Understanding the aggregation induced emission enhancement for a compound with excited state intramolecular proton transfer character

Article abstract of DOI:10.1039/c0cp01181a

A few of excited state intramolecular proton transfer (ESIPT) compounds have been discovered for their aggregation induced emission enhancement (AIEE). To understand the AIEE mechanism, an ESIPT compound BTHPB (N-(4-(benzo[d] thiazol-2-yl)-3-hydroxyphenyl

Full text of DOI:10.1039/c0cp01181a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
PCCP
Cite this: Phys. Chem. Chem. Phys., 2011, 13, 2044–2051
www.rsc.org/pccp
PAPER
Understanding the aggregation induced emission enhancement for a
compound with excited state intramolecular proton transfer characterw
Rui Hu,a Shayu Li,*a Yi Zeng,b Jinping Chen,b Shuangqing Wang,a Yi Li*b and
Guoqiang Yang*a
Received 14th July 2010, Accepted 4th November 2010
DOI: 10.1039/c0cp01181a
A few of excited state intramolecular proton transfer (ESIPT) compounds have been discovered
for their aggregation induced emission enhancement (AIEE). To understand the AIEE
mechanism, an ESIPT compound BTHPB (N-(4-(benzo[d]thiazol-2-yl)-3-hydroxyphenyl)-
benzamide) with simple structure was designed and synthesized. BTHPB showed apparent AIEE
property and the emission efficiency was observed as high as 0.27 in the aggregates. On the basis
of viscochromism experiments and calculations employing the linear coupling model, the
restriction of the rotation between the two subunits taken place in ESIPT was considered as the
main factor for the AIEE. The micro- and femtosecond transient absorption experiments offered
evidence for the considerations. Additionally, we also observed a negative effect of aggregation on
the fluorescence emission in the system. So the AIEE of ESIPT compound BTHPB originated
from the combination effects of positive and negative factors induced by the aggregation.
Although to control the self-quenching or nonradiative
1. Introduction
processes in aggregates is still a challenge for the luminescent
Luminescent organic aggregates such as amorphous solids,
materials, much effort has been spent to mitigate these notorious
nano- or microcrystals are found to be much more photostable
effects and a few new luminogens that can emit efficiently in
than isolated molecules.1 On account of other advantages
the aggregate state have been discovered.9–11 Successful
including facilitation of structural design, low cost, and unique
development was pioneered by Tang et al., who synthesized
properties, these aggregates have attracted considerable
a series of aromatic siloles that show unique enhanced
emission rather than fluorescence quenching in aggregates.12–17
interest for their potential applications in fluorescence sensors,
flat panel display and illumination, etc.2–5 In all of these
Different type of compounds with aggregation induced
applications, luminescent intensity of the aggregates is a key
emission enhancement (AIEE) characteristics have been reported
by several research groups.18–24 These compounds include
factor to evaluate their performance and even predominantly
determines their applicable extension. Unfortunately, most
aromatic siloles, arylethene derivatives, fluorenone excimers,
of the luminescent organic compounds are usually weak
dibenzosuberenylidene,
diphenyldibenzofulvenes,
pyran
or even non-emitting in their solid state, in contrast to the
strong fluorescence in their dilute solutions.6 This behavior
is generally attributed to the so-called concentration
quenching effect in that nonradiative deactivation of the
excited state is enhanced by strong intermolecular vibronic
interactions such as exciton coupling and excimer formation in
the solid state.7,8
derivatives, as well as conjugation polymers etc. According
to the structure of the compounds, the enhanced emission has
been attributed to single or combined effects of molecular
planarization, restricted group rotations in the molecules,
prevention of exciton diffusion and J-aggregates formation.25
To date, examples of the AIEE systems are still quite limited,
especially concerning the enormously increasing number of
luminescent organic compounds. So it is desirable to develop
more compounds with AIEE for their application possibility.
We have developed a novel class of AIEE material,
the excited state intramolecular proton transfer (ESIPT)
compounds.26 ESIPT compounds are of great interest from
a basic research viewpoint as well as for their various
applications.27 Several ESIPT compounds have been applied
in optic/optoelectronic devices, such as tunable solid-state
lasers and organic light emitting devices (OLEDs).28–30
In ESIPT compounds, an excited state tautomeric reaction
a Beijing National Laboratory for Molecular Sciences,
Key laboratory of Photochemistry, Institute of Chemistry,
Chinese Academy of Sciences, Beijing 100190, China.
