Electrochemical, computational, and photophysical characterization of new luminescent dirhenium-pyridazine complexes containing bridging or or SR anions

Article abstract of DOI:10.1021/ic202284a

A series of [Re₂(μ-ER)₂(CO)₆(μ-pydz) ] complexes have been synthesized (E = S, R = C₆H₅, 2; E = O, R = C₆F₅, 3; C₆H₅, 4; CH ₃, and 5; H, 6), starting either from [Re(CO)₅O ₃SCF₃] (for 2 and 4), [Re₂(μ-OR) ₃(CO)₆]^- (for 3 and 5), or [Re ₄(μ₃-OH)₄(CO)₁₂] (for 6). Single-crystal diffractometric analysis showed that the two μ-phenolato derivatives (3 and 4) possess an idealized C₂ symmetry, while the μ-benzenethiolato derivative (2) is asymmetrical, because of the different conformation adopted by the phenyl groups. A combined density functional and time-dependent density functional study of the geometry and electronic structure of the complexes showed that the lowest unoccupied molecular orbital (LUMO) and LUMO+1 are the two lowest-lying π* orbitals of pyridazine, whereas the highest occupied molecular orbitals (HOMOs) are mainly constituted by the "t_2g" set of the Re atoms, with a strong Re-(μ-E) π* character. The absorption spectra have been satisfactorily simulated, by computing the lowest singlet excitation energies. All the complexes exhibit one reversible monoelectronic reduction centered on the pyridazine ligand (ranging from -1.35 V to -1.53 V vs Fc⁺|Fc). The benzenethiolato derivative 2 exhibits one reversible two-electron oxidation (at 0.47 V), whereas the OR derivatives show two close monoelectronic oxidation peaks (ranging from 0.85 V to 1.35 V for the first peak). The thioderivative 2 exhibits a very small electrochemical energy gap (1.9 eV, vs 2.38-2.70 eV for the OR derivatives), and it does not show any photoluminescence. The complexes containing OR ligands show from moderate to poor photoluminescence, in the range of 608-708 nm, with quantum yields decreasing (ranging from 5.5% to 0.07%) and lifetimes decreasing (ranging from 550 ns to 9 ns) (3 > 4 > 6 ≈ 5) with increasing emission wavelength. The best emitting properties, which are closely comparable to those of the dichloro complex (1), are exhibited by the pentafluorophenolato derivative (3).

Full text of DOI:10.1021/ic202284a

Article
pubs.acs.org/IC
Electrochemical, Computational, and Photophysical Characterization
of New Luminescent Dirhenium−Pyridazine Complexes Containing
Bridging OR or SR Anions
Alessio Raimondi,† Monica Panigati,*,† Daniela Maggioni,† Laura D’Alfonso,# Pierluigi Mercandelli,*,‡
Patrizia Mussini,*,§ and Giuseppe D’Alfonso†
†Department of Inorganic, Metallorganic and Analytical Chemistry “Lamberto Malatesta”, Universita
̀
degli Studi di Milano, and UdR
INSTM of Milano, Via Venezian 21, 20133 Milano, Italy
‡Department of Structural Chemistry and Inorganic Stereochemistry, Universita
Milano, Italy
̀
degli Studi di Milano, Via Venezian 21, 20133
§Department of Physical Chemistry and Electrochemistry, Universita
̀
degli Studi di Milano, Via Golgi 19, 20133 Milano, Italy
#Department of Physics, Universita
̀
di Milano-Bicocca, Piazza della Scienza 3, 20126 Milano, Italy
S
* Supporting Information
ABSTRACT: A series of [Re2(μ-ER)2(CO)6(μ-pydz)] complexes have been synthesized (E = S, R = C6H5, 2; E = O, R = C6F5,
3; C6H5, 4; CH3, and 5; H, 6), starting either from [Re(CO)5O3SCF3] (for 2 and 4), [Re2(μ-OR)3(CO)6]− (for 3 and 5), or
[Re4(μ3-OH)4(CO)12] (for 6). Single-crystal diffractometric analysis showed that the two μ-phenolato derivatives (3 and 4)
possess an idealized C2 symmetry, while the μ-benzenethiolato derivative (2) is asymmetrical, because of the different
conformation adopted by the phenyl groups. A combined density functional and time-dependent density functional study of the
geometry and electronic structure of the complexes showed that the lowest unoccupied molecular orbital (LUMO) and LUMO
+1 are the two lowest-lying π* orbitals of pyridazine, whereas the highest occupied molecular orbitals (HOMOs) are mainly
constituted by the “t2g” set of the Re atoms, with a strong Re−(μ-E) π* character. The absorption spectra have been satisfactorily
simulated, by computing the lowest singlet excitation energies. All the complexes exhibit one reversible monoelectronic reduction
centered on the pyridazine ligand (ranging from −1.35 V to −1.53 V vs Fc+|Fc). The benzenethiolato derivative 2 exhibits one
reversible two-electron oxidation (at 0.47 V), whereas the OR derivatives show two close monoelectronic oxidation peaks
(ranging from 0.85 V to 1.35 V for the first peak). The thioderivative 2 exhibits a very small electrochemical energy gap (1.9 eV,
vs 2.38−2.70 eV for the OR derivatives), and it does not show any photoluminescence. The complexes containing OR ligands
show from moderate to poor photoluminescence, in the range of 608−708 nm, with quantum yields decreasing (ranging from
5.5% to 0.07%) and lifetimes decreasing (ranging from 550 ns to 9 ns) (3 > 4 > 6 ≈ 5) with increasing emission wavelength. The
best emitting properties, which are closely comparable to those of the dichloro complex (1), are exhibited by the
pentafluorophenolato derivative (3).
transfer (3MLCT) excited states, which can be widely tuned by
INTRODUCTION
■
variations of the ancillary L ligands or of the substituents on the
diimine ligand,3 attaining, in some cases, very high photo-
luminescence quantum yields (PLQY).4 The presence of a
Luminescent transition-metal complexes are currently receiving
much attention, because their favorable properties allow a
variety of uses, particularly in the fields of bioimaging1 and
electroluminescent devices.2
Rhenium carbonyl-diimine complexes, [Re(CO)3(N,N)L],
exhibit photoluminescence from triplet metal-to-ligand-charge
Received: October 21, 2011
Published: February 23, 2012
© 2012 American Chemical Society
2966
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
single electron-accepting diimine ligand, which removes the
problem of the excited-state localization that is ever-present in
other polypyridine systems, makes the rhenium(I) complexes
among the more-investigated organometallic systems, useful
also for fundamental studies.5
Scheme 1. The Different Synthetic Routes (i−v) Leading to
Complexes 1−6
We are currently developing a new family of luminescent
rhenium(I) complexes, with the general formula [Re2(μ-
X)2(CO)6(μ-1,2-diazine)] (where X = hydrides6 or halides;7,8
see Chart 1). Wavelengths, lifetimes, and quantum yields of the
Chart 1
= O or S) and stoichiometric pyridazine (pydz), in the presence
of a base B, according to eq 1.
emission dramatically vary upon varying the diazine sub-
stituents and the ancillary ligands X.6−8 The derivatives with X
= Cl and two alkyl groups in the β positions of the diazine give
the highest PLQYs, up to 0.53 in toluene solution at room
temperature.8 By contrast, the presence of substituents in the α
positions nearly zeroes the emission, because of steric
hindrance with the carbonyl ligands, which reduces the stiffness
of the complex, promoting nonradiative deactivation pathways.7
The replacement of chlorides by bulkier bromides and iodides
also freezes the emission (in solution at room temperature),7 in
agreement with the decreased rigidity of the coordination
sphere upon increasing the bulkiness of the bridging ligands.
Therefore, we decided to investigate the photophysical
properties of diazine complexes bearing polyatomic bridging
ligands, with a twofold objective. First, we were interested in
learning how to exploit changing the ancillary ligands to control
the highest occupied molecular orbital−lowest unoccupied
molecular orbital (HOMO−LUMO) gap, and then modulating
the position and intensity of the emission. Moreover, the raising
of the HOMO level could give low-band-gap complexes, which
are of interest for efficient light harvesting in solar cells.
