A triphenylamine/bis(terpyridine)IrIII dyad for the assembly of charge-separation constructs with improved performances

Article abstract of DOI:10.1002/ejic.200700669

A new dyad DII-Ir consisting of a triphenylamine donor and a bis(terpyridine)Ir^III acceptor separated by a bridge composed of two benzamide groups has been synthesized. The electrochemical and photophysical properties of the dyad have been compared to those of a corresponding dyad D-Ir characterized by a bridge connecting the donor and the acceptor consisting of a single benzamide unit. We show that, whereas the charge-separation steps are still very efficient in the long dyad (≥99%), charge recombination is slowed by a factor of three with respect to the short dyad. This encourages the use of this dyad in the assembly of a more elaborate array DII-Ir-A, expected to overcome the disappointingly low charge-separation yield of a previously studied D-Ir-A short triad. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejic.200700669

FULL PAPER
DOI: 10.1002/ejic.200700669
A Triphenylamine/Bis(terpyridine)IrIII Dyad for the Assembly of Charge-
Separation Constructs with Improved Performances
Lucia Flamigni,*[a] Barbara Ventura,[a] Etienne Baranoff,[b] Jean-Paul Collin,*[b] and
Jean-Pierre Sauvage*[b]
Keywords: Charge separation / Electron transfer / Iridium / Photochemistry / Supramolecular chemistry
A new dyad DII-Ir consisting of a triphenylamine donor and
the long dyad (Ն99%), charge recombination is slowed by a
a bis(terpyridine)IrIII acceptor separated by a bridge com- factor of three with respect to the short dyad. This encour-
posed of two benzamide groups has been synthesized. The
electrochemical and photophysical properties of the dyad
have been compared to those of a corresponding dyad D-
Ir characterized by a bridge connecting the donor and the
acceptor consisting of a single benzamide unit. We show that,
whereas the charge-separation steps are still very efficient in
ages the use of this dyad in the assembly of a more elaborate
array DII-Ir-A, expected to overcome the disappointingly low
charge-separation yield of a previously studied D-Ir-A short
triad.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
high excited-state energy level prevented draining of the
electronic energy of the other components due to energy
transfer, and the relative ease of reduction of the complex
unit provided a high driving force for the primary photo-
induced electron transfer starting the sequence. By using
arrays with a central bis(terpyridine)IrIII complex and ap-
pended porphyrins with suitable metallation to provide a
good electron donor [free-base or (porphyrin)zinc complex]
or a good electron acceptor [(porphyrin)gold(III) complex],
respectively, we succeeded in producing a CS state with a
lifetime of 450 ns and 100% yield.[2d,6]
Introduction
The process of producing long-lived charge-separated
(CS) states is the basic event in artificial photosynthesis,
which aims at efficiently converting light into chemical en-
ergy. In spite of the impressive results achieved in the last
decade in multi-component arrays,[1] genuine long-lived CS
states are still a valuable goal, especially when they involve
less common or unexplored chromophores with respect to
more conventional tetrapyrrolic or fullerene derivatives.[2]
The most convenient and better tested strategy for produc-
ing genuine and indisputable CS states is that of using
weakly coupled multi-component structures and of moving
the electron, through a sequence of electron-transfer events,
to molecular components far away from the hole. The first
electron-transfer step is driven by light, whereas the follow-
ing ones are ground-state electron-transfer reactions. A few
years ago we started to use (terpyridine-type)IrIII complexes
as functional assembling units of multi-component struc-
tures having either the role of electron relay[2d,3] or of pho-
tosensitizer.[4] The use of this type of complexes proved to
be very valuable both for the high energy level (ca. 2.5 eV)
of its excited state, a ligand-centered triplet (3LC) occasion-
ally with some charge-transfer triplet (3CT) character, and
for the relative ease of reduction, –0.75 V vs. SCE.[5] The
More recently, we studied a D-Ir-A triad, where D is a
triphenylamine electron donor, A is a naphthalenebis-
(imide) electron acceptor, and Ir is a bis(terpyridine)IrIII
complex (Scheme 1).
The lifetime of the CS state with opposite charges local-
ized at the extremities of the array, produced upon light
absorption by either the D or Ir unit is remarkable, of the
order of 120 µs at ambient temperature in air-free acetoni-
trile (100 µs in air-saturated samples), but the yield of
charge separation is quite modest, of the order of 10%.[7]
The reason for this inefficiency is that the intermediate CS
state D+-Ir–-A, formed with a 100% yield by a primary elec-
tron-transfer step originating from either an excited D unit
(1D) or excited Ir unit (3Ir), can deactivate to the ground
state faster (k = 1.4ϫ1010 s–1) than the rate at which it un-
dergoes a further electron-transfer step (k = 2.4ϫ109 s–1)
leading to the final CS state D+-Ir-A–. We reasoned that
increasing the D to Ir separation from ca. 1.6 nm in the
original system to a distance of ca. 2.3 nm by inserting a
further benzamide linker as in DII-Ir (Scheme 2), could be
a possible strategy to improve the final CS yield in a triad
DII-Ir-A.
[a] Istituto ISOF-CNR,
Via P. Gobetti 101, 40129 Bologna, Italy
E-mail: flamigni@isof.cnr.it
[b] Laboratoire de Chimie Organo-Minérale, UMR 7177 CNRS,
Université Louis Pasteur,
Institut Le Bel, 4, rue Blaise Pascal, 67070 Strasbourg, France
E-mail: jpcollin@chimie.u-strasbg.fr
sauvage@chimie.u-strasbg.fr
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2007, 5189–5198
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5189

L. Flamigni, B. Ventura, E. Baranoff, J.-P. Collin, J.-P. Sauvage
FULL PAPER
Scheme 1. Structures of the previously reported triad D-Ir-A and related models.
more complex array. Therefore, the present report deals
with the synthesis and photophysical properties of the dyad
DII-Ir in order to evaluate its suitability for the construc-
tion of a triad DII-Ir-A with improved properties with re-
spect to D-Ir-A.
Results and Discussion
The systems studied in this work are the dyad DII-Ir and
the pertinent models DII and Ir1, reported in Scheme 2.
Whereas complex Ir1 had been previously examined and its
properties will be only briefly summarized, DII and DII-Ir
will here be examined in detail, and their properties will be
compared with those of the short counterparts D and D-Ir
previously reported.[7,8] In the following discussion the
terms Ir and DII (not bold) will be used to indicate the Ir
metal complex (with model Ir1) and the donor units (with
model DII), respectively, within the dyad.
Scheme 2. Structures of the dyad DII-Ir and models reported in
this work.