E-mail: gqyang@iccas.ac.cn, shayuli@iccas.ac.cn;
Fax: +86-10-82617315; Tel: +86-10-82617263
b Key Laboratory of Photochemical Conversion and Optoelectronic
Materials, Technical Institute of Physics and Chemistry,
Chinese Academy of Sciences, Beijing 100190,
P. R. China. E-mail: yili@mail.ipc.ac.cn
w Electronic supplementary information (ESI) available: Spectra and a
SEM image. See DOI: 10.1039/c0cp01181a
c
2044 Phys. Chem. Chem. Phys., 2011, 13, 2044–2051
This journal is the Owner Societies 2011

View Article Online
concerning a proton transfer from a proton donor to an
acceptor group occurs via an intramolecular H-bond or
H-bonding bridge, which leads to an intrinsic peculiar four-level
photophysical cycle (absorption-proton transfer-emission-
back proton transfer) after photoexcitation.31 As a result, an
abnormal large Stokes shift is observed, which diminishes the
self-absorption and renders the possibility of intense emission
for these compounds in the solution with high concentration
or aggregates. So ESIPT compounds are expected to be
potential intrinsic luminescent materials.
with vigorous stirring for 20 s at room temperature, then the
solution was immediately taken for fluorescence, absorption
and fluorescent decay detection.
Spectroscopy characterization
UV-vis absorption spectra were recorded on a Hitachi U-3010
spectrophotometer. Fluorescence spectra were recorded using
a Hitachi F-4500 fluorescence spectrophotometer. The time-
resolved fluorescence spectra and fluorescence lifetime were
investigated with an Edinburgh FL900 spectrophotometer and
a pulsed LED from 370 nm was used to excite the samples.
The nanosecond transient absorption experiments were
investigated by Edinburgh Instruments LP920. The excitation
light was the harmonic of the Nd:YAG laser (Continuum
Surelite, 355 nm and 7 ns fwhm). A pulsed xenon arc lamp was
used to provide the analyzing light. The femtosecond transient
absorption spectra were recorded with ExciPro pump–probe
spectrometer. The pump light (170 fs, 360 nm, 1 kHz) was
provided by a combination of Coherent Chameleon Ultra,
Legend Elite and OPerA Solo. All the analyzed spectra were
corrected for a chirp of a white light continuum.
In previous works, the mechanisms of AIEE for ESIPT
compounds were preliminary investigated and explained as the
effects of combining J-aggregation and restriction of intra-
molecular rotation. In this work, we extend our earlier study
with several experimental and computational approaches to
understand the AIEE process. A new ESIPT compound, N-(4-
(benzo[d]thiazol-2-yl)- 3-hydroxyphenyl)benzamide (BTHPB),
with AIEE feature was designed and synthesized. The
flash photolysis and femtosecond transient absorption and
calculations employing the linear coupling model were carried
out to help us explore what is the most important factor for
the emission enhancement behavior in the aggregates of
ESIPT compounds.
Computational methods
Geometry optimization for the ground state of the BTHPB
molecule was carried out using the TURBOMOLE suite with
the density functional theory.34,35 Becke’s three-parameters
hybrid method, using the Lee–Yang–Parr correlation
functional (B3LYP), was employed here. The structures of
the lowest excited state for BTHPB keto tautomer were
obtained by TDDFT (time-dependent density functional
theory) methods, employing the hybrid functional B3LYP.
The calculations of vibration frequencies and normal coordinates
were carried out at the TDDFT structure. Employing the
linear coupling model, we investigated the major routes of
internal conversion of the excited states. For all the above
calculations we employed the 6-31G(d) basis set for carbon,
hydrogen, oxygen, nitrogen, and sulfur atoms.
Experimental
Materials
Benzoic acid and 1,10-carbonyldiimidazole (CDI) were
obtained from Beijing Chemical Reagents. Cyclohexane,
paraffin oil, ethyl ether, CH3CN and THF were AR grade
received from ACROS. Polystyrene (PS) was obtained from
ACROS. All chemicals were used directly without further
purification.
Synthesis of BTHPB
2-(4-Amino-2-hydroxyphenyl)benzothiazole (AHBA) was
synthesized according to ref. 32.
BTHPB
(N-(4-(benzo[d]thiazol-2-yl)-3-hydroxyphenyl)-
benzamide). The preparation procedure was performed
according to ref. 33. 602 mg benzoic acid (5 mmol) and
0.972 g CDI (6 mmol) were mixed at a dry argon atmosphere
in 30 ml toluene then refluxed for 6 h. When the temperature
cooled down, 1.224 g AHBA (5 mmol) was added to the
solution, the reaction system was stirred again and refluxed
overnight. The deposit was collected with a good yield, and
then purified by repeatedly washing with CH2Cl2/CH3OH
Results
Absorption and fluorescence spectroscopy
Solid BTHPB is soluble more or less in all conventional
organic solvents. The absorption spectra of BTHPB in cyclo-
hexane, ethyl ether, THF and CH3CN at room temperature
were shown in Fig. 1a and S1a.w The data in protic and
halogen solvents were excluded to avoid specific solute–solvent
interactions. From the spectra, it could be seen that the
absorbance of BTHPB seldom changed in both shape and
position with increasing the solvent polarity, which indicated
that the BTHPB molecule had no apparent conjugation-
structural changes in the ground state in different solvents.
