Effects of multibranching on 3-hydroxyflavone-based chromophores and the excited-state intramolecular proton transfer dynamics

Article abstract of DOI:10.1021/jp105542z

A series of one-, two-, and three-branched chromophores based on 3-hydroxyflavones (1-3) have been synthesized as the first example of multibranched chromophores demonstrating excited-state intramolecular proton transfer (ESIPT). Coupling between the 3-hy

Full text of DOI:10.1021/jp105542z

10412
J. Phys. Chem. A 2010, 114, 10412–10420
Effects of Multibranching on 3-Hydroxyflavone-Based Chromophores and the Excited-State
Intramolecular Proton Transfer Dynamics
Chao-Chen Lin, Chyi-Lin Chen, Min-Wen Chung, Yi-Ju Chen, and Pi-Tai Chou*
Department of Chemistry, National Taiwan UniVersity, No. 1, Section 4, RooseVelt Road,
Taipei 10617, Taiwan, R.O.C.
ReceiVed: June 16, 2010; ReVised Manuscript ReceiVed: August 12, 2010
A series of one-, two-, and three-branched chromophores based on 3-hydroxyflavones (1-3) have been
synthesized as the first example of multibranched chromophores demonstrating excited-state intramolecular
proton transfer (ESIPT). Coupling between the 3-hydroxyflavone branches connected by an electron-
donating triphenylamine core is manifested in the red-shifted and asymmetric absorption band of 2,
whereas the absorption of 3 is governed by the divided donor strength. Their excited-state charge-transfer
(ESCT)-coupled ESIPT dynamics is investigated via femtosecond fluorescence upconversion and is proved
to be well correlated with the ratio of normal/tautomer emission in the fluorescence spectra. For 1 and
2, with increased donor strength compared with the 4′-N,N-dialkylamino-3-hydroxyflavone analogue,
ESIPT appears to cease in the more polar solvent of acetonitrile. Nevertheless, similar dependence of
1-3 on solvent polarity signifies resembling charge-transfer character at the normal excited states (N*),
despite their varying structures. As evidenced by the theoretical approach, the frontier orbitals of
vibrationally relaxed (geometry-optimized) N*, from which fluorescence and ESIPT should take place,
are localized on one specific branch, leading to similar emission patterns and dynamics, whereas the
orbitals contributing to Franck-Condon excitation (absorption) spread over the entire molecule. The
localization is found to be facilitated by rotation of a specific branch pivoting on the central nitrogen
atom, while planarity is maintained within each 3-hydroxyflavone chromophore.
1. Introduction
(SR) of N* toward N* (the equilibrium configuration), a
eq
competing ESIPT process takes place.3 The feature can be
attributed to the resembling dipolar vectors of T* and N in
both magnitude and orientation (15°) such that prior to
3-Hydroxyflavone has long been a prototypical case of
excited-state intramolecular proton transfer (ESIPT).1 Upon
S0 f S1 (ππ*) excitation (λmax ∼340 nm in cyclohexane),
ultrafast ESIPT is executed (<100 fs), resulting in the steady-
state observation of only a largely Stokes-shifted tautomer
emission band. Substitution at the 4′ position with a dialkyl-
amino group renders an ideal example for studying excited-
state charge-transfer (ESCT)-coupled ESIPT,2,3 along with
other systems, such as 7-N,N-diethylamino-3-hydroxyflavone4
and p-N,N-ditolylaminosalicylaldehyde.5 The amino group
serves as an electron donor, of which the lone pair electrons
can be viewed as being resonated to the carbonyl oxygen
upon optical excitation, creating a large dipole moment for
the normal excited state (N*) relative to the tautomer excited
state (T*) and their respective ground states (N and T). From
the theoretical work of Hynes and co-workers,6 equilibrium
between the moving proton and the surrounding solvent
molecules is established at any instant so that the reaction
activation energy (∆G†) is determined by solvent reorganiza-
tion rather than a proton migration barrier and, hence, the
adoption of solvent orientation (polarization) instead of proton
position as the primary reaction coordinate. Moreover, N*
is stabilized more as the solvent polarity is increased, thereby
forming a larger solvent-induced barrier. For example, the
barrier of 4′-N,N-dimethylamino-3-hydroxyflavone in dichlo-
romethane is estimated to be 4.6 kcal/mol.7 Even more
intriguingly, it is discovered via the analysis of temporal
spectral evolution within 3 ps that during solvent relaxation
complete solvent relaxation (reorientation) to N* , the mol-
eq
ecule undergoes a possibly adiabatic type of proton transfer
(∼1.5 ps) in the absence of a solvent-induced barrier (Scheme
1).3 It is also noteworthy that recent experimental advances
in excited-state intramolecular proton-coupled electron trans-
fer (PCET) reactions have been reviewed by Hsieh et al.8
In continuation of the effort, substitution at the 4′ position is
further replaced with the diphenyl amino group (1), in hopes
of increasing the donor strength. In the meantime, the resulting
segment of a triphenylamine moiety suffices for an electron
donor that has been used extensively as a building block for
multibranched and dendritic structures.9-11 On the basis of the
triphenylamine donor core and the dipolar nature of 4′-N,N-
diphenylamino-3-hydroxyflavone, the corresponding series of
two- (V-shaped) and three-branched (trigonal) quadrupolar and
octupolar molecules are crafted (2 and 3; see Scheme 2) to
explore the effect of branching on steady-state spectra as well
as on the ESIPT dynamics. Throughout this study, the electronic
coupling between branches can be clearly resolved in the
absorption spectra of 2 and 3.