Second, the replacement of chlorides by polyatomic ligands
might also provide other sites, besides the diazine ring, for
functionalizing the dirhenium complexes, which might be useful
for application as luminescent bioprobes, where the linking to
suitable biomolecules is most-often required.
2[Re(CO) (OTf)] + pydz + 2REH + 2B
5
+
−
→ [Re (μ‐ER) (CO) (μ‐pydz)] + 2BH OTf
2
2
6
(1)
+ 4CO
Bis(dimethylamino)naphthalene (DMAN) was chosen as base,
because it cannot compete with diazine for rhenium
coordination.
This route worked satisfactorily with C6H5SH, in THF
solution, and gave the bis(thiophenolato) derivative [Re2(μ-
SC6H5)2(CO)6(μ-pydz)] (2) in ca. 30% isolated yields. In the
case of C6H5OH, reaction 1 also worked, but cleanly gave
[Re2(μ-OC6H5)2(CO)6(μ-pydz)] (4), using only molten
phenol as a solvent.
In contrast, attempts to obtain the bis(pentafluorophenolato)
derivative [Re2(μ-OC6F5)2(CO)6(μ-pydz)] (3) with the same
procedure afforded a mixture of products in which the expected
derivative was present in relatively low concentration (<30%
from 1H NMR analysis). Then, 3 was better obtained via a two-
step procedure. At first, the dinuclear anion [Re2(μ-
OC6F5)3(CO)6]− was formed, by reacting [Re(CO)5(OTf)]
with pentafluophenol in the presence of a base (B = DMAN,
see eq 2a), then the anion was treated with 1 equiv of pydz,
resulting in the replacement of one phenolate by a diazine
molecule (see eq 2b).
Here, we report on the synthesis and on the electrochemical,
computational, and photophysical characterization of five new
[Re2(μ-ER)2(CO)6(μ-pydz)] complexes, containing pyridazine
(pydz) as a chromophoric ligand and ER− groups as bridging
ancillary ligands (E = S, R = C6H5, 2; E = O, R = C6F5, 3; C6H5,
4; CH3, and 5; H, 6).
2[Re(CO) (OTf)] + 3C F OH + 3B
5
6 5
−
+
→ [Re (μ‐OC F ) (CO) ] + 4CO + 3HB
2
6 5 3
6
−
RESULTS AND DISCUSSION
(2a)
(2b)
+ 2OTf
■
Synthesis of the New Complexes. The previously
reported [Re2(μ-Cl)2(CO)6(μ-pydz)] complex 1, as well as
all the other complexes [Re2(μ-X)2(CO)6(μ-1,2-diazine)] (X =
Cl, Br, I), were obtained by refluxing the proper [Re(CO)5X]
starting material with the stoichiometric amount of 1,2-diazine.
This method cannot be used for the preparation of the title
derivatives, since [Re(CO)5(OR)] complexes are unstable and
elusive species.9 Therefore, we developed alternative routes,
using different starting materials, as summarized in Scheme 1.
In a first approach, the [Re(CO)5(OTf)] complex,10
−
[Re (μ‐OC F ) (CO) ] + pydz
6 5 3
2
6
−
→ [Re (μ‐OC F ) (CO) (μ‐pydz)] + C F O
6 5 2 6 5
2
6
To prepare the [Re2(μ-OCH3)2(CO)6(μ-pydz)] species (5),
we used the dinuclear [Re2(μ-OCH3)3(CO)6]− anion directly
as starting material, which can be prepared in good yields, as
NEt4 salt, from the reaction of [Re2(CO)10] with bases.11
+
containing the labile triflate CF3SO3 anion (OTf−), was
−
Differently from the bis(pentafluorophenolato) derivative, in
this case, the addition of 1 equiv of triflic acid was necessary to
used. The complex was treated with the proper REH reagent (E
2967
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
promote the elimination of one of the bridging methoxo groups
(see eq 3).
−
[Re (μ‐OCH ) (CO) ] + pydz + TfOH
2
3 3
6
→ [Re (μ‐OCH ) (CO) (μ‐pydz)] + CH OH
2
3 2
6
3
−
(3)
+ OTf
This agrees with the higher nucleophilicity and lower bulkiness
−
−
of OCH3 , with respect to OC6F5 .
Finally, a still different method was developed to prepare the
hydroxo [Re2(μ-OH)2(CO)6(μ-pydz)] derivative (6). The
cubane-like [Re4(μ3-OH)4(CO)12] complex12 (Scheme 1)
was used as starting material, and it was reacted with 2 equiv
of pydz, in water solution and under microwave activation,
resulting in the clean formation of 6, according to eq 4.
[Re (μ ‐OH) (CO) ] + 2pydz
4
4
12
3
→ 2[Re (μ‐OH) (CO) (μ‐pydz)]
Figure 1. Crystal structures of [Re2(μ-SC6H5)2(CO)6(μ-pydz)] (2)
and [Re2(μ-OC6F5)2(CO)6(μ-pydz)] (3): frontal and lateral views
(upper line and lower line, respectively). Displacement ellipsoids are
drawn at the 50% probability level. (Color code: C, gray; H, white; F,
lime green; N, blue; O, red; Re, magenta; and S, yellow.)
(4)
2
2
6
This reaction resembles the previously reported fragmentation
of the unsaturated tetranuclear cluster [Re4(μ3-H)4(CO)12]
with 2 equiv of pydz,6 even if, in this case, only the [2 + 2]
fragmentation pathway was observed, quantitatively affording
the dinuclear saturated derivative 6.
structure of 4 is depicted in the Supporting Information (Figure
S3).
The nature of the reaction products was unambiguously
established by the spectroscopic and analytical characterization,
and confirmed by the X-ray analysis of complexes 2, 3, and 4.
All the products exhibit in the ν(CO) regions of the IR spectra
the four bands pattern typical of this class of compounds,7 with
a shift to low wavenumbers upon increasing the donor power of
the bridging ligand (this point will be better discussed below).
In these dinuclear species, each Re atom attains a distorted
octahedral coordination and bears three terminal carbonyl
ligands in a facial arrangement, one of the N atoms of the
bridging pyridazine ligand, and the S (or O) atoms of the two
bridging benzenethiolato (or phenolato) ligands. The two μ-
phenolato derivatives 3 and 4 possess an idealized C2 symmetry
(although only 3 lies on a crystallographic 2-fold axis), while
the thio-derivative 2 is asymmetrical, because of the different
conformation adopted by the phenyl groups of the two bridging
benzenethiolato ligands.
1
The H NMR data agree with the expected structures. In
particular, the two resonances of the bridging pyridazine are
downfield shifted by ∼0.5−0.6 ppm, with respect to the free
ligand, as observed in related complexes.7
Variable-temperature 19F NMR spectra gave some informa-
tion on the solution structure and dynamics of complex 3. The
low-temperature (T < 221 K) spectra are consistent with the C2
solid-state structure (see below), with five different signals for
the five fluorine atoms of the two (equivalent) pentafluor-
ophenolato groups. The chemical-shift difference between the
two ortho resonances (>5 ppm) is much larger than that
between the two meta resonances (<1 ppm), because of the
shielding effect of the diazine π-electrons on one of the ortho F
atoms, as evidenced by the 1H−19F HOESY experiment shown
in Figure S1 in the Supporting Information. Upon increasing
the temperature, the ortho and meta signals broadened (see
Figure S2 in the Supporting Information), indicating the
rotation of each C6F5 ring around the C−O axis, leading to
equalization of the two ortho and two meta positions. Band
shape analysis on the 19F spectra provided the value of the
activation energy for this process (42.4 kJ mol−1). In the case of
the phenolato derivative (4), the 1H NMR spectra did not show
either splitting or broadening of the ortho and meta signals
upon reducing the temperature, indicating that the rotation of
the C6H5 rings is faster than that of their C6F5 counterparts,
because of the lower steric hindrance with the CO ligands.