All absorption bands with maxima at B350 nm were assigned
to the p–p* transition that corresponded to the coupling
between benzothiazole and the hydroxyphenyl ring.26 The
absorbance of BTHPB in water was also shown in Fig. 1a.
The absorption band at B350 nm decreased significantly
probably due to the formation of large particles, which were
in agreement with the SEM observations (see in Fig. S2).w
(V/V
= 10:1) to give a white solid (yield 74%).
1H-NMR(400 MHz, d6-DMSO, ppm) d = 11.67 (s, 1H),
10.48 (s, 1H), 8.13 (t, J = 8.92, 2H), 8.02 (d, J = 8.08, 1H),
7.98 (d, J = 7.28, 2H), 7.80 (d, J = 1.64, 1H), 7.64-7.51
(m, 4H), 7.42(t, J = 8.32, 2H). EI-MS (m/z): 346 (m+). Anal.
Calcd. for C20H14N2O2S: C, 69.35; H, 4.07; N, 8.09, found: C,
69.20; H, 4.11; N, 8.17. And the molecular structure were
shown in Scheme 1.
Preparation of the aggregates
The aggregates of BTHPB in water were prepared by once
injecting 50 mL THF solution (2.0 ꢀ 10ꢁ3 M) into 10 mL H2O
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 2044–2051 2045

View Article Online
molecules undergoing ESIPT reactions.37 However, emission
of BTHPB aggregates in water exhibited large a enhancement
from its monodisperse solutions though energy and band
shape were consistent. The luminescent quantum efficiency
of BTHPB aggregates was as high as 0.27 (t = 7.0 ns), which
was nearly 15-fold higher than that of the solution in
cyclohexane. The same emission energy and shape suggested
that the fluorescence in cyclohexane and aggregates came from
the transition between the same excited state and ground state
structures of cis-keto form. The different quantum efficiency
implied the different formation processes and/or radiationless
decay routes of the excited states.
Scheme 1 Molecular structure of BTHPB.
The fluorescence spectra of BTHPB in various solvents were
shown in Fig. 1b and S1b.w Unlike the absorbance changes,
BTHPB fluorescence responses to solvents exhibited a moderate
change in both shape and intensity. In cyclohexane, BTHPB
showed one emission band with a maximum at 509 nm. In
contrast, two emission bands peaking at 400 and 509 nm were
observed in other polar solvents. As was well known, ESIPT
compounds could form intermolecular H-bonds with protic or
polar solvents, which interrupted the ESIPT process and
induced a local emission from the enol form.36 So the emission
with a small Stokes shift, around 400 nm, was assigned from
the enol excited state with intermolecular hydrogen bond
configuration. With regard to the emission at 509 nm, the
large Stokes shift indicated that it could be assigned to the ESIPT
emission. It was worthwhile to note that almost no changes in
both the energy region and the band shape of the ESIPT
emission were observed with increasing the solvent polarity,
suggesting that the excited state of cis-keto was the form without
intramolecular charge transfer character. The fluorescence
spectrum of BTHPB aggregates is shown in Fig. 1b. Similar to
that in the cyclohexane solution, no enol but only ESIPT
emission was observed. Additionally, the shapes of excitation
spectra monitored at 509 nm (shown in Fig. S3w) resembled
closely those of the absorption spectra either in cyclohexane or in
aggregates, respectively. This indicated that all high-level excited
states can only relax to the same lowest excited state.
The quantum yield of the transition from enol* to cis-keto*
was generally very high due to the ultrafast ESIPT process in
comparison with other decay processes, and no enol emissions
were observed in cyclohexane solution or aggregates. We thus
mainly focused on the emission from the cis-keto* form in the
following discussion.
Discussion
In general, AIEE should be achieved easily if the aggregation
(i) resulted in protecting the fluorophore from fluorescence
quenching that normally took place for the dissolved
molecules (e.g. thermal motions, intersystem crossing), and
(ii) mitigated nonradiative deactivation processes that
normally took place in the aggregates (e.g. exciton coupling,
exciton diffusion).25 In the following section, the major routes
of the radiationless decay process of monodisperse BTHPB
were explored firstly, and then the effects of aggregation on
nonradiative deactivation processes were discussed. And
finally the origin of AIEE phenomena was investigated.
