Energy-transfer mechanism in photoluminescent terbium(III) complexes causing their temperature-dependence

Article abstract of DOI:10.1246/bcsj.80.1492

Photoluminescence of Terbium(III) complexes was investigated as a function of temperatures in the range of 80-280K for [Tb(bfa)₃(H ₂O)₂] (bfa: 4,4,4-trifluoro-l-phenyl-l,3-butanedionato), [Tb(hfa)₃(H₂O)₃] (hfa: hexafluoroacetyl- acetonato), [Tb(tfa)₃(H₂O)₂] (tfa: trifluoroacetylacetonato), [Tb(acac)₃(H₂O)₃] (acac: acetylacetonato), and [Tb(hfa)₃-(tppo)₂] (tppo: triphenylphosphine oxide). These complexes were classified into the two groups with different temperature-dependences. The first group consisting of [Tb(bfa)₃(H₂O)₂], [Tb(tfa)₃(H ₂O)₂], and [Tb(acac)₃(H₂O) ₃] showed a dependence determined by the energy gap between the excited triplet state of the ligand and the emitting level of ter-bium(III) ion. In contrast, for [Tb(hfa)₃(H₂O)₃] and [Tb(hfa)₃(tppo)₂] containing hfa as a ligand, not only the energy gap but also the energy barriers of the Forward energy transfer from the ligand to terbium(III) ion and Back energy transfer from terbium(III) ion to the ligand were taken into account for understanding their dependences. These results are discussed based on the re-orientation of the complexes accompanied by the forward and back energy transfer processes using DFT calculations.

Full text of DOI:10.1246/bcsj.80.1492

1492 Bull. Chem. Soc. Jpn. Vol. 80, No. 8, 1492–1503 (2007)
ꢀ 2007 The Chemical Society of Japan
Energy-Transfer Mechanism in Photoluminescent Terbium(III)
Complexes Causing Their Temperature-Dependence
ꢀ1
ꢀ1;3
Shinya Katagiri,1 Yasunori Tsukahara, Yasuchika Hasegawa,2 and Yuji Wada
1Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka 565-0871
2Graduate School of Material Science, Nara Institute of Science and Technology,
8916-5 Takayama-cho, Ikoma 630-0192
3Department of Applied Chemistry, Graduate School of Natural Science and Technology, Okayama University,
3-1-1 Tsushima-Naka, Okayama 700-8530
Received October 27, 2006; E-mail: yuji-w@cc.okayama-u.ac.jp
Photoluminescence of Terbium(III) complexes was investigated as a function of temperatures in the range of 80–
280 K for [Tb(bfa)3(H2O)2] (bfa: 4,4,4-trifluoro-1-phenyl-1,3-butanedionato), [Tb(hfa)3(H2O)3] (hfa: hexafluoroacetyl-
acetonato), [Tb(tfa)3(H2O)2] (tfa: trifluoroacetylacetonato), [Tb(acac)3(H2O)3] (acac: acetylacetonato), and [Tb(hfa)3-
(tppo)2] (tppo: triphenylphosphine oxide). These complexes were classified into the two groups with different tempera-
ture-dependences. The first group consisting of [Tb(bfa)3(H2O)2], [Tb(tfa)3(H2O)2], and [Tb(acac)3(H2O)3] showed a
dependence determined by the energy gap between the excited triplet state of the ligand and the emitting level of ter-
bium(III) ion. In contrast, for [Tb(hfa)3(H2O)3] and [Tb(hfa)3(tppo)2] containing hfa as a ligand, not only the energy gap
but also the energy barriers of the ‘‘Forward energy transfer’’ from the ligand to terbium(III) ion and ‘‘Back energy trans-
fer’’ from terbium(III) ion to the ligand were taken into account for understanding their dependences. These results are
discussed based on the re-orientation of the complexes accompanied by the forward and back energy transfer processes
using DFT calculations.
Lanthanide ions have been extensively investigated due to
their superior emission properties.1–6 Their special emission
properties are attributed to optical transitions within the f
manifold. The f-electrons are well shielded by the 5s, 5p-orbit-
als from the chemical environments and therefore retain most
of their atomic characters. As a consequence, the spectral po-
sitions of the emission lines are almost independent of the sur-
roundings, and the emission spectra as well as the absorption
spectra consist of extremely sharp lines. Thus, lanthanide ions
have been applied as the luminescent materials. However, f–f
transitions are forbidden by odd parity and many of them are
also spin forbidden.7 So, lanthanide ions have the weak ab-
sorption. Therefore, many studies have focused on photosensi-
tized emission of lanthanide complexes in order to overcome
the weak absorption due to the forbidden f–f transition. Candi-
dates for the photosensitizer are organic molecules and in-
organic compounds. Lanthanide complexes having organic
ligands have been selected as a photosensitizer, and a zeolite
system encapsulating a photosensitizer and lanthanide ions
has been also reported as a photosensitization system showing
special rainbow colored-emission.8 Luminescent lanthanide
materials are applied for a display and lighting apparatus.
Furthermore, europium chalcogenides have been examined
as semiconductors, since they are expected for practical appli-
cations as recording mediums, LASER devices and photoiso-
lators.9,10
ment engineering, and energy technology11 so that two-dimen-
sional or multi-dimensional mapping of temperatures and pres-
sures in real time can be achieved with no contact measure-
ments. The first luminescent dyes having temperature-sensing
property have been reported by Demas and DeGraff in
1988.12 rhenium complexes,13,14 ruthenium complexes,15 plati-
num complexes,16 and metal–porphyrins have also been em-
ployed as the temperature-sensing probe molecules.17
Lanthanide complexes are suitable as non-contact sensing
molecules, because of the strong and sharp emission with a
long lifetime.18 However, normally the f–f electrons are well
shielded by the 5s, 5p-orbitals from the chemical environment,
so the lanthanide compounds are less affected by their environ-
ments. Therefore, a strategy for making the temperature sensi-
tivity is required, and the chemical structures of ligands play
important roles in the strategy.
Amao and co-workers have reported the first temperature-
sensitive dye using europium(III) complexes in 2003.19 They
have prepared a thin film containing (1,10-phenanthroline)tris-
[4,4,4-trifluoro-1-(2-thienyl)-1,3-butanedionato]europium(III)
([Eu(tta)3(phen)]) by the Langmuir–Blodgett technique com-
posed of poly(N-dodecylacrylamide) (pDDA). The tempera-
ture-sensitivity of the europium(III) complex in the polymer
film has been reported to be 0.78% ꢁCꢂ1. Recently, Khalil
et al. also have prepared the high-performance europium(III)
complex for temperature sensitive paint (TSP) in 2004 (tem-
perature-sensitivity: 4.42% ꢁCꢂ1).20 Khalil et al. have used
thenoyltrifluroacetone, 1,1,1-trimethyl-5,5,6,6,7,7,7-heptafluoro-
2,4-heptanedione, 1-phenanthren-3-yl-3-phenanthren-9-yl-pro-
Functional probe molecules to measure temperatures/pres-
sures of material surfaces with no-contact are desired for
studies on fluid dynamics, aeronautical engineering, environ-
Published on the web August 10, 2007; doi:10.1246/bcsj.80.1492