More significantly, as we demonstrate for the first time the
integration of ESIPT into multibranched chromophores, the
emission spectra and related dynamics are studied in various
solvents to illustrate the impact of solvent polarity on the ESIPT
rate. Similar solvent-dependent behavior indicates that the
normal excited states (N*), where proton transfer is initiated,
* Corresponding author. E-mail: chop@ntu.edu.tw.
10.1021/jp105542z  2010 American Chemical Society
Published on Web 09/07/2010

3-Hydroxyflavone-Based Chromophores
J. Phys. Chem. A, Vol. 114, No. 38, 2010 10413
SCHEME 1: The Proposed Relaxation Mechanisma of 4′-N,N-Dialkylamino-3-hydroxyflavone in Polar Solventsa
a See text for the definition of states and relaxation processes. Also note that a pyrylium cationic structure is commonly drawn for the
3-hydroxyflavone tautomer.
SCHEME 2: Synthesis of Compounds 1-3
possess comparable dipolar vectors, despite the varying struc-
tures. It has been elucidated with quantum-chemical excited-
state calculations that geometry optimization gives rise to the
localization of frontier orbitals on one branch rather than being
delocalized over the entire molecule.9 Compared to multi-
branched chromophores incorporating trans-stilbene linkage,
which is intrinsically twisted at the ground state,9,10a,c,e through
the utilization of compact planar 3-hydroxyflavone moieties, it
is revealed that rotation of a whole single branch during
vibrational relaxation causes the localization of emission
chromophore.
KOH solution (2.9 equiv). The mixture was stirred at 50 °C for
3 h, cooled to room temperature, and then stayed overnight.
The obtained chalcone salts were poured into water and
neutralized to pH 6-7 with 1 M HCl; and the precipitates, which
could be crystallized in EtOH, were subsequently filtered,
washed with EtOH, and dried. Second, oxidative cyclization
was conducted with addition to aqueous 30% H2O2 (8-11
equiv) and 4 M NaOH (5.0 equiv) in a 1:1 mixture of EtOH
and THF (20 mL/mmol chalcone) at 0 °C. The reaction mixture
was stirred at room temperature overnight, then acidified again
with 1 M aqueous HCl and filtered off to yield the corresponding
chromones 1-3.
All reactions were performed under nitrogen. Solvents were
distilled from appropriate drying agents prior to use. Com-
mercially available reagents were used without further purifica-
tion. All reactions were monitored by TLC with Merck
precoated glass plates (0.20 mm with fluorescent indicator
UV254) and were visualized with UV light irradiation at 254/
366 nm. Flash column chromatography was carried out using
2. Experimental Section
2.1. Synthesis. Scheme 2 depicts the synthetic procedures
for compounds 1-3, in which the first step involved the
formation of chalcones via Claisen-Schmidt condensation12 of
commercially available compounds 4, 5, and 6 with o-hydroxy-
acetophenone (1.05 equiv, 6 mL/mmol acetophenone) in a
mixture of 25 mL of dried EtOH and 5 mL of 50% aqueous

10414 J. Phys. Chem. A, Vol. 114, No. 38, 2010
Lin et al.
silica gel from Merck (230-400 mesh). The 1H and 13C NMR
spectra were obtained on a 400 MHz Bruker spectrometer, and
the chemical shifts were reported relative to CDCl3. Mass spectra
were obtained on a JEOL SX-102A instrument operating in
electron impact (EI) mode. Elementary analyses were performed
on a CHN analyzer.
2-(4-(Diphenylamino)phenyl)-3-hydroxy-4H-chromen-4-
one (1). 1H NMR δ 8.23-8.20 (dd, J ) 21 Hz, 1H), 8.12-8.09
(m, 2H), 7.68-7.63(m, 1H), 7.54-7.52(d, J ) 21 Hz, 1H),
7.40-7.36 (m, 1H), 7.31-7.27 (m, 4H), 7.16-7.07 (m, 8H).
13C NMR: δ 172.7, 146.6, 137.5, 133.1, 129.5, 129.3, 128.9,
126.1, 125.4, 125.2, 124.2, 124.0, 123.3, 122.0, 121.0, 119.2,
118.0. EI-MS (MH+): 406.14. Anal. Calcd for C27H19NO3: C,
79.98; H, 4.72; N, 3.45; O, 11.84. Found: C, 79.95; H, 4.70; N,
3.47; O, 11.81.
2,2′-(4,4′-(Phenylazanediyl)bis(4,1-phenylene))bis(3-hydroxy-
4H-chromen-4-one) (2). 1H NMR δ 8.24-8.22 (d, J ) 21 Hz,
2H), 8.18-8.16 (m, 4H), 7.69-7.66 (m, 2H), 7.56-7.54 (d, J
) 21 Hz, 2H), 7.41-7.33 (m, 4H), 7.26-7.17 (m, 4H), 6.99
(br, 3H). 13C NMR: δ 172.8, 155.1, 148.4, 143.1, 137.7, 133.3,
131.6, 129.6, 128.8, 126.1, 125.3, 124.9, 124.3, 123.5, 122.8,
120.6, 118.0. EI-MS (MH+): 566.15. Anal. Calcd for
C36H23NO6: C, 76.45; H, 4.10; N, 2.48; O, 16.97. Found: C,
76.48; H, 4.12; N, 2.42; O, 16.93.
