Ruthenium(II)-polyimine-coumarin light-harvesting molecular arrays: Design rationale and application for triplet-triplet-annihilation-based upconversion

Article abstract of DOI:10.1002/chem.201101377

Ru^II-bis-pyridine complexes typically absorb below 450 nm in the UV spectrum and their molar extinction coefficients are only moderate (Iμ<16000M^-1cm^-1). Thus, Ru ^II-polyimine complexes that show intense visible-light absorptions are of great interest. However, no effective light-harvesting ruthenium(II)/organic chromophore arrays have been reported. Herein, we report the first visible-light-harvesting Ru^II-coumarin arrays, which absorb at 475 nm (Iμ up to 63300M^-1cm^-1, 4-fold higher than typical Ru^II-polyimine complexes). The donor excited state in these arrays is efficiently converted into an acceptor excited state (i.e., efficient energy-transfer) without losses in the phosphorescence quantum yield of the acceptor. Based on steady-state and time-resolved spectroscopy and DFT calculations, we proposed a general rule for the design of Ru ^II-polypyridine-chromophore light-harvesting arrays, which states that the ¹IL energy level of the ligand must be close to the respective energy level of the metal-to-ligand charge-transfer (MLCT) states. Lower energy levels of ¹IL/³IL than the corresponding ¹MLCT/³MLCT states frustrate the cascade energy-transfer process and, as a result, the harvested light energy cannot be efficiently transferred to the acceptor. We have also demonstrated that the light-harvesting effect can be used to improve the upconversion quantum yield to 15.2% (with 9,10-diphenylanthracene as a triplet-acceptor/annihilator), compared to the parent complex without the coumarin subunit, which showed an upconversion quantum yield of only 0.95%. Copyright

Full text of DOI:10.1002/chem.201101377

FULL PAPER
DOI: 10.1002/chem.201101377
Ruthenium(II)–Polyimine–Coumarin Light-Harvesting Molecular Arrays:
Design Rationale and Application for Triplet–Triplet-Annihilation-Based
Upconversion
Wanhua Wu,[a] Shaomin Ji,[a] Wenting Wu,[a] Jingyin Shao,[a] Huimin Guo,[a]
Tony D. James,[b] and Jianzhang Zhao*[a]
Abstract: RuII–bis-pyridine complexes
typically absorb below 450 nm in the
UV spectrum and their molar extinc-
tion coefficients are only moderate (e<
16000mꢀ1 cmꢀ1). Thus, RuII–polyimine
complexes that show intense visible-
light absorptions are of great interest.
However, no effective light-harvesting
into an acceptor excited state (i.e., effi-
cient energy-transfer) without losses in
the phosphorescence quantum yield of
the acceptor. Based on steady-state
and time-resolved spectroscopy and
DFT calculations, we proposed a gener-
al rule for the design of RuII–polypyri-
ligand charge-transfer (MLCT) states.
1
Lower energy levels of IL/3IL than the
corresponding 1MLCT/3MLCT states
frustrate the cascade energy-transfer
process and, as a result, the harvested
light energy cannot be efficiently trans-
ferred to the acceptor. We have also
demonstrated that the light-harvesting
effect can be used to improve the up-
conversion quantum yield to 15.2%
(with 9,10-diphenylanthracene as a trip-
let-acceptor/annihilator), compared to
the parent complex without the cou-
marin subunit, which showed an upcon-
version quantum yield of only 0.95%.
dine–chromophore
light-harvesting
1
ruthenium(II)/organic
chromophore
arrays, which states that the IL energy
level of the ligand must be close to the
respective energy level of the metal-to-
arrays have been reported. Herein, we
report the first visible-light-harvesting
RuII–coumarin arrays, which absorb at
475 nm (e up to 63300mꢀ1 cmꢀ1, 4-fold
higher than typical RuII–polyimine
complexes). The donor excited state in
these arrays is efficiently converted
Keywords: coumarin
·
light-har-
vesting · phosphorescence · photo-
chemistry · ruthenium
Introduction
plexes, such as [RuACHTUNGRTENU(NG bpy)3]CATHUNGTREN[NUNG PF6]2, are in the blue spectroscop-
ic region (below 450 nm) and the molar absorption coeffi-
cient (e) values are only moderate. Therefore, it is important
to be able to construct RuII–polypyridine complexes that
display light-harvesting ability in the visible-light range and
efficiently transfer the harvested excitation energy to the co-
ordination center (energy acceptor) to initiate cascade pho-
tophysical processes involving the triplet excited states, such
as photoinduced-electron-transfer/charge-separation, photo-
oxidation, and photovoltaics.[7,10–12]
Ruthenium(II)–polypyridine complexes are versatile materi-
als for applications in artificial photosynthesis, dye-sensi-
tized solar cells, photocatalysis, luminescent molecular
probes, and photoresponsive molecular devices.[1–9] RuII–pol-
ypyridine complexes, such as [RuACHTUNRGTNENG(U bpy)3]ACHTUNTGREN[NUNG PF6]2, have particu-
larly adventitious photophysics in that they have long phos-
phorescence lifetimes (from a few hundreds ns to ms, owing
to the metal-to-ligand charge transfer triplet excited state,
that is, 3MLCT), large Stokes shifts (about 150 nm), and
long emission wavelengths (beyond 600 nm). Unfortunately,
the UV/Vis absorptions of typical RuII–polyimine com-
Previous attempts at preparing RuII–chromophore light-
harvesting arrays that are effective in the visible-light spec-
trum have demonstrated their elusive photophysics; that is,
3
very often, the photophysics of the MLCT emission or the
3
1MLCT! MLCT energy-transfer collapsed with the intro-
duction of an organic chromophore. For example, attempts
to enhance the visible-light absorption and to funnel the
[a] W. Wu, S. Ji, W. Wu, J. Shao, Dr. H. Guo, Prof. J. Zhao
State Key Laboratory of Fine Chemicals
Dalian University of Technology
Dalian 116024 (P. R. China)
3
energy to the emissive MLCT state with BODIPY (boron–
dipyrromethene) as the antenna/energy donor have been
unsuccessful, as was the case with a recently reported IrIII–
BODIPY molecular array.[11] BODIPY is well-known for its
high fluorescence quantum yield and intense absorption in
the visible spectrum (unsubstituted BODIPY shows an in-
tense absorption at about 500 nm, e=79000mꢀ1 cmꢀ1).
Whilst the naphthalimide chromophore does function as an
energy donor to RuII complexes, the molar extinction coeffi-
Fax : (+86)41184986236
E-mail: zhaojzh@dlut.edu.cn
[b] Prof. T. D. James
Department of Chemistry
University of Bath
Bath BA2 7AY (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101377.
Chem. Eur. J. 2012, 18, 4953 – 4964
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4953