Nonradiative decay routes of monodisperse BTHPB
For the normal organic compounds, internal conversion (IC),
intersystem crossing (ISC) from singlet state to triplet state,
and energy transfer to surrounding molecules (ET) were three
mainly possible nonradiative decay processes for the dissolved
molecules.38 When BTHPB molecules were dispersed in
solvent, the absorbance of either cyclohexane or BTHPB itself
was far from the emission band and located in high
energy area, diminishing the possibility of ET with a trivial
Enhanced emission phenomenon
All the dilute solutions of BTHPB in four organic solvents
presented weak fluorescence. Among them, BTHPB in
cyclohexane gave the highest quantum efficiency, just only
0.018 (t = 0.23 ns). The lower fluorescent quantum efficiencies
in other polar solvents were probably induced by the inter-
molecular hydrogen bonds that decreased the number of
Fig. 1 Absorption and fluorescence spectra of BTHPB in cyclohexane and water. Concentration: 1.0 ꢀ 10ꢁ5 M. Excitation at 360 nm.
c
2046 Phys. Chem. Chem. Phys., 2011, 13, 2044–2051
This journal is the Owner Societies 2011

View Article Online
mechanism. And the average distance between two BTHPB
was several tens of nanometres in the dilute solution, which
first-principles linear-coupling-model calculations had been
proven as a reliable qualitative method more than a quantitative
one to explore IC nonradiative decay processes.43 Herein, we
investigated the vibrational modes with noticeable displacement
and reorganization energy.
prevented possible ET with Forster and Dexter mechanisms.
¨
So the ET process was rationally neglected. In addition,
intersystem crossing was not important for the HBT-type
compounds without heavy atoms at room temperature and
the absence of apparent spin–orbit coupling effect.39 Therefore
the IC route was the major component of a nonradiative decay
process for BTHPB solution. In the IC process, the energy of
the electronically excited state was relaxed to various
vibrational modes of the molecule, that is, torsions, rotations
and vibrations served as nonradiative channels for the excited
species to decay.40 In general, only molecular torsions and
rotations but not vibrations were restrained effectively by tight
packing of the molecules. Thereby, the AIEE of BTHPB
possibly originated from a restriction of molecular torsions
and rotations.
Under the displaced harmonic approximation, applying the
Born–Oppenheimer adiabatic approximation, the internal
conversion radiationless rate was determined by the
Huang-Rhys factor (see the theoretical details in ref. 44).
The Huang-Rhys factor characterized the modification of
vibrational quanta during transition from one electronic state
to another, which was given as:
ojDQ2j
Sj ¼
2ꢀh
DQj was the displacement of jth normal mode, and oj was
the vibrational frequency of jth normal mode. And the
reorganization energy of jth normal mode was
To verify this hypothesis, BTHPB was dissolved in the
solvents with different viscosities. Consideration was given to
the thought that higher viscosity of the solvent could hamper
the torsions and rotations of the molecules. Mixtures of
cyclohexane and paraffin oil were chosen as the solvents to
study the viscosity effect on the emission of BTHPB. As
expected, the fluorescence of BTHPB, shown in Fig. 2b, was
enhanced in turn with increasing the viscosity of the solvent
while the absorbance was consistent, as shown in Fig. 2a. The
emission quantum efficiency of BTHPB in paraffin oil (80%
volume ratio) was calculated as 0.050, which was nearly 3-fold
higher than that of cyclohexane solution. Taking the fact that
the viscosity of paraffin oil was only 15.3 cP, much lower than
that of normal organic soft-material,41 the restrictions of the
torsional and rotational motions would be much more
apparent in the aggregates the viscosity of which was estimated
reasonably to at least 1011 cP (pitch, 2.3 ꢀ 1011 cP).42
Therefore, the restriction of the intramolecular motions played
a key role for AIEE phenomena of BTHPB, similar to other
AIEE compounds.
lj = Sjojꢀh
In Table 1, the Huang-Rhys factor and the reorganization
energy of the cis-keto excited-state tautomer between the S1
and S0 states with noticeable displacement were given.
The modes with frequency of 1372, 1373 and 1412 cmꢁ1 had
major contributions to the reorganization energy. This
indicated that the three vibrations were the most efficient
internal conversion deactive routes of cis-keto excited state.
The first two modes were assigned to in-plane bending motions
of C–H in the cyclohexadienone ring. The third mode
represented a mixing of aromatic skeletal stretching vibration
and C–N (amide) stretching vibration. Additionally, it could
be seen from Table 1 that the other modes with large
reorganization energy (>100 cmꢁ1) could be assigned to
rotating, swing or bending motions, except the last mode that
belonged to stretching vibration of N–H in the thiazole ring. It
is worthwhile to note that the mode 19 cmꢁ1 with large
reorganization energy corresponded to rotation around the
bond between dihydrobenzo[d]thiazole and the cyclohexadienone
ring, as shown in Fig. 3. The huge Huang-Rhys factor of this
mode suggested that the rotation also contributed largely to
the IC process.43 As mentioned above, only rotations and
Although the restriction of torsions and rotations by tight
packing in the aggregate was very important for the AIEE
phenomenon, the main nonradiative channels for the excited
species could not be deduced only from the fluorescent
viscosity effects and thus remained unclear. In previous works,
Fig. 2 Absorption and fluorescence spectra of BTHPB in mixtures of cyclohexane and paraffin oil. Concentration: 1.0 ꢀ 10ꢁ5 M. Excitation at
360 nm.