Solid-State Structures. The crystal structures of the μ-
benzenethiolato derivative (2) and of the two μ-phenolato
derivatives (3 and 4) have been determined. Figure 1 shows
frontal and lateral views of the two complexes 2 and 3; the
Despite the smaller Re−S−Re bond angle in 2 (91.0°), with
respect to the corresponding Re−O−Re angles in 3 and 4
(104.4 and 103.8°), the Re−Re distance is much longer in the
sulfur-containing derivative (3.57 Å in 2 vs 3.39 and 3.35 Å in 3
and 4), as a consequence of the larger covalent radius of sulfur,
compared to oxygen. The Re−S bond distance in 2 (2.50 Å) is
indeed ca. 0.35 Å longer than the Re−O bond distances in 3
and 4 (2.15 and 2.13 Å).13
The coordination geometry of the O atoms of the μ-
phenolato ligands is almost exactly planar (sum of the bond
angles: 359.9° and 358.8° in 3 and 4), suggesting the presence
of a partial double bond character in the Re−O bonds. In
contrast, the S atoms of the μ-benzenethiolato ligands in 2
show a pyramidal geometry (sum of the bond angles: 314.6°).
As a consequence, the phenyl substituents in the thio-derivative
2 are not constrained to lie in the plane defined by the Re−(μ-
S)−Re atoms, leading to the possible existence of geometric
isomers in which the phenyl groups are directed either toward
or away from the μ-pyridazine ligand. Both the orientations can
be observed in the molecular conformation adopted by the
molecule in the solid state (see the lateral view of 2 in Figure
1). However, the presence in solution of other (more
symmetrical) isomers can be foreseen.
Computational Study. The new rhenium complexes 2−6
were studied by means of density functional and time-
dependent density functional (TD-DFT) computations. The
2968
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
Figure 2. Partial molecular orbital diagram for the complexes [Re2(μ-X)2(CO)6(μ-pydz)], X = Cl (1), SC6H5 (2), OC6F5 (3), OC6H5 (4), OCH3
(5), and OH (6). Orbitals are colored according to their main component: rhenium “t2g” orbitals (dark green), pyridazine π* orbitals (light blue),
rhenium “eg” orbitals (light green), and phenyl π orbitals (black, in complexes 2−4). Arrows correspond to electronic transitions involved in the
absorption spectra (computed TD-DFT wavelengths are given). In particular, for each species, values on the left and the right refer to the d−d and
the MLCT absorption, respectively.
optimized structures of complexes 3−6 possess the highest
possible symmetry for these species (C2v), while the
benzenethiolato derivative 2 belongs to the C2 point group.14
As already found for 1,7,8 LUMO and LUMO+1 are the two
lowest-lying π* orbitals of the coordinated pyridazine. The
following four MO (from LUMO+2 to LUMO+5) are the “eg”
set of the two Re atoms, showing, in addition, a large CO π*
character.
For species 5 and 6, the six HOMOs are the “t2g” set of the
two Re atoms in a pseudo-octahedral environment. In
particular, the three highest-lying orbitals show a strong Re−
(μ-O) π* character, in close analogy with the halogenated
derivatives previously described.7 For species 2−4, ten
HOMOs must be taken into account, since the energy of the
“t2g” orbitals of the Re atoms is similar to those of the four
highest-lying π orbitals of the two phenyl groups of the
phenolato (or pentafluorophenolato) bridging ligands. For
species 2, in addition, a partial mixing of these two types of MO
is observed and some of the “t2g” orbitals are partially
delocalized over the phenyl moieties. Molecular orbital energies
are listed in Table S1 in the Supporting Information and partial
orbital diagrams are depicted in Figure 2. A views of the most
relevant molecular orbitals for complex 6 is reported in Figure
S4 in the Supporting Information, while a comparison of the
HOMOs of species 1, 2, 4, and 6 can be found in Figure S5 in
the Supporting Information.
The absorption spectrum of all the complexes, down to 230
nm, was simulated by computing the lowest 50 singlet
excitation energies. Selected transitions are reported in Table
S2 in the Supporting Information and depicted in Figure 2. The
results will be discussed in the section Photophysical
Characterization.
Electrochemical Characterization. Figure 3 shows the
results of cyclic voltammetry (CV) analyses of complexes 2−6
in acetonitrile solution, including, for comparison, the
previously published data concerning the dichloro complex 1
and the analogous dibromo and diiodo derivatives, that
hereafter will be indicated as 1′ and 1″, respectively.7 The
most significant CV features are reported in Table 1, while a
more-extensive selection of the CV data is provided in Table S3
of the Supporting Information, where the voltammograms
recorded in CH2Cl2 solution also are reported (see Figure S6 in
the Supporting Information).
First Reduction Peaks. The first reduction peaks must be
localized for all complexes on their common pyridazine ligand,
consistent with our previous work.7 Accordingly, such peaks are
quite similar for all complexes, both in shape and potential,
while they strongly differ from the free ligand one. In fact, the
first reduction peak of pyridazine, located at a very negative
potential (−2.6 V vs Fc+|Fc), is chemically irreversible and
accounts for an uptake of two electrons,7 while the first
reduction peaks of the complexes are much more positive (by
1.05−1.25 V), implying that the coordination makes the diazine
ligand much more electron-poor. Moreover, the reduction of all
complexes is monoelectronic and reversible, both from the
chemical point of view (symmetrical return peaks, stable
In all the species but 2, the HOMO is a totally symmetric
orbital, analogous to those previously found in the halogenated
derivatives.7 The HOMO of 2 has a very similar composition;
however, it is antisymmetric, with respect to the C2 axis.
2969
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
in agreement with the regular decrease of the diffusion
coefficients of the investigated complexes.
The radical anion formation seems to be significantly favored
by the more-polar CH3CN solvent, with respect to CH2Cl2, in
both thermodynamic and kinetic terms, since the normalized
reduction potentials are significantly less negative, and the
electrochemical reversibility criterion (Ep − Ep/2 ≈ 0.057) are
met much more strictly (see Table S3 in the Supporting
Information).
First Oxidation Peaks. In our former study on dirhenium−
diazine complexes with halide ancillary ligands,7 the first
oxidation site was found to be localized on the metal core, and
the corresponding oxidation peak to be chemically and
electrochemically reversible and bielectronic, pointing to
simultaneous two-electron loss from the dinuclear metal core.
In contrast, in the series of the OR derivatives reported here
(3−6), a close sequence of two monoelectronic peaks is
observed (the number of exchanged electrons being confirmed
by comparison of the oxidation limiting currents with the
reduction ones; see Table S3 in the Supporting Information).
The first peak tends to chemical irreversibility at low scan rates
(no return peak, half-peak widths lower than in the reversible
case, consistent with an EC mechanism), while chemical
reversibility improves with increasing scan rate, at least in the
OCH3 and OH complexes 5 and 6 (see, for instance, the gray
curve in the OH case in Figure 3, corresponding to 2 V s−1).
The much more pronounced reversibility for the OCH3
derivative 5, qualitatively recognizable in Figure 3, has been
confirmed by the quantitative estimate of the pseudo-first-order
rate constants (k′) for the chemical reaction of the product of
the first monoelectronic oxidation step (see the Experimental
Section and Figure S7 in the Supporting Information). In
contrast, the two phenolato derivatives 3 and 4 show a
combination of an irreversible first peak followed by a reversible
second peak, in the entire scan rate range explored. These
features resemble the usual oxidative CV patterns of alkyl−aryl
ethers,15 although it could be only a coincidence.
Figure 3. Normalized CV features for the complexes investigated in
this work and the three complexes with halide ligands previously
investigated 1, 1′, 1″ (green lines),7 recorded on a glassy carbon (GC)
electrode at 0.2 V s−1 in acetonitrile solution. The gray line for 6 refers
to a scan at 2 V s−1.