2,2′,2′′-(4,4′,4′′-Nitrilotris(benzene-4,1-diyl))tris(3-hydroxy-
4H-chromen-4-one) (3). 1H NMR δ 8.27-8.23 (m, 6H),
7.81-7.79 (d, J ) 22 Hz, 3H), 7.71-7.58 (m, 3H), 7.45-7.41
(m, 3H), 7.36-7.28 (m, 6H). 13C NMR: δ 172.6, 155.3, 133.6,
131.3, 129.2, 128.9, 125.5, 124.9, 124.5, 123.6, 122.6, 120.6,
118.1. EI-MS (MH+): 725.16. Anal. Calcd for C45H27NO9: C,
74.58; H, 3.62; N, 1.93; O, 19.87; Found: C, 74.57; H, 3.60;
N, 1.92; O, 19.85.
half-maximum (fwhm) of ∼150 fs, which was chosen as a
response function of the system. A waveplate was placed in
the pump beam path to ensure that the polarization was set at
the magic angle (54.7°) with respect to that of the gate pulse to
eliminate fluorescence anisotropy.
2.3. Computational Methodology. Ground-state and the
lowest singlet excited-state (S1) geometries were optimized by
density functional theory (DFT) and analytic time-dependent
DFT (TDDFT)15 with the B3LYP16 hybrid functional, which
allows for DFT-quality optimizations in the excited state. The
6-31G(d)17 basis set was employed for all atoms. Vibrational
frequencies were then calculated on the basis of their optimized
geometries to verify that each of the calculated geometies is
the globally optimized structure. Time-dependent DFT (TD-
DFT)18 calculations using the B3LYP functional were then
performed on the basis of the optimized structures at ground
states to probe the absorptive properties. Typically, the lowest
10 singlet roots of the nonhermitian eigenvalue equations were
obtained to determine the vertical excitation energies. Oscillator
strengths were deduced from the dipole transition matrix
elements. All calculations were carried out using Gaussian 09
program.19
3. Results and Discussion
3.1. Steady-State Absorption Spectra. The steady-state
absorption spectra of compounds 1-3 in various solvents are
shown in Figure 1, and the extinction coefficients in dichlo-
romethane are plotted in Figure 2. Considering the low
ionization energy for the amino substituent, the S0 f S1
excitation of 1 should be ascribed to the charge-transfer
transition from diphenylamine (electron donor) to the carbonyl
oxygen (electron acceptor).3 Vibronic structure is observed in
cyclohexane, conveying that the charge-transfer transition is ππ*
by nature, as in the parent 3-hydroxyflavone molecule.
2.2. Spectroscopic Measurements. Steady-state absorption
and emission spectra were recorded by a Hitachi (U-3310)
spectrophotometer and an Edinburgh (FS920) fluorometer,
respectively, for which the wavelength-dependent response had
been corrected. Nanosecond lifetime studies were performed
with an Edinburgh FL 900 photon-counting system equipped
with a hydrogen-filled lamp as the excitation source. Coumarin
480 (λem ) 471 nm, Exciton) in ethanol, with a Φ of 0.95, served
as the reference for measuring emission quantum yields.13a
Emission quantum yields were calculated using
Meanwhile, in polar solvents, the vibronic feature is concealed
by inhomogeneous broadening. Going from 1 to 3, extinction
coefficients at the first absorption peak (εmax; see Table 1) in
dichloromethane are in the ratio of 1.0:1.8:2.4, which is not
strictly proportional to the number of branches, implying that
upon absorption, the transitions are delocalized among branches
for 2 and 3. The delocalization, or coupling of branching
chromophores, is also evident in the peak wavelength and shape
of the lowest energy absorption bands for 2. Obviously, the first
absorption peak (λmax) of 2 (429 nm) is red-shifted from that of
1 (404 nm), which is attributed to coupling between the S1 states
of each separate branch, resulting in one stabilized and one
destabilized state. These states eventually become the new S1
(represented as S′) and S′, being slightly higher in energy (vide
Is Ar ns2
Φs ) Φ
(1)
r Ir A
2
s nr
by comparing the integrated areas (I, where the subscripts s and
r denote sample and reference, respectively) under the fluores-
cence curves with similar absorbance (A) at the same excitation
wavelength and were corrected for the refractive indices (n) of
the solvents used.13b
The fluorescence upconversion measurements were performed
using a femtosecond optically gated system (FOG-100, CDP),
details of which have been described in previous reports.14
Briefly, the fundamental of a Ti:sapphire laser (Tsunami, Spectra
Physics) at 800 nm was used to produce second harmonics (SH)
as the excitation source. The resulting fluorescence and the
optical delayed remaining fundamental pulses were collected
and focused on a BBO type-I crystal (0.5 mm) for sum-
frequency generation. The upconverted signal was then separated
by an F/4.9 (f ) 380 mm) monochromator and detected via a
photon-counting PMT (R1527P, Hamamatsu). The cross-cor-
relation between SH and the fundamental had a full width at
1
2
infra). Moreover, the observed absorption band of 2 is distinc-
tively asymmetric owing to the differing transition dipole
moments of S1′ and S2′.9 This viewpoint can be supported by
the computed oscillator strengths listed in Table 2 (f ) 0.93
and 0.18 for S1′ and S′2, respectively).
In stark contrast, λmax of 3 is blue-shifted to 396 nm,
suggesting that multibranching leads to not only coupling
between states but also other effects. Since for 4′-N,N-
dialkylamino-3-hydroxyflavone or 1, the HOMO resides at the
amino group, as multibranching is introduced, the donating
strength is divided, and the HOMO energy level should be
lowered, resulting in a larger S0-S1 energy gap. Therefore,
multibranching introduces two opposing effects; that is, splitting
of energy levels, which decreases the S0-S1 energy gap; and
weakening of the donor strength, which increases the S0-S1
energy gap. The impact of weakened donor strength should be

3-Hydroxyflavone-Based Chromophores
J. Phys. Chem. A, Vol. 114, No. 38, 2010 10415
Figure 1. Static absorption and fluorescence spectra of (A) 1, (B) 2, and (C) 3 in cyclohexane, benzene, dichloromethane, and acetonitrile at 298
K.
