Improving the efficiency of the photoinduced charge-separation process in a rhenium(I)-zinc porphyrin dyad by simple chemical functionalization

Article abstract of DOI:10.1021/ic302663c

We demonstrate here that, whereas the rhenium(I)-zinc porphyrin dyad fac-[Re(CO)₃(bpy)(Zn·4′MPyP)](CF₃SO ₃) [1; 4′MPyP = 5-(4′-pyridyl)-10,15,20- triphenylporphyrin] shows no evidence for photoinduced electron transfer upon excitation in the visible region because the charge-separated state ZnP ⁺-Re^- is almost isoenergetic with the singlet excited state of the zinc porphyrin (ΔG = -0.05 eV), the introduction of electron-withdrawing ethyl ester groups on the bpy ligand significantly improves the thermodynamics of the process (ΔG = -0.42 eV). As a consequence, in the new dyad fac-[Re(CO)₃(4,4′-DEC-bpy)(Zn·4′MPyP) ](CF₃SO₃) (4; 4,4′-DEC-bpy = 4,4′- diethoxycarbonyl-2,2′-bipyridine), an efficient and ultrafast intramolecular electron-transfer process occurs from the excited zinc porphyrin to the rhenium unit upon excitation with visible light. Conversely, the introduction of electron-donor tert-butyl groups on the meso-phenyl moieties of the zinc porphyrin has a negligible effect on the photophysics of the system. For dyad 4, the time constants for the charge-separation and charge-recombination processes in solvents of different polarity (PrCN, DCM, and toluene) were measured by an ultrafast time-resolved absorption technique (λ_exc = 560 nm).

Full text of DOI:10.1021/ic302663c

Article
pubs.acs.org/IC
Improving the Efficiency of the Photoinduced Charge-Separation
Process in a Rhenium(I)−Zinc Porphyrin Dyad by Simple Chemical
Functionalization
Teresa Gatti,†,‡ Paolo Cavigli,† Ennio Zangrando,† Elisabetta Iengo,*,† Claudio Chiorboli,§
and Maria Teresa Indelli*,⊥
†Department of Chemical and Pharmaceutical Sciences, University of Trieste, Via L. Giorgieri 1, 34127 Trieste, Italy
§ISOF-CNR Sezione di Ferrara, Via L. Borsari 40, 44100 Ferrara, Italy
⊥Department of Chemistry, University of Ferrara, Via L. Borsari 46, 44100 Ferrara, Italy
S
* Supporting Information
ABSTRACT: We demonstrate here that, whereas the rhenium(I)−
zinc porphyrin dyad fac-[Re(CO)3(bpy)(Zn·4′MPyP)](CF3SO3) [1;
4′MPyP = 5-(4′-pyridyl)-10,15,20-triphenylporphyrin] shows no
evidence for photoinduced electron transfer upon excitation in the
visible region because the charge-separated state ZnP+−Re− is almost
isoenergetic with the singlet excited state of the zinc porphyrin (ΔG
= −0.05 eV), the introduction of electron-withdrawing ethyl ester
groups on the bpy ligand significantly improves the thermodynamics
of the process (ΔG = −0.42 eV). As a consequence, in the new dyad
fac-[Re(CO)3(4,4′-DEC-bpy)(Zn·4′MPyP)](CF3SO3) (4; 4,4′-
DEC-bpy = 4,4′-diethoxycarbonyl-2,2′-bipyridine), an efficient and
ultrafast intramolecular electron-transfer process occurs from the
excited zinc porphyrin to the rhenium unit upon excitation with
visible light. Conversely, the introduction of electron-donor tert-butyl groups on the meso-phenyl moieties of the zinc porphyrin
has a negligible effect on the photophysics of the system. For dyad 4, the time constants for the charge-separation and charge-
recombination processes in solvents of different polarity (PrCN, DCM, and toluene) were measured by an ultrafast time-resolved
absorption technique (λexc = 560 nm).
INTRODUCTION
■
In the field of photocatalysis, the preparation of simple, easy-to-
make, and robust dyads and triads is a topic of great research
interest. Such systems, in fact, might afford efficient exploitation
of solar light for performing tantalizing, but so far elusive,
reactions, such as CO2 or water reduction.1 The minimal
system to be used for a photoinduced reduction process is a
Figure 1. Schematic representation of the photoinduced electron-
transfer process in a generic dyad.
dyad formed by a chromophore unit, which must also be a
good electron donor in its excited state (D), connected to a
good electron-acceptor unit (A). In other words, the photo-
active unit must absorb light, preferentially in the visible region,
and then rapidly transfer an electron from its excited state to
electron-donor moieties, is directly coordinated to one or more
peripheral electron-acceptor fac-[Re(CO)3(bpy)]+ fragments.3
One example is fac-[Re(CO)3(bpy)(Zn·4′MPyP)](CF3SO3)
[1; 4′MPyP = 5-(4′-pyridyl)-10,15,20-triphenylporphyrin and
bpy = bipyridine]. These compounds proved to be very stable
in solution, also upon excitation, and were structurally
characterized in solution and in the solid state. However, a
detailed photophysical study showed no evidence for photo-
induced electron transfer in dyad 1 because the charge-
separated state ZnP+−Re− is almost isoenergetic with the
the other fragment; the latter thus becomes a powerful reducing
agent capable of performing a rapid reduction of the substrate
(Figure 1). For an efficient process, the rate of the reduction
must be faster than the recombination step. In particular, dyads
in which a rhenium(I) tricarbonyldiimine complex is bound to
a photosensitizer, of either organic or inorganic nature, that
absorbs in the visible range are actively investigated as catalysts
for the photoreduction of CO2 to CO.2
In this context, in recent years we described simple
rhenium(I) porphyrin conjugates, in which a zinc pyridylpor-
phyrin chromophore, which behaves as both photoactive and
Received: December 5, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Photoinduced Electron-Transfer Process in the Rhenium(I)−Zinc Porphyrin Dyad 1
Figure 2. Functionalized rhenium(I)−zinc porphyrin dyads 2−4 with a numbering scheme for the NMR spectra.