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 2044–2051 2047

View Article Online
Table 1 Huang-Rhys factor and the reorganization energy for the
noticeable displaced vibration modes of cis-keto tautomer of BTHPB
assigned to a triplet state. By analogy with other ESIPT
compounds, the transient signals were attributed to the
absorbance of a trans-keto photoproduct.51 This absorption
from the transient photoproduct proved that the rotation
between the dihydrobenzo[d]thiazole group and the cyclo-
hexadienone ring was present in the cyclohexane solution. In
the aggregates, however, no transient absorption signals were
detected, suggesting the restriction of the rotation by the
physical constraint in the rigid matrix. The transient absorption
measurements thus offered a solid experimental support to our
consideration that the restriction of the rotation between the
two subunits that took place in the ESIPT process was mainly
responsible for the AIEE.
oj/cmꢁ1
Sj
lj/cmꢁ1
19
112
214
303
317
532
561
987
1099
1118
1132
1248
1312
1372
1373
1412
1495
1497
3138
13.12
0.4622
0.747
0.2895
0.2068
0.3321
0.211
0.2261
0.1236
0.0683
0.0609
0.0769
0.0739
0.3156
0.2496
0.5699
0.1136
0.0773
0.0934
245
51
159
87
65
176
118
223
135
76
68
96
96
433
342
805
169
115
293
To further investigate the difference of the ESIPT cycle
between the solution and the aggregate, femtosecond transient
absorption was also carried out. It was well-accepted that the
solvent had no effects on the ESIPT process and on the
efficiency of the creation of photochromic tautomers.52 We
thus chose THF (as a better solvent for BTHPB) and a
polystyrene matrix instead of cyclohexane and aggregates
(bad signal/noise ratio), respectively, because of the limitation
of the detection sensitivity.
torsions could be constrained effectively by packing, thereby
we considered that the restriction of intramolecular motions,
especially the rotation of the two subunits which took place in
the ESIPT process was the main cause for AIEE.
Spectral traces selected for several time delays between the
pump and probe were shown in Fig. 5. During short time, two
positive bands and an obvious negative band were observed in
the BTHPB solution. The positive absorption band around
660 nm was generated as the same time with the negative band.
The kinetic curve (see in Fig. S5w) was well fitted with the
convolution of the instrumental function at the probe
wavelength of 660 nm, showing a mono-exponential decay
with a lifetime of 40 ps. According to the literature, it was
assigned to the absorbance of S1 (p,p*) cis-keto state.51 The
decay kinetic curve (shown in Fig. S5w) of the negative band
was well fitted with a 38 ps mono-exponential decay that was
quite consistent with the lifetime of S1 keto state as mentioned
above. Additionally, because the position of the residual
negative band was close to the steady-state emission, it was
assigned to the stimulated emission from S1 (p,p*) of the
cis-keto form. Further analysis of the decay curve indicated
that the formation of the stimulated emission band was very
fast even within the time range of the instrument function.
This ultrafast process was commonly regarded as ESIPT
process. Another positive band in the 440–500 nm range was
assigned to the absorbance of the trans-keto photoproduct,
which remained nearly constant after its generation during our
measurement time range. Quite different from that in BTHPB
solution, only two absorption bands of BTHPB were observed
in the polystyrene matrix: the negative band around 510 nm
and the positive band at 660 nm. The absorption in the range
of 440–500 nm of the trans-keto photoproduct was not
detected. Similar to the results of the flash photolysis
experiment, the femtosecond absorption spectra also
confirmed that the trans-keto photoproduct was only formed
in the solution but not in the solid matrix, suggesting no
generation of trans-keto tautomer occurred.
Aggregation effects on nonradiative decay processes
The generally accepted ESIPT mechanism of HBT-type
ESIPT compounds is shown in Scheme 2. After the ESIPT,
cis-keto* relaxed with two competitive processes: radiative
decay to the cis-keto ground state and nonradiative decay to
the trans-keto photoproduct.45–49 The cis-keto species
returned rapidly through ground state intramolecular proton
transfer (GSIPT) to the initial enol tautomer and completed
the four-level photophysical cycle, whereas the trans-keto
photoproduct was stable from a few tens of microseconds to
milliseconds.50 We noted that the trans-keto form was
transformed from the cis-keto form by rotating 1801 via a
bond between the dihydrobenzo[d]thiazole group and the
cyclohexadienone ring. The rotation was considered the main
nonradiative decay route, as mentioned in the previous
discussion. Therefore, the block to rotation or not was
expected to be expressed by number changes of trans-keto
photoproduct. Transient absorption with flash photolysis
experiments were thus carried out to check our considerations.