S. Katagiri et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007) 1493
becomes larger when the excited triplet state of the ligand is
closer to the emitting level of terbium(III) ion, resulting in
the increase of the emission quantum yield. However, the
EG value that gives the maximum emission quantum yield is
not zero but has a certain value. Contribution of BET becomes
more, leading to a decrease in the emission quantum yield,
when the excited triplet state of the ligand is closer to the emit-
ting level of terbium(III) ion, if BET is considered. Therefore,
a suitable EG value giving the maximum emission quantum
yield should be determined by both EG and contribution of
BET. Furthermore, due to the temperature-dependence of BET
caused by an energy barrier present in this process, the suitable
EG value giving the maximum emission quantum yield should
be changed with the temperature change. Since the emission
quantum yield increases with decreasing temperatures due to
the less contribution of BET at low temperatures, the maxi-
mum value of the emission quantum yield should be observed
at a certain value of EG, which should change with respect to
temperature.
Intersystem Crossing (ISC)
S1
Back Energy Transfer (BET)
T1
5D4
Energy Gap
(EG)
Vibrational Excitation
Forward Energy
Transfer (FET)
Nonradiative Transition
7FJ
S0
π
-
π
∗ transition of the ligands
f-f transition of Tb(III)
Scheme 1. Energy diagram of terbium(III) complex showing
the energy-transfer processes.
pane-1,3-dione, and 4,4,5,5,6,6,6-heptafluoro-1-phenanthren-
3-ylhexane-1,3-dione as the ligands. They have measured
the decay lifetimes of the emission as a function of tempera-
tures from the three view points: (i) the type and number of
ligands in the complex, (ii) the chemical structures of the ma-
trix, and (iii) the concentrations of the europium complexes in
the polymer matrix. The origin of the temperature sensitivity
in these works is ascribed to changes in the vibrational excita-
tion depending on the temperatures.
In the present work, we investigated five terbium(III) com-
plexes with different ligands and show the different tempera-
ture-dependent changes in the photosensitized emission pro-
perties from the ultracold temperature of 80 to 280 K. We ob-
served the emission spectra and the emission lifetimes and ob-
tained the branching ratios of the five complexes at various
temperatures, and we discuss the changes of the symmetry of
the complex structures depending on the temperatures in rela-
tion to the transition probability of each emission bands of
We have proposed a terbium(III) complex as a temperature-
sensing probe functioning on a mechanistic base of the back
energy transfer from the emitting level of terbium(III) ion to
the excited triplet state of the ligand (abbreviated as BET) in
2004.21 The mechanism is schematically shown in Scheme 1.
The photosensitized emission is initiated by the excitation of
the ground state of the ligand containing a photosensitizer to
7
terbium(III) ion (5D4 ! FJ: J ¼ 0; 1; 2; 3; 4; 5; 6). The radia-
tive rate could be compared among the complexes from the
emission lifetimes at the ultracold temperature (80 K), because
BET was fully suppressed at ultracold temperature. In addi-
tion, we proposed another factor causing the temperature sen-
sitivity in addition to BET, which is ‘‘forward energy transfer,’’
or the energy transfer from the excited triplet ligand to ter-
bium(III) ion. We carried out the quantum chemical calcula-
tions of the ligand in the ground state and in the excited triplet
state and clarified the origin of the energy barriers present in
the forward and back energy transfer processes. Thus, we can
fully explain the temperature-dependence of the emission
properties of the complexes systematically, and this will con-
tribute to the design of the lanthanide complexes for the con-
trolled temperature-dependent emission.
ꢀ
its excited singlet state (ꢀ–ꢀ transition), followed by inter-
system crossing to the excited triplet state. Then, the excitation
energy of the excited triplet state is transferred to the lan-
thanide ions, resulting in the emission. The right side of
Scheme 1 shows the 4f orbital of terbium(III) ion. The most
important step in this process should be the energy transfer be-
tween the excited triplet state (T1) and the emitting level of
terbium(III) ion (5D4).22,23 In general, the energy transfer be-
tween the excited triplet state of the ligand and the emitting
level of Tb ion is described by the Dexter mechanism.24 The
excited energy of terbium(III) ion is dissipated in the two pro-
cesses; the radiative transition and the nonradiative transition.
The nonradiative transition consists of the vibrational excita-
tion, BET and the concentration quenching.20
Experimental
Materials.
Terbium(III) acetate tetrahydrate (99.9%),
terbium(III) chloride hexahydrate, 1,1,1,5,5,5-hexafluoro-2,4-pen-
tanedione (HFA), triphenylphosphine oxide (TPPO), and 4,4,4-
trifluoro-1-phenyl-1,3-butanedione (BFA) were purchased from
Wako Pure Chemical Industries Ltd. Isopentane, diethyl ether,
and ethanol were obtained from NACALAI TESQUE, INC. Tri-
fluoroacetylacetone (TFA) and acetylacetone (acac) were pur-
chased from TOKYO KASEI KOGYO CO., LTD. They were
used for the experiments as supplied.
Apparatus. Infrared spectra used to identify the synthesized
materials were obtained with a Perkin-Elmer FT-IR 2000 spectro-
meter. Elemental analyses were performed with a Perkin-Elmer
240C. 19F NMR data were obtained with a JEOL EX-270 spectro-
meter. 19F NMR chemical shifts were determined using hexa-
fluorobenzene as an external standard (ꢂ ꢂ162:0 (s, Ar–F)).
Temperature-dependence is expected for the vibrational ex-
5
citation and BET in the nonradiative transition from the D4
level of terbium(III) ion. However, most of the works using
the temperature-sensing molecules have been carried out in the
temperature range of 273–330 K. We need to understand the
relation of BET and the energy gap (EG) between the excited
triplet state of the ligand and the emitting level of terbium(III)
ion by investigating the emission properties in a wider range
including ultracold temperatures. The systematic measure-
ments of the emission quantum yields of the lanthanide com-
plexes having ꢁ-diketone as a ligand from ultracold tempera-
ture (77 K) to 300 K have been carried out in 1970 by Sato
and Wada.25 They have reported that the energy-transfer rate