singlet excited state of the zinc porphyrin unit. Nevertheless, in
the presence of pyridine (which binds axially to zinc and thus
affects the porphyrin oxidation potential), a moderately efficient
photoinduced electron-transfer process from the zinc porphyrin
to the Re(bpy) unit was observed upon excitation in the visible
region (Scheme 1). This result led us to predict that an
appropriate functionalization of the bpy ligand and/or of the
porphyrin chromophore, with a judicious choice of electron-
donor and electron-withdrawing substituents, might improve
the thermodynamics, and thus the efficiency, of the photo-
induced electron-transfer process.3
Here we report on the results obtained following this
approach and demonstrate that, whereas the introduction of
electron-donor tert-butyl (tBu) groups (either three or six) on
the meso-phenyl moieties of the zinc porphyrin has a negligible
effect on the photophysics of the system [dyads fac-
[Re(CO)3(bpy)(Zn·4′MPytBuP)](CF3SO3) (2) and fac-[Re-
(CO)3(bpy)(Zn·4′MPydtBuP)](CF3SO3) (3); Figure 2], the
introduction of electron-withdrawing ethyl ester groups on the
bpy ligand led to the expected results, and the dyad fac-
[Re(CO)3(4,4′-DEC-bpy)(Zn·4′MPyP)](CF3SO3) (4; 4,4′-
DEC-bpy = 4,4′-diethoxycarbonyl-2,2′-bipyridine; Figure 2)
efficiently yields the charge-separated state ZnP+−Re− upon
excitation with visible light.
EXPERIMENTAL SECTION
■
Instrumental Methods. Mono- and bidimensional (H−H COSY)
1H NMR spectra were recorded on a JEOL Eclipse 400FT (400 MHz)
or a Varian 500 (500 MHz) spectrometer. All spectra were run at
1
room temperature; H chemical shifts were referenced to the peak of
residual nondeuterated solvent (δ = 7.26 ppm for CDCl3, 2.04 ppm for
acetone-d6, 3.30 ppm for CD3OD, and 4.33 ppm for CD3NO2).
Electrospray ionization mass spectrometry (ESI-MS) measurements
were performed on a Perkin-Elmer APII spectrometer at 5600 eV by
Dr. Fabio Hollan, Department of Chemical and Pharmaceutical
Sciences, University of Trieste, Italy. Solution (CHCl3) IR spectra
were recorded in 0.1 mm cells with NaCl windows on a Perkin-Elmer
Fourier transform infrared/Raman 2000 instrument in the trans-
mission mode. UV/vis absorption spectra were recorded with a Jasco
V-570 UV/vis/near-IR spectrophotometer. Emission spectra were
acquired on a Spex-Jobin Ivon Fluoromax-2 spectrofluorimeter,
equipped with Hamamatsu R3896 tubes. Nanosecond emission
lifetimes were measured using a TCSPC apparatus (PicoQuant
Picoharp300) equipped with subnanosecond LED sources (280−600
nm range; 500−700 ps pulse width) powered with a PicoQuant PDL
800-B variable-pulsed (2.5−40 MHz) power supply. The decays were
analyzed by means of PicoQuant FluoFit Global Fluorescence Decay
Analysis Software. Femtosecond time-resolved experiments were
performed using a pump−probe setup based on a Spectra-Physics
Hurricane Ti:sapphire laser source and an Ultrafast Systems Helios
spectrometer.4 The 560 nm pump pulses were generated with a
Spectra Physics 800 optical parametric amplifier. Probe pulses were
B
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
obtained by continuum generation on a sapphire plate (useful spectral
range: 450−800 nm). The effective time resolution was ca. 300 fs, the
temporal chirp over the white-light 450−750 nm range was ca. 200 fs,
and the temporal window of the optical delay stage was 0−1000 ps.
The time-resolved spectral data were analyzed with Ultrafast Systems
Surface Explorer Pro software.
Crystallographic data for compound 2 were collected at the X-ray
diffraction beamline of the synchrotron Elettra (Trieste) at 100 K, λ =
1.0000 Å; those for fac-[Re(CO)3(4,4′-DEC-bpy)(py)](CF3SO3) (7)
were collected on a Nonius DIP-1030H single-crystal diffractometer
(Mo Kα radiation, λ = 0.71073 Å) at room temperature. Cell
refinement, indexing, and scaling of the data sets were performed using
programs Denzo and Scalepack.5 Both structures were solved by direct
methods and subsequent Fourier analyses6 and refined by the full-
matrix least-squares method based on F2 with all observed reflections.6
The triflate anion in 2 was found to be disordered over two positions,
each refined with an occupancy of 0.50. Hydrogen atoms were placed
at calculated positions. All of the calculations were made using WinGX,
version 1.80.05.7
Materials. Tetraphenylporphyrin (TPP) and the corresponding
tBu-substituted TPPs 5,10,15,20-tetrakis(4′-tert-butylphenyl)-
porphyrin (tBuTPP) and 5,10,15,20-tetrakis(3′,5′-tert-butylphenyl)-
porphyrin (dtBuTPP) and 5-(4′-pyridyl)-10,15,20-triphenylporphyrin
(4′MPyP) and the corresponding tBu-substituted MPyPs 5-(4′-
pyridyl)-10,15,20-tris(4′-tert-butylphenyl)porphyrin (4′MPytBuP)
and 5-(4′-pyridyl)-10,15,20-tris(3′,5′-tert-butylphenyl)porphyrin
(4′MPydtBuP) were synthesized and purified as described in the
literature.8 The substituted ligand 4,4′-diethoxycarbonyl-2,2′-bipyr-
idine (4,4′-DEC-bpy) was also prepared as described in the literature.9
Synthesis of the Rhenium(I) Complexes and Rhenium(I)
Porphyrin Conjugates. Preparation of the fac-[Re(CO)3(4,4′-DEC-
bpy)(DMSO-O)](CF3SO3) (6) precursor was similar to that
previously described by us for fac-[Re(CO)3(bpy)(DMSO-O)]-
(CF3SO3) (5).10 The dyads 2 and 3 were prepared using procedures
similar to that described by us for 1.3 The dyad 4 and model
compounds 7 and fac-[Re(CO)3(bpy)(py)](CF3SO3) were similarly
prepared from complexes 6 and 5, respectively.
fac-[Re(CO)3(bpy)(4′MPytBuP)](CF3SO3) (8). 4′MPytBuP (90
mg, 0.12 mmol) and 5 (60 mg, 0.09 mmol) were dissolved in CHCl3
(25 mL). The mixture was heated to reflux for 8 h and then
concentrated in vacuo to ca. 15 mL. The product precipitated as a
purple solid upon the addition of diethyl ether until saturation and was
collected by filtration, washed with diethyl ether, and vacuum-dried.