As shown in Fig. 4, transient absorption spectra of BTHPB
in cyclohexane exhibited a positive band from 400 to 500 nm
with a maximum at 450 nm. The band lifetime was too long to
relax to zero in our instrument and could only be estimated
as B500 ms at all wavelengths. These decay kinetics were
insensitive to the presence of oxygen and thus could not be
All the experimental and calculation results described above
proved that the AIEE of BTHPB was originated from the
restriction of the intramolecular rotation process. However,
aggregation was not only a random packing process and the
Fig. 3 Normal mode displacement vectors of the vibration mode
19 cmꢁ1
.
c
2048 Phys. Chem. Chem. Phys., 2011, 13, 2044–2051
This journal is the Owner Societies 2011

View Article Online
quantum efficiency of the cis-keto* emission should be
expressed as:
F = FESIPT ꢀ FF(K) = FESIPT ꢀ kF(K) ꢀ tF(K)
(1)
Thus, the ratio of quantum efficiency for BTHPB in the
solution and the aggregates could be predicted as:
taFgðgKrÞ ꢀ kaggr
aggr
ESIPT
sol
Faggr
Fsol
F
FðKÞ
¼
ꢀ
ð2Þ
sol
tsFoðlKÞ ꢀ k
F
ESIPT
FðKÞ
The inherent radiation decay rate kF could be estimated,
derived by the equation kf = f ꢀ E2/1.4992, by using the
Einstein spontaneous emission coefficient, where f and E
were the oscillator strength and the energy of excited state,
respectively.44 Taking into account the fact that both
fluorescence energy and spectra shape of cis-keto emission
were consistent in aggregates and solutions, the oscillator
strength and energy of cis-keto* could be considered as
identical in both condensed states. In this case, eqn (2) could
be simplified. After substituting F and t into the equation, the
Scheme 2 Mechanism of a photoinduced process for HBT-type
ESIPT compounds.
aggr
sol
relationship between F
and F
was expressed as:
ESIPT
ESIPT
sol
ESIPT
F
E 2FEagSgIrPT, suggesting that the aggregation process
induced the decrease of the ESIPT yield.
In general, ESIPT was an extremely fast process that kESIPT
is about 1013 sꢁ1 in solution and 1012 sꢁ1 in the aggregates.
Neither the radiative decay (rate constant is about 108–109 sꢁ1
nor internal conversion (rate constant is about 109–1010 sꢁ1
)
)
was able to compete effectively with the ESIPT process.53 The
mechanism of a decay process fast enough to compete with
ESIPT was not very clear for the BTHPB aggregates. Taking
the fact that the spectrum overlap of absorbance and emission
of the enol tautomer was very small, the common exciton
Fig. 4 Time-resolved transient absorption of BTHPB in cyclohexane,
monitored by Edinburgh Instruments LP920.
diffusion with the Forster mechanism could be excluded.
¨
molecules located in aggregates could not be considered
simply as in a rigid matrix. As mentioned above, aggregation
induced the emission up to 15-fold enhancement. On the other
hand, we noticed that the fluorescent lifetime from cis-keto*
increased about 30-fold (from 0.23 ns to 7.0 ns) after the
aggregation process.
Herein, we presumed that the decrease of FESIPT resulted from
the strong intermolecular interactions of the enol forms
in aggregates on account of the considerable changes of
absorbance in the aggregates.18,54
Thus, two opposite effects were observed by the
aggregation: (i) the intermolecular interactions induced a
slight decrease of FESIPT, and (ii) restricted intramolecular
rotation induced an increase of FF(K). The combination of the
On the basis of photophysical principles and the specific
four-level photophysical cycle of ESIPT compounds, the
Fig. 5 Time-resolved transient absorption of BTHPB in (a) THF and in (b) polystyrene with 360 nm excitation, monitored by pump–probe
setups.
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 2044–2051 2049

View Article Online
two effects led to a significant emission enhancement in the
BTHPB aggregates.
16 H. Tong, Y. Q. Dong, M. Haussler, J. W. Y. Lam, H. H. Y. Sung,
I. D. Williams, J. Z. Sun and B. Z. Tang, Chem. Commun., 2006,
1133–1135.
17 A. J. Qin, C. K. W. Jim, Y. H. Tang, J. W. Y. Lam, J. Z. Liu,
F. Mahtab, P. Gao and B. Z. Tang, J. Phys. Chem. B, 2008, 112,
9281–9288.
Conclusion
18 F. Wang, M. Y. Han, K. Y. Mya, Y. B. Wang and Y. H. Lai,
J. Am. Chem. Soc., 2005, 127, 10350–10355.
19 J. Dong, K. M. Solntsev and L. M. Tolbert, J. Am. Chem. Soc.,
2009, 131, 662–670.