cient (e) at 430 nm is only enhanced by a factor of 2.0 (e=
2.0ꢁ104 mꢀ1 cmꢀ1) and the emission of the array is much
weaker than that of the parent complex.[13] To the best of
our knowledge, no successful visible-light-harvesting RuII–
polyimine complexes have been reported in which the total
absorption is increased and in which the excited state of the
energy donor is efficiently converted into the acceptor excit-
ed state (i.e., efficient energy-transfer) without losses in the
quantum yield of the acceptor.
We have been interested in the photophysics of transition-
metal complexes, such as those containing RuII, for
a while.[9,14] To elucidate the photophysics of RuII–organic-
chromophore light-harvesting arrays, herein, we set out to
design visible-light-harvesting arrays based on RuII com-
plexes that showed intense absorption beyond 450 nm, by
using coumarin as the antenna/donor and a MLCT chromo-
phore as the acceptor. With one of these RuII–chromophore
arrays, the total absorption was increased and the excited
state of the energy donor was efficiently converted into the
excited state of the acceptor (i.e., efficient energy-transfer)
without a loss in the quantum yield of the acceptor (e=
6.33ꢁ104 mꢀ1 cmꢀ1 at 470 nm).
desired direct interactions (such as intramolecular collision)
between the coumarin moiety and the MLCT chromophore
(RuII coordination center). The interaction between the cou-
marin chromophore and the RuII-coordination center may
complicate the photophysics of the arrays. The closed-ring
form of the coumarin subunit in complex Ru-3 will have
3
a slightly different IL state compared to that of Ru-2. Thus,
the photophysical properties of complexes Ru-2 and Ru-3
may be different. Owing to intramolecular hydrogen-bond-
ing interactions, we proposed that the coumarin subunits in
complexes Ru-2 and Ru-3 were coplanar with the Phen
ligand. To elucidate the effect of intramolecular hydrogen
bonds on the photophysical properties, the coumarin subunit
and the phen/imidazole ligand were separated in complex
Ru-4. We observed different photophysical properties for
the four complexes. DFT/TDDFT calculations were carried
3
out to check the IL triplet excited state to help the design
of light-harvesting arrays (see the Supporting Information).
Enhanced visible-light absorption and efficient energy-trans-
fer from the antenna to the energy acceptor: The S0!
1MLCT absorption band in parent complex Ru-1 was ob-
served at 450 nm (Figure 1). Complexes Ru-2, Ru-3, and
Ru-4 showed substantially more-intense absorption at
450 nm and at higher wavelengths. For example, the e value
of Ru-2 (6.33ꢁ104 mꢀ1 cmꢀ1) was about 4.0-fold higher than
that of Ru-1 (1.60ꢁ104 mꢀ1 cmꢀ1). The UV/Vis absorptions of
the ligands were also studied and intense absorptions were
found in the visible range for ligands L2, L3, and L4 (see
the Supporting Information). DFT/TDDFT calculations of
the ligands clearly demonstrated that the energy level of the
We demonstrated that the energy levels of the S1 state of
the energy donor (coumarin ligand) must be higher than
those of the MLCT acceptor because the energy-transfer
1
3
1IL! MLCT! MLCT pathway and light-harvesting be-
1
comes inefficient when the IL energy level is lower than
the MLCT state.
1
As a preliminary application of the light-harvesting array,
these complexes were used as triplet photosensitizers for
triplet–triplet annihilation (TTA) upconversion.[15] By using
laser excitation at 473 nm (power density: 71 mWcmꢀ2; note
that terrestrial solar radiation is 100 mWcmꢀ2), the mixed
sensitizer/annihilator solution displayed a blue emission at
400 nm, as a direct result of the light-harvesting effect. Up-
conversion quantum yields of up to 15.2% were observed
(non-coherent light with low power density was sufficient to
sensitize the upconversion). Under the same experimental
conditions, the parent complex, which did not show the
light-harvesting effect, did not show significant upconver-
sion.
1
vertical IL excited states could be precisely predicted (see
the Supporting Information) and that this information will
be helpful for the design of light-harvesting RuII arrays.
The absorption maxima of complex Ru-2 in the visible
spectrum (475 nm) was close to that of Ru-1 (Figure 1a);
1
thus, the energy level of the Frank–Codon IL state of the
coumarin ligand of complex Ru-2 was close to that of the in-
1
trinsic MLCT state. However, for complex Ru-3, a slightly
red-shifted absorption (485 nm) was observed; thus, the
1
energy level of IL state of Ru-3 was probably lower than
1
the MLCT state. Conversely, complex Ru-4 showed blue-
shifted absorption bands compared to complexes Ru-2 and
Ru-3, which could be due to the lack of intramolecular hy-
drogen-bonding interactions or to the decreased intramolec-
ular-charge-transfer (ICT) effect of the coumarin ligand of
Ru-4. The absorption differences between the complexes in
the visible range (400–550 nm) were mainly due to the cou-
marin ligand, based on the DFT/TDDFT calculations on the
ligands (see the Supporting Information).
All of the complexes emitted at 608 nm (Figure 1b) and
the fluorescence was completely quenched in all cases. Thus,
we proposed that these complexes shared the same emissive
3MLCT triplet excited states. However, the complexes
showed different emission intensities. Notably the emission
intensity of complex Ru-2 was 5.4 times that of Ru-1 on ex-
citation at 490 nm, which clearly demonstrated the effective
Results and Discussions
Design of the light-harvesting array: Coumarin is well-
known for its intense absorption in the visible spectrum.[16]
Thus, we designed molecular arrays Ru-2, Ru-3, and Ru-4
(Scheme 1) that contained coumarin subunits because the
energy level of the S1 and T1 states of the ligands were close
1
3
to the MLCT and MLCT states of the RuII complexes, re-
spectively (the energy levels of the T1 excited states of the
ligands were calculated by using DFT/time-dependent-DFT
(TDDFT) methods, see the Supporting information).[9,17]
Complex Ru-1 was used as a model complex in the study.
All of the ligands and complexes were obtained in satisfac-
tory yields. Imidazole was used as a rigid linker to avoid un-
4954
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4953 – 4964

Ruthenium(II)–Polyimine–Coumarin Molecular Arrays
FULL PAPER
Scheme 1. Synthesis of Ru-1 and RuII–coumarin arrays Ru-2, Ru-3, and Ru-4: a) 1) CH
2ACHTUNRGTNEUNG(COOEt)2, EtOH, reflux, 6 h; 2) HCl/AcH, reflux, 6 h; b) POCl3,
DMF, 0–58C; c) KBr, H2SO4/HNO3, 1008C, 3 h; d) CH3COOH, reflux, 6 h; e) BrCH2CH2CH2Cl, Na2CO3, 1008C, 11 h; f) HI, HCl, reflux, 60 h;
g) CH3COOH, reflux, 6 h; h) EtOH, RT, 2 h, 2,2’-bipyridine, 1008C, 24 h; i) Br2, acetic acid, RT; j) 4-formylphenylboronic acid, [PdACHTNURTGNE(UNG PPh3)4], ethanol/tolu-
ene/water (1:1:2, v/v/v).
light-harvesting of Ru-2 in the visible spectrum because Ru-
2 had a higher e value at the excitation wavelength
(490 nm). Thus, the emission of Ru-2 was due to its light-
harvesting effect, thereby indicting an efficient cascade
energy-transfer. In other words, the total absorption was in-
creased and the donor excited state was efficiently convert-
Chem. Eur. J. 2012, 18, 4953 – 4964
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4955

J. Zhao et al.
sorptions in the UV range (l<400 nm).[18] Whilst another
RuII complex has shown moderate e enhancement in the
visible spectrum (e=5.0ꢁ104 mꢀ1 cmꢀ1 at 427 nm, a 5.0-fold
3
increase over the model complex), the MLCT emission was
greatly reduced compared to the model complex.[13] More-
over, RuII–chromophore arrays that contained BODIPY as
the antenna failed to show any phosphorescence.[10,11] For
the BODIPY-incorporated RuII arrays, we proposed that the
1IL state energy level was lower than 1MLCT state; thus,
BODIPY was unable to funnel the absorbed excitation
1
3
energy to the MLCT state. Furthermore, the MLCT excit-
3
ed state was quenched by the IL state that was localized on
the BODIPY moiety. We believe that this general rule will
be useful for the design of efficient RuII-based light-harvest-
ing arrays that work in the visible range.
The weak emission of complex Ru-3 may be caused by
1
non-efficient upward IC from 1IL! MLCT (1IL energy
level was lower than 1MLCT), or by the lower-lying 3IL
3
state. The lower energy level of the IL state may be respon-
sible for the weak emission.[9,13,19] For complex Ru-4, we
3
proposed that the IL energy level of the ligand was higher
3
than MLCT (supported by the UV/Vis absorption spectra);
thus, the intrinsic MLCT photophysics were not perturbed
3
Figure 1. UV/Vis absorption and phosphorescence spectra of arrays Ru-1,
Ru-2, Ru-3, and Ru-4. a) UV/Vis absorption spectra (1.0ꢁ10ꢀ5 moldmꢀ1
in MeCN); b) visible-light-harvesting effect: emission spectra of the com-
plexes (lex =490 nm, 2.0ꢁ10ꢀ6 moldmꢀ3 in MeCN). The solution was
purged with Ar for 30 min before the measurements were taken. All
emission spectra were measured under the same conditions (208C).
and the emission of Ru-4 was similar to that of Ru-1. Fur-
thermore, the absorption of Ru-4 at 490 nm was not signifi-
cantly improved compared to that of Ru-1 (Figure 1a).
Furthermore, the DFT/TDDFT calculations on the li-
gands predicted different energy levels of the T1 excited
states: 2.70 eV (460 nm), 2.00 eV (620 nm), 1.96 eV
(631 nm), and 2.16 eV (573 nm) for ligands L-1, L-2, L-3,
and L-4, respectively (see the Supporting Information). We
demonstrated that the DFT/TDDFT method was helpful for
the rational design of light-harvesting RuII–fluorophore
arrays with matched energy levels. The photophysical prop-
erties of the ligands and the complexes are summarized in
Table 1.
ed into the acceptor excited state (i.e., efficient energy-
transfer) without any loss in the acceptor phosphorescence
quantum yield. Complex Ru-3 showed the weakest emission
and the emission of Ru-4 was similar to Ru-1.
To the best of our knowledge, complex Ru-2 is the first
example of an effective RuII–chromophore light-harvesting
array that shows intense absorption in the visible range
(above 450 nm) and in which the harvested excitation
energy is efficiently transferred to the energy acceptor (the
coordination center). RuII–polypyridine–coumarin (without
the NEt2 moiety) arrays that contain carbazole moieties
have been reported, but they only showed more-intense ab-
UV/Vis absorption and excitation spectra of the complexes:
Demonstration of the light-harvesting effect: To evaluate
the light-harvesting effect, we compared the UV/Vis absorp-
tion spectra and the phosphorescence excitation spectra of
Table 1. Photophysical parameters of the ligands and the RuII complexes.
labs [nm][a]
e[b]
Emission properties
I0/I100
[a]
[e]
[h]
[h]
lem
F [%]
t [ns][f]
kr [sꢀ1
]
knr [sꢀ1
]
L-1
L-2
L-3
L-4
Ru-1
Ru-2
Ru-3
275
442
481
418
457
475
485
3.16
5.08
3.95
2.33
1.62
6.33
6.05
421
502
512
495
608
607
608
10.2[c]
1.2
1.3
1.2
1.3
35
4.81
2.56
2.57
2.85
1125
3785
2490
2.12ꢁ107
1.46ꢁ108
1.16ꢁ108
1.75ꢁ107
1.26ꢁ105
2.33ꢁ104
1.81ꢁ103
1.87ꢁ108
2.45ꢁ108
2.74ꢁ108
3.33ꢁ108
7.63ꢁ105
2.41ꢁ105
4.00ꢁ105
37.3[c]
29.7[c]
5.0[c]
14.2[d]
8.8[d]
765
62
0.5 (485 nm)[d]
0.8 (450 nm)[d]
6.0[d]
Ru-4
423
6.53
607
171
3109
1.93ꢁ104
3.02ꢁ105
[a] Solvent: MeCN (1.0ꢁ10ꢀ5 moldmꢀ3). [b] Molar extinction coefficient at the absorption maxima, e: 104 mꢀ1 cmꢀ1. [c] Quinine sulfate used as the stan-
dard (F=0.547 in 0.05m sulfuric acid). [d] Deaerated solution with [Ru(phen)(bpy)2][PF6]2 (F=0.06 in MeCN) as the standard. [e] Ratio of the emission
intensities of the complexes under N2 (I0) and O2 atmospheres (I100). [f] Luminescence lifetimes: 1.0ꢁ10ꢀ5 moldmꢀ3 in MeCN. Measured in deaerated so-
lution. [g] Radiative deactivation rate constant (kr) and non-radiative deactivation rate constant (knr). kr =Fem/tem, knr =(1ꢀFem)/tem
A
R
ACHTUNGTRENNUNG
.
4956
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4953 – 4964