Yield: 87 mg (71%). 1H NMR (δ, 400 MHz, acetone-d6): 9.70 (d, J =
5.3 Hz, 2H, H6,6′), 9.03 (d, J = 6.4 Hz, 2H, H2,6), 8.95 (d, J = 8.3 Hz,
2H, H3,3′), 8.85 (m, 8H, βH), 8.61 (t, J = 7.7 Hz, 2H, H4,4′), 8.34 (d,
J = 6.5 Hz, 2H, H3,5), 8.18 (d, J = 8.1 Hz, 2H, oH), 8.15 (d, J = 8.0
Hz, 4H, oH), 8.12 (t, J = 8.0 Hz, 2H, H5,5′), 7.85 (m, 6H, mH), 1.60
(s, 27H, tBu), −2.86 (s, 2H, NH). ESI-MS. Calcd for C68H61N7O3Re
([M − CF3SO3]+): 1210.5. Found: 1210.4. IR (cm−1, CHCl3): 2038
(s, νCO fac
̃
), 1930 (s, br, νCO fac
̃
).
fac-[Re(CO)3(bpy)(4′MPydtBuP)](CF3SO3) (9). 4′MPydtBuP
(180 mg, 0.18 mmol) and 5 (90 mg, 0.14 mmol) were dissolved in
CHCl3 (20 mL). The mixture was heated to reflux for 8 h and then
concentrated in vacuo to ca. 8 mL. The dropwise addition of n-hexane
(5 mL) induced precipitation of the product as a microcrystalline
purple solid, which was collected by filtration and recrystallized from
acetone/n-hexane. Yield: 150 mg (76%). 1H NMR (δ, 400 MHz,
CDCl3): 9.26 (d, J = 4.7 Hz, 2H, H6,6′), 9.17 (d, J = 8.4 Hz, 2H,
H3,3′), 8.90 (m, 6H, βH), 8.61 (d, 2 J = 4.9 Hz, H, βH), 8.53 (d, J =
6.5 Hz, 2H, H2,6), 8.48 (t, J = 7.2 Hz, 2H, H4,4′), 8.21 (d, J = 6.5 Hz,
2H, H3,5), 8.03 (d, J = 1.8 Hz, 2H, oH), 8.01 (d, J = 1.8 Hz, 4H, oH),
7.83 (t, J = 6.5 Hz 2H, H5,5′), 7.78 (m, 3H, pH), 1.50 (s, 54H, tBu),
−2.78 (s, 2H, NH). ESI-MS. Calcd for C80H85N7O3Re ([M −
CF3SO3]+): 1378.6. Found: 1378.9. IR (cm−1, CHCl3): 2038 (s,
νCO fac
̃
), 1930 (s, br, νCO fac
̃
).
fac-[Re(CO)3(4,4′-DEC-bpy)(4′MPyP)](CF3SO3) (10). 4′MPyP
(50 mg, 0.09 mmol) and 6 (60 mg, 0.08 mmol) were dissolved in
CHCl3 (15 mL), and the solution was heated to reflux for 8 h. After
concentration to ca. 8 mL, n-hexane (10 mL) was added to induce
precipitation of the product as a purple solid, which was collected by
filtration, washed thoroughly with n-hexane, and vacuum-dried. Yield:
1
70 mg (72%). H NMR (δ, 400 MHz, CDCl3): 9.58 (d, J = 5.6 Hz,
2H, H6,6′), 9.15 (s, 2H, H3,3′), 8.83 (m, 6H, βH), 8.76 (d, J = 6.2 Hz,
2H, H2,6), 8.65 (d, J = 3.3 Hz, 2H, βH), 8.43 (d, J = 4.7 Hz, 2H,
H5,5′), 8.28 (d, J = 6.1 Hz, 2H, H3,5), 8.19 (d, J = 8.0 Hz, 2H, oH),
8.17 (d, J = 7.5 Hz, 4H, oH), 7.74 (m, 9H, m+pH), 4.55 (q, J = 7.1 Hz,
4H, CH2), 1.47 (t, J = 7.1 Hz, 6H, CH3), −2.88 (s, 2H, NH). ESI-MS.
Calcd for C62H45N7O7Re ([M − CF3SO3]+): 1186.3. Found: 1186.5.
fac-[Re(CO)3(4,4′-DEC-bpy)(DMSO-O)](CF3SO3) (6). fac-[Re-
(CO)3(DMSO-O)3](CF3SO3) (100 mg, 0.15 mmol) and 4,4′-DEC-
bpy (50 mg, 0.17 mmol) were dissolved in acetone (15 mL). The
solution, initially colorless, was refluxed for 1 h under an argon
atmosphere and turned yellow. Concentration to ca. 7 mL followed by
the addition of n-hexane (3 mL) induced in a few days the formation
of microcrystals, which were removed by filtration, washed with n-
IR (cm−1, CHCl3): 2037 (s, νCO fac
νCO).
