Evidence for two-photon absorption-induced ESIPT of chromophores containing hydroxyl and imino groups

Article abstract of DOI:10.1002/cphc.201100885

This paper presents experimental and theoretical investigations into excited-state intramolecular proton transfer (ESIPT) in new chromophores with hydroxyl and imino groups under one- and two-photon excitation. The results show that internal hydrogen bond

Full text of DOI:10.1002/cphc.201100885

DOI: 10.1002/cphc.201100885
Evidence for Two-Photon Absorption-Induced ESIPT of
Chromophores Containing Hydroxyl and Imino Groups**
Fang Gao,* Xiaojuan Ye, Hongru Li,* Xiaolin Zhong, and Qi Wang[a]
This paper presents experimental and theoretical investigations
into excited-state intramolecular proton transfer (ESIPT) in new
chromophores with hydroxyl and imino groups under one-
and two-photon excitation. The results show that internal hy-
drogen bonding exhibits a remarkable influence on the maxi-
mum absorption wavelength of 2-[(4’-N,N-diethylaminodiphe-
nylethylene-4-ylimino)methyl]phenol (C1) and 2-[(4’-methoxyl-
diphenylethylene-4-ylimino)methyl]phenol (C3). Compounds
C1 and C3 exhibit well-separated dual fluorescence emission
bands under one- and two-photon excitation. The second fluo-
rescence peaks of C1 and C3 are characterized by much larger
Stokes shift than the first normal peaks (ca. 140 vs. 30 nm).
4-[(4’-N,N-Diethylaminodiphenylethylene-4-ylimino)methyl]phe-
nol (C2) and 4-[(4’-methoxyldiphenylethylene-4-ylimino)-
methyl]phenol (C4) display single emission bands with small
Stokes shifts (ca. 30 nm) in various solvents under one- and
two-photon excitation. Furthermore, the first emission maxima
of C1 and C3 are almost identical to the maximum fluores-
cence emission wavelengths of C2 and C4, respectively. These
results show that C1 and C3 can undergo ESIPT via a reason-
able six-membered ring, while there is no ESIPT in C2 and C4
under one- and two-photon excitation. Compounds C1 and C2
have larger two-photon absorption cross-sections under vari-
ous near-infrared laser frequencies tuned from 700 to 880 nm.
Molecular geometry optimization of the phototautomers (enol
and keto) was performed to analyze the experimental results.
The possibility of using these chromophores for metal ions as
chemosensors of was thoroughly investigated. In DMF C3 ex-
hibits excellent sensing responses to Zn2+ and Fe3+ ions
through a greatly increased greatly and a largely reduced
emission, respectively. In methanol disappearance of ESIPT
emission with added Zn2+ ions confirms its existence. The
binding constants of C3 with Zn2+ and Fe3+ ions in DMF are
also estimated.
1. Introduction
Excited-state intramolecular proton transfer (ESIPT) has re-
ceived considerable experimental and theoretical attention be-
cause it is regarded as a significantly important phenomenon
in chemical and biological systems.[1–6] ESIPT is achieved by an
ultrafast femtosecond photoinduced enol–keto tautomeriza-
tion process as the intramolecular hydrogen-bonded molecule
is excited; this represents a four-level cyclic reaction occurring
via five- or quasi-six-membered ring, and even in rare cases
seven-membered rings.[7–9] Such definite conformational
changes are caused by the stable enol form (E) in the ground
state and the stable keto form (K) in the excited state, resulting
in absorption from E!E* and emission from K*!K, and thus,
dual, well-separated emission bands and exceptionally and ab-
normally large Stokes shift without self-absorption; this has
been found in a range of applications, including photochromic
materials,[10] fluorescence probes,[11] and electroluminescent
materials.[12] Most reported ESIPT compounds are small, which
could be favorable for a barrierless or negligible energy barrier
E!K transition in the excited state. It is necessary to develop
new ESIPT molecules with larger chemical structures. One of
the main concerns is whether ESIPT could be inhibited by in-
creased size, so it is quite difficult to look for an enlarged
ESIPT molecule with a finite and well-defined barrier, although
this is important to gain detailed insights into the reaction po-
tential energy surfaces.[7,8] Few such ESIPT molecules have
been reported due to the great challenge.[13] To the best of our
knowledge, no strong evidence has been observed to demon-
strate two-photon absorption (TPA)-induced ESIPT so far.
An efficient strategy could be to develop enlarged TPA ESIPT
molecules if the ESIPT moiety were attached to chromophore
groups through covalent bonds. ESIPT in salicylidene methyla-
mine makes it an attractive target for constructing larger mole-
cules.[11d,14] Molecular geometry optimization showed that new
chromophores, 2-[(4’-N,N-diethylaminodiphenylethylene-4-yli-
mino)methyl]phenol (C1) and 2-[(4’-methoxyl-diphenyl-
ethylene-4-ylimino)methyl]phenol (C3), were likely to show
ESIPT with a finite and well-defined barrier through a quasi-six-
membered ring (Scheme 1),[15] in which a hydroxyl group is in
the ortho position. Strong experimental evidence of ESIPT
under one- and two-photon excitation was obtained for both
chromophores. Furthermore, the S0 and S1 electronic states of
2-[(4’-methoxyldiphenylethylene-4-ylimino)methyl]phenol (C2)
and 4-[(4’-methoxyldiphenylethylene-4-ylimino)methyl]phenol
[a] Prof. Dr. F. Gao, X. Ye, Prof. Dr. H. Li, X. Zhong, Q. Wang
College of Chemistry and Chemical Engineering
Chongqing University
Chongqing 400044 (China)
E-mail: fanggao1971@gmail.com
hrli@cqu.edu.cn
[**] ESIPT=excited-state intramolecular proton transfer.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cphc.201100885.
ChemPhysChem 2012, 13, 1313 – 1324
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1313