Figure 2. Molar extinction coefficients of compounds 1-3 in
dichloromethane.
TABLE 1: Absorption Properties of Compounds 1-3 in
Dichloromethane
λmax (nm)
εmax (M-1 cm-1
)
1
2
3
404
429
396
2.33 × 104
4.32 × 104
5.53 × 104
Figure 3. Optimized ground state geometry of compounds 1-3.
TABLE 2: Calculated Energy Levels, Oscillator Strengths
(f), and Orbital Transition Analyses of Lower-Lying
Transitions (Absorption) for Compounds 1-3
state
λcal (nm)a
fb
assignments
dipolar branch maintains planarity in 2 and 3. The intact
3-hydroxyflavone chromophore with dihedral angle between the
phenyl group and the chromenone moiety of <0.5°, compared
with nonplanar (twisted) branches,9,10 raises the S1 energy of 3
(represented as S′1′) to an extent that trumps the stabilizing
interaction due to coupling between S1 states on each branch.
Conversely, the latter effect dominates in 2. It is also worthy to
note that this trend is not reflected in the calculated absorption
wavelength (λcal) using TDDFT (Table 2). Nevertheless, it can
be seen that the coupling effect is less obvious going from 2 to
3. On the other hand, coupling among branches in 3 produces
two degenerate lowest-lying states, S′1′, with f ) 0.75 and S2′′
with f ) 0.76, and one higher lying state, S′3′ with vanishing
oscillator strength (f ) 0.0007).9 Thus, the observed first
absorption band of 3 returns to being symmetric, as opposed to
that of 2, indicative of a 3-fold symmetry axis.
1
2
S1
S1′
S2′
S1′′
S2′′
S3′′
431.1
462.6
421.3
468.6
466.9
411.4
0.6465
0.9256
0.1793
0.7507
0.7582
0.0007
HOMO f LUMO (99%)
HOMO f LUMO (99%)
HOMO f LUMO+1 (97%)
HOMO f LUMO (97%)
HOMO f LUMO+1 (97%)
HOMO f LUMO+2 (95%)
3
a Calculated absorption onset. b Oscillator Strength.
more prominent in our case of compounds 1-3 because of the
planarity and, hence, better electron-withdrawing ability through
the intact conjugation of each branch (Figure 3); that is, the
3-hydroxyflavone chromophore. With the triphenylamine core,
1-3 all exhibit propeller-like geometry, in which each branch
is twisted with respect to the plane defined by the central
nitrogen atom and the three connecting carbons, whereas each

10416 J. Phys. Chem. A, Vol. 114, No. 38, 2010
Lin et al.
TABLE 3: Emission and Relaxation Dynamics of Compounds 1-3 in Selected Solvents
solvent
emission λmax (Φ)
early dynamics (ps)a
population decay (ns)
1
cyclohexane
N: 434 nm (0.0070)
T: 568 nm (0.52)
N: 467 nm (0.039)
T: 572 nm (0.52)
N: 513 nm (0.11)
T: 589 nm (0.54)
N: 590 nm (0.14)
480 nm [τ1: 7.67 (0.98)]
650 nm [τ1: 7.28 (-0.55)]
480 nm [τ1: 23.3 (0.99)]
2.96
2.89
3.56
1.44
3.02
2.94
3.35
2.89
benzene
650 nm [τ1: 23.9 (-0.46)]
dichloromethane
acetonitrile
480 nm [τ1: 3.18 (0.69), τ2: 97.1 (0.30)]
650 nm [τ1: 2.94 (-0.19), τ2: 98.5 (-0.22)]
480 nm [τ1: 0.96 (0.92)]
650 nm [τ1: 0.51 (-0.50)]
480 nm [τ1: 13.7 (0.85)]
2
3
cyclohexane
dichloromethane
cyclohexane
dichloromethane
N: 444 nm (0.052)
T: 568 nm (0.49)
N: 513 nm (0.18)
T: 583 nm (0.44)
N: 448 nm (0.018)
T: 565 nm (0.52)
N: 484 nm (0.12)
T: 581 nm (0.52)
480 nm [τ1: 2.48 (0.85), τ2: 259 (0.10)]
480 nm [τ1: 10.1 (0.88)]
480 nm [τ1: 1.57 (0.51), τ2: 88.3 (0.37)]
a Numbers in parentheses are the fitted preexponential factors, with negative values indicating rise time constants. Note that the sum of
preexponential factors may not be equal to 1 due to exclusion of the longer population decay component.
3.2. Fluorescence Spectra and Femto-Picosecond Relax-
ation Dynamics. Fluorescence spectra of compounds 1-3 are
depicted in Figure 1, and the relevant data as well as detailed
fitting parameters for relaxation dynamics are listed in Table
3. In cyclohexane, largely Stokes-shifted tautomer emission
centering around 570 nm predominates, and the charge-
transfer normal emission around 440 nm is still visible,
implying fast but finite ESIPT in nonpolar solvent. The
emission quantum yields of all compounds (e.g., Φ ) 0.52
for 1 in cyclohexane) are more than twice the value of 4′-
N,N-diethylamino-3-hydroxyflavone (e.g., Φ ) 0.21 in cy-
clohexane),3 accompanied by longer population decay times;
for instance, 2.96 ns for 1 relative to 1.28 ns for 4′-N,N-
diethylamino-3-hydroxyflavone in cyclohexane. This en-
hancement can be rationalized with suppressed rotation of
the phenyl rings caused by steric hindrance and thus a reduced
nonradiative decay rate (knr), whereas the motions of flexible
alkyl groups channel into nonradiative deactivation.
and acetonitrile. The associated kinetics for [N*] and [T*] are
outlined as follows.