Zinc Adducts. The insertion of zinc into the free-base porphyrins
̃
), 1931 (s, br, νCO fac
̃
), 1733 (m,
̃
1
hexane, and vacuum-dried. Yield: 77 mg (65%). H NMR (δ, 400
MHz, CD3NO2): 9.32 (d, J = 5.7 Hz, 2H, H6,6′), 9.08 (s, 2H, H3,3′),
8.25 (d, J = 5.6 Hz, 2H, H5,5′), 4.52 (q, J = 7.1 Hz, 4H, CH2), 2.60 (s,
6H, DMSO-O), 1.45 (t, J = 7.1 Hz, 6H, CH3).
and rhenium(I) pyridylporphyrin conjugates was performed according
to this general procedure: a concentrated CHCl3 solution of each
adduct was treated overnight with an 8-fold molar excess of
Zn(CH3COO)2·2H2O dissolved in a minimum amount of methanol
(MeOH). The solvent was evaporated in vacuo and the solid
redissolved in CHCl3. The product was precipitated by the addition of
n-hexane, removed by filtration, washed thoroughly with cold MeOH
and diethyl ether, and vacuum-dried. Yield: >85%. The compounds
fac-[Re(CO)3(4,4′-DEC-bpy)(py)](CF3SO3) (7). To a solution of
complex 6 (50 mg, 0.06 mmol) in CH2Cl2 (20 mL) was added
pyridine (30 μL, 0.3 mmol). The solution was stirred for 1 day at room
temperature and then concentrated in vacuo to ca. 5 mL. Yellow
microcrystals of pure product formed upon the dropwise addition of n-
hexane until saturation and were removed by filtration, washed with n-
hexane, and vacuum-dried. Yield: 85%. Crystals of 7 suitable for X-ray
diffraction were obtained by the slow diffusion of n-hexane into a
1
were characterized by H NMR and IR spectroscopy and ESI-MS.
fac-[Re(CO)3(bpy)(Zn·4′MPytBuP)](CF3SO3) (2). Yield: 85%. H
1
NMR (δ, 400 MHz, CD3OD): 9.54 (d, J = 5.6 Hz, 2H, H6,6′), 8.85
(d, J = 4.7 Hz, 2H, H3,3′), 8.82 (m, 6H, βH), 8.76 (d, J = 6.5 Hz, 2H,
H2,6), 8.57 (d, J = 4.7 Hz, 2H, βH), 8.44 (t, J = 8.0 Hz, 2H, H4,4′),
8.19 (d, J = 6.5 Hz, 2H, H3,5), 8.07 (d, J = 5.9 Hz, 2H, oH), 8.05 (d, J
= 5.9 Hz, 4H, oH), 7.97 (t, J = 6.7 Hz, 2H, H5,5′), 7.77 (m, 6H, mH),
1.60 (s, 18H, tBu), 1.59 (s, 9H, tBu). ESI-MS. Calcd for
C68H59N7O3ZnRe ([M − CF3SO3]+): 1272.3. Found: 1272.5. IR
1
dichloromethane (DCM) solution of the complex. H NMR (δ, 400
MHz, CDCl3): 9.27 (d, J = 5.6 Hz, 2H, H6,6′), 9.02 (s, 2H, H3,3′),
8.31 (m, 4H, H5,5′ + H2,6 py), 7.83 (t, J = 7.7 Hz, 1H, H4 py), 7.47
(t, J = 7.6 Hz, 2H, H3,5 py), 4.57 (q, J = 7.2 Hz, 4H, CH2), 1.50 (t, J =
7.2 Hz, 6H, CH3). IR (cm−1, CHCl3): 2038 (s, νCO fac
̃
), 1945 (s, br,
νCO fac
̃
), 1935 (s, br, νCO fac
̃
), 1726 (m, νCO
̃
).
Crystal data: C25H21F3N3O10ReS, MW = 798.71, triclinic, space
(cm−1, CHCl ): 2036 (s, ν
Crystals of 2 suitable for X-ray diffraction were obtained by the slow
diffusion of n-hexane into a DCM solution of the compound. Crystal
̃
̃
), 1932 (s, br, νCO fac).
3
CO fac
group P1, a = 10.710(2) Å, b = 15.457(2) Å, c = 18.150(3) Å, α =
̅
80.300(11)°, β = 80.564(12)°, γ = 79.363(9)°, V = 2883.3(8) Å3, Z =
4, Dc = 1.840 g/cm3, μ(Mo Kα) = 4.366 mm−1, F(000) = 1560, θ
range = 1.35−29.13°. Final R1 = 0.0434, wR2 = 0.1148, and S = 0.937
for 779 parameters and 15035 unique reflections [R(int) = 0.0420], of
which 9292 with I > 2σ(I), max positive and negative peaks in ΔF map
1.048 and −0.935 e/Å3.
data: C H F N O ReSZn, MW = 1422.86, triclinic, space group P1,
̅
69 59
3
7
6
a = 8.7260(8) Å, b = 9.8370(9) Å, c = 40.956(3) Å, α = 84.227(10)°, β
= 84.429(11)°, γ = 72.995(7)°, V = 3336.3(5) Å3, Z = 2, Dc = 1.416 g/
cm3, μ = 4.750 mm−1, F(000) = 1436, θ range = 2.11−26.76°. Final R1
= 0.0925, wR2 = 0.2276, and S = 1.194 for 838 parameters and 5036
C
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 2. Synthesis of Dyad 4, with the Yield of Isolated Product in Each Step
a
Reagents and conditions: (i) 4,4′-DEC-bpy, acetone, reflux; (ii) 4′MPyP, CHCl3, reflux; (iii) Zn(OAc)2·2H2O, CHCl3/MeOH, room temperature.
unique reflections [R(int) = 0.0640], of which 3766 with I > 2σ(I),
max positive and negative peaks in ΔF map 1.538 (close to the
disordered triflate) and −1.731 e/Å3.
fac-[Re(CO)3(bpy)(Zn·4′MPydtBuP)](CF3SO3) (3). Yield: 87%.
1H NMR (δ, 500 MHz, CD3OD): 9.54 (d, J = 5.7 Hz, 2H, H6,6′), 8.83
(m, 6H, βH), 8.77 (m, 4H, H2,6 + H3,3′), 8.60 (d, J = 4.7 Hz, 2H,
βH), 8.45 (t, J = 8.2 Hz, 2H, H4,4′), 8.23 (d, J = 6.5 Hz, 2H, H3,5),
8.07 (d, J = 1.7 Hz, 2H, oH), 8.05 (d, J = 1.7 Hz, 4H, oH), 7.98 (t, J =
6.7 Hz, 2H, H5,5′), 7.89 (m, 3H, pH), 1.54 (s, 54H, tBu). ESI-MS.
Calcd for C80H83N7O3ZnRe ([M − CF3SO3]+): 1440.5. Found:
1440.5. IR (cm−1, CHCl3): 2036 (s, νCO fac
1926 (s, br, νCO fac).