Decrease of both charge-separation (1DII-Ir or DII-3Ir
Ǟ DII+-Ir–) and charge-recombination (DII+-Ir– Ǟ D-Ir)
rates are expected after increasing the D–Ir distance but,
whereas a slight decrease in the 100% efficiency of the
charge separation step can be afforded without being too
detrimental to the overall CS process, a decrease in the rate
Synthesis
The dyad DII-Ir has been prepared according to the reac-
of charge recombination in the dyad DII-Ir can have impor- tions shown in Scheme 3. The terpyridine 2 was prepared
tant effects on the overall charge separation over the ex- in two steps by homologation of the already known 4Ј-(4-
treme components in the derived triad DII-Ir-A. Since the aminophenyl)-2,2Ј:6Ј,2ЈЈ-terpyridine[9] with 4-nitrobenzoyl
synthesis of the latter requires a significant synthetic effort, chloride and then by reduction of the nitro group with hy-
we decided to verify first if the properties of the component drazine in the presence of a catalytic amount of palladium
dyad could suggest better performances for the derived on charcoal. The terpyridine 3, bearing a triarylamine do-
triad, before undertaking the demanding synthesis of the nor group linked to the terpy fragment by two amide brid-
5190
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5189–5198

A Triphenylamine/Bis(terpyridine)IrIII Dyad
Scheme 3. Synthesis of the dyad DII-Ir.
ges, was prepared by condensation of the terpyridine 2 with Absorption Spectroscopy
4-[bis(4-methoxyphenyl)amino]benzoic acid, in the presence
The absorption spectra in acetonitrile of the components,
of DMAP and EDC (peptidic coupling).[10] The dyad
their superposition, and the experimental spectrum of the
array are reported in Figure 1. There is, within experimental
errors, a good overlap of the experimental spectrum of DII-
Ir with the calculated superposition of the model compo-
nents in the 250–350 nm range, whereas the array spectrum
differs slightly from the calculated one above 350 nm. In
fact, it displays a slightly higher absorbance and a broaden-
ing on the low energy side. Quite remarkably, the broaden-
ing toward the low-energy range of the spectrum is less pro-
nounced with respect to the short dyad D-Ir, characterized
by an absorption extending over 500 nm, see inset of Fig-
ure 1. In order to understand this behavior we have to recall
that, whereas in (tpy)2Ir3+-type complexes the low-energy
absorption band is in general a predominantly LC transi-
tion, complexes of the (tpy)2Ir3+-type with benzamide sub-
DII-Ir was obtained in 70% yield by reaction of the pre-
cursor [4Ј-(3,5-di-tert-butylphenyl)-2,2Ј:6Ј,2ЈЈ-terpyridine]-
[5a]
IrCl3
with the terpyridine 3 in ethylene glycol at 160 °C
during 20 min.
Electrochemistry
The redox characteristics of the reference compounds Ir1
and DII as well as that of the dyad DII-Ir examined by
cyclic voltammetry in CH3CN are reported with the values
previously obtained for D and D-Ir in Table 1. One-electron
reversible waves were observed for all the redox processes,
in the dyad they were easily assigned to the corresponding
individual components. A few mV of difference with respect
to the short dyad D-Ir are observed in the redox couples of
the donor group (D+/D) and of the complex (terpy/terpy–).
Clearly, the presence of a second amide bridge in the dyad
DII-Ir weakens even more the electronic coupling between
the two redox entities.
Table 1. Cyclic voltammetry data of the reference compounds Ir1,
DII, and the dyad DII-Ir in acetonitrile, 0.1  nBu4NPF6, with SCE
as a reference. Values for dyad D-Ir and model D are also re-
ported.[7]
E1/2 (V vs. SCE)
D+/D
terpy/terpy–
Ir1
–0.76
D
0.80
0.75
0.76
0.76
D-Ir
DII
DII-Ir
–0.75
–0.78
Figure 1. Absorption spectra of the model components and of the
arrays. Ir1 (thin line), DII (dotted line), DII-Ir (thick line), Ir1 +
DII (dashed line). Inset: DII-Ir (thick line), D-Ir (dash-doted line).
Eur. J. Inorg. Chem. 2007, 5189–5198
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5191

L. Flamigni, B. Ventura, E. Baranoff, J.-P. Collin, J.-P. Sauvage
FULL PAPER
stituents were recently demonstrated to exhibit photophysi- residual luminescence in DII-Ir could be due to some very
cal properties at room temperature typical of a CT state, minor, unbound component in the dyad and not to a genu-
either of metal-to-ligand charge-transfer nature (MLCT) or ine quenched unit in the array; the definitive answer on the
of intraligand charge-transfer nature (ILCT).[5,11,12] The effective quenching of the component in the dyad can only
presence of an amine electron donor as a substituent can be provided by the time-resolved experiments, discussed be-
strongly enhance the ILCT nature of the transition, with low.
the terpyridine acting as an electron acceptor and the amine
as an electron donor, and further shift the absorbance to
the red. From this point of view, interposition of a further
amide spacer between the terpyridine acceptor and the aro-
matic amine donor has the effect of increasing the energy
of the band, moving it further to the blue, as noticed in the
inset of Figure 1 for DII-Ir compared to D-Ir.
If we ignore the small perturbation of the spectrum of
the Ir unit discussed above, there is no evidence from the
spectroscopic data of electronic interactions in the array
with respect to the models, in agreement with the electro-
chemical data. This allows us to use a localized description
of the individual subunits in the subsequent discussion of
Figure 2. Room-temperature luminescence spectra in acetonitrile.
the photoinduced energy and electron-transfer reactions.
Left panel: excitation at 330 nm, 50% excitation of DII and Ir com-
ponents, DII (solid line) A = 0.056, DII-Ir (dashed line) A = 0.12.
Right panel: selective excitation at 430 nm of Ir component, Ir1
(solid line), A = 0.13, DII-Ir (dashed line) A = 0.13. The ab-
sorbance of the solutions is arranged to provide the absorption of
the same number of photons in the model and in the corresponding
unit in the array.
Luminescence Spectroscopy
DII emits strongly, with a band maximum at 524 nm,
slightly blueshifted with respect to the short analog D (λmax
= 540 nm); Ir1 emits moderately with a maximum at
570 nm (Table 2).