1494 Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007)
Energy Transfer in Terbium(III) Complexes
Preparation of Terbium(III) Complexes. Preparation of
Triaquatris(hexafluoroacetylacetonato)terbium(III) ([Tb(hfa)3-
(H2O)3]): Terbium(III) acetate tetrahydrate (5.0 g, 12.3 mmol)
was dissolved in distilled water (20 mL) by stirring. 1,1,1,5,5,5-
Hexafluoro-2,4-pentanedione (7 g, 33.6 mmol) was added drop-
wise to the above solution. The mixture produced a precipitation
of white green powder after stirring for 3 h. The reaction mixture
was filtered. The resulting white green needle crystals were
recrystallized from methanol/water.7,26 Yield: 70%. IR(KBr):
1
3
2
CF3
CF3
CF3
H
H
H
H
O
O
O
O
O
Tb
Tb
O
Tb
3
2
2
3
3
4
CH3
CF3
H
H
O
H
H
O
1650 (st, C=O), 1255–1141 (st, C–F) cmꢂ1 19F NMR (CD3-
.
O
O
Tb
O
O
3
COCD3) ꢂ ꢂ61:22 (s, CF3). Anal. Calcd for C15H9F18O9Tb: C,
21.60; H, 1.09%. Found: C, 21.47; H, 1.34%.
CH3
CH3
3
3
Preparation of Tris(hexafluoroacetylacetonato)bis(triphen-
ylphosphine oxide)terbium(III) ([Tb(hfa)3(tppo)2]): Metha-
nol (50 mL) containing [Tb(hfa)3(H2O)2] (1.64 g, 2.04 mmol)
and triphenylphosphine oxide (TPPO) (1.22 g, 4.38 mmol) was re-
fluxed under stirring for 12 h. The reaction mixture was concen-
trated using a rotary evaporator. Reprecipitation by addition of
excess hexane produced crude crystals, which were washed in
toluene several times. Recrystallization from hot methanol gave
green translucent crystals of [Tb(hfa)3(tppo)2].9 Yield: 56%. Anal.
Calcd for C51H33F18O8P2Tb: C, 45.83; H, 2.49%. Found: C,
45.58; H, 2.49%.
5
CF3
CF3
O
P O
Tb
O
2
3
Fig. 1. Chemical structures of 1 = [Tb(bfa)3(H2O)2], 2 =
[Tb(hfa)3(H2O)3], 3 = [Tb(tfa)3(H2O)2], 4 = [Tb(acac)3-
(H2O)3], and 5 = [Tb(hfa)3(tppo)2].
Table 1. Triplet State Energies of Gd ꢁ-Diketonates
Preparation of Tris(acetylacetonato)triaquaterbium(III)
([Tb(acac)3(H2O)3]): Terbium(III) acetate tetrahydrate (5.0 g,
12.3 mmol) was dissolved in distilled water (20 mL) by stirring.
2,4-Pentanedione (3.74 g, 37.4 mmol) was added dropwise to the
above solution. The pH value of this solution was adjusted at
pH 7 by adding NH3 aqueous solution. The mixture produced a
white precipitate after stirring for 3 h. The reaction mixture was
filtered. The resulting white needle crystals were recrystallized
from methanol/water.27 Yield: 50%. Anal. Calcd for C15H27O9-
Tb: C, 35.31; H, 5.33%. Found: C, 35.66, H, 5.23%.
Preparation of Diaquatris(benzoyltrifluoroacetylacetonato)-
terbium(III) ([Tb(bfa)3(H2O)2]): To ethanol (15 mL) were add-
ed 4,4,4-trifluoro-1-phenyl-1,3-butanedione (BFA) (0.65 mg, 3.0
mmol), and then sodium hydroxide (1 M, 3 mL). The mixture was
heated to 60 ꢁC and stirred. This solution was dropped into water
(10 mL) containing terbium(III) chloride hexahydrate (0.39 g, 1.04
mmol). Ethanol in the mixture was removed by evaporation after
stirring for 3 h. White-yellow powder was obtained. The reaction
mixture was filtered, and the powder was purified with ethanol/
water.27 Yield: 34%. Anal. Calcd for C30H22F9O8Tb: C, 42.87;
H, 2.64%. Found: C, 42.73, H, 2.62%.
Preparation of Diaquatris(trifluoroacetylacetonato)ter-
bium(III) ([Tb(tfa)3(H2O)2]): Terbium(III) acetate tetrahydrate
(3.78 g, 8.3 mmol) was dissolved in distilled water (30 mL) by stir-
ring. Trifluoroacetylacetone (4.07 g, 26.4 mmol) was added drop-
wise to the above solution. The mixture was heated to 60 ꢁC
and stirred. The mixture produced a white green precipitate after
stirring for 5 h. The reaction mixture was filtered. The resulting
white needle crystals were recrystallized in methanol/water.7,26
Yield: 64%. Anal. Calcd for C15H16F9O8Tb: C, 27.54; H, 2.47%.
Found: C, 27.44; H, 2.40%.
Optical Measurements. The emission spectra were measured
on a SPEX-Fluorolog ꢃ-3 (JOBIN YVON) system equipped with
low-temperature options. Low temperatures were achieved by
means of a nitrogen bath cryostat (Oxford Instruments, Optistat
DN) and a temperature controller (Oxford Instruments, ITC 502S)
(80–300 K). The excitation spectra were measured by HITACHI
F-4500 spectrophotometer. The absorption spectra were measured
by a JASCO V-570 spectrometer. Decay profiles monitored at 543
Complex
The triplet state energy/cmꢂ1
[Gd(acac)3(H2O)3]
[Gd(tfa)3(H2O)2]
[Gd(hfa)3(H2O)3]
[Gd(bfa)3(H2O)2]
24800
22800a)
22200
21400a)
a) The value from literature.25
nm were recorded with a photomultiplier coupled to a Tektronix
TDS3052 oscilloscope upon excitation with the third harmonic
of a Nd:YAG laser (355 nm).9 Each optical measurement was car-
ried out in EPA (ethanol:isopentane:ether = 2:5:5) in the temper-
ature range of 80–280 K using a 1 cm rectangular quartz cell.
Results
Photosensitized Luminescence.
The structures of the
terbium(III) complexes employed in this work are shown in
Fig. 1. We selected ꢁ-diketone as a common structure in the
ligands, because we were able to control the properties of
the complexes by modifying the substituents at both ends of
the ligands (4 modified ligands). In Table 1 are listed the levels
of the excited triplet states of the 4 ligands determined by
measuring the phosphorescence of gadolinium(III) complexes
of each ligand.25 In addition, we synthesized complex 5 by
substituting the three coordinating water molecules of complex
2 with triphenylphosphine oxide (TPPO) in order to investigate
the effect of the structure of the second ligand on the emission
properties. When the EG (expressed as ꢀ) between the excited
triplet state of the ligand and the emitting level of terbium(III)
ion are less than ꢀ ¼ 1850 cmꢂ1, BET becomes appreciable at
room temperature.28 hfa of complex 2 has an EG of ꢀ ¼
1700 cmꢂ1, so BET should be important in the emission. We
selected bfa (ꢀ ¼ 900 cmꢂ1) having EG smaller than hfa’s,
tfa (ꢀ ¼ 2300 cmꢂ1) having EG a little larger than ꢀ ¼
1850 cmꢂ1, and acac (ꢀ ¼ 4300 cmꢂ1) having a large EG.
The emission spectrum of complex 3 in EPA at 300 K
(Fig. 2) was a typical one of a terbium(III) ion. The emission