[N*(t)] ) [N*(0)] B e-t/τ + B2e-t/τ
1
2
(
)
1
(3)
(4)
-t/τ1
+ e-t/τ
2
(
)
[T*(t)] ) [N*(0)]B1 -e
1
τ1
) k1 = kPT + k-PT
kPT
k-PT
1
τ2
) k2 ) kT
+ kN
(
)
(
)
kPT + k-PT
kPT + k-PT
Take 1 as a prototype, when the solvent polarity is increased,
the ratio of normal emission increases correspondingly and shifts
to longer wavelength, until in acetonitrile, only one emission band
remains. Supposedly, ESIPT does not occur in acetonitrile for 1,
which can be confirmed with the absence of proton transfer
dynamics in addition to solvent relaxation (SR) and the population
decay (vide infra). With femtosecond fluorescence upconversion,
the dynamics of normal emission at 480 nm for 1 in cyclohexane
(Figure 4) is determined to be composed of a 7.67 ps fast decay
(98%) and a residual long-lived component.
kPT
B1 =
kPT + k-PT
(5)
k-PT
B2 =
kPT + k-PT
The relations hold true if the forward and reverse rates of
ESIPT, kPT and k-PT, are much faster than the population decay
rates for both states, kN and kT. The same phenomenon is
observed in benzene for 1, with a 23.3 ps fast decay (99%) at
480 nm normal emission and a correlated 23.9 ps rise (-46%)
at 650 nm tautomer emission, plus similar population decays
for the two bands. Nonetheless, k-PT for 1 in both cyclohexane
and benzene is relatively slow compared to kPT so that B2 (see
eq 5) is minute (Table 3).
Relaxation dynamics in dichloromethane is drastically dif-
ferent in that at least three time constants are apparently needed
(Figure 4) to achieve good convoluted fits. For example, upon
monitoring at the 480 nm normal emission of 1, two fast decays,
τ1 ) 3.18 ps (69%) and τ2 ) 97.1 ps (30%), are resolved in
On the other hand, the tautomer emission dynamics at 650
nm consists of a 7.28 ps rise time constant (-55%) plus a long
population decay. After acquiring the population decays, it is
found that the two time constants (τ1 and the population decay)
for tautomer emission resemble those of normal emission within
experimental error. Bringing in the fact that the magnitudes of
the two pre-exponential factors for the tautomer emission are
roughly the same, 1 follows the same kinetic model as
4′-N,N-dialkylamino-3-hydroxyflavone,2,3 in which fast equi-
librium is established between normal (N*) and tautomer (T*)
excited states in more polar solvents, such as dichloromethane

3-Hydroxyflavone-Based Chromophores
J. Phys. Chem. A, Vol. 114, No. 38, 2010 10417
by similar dipolar vectors of the tautomer excited state (T*)
and the normal form ground state (N). Accordingly, at early
times subsequent to excitation, the solvent orientation (polariza-
tion) remains favorable for N and T*, allowing execution of
ultrafast ESIPT prior to the formation of a solvent-induced
barrier. At times later than 30 ps, when solvent relaxation is
nearly complete, the spectral feature displays a decrease in
normal emission and an increase in tautomer emission, which
can be rationalized by the slower ESIPT rate due to different
polarizations between equilibrated normal (N* ) and tautomer
eq
(T* ) excited states (Scheme 1) and the existence of a solvent-
eq
induced barrier.
The most significant distinction between 1 and 4′-N,N-
diethylamino-3-hydroxyflavone lies in that only one emission
band exists in acetonitrile for 1, presumably the normal emission,
signifying the prohibition of ESIPT. This conclusion is validated
with the observations of ultrafast solvent relaxation (e.g. 0.96
ps decay at 480 nm versus 0.51 ps rise at 650 nm) and
population decay (1.44 ns) only, with no intermediate kinetic
processes. Such results act in accordance with the increased
donating strength of -NPh2 relative to -NR2, which on one
hand creates a larger dipole moment in the charge-transfer
normal excited state (N*), shifting the equilibrium position of
N* farther away from that of T* and in turn resulting in a larger
eq
eq
solvent induced barrier, so that the rate of ESIPT could not
compete with other relaxation pathways of N*.8,20
Alternatively, it may indicate that N* of 1 is stabilized more
than its dialkylamino analogue, such that N* is much lower in
eq
energy than T* , leading to thermally unfavorable N* f T*
eq
eq
eq
ESIPT. Note that the prohibition of ESIPT is also observed for
2 in acetonitrile. Nevertheless, ESIPT recovers for 3 in aceto-
nitrile, which is supported by the steady-state observation of
dual emission, indirectly hinting the greater influence of
weakened donor strength in 3 due to multibranching, as
proposed in earlier sections.
Last but not least, the dynamics inspected with fluorescence
upconversion for the whole series of compounds 1-3 is always
consistent with the steady-state measurements. In other words,
a faster ESIPT rate is associated with a minor proportion of the
normal emission. For instance, the ESIPT dynamics of 1-3 in
cyclohexane (Figure 5) is on the order of 1 (7.67 ps-1) > 3
(10.1 ps-1) > 2 (13.7 ps-1), with 1 having the smallest ratio of
normal emission and 2 having the largest. In dichloromethane
(Figure 6), ESIPT dynamics is 3 (88.3 ps-1) > 1 (97.1 ps-1) >
2 (259 ps-1), consistent with the trend of an increasing normal
emission ratio.