Luminescence spectra of the models Ir1 and DII at 77 K
in a glassy butyronitrile matrix display emission maxima at
414 and at 510 nm, respectively, with a remarkable hypso-
chromic shift from fluid (295 K) to rigid (77 K) solvent (see
Table 2). This can be explained by a strong CT character
for the 1DII emitting excited state which is less stabilized in
a rigid medium; in the case of Ir1, a change in the spectral
profile is also detected (compare Figures 2 and 3), indicative
of a switching of the excited-state nature from a room-tem-
perature CT to a predominantly ligand-centered state in the
glass, as previously discussed.[11]
Table 2. Luminescence properties in air-equilibrated nitrile sol-
vents. Acetonitrile at 295 K and butyronitrile at 77 K.
State
295 K
τ
77 K
τ
λmax
Φem
λmax
[nm]
E
[nm]
[ns]
[ns] [eV]
2.1ϫ10–3 510 80000 2.43
Ir1[a,b] 3Ir1
570
540 0.450
490
D[b]
1D
0.012
422
2.6
–
–
2.93
2.40
D-Ir[b] D-3Ir
1D-Ir
–
Յ6ϫ10–5 515[d]
ϽϽ0.01 Յ6ϫ10–5
–
414
516
DII[c] 1DII
524
1.0
–
0.01
0.073
2.0
2.99
2.40
3.10
DII-Ir DII-3Ir[a] 570
6ϫ10–5
1DII-Ir[c]
5ϫ10–4 400[d]
[a] Excitation at 430 nm for steady state, for time-resolved determi-
nations excitation at 373 nm. [b] From ref.[7] [c] Excitation at
330 nm for steady state, for time-resolved determinations excitation
at 373 nm, 331 nm, or 355 nm. [d] Weak signal.
The luminescence spectra of the array DII-Ir at room
temperature upon selective excitation of the Ir unit at
430 nm is compared to those of the model Ir1 absorbing
the same number of photons as the unit in the array (Fig-
ure 2, right panel). In the left panel of Figure 2 the lumines-
cence detected in the dyad upon excitation at 330 nm, corre-
sponding to excitation of both DII and Ir complex units
(absorbing 50% each), is reported and compared to the lu-
minescence from a solution of DII absorbing the same
number of photons as the unit in the array. The emission
of Ir and DII units in the dyad are nearly completely
quenched, to ca. 3% and less than 1%, respectively, indicat-
ing the occurrence of an efficient quenching process involv-
Figure 3. Luminescence spectra in butyronitrile at 77 K. Left panel:
excitation at 330 nm, 50% excitation of DII and Ir components,
DII (solid line) A = 0.1, Ir1 (dotted line) A = 0.1, DII-Ir (dashed
line) A = 0.2. Right panel: selective excitation at 430 nm of Ir com-
ponent, Ir1 (solid line) A = 0.08, DII-Ir (dashed line) A = 0.08.
The absorbance of the solutions is arranged to provide the absorp-
tion of the same number of photons in the model and in the corre-
3
1
ing both Ir and DII. It should be noticed that the weak sponding unit in the array.
5192
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5189–5198

A Triphenylamine/Bis(terpyridine)IrIII Dyad
Luminescence experiments on the models and DII-Ir in
Time-resolved experiments at 77 K in butyronitrile glass
butyronitrile glass at 77 K show that nearly no quenching allowed the following parameters to be measured. The life-
of the Ir unit occurs at 77 K, whereas the DII unit is still time of the models D and DII are 2.6 and 2 ns, respectively,
strongly quenched also at 77 K, as evidenced in Figure 3 showing a slight temperature dependence. On the contrary,
which compares the luminescence from DII and Ir1 solu- the lifetime of Ir1 at 77 K is 80 µs, much longer than at
tions to that from DII-Ir after selective excitation of the Ir room-temperature (Table 2). This has to be assigned to a
complex unit at 430 nm in the right panel, or of the DII switch in the nature of the excited state when the medium
3
and Ir components (absorbing 50% each) at 330 nm in the becomes rigid from CT, very likely of intra-ligand charge-
left panel. With respect to the results previously reported transfer nature, to a 3LC, as already discussed above. In the
for the short corresponding dyad D-Ir both at room tem- dyad DII-Ir, the lifetime of Ir measured with a time-corre-
perature and at 77 K, in DII-Ir the luminescence quenching lated single photon apparatus is 56 µs compared to the
is less effective. This is in agreement with a slower electron 80 µs of the model, confirming a very modest quenching in
transfer over the longer bridge connecting the donor to the agreement with steady-state experiments. The DII unit de-
Ir complex.
cay in the dyad DII-Ir at 77 K was still extremely fast and
The luminescence lifetimes of the model Ir1 in air-equil- of the same order as the exciting profile; an accurate deter-
ibrated acetonitrile solution determined by a single photon mination of the lifetime was precluded by the strong light
nanosecond apparatus, is 490 ns.[7] Any attempt to deter- scattering in the glass. From the short counterpart D-Ir no
mine the lifetime of the Ir unit in the dyad was unsuccessful luminescence could be detected similarly to what occurs at
at room temperature, neither a single-photon counting tech- room temperature.
nique with 0.3 ns resolution, nor a streak-camera detection
with 10 ps resolution allowed any emission to be registered.
Given the very low radiative rate constant of the Ir com-
Transient Absorption Spectroscopy
plex, kr ≈ 5ϫ103 s–1 (kr = φ/τ), a technique like the single-
shot streak camera is not convenient for detecting such low
Valuable information on the products of electron transfer
3
emissions, therefore we cannot exclude for the Ir compo- can be gained by transient absorption techniques. The
nent in DII-Ir a lifetime shorter than 0.3 ns. Only the time- chemically generated cation DII+ (see Exp. Sect. for details)
resolved absorption data discussed below will clear this has a peak at 760 nm (ε ≈ 27400) and a minor broad band
point. The strong emission of DII was time-resolved by a at 590 nm, in agreement with previous reports.[13] It should
picosecond streak camera after excitation at 355 nm and be noted that at the excitation wavelength, 355 nm, both
displays a lifetime of 1.0 ns. This luminescence is quenched units of the DII-Ir dyad absorb light with the following
in DII-Ir and displays a time profile very similar to the in- probability (see Figure 1): 60% for Ir, 40% for D.
strumental response (Figure 4) from which a lifetime of ca.