20 H. Tong, Y. Q. Dong, Y. N. Hong, M. Haussler, J. W. Y. Lam,
H. H. Y. Sung, X. M. Yu, J. X. Sun, I. D. Williams, H. S. Kwok
and B. Z. Tang, J. Phys. Chem. C, 2007, 111, 2287–2294.
21 H. Tong, Y. Dong, M. Haussler, Y. Hong, J. W. Y. Lam, H. H.
Y. Sung, I. D. Williams, H. S. Kwok and B. Z. Tang, Chem. Phys.
Lett., 2006, 428, 326–330.
22 K. Itami, Y. Ohashi and J. Yoshida, J. Org. Chem., 2005, 70,
2778–2792.
23 Z. X. Wang, H. X. Shao, J. C. Ye, L. Tang and P. Lu, J. Phys.
Chem. B, 2005, 109, 19627–19633.
24 Y. Liu, X. T. Tao, F. Z. Wang, X. N. Dang, D. C. Zou, Y. Ren and
M. H. Jiang, J. Phys. Chem. C, 2008, 112, 3975–3981.
25 J. F. Lamere, N. Saffon, I. Dos Santos and S. Fery-Forgues,
Langmuir, 2010, 26, 10210–10217.
26 Y. Qian, S. Y. Li, G. Q. Zhang, Q. Wang, S. Q. Wang,
H. J. Xu, C. Z. Li, Y. Li and G. Q. Yang, J. Phys. Chem. B,
2007, 111, 5861–5868.
Although there are several reports on the AIEE behavior of
organic compounds, there have been few works focusing on
the AIEE mechanism of the ESIPT molecule. The novel
ESIPT compound BTHPB presented an obvious AIEE
phenomenon. With the viscosity effects, micro- and femto-
second transient absorption experiments and quantum
chemical calculations, the predominant factor of AIEE for
BTHPB was determined to the restriction of the rotation
between the dihydrobenzo[d]thiazole group and the cyclo-
hexadienone ring that undertook the ESIPT process. The
negative factor induced by the aggregation was assigned to
the intermolecular interactions. It indicated that the efficiency
of the ESIPT process could not be simply assumed as 100%
and the total AIEE observation is the combination of the
positive and negative effects. The results implied that designing
luminescent compounds with higher emission efficiency in the
solid (or aggregates) should be balanced by the intermolecular
interactions.
27 S. J. Lim, J. Seo and S. Y. Park, J. Am. Chem. Soc., 2006, 128,
14542–14547.
28 S. Park, J. Seo, S. H. Kim and S. Y. Park, Adv. Funct. Mater.,
2008, 18, 726–731.
29 K. Sakai, T. Tsuzuki, Y. Itoh, M. Ichikawa and Y. Taniguchi,
Appl. Phys. Lett., 2005, 86, 081103.
30 S. M. Chang, Y. J. Tzeng, S. Y. Wu, K. Y. Li and K. L. Hsueh,
Thin Solid Films, 2005, 477, 38–41.
31 P. K. Sengupta and M. Kasha, Chem. Phys. Lett., 1979, 68,
382–385.
32 B. Ermanno, S. Piero, M. Mario and P. Macro, J. Heterocycl.
Chem., 1983, 20, 1517–1521.
Acknowledgements
We are grateful to the National Natural Science Foundation
of China (Grant Nos. 20703049, 20733007, 20873165,
50973118) and the National Basic Research Program
(2007CB808004, 2009CB930802).
33 M. K. Chmielewski, V. Marchan, J. Cieslak, A. Grajkowski,
V. Livengood, U. Munch, A. Wilk and S. L. Beaucage, J. Org.
Chem., 2003, 68, 10003–10012.
34 P. Deglmann, F. Furche and R. Ahlrichs, Chem. Phys. Lett., 2002,
362, 511–518.
35 P. Deglmann and F. Furche, J. Chem. Phys., 2002, 117, 9535–9538.
36 A. S. Klymchenko, V. G. Pivovarenko and A. P. Demchenko,
J. Phys. Chem. A, 2003, 107, 4211–4216.
37 S. Y. Li, Q. Wang, Y. Qian, S. Q. Wang, Y. Li and G. Q. Yang,
J. Phys. Chem. A, 2007, 111, 11793–11800.
38 N. J. Turro, in Modern Molecular Photochemistry, Benjamin
Cummings Publishing, Menlo Park, CA, 1978, ch. 9.
39 W. Al Soufi, K. H. Grellman and B. Nickel, Chem. Phys. Lett.,
1990, 174, 609–616.
40 M. Bixon and J. Jortner, J. Chem. Phys., 1968, 48,
715–726.
41 P. E. Baillagou and D. S. Soong, Polym. Eng. Sci., 1985, 25,
232–244.
42 R. Edgeworth, B. J. Dalton and T. Parnell, Eur. J. Phys., 1984, 5,
198–200.
References
1 Y. N. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Commun., 2009,
4332–4353.