S. Katagiri et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007) 1495
had a shoulder at 340 nm.
The absorption spectra of the five terbium(III) complexes at
bands were observed at 490, 545, and 583 nm, attributed to
the f–f transitions 5D4 ! F6, 5D4 ! F5, and 5D4 ! 7F4,
respectively. The other complexes were also found to have
the typical emission spectrum of a terbium(III) ion.
7
7
ꢀ
300 K are shown in Fig. 4. The ꢀ–ꢀ transitions of the four
ligands were observed. Therein, the rising edges of the spectra
were located at 375, 360, 340, 330, and 360 nm for complexes
1, 2, 3, 4, and 5, respectively. It was concluded from the coin-
cidence between the absorption and the excitation spectra that
The excitation spectra of the five terbium(III) complexes at
300 K are shown in Fig. 3. The spectra were monitored at
543 nm, which is attributed to the f–f transition (5D4 ! F5)
7
ꢀ
of a terbium(III) ion. The peaks of the excitation bands of
complexes 1, 2, 3, and 4 were observed at 365, 336, 325,
and 315 nm, respectively. The excitation band of complex 5
the emission of terbium(III) ion is induced by the ꢀ–ꢀ ab-
sorption of the ligands. Therefore, energy transfer from the ex-
cited state of the ligand to terbium(III) ion leading to the pho-
tosensitized emission of terbium(III) ion has been confirmed.
Temperature-Dependent Photosensitized Luminescence.
The emission spectra of the five terbium(III) complexes were
measured over a wide range of temperatures (80–280 K), as
shown in Fig. 5. The emission intensities normalized by the
intensities at the maxima obtained at 80 K for each complex
are plotted against the temperatures in Fig. 6. We abandoned
the comparison of the absolute intensities, because we needed
to change the slits of the optical measurement set-up for each
terbium(III) complex due to widely varied emission intensities
of the complexes.
The temperature-dependence of the emission spectrum of
complex 1 in EPA is shown in Fig. 5a. The excited triplet state
of the bfa was closest to the 5D4 level of terbium(III) ion
among the five complexes employed in this work (EG of com-
plex 1 was ꢀ ¼ 900 cmꢂ1). Therefore, the emission of com-
plex 1 should be affected by BET. Actually the emission spec-
trum of complex 1 was very weak at 180–280 K due to the
energy dissipation through BET. The emission intensities of
480
520
560
600
Wavelength / nm
Fig. 2. Emission spectrum of 3 in EPA (ex. at 330 nm) at
300 K (1.02 mM).
a
b
c
250 300 350 400 450
Wavelength / nm
250 300 350 400 450
Wavelength / nm
250 300 350 400 450
Wavelength / nm
d
e
250 300 350 400 450
Wavelength / nm
250 300 350 400 450
Wavelength / nm
Fig. 3. Excitation spectra of a) 1 (0.014 mM), b) 2 (0.0084 mM), c) 3 (0.011 mM), d) 4 (0.022 mM), e) 5 (0.0067 mM) in EPA
7
at 300 K monitored at 543 nm (5D4 ! F5).

1496 Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007)
Energy Transfer in Terbium(III) Complexes
0.6
0.8
0.6
0.4
0.2
0.0
a
b
c
0.6
0.4
0.2
0.0
0.4
0.2
0.0
250 300
350 400 450
Wavelength / nm
250 300 350 400 450
250 300 350 400 450
Wavelength / nm
Wavelength / nm
d
e
0.2
0.4
0.2
0.0
0.1
0.0
250 300 350 400 450
250 300 350 400 450
Wavelength / nm
Wavelength / nm
Fig. 4. Absorption spectra of a) 1 (0.014 mM), b) 2 (0.0084 mM), c) 3 (0.011 mM), d) 4 (0.022 mM), e) 5 (0.0067 mM)
in EPA at 300 K.
complex 1 dramatically increased with a decrease in the tem-
perature below 160 K (Fig. 6a). This can be understood as a
result of the suppression of BET below 160 K.
the suppression of BET. However, BET should not be so domi-
nant when EG of ꢀ ¼ 2300 cmꢂ1 was considered. The critical
value for the occurrence of BET proposed by Latva et al.
seems to vary depending on the structures of the complexes.28
The temperature-dependence of the emission spectrum of
complex 4 in EPA is shown in Fig. 5d. The spectra were noisy
due to weak emission intensities. EG of complex 4 was ꢀ ¼
4300 cmꢂ1. Therefore, the emission intensity of complex
4 should not be affected by BET. The emission intensity of
complex 4 mildly increased with a decrease in temperature.
(Fig. 6d) The temperature-dependence of the emission intensi-
ty of complex 4 can be understood as a result of the suppres-
sion of the vibrational excitation.29,30
In summary, the temperature-dependences of the emission
intensities of complex 1 and 3 showed drastic effect of BET,
i.e., the changes in the emission intensities dramatically in-
creased upon suppression of BET. The results obtained here
have led to an important conclusion that the suppression of
BET contributes more effectively to the temperature-depen-
dence of the emission intensity of terbium(III) complexes than
that of the vibrational excitation.
The emission spectra of complex 2 in EPA depending on
temperature are shown in Fig. 5b. BET can occur at room tem-
perature, when it is taken into account that EG of complex 2 is
ꢀ ¼ 1700 cmꢂ1. The emission intensity of complex 2 increas-
ed in the temperature range from 280 to 240 K and then de-
creased in the temperature range from 240 to 160 K. We called
the change the volcano-shaped. The top of the volcano-shape
was located at 240 K (Fig. 6b). The volcano-shape dependence
in the emission intensities on the temperatures were observed
only for complexes 2 and 5 having hfa as a ligand, and not ob-
served for the complexes 1, 3, and 4. The emission intensities
of complex 2 dramatically increased with a decrease in the
temperatures below 160 K.
The emission spectra of complex 5 in EPA depending on
temperature are shown in Fig. 5e. The change in the emission
intensity of complex 5 was also showed the volcano-shaped in
the temperature range from 160 to 280 K (Fig. 6e). The tem-
perature-dependences of the complexes 2 and 5 having hfa
showed the volcano-shape, as shown in Figs. 6b and 6e.
The temperature-dependence of the emission spectrum of
complex 3 in EPA is shown in Fig. 5c. The emission intensi-
ties of complex 3 dramatically increased with a decrease in
the temperature from 280 to 200 K. The changes of the emis-
sion intensity of complex 3 became less below 200 K (Fig. 6c)
than above the temperature. The increase observed in the re-
gion from 280 to 200 K might be understood as a result of
Lifetime Measurement. We measured the decays of the
emission of the five terbium(III) complexes. Time profiles of
7
the 5D4 ! F5 emission (543 nm) of terbium(III) ion were ob-
served, as shown in Fig. 7. The linear decays in the plot of
logarithm of the emission intensity against the time were
obtained for the complexes giving strong emission, but the
fluctuation could not be removed for those giving weak emis-
sion. Thus, we used only the initial stages of the profiles for