The end of pulse transient absorption spectra detected
10 ps can be calculated by deconvolution with the instru- with picosecond resolution of optically matched solutions
mental profile. This value is in reasonable agreement with of the models and DII-Ir are reported in Figure 5. Ir1 dis-
steady-state data, which indicates a reduction in the emis- plays a strong absorption band with a broad maximum
3
sion yield of the order of 1% in passing from the DII model around 780 nm assigned to Ir1, stable on the 3 ns window
to the dyad DII-Ir. In the case of the short dyad D-Ir, no of the experiment. DII displays a sharp band at 750 nm
luminescence could be detected from the D component, with a lifetime of ca. 900 ps, in reasonable agreement with
confirming a more efficient quenching due to electron the luminescence lifetime of 1.0 ns, indicating that the de-
1
transfer responsible for charge separation and allowing us tected species is the singlet excited state DII.[14] The time-
to set an upper limit to the lifetime of τ ϽϽ 10 ps.
resolved spectrum of DII-Ir has a maximum at 765 nm
which, for similarity with the spectrum of D+-Ir– (λmax
=
765 nm) and with the chemically generated spectrum of
DII+, is assigned to the CS state DII+-Ir–. No trace of the
spectral features of DII, which has in fact been shown by
luminescence experiments to decay with a lifetime of 10 ps,
i.e. during the laser pulse, is detected. Similarly, no ab-
3
sorbance ascribable to Ir, which would be easily identifi-
able because of its broad and intense spectrum, appears at
the end of the pulse. This set up the unsolved problem of
the lifetime of DII-3Ir species, which could not be unambig-
uously determined by luminescence techniques (see above).
By transient absorbance experiments we can set an upper
limit of 20 ps for the lifetime of DII-3Ir, a longer lifetime of
this species would in fact cause the appearance in the spec-
trum registered at the end of the pulse of the typical fea-
Figure 4. Luminescence registered at 520 nm from DII-Ir acetoni-
trile solution after excitation with a laser pulse (35 ps, 355 nm,
1 mJ) (full circles) with the instrumental pulse profile (dotted line)
and the fitting (solid line).
3
tures of the Ir absorbance (see Figure 5) which are, on the
contrary, absent. The time evolution of the spectrum, re-
Eur. J. Inorg. Chem. 2007, 5189–5198
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5193

L. Flamigni, B. Ventura, E. Baranoff, J.-P. Collin, J.-P. Sauvage
FULL PAPER
ported in Figure 6 with the extracted time profile, allows us has a lifetime of 440 ns, in good agreement with the lumi-
to derive a lifetime of the CS state of 210 ps. The ab- nescence lifetime.
1
sorbance decays to the baseline without leaving any residual
The singlet excited state D derived from the component
absorbance in the longer timescale, as confirmed by nanose- D displayed an identical absorption spectrum but a shorter
cond flash-photolysis experiments on DII-Ir. No residual lifetime of 490 ps and, as far as the CS state is concerned,
absorbance in the nanosecond time scale can be detected DII+-Ir– has a nearly identical spectrum but a much longer
for DII either, on the contrary nanosecond flash-photolysis lifetime than D+-Ir– (70 ps).
experiments allow us to detect in Ir1 solutions the full evol-
ution of the Ir1 species in air-equilibrated samples which DII-Ir are collected in Table 3.
The transient absorbance data for the models, D-Ir and
3
Photoinduced Processes
Scheme 4 shows the energy level diagrams of the present
[7]
DII-Ir dyad and of the corresponding short dyad D-Ir
and can help to discuss and compare the processes in the
two cases.
These diagrams can be drawn from the luminescence
data collected in Table 2 and from the electrochemical data
reported in Table 1. The excited-state energies are derived
from the maxima of the emission at 77 K (Table 2), whereas
the charge-separated state energy levels are calculated by
simply adding the energy necessary to oxidize the potential
donor and the energy necessary to reduce the potential ac-
ceptor in acetonitrile (Table 1) without further corrections.
Both dyads are made of an easy to oxidize D or DII
component (E1/2 = +0.75 V and E1/2 = +0.76 V in the perti-
nent dyads, respectively) and of the iridium complex which
can be reduced at low potentials (E1/2 = –0.75 V in the short
and E1/2 = –0.78 V in the long dyad) yielding charge-sepa-
rated states D+-Ir– or DII+-Ir– with an energy of 1.51 and
1.54 eV, respectively (Table 1). These are characterized by
an extra electron localized on the metal complex, most
probably on the ligand, and a hole on the triphenylamine
donor. Charge separation is feasible both from the excited
Figure 5. End of pulse absorption spectra of Ir1 (solid line), DII
(dotted line), and DII-Ir (dashed line) in acetonitrile solutions upon
excitation at 355 nm (2.8 mJ/pulse, 35 ps pulse). All solutions had
A = 0.51 at the exciting wavelength.
1
donor, DII-Ir (1D-Ir) and from the excited acceptor DII-
3Ir (D-3Ir) but the excited state quenching mechanism in the
two cases will be different. When Ir is excited, a reductive
quenching of its excited state will occur, i.e. the electron will
move from the HOMO localized on the donor to the
HOMO localized on the Ir complex, with ∆G0 = –0.9 and
–0.86 eV for D-Ir and DII-Ir, respectively. When the donor
is excited, an electron will move from the LUMO localized
on DII to the LUMO on the Ir complex acceptor, with ∆G0
= –1.43 eV for D-Ir and ∆G0 = –1.56 eV for DII-Ir. Both
types of electron transfer take place at room temperature,
as testified by the quenching of the luminescent states local-
ized on both units as shown in Figures 2 and 4.
Figure 6. Time evolution of the transient spectrum of DII-Ir in ace-
tonitrile (2.5 mJ, A = 0.5); spectra taken with delays of 165 ps. In
the inset, the time evolution of the absorbance at 765 nm and ex-
ponential fitting are shown.
Table 3. Transient absorption data at room temperature (excitation
at 355 nm) in air-equilibrated acetonitrile solutions of DII-Ir, D-Ir,
and models.
State
λmax
τ
[nm]
[ns]
We cannot exclude the possibility of an energy transfer
1
from DII-Ir to DII-3Ir, which is spin-forbidden but could
Ir1[a]
D[a]
3Ir1
780
750
–
440
0.49
1D
be made possible by the strong spin-orbit coupling induced
by the heavy Ir ion. The overall result would be undistin-
guishable from a direct electron transfer from the excited
state localized on the donor; however, the 77 K experiment
seems to indicate that such an energy transfer (expected to
be little affected by temperature) does not take place. In the
latter case in fact we would expect a sensitization of the Ir
luminescence upon excitation of the DII unit at 330 nm,
D-Ir[a]
D-3Ir
1D-Ir
D+-Ir–
1DIIc
DII-3Ir
1DII-Ir
DII+-Ir–
Ͻ0.02
Ͻ0.02
0.070
0.90
Ͻ0.02
Ͻ0.02
0.210
–
765
750
–
DII
DII-Ir
765
[a] From ref.[7]
5194
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5189–5198

A Triphenylamine/Bis(terpyridine)IrIII Dyad
Scheme 4. Energy-level diagrams and schematic representation of the processes for D-Ir and DII-Ir.