Ruthenium(II)–Polyimine–Coumarin Molecular Arrays
FULL PAPER
the complexes (Figure 2). It is well-known that superimposa-
ble excitation/UV/Vis absorption spectra indicate an effi-
cient light-harvesting effect.[13,18]
crepancy between the UV/Vis absorption and the excitation
1
spectrum was due to a frustrated 1IL! MLCT transition,
1
owing to the energy level of the IL state being lower than
1
that of the MLCT state (Figure 1a). Instead, we proposed
1
1
that the undesired MLCT! IL energy-transfer was possi-
ble for complex Ru-3.
For complex Ru-4, the excitation/UV/Vis absorption pro-
file was reversed. The excitation spectrum was red-shifted
compared to the UV/Vis absorption spectrum and the exci-
tation maximum was also located at 450 nm. The non-effi-
1
1
3
cient IL! MLCT! MLCT transition at 420 nm may have
1
been due to significant non-radiative IL!S0 decay of the
ligand (knr =3.33ꢁ108 sꢀ1, F=5.0%; Table 1). A similar phe-
nomenon was observed for a RuII–naphthalimide array.[12]
Furthermore, the lack of intramolecular hydrogen bonding
in compound Ru-4 may influence the photophysics.
Illustration of the photophysical process by using a Jablonski
diagram: The photophysics of the arrays, that is, the emis-
sion wavelength, the phosphorescence lifetimes, and the
light-harvesting effect, were summarized by a qualitative Ja-
blonski diagram (Scheme 2). Complex Ru-1 displayed typi-
cal MLCT photophysics and the light-harvesting effect was
operating for this model complex; that is, the intense ab-
sorption of the imidazole ligand in the UV range (at about
286 nm) was efficiently funneled to the 1MLCT state be-
cause the energy level of the imidazole-localized singlet
Figure 2. Comparison of the normalized UV/Vis absorption and the exci-
tation spectra of: a) Ru-1, b) Ru-2, c) Ru-3, and d) Ru-4 (1.0ꢁ
10ꢀ5 moldmꢀ3 in MeCN, 208C). The excitation spectra of arrays Ru-1,
Ru-2, and Ru-3 were recorded at lem =608 nm; the excitation spectrum
of array Ru-4 was recorded at lem =607 nm.
1
state (1IL) was much higher than the MLCT state, thereby
1
3
For complex Ru-1, the excitation spectrum was superim-
posable on the UV/Vis spectrum (Figure 2a). This result
was reasonable because the 450 nm band was due to the
making the 1IL! MLCT! MLCT process efficient. Fur-
3
thermore, the energy level of the IL state was much higher
that the MLCT state. Therefore, there was no interaction
3
1
S0! MLCT transition, which effectively excited the phos-
between these two triplet states, and the intrinsic photophy-
phorescence. The yield of 1MLCT! MLCT intersystem
sics of the [RuACTHNGUTERNNUG
(bpy)3] complex was retained.[17]
3
crossing (ISC) is known to be close to unity.[20,21]
1
1
For Ru-2, both S0! MLCT and S0! IL transitions con-
tributed to the absorption bands at 450 nm (Figure 1a) and
the former transition efficiently excited the phosphorescence
1
3
through a S0! MLCT! MLCT process. However, the fate
1
of the S0! IL process was dictated by the relative energy
1
1
1
3
levels of the IL and MLCT states (a direct IL! MLCT
process was strongly forbidden, owing to the different spin
1
manifold of the states involved in this transition). The IL!
1MLCT process was only efficient if the energy level of the
1IL state of the donor (fluorophore) was higher than the
1MLCT state of the acceptor (RuII-coordination center).
The excitation spectrum of complex Ru-2 was superimposa-
1
ble with its UV/Vis absorption (Figure 2b). Thus, the IL!
1MLCT! MLCT process and the light-harvesting effect of
Scheme 2. Qualitative Jablonski diagram showing the rule for effective
light-harvesting of the RuII–polypyridine–chromophore arrays (Ru-2
shown as an example). The relative energy levels of the states may vary
from complex to complex. GS is the ground state (S0), 1MLCT* is the
metal-to-ligand-charge-transfer excited state (singlet), ISC is intersystem
crossing, 3MLCT is the metal-to-ligand-charge-transfer excited state (trip-
let), TTE is triplet–triplet equilibrium (here, it is the equilibrium between
3
Ru-2 were efficient.
Interestingly, the excitation spectrum of complex Ru-3
was different from its UV/Vis absorption (Figure 2c). The
excitation maximum was blue-shifted when compared to the
UV/Vis absorption, which indicated that the absorption at
480 nm was inefficient to excite the phosphorescence. This
conclusion was supported by the lower phosphorescence
quantum yield of Ru-3 with excitation at 485 nm than that
with 450 nm excitation (Table 1). We proposed that the dis-
3
3
3
the MLCT and IL states), IC is internal conversion, IL is the coumarin-
localized triplet excited state, 1IL is the coumarin-localized singlet excited
state, Allowed processes are marked by a tick. The spin state of the excit-
ed electronic states or the relative energy levels dictates the possibility of
a process.
Chem. Eur. J. 2012, 18, 4953 – 4964
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4957