̃
), 1939 (s, br, νCO fac
̃
),
̃
fac-[Re(CO)3(4,4′-DEC-bpy)(Zn·4′MPyP)](CF3SO3) (4). Yield:
92%. 1H NMR (δ, 400 MHz, CDCl3): 9.49 (d, J = 5.7 Hz, 2H,
H6,6′), 9.12 (s, 2H, H3,3′), 8.90 (m, 6H, βH), 8.69 (d, J = 5.1 Hz, 2H,
βH), 8.59 (d, J = 6.4 Hz, 2H, H2,6), 8.38 (d, J = 5.5 Hz, 2H, H5,5′),
8.28 (d, J = 6.1 Hz, 2H, H3,5), 8.18 (d, J = 8.2 Hz, 2H, oH), 8.16 (d, J
= 7.2 Hz, 4H, oH), 7.73 (m, 9H, m+pH), 4.52 (q, 4H, CH2), 1.47 (t,
6H, CH3). ESI-MS. Calcd for C62H43N7O7ZnRe ([M − CF3SO3]+):
1
Figure 3. Downfield region of the H NMR spectrum of conjugate 4
(CDCl3, 400 MHz). See Figure 2 for the labeling scheme.
1248.2. Found: 1248.4. IR (cm−1, CHCl3): 2038 (s, νCO fac
̃
), 1945 (s,
upfield compared to the unbound porphyrin because the
protons fall into the shielding cone of the adjacent bpy ligand.10
The X-ray molecular structure of the cation of 2 is shown in
Figure 4. The coordination bond lengths and angles are
unexceptional and in agreement with those of similar
rhenium(I) porphyrin conjugates previously described by us.3
As a model for the rhenium acceptor unit, we also prepared
complex 7 by treatment of the common intermediate 6 with a
slight excess of pyridine in a DCM solution at room
temperature. Complex 7 was characterized in solution by H
NMR spectroscopy and in the solid state by X-ray
crystallography (see the Supporting Information).
Electrochemical Behavior of Dyads 3 and 4. The
electrochemical properties of dyads 3 and 4 were examined by
cyclic voltammetry (CV) in a DCM solution [electrochemical
window from +1.0 to −1.5 V, with saturated calomel electrode
(SCE) as the reference electrode], focusing on the potential
values for the first oxidation and reduction processes. For
br, νCO fac
̃
), 1935 (s, br, νCO fac
̃
), 1726 (m, νCO
̃
).
RESULTS AND DISCUSSION
■
The new dyads 2−4 were conveniently prepared via an efficient
stepwise procedure from the cationic rhenium(I) precursor fac-
[Re(CO)3(DMSO-O)3](CF3SO3); the bpy (or substituted
bpy) ligand was introduced first, followed by the pyridylpor-
phyrin. The last step was the insertion of zinc into the
porphyrin core of the conjugate. As an example, the synthesis of
dyad 4 is reported in Scheme 2.
1
Dyads 2−4 were characterized by 1H NMR spectroscopy and
ESI-MS. Figure 3 shows, as an example, the 1H NMR spectrum
of conjugate 4; all the resonances were assigned by means of
conventional H−H COSY experiments. It is worth noting that
the resonance of the two pyridyl protons adjacent to the
nitrogen atom (H2,6), which typically is shifted downfield upon
binding to the {fac-Re(CO)3}+ fragment, in this case is shifted
D
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
almost identical also in the dyads: compare the oxidation
potentials of 1 and 3 and the reduction potentials of 1 and 4.
Photophysical Behavior of Dyads 2−4. A detailed
photophysical investigation on the new dyads 2−4 was carried
out in a DCM solution by stationary and time-resolved
absorption and emission spectroscopy. For dyad 4, some
kinetic measurements were also performed in propionitrile and
toluene for comparative purposes.
The absorption spectra of the three dyads are practically
identical and dominated in the whole spectral region by the
zinc porphyrin component, with two typical Q bands between
500 and 700 nm in addition to the Soret band at ca. 400 nm
(see the SI). The rhenium unit does not absorb in the visible
region, while in the UV region, it exhibits a metal-to-ligand
charge-transfer Re → bipy band at ca. 350 nm, hidden by the
porphyrin Soret band, and intense ligand-centered transitions
below 330 nm (see the SI).
The results of the emission experiments are collected in
Table 2. Following excitation in the visible region, dyads 2 and
Figure 4. Molecular structure of the complex cation of 2 with a
selected atom-labeling scheme. Coordination bond distances (Å): Zn−
N1 2.031(12), Zn−N2 2.058(16), Zn−N3 2.020(13), Zn−N4
2.026(16), Re−C1x 1.741(18), Re−C2x 1.727(19), Re−C3x
1.715(15), Re−N5 2.094(16), Re−N6 2.19(2), Re−N7 2.09(2).
Table 2. Emission Properties of Dyads 1−4 and of the Model
a
Compounds
b
c
b
compound
λmax(em), nm
Φ0/Φ
τ, ns
τ0/τ
Zn·TPP
599, 645
601, 648
597, 647
607, 650
613, 650
610, 650
596, 644
1.7
1.6
Zn·tBuTPP
Zn·dtBuTPP
comparison, the zincated meso-tetraphenylporphyrins (Zn·TPP,
Zn·tBuTPP, and Zn·dtBuTPP) and the rhenium model
complexes fac-[Re(CO)3(bpy)(py)](CF3SO3) and 7 were
studied in the same experimental conditions. The results are
summarized in Table 1. The data for dyad 1, previously
investigated by us,3b are also reported for comparative
purposes. The results obtained for the reference porphyrins
(Table 1) clearly show that the introduction of the electron-
donor tBu substituents on the meso-phenyl rings has only a
small effect on the oxidation potential of the porphyrin.
Conversely, the presence of the electron-withdrawing ethyl
ester groups on the bpy ligand causes a large positive shift (ca.