which does not occur (see Figure 3). Therefore, the follow- Thus, the insertion of a further benzamide group between
ing discussion will not take the spin-forbidden energy trans- the donor and acceptor is not too detrimental to the
fer process into account.
charge-separation process at room temperature. This is dif-
1
Whereas we could measure the reaction rate of DII-Ir ferent at 77 K, where the absence of quenching of the Ir
Ǟ DII+-Ir–, kcs = 1011 s–1, corresponding to the lifetime of component and the persistence of the quenching of the D
10 ps, for the DII-3Ir Ǟ DII+-Ir– reaction we could only set unit in the dyad at 77 K indicate that electron transfer from
a lower limit, kcs Ͼ 5ϫ1010 s–1, based on the results of the DII-3Ir is inefficient and only electron transfer from DII-
1
transient absorbance (see above). In the short dyad D-Ir, Ir continues to be operative with a still remarkable rate of
similar results were obtained for the D-3Ir Ǟ D+-Ir– reac- the order of 20 ps (Figure 3 and Table 2). This can be
tion, for which an identical lower limit of the rate constant understood if one considers that the rigidity of the medium,
kcs Ͼ 5ϫ1010 s–1, corresponding to the time resolution of preventing solvent molecule reorganization around the CS
1
the transient absorbance experiment, was set. For the D- state, destabilizes the CS level by a value which can be quite
Ir Ǟ D+-Ir– step, the improved time resolution of the appa- high[16] and decreases the driving force for electron-transfer
ratus with respect to previously reported data (kcs
Ͼ
reactions originating from the DII-3Ir and the 1DII-Ir
5ϫ1010 s–1)[15] allowed us to set a limit of kcs Ͼ 1011 s–1. states. Under these conditions, the lowest-lying DII-3Ir (ca.
The driving force of the charge-separation reactions are not 2.4 eV) is unable to transfer an electron, whereas the higher
very different in going from the short to the long dyad, 1DII-Ir (ca. 3 eV) has sufficient driving force to transfer the
but the distance between the donor and acceptor changes electron also in rigid media at 77 K.
remarkably, from 1.65 to 2.3 nm. Therefore, a slower charge
Once the charge-separated state DII+-Ir– is formed, it de-
1
1
separation step DII-Ir Ǟ DII+-Ir– compared to D-Ir Ǟ cays to the ground state with a lifetime of 210 ps, three-
D+-Ir– can be measured. On the contrary, for the charge times shorter than D+-Ir– which has a lifetime of 70 ps. The
separation starting from the excited state of the Ir metal driving force of the charge-recombination reaction is high,
complex, either DII-3Ir Ǟ DII+-Ir– or D-3Ir Ǟ D+-Ir–, it is leading us to assume a Marcus-inverted behavior, i.e. the
difficult to make comparisons since the same lower limit rate slows as the driving force increases.[17] The driving force
for the rate constant has been determined. Overall, a slight is slightly higher for DII-Ir compared to D-Ir (∆G0
=
decrease in the rate of charge separation can be registered –1.54 eV for the long dyad against ∆G0 = –1.50 eV), but
in DII-Ir compared to D-Ir; however, the yield of charge most importantly the insertion of a further benzamide
separation – φcs = kcs/Σk = kcs/(kcs + 1/τ0), where τ0 is the group is effective in electronically decoupling the donor and
lifetime of the excited state in the models – is still of the acceptor units in the long dyad and decreases the charge-
1
order of 99% when starting from DII-Ir (τ0 = 1 ns) and recombination rate. The increase from 70 to 210 ps lifetime
nearly 100% when starting from DII-3Ir (τ0 = 490 ns in air). in passing from D-Ir to the long DII-Ir dyad, suggests that
Eur. J. Inorg. Chem. 2007, 5189–5198
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5195

L. Flamigni, B. Ventura, E. Baranoff, J.-P. Collin, J.-P. Sauvage
FULL PAPER
recorded with a Bruker MicroTOF instrument. Cyclic voltammetry
experiments were performed using an EG&G 273A potentiostat, a
Pt working electrode, a Pt counter-electrode, a KCl-saturated calo-
mel electrode (SCE) as a reference, and 0.1  nBu4NPF6 in CH3CN
as the supporting electrolyte. The typical sweep rate was 100 mV/s.
a newly synthesized long triad DII-Ir-A could be more ef-
ficient in producing charge separation over the terminal
components. In the D-Ir-A triad, as discussed above, the
poor efficiency of charge separation was due to the fast step
(k = 1.4ϫ1010 s–1) preferably leading D+-Ir–-A to the
ground state rather than to the fully CS state D+-Ir-A– by
a further electron transfer step (k = 2.4ϫ109 s–1). The de-
crease of the charge recombination step in DII+-Ir–-A (k =
4.8ϫ109 s–1), assuming that the rest of the triad is un-
changed with respect to the previously studied D-Ir-A triad,
would allow us to reach an efficiency of charge separation
of the order of 30%. This would be a noticeable improve-
ment with respect to the yield of ca. 10% achieved in the
short triad D-Ir-A and, if we consider that due to the in-
creased donor–acceptor distance an even longer lifetime of
the final CS would be expected in a long triad DII-Ir-A,
the demanding synthesis of this structure could be under-
taken with some degree of confidence in satisfactory per-
formances.
Starting Compounds: [4Ј-(3,5-Di-tert-butylphenyl)-2,2Ј:6Ј,2ЈЈ-terpy-
ridine]IrCl3,[5a] tpy-ph-NH2,[9] 4-[bis(4-methoxyphenyl)amino]ben-
zoic acid, DII and Ir1[7] were prepared according to the published
procedures.
1: terpy-ph-NH2 (500 mg, 1.54 mmol) was dissolved in dimethyl-
acetamide (8 mL), after which some pyridine (2 mL) was added.
Upon addition of p-nitrobenzoyl chloride (410 mg, 2.2 mmol, ex-
cess), the color of the solution immediately turned from pale yellow
to deep yellow. After stirring at room temperature for 28 h, the
reaction mixture was added dropwise to cold Et2O and was placed
in the freezer for 30 min to induce precipitation. The dark yellow
precipitate of 1 was filtered off and dried under vacuum. Yield:
95% (700 mg). C28H19N5O3 (473.48): calcd. C 71.03, H 4.04, N
14.79; found C 71.45, H 4.03, N 14.47. 1H NMR ([D6]DMSO,
200 MHz): δ = 10.87 (s, 1 H, NH), 8.86–8.83 (m, 6 H, 6-H, 3-H,
3Ј-H, 5Ј-H), 8.44–8.08 (m, 10 H, o1-H, m1-H, o2-H, m2-H, 4-H),
7.72–7.65 (m, 2 H, 5-H) ppm. 13C NMR ([D6]DMSO, 75 MHz): δ
= 164.1, 153.5, 152.6, 149.6, 149.3, 147.6, 147.5, 140.5, 140.3, 132.0,
129.4, 127.7, 125.6, 123.6, 122.4, 120.9, 118.9 ppm. ES-MS: m/z
(calcd.) = 474.16 (474.15) [M + H]+.