J. Zhao et al.
1
Because the energy level of the IL state was close to that
we proposed that the 3IL excited state of Ru-3 was at
a lower energy level than the MLCT state, which led to the
1
3
of the MLCT state for complex Ru-2, it displayed the light-
harvesting effect, but in a more red-shifted region than Ru-
1. Therefore, on absorption by the coumarin ligand at
475 nm (e=63300 cmꢀ1 molꢀ1 dm3), the energy of the 1IL
weak emission of Ru-3.
Spin-density surfaces of the RuII complexes: The IL excited
states of complexes Ru-2, Ru-3, and Ru-4 were confirmed
by DFT calculations on the spin-density surface of the T1
state (Figure 4). For complex Ru-1, the spin density of the
3
1
state was efficiently transferred to the MLCT state and fi-
3
1
nally to the emissive MLCT state (Figure 1b). Direct IL!
3IL transition was also possible.
A longer phosphorescence lifetime was observed for com-
plex Ru-2 (3.8 ms) versus Ru-1 (1.1 ms). We proposed that
3
the energy level of the IL state of complex Ru-2 was close
3
to the emissive MLCT state. Thus, a triplet–triplet equilibri-
um (TTE) was established.[9,17] The 3IL state was usually
a long-lived non-radiative excited state and its lifetime was
3
much longer than that of the MLCT state;[17] thus, it served
as an energy reservoir to funnel energy to the emissive
3
3MLCT state, and the lifetime of the MLCT state was pro-
longed. A lower phosphorescence quantum yield (F=0.09)
was observed for complex Ru-2 compared to Ru-1 (F=
0.14).
Nanosecond time-resolved transient difference absorption
spectroscopy: The nanosecond time-resolved transient ab-
sorptions of the complexes were also studied (Figure 3). For
complex Ru-1, typical transient absorption of the RuII–poly-
imine complexes was observed (Figure 3a), with the bleach-
ing of the ground state at 450 nm (the bleaching at 600 nm
was due to the emission of complex Ru-1). For complex Ru-
Figure 4. Spin-density distribution for the lowest triplet state of arrays
Ru-1, Ru-2, Ru-3, and Ru-4. Calculated at the DFT/B3LYP/6-31G(d)/
LANL2DZ level with Gaussian 09W.[34]
2, both the ground-state bleach of the MLCT and the intra
A
ACHTUNGERTNlNUNG i-
triplet state was distributed over the RuII center and the bpy
ligands, which was in agreement with the MLCT excited
state of typical RuII–polyimine complexes. However, for
complexes Ru-2, Ru-3, and Ru-4, the spin density was
mainly distributed over the coumarin moiety and the Phen
Furthermore, transient absorption beyond 500 nm was ob-
served, which was tentatively assigned to the coumarin
moiety. For complexes Ru-3 and Ru-4, similar transient ab-
sorption profiles were observed. The bleaching of Ru-3 at
3
3
480 nm was due to the IL state, which indicated that the IL
ligand. Thus, the IL state was the major component of the
component in the triplet excited state was dominant. Thus,
triplet excited states of these complexes.
Above all, the unique photo-
physics of complex Ru-3, that
is, weak emission and a blue-
shifted excitation spectrum
compared to its UV/Vis absorp-
tion spectrum (Figure 1b and
Figure 2c), were also rational-
ized by the Jablonski diagram
(Scheme 2), for which the
ligand-centered 1IL and 3IL
states had lower energy levels
than the 1MLCT and 3MLCT
states of the coordination
center, respectively. Two path-
ways were possible for the
light-harvesting
effect:
the
and
1
3
1IL! MLCT! MLCT
3
3
1IL! IL! MLCT; thus, re-
versed
energy-transfer
of
1MLCT! IL
was possible.
1
Figure 3. Nanosecond time-resolved transient absorption of: a) Ru-1, b) Ru-2, c) Ru-3, and d) Ru-4 (1.0ꢁ
10ꢀ5 m, in MeCN, 208C).
Moreover, the low phosphores-
4958
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4953 – 4964

Ruthenium(II)–Polyimine–Coumarin Molecular Arrays
FULL PAPER
cence quantum yield of complex Ru-3 (F=0.0045) was ra-
(longer wavelength), has attracted much attention owing to
its applications in optoelectric materials, molecular probes,
photovoltaics, and artificial photosynthesis.[2,9,23]
3
tionalized by the lower energy level of the IL state than the
3MLCT state of the array. As a result, no efficient TTE was
3
3
established between the MLCT and the IL states. The radi-
ative rate constant (kr =1.81ꢁ103 sꢀ1) was much smaller
than the other complexes. Furthermore, the energy gap of
Currently, several approaches are available for light up-
conversion. For example, two-photon absorption (TPA) dyes
can absorb at longer wavelength and produce fluorescence
emissions at shorter wavelength. However, TPA can usually
only be achieved with coherent laser irradiation, with typical
power densities of the order of MWcmꢀ2 (ꢁ106 Wcmꢀ1).[24]
This value is well beyond the power density of a normal
light source, or of terrestrial solar radiance (typically
0.10 Wcmꢀ2). Furthermore, the emission wavelength of
a TPA dye cannot be readily tuned because it is difficult to
tailor the chromophore to tune the emission wavelength,
and at the same time, to maintain a large TPA cross-section.
Recently, a new method for light upconversion by using
TTA has emerged (see Scheme 3).[25–32] In this approach, an
energy donor (sensitizer) is excited first. Next, the triplet
state (T1) of the sensitizer is populated through ISC, and the
energy is transferred (triplet–triplet energy-transfer, TTET;
[Eq. (1a)]) to the triplet state of the accepter. Then, two
1
3
the IL and the IL states may have also been important for
the photophysics of the complexes. Because the 3IL state
was a non-emissive excited state (the existence of this dark
state was supported by the upconversion experiments, see
below), the phosphorescence quantum yield of complex Ru-
3 was greatly reduced. The photophysics of complex Ru-4
was similar to Ru-1.
We demonstrated that the effective absorption of the
RuII–polypyridine–chromophore arrays (absorption bands
that could either lead to phosphorescence or to produce
a triplet excited state), could not be extended into the red-
region of the spectrum without limitations. The extent to
which this restriction was imposed directly correlated with
1
the energy level of the MLCT state of the RuII complex.
The apparent intense absorption in the visible range (such
as with complex Ru-3) would not be funneled efficiently to
the emissive 3MLCT state if the 1IL energy level of the
donor (a light-harvesting antenna-localized singlet state, for
example, coumarin for Ru-3) was lower than that of the
1MLCT state of the acceptor. In that case, the energy level
3
molecules in triplet state A* annihilate into a molecule in
a singlet excited state (1A*), and another molecule in the
ground state (1A) [Eq. (1b)]. Fluorescence light is then emit-
ted from the singlet excited state (1A*; [Eq. (1c)]; see also
Scheme 3). This kind of upconversion can be achieved with
non-coherent excitation at very low power density, that is,
much less than 100 mWcmꢀ2, thereby making this method
particularly suitable for a wide range of applications.[24] Fur-
thermore, the modular feature of the TTA upconversion
makes the absorption and the emission wavelength of the
TTA system amenable to wavelength tuning by judicious se-
lection of donor (sensitizer)/acceptor (annihilator) pairs.
3
3
of the IL state may have often been lower than the MLCT
state. This role was used to rationalize many unsatisfactory
attempts at the preparation of light-harvesting arrays based
on RuII–polypyridine complexes and chromophores, such as
BODIPY.[11,22] Herein, we propose that, among the other
reasons, BODIPY cannot serve as an efficient light-harvest-
ing unit to induce intense phosphorescence for the RuII
1
arrays because the energy level of the IL state of BODIPY
3
3
1
3
*
*
ð1aÞ
ð1bÞ
ð1cÞ
Ru þDPA ! Ruþ DPA ðTTETÞ
(about 500 nm, 2.48 eV) is simply lower than that of the
1MLCT state (450 nm, 2.76 eV). Another energy-transfer
3
1
1
3
*
*
*
DPA þ DPA ! DPA þDPA ðTTAÞ
pathway, IL! IL, is most-probably frustrated because the
heavy-atom effect on the organic chromophore in this kind
of dyad is usually weak. This general rule will be useful for
the design of light-harvesting RuII arrays that show intense
absorption in the visible spectrum and, at the same time, in-
*
DPA ! DPAþhv ðfluorescence by emissionÞ
However, the development of TTA-based light upconver-
tensified MLCT emission. However, one should be careful
sion is still in its infancy. For example, the RuII complexes
used in the TTA up-conversion are the parent complexes,
which typically show poor absorption in the visible spec-
trum.[15,30] Therefore, we set out to investigate the effect of
light-harvesting in complexes Ru-2, Ru-3 and Ru-4 on the
efficiency of upconversion.
3
to apply this rule to complexes with large p-conjugated
frameworks.[9]
Application of the visible-light-harvesting arrays: Triplet–
triplet-annihilation-based upconversion: These new RuII-
based light-harvesting arrays showed intense absorption in
the visible spectrum and efficient population of the triplet
excited states. Thus, these complexes could be used for pho-
tochemical processes that involve triplet–triplet energy-
transfer. As a preliminary application of these visible-light-
harvesting RuII–polypyridine–coumarin arrays, we studied
the triplet–triplet annihilation (TTA) upconversion ability of
these complexes as triplet photosensitizers. Light upconver-
sion, which is the observation of an emission at a higher
energy (shorter wavelength) than the excitation energy
We selected 9,10-diphenylanthracene (DPA) as the ac-
ceptor/annihilator for the TTA experiments because the
triplet-state energy of DPA (1.77 eV, 700 nm) was lower
than the triplet energy level of the donor/sensitizer (for ex-
ample, 2.07 eV, 600 nm), which matched the requirements
for the TTA upconversion system.[15]
We observed blue fluorescence emission of DPA (at
400 nm and longer wavelength) with 473 nm laser excitation
of mixtures of Ru-2, Ru-3, and Ru-4 with DPA, respectively
(Figure 5).[15] The typical excitation power was 5 mW (power
Chem. Eur. J. 2012, 18, 4953 – 4964
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4959