Figure 4. Time-resolved sum frequency signal of fluorescence at (o)
480 and (∆) 650 nm with the gate pulse (800 nm) for 1 in (A)
cyclohexane, (B) benzene, (C) dichloromethane, and (D) acetonitrile.
Solid lines are the best-fitting curves.
Overall, similar ESIPT behavior of 1-3 suggests that the
normal excited states (N*) in which ESIPT originates possess
analogous dipolar vectors. Nonetheless, this disagrees with the
consequence of electronic coupling between branches in 2 and
3, which may lead to cancellation of dipole moments owing to
the relatively more symmetric V-shape and trigonal structures.
To resolve this contradiction, theoretical approaches are then
pursued to sort out how the delocalized Franck-Condon states
addition to the longer population decay, with τ1 ascribed to SR
and τ2 to ESIPT. Solvent relaxation is definitely expected for
N* owing to its significant charge separation. Interestingly, two
rises with similar time scales, τ1 ) 2.94 ps (-19%) and τ2 )
98.5 ps (-22%) are obtained at 650 nm tautomer emission, even
though the dipole moment of T* was estimated to be somewhat
smaller.3
Through analysis of the temporal spectral evolution of 4′-
N,N-diethylamino-3-hydroxyflavone,3 it was discovered that the
tautomer emission band emerged alongside the continuously red-
shifting charge-transfer normal emission band within ∼1 ps,
hinting an ESIPT pathway that is occurring simultaneously with
solvent relaxation. This ultrafast ESIPT process is promoted
*
(NFC) of 1-3 generated upon absorption evolves into states with
resembling dipolar vectors through vibrational relaxation.
3.3. Theoretical Approaches. The frontier orbitals of com-
pounds 1-3 for absorption (Franck-Condon excitation) are
calculated first with TDDFT and those of the lowest-lying
excited states (S1 for 1, S1′ for 2, and S′1′ for 3; see Table 2) are
illustrated in Figure 7. These states, in particular, can be well-
represented as transitions between a pair of orbitals with >97%
contribution. Delocalization among branches is unambiguous
in 2 and 3, with HOMO being located primarily around the

10418 J. Phys. Chem. A, Vol. 114, No. 38, 2010
Lin et al.
Figure 6. Time-resolved sum-frequency signal of fluorescence at 480
nm with the gate pulse (800 nm) for (A) 1, (B) 2, and (C) 3 in
dichloromethane.
Figure 5. Time-resolved sum-frequency signal of fluorescence at 480
nm with the gate pulse (800 nm) for (A) 1, (B) 2, and (C) 3 in
cyclohexane.
supplement, the dipole moments of the emitting states for 1-3
are superimposed in Figure 7, and their magnitudes are estimated
to be 3.4, 6.8, and 5.9 D, respectively. With the dipolar vectors
evidently not canceled out for 2 and 3 despite their symmetric
structures, it is verified that similar solvent-dependent ESIPT
is executed at the localized S1 states created after vibrational
relaxation.
triphenylamine core and LUMO onto all of the 3-hydroxyflavone
chromophores. For 3, owing to degeneracy of the lowest-lying
excited states, the LUMO should be a linear combination of
two orbitals, each falling on two dipolar branches. On the other
hand, after geometry optimization at S′1 and S1′′, designated as
the emission states, in which ESIPT originates and competes
with the normal emission, the frontier orbitals are found to be
funneled into specific branches. The localization is a conse-
quence of vibrational relaxation, which is also perceived in the
Frenkel exciton model.21 As a matter of fact, localized emitting
states have been explored in several multibranched systems; for
example, trigonal structures with triphenylamine or triphenyl-
benzene cores.22
In the case of 2 and 3, vibrational relaxation is clearly
manifested in the rotation of one single branch, with the electron
configuration adjusting instantaneously such that LUMO resides
solely on the rotated branch, while HOMO is on another
unchanged branch including the central N atom. Presumably,
in the emitting state, the nitrogen nonbonding orbital resonates
efficiently only with the unchanged 3-hydroxyflavone chro-
mophore, constituting an effective electron-donating segment.
In contrast, in the absorbing state, which adopts the optimized
propeller-like geometry of the ground state (Figure 3), the
nitrogen lone pair is equally shared by all branches. As a
4. Conclusion
In this work, we have successfully designed a series of one-
(dipolar), two- (quadrupolar), and three-branched (octupolar)
molecules 1-3 bearing triphenylamine as the electron donor
core and 3-hydroxyflavone chromophores at the para positions
as accepting branches. Akin to the previously studied 4′-N,N-
dialkylamino-3-hydroxyflavone, excited-state charge-transfer
coupled excited-state intramolecular proton transfer is demon-
strated via emission spectra and the corresponding dynamics.
When the solvent polarity is increased, the ESIPT rate decreases
owing to a larger solvent-induced barrier. The initial 1-3 ps
formation time of the tautomer emission observed in dichlo-
romethane is facilitated by the resembling dipolar vectors of
the normal form ground state (N) and the tautomer excited state
(T*) before and during SR. In acetonitrile, the solvent-induced
barrier is too large to allow for ESIPT in 1 and 2, and only the
normal emission bands are visible.

3-Hydroxyflavone-Based Chromophores
J. Phys. Chem. A, Vol. 114, No. 38, 2010 10419
Figure 7. Frontier orbitals of compounds 1-3 for absorption and emission at the respective S1 states. Brown arrows indicate dipolar vectors at the
emission states.