S. Katagiri et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007) 1497
a
b
200
200
100
100
Temperature / K
Temperature / K
480 520 560 600
Wavelength / nm
500
550
600
650
Wavelength / nm
c
d
250
200
250
200
150
100
150
100
Temperature / K
Temperature / K
480 520 560 600
Wavelength / nm
480
520
560
Wavelength / nm
e
200
100
Temperature / K
480 520 560 600
Wavelength / nm
Fig. 5. Temperature-dependence of the emission spectra of a) 1 (Ex. at 365 nm, slit 5 nm for excitation/4 nm for emission),
b) 2 (Ex. at 336 nm, slit 5 nm for excitation/5 nm for emission), c) 3 (Ex. at 330 nm, slit 7 nm for excitation/4 nm for emission),
d) 4 (Ex. at 315 nm, slit 15 nm for excitation/15 nm for emission), and e) 5 (Ex. at 336 nm, slit 5 nm for excitation/5 nm for
emission, in EPA). The concentrations were 15.5 mM for 1, 16.8 mM for 2, 18.3 mM for 3, 13.7 mM for 4, 16.5 mM for 5.
obtaining the lifetimes by fitting to the linear dependence. The
emission lifetimes of the terbium(III) complexes in EPA are
plotted against the temperatures in Fig. 8.
region. The emission lifetime of complex 5 steadily increased
with a decrease in the temperature.
The temperature-dependence of the lifetime of the emission
of the terbium(III) complexes can be understood by the contri-
bution of the vibrational excitation and BET depending on
temperatures. The temperature-dependence of BET should be
The emission lifetimes of the five terbium(III) complexes at
280 K were spread over a wide range but were similar (around
600–900 ms) at 80 K, at which BET and vibrational excitation
should be suppressed. The emission lifetime of complex 2 was
largest at 80 K among the five complexes. The emission life-
time of complex 1 was very short without an appreciable
change in the temperature range of 180–280 K and greatly
increased below 180 K, showing a steady value below 125 K
(650 ms). The emission lifetime of complex 2 was very short
and did not change in the temperature range of 180–280 K,
in contrast to complex 1. The emission lifetime of complex
2 greatly increased below 150 K. The emission lifetime
of complex 3 increased with a decrease in the temperature
between 220 and 280 K and slightly changed below 200 K.
The emission lifetime of complex 4 did not show a large
change and stayed near 800 ms over the whole temperature
5
related to the energy gap from the D4 level of TbIII to the
excited triplet state of the ligand. Therefore, the difference
observed for the temperature-dependence of the emission life-
times of the five complexes can be attributed to the different
EG values.
Energy transfer from T1 to the 5D4 level is so fast that
we could not observe the gradual rise of emission in Fig. 7.
Therefore, the emission lifetime (ꢃ) is given by the following
equation:
1
ꢃ ¼
;
ð1Þ
kr þ knr[Tb] þ kv þ kBET
where kr, knr[Tb], kv, and kBET are the rate constants of

1498 Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007)
Energy Transfer in Terbium(III) Complexes
1.0
1.0
0.5
0.0
1.0
a
b
c
0.5
0.0
0.5
0.0
100 150 200 250
100 150 200 250
Temperature / K
100 150 200 250
Temperature / K
Temperature / K
1.0
0.5
0.0
1.0
d
e
0.5
0.0
100 150 200 250
100 150 200 250
Temperature / K
Temperature / K
Fig. 6. Temperature-dependence of the emission intensity at 543 nm of a) 1, b) 2, c) 3, d) 4, and e) 5 in EPA. The emission
intensities were normalized by the intensities at the maxima obtained at 80 K for each complex.
radiative transition, concentration quenching, vibrational exci-
tation, and BET, respectively. Here, the emission intensity of
Tb nitrate in EPA (excited at 370 nm) was very weak in the
temperature range of 80–270 K (Fig. 9). The temperature-
dependence of the emission intensity of terbium(III) ion in
its nitrate slightly changed, indicating that the temperature-
dependence of the f–f transition in terbium(III) ion was neg-
ligible as shown in the inset of Fig. 9. Thus, we assume that
kr and knr are not dependent on the temperatures. From
Eq. 1, only kv and kBET should decrease with a decrease in
the temperature. Accordingly, the ꢃ increases. ꢃ is determined
mainly by kBET, because the temperature-dependence of kv in
terbium(III) complex is less than that of kBET. Less tempera-
ture-dependence of kv has been described in the section of
‘‘Temperature-Dependent Photosensitized Luminescence.’’
Branching Ratio. In the case of EuIII, the radiative rates
of europium(III) complexes are linked to geometric structures.7
If there is no inversion symmetry at the lanthanide ion sites,
uneven ligand-field components can mix with opposite-parity
states in 4fn-configuration levels. Electric-dipole transitions
are then no longer strictly forbidden in the ligand fields, result-
ing in faster radiative rates.31 We wish to discuss the radiative
rates and the molecular structures of terbium(III) complexes in
the same way as europium(III) complex.9 Thus, the branching
ratios evaluated by the emission spectra of the terbium(III)
complexes were compared in order to discuss the radiative
rates in relation to the symmetry of the ligand fields here again.
The emission spectra of the europium(III) complexes were
is known as a magnetic-dipole transition independent of the
changes of the ligand fields. In the case of terbium(III) com-
5
7
plex, the referential peaks chosen were the D4 ! F5 (543
7
nm) and 5D4 ! F6 transition (488 nm), corresponding to
7
the magnetic-dipole transition (5D4 ! F5) of ꢀJ ¼ 1 and
7
the electric-dipole transition (5D4 ! F6) of ꢀJ ¼ 2, respec-
tively. Here, the spectra were normalized with respect to the
7
5D4 ! F5 transition corresponding to the magnetic-dipole
transition being independent of the ligand fields. The branch-
ing ratio was defined as that of the intensity of the electric-
dipole transition to the magnetic-dipole transition. Asymmetry
of the ligand field is reflected as increase in the branching
ratio, resulting in increase in the radiative rate.7
The emission spectra of complex 3 normalized with respect
to the magnetic dipole transition at 80, 100, 200, and 280 K
are shown in Fig. 10C. No change in the branching ratios
was observed at these different temperatures. So, the ligand
field should not be changed by changing the temperature.
Therefore, kr should be independent of the temperatures. The
branching ratios for complexes 1 and 4 did not show any
change with the change in the temperature as well as for com-
plex 3 (Figs. 10A, 10C, and 10D), leading to the same discus-
sion on the ligand field and kr. Thus, it was concluded that the
geometric structures of complex 1, 3, and 4 do not change with
temperature. On the other hand, though the branching ratios of
complex 2 and 5 showed no change in the temperature range of
180–280 K, they increased with a decrease in the temperature
below 180 K (Figs. 10B and 10E), indicating the structural
change below this temperature.
7
normalized with respect to the 5D0 ! F1 transition, which