0.4 V) in the reduction potential of the rhenium unit, which
thus becomes significantly easier to reduce. In the cyclic
voltammogram of dyad 4, the first oxidation process is assigned
to zinc porphyrin, whereas the first reduction process involves
the bpy ligand of the rhenium unit (see the SI). As expected by
a comparison with similar systems,3,11 this voltammetric
behavior is a good superposition of those of the molecular
components, indicating the supramolecular nature of the
conjugates investigated. Accordingly, the effects induced by
functionalization on the model compounds are found to be
1.7
d
1
1.6
2.0
1.9
8.3
0.9
2.0
2.3
2
3
4
0.7
0.6
2.8
<0.25
>6.8
a
Selective excitation of the zinc porphyrin component in the Q-band
b
region (λexc = 550 nm), room temperature, and DCM solution. Φ0
and τ0 are respectively the fluorescence quantum yield and lifetime of
c
the corresponding zinc porphyrin model. Estimated error: 0.1 ns.
d
From ref 3.
3 showed typical porphyrin-based fluorescence. Comparative
experiments carried out on optically matched solutions of 2 and
3g and of the corresponding zinc porphyrin models at an
excitation wavelength of 550 nm, clearly evidenced that the
moderate fluorescence quenching observed for the rhenium
dyads (both intensity and lifetime) is attributable, as previously
discussed for 1,3 to the heavy-atom effect. These results are
consistent with CV measurements (Table 1; compare 1 and 3)
showing that the introduction of the electron-donor tBu
substituents on the meso-phenyl rings has little effect on the
a
Table 1. Electrochemical Data for Dyads 3 and 4 and for the Reference Compounds
b
c
compound
E1/2(red), V
E1/2(ox), V
E0−0(S1), eV
ΔG, eV
Zn·TPP
Zn·tBuTPP
Zn·dtBuTPP
−1.38
−1.50
−1.50
−1.13
−0.72
−1.16
−1.15
−0.71
+0.80
+0.75
+0.74
fac-[Re(CO)3(bpy)(py)](CF3SO3)
fac-[Re(CO)3(4,4′-DEC-bpy)(py)](CF3SO3) (7)
fac-[Re(CO)3(bpy)(Zn·4′MPyP)](CF3SO3) (1)
fac-[Re(CO)3(bpy)(Zn·4′MPydtBuP)](CF3SO3) (3)
fac-[Re(CO)3(4,4′-DEC-bpy)(Zn·4′MPyP)](CF3SO3) (4)
d
+0.83
+0.79
+0.88
2.04
2.03
2.02
−0.05
−0.09
−0.42
a
All measurements were made in argon-purged DCM solutions at 298 K [0.1 M TBA(PF6) as the supporting electrolyte, scan rate 200 mV/s, SCE
as the reference electrode, and glassy carbon as the working electrode]. Half-wave potentials calculated as an average of the cathodic and anodic
b
c
peaks (ΔEp = 60−80 mV). Estimated from the fluorescence spectrum. Free-energy change for photoinduced electron transfer from the excited
(S1) zinc porphyrin unit to the rhenium unit.17 From ref 3b.
d
E
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Transient spectral changes obtained for dyad 4 in a DCM solution (λexc= 560 nm) in the 0−10 ps range (left) and in the 10−400 ps range
(right).
oxidation potential of the porphyrin. As a consequence, the
driving force for the photoinduced electron-transfer process
from the singlet excited state of zinc porphyrin to the rhenium
unit improves negligibly on going from 1 to 3 (ΔG = −0.05 eV
for 1 vs ΔG = −0.09 for 3). Accordingly, also in dyads 2 and 3,
this process does not occur. By contrast, a strong quenching of
the fluorescence intensity and lifetime (with respect to the
reference Zn·TPP) was observed for dyad 4 (see the SI and
Table 2).12 The lifetime of the residual emission is shorter
(<250 ps) than the instrumental response. These results clearly
indicate that in dyad 4 there is an additional and efficient
(100%) process for deactivation of the singlet excited state of
zinc porphyrin that involves the rhenium(I) unit.
Ultrafast absorption measurements were performed to obtain
information on the quenching mechanism. When excitation of
4 was carried out at 560 nm, where light is selectively absorbed
by the zinc porphyrin, the spectral changes shown in Figure 5
Figure 6. Kinetic analysis of the changes of the transient spectrum of 4
at 660 nm in DCM.
were observed. The behavior is clearly biphasic, with different
that in dyad 4, in which 4,4′-DEC-bpy replaces bpy on the
spectral changes taking place in the 0−10 and 10−400 ps time
rhenium(I) acceptor fragment (compare 1 and 4), an efficient
ranges (Figure 5). The initial spectrum of Figure 5, taken
and ultrafast intramolecular electron-transfer process occurs
immediately after the excitation pulse (t = 0.9 ps), is typical for
from the excited zinc porphyrin to the rhenium unit. These
results can be explained on the basis of the energy level diagram
the zinc porphyrin singlet excited state:13 a broad intense
(Figure 7) in which the charge-separated state Zn−Re−, whose
energy is calculated at 1.6 eV from the electrochemical results
(Table 1), lies lower than the local singlet state (S1) of zinc
porphyrin [E(S1) = 2.02 eV, estimated from the fluorescence
spectrum]. Thus, in dyad 4 the photoinduced electron-transfer
process from excited zinc porphyrin to the rhenium unit has
ΔG = −0.42 eV (i.e., ca. 0.4 eV more negative than that for 1−
3). Consistent with the fast rate measured by ultrafast
spectroscopy (time constant = 4 ps), the charge-separation
process falls in the normal Marcus region according to the
standard electron-transfer theory.16 Conversely, the rate for the
charge-recombination process to the ground state is much
slower (time constant = 96 ps) because it belongs to the
Marcus inverted region (ΔG = −1.6 eV). Since the energy of
the charge-separated state depends strongly on the solvent
polarity, photophysical characterization of 4 was repeated in a
solvent of higher polarity (propionitrile, PrCN) and in one of
lower polarity (toluene) compared to DCM. Stationary
emission measurements clearly indicated that in both solvents,
as in DCM, fluorescence of zinc porphyrin was almost
completely quenched. Ultrafast laser experiments in PrCN
and toluene gave transient spectral changes qualitatively similar
to those observed in DCM (see the SI), indicating that, in each
positive absorption in the 450−550 nm region and a relatively
flat positive absorption throughout the 550−750 nm visible
region with a superimposed bleaching of the ground-state Q
bands at 565 (hidden within the excitation window) and 610
nm. Additional apparent bleaches at 625 and 670 nm are caused
by stimulated emission. In the 1−10 ps time interval (Figure 5,
left), the spectral changes show a substantial recovery of the
bleach in the 520−680 nm range, accompanied by an increase
in the optical density of the positive absorption in the 620−700
nm range, where at 660 nm the typical absorption band of the
radical cation of zinc porphyrin can be easily recognized.14 We
attribute these spectral changes to the formation of the charge-
separated product Zn+−Re− in which the zinc porphyrin
chromophore is oxidized and the rhenium unit is reduced. In
the longer time range (Figure 5, right), the spectroscopic
signatures of the charge-separated state disappear with a
uniform decay of the whole spectrum toward the initial baseline
with isosbestic points at zero differential absorbance. This
clearly indicates that Zn+−Re− converts quantitatively by
charge recombination to the ground state.15 Kinetic analysis
of the spectral changes at 660 nm yields a time constant of 4 ps
for charge separation and 96 ps for the charge-recombination
process (Figure 6). These ultrafast experiments clearly show
F
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
leads to a noticeable increase of the time constants for both
processes. Overall, the observed electron-transfer kinetics is
satisfactorily accounted for in terms of standard electron-
transfer theory.