Second, delocalization of excitation or electronic coupling
between branches is clearly shown in the absorption spectra.
Moreover, with theoretical approaches, it is revealed that upon
vibrational relaxation at the S1 states, rotation of one single
branch fosters the localization of transition orbitals, with LUMO
located merely on the rotated 3-hydroxyflavone chromophore
and HOMO on another unchanged branch. This localization
results in similar ESIPT behavior for 1-3 despite their differing
structures. In view of the application, we have provided a new
design strategy for quadrupolar or octupolar two-photon absorb-
ing (TPA) chromophores, especially in the fields of two-photon
imaging,23,24 and the related TPA studies are currently in
progress.
References and Notes
(1) (a) Sengupta, P.; Kasha, M. Chem. Phys. Lett. 1979, 68, 382. (b)
Strandjord, A. J. G.; Courtney, S. H.; Friedrich, D. M.; Barbara, P. F. J.
Phys. Chem. 1983, 87, 1125. (c) McMorrow, D.; Kasha, M. J. Phys. Chem.
1984, 88, 2235. (d) Schwartz, B. J.; Peteanu, L. A.; Harris, C. B. J. Phys.
Chem. 1992, 86, 3591. (e) Ameer-Beg, S.; Ormson, S. M.; Brown, R. G.;
Matousek, P.; Towrie, M.; Nibbering, E. T. J.; Foggi, P.; Neuwahl, F. V. R.
J. Phys. Chem. A 2001, 105, 3709.
(2) (a) Shynkar, V. V.; Me´ly, Y.; Duportail, G.; Pie´mont, E.; Klym-
chenko, A. S.; Demchenko, A. P. J. Phys. Chem. A 2003, 107, 9522. (b)
Swinney, T. C.; Kelley, D. F. J. Chem. Phys. 1993, 99, 211. (c) Douhal,
A.; Sanz, M.; Carranza, M. A.; Organero, J. A.; Santos, L. Chem. Phys.
Lett. 2004, 394, 54. (d) Ameer-Beg, S.; Ormson, S. M.; Poteau, X.; Brown,
R. G.; Foggi, P.; Bussotti, L.; Neuwahl, F. V. R. J. Phys. Chem. A 2004,
108, 6938.
(3) Chou, P.-T.; Pu, S.-C.; Cheng, Y.-M.; Yu, W.-S.; Yu, Y.-C.; Hung,
When branches capable of ESIPT are integrated, the interfer-
ing autofluorescence can be bypassed, since there is a large
Stokes shift between the absorption and tautomer emission. If
the conjugation is further elongated, the largely Stokes-shifted
tautomer emission could even be tuned to near-infrared,
providing better penetration without a large increase in molecular
weight. Furthermore, via coupling with ESCT, emission spectra
of the fabricated TPA chromophores are essentially sensitive
to the local solvent polarity or hydrogen bonding environments,
serving as prospective molecular probes.
F.-T.; Hu, W.-P. J. Phys. Chem. A 2005, 109, 3777.
(4) Cheng, Y.-M.; Pu, S.-C.; Yu, Y.-C.; Chou, P.-T.; Huang, C.-H.;
Chen, C.-T. J. Phys. Chem. A 2005, 109, 11696.
(5) Chou, P.-T.; Yu, W.-S.; Cheng, Y.-M.; Pu, S.-C.; Yu, Y.-C.; Lin,
Y.-C.; Huang, C.-H.; Chen, C.-T. J. Phys. Chem. A 2004, 108, 6487.
(6) (a) Kiefer, P. M.; Hynes, J. T. J. Phys. Chem. A 2002, 106, 1834.
(b) Kiefer, P. M.; Hynes, J. T. J. Phys. Chem. A 2002, 106, 1850.
(7) Roshal, A. D.; Organero, J. A.; Douhal, A. Chem. Phys. Lett. 2003,
379, 53.
(8) Hsieh, C.-C.; Jiang, C.-M.; Chou, P.-T. Acc. Chem. Res. 2010, DOI:
10.1021/ar1000499.
(9) Katan, C.; Terenziani, F.; Mongin, O.; Werts, M. H. V.; Porre`s,
L.; Pons, T.; Mertz, J.; Tretiak, S.; Blanchard-Desce, M. J. Phys. Chem. A
2005, 109, 3024.
(10) (a) Chung, S.-J.; Kim, K.-S.; Lin, T.-C.; He, G. S.; Swiatkiewicz,
J.; Prasad, P. N. J. Phys. Chem. B 1999, 103, 10741. (b) He, G. S.;
Swiatkiewicz, J.; Jiang, Y.; Prasad, P. N.; Reinhardt, B. A.; Tan, L.-S.;
Acknowledgment. We thank the National Science Council
of Taiwan for financial support and the National Center for
High-Performance Computing for computer time and facilities.

10420 J. Phys. Chem. A, Vol. 114, No. 38, 2010
Lin et al.
Kannan, R. J. Phys. Chem. A 2000, 104, 4805. (c) Porre`s, L.; Mongin, O.;
Katan, C.; Charlot, M.; Pons, T.; Mertz, J.; Blanchard-Desce, M. Org. Lett.
2004, 6, 47. (d) Yoo, J.; Yang, S. K.; Jeong, M.-Y.; Ahn, H. C.; Jeon,
S.-J.; Cho, B. R. Org. Lett. 2003, 5, 645. (e) Kato, S.-i.; Matsumoto, T.;
Shigeiwa, M.; Gorohmaru, H.; Maeda, S.; Ishi-i, T.; Mataka, S. Chem.sEur.