S. Katagiri et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007) 1499
0
-1
-2
-3
0
a
b
c
0
-1
-2
-1
-2
-3
x10−6
-4
0
1
2
3
4
0.0 0.5 1.0 1.5 2.0
Time / ms
0
1
2
Time / ms
Time / ms
0
d
e
-1
-1
-2
-3
-4
-2
-3
0.0
0.5
Time / ms
1.0
0
2
4
6
Time / ms
Fig. 7. Temperature-dependence of the emission decay of a) 1 (1.2 mM) (inset: 180–280 K), b) 2 (0.28 mM), c) 3 (1.0 mM),
d) 4 (1.3 mM), and e) 5 (1.2 mM) in EPA in the temperature range of 80–280 K.
270 K
1.0
800
0.5
600
µ
0.0
400
100 200
Temperature / K
80 K
200
0
100
150
200
250
Temperature / K
480 500 520 540 560 580 600
Wavelength / nm
Fig. 8. Temperature-dependence of ꢃ of 1 ( ), 2 ( ), 3
(
), 4 ( ), and 5 ( ) in EPA obtained from Fig. 7.
Fig. 9. Emission spectra of Tb(NO3)3 in EPA (0.063 M)
observed between 80 and 270 K, Ex. at 370 nm, slit 5 nm
for excitation/5 nm for emission. Inset: Temperature-
dependence of the emission intensities observed at
543 nm.
Emission spectra of complexes 1, 3, and 4 at 80 K normal-
7
ized with respect to the 5D4 ! F5 transition in each spectrum
are shown in Fig. 11. The order of the branching ratios ob-
tained from Fig. 11 was complex 1 > complex 4 ; complex
3, indicating that the symmetry of these complexes should
decrease in this order. Therefore, kr should increase in this
order, because the f–f transition is allowed upon degeneration
of the symmetry of the complex structure. Here, when BET is
completely suppressed at 80 K, ꢃ of the complexes should
be controlled only by kr from Eq. 1. The order of ꢃ of the
complexes at 80 K was complex 3 ; complex 4 > complex 1.
Therefore, the order of the value of kr evaluated from ꢃ was
complex 1 > complex 4 ; complex 3, agreeing with the order
of kr estimated from the branching ratio.
We excluded complexes 2 and 5 from the above discussion.
Because ꢃ of complexes 1, 3, and 4 reached a maximum and

1500 Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007)
Energy Transfer in Terbium(III) Complexes
1.0
1.0
1.0
0.5
0.0
A
B
C
0.5
0.0
0.5
0.0
80 K
180 K
480
500
520
540
480 500
520 540
560
480
500
520
540
560
Wavelength / nm
Wavelength / nm
Wavelength / nm
1.0
1.0
D
E
0.5
0.0
0.5
0.0
80 K
180 K
480
520
560
600
480
500
520
540
560
Wavelength / nm
Wavelength / nm
Fig. 10. Emission spectra of the complexes in EPA normalized with respect to the magnetic-dipole transition at various tempera-
tures. A) 1 at 80, 100, 200, and 280 K, B) 2 at 80, 100, 140, 160, 200, 240, and 280 K, C) 3 at 80, 100, 200, and 280 K, D) 4 at 80,
100, 200, and 280 K, and E) 5 at 80, 100, 140, 160, 200, 240, and 280 K. The concentrations were 15.5 mM for 1, 16.8 mM for 2,
18.3 mM for 3, 0.59 mM for 4, and 16.5 mM for 5.
when EG is large, resulting in the suppression of BET at higher
temperatures for the complexes having large EG. Therefore,
the complexes should undergo suppression of BET in the or-
der of complex 4 > complex 3 > complex 2 = complex 5 >
complex 1 when the temperature is decreased. However, EG
of complex 4 was very large, and therefore, its emission life-
time should increase above the measurement temperatures
employed in this work. No clear increase was observed in
the present work.
Since EG determines the temperature at which suppression
of BET occurs, the increment in the ꢃ should be in the order
of complex 3 > complex 2 > complex 1. However, the order
obtained in the present experiment was complex 3 > complex
1 > complex 2. The temperature-dependence of the ꢃ expect-
ed from EG was observed for complexes 1, 3, and 4, but com-
plex 2 showed unexpected behavior. The rapid increase in the
ꢃ was observed from 280 to 220 K for complex 3, which was
attributed to suppression of BET. The increase observed from
180 to 125 K for complex 1 should be also attributed to sup-
pression of BET. An increase in ꢃ was expected from 200–
250 K for complex 2 from EG, but a volcano-shaped depen-
dence was observed at this temperature region. In addition,
the rapid increase was shifted to the lower temperature region
of 80–150 K.
1.0
0.5
0.0
Complex 1
Complex 4
Complex 3
480
500
520
540
560
Wavelength /nm
Fig. 11. Emission spectra of complex 1 (15.5 mM), 3 (18.3
mM), and 4 (0.59 mM) at 80 K normalized with respect to
the magnetic-dipole transition.
complexes 2 and 5 showed an increasing behavior at 80 K
even, BET could still contribute to the emission properties of
these two complexes.
Discussion
In the previous section, it was shown that the branching
ratio observed in the emission of complex 2 having HFA as a
ligand increased a with decrease in the temperature, while
Dependence of the Emission Lifetimes on the Tempera-
tures. Thermal energy required for BET becomes higher