Conclusions
We have synthesized a new dyad DII-Ir made of a tri-
phenylamine donor and of a bis(terpyridine)IrIII acceptor
separated by a bridge composed of two benzamide groups,
and we have studied the photoinduced processes. The prop-
erties of the dyad have been compared to those of a corre-
sponding dyad D-Ir characterized by a bridge connecting
the donor and the acceptor consisting of a single benzamide
unit. We showed that, whereas the charge-separation steps
are still very efficient in the long dyad (Ն99%), charge re-
combination is slowed by a factor of three with respect to the
short dyad. This is promising and encourages the use of this
dyad in the assembly of the more elaborate array DII-Ir-A.
This would simply have a further benzamide unit in the
spacer connecting DII to Ir, with respect to the already
studied D-Ir-A triad of Scheme 1 for which an unsatisfac-
tory yield of ca. 10% of charge separation was reported. In
the long triad DII-Ir-A, a lifetime of the CS state DII+-Ir–-
A of the order of 210 ps would be expected, similar to the
dyad studied here. This would decrease the competition of
the recombination to the ground state with respect to the
further electron-transfer step leading to full charge separa-
tion over the extreme components with an efficiency of the
order of 30% with respect to the ca. 10% of the previously
studied D-Ir-A triad. It should also be stressed that, due to
the increased D–A distance, an even longer lifetime of the
final CS state, with respect to the 120 µs of the short D-Ir-
A triad, is expected.
2: Compound 1 (230 mg; 0.49 mmol) was suspended in refluxing
MeOH/THF (1:1, v/v; 100 mL). Pd/C 5% (108 mg) was added as a
solid, and a mixture of hydrazine monohydrate (0.5 mL) in MeOH
(5 mL) was added dropwise over a period of 10 min. After the ad-
dition, the colorless solution was refluxed for 40 min, cooled to
room temperature, filtered through Celite, and concentrated. After
one night in the refridgerator, the precipitate was filtered off and
washed with water, MeOH, and Et2O. Compound 2 was obtained
as a colorless solid. Yield: 97% (210 mg). C28H21N5O (443.50):
calcd. C 75.83, H 4.77, N 15.79; found C 74.14, H 4.63, N 15.20.
1H NMR ([D6]DMSO, 200 MHz): δ = 10.02 (s, 1 H, NH), 8.79
(dd, 4J = 4.7, 5J = 0.8 Hz, 2 H, 6-H), 8.75 (s, 2 H, 3Ј-H, 5Ј-H),
3
8.70 (d, J = 7.9 Hz, 2 H, 3-H), 8.06 (m, 2 H, 4-H), 8.00 (m, 2 H,
o1-H, m1-H), 7.79 (d, 3J = 8.6 Hz, 2 H, o2-H), 7.55 (ddd, 3J = 7.5,
5
3
4J = 4.7, J = 1.0 Hz, 2 H, 5-H), 6.64 (d, J = 8.6 Hz, 2 H, m2-H)
ppm. 13C NMR ([D6]DMSO, 75 MHz): δ = 161.5, 152.0, 151.5,
148.8, 145.7, 145.4, 137.7, 133.8, 127.8, 125.9, 123.5, 120.9, 117.3,
117.2, 116.9, 113.7, 108.9 ppm. ES-MS: m/z (calcd.) = 444.18
(443.17) [M + H]+.
3: DMAP (470 mg, 3.64 mmol, 3.3 equiv.) and EDC (420 mg,
2.20 mmol, 1.9 equiv.) were added to a solution of 4-[bis(4-meth-
oxyphenyl)amino]benzoic acid (634 mg, 1.81 mmol, 1.5 equiv.) in
freshly distilled CH2Cl2 (6 mL). After a few minutes of stirring, the
solution was deep yellow. 2 (500 mg, 1.13 mmol) and DMF (1 mL)
were added, and the reaction mixture was stirred under argon at
40 °C for 10 h. CH2Cl2 was removed under reduced pressure. Water
and saturated NaHCO3 solution (25 mL each) were added resulting
in a pale yellow precipitate which was filtered off and washed with
NaHCO3 solution, water and then with a mixture of CH2Cl2/Et2O
(1:1, v/v) to afford 3 as a pale yellow solid in 71% yield (626 mg,
0.81 mmol). C49H38N6O4 (774.86): calcd. C 75.95, H 4.94, N 10.85;
found C 74.97, H 4.76, N 10.46. 1H NMR ([D6]DMSO, 300 MHz):
Experimental Section
General: 1H NMR spectra were acquired either with a Bruker
WP 200 SY, a Bruker AC300, or a Bruker AM 400 spectrometer,
using the deuterated solvent as the lock and residual solvent as the
internal reference (CD3CN: δ = 1.96 ppm; [D6]DMSO: δ =
2.52 ppm). Electron spray ionization mass spectra (ESI-MS) were
δ = 10.37 (s, 1 H, NH), 10.23 (s, 1 H, NH), 8.78 (d, J = 4.1 Hz, 2
3
3
H, 6-H), 8.70 (s, 2 H, 3Ј-H, 5Ј-H), 8.68 (d, J = 7.9 Hz, 2 H, 3-H),
3
8.08–8.94 (m, 10 H, 4-H, o3-H, m3-H, o4-H, m4-H), 7.82 (d, J =
3
8.9 Hz, 2 H, o5-H), 7.54 (ddd, 3J = 7.5, J = 4.9, 4J = 1.1 Hz, 2 H,
3
5-H), 7.14 (d, 3J = 9.0 Hz, 4 H, a-H), 6.98 (d, J = 9.0 Hz, 4 H, b-
5196
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5189–5198

A Triphenylamine/Bis(terpyridine)IrIII Dyad
3
H), 6.74 (d, J = 8.9 Hz, 2 H, m5-H), 3.77 (s, 6 H, CH3) ppm. ES- tor were driven by customized software (Eurins) which also allowed
MS: m/z (calcd.) = 775.31 (775.30) [M + H]+.