CONCLUSIONS
■
Ready-to-make and highly robust dyads of the type fac-
[Re(CO)3(bpy)(Zn·pyridylporphyrin)](CF3SO3), in which the
zinc pyridylporphyrin chromophore behaves as photoactive and
electron-donor moieties and the fac-[Re(CO)3(bpy)]+ unit as
the electron acceptor, can be efficiently synthesized in a
stepwise manner from the cationic rhenium(I) precursor fac-
[Re(CO)3(DMSO-O)3](CF3SO3). The modular procedure
affords, with modest synthetic effort, tailor-made dyads whose
properties can be tuned by changing, via appropriate
functionalization, those of the bpy and porphyrin ligands.
As previously anticipated, we demonstrate here that a simple
chemical modification of dyad 1, i.e., functionalization of the
bpy ligand with electron-withdrawing ethyl ester groups in the
4 and 4′ positions, greatly improves, as the rhenium unit
becomes significantly easier to reduce, the thermodynamics of
the photoinduced electron-transfer process (ΔG = −0.42 eV vs
ΔG = −0.05 eV in 1). As a consequence, whereas dyad 1 shows
no evidence for the formation of the charge-separated state
ZnP+−Re− (unless when in the presence of excess pyridine), in
the new dyad 4, an efficient and ultrafast intramolecular
electron-transfer process occurs upon excitation with visible
light. On the contrary, the introduction of electron-donor tBu
groups (either three or six) on the meso-phenyl moieties of zinc
porphyrin has a negligible effect on the photophysics of the
system.
Figure 7. Energy level diagram for dyad 4 in a DCM solution.
case, a fast charge separation (i.e., formation of ZnP+−Re−),
followed by slower recombination, occurred. However, the
kinetics of the processes were found to be solvent-dependent.
This is particularly true for the charge-recombination process,
which slows down with decreasing solvent polarity. Kinetic
analysis of the spectral changes (biexponential fitting performed
at 660 nm; see the SI) gives the values of the time constants for
the charge-separation and charge-recombination processes
reported in Table 3.
Ultrafast time-resolved experiments performed at 560 nm on
dyad 4 in DCM afforded a time constant of 4 ps for the charge-
separation process and 96 ps for the subsequent charge-
recombination process. Moreover, experiments performed in
solvents of higher (propionitrile) or lower polarity (toluene)
compared to DCM consistently suggest that the charge-
separation process occurs in the normal Marcus region, whereas
the charge recombination belongs to the Marcus inverted
region. Clear evidence for the photoinduced electron-transfer
quenching of a zinc porphyrin unit covalently connected to a
Re(bpy) fragment was obtained by Perutz using time-resolved
ultrafast IR spectroscopy.11 As we already pointed out in a
previous paper,3 it is difficult to compare the kinetics of the
electron-transfer processes for the dyad described by Perutz
and 4 because, despite the similarity of the molecular
photoactive components, the two systems are structurally
different. However, in both cases, the lifetime of the charge-
separated state might be too short (tens of picoseconds) to be
useful for chemical reactions. Still, we are presently testing dyad
4 in the reaction of photoreduction of CO2 (in the presence of
an appropriate sacrificial electron donor). Our current work is
also aimed at improving the photophysical properties of these
dyads by further modifying the nature of the ligands, or by
introducing additional redox-active components, in order to
assemble triad systems with the final goal of increasing the
lifetime of the charge-separated state.
Table 3. Time Constants (τ) for the Charge-Separation (CS)
and Charge-Recombination (CR) Processes of Dyad 4 in
Different Solvents (ε is the Dielectric Constant)
a
a
solvent
τCS
,
ps
τCR
,
ps
ε
PrCN
DCM
toluene
5
25
27.7
9.1
4
96
22
130
2.4
a
Experimental error: 10%.
The observed effect of the solvent on the electron-transfer
rates can be qualitatively explained in terms of standard
electron-transfer theory as a consequence of the changes in the
driving force and reorganization energy.16 In DCM, where ΔG
= −0.42 eV, the charge-separation process, assuming that the
reorganization energy has a value of ca. 1 eV, falls in the normal
Marcus region, whereas the charge recombination belongs to
the Marcus inverted region. As the solvent polarity decreases,
the energy of the charge-separated state rises; as a consequence,
the driving force for charge recombination to the ground state
grows, while the reorganization energy diminishes. In the
Marcus inverted region, both factors act in the same direction,
and thus a strong decrease in the rate with decreasing solvent
polarity is predicted for the charge-recombination processes.