S. Katagiri et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007) 1501
a
c
b
d
50.0
: O
-50.0
: C
: H
: F
Fig. 12. Three-dimensional electrostatic potential views obtain by DFT calculation. a) the ground state of acac, b) the excited
triplet state of acac, c) the ground state of HFA, d) the excited triplet state of HFA. Electrostatic potential map from ꢂ50:0 to
ꢂ3
50.0 kcal molꢂ1 onto a 0.002 eA
.
˚
ꢀ
tential maps of the two ligands for both in the ground state
those of the other complexes having no HFA did not change by
the changes of the temperatures, indicating that the symmetry
of complex 2 should change with changes in the temperature.
Therefore, the temperature-dependence of ꢃ of complex 2
cannot be explained only by the idea of BET determined by EG.
Unusual Temperature-Dependence of the Emission Prop-
erty of [Tb(hfa)3(H2O)3]. We explained the temperature-
dependence of the photosensitized emission of the terbium(III)
complexes by the concept of BET determined by EG as de-
scribed above. However, we could not explain the behavior
of complex 2 having HFA as a ligand. Therefore, we introduce
another idea in order to understand the unusual behavior of
complex 2: presence of an energy barrier (EB) in the forward
energy transfer (FET) from the excited triplet state to ter-
bium(III) ion. This EB should be related to a change between
the starting structure of the complex and the final one, which
can be referred as ‘‘re-orientation’’ of the complex induced by
FET. This EB should depend on the chemical structures of
the ligands.32
EB of FET for complex 2 should be large compared to those
of the other complexes, resulting in the competitive relation
between FET and BET. This should be a reason why only
complex 2 showed unusual behavior. In the case of HFA
having –CF3 groups with an electron-withdrawing properties,
the electron configuration of the ligand is greatly different be-
tween the ground state and the excited triplet state. Therefore,
the chelating structures of the ligands with a terbium(III) ion
should be very different between the two situations as demon-
strated by the following section dealing with the quantum
chemical calculations.
ground state at DFT B3LPY/6-31G levels. Electrostatic po-
and the excited triplet state were obtained by a single point cal-
culation at the DFT B3LPY/6-31G levels on the assumption
ꢀ
that the geometries of the ligands in the excited triplet state did
not change from those in the ground state. Three-dimensional
electrostatic potential views of the two ligands are depicted in
Fig. 12 (a: S0 of acac, b: T1 of acac, c: S0 of HFA, d: T1 of
HFA). As shown in Figs. 12c and 12d, there was a large differ-
ence in the electrostatic potential between S0 and T1 states of
the hfa ligand, and it was more dramatic than those seen for
acac (Figs. 12a and 12b). The charges of the two O atoms
(ꢂ0:204) of the hfa ligand in T1 state was more positive than
that in S0 state (ꢂ0:275), indicating a change in the charge dis-
tribution caused by the excitation and the intersystem crossing.
On the other hand, the charges of the two O atoms (ꢂ0:325) of
the acac ligand in T1 state were also more positive than that in
S0 state (ꢂ0:345). However, the change in the charges of the O
atoms caused by the excitation and the intersystem crossing for
the hfa ligand was 0.071, which was larger than that (0.020)
for the acac ligand. The above results obtained by the DFT cal-
culations support the following discussion.
In the ground state, the negative charge of the HFA ligand is
less localized than in the acac ligand and is extended on the
whole molecule. Furthermore, the excitation and the intersys-
tem crossing enhance the delocalization of the charge in the
molecules. The induced change of the charge is more dramatic
for the hfa ligand than for the acac ligand therein, leading to an
idea that the re-orientation energies accompanied with FET
and BET are higher for complex 2 than complex 4, if the re-
orientation energies originate mainly from changes in the che-
lating structures of the complexes. These high re-orientation
energies should act as EB present in FET and BET for com-
plex 2. The barriers should be different from each other for
We carried out DFT calculations on acac (used as a ligand
in complex 4) and HFA (used as a ligand in complex 2) in the
singlet ground state (S0) and the excited triplet state (T1). Fully
optimized geometries were obtained for the ligands in the