us to acquire spectra at increasing time delays between the pump
and the probe. Typically, 200–500 laser shots were collected and
averaged to obtain a single spectrum at a specific time delay. Ki-
netic analyses were made by selecting the absorbance values of suc-
cessive time-resolved spectra at the selected wavelength and by ap-
plying standard iterative procedures. More details can be found
elsewhere.[19]
DII-Ir: A suspension of [4Ј-(3,5-di-tert-butylphenyl)-2,2Ј:6Ј,2ЈЈ-ter-
pyridine]IrCl3 (48 mg, 0.067 mmol) and 3 (51 mg, 0.066 mmol) in
ethylene glycol (15 mL) was homogenized under ultrasound during
25 min, then heated at 160 °C during 20 min under argon. The
crude product was precipitated with a saturated solution of KPF6,
filtered off, and purified on silica (CH3CN/H2O/KNO3, 100:0:0 to
100:7:0.7). After anion exchange, DII-Ir was obtained as a yellow
Nanosecond laser flash photolysis experiments were performed by
a system based on an Nd-YAG laser (JK Lasers, 355 nm, 1.5–
8.5 mJ, 18 ns pulse) previously described using a right-angle analy-
sis on the excited sample.[20] Experimental uncertainties are esti-
mated to be within 10% for lifetime determination involving simple
exponential, 20% for lifetime determinations involving multiple ex-
ponentials or more complex kinetics, 15% for quantum yields, 20%
for molar absorption coefficients and 3 nm for emission and ab-
sorption peaks. Molecular dimensions were estimated after MM2
energy minimization from CS Chem3D Ultra software.
1
solid. Yield: 70% (85 mg). H NMR (CD3CN, 400 MHz): δ = 9.15
(s, 1 H, 1-NH), 9.10 (s, 2 H, 3Ј-D, 5Ј-D), 9.04 (s, 2 H, 3Ј-H, 5Ј-H),
8.82 (s, 1 H, 2-NH), 8.78 (dd, 3J = 8.2, 4J = 0.7 Hz, 2 H, 3-H),
4
8.73 (dd, 3J = 8.2, J = 0.7 Hz, 2 H, 3-D), 8.29–8.22 (m, 6 H, 4-H,
3
4-D, o3-H), 8.19 (d, J = 8.9 Hz, 2 H, m3-H), 8.02 (d, 3J = 8.8 Hz,
4
3
2 H, o4-H), 7.99 (d, J = 1.7 Hz, 2 H, o-H), 7.90 (d, J = 8.8 Hz,
2 H, m4-H), 7.87 (t, 4J = 1.7 Hz, 1 H, p-H), 7.77–7.69 (m, 6 H, o5-
H, 6-H, 6-D), 7.52–7.47 (m, 4 H, 5-H, 5-D), 7.17 (d, 3J = 8.9 Hz, 4
3
3
H, a-H), 6.96 (d, J = 8.9 Hz, 4 H, b-H), 6.7 (d, J = 8.9 Hz, 2 H,
m5-H), 3.80 (s, 6 H, CH3), 1.52 (s, 18 H, tBu) ppm. HR ES-MS:
m/z (calcd.) = 1678.4467 (1678.4387) [M – PF6]+.
Supporting Information (see footnote on the first page of this arti-
cle): Atom numbering scheme for compounds 1, 2, 3, and DII-Ir.
The DII+ species was generated by the addition of one drop of Br2
to a solution of DII (2.5ϫ10–4 , 2 mm optical path) in acetoni-
trile, the calculated molar absorption coefficient assuming complete
conversion of the starting DII was ε = 27400 –1 cm–1 at 760 nm.
Acknowledgments
We thank Consiglio Nazionale delle Ricerche of Italy (PM-P04-
010, MACOL), Centre National de la Recherche Scientifique
(France), Ministero dell’Istruzione, dell’Universita’ e della Ricerca
of Italy, and COST D31 for financial support.
For the photophysical experiments, the solvents were spectrophoto-
metric-grade acetonitrile at 295 K and butyronitrile at 77 K. If not
otherwise specified, solutions were at ambient temperature and air-
equilibrated. Standard 10 mm fluorescence cells were used at 295 K
whereas experiments at 77 K made use of capillary tubes in a home-
made quartz Dewar filled with liquid nitrogen. Because of geomet-
rical irradiation conditions at 77 K the absolute quantum yield
could not be determined, but the relative emission yields could be
derived with some confidence. When necessary, solutions were
bubbled for 10 min with a stream of argon in home-modified
10 mm fluorescence cells. A Perkin–Elmer Lambda 9 UV/Vis spec-
trophotometer and a Spex Fluorolog II spectrofluorimeter were
used to acquire absorption and emission spectra. Emission quan-
tum yields were determined after correction for the photomultiplier
response, with reference to air-equilibrated (3,5-di-tert-butylphenyl-
2,2Ј;6Ј,2ЈЈ-terpyridine)2IrIII(PF6)3 with φem = 0.022 in air-saturated
acetonitrile.[5a] Luminescence lifetimes (τ) were obtained with IBH
single-photon counting equipment upon excitation at 373 and
331 nm from a pulsed diode source or with an apparatus based on
an Nd:YAG laser (35 ps pulse duration, 355 nm, 0.5 mJ) and a
Streak Camera with an overall resolution of 10 ps.[18] Transient ab-
sorbance in the picosecond range made use of a pump and probe
system based on an Nd-YAG laser (Continuum PY62/10, 35 ps
pulse). The third harmonic (355 nm) at a frequency of 10 Hz and
an energy of ca. 3 mJ/pulse was used to excite the samples whose
absorbance at the excitation wavelength was ca. 0.6. The residual
1064 nm light (ca. 40 mJ) was focused on a stirred 10 cm cell con-
taining a mixture of D2O/D3PO4 to produce a white light contin-
uum which was used as analyzing light. A computer-controlled op-
tical delay stage (Ealing) on the path of the excitation beam pro-
vided a delay between excitation and analysis. The analyzing light
was split into two parts probing irradiated and un-irradiated por-
tions, respectively, of the sample and crossed the sample cell in a
nearly collinear geometry with respect to the excitation beam. The
transmitted probes were fed through optical fibers into a spectro-
graph (Spectrapro 275, Acton) and were detected in two separate
regions of a CCD detector (Princeton Instruments). The control
units for the delay line, for the spectrograph and for the CCD detec-
[1] a) D. Gust, T. A. Moore, A. L. Moore, A. N. Macpherson, A.