On the other hand, for the charge-separation reactions that fall
in the normal Marcus region, the two effects tend to
compensate for each other, leading to prediction of much
weaker solvent effects. The experimental results shown in Table
3 are consistent with these theoretical considerations: whereas
the change from DCM to PrCN affects mainly the
recombination time constant, the low polarity of toluene
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional spectroscopic and electrochemical characterization
(ESI-MS, UV/vis, emission, CV, time-resolved experiments),
decay analysis, and X-ray crystallographic data in CIF format for
G
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
concentrations (0.5 M), was not observed, thus confirming that a
bimolecular electron-transfer process cannot occur because the lifetime
of the zinc porphyrin singlet excited state is too short.
7 and 2. This material is available free of charge via the Internet
at http://pubs.acs.org.
(13) Scandola, F.; Chiorboli, C.; Prodi, A.; Iengo, E.; Alessio, E.
Coord. Chem. Rev. 2006, 250, 1471−1496.
AUTHOR INFORMATION
Corresponding Author
*E-mail: eiengo@units.it (E.I.), idm@unife.it (M.T.I.).
■
(14) Fajer, J.; Borg, D. C.; Forman, A.; Dolphin, D.; Felton, R. H. J.
Am. Chem. Soc. 1970, 92, 3451−3459.
(15) The ultrafast results demonstrate that in 4 the deactivation
Present Address
process from the charge-separated state Zn+−Re−, initially formed in a
‡T.G.: Center for Nano Science and Technology@PoliMi,
Istituto Italiano di Tecnologia, Via G. Pascoli 70/3, 20133
Milano, Italy.
3
singlet spin state, toward formation of the *ZnP−Re triplet does not
favorably compete with the charge recombination to the ground state
because it requires a spin inversion (see Figure 7).
Notes
(16) (a) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811,
265−322. (b) Newton, M. D. Chem. Rev. 1991, 91, 767−792.
(17) ΔG values were obtained as the difference between the energy
of the singlet excited state of the porphyrin unit [E0−0(S1)] and the
energy of the charge-separated state ZnP+−Re− inferred from
electrochemical data (see Figure 7). No correction for the electrostatic
work term is required in this case because of the charge shift character
of the process involved.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from Prin2008 Grant 20085ZXFEE and
Fondazione Beneficentia Stiftung is gratefully acknowledged.
E.I. wants to acknowledge financial support from Reintegration
Grant PERG02-GA-2007-224823. We thank Prof. E. Alessio
and Prof. F. Scandola for helpful discussions.
REFERENCES
■
(1) (a) Barber, J. Chem. Soc. Rev. 2009, 38, 185−196. (b) Gust, D.;
Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2009, 42, 1890−1898.
(c) Fontacave, A. F.; Artero, V.; Pereira, A.; Fontacave, M. Dalton
Trans. 2008, 5567−5569. (d) Li, X.; Wang, M.; Zhang, S.; Pan, J.; Na,
́
Y.; Liu, J.; Åkermark, B.; Sun, L. J. Phys. Chem. B 2008, 112, 8198−
8202. (e) Esswein, A. J.; Nocera, D. G. Chem. Rev. 2007, 107, 4022−
4047.
(2) (a) Agarwal, J.; Fujita, E.; Schaefer, H. F.; Muckerman, J. T. J. Am.
Chem. Soc. 2012, 134, 5180−5186. (b) Windle, C. D.; Campian, M. V.;
̂
Duhme-Klair, A.-K.; Gibson, E. A.; Perutz, R. N.; Schneider, J. Chem.
Commun. 2012, 48, 8189−8191. (c) Yamamoto, Y.; Tamaki, Y.; Yui,
T.; Koike, K.; Ishitani, O. J. Am. Chem. Soc. 2012, 132, 11743−11752.
(d) Schneider, J.; Vuong, K. Q.; Calladine, J. A.; Sun, X.-Z.; Whitwood,
A. C.; George, M. W.; Perutz, R. N. Inorg. Chem. 2011, 50, 11877−
11889. (e) 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. (f) Morris, A. J.; Meyer, G. J.;
Fujita, E. Acc. Chem. Res. 2009, 42, 1983−1994.
(3) (a) Ghirotti, M.; Chiorboli, C.; Indelli, M. T.; Scandola, F.;
Casanova, M.; Iengo, E.; Alessio, E. Inorg. Chim. Acta 2007, 360,
1121−1130. (b) Casanova, M.; Zangrando, E.; Iengo, E.; Alessio, E.;
Indelli, M. T.; Scandola, F.; Orlandi, M. Inorg. Chem. 2008, 47,
10407−10418.
(4) Chiorboli, C.; Rodger, M. A. J.; Scandola, F. J. Am. Chem. Soc.
2003, 125, 483−491.
(5) Otwinowski, Z.; Minor, W. In Processing of X-ray Diffraction
Data Collected in Oscillation Mode. Methods in Enzymology,
Macromolecular Crystallography; Carter, C. W., Jr., Sweet, R. M.,
Eds.; Academic Press: New York, 1997; Vol. 276, Part A, pp 307−326.
(6) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, A64, 112−122.
(7) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838.
(8) (a) Fleischer, E. B.; Sacther, A. M. Inorg. Chem. 1991, 30, 3763−
3769. (b) Alessio, E.; Macchi, M.; Heath, S. L.; Marzilli, L. G. Inorg.
Chem. 1997, 36, 5614−5623.
(9) McCafferty, D. G.; Bishop, B. M.; Wall, C. G.; Hughes, S. G.;
Mecklenberg, S. L.; Meyer, T. J.; Erickson, B. W. Tetrahedron 1995, 51,
1093−1106.
(10) Casanova, M.; Zangrando, E.; Munini, F.; Iengo, E.; Alessio, E.
Dalton Trans. 2006, 5033−5045.
(11) (a) Gabrielsson, A.; Hartl, F.; Zhang, H.; Lindsay Smith, J. R.;
Towrie, M.; Vlcek, A., Jr.; Perutz, R. N. J. Am. Chem. Soc. 2006, 128,
4253−4266. (b) Gabrielsson, A.; Lindsay Smith, J. R.; Perutz, R. N.
Dalton Trans. 2008, 4259−4269.
(12) Quenching of the fluorescence of Zn·TPP (or Zn·tBuTPP or
Zn·dtBuTPP) in the presence of complex 7, even at high
H
dx.doi.org/10.1021/ic302663c | Inorg. Chem. XXXX, XXX, XXX−XXX