The picture described here seems to be strictly connected to
the presence of bridging O atoms. Indeed, the SC5H6 derivative
2 shows (in both of the investigated solvents) the reversible,
bielectronic oxidation peak already observed in the halide cases.
Another surprising feature of 2 is the value of its oxidation
potential (0.47 V), which is much lower than that of the
phenolato derivative (4) (1.1 V), despite the close similarity of
their HOMO energies (see Figure 2 and Table S1 in the
Supporting Information). Figure 4a shows a plot of the Ep
values (for compounds 2−6 and the related halide derivatives 1,
product) and the electrochemical one (facile transfer of a single
electron, which is taken into account by the ∼57 mV half-peak
widths together with the almost-zero Ep versus log ν slopes),
thus accounting for very fast formation of a stable radical anion
(see Table S3 in the Supporting Information). The normalized
convoluted peak currents (see the Experimental Section)
regularly decrease with increasing bulkiness of the X ligands,
a
Table 1. Selected CV Features for the [Re2(μ-X)2(CO)6(μ-pydz)] Complexes Investigated Here and in a Previous Work (ref 7)
compound
X
F
R
Ep max, red [V]
Ep max, I ox [V]
Ep max, II ox [V]
LUMO [eV]
HOMO [eV]
Eg [eV]
b
1
Cl
Br
I
0.42
0.45
0.42
0.30
n.a.
−0.19
−0.22
−0.24
−0.23
n.a.
−1.345
−1.365
−1.378
−1.431
−1.352
−1.416
−1.475
−1.527
1.315
1.163
0.944
0.473
1.345
1.100
0.911
0.854
−3.46
−3.44
−3.42
−3.37
−3.45
−3.38
−3.33
−3.27
−6.12
−5.96
−5.74
−5.27
−6.15
−5.90
−5.71
−5.65
2.66
2.53
2.32
1.90
2.70
2.52
2.39
2.38
b
1′
b
1″
2
3
4
5
6
SC6H5
OC6F5
OC6H5
OCH3
OH
1.463
1.254
1.114
1.070
0.37
0.29
0.33
−0.40
−0.56
−0.70
a
Peak potentials Ep (scan rate = 0.2 V s−1); electrochemical HOMO and LUMO energy levels and gaps calculated along the peak maxima. Potentials
are referred to the Fc+|Fc couple in the operating media (MeCN + 0.1 M TBAPF6). Values of the Swain−Lupton parameters F and R are also
b
provided for each group X. Data taken from ref 7.
2970
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
its inductive and resonance components (the F and R
parameters, respectively).16
These findings fully support the view that the oxidation
process is markedly different in the case of the oxygen-bridged
derivatives (which show two monoelectronic oxidation steps),
with respect to the other four complexes, which exhibit a single
bielectronic oxidation step. This might be related to the closer
Re(μ-X)2Re scaffold and to the harder nature of the bridging
ligands in the case of the OR derivatives, which could result in a
less-efficient stabilization of the cationic products, with respect
to the softer anions. A detailed mechanistic electrochemical
investigation will be necessary to throw light into this point.
The last column of Table 1 shows the electrochemical
HOMO−LUMO gap. The energy gap for the SR derivative 2
is, by far, the smallest in the series (lower than 2 eV, both along
the maxima and the onset criteria; see Table S3 in the
Supporting Information for the latter value), while the highest
energy gap is obtained, as expected, with the OC6F5 ligand,
which approximately corresponds to the chloride case in the
halide series. Actually, the effects of Cl and OC6F5 ligands seem
to be almost equivalent, from a thermodynamic point of view
(almost equal HOMO and LUMO levels and first oxidation
and reduction potentials, albeit the oxidative reaction
mechanism is different).
Photophysical Characterization. At room temperature,
in dichloromethane solution, all the complexes show spectra
dominated by two main absorption features in the ultraviolet−
visible (UV−vis) region (see Figure S9 in the Supporting
Information and Figure 5 for complex 3 in different solvents).
Figure 4. Plot of the first oxidation Ep values (measured in acetonitrile
solution) versus (a) the vacuum computed HOMO energies and (b)
the wavenumbers of the highest energy (totally symmetric) ν(CO)
band (acetonitrile solution).
1′, and 1″)7 versus the energies computed for the HOMO in
vacuo. Two trends can be clearly recognized (except the
aberrant behavior of the μ-hydroxo species 6, attributable to the
presence of specific hydrogen bond interactions with the
solvent that are not described by the calculation). The sulfur
derivative 2 perfectly fits with the trend of the complexes
bearing halide ancillary ligands, while the phenolato 4 agrees
with the trend of the OR derivatives.
An impressively similar two-trend plot (Figure 4b) is
obtained by correlating the measured oxidation potentials
with the ν(CO) values of the stretching modes, which are
reliable indicators of the electron density on the metal atom. In
this second correlation, both of the plotted quantities have been
measured in solution and then, in this case, the OH derivatives
6 also fit the trend of the other OR species.
Figure 5. Solvent effect on the absorption () and emission (- - -)
spectra of [Re2(μ-OC6F5)2(CO)6(μ-pydz)] (3). A comparison with
the computed excitation energies and oscillator strengths (vertical
lines) is reported. (Color legend: blue, toluene; red, dichloromethane;
green, acetonitrile.)
On the basis of the TD-DFT computations, the bands at higher
energy, whose position (ca. 270 nm) is independent of the
polarity of the solvent, can be attributed to the superposition of
many singlet d−d excitations from the “t2g” set to the “eg” set of
the two Re atoms. Some π−π* excitations of the pyridazine
ligand fall in the same range of energy (230−300 nm);
however, their transition probabilities are significantly lower
than those of the d−d ones (at least for the transitions lying in
the energetic range considered in the computational study). In
A similar correlation (see Figure S8 in the Supporting
Information) also is obtained by evaluating the variation of the
oxidation potentials versus one (R) of the two descriptors of
the substituents effects in organic reactions, developed by
Swain and Lupton for factoring the Hammett’s σ constants into
2971
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
Table S2 in the Supporting Information (and in Figure 2), only
the most-intense transition of A1 (or A) symmetry is reported.
Conductor-like polarizable continuum model (CPCM) calcu-
lations confirm that these transitions are almost unaffected by
the solvent.
The lower-energy absorption bands, which have a slightly
lower intensity, can be assigned to metal−ligand-to-ligand
charge transfer (MLLCT) transitions, taking into account the
significant contribution of the bridging ancillary ligands to the
metal-centered HOMO set. These bands arise from the
convolution of multiple transitions, as testified by the more-
or less-pronounced shoulders observed at lower frequencies. In
particular, the three transitions (from HOMO−n to LUMO or
LUMO+1 orbitals) already described for the halogenated
derivatives7 can also be identified for the new species 2−6, and
are listed in Table S2 in the Supporting Information (and in
Figure 2). The computed wavelengths are in fair agreement
with the experimental absorption maxima measured in a
toluene solution. The mean absolute deviation between the
experimental λmax and the wavelength for the more-intense
totally symmetric excitation computed in the gas phase is 24
nm. The agreement strongly improves (the mean absolute
deviation reduces to 9 nm) by including the toluene solvent in
the calculation.
The charge-transfer character of such transitions is supported
by their strong solvent dependence:17 indeed, as already
reported for the related Re2(μ-diazine) complexes,6−8 a blue
shift is observed upon increasing solvent polarity (for instance,
from 363 nm in toluene to 335 nm in MeCN for 3; see Figure
5).
Upon excitation at 450 nm (for 2−4) or 480 nm (for 5 and
6), at room temperature in diluted deoxygenated toluene
solutions, all the complexes with bridging OR groups show
broad, featureless emission in the red region of the visible
spectrum (range of 608−708 nm). The photoluminescence
spectra for complexes 3−6 are shown in Figure 6, and the data
are summarized in Table 2.