1502 Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007)
Energy Transfer in Terbium(III) Complexes
FET and BET, because the starting structures are different for
both processes.
References
G. A. Hebbink, L. Grave, L. A. Woldering, D. N.
1
Dependence of the emission lifetimes on temperatures ob-
served for complex 2 was different from that for complexes
1, 3, and 4. The dependence of complex 2 showed a volcano-
shape between 200 and 280 K and rapidly increased below
150 K. The branching ratio showed no change between 200
and 280 K, and increased below 200 K, indicating a structural
change in complex 2 below this temperature. The volcano-
shaped dependence is attributed to the competitive contribu-
tion of FET and BET between 200 and 280 K. When the tem-
perature is decreased, suppression of BET first should induce
an increase in ꢃ and then that of FET should lead to its de-
crease, resulting in the volcano-shape. A rapid increase in
the emission lifetime observed below 150 K is related to the
changed chelating structure of complex 2, which was con-
firmed by the discussion of the change of the branching ratio.
On the other hand, in the case of complexes 1, 3, and 4, the
re-orientation energy should be small, because they possess
methyl group or a phenyl group, which is less electron-with-
drawing than –CF3. Accordingly, EB is small. So, the tempera-
ture-dependence of complexes 1, 3, and 4 should be controlled
mainly by BET. Thus, the results were understandable from
EG for these complexes. For complex 5, the second ligands
were bulky. So, the re-orientation should need more thermal
energy. In consequence, the suppression of BET appeared in
the whole range temperature.
Reinhoudt, F. C. J. M. Van Veggel, J. Phys. Chem. A 2003,
107, 2483.
2
S. I. Klink, L. Grave, D. N. Reinhoudt, F. C. J. M. van
Veggel, M. H. V. Werts, F. A. J. Geurts, J. W. Hofstraat, J. Phys.
Chem. A 2000, 104, 5457.
3
S. I. Klink, G. A. Hebbink, L. Grave, F. C. J. M. van
Veggel, D. N. Reinhoudt, L. H. Slooff, A. Polman, J. W.
Hofstraat, J. Appl. Phys. 1999, 86, 1181.
4
S. I. Klink, G. A. Hebbink, L. Grave, P. G. B. O. Alink,
F. C. J. M. van Veggel, M. H. V. Werts, J. Phys. Chem. A
2002, 106, 3681.
5
L. Ying, A. Yu, X. Zhao, Q. Li, D. Zhou, C. Huang, S.
Umetani, M. Matasai, J. Phys. Chem. 1996, 100, 18387.
B. Yan, H. Zhang, S. Wang, J. Ni, J. Photochem.
Photobiol., A 1998, 116, 209.
Y. Hasegawa, M. Yamamuro, Y. Wada, N. Kanehisa, Y.
Kai, S. Yanagida, J. Phys. Chem. A 2003, 107, 1697.
Y. Wada, M. Sato, Y. Tsukahara, Angew. Chem., Int. Ed.
2006, 45, 1925.
6
7
8
9 K. Nakamura, Y. Hasegawa, Y. Wada, S. Yanagida, Chem.
Phys. Lett. 2004, 398, 500.
10 a) E. J. Schimitschek, E. G. K. Schwarz, Nature 1962, 196,
832. b) H. Samelson, A. Lempicki, C. Brecher, V. Brophy, Appl.
Phys. Lett. 1964, 5, 173. c) T. Kobayashi, S. Nakatsuka, T.
Iwafuji, K. Kuriki, N. Imai, T. Nakamoto, C. D. Claude, K.
Sasaki, Y. Koike, Y. Okamoto, Appl. Phys. Lett. 1997, 71,
2421. d) K. Kuriki, Y. Koike, Y. Okamoto, Chem. Rev. 2002,
102, 2347.
Conclusion
We observed the temperature-dependence of the photosensi-
tized emission of terbium(III) complexes having ꢁ-diketonato
ligands systematically varied in their structures over a wide
range of temperatures. Two decisive factors were shown to ef-
fect the temperature-dependence. The first was EG between
the excited triplet state and the emitting level of terbium(III)
ion. The second was the EB accompanying the FET and BET.
These EB are directly related to the re-orientation energy origi-
nated from the difference between the starting structures of the
complexes and the final ones in the energy-transfer processes.
When this difference is small, EB is also small. So, the tempera-
ture-dependence of the emission properties can be fully ex-
plained by EG. On the other hand, when this difference is large,
the temperature-dependence of the emission is seriously affect-
ed by the competition between FET and BET. Then, an unusual
behavior like a volcano-shaped temperature-dependence curve
is observed. Quantum chemical calculations supported that
the electron configurations of the ligands greatly change be-
tween the ground state and the excited triplet state especially
in the case of hfa, which has electron-withdrawing –CF3
groups. In this work, we showed a new strategy for the design
of the terbium(III) complexes with desired emission properties.
11 a) S. Shindo, Low-Speed Wind Tunnel Testing Technique,
ed. by CORONA PUBLISHING CO., LTD., Tokyo Japan,
1992, Chap. 1. b) Proceedings of MOSAIC International Work-
shop, Tokyo, Japan, 2003, November 4–9. c) K. Mitsuo, K. Asai,
A. Takahashi, H. Mizushima, Meas. Sci. Technol. 2006, 17, 1282.
d) K. Mitsuo, K. Asai, M. Hayasaka, M. Kameda, J. Visualization
2003, 6, 321.
12 J. N. Demas, B. A. DeGraff, J. Macromol. Sci., Chem.
1988, A25, 1189.
13 A. P. Zipp, L. Sacksteder, J. Streich, A. Cook, J. N. Demas,
B. A. DeGraff, Inorg. Chem. 1993, 32, 5629.
14 R. V. Slone, J. T. Hupp, C. L. Stern, T. E. Albrecht-
Schmitt, Inorg. Chem. 1996, 35, 4096.
15 K. Maruszewski, D. Andrzejewski, W. Strek, J. Lumin.
1997, 72–74, 226.
16 B. F. Carroll, J. P. Hubner, K. S. Schanze, J. M. Bedlek-
Anslow, J. Visualization 2001, 4, 121.
17 G. Khalil, M. Gouterman, S. Ching, C. Costin, L. Coyle, S.
Gouin, E. Green, M. Sadilek, R. Wan, J. Yearyean, B. Zelelow,
J. Porphyrins Phthalocyanines 2002, 6, 135.
´ ´
18 a) T. Gunnlaugsson, J. P. Leonard, K. Senechal, A. J.
Harte, J. Am. Chem. Soc. 2003, 125, 12062. b) J. B. Beck, S. J.
Rowan, J. Am. Chem. Soc. 2003, 125, 13922. c) R. K. Mahajan,
I. Kaur, R. Kaur, S. Uchida, A. Onimaru, S. Shinoda, H. Tsukube,
Chem. Commun. 2003, 2238. d) S. Petoud, S. M. Cohen, J.-C. G.
This work was supported by a Grant-in-Aid for Scientific
Research (No. 17750014) and a Grant-in-Aid for Scientific
Research on Priority Areas (417) (No. 17029038) from MEXT
of the Japanese government. One of the authors (S. Katagiri)
expresses his special thanks for the center of excellence
(21COE) program ‘‘Creation of Integrated EcoChemistry’’ of
Osaka University.
Bunzli, K. N. Raymond, J. Am. Chem. Soc. 2003, 125, 13324.
¨
e) P. B. Glover, P. R. Ashton, L. J. Childs, A. Rodger, M. Kercher,
R. M. Williams, L. De Cola, Z. Pikramenou, J. Am. Chem. Soc.
2003, 125, 9918. f) T. Gunnlaugsson, J. P. Leonard, Chem.
Commun. 2003, 2424.
19 M. Mitsuishi, S. Kikuchi, T. Miyashita, Y. Amao, J. Mater.

S. Katagiri et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 8 (2007) 1503
Chem. 2003, 13, 2875.
27 L. R. Melby, N. J. Rose, E. Abramson, J. C. Caris, J. Am.
Chem. Soc. 1964, 86, 5117.
20 G. E. Khalil, K. Lau, G. D. Phelan, B. Carlson, M.
Gouterman, J. B. Callis, L. R. Dalton, Rev. Sci. Instrum. 2004,
75, 192.
21 S. Katagiri, Y. Hasegawa, Y. Wada, S. Yanagida, Chem.
Lett. 2004, 33, 1438.
22 J. Yu, J. Lumin. 1998, 78, 265.
28 M. Latva, H. Takalo, V. M. Mukkala, C. Matachescu, J. C.
Rodriguez-Ubis, J. Kankare, J. Lumin. 1997, 75, 149.
29 Y. Hasegawa, M. Iwamuro, K. Murakoshi, Y. Wada, R.
Arakawa, T. Yamanaka, N. Nakashima, S. Yanagida, Bull. Chem.
Soc. Jpn. 1998, 71, 2573.
23 a) G. L. Closs, P. Piotrowiak, J. M. MacInnis, G. R.
Fleming, J. Am. Chem. Soc. 1988, 110, 2652. b) G. L. Closs,
M. D. Johnson, J. R. Miller, P. Piotrowiak, J. Am. Chem. Soc.
1989, 111, 3751.
24 D. L. Dexter, J. Chem. Phys. 1953, 21, 836.
25 S. Sato, M. Wada, Bull. Chem. Soc. Jpn. 1970, 43, 1955.
26 L. Liu, W. Li, Z. Hong, J. Peng, X. Liu, C. Liang, Z. Liu,
J. Yu, D. Zhao, Synth. Met. 1997, 91, 267.
30 Y. Hasegawa, T. Ohkubo, K. Sogabe, Y. Kawamura,
Y. Wada, N. Nakashima, S. Yanagida, Angew. Chem., Int. Ed.
2000, 39, 357.
31 F. Gan, Laser Materials, World Scientific, Singapore,
1995, p. 70.
32 F. Gutierrez, C. Tedeschi, L. Maron, J. P. Daudey, R.
Poteau, J. Azema, P. Tisnes, C. Picard, Dalton Trans. 2004, 1334.