J. Zhao et al.
must have been a triplet dark state that was populated by
the 473 nm laser excitation, and that it was this state by
which the TTA process was activated.[32a,c] Dark triplet
3
states, such as IL, are ideal as energy reservoirs for the trip-
let energy-transfer to the acceptor/annihilator.[32a,c,f,i,j] This
assumption is consistent with our photophysical studies
(Figure 1). Currently, we are actively working along this line
to devise transition-metal complexes with dark triplet excit-
ed states (i.e., non-luminescent, but the triplet state can be
populated by photoexcitation) as sensitizers to improve the
quantum yield of the TTA-based upconversion, because, in
this case, the phosphorescence emission of the triplet sensi-
tizer was actually competitive with the upconversion.
The enhancement in upconversion by light-harvesting was
demonstrated by comparing the results of complexes Ru-2
and Ru-4. Both complexes showed similar phosphorescence
quantum yields and lifetimes (Table 1). However, the upcon-
verted fluorescence emission intensity of complex Ru-2 was
about 3.0-fold larger than that of Ru-4; this ratio was in
agreement with the ratio of the molar extinction coefficients
of the two complexes at 473 nm. This result indicated that
the light-harvesting of Ru-2 was effective in producing the
triplet excited states of the complex.
Figure 5. Upconversion with the RuII complexes as sensitizers (in MeCN,
208C). a) Emission of upconverted DPA (4.0ꢁ10ꢀ5 m) and residual phos-
phorescence following selective excitation of the sensitizers (1.0ꢁ10ꢀ5 m)
at 473 nm (71 mWcmꢀ2). b) Emission of the complexes without DPA.
The upconversion was detectable by the naked eye
(Figure 6). Under these conditions, complex Ru-1 did not
density: 71 mWcmꢀ2), which was too low for simultaneous
two-photon absorption. The anti-Stokes shift of the upcon-
version was 0.48 eV. However, for Ru-1, no significant up-
converted fluorescence was observed under the experimen-
tal conditions (which was below the threshold for upconver-
sion with Ru-1). Excitation of DPA or the complexes alone
(473 nm laser) failed to produce any emission in the range
400–500 nm. Therefore, the fluorescence emission was the
result of TTA upconversion. Besides the fluorescence bands
in the range 400–500 nm, phosphorescence bands of the sen-
sitizer were simultaneously observed (centered at 600 nm),
which were due to the non-complete TTET (Figure 5a).[31]
Interestingly, although complex Ru-3 showed intense ab-
sorption that was similar to that of Ru-2, the upconversion
of Ru-3 was much weaker than Ru-2. This result demon-
strated that the light-harvesting effect of Ru-3 was not effec-
tive, that is, the energy of the absorbed light could not be ef-
ficiently transferred to a triplet excited state, either through
1
3
1
3
1IL! MLCT! MLCT or IL! IL pathways.
We measured the upconversion quantum yields (FUC) of
the complexes versus two independent standards: coumarin
6 (F=0.78 in EtOH) and 2,5-dibutyl-3,6-bis-phenyl-pyrrolo-
ACHTUNGTRENNUNG[3,4-c]pyrrole-1,4-dione (DPP; F=0.69 in DMSO). The ob-
served upconversion quantum yield of complex Ru-2 was
15.2%, which was about 16 times that of parent complex
Ru-1 and of benchmark complex [RuACHTUNTGRNENUG(dmb)3]ACHTUNTGERN[NUGN PF6]2 (dmb=
4,4’-dimethyl-2,2’-bipyridine).[15] We observed significantly
upconverted fluorescence with Ru-2/DPA, but little or no
upconverted fluorescence with [RuACHTUNTGRNENUG(dmb)3]ACHTUNTGRNEN[UGN PF6]2 or Ru-1.
Although the emission of Ru-3 was weaker than that of Ru-
4 (Figure 5b), the upconverted fluorescence of Ru-3/DPA
was significant. We believed that, with complex Ru-3, there
Figure 6. Photographs of the upconversion. Upconversion spectra of the
samples are shown in Figure 5. lex =473 nm (71 mWcmꢀ2), MeCN, 208C,
ACTHNUTRGNEUNG
c[sensitizers]=1.0ꢁ10ꢀ5 m, c[DPA]=4.0ꢁ10ꢀ5 m.
4960
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4953 – 4964

Ruthenium(II)–Polyimine–Coumarin Molecular Arrays
FULL PAPER
3
show significant upconversion. Therefore, the red phosphor-
escence of the complex was observed. For light-harvesting
array Ru-2, an intense purple color was observed in the
presence of DPA, owing to the efficient upconversion (fluo-
rescence was in blue region of the spectrum). For complexes
Ru-3 and Ru-4, upconversion were also observed but the ef-
ficiency was lower than that of Ru-2.
The efficiency of the TTET process was quantitatively
evaluated by quenching of the phosphorescence of the com-
plexes with DPA as the triplet quencher, that is, the emis-
sion of the complexes was reduced in the presence of DPA
(in this case, quenching of the Fçster energy-transfer was
impossible, owing to the non-overlapping absorption/emis-
sion spectra of the acceptor and the donor). Stern–Volmer
quenching curves were constructed with the quenching data
(Figure 7). Complex Ru-2 clearly gave the most-significant
light-harvesting would produce more MLCT excited states
(with complex Ru-2), whereas longer phosphorescence life-
times would enhance the TTET process (with complexes
Ru-2, Ru-3, and Ru-4), and both enhancements would im-
prove the TTA-based upconversion. The lifetime of complex
Ru-1 was below the threshold of the experimental condi-
tions (such as DPA concentration, laser power density, etc.).
Therefore, no significant upconverted fluorescence was ob-
served (for the upconversion of Ru-1 in the presence of
more-concentrated DPA, see the Supporting Information).
The higher upconversion efficiency of Ru-3 was due to the
3
non-emissive IL state, with which the TTA process could
also be initiated. The upconversion of Ru-4 was due to the
long-lived triplet excited state, with which the TTET and
TTA efficiency was improved. The upconversion of Ru-2
was even more efficient than that of Ru-4, owing to the en-
hanced light-harvesting effect of Ru-2 over Ru-4.
The principle photophysics of the TTA-based light up-
conversion of the arrays are summarized in Scheme 3, which
shows the effects of the light-harvesting of array Ru-2, the
extended phosphorescence lifetimes of array Ru-2, and the
3
populated non-emissive IL state of array Ru-3 on the TTA-
based upconversion efficiency. Two pathways are possible
1
3
for the light-harvesting effect: 1IL! MLCT! MLCT and
3
3
1IL! IL! MLCT.
Figure 7. Stern–Volmer plots generated from quenching of the emission
intensity of: [RuACHTUNGTRENNUNG =
(dmb)3]2+ (lex =460 nm), Ru-1 (lex =450 nm), Ru-2 (lex
450 nm), Ru-3 (lex =480 nm), and Ru-4 (lex =450 nm) measured as a func-
tion of DPA (triplet quencher) concentration in CH3CN, 208C.
quenching effect in the presence of DPA. The Stern–Volmer
quenching constant (Table 2) was 8.80ꢁ104 mꢀ1 and the bi-
molecular quenching constant was up to 2.32ꢁ1010 mꢀ1 sꢀ1,
which was close to the diffusion-controlled limit (109–
1010 mꢀ1 sꢀ1). For other complexes, lower quenching constants
were observed and the trend was in agreement with the life-
times of the complexes.
Scheme 3. Qualitative Jablonski diagram showing the sensitized TTA up-
conversion with RuII complexes as the triplet photosensitizers and DPA
as the triplet acceptor. The effect of the light-harvesting ability and the
phosphorescence lifetime of the RuII sensitizers on the efficiency of TTA
upconversion is also shown (Ru-2 shown as an example). E is energy, GS
Table 2. Parameters for the upconversion of the RuII complexes.
[b]
[c]
Ksv (ꢁ103) [mꢀ1
]
kq (ꢁ109) [mꢀ1 sꢀ1
]
FUC [%][d]
Ru-1
Ru-2
Ru-3
Ru-4
8.65
88.0
17.1
27.1
3.1
7.86
23.2
6.86
8.76
3.54
0.95
15.2
2.7
11.3
0.92
is the ground state (S0), 1IL* is the intra
ACTHNUTRGENNlUG iCAHTUNGTERNNUGgand singlet excited state (cou-
marin-localized), IC is internal conversion, 1MLCT* is the RuII based
metal-to-ligand charge-transfer singlet excited state, ISC is intersystem
crossing, 3MLCT* is the RuII-based metal-to-ligand-charge-transfer trip-
[RuACHTUNGTRENNUNG(dmb)3]
3
let excited state, IL* is the intra
li
N
[a] In deaerated CH3CN (1.0ꢁ10ꢀ5 m, 208C). [b] Stern–Volmer quenching
constant (KSV). [c] Bimolecular quenching constants (kq). [d] Upconver-
sion quantum yields (FUC).
calized). We proposed that the 3IL* and 3MLCT* of array Ru-2 were in
equilibrium, which prolonged the lifetime of the excited states. TTET is
triplet-triplet energy transfer, 3DPA* is the triplet excited state of DPA,
TTA is triplet–triplet annihilation, 1DPA* is the singlet excited state of
DPA. The emission bands observed for the sensitizers were the 3MLCT
emissive excited state (herein, the 3IL* was non-emissive). The emission
bands observed in the TTA experiment were the simultaneous 3MLCT*
emission (phosphorescence) and 1DPA* emission (fluorescence). The typ-
The upconversion efficiency was enhanced by improving
the light-harvesting ability and prolonging the phosphores-
cence lifetimes of the energy donors/sensitizers (Figure 5):
ical power density of the laser used in the upconversion was 71 mWcmꢀ2
.
Chem. Eur. J. 2012, 18, 4953 – 4964
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4961