Figure 6. Effect of the ancillary OR or SR ligands on the emission
spectra, recorded at room temperature in deoxygenated toluene
solution, upon excitation at 450 nm (for 2−4) or 480 nm (for 5 and
6). The emission intensity of the OC6F5 derivative 3 has been
decreased by a factor of 5 relative to the other complexes. Bands
indicated with * are artifacts due to the Raman scattering from the
toluene solvent.
of the bridging ancillary X ligands, provided that these do not
perturb the stiffness of the Re2(μ-X)2(μ-diazine) core (as it
occurs with the bulky Br derivative, whose knr value falls well
above the EGL line).
It is worth mentioning that complexes containing analogous
Re2(μ-ER)2(CO)6 fragments (E = O, S), joined by 4,4′-bipyridil
ligands in [Re2(μ-ER)2(CO)6(4,4′-bpy)2]2 molecular rectan-
gles, had been previously reported,18 and also in these cases
(very weak) emission in solution was observed only from some
of the OR derivatives (highest PLQY 0.039% in acetonitrile
solution).18d
It has been checked that the emission is independent of the
excitation wavelength and that all the complexes are photo-
stable. On the basis of previous results on related systems, the
emission can be confidently described as arising from triplet
metal-to-ligand charge transfer (3MLCT) states, as supported
by the red shift (for 3: λem = 608 nm in toluene, 620 nm in
dichloromethane, 650 nm in acetonitrile) and the strong
decrease of the emission intensity observed upon increasing the
solvent polarity (see Figure 5 for 3).
CONCLUSIONS
■
The synthesis of a series of neutral dinuclear rhenium
complexes of general formula [Re2(μ-ER)2(CO)6(μ-pydz)] (E
= O or S) has been performed. Several new synthetic routes
have been developed, all of which are more complex than those
used for the related dihalogen [Re2(μ-X)2(CO)6(μ-1,2-
diazine)] complexes, because of the unavailability of [Re-
(CO)5X] starting materials when X = ER.
The photophysical properties of the pentafluorophenolato
derivative 3 are closely comparable to those of the parent
dichloro-derivative 1 (see Table 2), in agreement with the close
similarity of their HOMO and LUMO levels, and then of their
reduction and oxidation potentials. The position of the
emission of the four complexes 3−6 is strongly affected by
the nature of the OR groups and red shifts upon increasing the
donor power of the ligand, following the trend of the
electrochemical HOMO−LUMO gaps. The photolumines-
cence quantum yields decrease in the same order (see Table
2 and Figure 6), according to the Energy Gap Law (EGL), and
the knr values roughly lie on the same line as the previously
reported dichloro derivatives containing pydz, monoalkylpyr-
idazines, and dialkylpyridazines (see Figure S10 in the
Supporting Information). This indicates that the nature of
the excited states and the pathways responsible for their
nonradiative deactivation are only slightly affected by the nature
The substitution of halides by OR groups does not modify
significantly the nature of the frontier orbitals and of the
absorption/emission processes, which maintain the previously
described metal−ligand-to-ligand charge transfer character. The
bridging OR groups are deeply involved in the HOMO set,
whose energy progressively rises on increasing the OR donor
power (except the deviant behavior of the OH derivative), with
consequent easier oxidation of the complex and reduced band
gap. Accordingly, a progressively red-shifted and weakened
emission is observed upon going from OC6F5 to OC6H5 to
OCH3 (and OH).
These results confirm that it is possible to modulate the
photophysical properties of the Re2-diazine complexes in a
twofold manner: the variation of the substituents on the
chromophoric moiety (the diazine ring) affects the LUMO
level, while the variation of the ancillary ligands modifies the
HOMO level.
2972
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
Table 2. Absorption and Emission Data in Toluene Solution of [Re2(μ-X)2(CO)6(μ-pydz)] Complexes
compound
X
λabs [nm]
ε [× 10−3 M−1 cm−1
]
λem [nm]
Φem [× 102]
τ [ns]
kr [× 10−5 s−1
]
knr [× 10−5 s−1
]
a
1
Cl
379
403
362
374
372
369
8.1
7.0
7.6
6.9
8.7
8.0
603
7.0
800
0.88
11.6
2
3
4
5
6
SC6H5
OC6F5
OC6H5
OCH3
OH
608
680
708
702
5.5
550
37
1.0
17.2
270
0.55
0.07
0.09
1.5
8.6
9.2
0.81
0.98
1160
1090
a
The data for the Cl derivative are taken from the Supporting Information in ref 19.
Isolated yield: 13 mg, 30%. IR ν(CO) (cm−1), CH2Cl2: 2035 w, 2018
s, 1936 m, 1912 m. H NMR (δ, CD2Cl2): 9.96 (pseudo t, 2H), 7.90
The behavior of the bis-pentafluorophenolato complex 3 is
1
comparable to that of the dichloro complex 1; therefore,
improvement in the photoluminescence quantum yields is
expected upon raising the LUMO levels, by replacing
pyridazine with the 4,5-dialkyl-pyridazine ligands, as previously
observed in the series of dichloro derivatives.8
(pseudo t, 2H), 7.18−7.12 (m, 10H). UV−vis (CH2Cl2) λ = 406 nm
(6.8 × 103 cm−1 M−1). Anal. Calcd: C, 31.50; H, 1.68; N, 3.34. Found:
C, 31.42; H, 1.74; N, 3.41. Red crystals suitable for X-ray analysis were
grown by slow evaporation from a CH2Cl2 solution at 300 K.
Synthesis of [Re2(μ-OC6F5)2(CO)6(μ-pydz)] (3). A sample of
[Re(CO)5OTf] (50 mg, 0.105 mmol) dissolved in freshly distilled
CH2Cl2 was treated at room temperature with 1.5 equiv of C6F5OH
(30 mg, 0.160 mmol) and 1.5 equiv of DMAN (34 mg, 0.160 mmol).
After refluxing for 5 h, IR monitoring showed two ν(CO) bands at
2023 cm−1 (s), 1913 cm−1 (vs), attributable to [DMANH]+[Re2(μ-
OC6F5)3(CO)6]−. The colorless solution was evaporated to dryness,
dissolved in freshly distilled toluene, treated with 0.5 equiv (with
respect to [Re(CO)5OTf]) of pyridazine (4.3 mg, 0.054 mmol), and
refluxed for 3 h. The color of the solution turned to dark yellow and IR
monitoring showed the disappearance of the bands of the intermediate
species, substituted by the bands of 3. 1H NMR analysis in CD2Cl2 of
the reaction mixture showed the presence mainly of 3, accompanied by
minor amounts of unidentified species. Separation by column
chromatography on SiO2 using dichloromethane/hexane 9:1 afforded
spectroscopically pure 3 (RF = 0.64). Isolated yields: 22 mg (47%). IR
ν(CO) (cm−1), CH2Cl2: 2047 w, 2033 s, 1944 s, 1916 s; 1H NMR (δ,
CD2Cl2): 9.85 (pseudo t, 2H), 8.27 (pseudo t, 2H); 19F NMR (δ,
CD2Cl2, 173 K): −156.7 (pseudo d, ortho), −161.8 (pseudo d, ortho),
−163.7 (pseudo t, meta), −164.3 (pseudo t, meta), −164.3 (pseudo t,
para). UV−vis (CH2Cl2) λ = 360 nm (7.5 × 103 cm−1 M−1). Anal.
Calcd: C, 26.78; H, 0.41; N, 2.84. Found: C, 26.92; H, 0.48; N, 2.89.
Yellow crystals suitable for X-ray analysis were grown by slow diffusion
of n-hexane into a CH2Cl2 solution, at 248 K.
