eV)
first excited states of C60 are not involved in the deactivation of the
singlet excited state *1ZnP−Re−C60, since neither the transient
absorption of the singlet or triplet first excited states of C60 nor
emissions originating form excited states of C60 were observed.
(22) Arrigo, A.; Santoro, A.; Indelli, M. T.; Natali, M.; Scandola, F.;
Campagna, S. Phys. Chem. Chem. Phys. 2014, 16, 818−826.
M
dx.doi.org/10.1021/ic502430e | Inorg. Chem. XXXX, XXX, XXX−XXX