J. Zhao et al.
product was purified by column chromatography on silica gel (CH2Cl2)
to give the light-yellow product (130.0 mg, yield: 79.4%). 1H NMR
(400 MHz, CDCl3): d=10.03 (s, 1H), 7.92 (m, 4H), 7.81 (s, 1H), 7.36 (d,
1H, J=9.0 Hz), 6.63 (d, 1H, J=8.5 Hz), 6.54 (s, 1H), 3.48–3.42 (m, 4H),
1.24 ppm (t, 6H, J=7.0 Hz).
Conclusion
We have reported the first effective visible-light-harvesting
RuII–polypyridine–coumarin array, which shows enhanced
absorption in the visible range and efficient energy-transfer
from the light-harvesting antenna to the coordination
center. The absorption of molecular array Ru-2 was in the
visible range (400–500 nm) and the molar extinction coeffi-
cient (e=6.33ꢁ104 mꢀ1 cmꢀ1 at 475 nm) was enhanced by 3.9-
fold compared to the parent complex (Ru-1). Previously, no
effective visible-light-harvesting RuII–chromophore had
been reported with efficient energy-transfer from the light-
harvesting antenna to the coordination center and without
loss in the acceptor phosphorescent quantum yield. We also
proposed a general rule to help the design of visible-light-
harvesting RuII–organic-chromophore arrays. This rule
General method for synthesis of the imidazole-linked phenanthroline li-
gands: Under an Ar atmosphere, 1,10-phenanthroline-5,6-dione
(100.0 mg, 0.48 mmol), the corresponding aldehyde (1.2 equiv), and am-
monium acetate (685.7 mg, 9.62 mmol) were dissolved in glacial acetic
acid (20 mL) and the mixture were heated to reflux for 6 h. After com-
pletion of the reaction, the mixture was cooled to RT and a majority of
the solvent was removed under reduced pressure. Next, a dilute solution
of NaOH was to adjust the pH value to about 7.0. A yellow precipitate
formed which was collected, washed with water, and dried under vacuum
overnight. The crude product was purified by column chromatography on
silica gel (CH2Cl2/MeOH, 10:1 v/v).
Compound L1: Compound L1 was obtained as a pale white precipitate
and purified by recrystallization (106.6 mg, 75.0%). 1H NMR (400 MHz,
[D4]MeOH): d=8.74 (d, 1H, J=3.9 Hz), 8.59–8.32 (m, 1H), 7.98 (d, 1H,
J=7.8 Hz), 7.48–7.41 (m, 3H), 4.87 ppm (m, 5H); 13C NMR (100 MHz,
[D4]MeOH): d=151.42, 147.42, 143.14, 129.77, 129.63, 128.81, 126.47,
123.10 ppm; HRMS (ESI): m/z calcd for [C19H12N4+H]+: 297.1140;
found: 297.1135.
Compound L2: Yellow solid (130.0 mg, 62.1%); 1H NMR (400 MHz,
CDCl3): d=9.16 (d, 2H, J=4.2 Hz), 8.99 (s, 1H), 8.78 (d, 2H, J=
5.8 Hz), 7.72–7.69 (m, 2H), 7.48 (d, 1H, J=8.0 Hz), 6.66–6.63 (m, 1H),
6.50 (d, 1H, J=2.3 Hz), 3.48–3.43 (m, 4H), 1.26 ppm (t, 6H, J=7.0 Hz);
13C NMR (100 MHz, CDCl3): d=162.1, 156.4, 151.7, 148.2, 146.8, 144.3,
141.5, 130.1, 129.4, 122.9, 110.0, 108.7, 107.7, 96.8, 45.1, 12.5 ppm; HRMS
(ESI): m/z calcd for [C26H21N5O2+H]+: 436.1774; found: 436.1767.
1
3
stated that the energy levels of the IL and IL state (intra
ACHTUNGERTNlNUNG i-
ACHTUNGTRENNUNG
close to the energy levels of the intrinsic S1 and T1 states of
the MLCT acceptor, otherwise the 1IL! MLCT energy-
1
transfer is inefficient (in some circumstances, the direct
1IL! IL transition is frustrated owing to a weak heavy-
3
atom effect). Thus, a limitation was imposed on the attempts
to shift the effective absorption of the RuII–organic-chromo-
phore arrays to the red-end of the spectrum. As a prelimina-
ry application of the light-harvesting RuII complex, triplet–
triplet annihilation upconversion was performed, in which
the complexes were used as triplet sensitizers, and the up-
conversion quantum yield was up to 15.2% (about 16.5
times that of the parent complexes, which did not show
a light-harvesting effect, under the same experimental con-
ditions). These results will be important for the design of ef-
ficient RuII–organic-chromophore light-harvesting arrays, of
which the Ru-2 array reported here is the first successful ex-
ample.
Compound L3: Yellow solid (70.0 mg, 32.0%); 1H NMR (400 MHz,
CDCl3): d=9.15 (d, 2H, J=4.2 Hz), 8.88–8.51 (m, 3H), 7.78–7.75 (m,
2H), 7.12 (s, 1H), 3.40–3.35 (m, 4H), 2.91 (t, 2H, J=5.9 Hz), 2.82 (t, 2H,
J=6.0 Hz), 2.04–2.01 ppm (m, 4H); HRMS (ESI): m/z calcd for
[C28H21N5O2+Na]+: 482.1593; found: 482.1586.
Compound L4: The crude product was washed several times with
CH2Cl2. 1H NMR (400 MHz, CDCl3): d=9.25 (m, 2H), 9.04 (m, 2H),
8.32ACTHNUTRGENNG(U m, 2H), 8.04–7.91ACHTUNGTNER(NUGN m, 4H), 7.46 (t, 2H, J=9.3 Hz), 6.74 (d, 1H, J=
8.0 Hz), 6.57 (d, 1H, J=8.2 Hz), 3.51–3.48 (m, 4H), 1.24 ppm (t, 6H, J=
6.8 Hz); HRMS (ESI): m/z calcd for [C32H25N5O2+H]+: 512.2087; found:
512.2090.
General method for the preparation of the RuII complexes: [RuCl2-
AHCTUNGERTG(NNUN cymene)] (0.5 equiv) and ligands L1–L4 (1.0 equiv) were dissolved in
EtOH (5.0 mL). The mixture was stirred at RT under a N2 atmosphere
for 2 h. The reaction was monitored by TLC. Next, a solution of 2,2’-bi-
pyridine (bpy, 2.0 equiv) in water (10 mL) was added and the mixture
was heated to reflux for 22 h. After cooling, the solution was concentrat-
ed under reduced pressure and treated with a saturated aqueous solution
of NH4PF6. A red precipitate formed which was collected by filtration.
The crude product was then purified by column chromatography on silica
gel (MeCN/water/saturated aqueous NaNO3, 100/9/1 v/v) and treated
again with a saturated aqueous solution of NH4PF6. A precipitate formed
which was collected, washed with water, and dried under vacuum.
Experimental Section
General analytical measurements: NMR spectra were recorded on
a Bruker 400 MHz spectrometer. CDCl3 or CDCl3/CD3OD were used as
the solvents and tetramethylsilane (TMS) was used as a reference (d=
0.00 ppm). High-resolution mass spectra (HRMS) were determined on
a LC/Q-TOF MS system (UK). Fluorescence spectra were measured on
a F4500 (Hitachi) or a CRT 970 spectrofluorometer. Fluorescence life-
times were measured with a Fluoromax-4 spectrofluorometer (Horiba
Jobin Yvon). Absorption spectra were recorded on a Perkin–Elmer-
Lambda-35 UV/Vis spectrophotometer. Compounds 2, 3, 5, 6, 7, 8, 9, and
10 were synthesized according to literature procedures.[33] The spin-densi-
ty surface of the RuII complexes were optimized in their triplet state geo-
metries at the B3LYP/6–31G(d)/LANL2DZ level with Gaussian 09W.[34]
1
Ru-1: Red solid (36.7 mg, 75.0%); H NMR (400 MHz, [D6]acetone): d=
13.38 (s, 1H), 9.18 (d, 1H, J=8.0 Hz), 9.07 (d, 1H, J=8.0 Hz), 8.86–8.80
(m, 4H), 8.37–8.32 (m, 4H), 8.26 (t, 2H, J=7.7 Hz), 8.24–8.12 (m, 4H),
7.96–7.95 (m, 4H), 7.65–7.59 (m, 5H), 7.40 ppm (m, 2H). 13C NMR
(100 MHz, [D6]acetone): d=158.3, 158.1, 153.8, 152.8, 152.6, 151.3, 151.1,
138.8, 138.8, 131.6, 131.4, 131.2, 131.1, 130.3, 130.1, 128.7, 128.6, 127.4,
Synthesis of compound 11: After degassing a mixture of EtOH (4.0 mL)/
tolueneACHTUNGTRENNUNG(4 mL)/water (8.0 mL), compound 10 (150.0 mg, 0.51 mmol), 4-
formylphenylboronic acid (153.0 mg, 1.02 mmol), and K2CO3 (211.0 mg,
1.53 mmol) were added, the flask was placed under vacuum and back-
filled with argon several times. Then, [PdACHTNUTRGNE(UNG PPh3)4] (30.0 mg, 0.026 mmol,
5.0 mol%) was added, the reaction mixture was heated at 808C for 8 h.
After being cooled to RT, the solvent was evaporated under reduced
pressure and the organic phase was extracted with CH2Cl2 and dried over
Na2SO4. The solvent was removed under reduced pressure and the crude
127.2, 126.9, 125.3, 125.2 ppm; HRMS (ESI): m/z calcd for [(M-2PF6)2+
/
2]: 355.0735; found: 355.0754.
1
Ru-2: Red solid (70.0 mg, 53.4%): H NMR (400 MHz, [D6]acetone): d=
12.81 (s, 1H), 9.47 (m, 1H), 9.16 (m, 1H), 9.03 (m, 1H), 8.88–8.82 (m,
4H), 8.38 (d, 1H, J=5.0 Hz), 8.34 (d, 1H, J=5.0 Hz), 8.27 (t, 2H, J=
8.0 Hz), 8.20–8.13 (m, 4H), 7.98–7.91 (m, 4H), 7.78–7.72 (m, 1H), 7.65 (t,
2H, J=6.5 Hz), 7.40 (t, 2H, J=6.5 Hz), 6.89 (m, 1H), 6.65 (m, 1H),
3.63–3.61 (m, 4H), 1.29–1.26 ppm (m, 6H); 13C NMR (100 MHz,
4962
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4953 – 4964