Synthesis of [Re2(μ-OC6H5)2(CO)6(μ-pydz)] (4). Molten phenol
was used as the solvent, as follows: a sample of [Re(CO)5OTf] (60
mg, 0.126 mmol) was treated in a Schlenk tube with C6H5OH (6 g),
0.65 equiv of pyridazine (6.6 mg, 0.082 mmol), 0.9 equiv of Na2CO3
(12 mg, 0.113 mmol), and then heated at 55 °C, causing melting of
phenol. The orange mixture was maintained at this temperature for 20
h, and then unreacted C6H5OH was sublimed under vacuum, leaving a
yellow-orange residue. The solid was suspended in THF, then the
mixture was filtered and chromatographed on SiO2, with dichloro-
methane/ethyl acetate (9:1). The expected product 4 was eluted first
(RF = 0.73, isolated yield 11 mg, 20%.) IR ν(CO) (cm−1), CH2Cl2:
2036 w, 2021 s, 1927 s, 1903 s; 1H NMR (δ, CD2Cl2): 9.83 (pseudo t,
2H), 8.20 (pseudo t, 2H), 7.31−7.26 (m, 4H), 7.00−6.92 (m, 6H).
UV−vis (CH2Cl2) λ = 372 nm (6.9 × 103 cm−1 M−1). Anal. Calcd: C,
32.75; H, 1.75; N, 3.47. Found: C, 32.58; H, 1.59; N, 3.53. Yellow
crystals for X-ray analysis were obtained by slow diffusion of n-hexane
into a CH2Cl2 solution, at 248 K.
However, a perfluorophenolate unit is of limited interest for
bioconjugation. Therefore, the low emission properties of the
complexes bearing OR groups different from OC6F5, together
with the not-straightforward synthesis of this family of [Re2(μ-
OR)2(CO)6(μ-pydz)] complexes, indicate that this approach
might not be ideal for conjugating a luminescent Re2-diazine
group to biomolecules bearing ROH moieties. Different routes
will be investigated, involving inter alia mixed [Re2(μ-Cl)(μ-
OR)(CO)6(μ-diazine)] derivatives, which are expected to
exhibit higher PLQYs than the corresponding bis-(OR)
derivatives. Moreover, the strong enhancement of the
luminescence on going from organic to aqueous solution,
exhibited by some [Re2(μ-OR)2(CO)6(4,4′-bpy)2]2 rectangles
bearing long alkyl chains,18d suggests that analogous self-
aggregation phenomena might positively perturb the emissive
behavior of our systems when bonded to large biomolecules.
As far as the benzenethiolate derivative (2) is concerned,
despite its lack of luminescent properties, it appears to be of
interest, from a different point of view. Its low energy band gap
and its perfectly reversible oxidation and reduction processes
might find application for current generation by light
harvesting. We are currently investigating the modulation of
the electrochemical and light absorption properties of similar
[Re2(μ-SR)2(CO)6(μ-1,2-diazine)] complexes upon varying
the nature of the SR groups and of the diazine substituents.
EXPERIMENTAL SECTION
■
All of the reactions were performed under N2, using the Schlenk
technique and solvents deoxygenated and dried by standard methods.
Pyridazine (pydz) was used as received from Aldrich. [Re-
(CO)5(OTf)],10 NEt4[Re2(μ-OCH3)3(CO)6],11 and [Re4(μ3-
OH)4(CO)12]12 were prepared according to literature methods.
Elemental analyses were carried out in the Department of Inorganic,
Metallorganic and Analytical Chemistry of the University of Milan. EI-
1
HRMS were acquired on a VG Autospec M246 spectrometer. H and
19F NMR spectra were recorded on Bruker DRX300 or DRX400
spectrometers. IR spectra were acquired on Bruker Vector22 FT
instrument. Electronic absorption spectra were recorded on Agilent
8543 spectrophotometer, at room temperature.
Synthesis of [Re2(μ-OCH3)2(CO)6(μ-pydz)] (5). A sample of
NEt4[Re2(μ-OCH3)3(CO)6] (43 mg, 0.057 mmol) dissolved in 5 mL
of freshly distilled CH2Cl2, was treated with 1 equiv of pyridazine (4.4
mg, 0.057 mmol) and 1 equiv of triflic acid CF3SO3H (8.5 mg, 0.057
mmol). The mixture was stirred at room temperature for 6 h, than the
orange-red solution was evaporated to dryness. Separation by column
chromatography on SiO2 using dichloromethane/ethyl acetate (9:1)
afforded, besides the expected [Re2(μ-OCH3)2(CO)6(μ-pydz)] (RF =
0.71, compound 5), also two other species in which the methoxo
groups had been partially or completely replaced by OH groups:
[Re2(μ-OCH3)(μ-OH)(CO)6(μ-pydz)] (RF = 0.24) and [Re2(μ-
Synthesis of [Re2(μ-SC6H5)2(CO)6(μ-pydz)] (2). A sample of
[Re(CO)5(OTf)] (50 mg, 0.105 mmol) dissolved in freshly distilled
THF was treated with 0.5 equiv of pyridazine (4.2 mg, 0.053 mmol)
and then with 1 equiv of C6H5SH (12 mg, 0.105 mmol) and 1 equiv of
bis(dimethylamino)naphthalene (DMAN) (22.9 mg, 0.105 mmol).
The reaction mixture was refluxed for 2.5 h. A white precipitate was
separated and the supernatant solution was evaporated to dryness.
Separation by column chromatography on SiO2 using ethyl acetate/
dichloromethane 1:19 afforded spectroscopically pure 2 (RF = 0.90).
2973
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
OH)2(CO)6(μ-pydz)] (RF = 0.14, compound 6: it was isolated by
elution with neat ethyl acetate). [Re2(μ-OCH3)2(CO)6(μ-pydz)] (5).
Isolated yield: 16 mg (41%). IR ν(CO) (cm−1), CH2Cl2: 2027 w, 2010
concentration in each substrate, deareated by N2 bubbling, with 0.1 M
tetrabutylammonium hexafluorophosphate (TBAPF6, Fluka) as the
supporting electrolyte, at 298 K. The ohmic drop has been
compensated by the positive feedback technique.23 The experiments
were carried out using an AUTOLAB PGSTAT potentiostat
(EcoChemie, The Netherlands) run using a personal computer
(PC) with GPES software. The working electrode was a glassy carbon
one (AMEL, diameter = 1.5 mm) cleaned by diamond powder
(Aldrich, diameter = 1 μm) on a wet cloth (STRUERS DP-NAP); the
counterelectrode was a platinum wire; the reference electrode was an
aqueous saturated calomel electrode, having, in our working medium, a
difference of −0.385 V vs the Fc+|Fc couple (the intersolvental redox
potential reference currently recommended by IUPAC)24 and +0.032
V vs the Me10Fc+|Me10Fc couple (an improved intersolvental reference
under investigation).25 The normalized convoluted peak currents have
been computed by the relationship
1
s, 1915 s, 1892 s; H NMR (δ, CD2Cl2): 9.69 (pseudo t, 2H), 8.06
(pseudo t, 2H), 4.22 (s, 6H). UV−vis (CH2Cl2) λ = 370 nm (8.6 ×
103 cm−1 M−1). Anal. Calcd: C, 21.11; H, 1.48; N, 4.10. Found: C,
20.94; H, 1.37; N, 4.10.
Synthesis of [Re2(μ-OH)2(CO)6(μ-pydz)] (6). A sample of
[Re4(μ3-OH)4(CO)12] (150 mg, 0.131 mmol) dissolved in 3 mL of
water was treated with 2 equiv of pyridazine (21.0 mg, 0.262 mmol)
and heated at 100 °C for 4 h in a microwave oven, affording an orange
precipitate and a yellow solution. The entire suspension was
chromatographed on a SiO2 column, with ethyl acetate/dichloro-
methane (1:1), affording spectroscopically pure 6. Isolated yield: 111
mg, 65%. IR ν(CO) (cm−1), CH2Cl2: 2029 w, 2013 s, 1917 s, 1892 s;
1H NMR (δ, CD2Cl2): 9.75 (pseudo t, 2H), 8.07 (pseudo t, 2H), 1.14
(s, 2H). UV−vis (CH2Cl2) λ = 368 nm (8.0 × 103 cm−1 M−1). EI-
HRMS: m/z 655.923460, calcd for C10H6N2O8Re2: 655.923945. Anal.