J. 2006, 12, 2303.
(11) (a) Drobizhev, M.; Karotki, A.; Rebane, A.; Spangler, C. W. Opt.
Lett. 2001, 26, 1081. (b) Adronov, A.; Fre´chet, J. M. J.; He, G. S.; Kim,
K.-S.; Chung, S.-J.; Swiatkiewicz, J.; Prasad, P. N. Chem. Mater. 2000,
12, 2838. (c) Varnavski, O.; Yan, X.; Mongin, O.; Blanchard-Desce, M.;
Goodson, T., III. J. Phys. Chem. C 2007, 111, 149.
(12) (a) Chou, P. T.; Martinez, M. L.; Clements, J. H. J. Phys. Chem.
1993, 97, 2618. (b) Chou, P. T.; Martinez, M. L.; Clements, J. H. Chem.
Phys. Lett. 1993, 204, 395.
(13) (a) Jones, G.; Jackson, W. R.; Choi, C. Y.; Bergmark, W. R. J.
Phys. Chem. 1985, 89, 294. (b) Lakowicz, J. R. Principles of Fluorescence
Spectroscopy; Plenum Press: New York, 1983.
(14) (a) Chou, P. T.; Chen, Y. C.; Yu, W. S.; Chou, Y. H.; Wei, C. Y.;
Cheng, Y. M. J. Phys. Chem. A 2001, 105, 1731. (b) Chou, P. T.; Yu,
W. S.; Cheng, Y. M.; Pu, S. C.; Yu, Y. C.; Lin, Y. C.; Huang, C. H.; Chen,
C. T. J. Phys. Chem. A 2004, 108, 6487.
(15) (a) Furche, F.; Ahlrichs, R. J. Chem. Phys. 2002, 117, 7433. (b)
Scalmani, G.; Frisch, M. J.; Mennucci, B.; Tomasi, J.; Cammi, R.; Barone,
V. J. Chem. Phys. 2006, 124, 094107, 1-15.
(16) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (b)
Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.;
Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.;
Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.;
Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.;
Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts,
R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth,
G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas,
¨
O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.
(20) Hsieh, C.-C.; Chen, K.-Y.; Hsieh, W.-T.; Lai, C.-H.; Shen, J.-Y.;
Jiang, C.-M.; Duan, H.-S.; Chou, P.-T. ChemPhysChem 2008, 9, 2221.
(21) (a) Fidder, H.; Knoester, J.; Wiersma, D. A. J. Chem. Phys. 1991,
95, 7880. (b) Poliakov, E. Y.; Chernyak, V.; Tretiak, S.; Mukamel, S.
J. Chem. Phys. 1999, 110, 8161.
(22) (a) Katan, C.; Tretiak, S.; Werts, M. H. V.; Bain, A. J.; Marsh,
R. J.; Leonczek, N.; Nicolaou, N.; Badaeva, E.; Mongin, O.; Blanchard-
Desce, M. J. Phys. Chem. B 2007, 111, 9468. (b) Terenziani, F.; Le
Droumaguet, C.; Katan, C.; Mongin, O.; Blanchard-Desce, M. ChemPhy-
sChem 2007, 8, 723. (c) Katan, C.; Charlot, M.; Mongin, O.; Le
Droumaguet, C.; Jouikov, V.; Terenziani, F.; Badaeva, E.; Tretiak, S.;
Blanchard-Desce, M. J. Phys. Chem. B 2010, 114, 3152. (d) Lahankar, S. A.;
West, R.; Varnavski, O.; Xie, X.; Goodson, T., III; Sukhomlinova, L.;
Twieg, R. J. Chem. Phys. 2004, 120, 337. (e) Verbouwe, W.; Van der
Auweraer, M.; De Schryver, F. C.; Piet, J. J.; Warman, J. M. J. Am. Chem.
Soc. 1998, 120, 1319.
(17) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209.
(18) (a) Jamorski, C.; Casida, M. E.; Salahub, D. R. J. Chem. Phys.
1996, 104, 5134. (b) Petersilka, M.; Grossmann, U. J.; Gross, E. K. U.
Phys. ReV. Lett. 1996, 76, 1212. (c) Bauernschmitt, R.; Ahlrichs, R.;
Hennrich, F. H.; Kappes, M. M. J. Am. Chem. Soc. 1998, 120, 5052. (d)
Casida, M. E. J. Chem. Phys. 1998, 108, 4439. (e) Stratmann, R. E.;
Scuseria, G. E.; Frisch, M. J. J. Chem. Phys. 1998, 109, 8218.
(19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson,
G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.;
Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.;
(23) (a) Larson, D. R.; Zipfel, W. R.; Williams, R. M.; Clark, S. W.;
Bruchez, M. P.; Wise, F. W.; Webb, W. W. Science 2003, 300, 1434. (b)
Ko¨ler, R. H.; Cao, J.; Zipfel, W. R.; Webb, W. W.; Hanson, M. R. Science
1997, 276, 2039. (c) Denk, W.; Strickler, J. H.; Webb, W. W. Science 1990,
248, 73. (d) He, G. S.; Markowicz, P. P.; Line, P.-C.; Prasad, P. N. Nature
1999, 415, 767. (e) Helmchen, F.; Denk, W. Nat. Methods 2005, 2, 932.
(24) Velusamy, M.; Shen, J.-Y.; Lin, J. T.; Lin, Y.-C.; Hsieh, C.-C.;
Lai, C.-H.; Lai, C.-W.; Ho, M.-L.; Chen, Y.-C.; Chou, P.-T.; Hsiao, J.-K.
AdV. Funct. Mater. 2009, 19, 2388.
JP105542Z