Lopez, J. M. DeGraziano, I. Gouni, E. Bittersman, G. R. Seely,
F. Gao, R. A. Nieman, X. C. Ma, L. J. Demanche, S.-C. Hung,
D. K. Luttrull, S.-J. Lee, P. K. Kerrigan, J. Am. Chem. Soc.
1993, 115, 11141–11152; b) K. A. Joliffe, T. D. Bell, K. P. Ghig-
gino, S. J. Langford, M. N. Paddon-Row, Angew. Chem. 1998,
110, 960–963; Angew. Chem. Int. Ed. Engl. 1998, 37, 916–919;
c) H. Imahori, D. M. Guldi, K. Tamaki, Y. Yoshida, C. Luo,
Y. Sakata, S. Fukuzumi, J. Am. Chem. Soc. 2001, 123, 6617–
6628; d) H. Imahori, Y. Sekiguchi, Y. Kashiwagi, T. Sato, Y.
Araki, O. Ito, H. Yamada, S. Fukuzumi, Chem. Eur. J. 2004,
10, 3184–3196.
[2] a) T. J. Meyer, Acc. Chem. Res. 1989, 22, 163–170; b) J. H. Al-
strum-Acevedo, M. K. Brennaman, T. J. Mayer, Inorg. Chem.
2005, 44, 6802–6827; c) L. De Cola, V. Balzani, F. Barigelletti,
L. Flamigni, P. Belser, A. von Zelewsky, M. Frank, F. Vogtle,
Inorg. Chem. 1993, 32, 5228–5238; d) I. M. Dixon, J.-P. Collin,
J.-P. Sauvage, L. Flamigni, Inorg. Chem. 2001, 40, 5507–5517;
e) C. Chiorboli, S. Fracasso, F. Scandola, S. Campagna, S. Ser-
roni, R. Konduri, F. M. MacDonnell, Chem. Commun. 2003,
14, 1658–1659; f) S. Chakraborty, T. J. Wadas, H. Hester, R.
Schmehl, R. Eisenberg, Inorg. Chem. 2005, 44, 6865–6878; g)
M. Borgström, N. Shaikh, O. Johansson, M. F. Anderlund, S.
Styring, B. Åkermark, A. Magnuson, L. Hammarström, J. Am.
Chem. Soc. 2005, 127, 17504–17515.
[3] a) L. Flamigni, I. M. Dixon, J.-P. Collin, J.-P. Sauvage, Chem.
Commun. 2000, 2479–2480; b) I. M. Dixon, J.-P. Collin, J.-P.
Sauvage, F. Barigelletti, L. Flamigni, Angew. Chem. Int. Ed.
2000, 39, 1292–1295.
[4] a) L. Flamigni, G. Marconi, I. M. Dixon, J.-P. Collin, J.-P.
Sauvage, J. Phys. Chem. B 2002, 106, 6663–6671; b) E. Baran-
off, K. Griffiths, J.-P. Collin, J.-P. Sauvage, B. Ventura, L. Fla-
migni, New J. Chem. 2004, 28, 1091–1095.
[5] a) J.-P. Collin, I. M. Dixon, J.-P. Sauvage, J. A. G. Williams, F.
Barigelletti, L. Flamigni, J. Am. Chem. Soc. 1999, 121, 5009–
5016; b) I. M. Dixon, J.-P. Collin, J.-P. Sauvage, L. Flamigni,
S. Encinas, F. Barigelletti, Chem. Soc. Rev. 2000, 29, 385–391.
Eur. J. Inorg. Chem. 2007, 5189–5198
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
5197

L. Flamigni, B. Ventura, E. Baranoff, J.-P. Collin, J.-P. Sauvage
FULL PAPER
[6]
[7]
[8]
[9]
[10]
10 ps, due to a better deconvolution procedure of the signal
with the instrumental response.
E. Baranoff, J.-P. Collin, L. Flamigni, J.-P. Sauvage, Chem. Soc.
Rev. 2004, 33, 147–155.
L. Flamigni, E. Baranoff, J.-P. Collin, J.-P. Sauvage, Chem. Eur.
J. 2006, 12, 6592–6606.
E. Baranoff, I. M. Dixon, J.-P. Collin, J.-P. Sauvage, B. Ventura,
L. Flamigni, Inorg. Chem. 2004, 43, 3057–3066.
P. Lainé, F. Bedioui, P. Ochsenbein, V. Marvaud, M. Bonin, E.
Amouyal, J. Am. Chem. Soc. 2002, 124, 1364–1277.
J. C. Sheehan, P. A. Cruickshank, G. L. Broshart, J. Org. Chem.
1961, 26, 2525–2528.
[16] a) Values as high as 0.8 eV have been reported, see, for exam-
ple: G. L. Gaines III, M. P. O’Neil, W. A. Swec, M. P. Niem-
czyk, M. R. Wasielewski, J. Am. Chem. Soc. 1991, 113, 719–
721; b) L. Hammarström, F. Barigelletti, L. Flamigni, N. Arm-
aroli, A. Sour, J.-P. Collin, J.-P. Sauvage, J. Am. Chem. Soc.
1996, 118, 11972–11973 and references cited therein.
[17] a) R. A. Marcus, N. Sutin, Biochim. Biophys. Acta 1985, 811,
265; b) R. A. Marcus, Angew. Chem. Int. Ed. Engl. 1993, 32,
1111.
[11] L. Flamigni, B. Ventura, F. Barigelletti, E. Baranoff, J.-P. Col-
lin, J.-P. Sauvage, Eur. J. Inorg. Chem. 2005, 1312–1318.
[12] W. Leslie, A. S. Bastanov, J. A. K. Howard, J. A. G. Williams,
Dalton Trans. 2004, 623–631.
[13] R. I. Walter, J. Am. Chem. Soc. 1966, 88, 1923–1930.
[14] Some instability was displayed by the DII solution under the
harsh laser energy conditions of the pump and probe experi-
ment.
[18] L. Flamigni, J. Phys. Chem. 1993, 97, 9566–9572.
[19] a) B. Ventura, A. Degli Esposti, B. Koszarna, D. T. Gryko, L.
Flamigni, New J. Chem. 2005, 29, 1559–1566; b) L. Flamigni,
N. Armaroli, F. Barigelletti, V. Balzani, J.-P. Collin, J.-O. Dal-
bavie, V. Heitz, J.-P. Sauvage, J. Phys. Chem. B 1997, 101, 5936–
5943.
[20] a) L. Flamigni, J. Phys. Chem. 1992, 96, 3331–3337; b) L. Fla-
migni, J. Chem. Soc. Faraday Trans. 1994, 90, 2331–2336.
Received: June 28, 2007
[15] Since previously reported experiments the time resolution of
the luminescence experiments was improved from 20 ps to
Published Online: October 2, 2007
5198
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 5189–5198