Calcd: C, 18.35; H, 0.92; N, 4.28. Found: C, 18.56; H, 1.04; N, 4.21.
Single-Crystal X-ray Diffraction Analysis. Crystal Data for
2·CH2Cl2. C23H16Cl2N2O6Re2S2, Mr = 923.80, orthorhombic,
space group Pbca (No. 61), a = 18.041(2) Å, b = 14.442(2) Å, c
= 20.886(2) Å, V = 5441.8(11) Å3, Z = 8, dcalc = 2.255 g cm−3, T
= 100(2) K, crystal size = 0.35 mm × 0.27 mm × 0.23 mm, μ =
9.281, λ = 0.71073 Å. Refinement of 334 parameters on 7883
independent reflections out of 68 104 measured reflections (Rint
= 0.0346, Rσ = 0.0191, 2θmax = 60.0°) led to R1 = 0.0231 (I >
2σ(I)), wR2 = 0.0545 (all data), and S = 1.087, with the largest
peak and hole of 0.940 and −0.610 e Å−3.
i
c
I
L
Ac
L
1/2
= nFD
=
where IL is the limiting current, iL the limiting current density, A the
electrode surface, c the substrate concentration in the bulk, n the
number of transferred electrons, F the Faraday constant (F = 96485 C
mol−1 e−1), and D the diffusion coefficient. The pseudo-first-order k′
constants for the chemical reaction of the P product of the first
monoelectronic oxidation step have been estimated from the following
equations:
⎛
⎝
⎞
⎠
⎛
ln⎜
⎝
⎞
I
[P]
p,b
t
⎜
⎜
⎟
⎟
⎟ = −k′t = ln
Crystal Data for 3. C22H4F10N2O8Re2, Mr = 986.67, orthorhombic,
space group Fdd2 (No. 43), a = 19.395(2) Å, b = 25.618(2) Å, c =
10.262(2) Å, V = 5098.8(12) Å3, Z = 8, dcalc = 2.571 g cm−3, T =
150(2) K, crystal size = 0.14 mm × 0.11 mm × 0.11 mm, μ = 9.613, λ
= 0.71073 Å. Refinement of 199 parameters on 3694 independent
reflections out of 27 975 measured reflections (Rint = 0.0606, Rσ =
0.0374, 2θmax = 60.0°) led to R1 = 0.0243 (I > 2σ(I)), wR2 = 0.0507
(all data), and S = 1.017, with the largest peak and hole of 0.894 and
−0.635 e Å−3.
[P]
I
p,f
⎠
0
where
(E − E ) + (E − E )
p,f p,b
λ
λ
t =
v
Here, Ep,f is the peak potential of the forward peak, Ip,f the current of
the forward peak, Ep,b the peak potential of the backward peak, Ip,b the
current of the backward peak, Eλ the inversion potential in the positive
potential scan, and v the potential scan rate.
Crystal Data for 4·1/2CH2Cl2. C22.5H15ClN2O8Re2, Mr = 849.21,
monoclinic, space group P21/c (No. 14), a = 8.860(2) Å, b =
12.518(2) Å, c = 23.094(2) Å, β = 96.10(2)°, V = 2546.8(7) Å3, Z = 4,
dcalc = 2.215 g cm−3, T = 295(2) K, crystal size = 0.19 mm × 0.07 mm
× 0.05 mm, μ = 9.650, λ = 0.71073 Å. Refinement of 325 parameters
on 7447 independent reflections out of 57 778 measured reflections
(Rint = 0.0540, Rσ = 0.0303, 2θmax = 60.0°) led to R1 = 0.0313 (I >
2σ(I)), wR2 = 0.0635 (all data), and S = 1.019, with the largest peak
and hole of 0.985 and −0.731 e Å−3.
Photophysical Measurements. Steady-state fluorescence meas-
urements and quantum yield determinations have been performed on
a Cary eclipse fluorescence spectrometer (Varian, Inc., Agilent
Technologies, Santa Clara, CA). The solutions have been prepared
under N2 directly in the quartz cuvettes, using the Schlenk technique,
and were further deoxygenated by bubbling N2 for 20 min just before
measurements. The emission intensities have been normalized to an
absorption value of 0.1. Quantum yields have been determined by
comparison with the emission of [Ru(bpy)3]Cl2 in water (Φ = 0.04).26
Dynamic fluorescence measurements have been performed with a
frequency modulated phase fluorometer (Digital K2, I.S.S., Urbana,
IL). The excitation was accomplished by the 9 mW output of a 378-
nm diode laser (I.S.S., Urbana, IL). At least 15 data points at
logarithmically spaced frequencies in the range of 0.3−30 MHz with a
cross-correlation frequency of 400 Hz (A2D, ISS, Inc., USA) have
been acquired for lifetime measurements. The convenient accuracies
for phase angles and modulation ratios have been 0.2° and 0.004,
respectively. Lifetime measurements have been performed under the
magic-angle conditions27 and a 535-nm long-pass filter (Andover Co.)
has been employed in order to cut light scattering. A solution of
glycogen in doubly distilled water has been used as a reference
sample.28 Lifetime data fitting has been accomplished by an ISS
routine, based on the Marquardt least-squares minimization, with a
two-exponential decay scheme, in order to take into account the
scattering contribution to the overall signal. The fit of the fluorescence
intensity decay (F(t)) yields the lifetime values τi, together with the
corresponding fractional intensities (fi):
CCDC-830740−830742 contains the crystallographic data for 2−4.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre.
Computational Details. Ground-state geometries were optimized
by means of density functional calculations. The parameter-free hybrid
functional PBE020 was employed along with the standard valence
double-ζ polarized basis set 6-31G(d,p) for C, H, Cl, N, and O. For Re
and S, the Stuttgart−Dresden effective core potentials were employed,
along with the corresponding valence triple-ζ basis set.
All the calculations were done assuming C2v (for 3−6) or C2 (for 2)
symmetry. The nature of all the stationary points was checked by
computing vibrational frequencies and all the species were found to be
true minima.
In order to simulate the absorption electronic spectrum down to
230 nm, the lowest 50 singlet excitation energies were computed by
means of time-dependent density functional calculations. Calculations
were done also in the presence of solvent (toluene, used in most of the
photophysical characterizations), as described by the conductor-like
polarizable continuum model (CPCM).21
All the calculations were done with Gaussian 03.22
Electrochemical Measurements. The cyclovoltammetric analysis
has been performed at scan rates typically ranging from 0.02 V s−1 to
10 V s−1, in HPLC-grade acetonitrile solutions at 0.00025−0.001 M
−t/τ
i
F(t) = ∑ α e
i
2974
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975

Inorganic Chemistry
Article
and
(12) Herberhold, M.; Suss, G.; Ellermann, J.; Gabelein, H. Chem. Ber
̈
̈
1978, 111, 2931−2941.
α τ
i
i
f =
(13) Geometric parameters found in 2−4 are unexceptional and
similar to those reported for analogous species containing a Re−(μ-
OR)−Re or a Re−(μ-SR)−Re fragment (Cambridge Structural
Database 5.32). Bond distances and angles reported in the text are
average values; the largest standard uncertainty on bond distances and
angles are 0.012 Å and 0.5°, respectively.
(14) The C2v geometry of 2 possesses two imaginary frequencies (34i
and 31i). This conformation resides 4.6 kJ mol−1 higher than the C2
minimum. In addition, starting the optimization from the solid-state
geometry an asymmetric minimum was found, whose energy is only
negligibly different from those of the C2 minimum. Since MO energies
and excitations are almost identical for the C2 and the C1 conformers,
only the values computed for the symmetric species will be reported
and discussed.
(15) (a) Wendt, H.; Bitterlich, S. Electrochim. Acta 1992, 37, 1951−
1958. (b) Said, A. H.; Mhalla, F. M.; Amatore, C.; Verpeaux, J. N. J.