Ruthenium(II)–Polyimine–Coumarin Molecular Arrays
FULL PAPER
[D6]acetone): d=160.7, 157.5, 157.3, 157.1, 152.5, 152.1, 152.0, 150.5,
149.9, 149.7, 146.0, 145.4, 143.0, 138.1, 137.9, 137.3, 130.9, 130.8, 127.9,
127.8, 126.5, 126.2, 126.0, 125.9, 124.4, 124.4, 121.7, 110.3, 108.5, 107.2,
96.4, 44.7, 11.9 ppm; HRMS (ESI): m/z calcd for [(M-2PF6)2+/2]:
424.6057; found: 424.6074.
[2] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. V. Ze-
lewsky, Coord. Chem. Rev. 1988, 84, 85–277.
[3] M. R. Gill, J. Garcia-Lara, S. J. Foster, C. Smythe, G. Battaglia, J. A.
Thomas, Nat. Chem. 2009, 1, 662–667.
[4] W. Chen, F. N. Rein, R. C. Rocha, Angew. Chem. 2009, 121, 9852–
9855; Angew. Chem. Int. Ed. 2009, 48, 9672–9675.
1
Ru-3: Red solid (15.1 mg, 11.9%); H NMR (400 MHz, [D6]acetone) d=
[5] C. Y. Chen, J. G. Chen, S. J. Wu, J. Y. Li, C. G. Wu, K. C. Ho,
Angew. Chem. 2008, 120, 7452–7455; Angew. Chem. Int. Ed. 2008,
47, 7342–7345.
[6] L. Sun, M. Burkitt, M. Tamm, M. K. Raymond, M. Abrahamsson,
D. LeGourrierec, Y. Frapart, A. Magnuson, P. H. Kenez, P. Brandt,
A. Tran, L. Hammarstrçm, S. Styring, ꢂ. Bjçrn, J. Am. Chem. Soc.
1999, 121, 6834–6842.
[7] J. Gu, J. Chen, R. H. Schmehl, J. Am. Chem. Soc. 2010, 132, 7338–
7346.
[8] V. W. Manner, J. M. Mayer, J. Am. Chem. Soc. 2009, 131, 9874–
9875.
[9] a) S. Ji, W. Wu, W. Wu, P. Song, K. Han, Z. Wang, S. Liu, H. Guo, J.
Zhao, J. Mater. Chem. 2010, 20, 1953–1963; b) J. Zhao, S. Ji, W. Wu,
W. Wu, H. Guo, J. Sun, H. Sun, Y. Liu, Q. Li, L. Huang RSC Adv.
2012, 2, 1712–1728.
[10] F. Odobel, Zabri, Inorg. Chem. 2005, 44, 5600–5611.
[11] a) M. Galletta, S. Campagna, M. Quesada, G. Ulrich, R. Ziessel,
Chem. Commun. 2005, 4222–4224; b) G. Ulrich, R. Ziessel, A. Har-
riman, Angew. Chem. 2008, 120, 1202–1219; Angew. Chem. Int. Ed.
2008, 47, 1184–1201; c) A. A. Rachford, R. Ziessel, T. Bura, P. Re-
tailleau, F. N. Castellano, Inorg. Chem. 2010, 49, 3730–3736.
[12] G. J. Ryan, S. Quinn, T. Gunnlaugsson, Inorg. Chem. 2008, 47, 401–
403.
[13] a) D. S. Tyson, D. S. Tyson, C. R. Luman; X. Zhou, F. N. Castellano,
Inorg. Chem. 2001, 40, 4063–4071; X. Zhou, F. N. Castellano, Inorg.
Chem. 2001, 40, 4063–4071; b) V. Balzani, P. Ceroni, M. Maestri and
V. Vicinelli, Curr. Opin. Chem. Biol. 2003, 7, 657–665.
[14] S. Ji, H. Guo, X. Yuan, X. Li, H. Ding, P. Gao, C. Zhao, W. Wu, W.
Wu, J. Zhao, Org. Lett. 2010, 12, 2876–2879.
12.74 (s, 1H), 9.38 (d, 1H, J=8.3 Hz), 9.09 (d, 1H, J=8.1 Hz), 8.86–8.78
(m, 5H), 8.34–8.29 (m, 2H), 8.23 (t, 2H, J=8.0 Hz), 8.16–8.10 (m, 4H),
7.95–7.86 (m, 4H), 7.61 (t, 2H, J=6.1 Hz), 7.37 (t, 2H, J=6.8 Hz), 7.27
(s, 1H), 3.42–3.38 (m, 4H), 2.82–2.78 (m, 4H), 1.99–1.94 ppm (m, 4H).
13C NMR (100 MHz, [D6]acetone): d=161.7, 158.3, 158.1, 152.9, 152.9,
152.9, 151.2, 150.9, 150.7, 148.9, 146.8, 146.1, 143.9, 138.9, 138.8, 138.2,
131.9, 131.7, 131.6, 129.6, 128.7, 128.6, 127.6, 127.2, 126.9, 126.8, 126.8,
125.3, 125.2, 122.5, 120.9, 109.1, 106.5, 106.4, 50.8, 50.3, 21.8, 20.8,
20.7 ppm; HRMS (ESI): m/z calcd for [(M-2PF6)2+/2]: 436.6057; found:
436.6064.
Ru-4: Red solid (48.6 mg, 42.3%); 1H NMR (400 MHz, [D6]acetone):
d=9.07 (s, 2H), 8.85–8.82 (m, 4H), 8.36 (d, 2H, J=4.6 Hz), 8.31–8.24
(m, 4H), 8.20–8.14 (m, 5H), 8.01–7.92 (m, 6H), 7.64 (t, 2H, J=6.5 Hz),
7.55 (d, 1H, J=8.8 Hz), 7.42 (m, 2H), 6.80 (d, 1H, J=8.8 Hz), 6.54 (s,
1H), 3.58–3.53 (m, 4H), 1.24 ppm (t, 6H, J=8.0 Hz); 13C NMR
(100 MHz, [D6]acetone): d=161.2, 158.3, 158.1, 157.4, 153.7, 152.9, 152.9,
152.0, 151.2, 142.1, 139.2, 138.9, 138.8, 130.6, 129.5, 129.0, 128.7, 128.6,
127.1, 125.3, 125.2, 119.4, 110.2, 109.7, 97.2, 45.3, 12.7 ppm; HRMS (ESI):
m/z calcd for [(M-2PF6)2+/2]: 462.6213; found: 462.6208
Upconversion: The upconversion were carried out with two continuous-
wave diode-pumped solid-state lasers (DPSSLs) with outputs at 473 and
532 nm, respectively (the power was tunable in the range of 0–200 mW).
A reflecting lens was used to direct the laser beam into the sample cell
so that the laser beam was not directed toward the detector window of
the fluorospectrometer. A small black box was used to trap the laser
beam behind the sample cell to suppress the scattered laser; thus, no
notch filter was necessary. All the solutions were purged with Ar for
30 min before the measurements were taken. The typical laser power was
5 mW (the diameter of the laser spot was 3 mm; thus, the power density
of the laser was about 71 mWcmꢀ2). The upconversion quantum yields
were determined with two dyes: coumarin 6 (F=0.78 in EtOH) and 2,5-
[15] R. R. Islangulov, D. V. Kozlov, F. N. Castellano, Chem. Commun.
2005, 3776–3778.
[16] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd ed,
Kluwer Academic/Plenum Publishers, New York, 1999.
[17] N. D. McClenaghan, L. Yeydet, B. Mauber, M. T. Indelli, S. Cam-
pagna, Coord. Chem. Rev. 2005, 249, 1336–1350.
[18] D. S. Tyson, F. N. Castellano, Inorg. Chem. 1999, 38, 4382–4383.
[19] X. Y. Wang, A. D. Guerzo, R. H. Schmehl, J. Photochem. Photobiol.
C 2004, 5, 55–77.
[20] D. S. Tyson, F. N. Castellano, J. Phys. Chem. A 1999, 103, 10955–
10960.
dibutyl-3,6-bis-phenyl-pyrroloACHTUNTRGNEU[GN 3,4-c]pyrrole-1,4-dione (DPP; F=0.69 in
DMSO) as the quantum-yield standards, and the quantum yields were
calculated by using [Eq. (2)],[25b] where FUC, Aunk, Iunk, and hunk represent
the quantum yield, absorbance, integrated photoluminescence intensity,
and the refractive index of the samples, respectively. The upconversion
quantum yields determined with the two standards were basically the
same with relative errors permissible; the final value was the average of
the two.
[21] D. S. Tyson, K. B. Henbest, J. Bialecki, F. N. Castellano, J. Phys.
Chem. A 2001, 105, 8154–8161.
[22] F. Nastasi, F. Puntoriero, S. Campagna, S. Diring, R. Ziessel, Phys.
Chem. Chem. Phys. 2008, 10, 3982–3986.
Astd Iunk hunk
Aunk Istd hstd
2
ð2Þ
ꢀUC ¼ 2ꢀstd
ð
Þð Þð
Þ
[23] a) H. C. Chen, C. Y. Hung, K-H. Wang, H-L. Chen, W. S. Fann, F. C.
Chien, P. Chen, T. J. Chow, C. P. Hsu, S. S. Sun, Chem. Commun.
2009, 4064–4066; b) R. S. Khnayzer, J. Blumhoff, J. A. Harrington,
A. Haefele, F. Deng, F. N. Castellano, Chem. Commun. 2012, 48,
209–211.
Photographs of the upconversion were taken with a Samsung NV5 CCD
digital camera. The exposure time was 1 s (the default value of the digital
camera under the experimental conditions).
[24] S. Baluschev, V. Yakutkin, T. Miteva, Y. Avlasevich, S. Chernov, S.
Aleshchenkov, G. Nelles, A. Cheprakov, A. Yasuda, K. Mꢃllen, G.
Wegner, Angew. Chem. 2007, 119, 7837–7840; Angew. Chem. Int.
Ed. 2007, 46, 7693–7696.
Acknowledgements
We thank the NSFC (20634040, 20972024, and 21073028), the Fundamen-
tal Research Funds for the Central Universities (DUT10ZD212), the
Royal Society (UK) and the NSFC (21011130154), the Ministry of Edu-
cation (SRFDP-200801410004 and NCET-08–0077), the State Key Labo-
ratory of Fine Chemicals (KF0802), the Education Department of Liaon-
ing Province (2009T015), the PCSIRT (IRT0711), and Dalian University
of Technology (HaiTian Scholar funding for T.D.J.) for financial support.
[25] a) R. R. Islangulov, J. Lott, C. Weder, F. N. Castellano, J. Am. Chem.
Soc. 2007, 129, 12652–12653; b) T. N. Singh-Rachford, F. N. Castel-
lano, Coord. Chem. Rev. 2010, 254, 2560–2573.
[26] T. N. Singh-Rachford, A. Haefele, R. Ziessel, F. N. Castellano, J.
Am. Chem. Soc. 2008, 130, 16164–16165.
[27] D. Y. Kim, D. W. Song, N. Chopra, P. D. Somer, F. So, Adv. Mater.
2010, 22, 2260–2263.
[28] a) Y. Y. Cheng, B. Fꢃckel, T. Khoury, R. G. C. R. Clady, M. J. Y.
Tayebjee, N. J. Ekins-Daukes, M. J. Crossley, T. W. Schmidt, J. Phys.
Chem. Lett. 2010, 1, 1795–1799; b) Y. Yap Cheng, T. Khoury,
[1] K. Kalyanasundaram, Coord. Chem. Rev. 1982, 46, 159–244.
Chem. Eur. J. 2012, 18, 4953 – 4964
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4963