2 W. Chen, J. O. Bovin, A. G. Joly, S. P. Wang, F. H. Su and
G. H. Li, J. Phys. Chem. B, 2004, 108, 11927–11934.
3 A. Hagfeldt and M. Gratzel, Chem. Rev., 1995, 95, 49–68.
4 J. Y. Kim and F. E. Osterloh, J. Am. Chem. Soc., 2005, 127,
10152–10153.
5 A. N. Shipway, E. Katz and I. Willner, ChemPhysChem, 2000, 1,
18–52.
6 B. K. An, S. K. Kwon, S. D. Jung and S. Y. Park, J. Am. Chem.
Soc., 2002, 124, 14410–14415.
7 J. B. Birks, in Photophysics of Aromatic Moelcules, Wiley, London,
1970.
8 J. B. Birks and M. Leite, J. Phys. Pt. B, 1970, 3, 513.
9 R. Deans, J. Kim, M. R. Machacek and T. M. Swager, J. Am.
Chem. Soc., 2000, 122, 8565–8566.
43 Q. Peng, Y. P. Yi, Z. G. Shuai and J. S. Shao, J. Am. Chem. Soc.,
2007, 129, 9333–9339.
10 W. Holzer, A. Penzkofer, R. Stockmann, H. Meysel, H. Liebegott
and H. H. Horhold, Polymer, 2001, 42, 3183–3194.
¨
44 S. W. Yin, Q. Peng, Z. Shuai, W. H. Y. Fang, Y. H. Wang and
Y. Luo, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 73,
DOI: 10.1103/PhysRevB.73.205409.
11 C. Belton, D. F. O’Brien, W. J. Blau, A. J. Cadby, P. A. Lane,
D. D. C. Bradley, H. J. Byrne, R. Stockmann and H. H. Horhold,
Appl. Phys. Lett., 2001, 78, 1059–1061.
45 W. E. Brewer, M. L. Martinez and P. T. Chou, J. Phys. Chem.,
1990, 94, 1915–1918.
46 M. A. Rıos and M. C. Rıos, J. Phys. Chem. A, 1998, 102,
´ ´
12 J. D. Luo, Z. L. Xie, J. W. Y. Lam, L. Cheng, H. Y. Chen,
C. F. Qiu, H. S. Kwok, X. W. Zhan, Y. Q. Liu, D. B. Zhu and
B. Z. Tang, Chem. Commun., 2001, 1740–1741.
1560–1567.
47 R. de Vivie-Riedle, V. De Waele, L. Kurtz and E. Riedle, J. Phys.
Chem. A, 2003, 107, 10591–10599.
13 J. W. Chen, C. C. W. Law, J. W. Y. Lam, Y. P. Dong, S. M. F. Lo,
I. D. Williams, D. B. Zhu and B. Z. Tang, Chem. Mater., 2003, 15,
1535–1546.
48 A. J. A. Aquino, H. Lischka and C. Hattig, J. Phys. Chem. A, 2005,
¨
109, 3201–3208.
14 J. W. Chen, H. Peng, C. C. W. Law, Y. P. Dong, J. W. Y. Lam,
I. D. Williams and B. Z. Tang, Macromolecules, 2003, 36,
4319–4327.
15 J. W. Chen, B. Xu, X. Y. Ouyang, B. Z. Tang and Y. Cao, J. Phys.
Chem. A, 2004, 108, 7522–7526.
49 S. R. Va
F. Rodrı
1814–1826.
zquez, M. C. R. Rodrıguez, M. Mosquera and
´ ´
guez-Prieto, J. Phys. Chem. A, 2007, 111,
´
c
2050 Phys. Chem. Chem. Phys., 2011, 13, 2044–2051
This journal is the Owner Societies 2011

View Article Online
50 M. Barbatti, A. J. A. Aquino, H. Lischka, C. Schriever,
S. Lochbrunner and E. Riedle, Phys. Chem. Chem. Phys., 2009,
11, 1406–1415.
51 M. Sliwa, N. Mouton, C. Ruckebusch, S. Aloise, O. Poizat,
G. Buntinx, R. Metivier, K. Nakatani, H. Masuhara and
T. Asahi, J. Phys. Chem. C, 2009, 113, 11959–11968.
52 M. Zio
´
"ek, J. Kubicki, A. Maciejewski, R. Naskre˛cki and
A. Grabowska, J. Chem. Phys., 2006, 124, 124518.
53 B. Valeur, in Molecular Fluorescence - Principles & Applications,
John Wiley and Sons Ltd., 2002.
54 Y. Kawamura, J. Brooks, J. J. Brown, H. Sasabe and C. Adachi,
Phys. Rev. Lett., 2006, 96, DOI: 10.1103/PhysRevLett.96.017404.
c
This journal is the Owner Societies 2011
Phys. Chem. Chem. Phys., 2011, 13, 2044–2051 2051