Electroanal. Chem. 1999, 464, 85−92. (c) Said, A. H.; Mhalla, F. M.;
Amatore, C.; Thouin, L.; Pebay, C.; Verpeaux, J. N. J. Electroan. Chem.
2002, 537, 39−46.
(16) (a) Swain, C. G.; Lupton, E. C. Jr. J. Am. Chem. Soc. 1968, 90,
4328−4337. (b) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91,
165−195.
(17) Giordano, P. J.; Wrighton, M. S. J. Am. Chem. Soc. 1979, 101,
2888−2897.
(18) (a) Benkstein, K. D.; Hupp, J. T.; Stern, C. L. Inorg. Chem. 1998,
37, 5404−5405. (b) Woessner, K. D.; Helms, J. B.; Shen, Y.; Sullivan,
B. P. Inorg. Chem. 1998, 37, 5406−5407. (c) Marimaran, B.;
Rajendran, T.; Lu, Y. L.; Lee, G. H.; Peng, S. M.; Lu, K. L. J. Chem.
Soc., Dalton Trans. 2001, 515−517. (d) Marimaran, B.; Thanasekaran,
P.; Rajendran, T.; Lin, R. J.; Chang, I. J.; Lee, G. H.; Peng, S. M.;
Rajagopal, S.; Lu, K. L. Inorg. Chem. 2002, 42, 5323−5325.
(19) Quartapelle Procopio, E.; Mauro, M.; Panigati, M.; Donghi, D.;
Mercandelli, P.; Sironi, A.; D’Alfonso, G.; De Cola, L. J. Am. Chem. Soc.
2010, 132, 14397−14399.
i
∑ α τ
i
i
where αi represents the pre-exponential factors.
ASSOCIATED CONTENT
■
S
* Supporting Information
Three tables with details concerning the computations and the
electrochemical characterization; 10 figures showing details
about the NMR spectra, the crystal structures, the isodensity
surfaces of some relevant molecular orbitals, the electro-
chemical parameters, the absorption spectra in dichloro-
methane, and the EGL correlation; crystallographic data in
CIF format, and optimized coordinates in Tripos Mol2 format.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: monica.panigati@unimi.it.
ACKNOWLEDGMENTS
■
G.D., D.M., and P.M. thank the Italian MIUR for financial
support (PRIN 2009 PRAM8L Innovative materials for organic
and hybrid photovoltaics).
REFERENCES
■
(1) (a) Fernandez-Moreira, V.; Thorp-Greenwood, F. L.; Coogan, M.
P. Chem. Commun. 2010, 46, 186−202. (b) Lo, K. K.-W. Top.
Organomet. Chem. 2010, 29, 115−158. (c) Lo, K. K.-W.; Hui, W. K.;
Chung, C. K.; Tsang, K. H.-K.; Ng, D. C.-M.; Cheung, K.-K. Coord.
Chem. Rev. 2005, 249, 1434−1450.
(20) Called PBE1PBE in Gaussian. (a) Adamo, C.; Barone, V. J.
Chem. Phys. 1999, 111, 6158−6170. (b) Perdew, J. P.; Burke, K.;
Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865−3868. (c) Perdew, J. P.;
Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78, 1396.
(21) (a) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995−
2001. (b) Cossi, M.; Barone, V. J. Chem. Phys. 2001, 115, 4708−4717.
(c) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669−681.
(2) Yersin, H.; Rausch, A. F.; Czerwieniec, R.; Hofbeck, T.; Fisher, T.
Coord. Chem. Rev. 2011, 255, 2622−2652.
(3) See for instance (a) Kumar, A.; Sun, S.-S.; Lees, A. J. Top.
Organomet. Chem. 2010, 29, 1−35. (b) Yam, V. W.-W.; Wong, K. M.-
C. Chem. Commun. 2011, 47, 11579−11592.
(4) See for instance (a) Villegas, J. M.; Stoyanov, S. R.; Huang, W.;
Rillema, D. P. Inorg. Chem. 2005, 44, 2297−2309. (b) Tseng, Y.-H.;
Bhattacharya, D.; Lin, S.-H.; Thanasekaran, P.; Wu, J.-Y.; Lee, L.-W.;
Sathiyendiran, M.; Ho, M.-L.; Chung, M.-W.; Hsu, K.-C.; Chou, P.-T.;
Lu, K.-L. Inorg. Chem. 2010, 49, 6805−6807.
(22) Gaussian 03 (revision C.02), Gaussian, Inc., Wallingford, CT,
2004.
(23) Bard, A. J.; Faulkner, L. R. Electrochemical Methods.
Fundamentals and Applications; Wiley: New York, 2002, pp 648−650.
(24) (a) Gritzner, G.; Kuta, J. Pure Appl. Chem. 1984, 56, 461−466.
(b) Gritzner, G. Pure Appl. Chem. 1990, 62, 1839−1858.
(25) (a) Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.;
Lay, P. A.; Masters, A. F.; Phillips, L. J. Phys. Chem. B 1999, 103,
(5) (a) Nahhas, A. E.; Cannizzo, A.; van Mourik, F.; Blanco-
Rodriguez, A. M.; Zal
114, 6361−6369. (b) Bakova,
Zalis, S. Coord. Chem. Rev. 2011, 255, 975−989. (c) Gindensperger, E.;
́
is,
̌
S.; Vlce
̌
k, A.; Chergui, M. J. Phys. Chem. 2010,
́
R.; Chergui, M.; Daniel, C.; Vlce
̌
k, A.;
́
̌
Koppel, H.; Daniel, C. Chem. Commun. 2010, 46, 8225−8227.
̈
(d) Kayanuma, M.; Daniel, C.; Koppel, H.; Gindensperger, E. Coord.
̈
́
6713−6722. (b) Ruiz, J.; Astruc, D. C. R. Acad. Sci., Ser. IIc: Chem.
Chem. Rev. 2011, 255, 2693−2703.
1998, I, 21−27. (c) Falciola, L.; Gennaro, A.; Isse, A. A.; Mussini, P.
R.; Rossi, M. J. Electroanal. Chem. 2006, 593, 47−56.
(26) Ishida, H.; Tobita, S.; Hasegawa, Y.; Katoh, R.; Nozaki, K.
Coord. Chem. Rev. 2010, 254, 2449−2458.
(6) Panigati, M.; Donghi, D.; D’Alfonso, G.; Mercandelli, P.; Sironi,
A.; Mussini, P.; D’Alfonso, L. Inorg. Chem. 2006, 45, 10909−10921.
(7) Donghi, D.; D’Alfonso, G.; Mauro, M.; Panigati, M.; Mercandelli,
P.; Sironi, A.; Mussini, P.; D’Alfonso, L. Inorg. Chem. 2008, 47, 4243−
4255.
(27) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: New York, 1999.
(28) Collini, M.; Chirico, G.; Baldini, G.; Bianchi, M. E. Biopolymers
1995, 36, 211−225.
(8) Mauro, M.; Quartapelle Procopio, E.; Sun, Y.; Chien, C. H.;
Donghi, D.; Panigati, M.; Mercandelli, P.; Mussini, P.; D’Alfonso, G.;
De Cola, L. Adv. Funct. Mater. 2009, 19, 2607−2614.
(9) See Lucenti, E.; D’Alfonso, G.; Macchi, P.; Maranesi, M.;
Roberto, D.; Sironi, A.; Ugo, R. J. Am. Chem. Soc. 2006, 128, 12054−
12055 and references therein..
(10) Schmidt, S. P.; Nitschke, J.; Trogler, W. C. Inorg. Synth. 1989,
26, 113−117.
(11) (a) Ciani, G.; D’Alfonso, G.; Freni, M.; Romiti, P.; Sironi, A. J.
Organomet. Chem. 1978, 152, 85−94. (b) Jiang, C.; Wen, Y.-S.; Liu, L.-
K.; Hor, T. S. A.; Yan, Y. K. Organometallics 1998, 17, 173−181.
2975
dx.doi.org/10.1021/ic202284a | Inorg. Chem. 2012, 51, 2966−2975