J. Zhao et al.
R. G. C. R. Clady, M. J. Y. Tayebjee, N. J. Ekins-Daukes, M. J. Cross-
ley, T. W. Schmidt, Phys. Chem. Chem. Phys. 2010, 12, 66–71.
[29] a) H. S. Mader, M. Link, D. E. Achatz, K. Uhlmann, X. Li, O. S.
Wolfbeis, Chem. Eur. J. 2010, 16, 5416–5424; b) A. Monguzzi, J.
Mezyk, F. Scotognella, R. Tubino, F. Meinardi, Phys. Rev. B 2008,
78, 195112–195111–195112–5; c) A. Monguzzi, R. Tubino, F. Mei-
nardi, Phys. Rev. B 2008, 77, 155122–1–15512–4.
2012, 18, 1961–1968; m) Y. Liu, Q. Li, J. Zhao, H. Guo, RSC Adv.
2012, 2, 1061–1067.
[33] a) D. Ray, P. K. Bharadwaj, Inorg. Chem. 2008, 47, 2252–2254; b) Y.
Sun, D. A. Lutterman, C. Turro, Inorg. Chem. 2008, 47, 6427–6434;
c) J. Van Gompel, G. B. Schuster, J. Org. Chem. 1987, 52, 1465–
1468; d) D. N. Lee, G. J. Kim, H. J. Kim, Tetrahedron Lett. 2009, 50,
4766–4768.
[30] T. N. Singh-Rachford, R. R. Islangulov, F. N. Castellano, J. Phys.
Chem. A 2008, 112, 3906–3910.
[31] T. N. Singh-Rachford, F. N. Castellano, J. Phys. Chem. Lett. 2010, 1,
195–200.
[34] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ꢄ.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc. Wallingford CT, 2009..
[32] a) J. Zhao, S. Ji, H. Guo, RSC Adv. 2011, 1, 937–950; b) S. Ji, W.
Wu, W. Wu, H. Guo, J. Zhao Angew. Chem. 2011, 123, 1664–1667;
Angew. Chem. Int. Ed. 2011, 50, 1626–1629; Angew. Chem. Int. Ed.
2011, 50, 1626–1629; c) S. Ji, H. Guo, W. Wu, W. Wu, J. Zhao
Angew. Chem. 2011, 123, 8433–8436; Angew. Chem. Int. Ed. 2011,
50, 8283–8286; d) W. Wu, W. Wu, S. Ji, H. Guo, J. Zhao, Dalton
Trans. 2011, 40, 5953–5963; e) H. Sun, H. Guo, W. Wu, X. Liu, J.
Zhao, Dalton Trans. 2011, 40, 7834–7841; f) Y. Liu, W. Wu, J. Zhao,
X. Zhang, H. Guo, Dalton Trans. 2011, 40, 9085–9089; g) W. Wu, J.
Sun, S. Ji, W. Wu, J. Zhao, H. Guo, Dalton Trans. 2011, 40, 11550–
11561; h) W. Wu, H. Guo, W. Wu, S. Ji, J. Zhao, J. Org. Chem. 2011,
76, 7056–7064; i) J. Sun, W. Wu, H. Guo, J. Zhao, Eur. J. Inorg.
Chem. 2011, 3165–3173; j) L. Huang, L. Zeng, H. Guo, W. Wu, W.
Wu, S. Ji, J. Zhao, Eur. J. Inorg. Chem. 2011, 4527–4533; k) W. Wu,
H. Guo, W. Wu, S. Ji, J. Zhao, Inorg. Chem. 2011, 50, 11446–11460;
l) W. Wu, H. Guo, J. Zhao, J. Sun, S. Ji, Z. Wang, Chem. Eur. J.
Received: May 5, 2011
Revised: January 19, 2012
Published online: March 8, 2012
4964
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4953 – 4964