F. Gao, H. Li et al.
Calculation Methods
The calculations were performed with the Gaussian 03 program
package. All calculations were carried out with the 6-31G** basis
set. The geometry optimizations of enol and keto forms of C2 and
C4 for the ground electronic state (S0) were carried out with the
HF method at the DFT level using the B3LYP method,[17–19] whereas
CIS was employed to optimize the geometries of the first singlet
excited state (S1) of the phototautomers.
Although the CIS method produces reliable geometries and force
fields, it predicts excitation energies that are too high (ca. 1 eV).[17]
To correct the errors and introduce the dynamic electron correla-
tion, DFT and time-dependent DFT (TD-DFT) were performed to
predict energies at HF- and CIS-optimized geometries of S0 and S1
states, respectively, such as DFT//HF or TDDFT//CIS (denoted as
single-point calculation//optimization method).[20]
Characterization of Compounds
NMR spectroscopy (1H and 13C) was performed at room tempera-
ture on a Bruker 500 MHz spectrometer with tetramethylsilane
(TMS) as an internal standard. Elemental analysis was performed
on a CE440 elemental analysis meter from Exeter Analytical. Melt-
ing points was determined by using a Beijing Fukai melting point
apparatus.
One- and Two- Photon Optical Measurements
UV/Vis absorption spectra (1ꢁ10ꢀ5 molLꢀ1) were recorded with a
Cintra spectrophotometer. Emission spectra (1ꢁ10ꢀ6–1ꢁ
10ꢀ5 molLꢀ1) were recorded with a Shimadzu RF-531PC spectro-
fluorophotonmeter. Rhodamine 6G in ethanol (F=0.94) and qui-
nine sulfate in 0.5m H2SO4 (F=0.546)[21] were used as references
to determine the fluorescence quantum yields of the compounds
reported herein. To avoid self-quenching of the fluorescence emis-
sion, a low concentration of the compounds was used to deter-
mine the fluorescence quantum yields, according to Equa-
tion (1):[22]
Scheme 1.
(C4), with a hydroxyl group at the para position, were sur-
veyed experimentally and theoretically. The ground electronic
state (S0) was modeled with the Hartree–Fock (HF) method
and at the DFT level using the B3LYP method, whereas config-
uration interaction singlet excitation (CIS) was employed to op-
timize the geometries of the first singlet excited state (S1) of
the phototautomers.
R
n20A0 If ðlfÞdlf
Ff ¼ F0f
ð1Þ
R
n2A If0ðlfÞdlf
Owing to the presence of heteroatoms, ESIPT molecules
could be used as chemosensors for specific metal ions, such as
Na+, Zn2+, Mg2+, Fe3+, and Ni2+, which play significant roles in
biology and the environment. We also fully evaluated the pos-
sibility of using these chromophores as chemosensors for spe-
cific metal ions.
in which n0 and n are the refractive indices of the solvents, A0 and
A are the optical densities at the excitation wavelength, Ff and Ff0
are the quantum yields, and the integrals denote the area of the
fluorescence bands for the reference and sample, respectively.
Two-photon excited fluorescence spectra, pumped by a Ti:sapphire
femtosecond laser (Spectra-Physics, Tsunami mode-locked, 80 MHz,
<130 fs, average power ꢁ700 mW) tuned by steps of 20 nm in
the range 700–880 nm were recorded with an Ocean Optics
USB2000 CCD camera with a detecting range of 180–880 nm. TPA
cross-sections (s) were determined by an up-conversion fluores-
cence method using fluorescein (5ꢁ10ꢀ4 molLꢀ1) in a solution of
NaOH (0.1 molLꢀ1) as a reference sample.[23] The samples were
bubbled with N2 for 15 min to eliminate O2 before detection. TPA
cross-sections of the compounds were determined by Equations (2)
and (3):[24]
Experimental Section
Reagents and Materials
Chromphores C1, C2, C3, and C4 were synthesized in our laborato-
ry (Scheme 1). Organic solvents were obtained from Chongqing
Medical and Chemical Corporation. Other chemicals and reagents
were purchased from Aldrich, unless otherwise specified. Organic
solvents were purified and dried by using well-known standard
laboratory techniques before use.[16] Others compounds were fur-
ther purified by distillation or recrystallization before use. Zn-
(OOCCH3)2·2H2O, Mg(OOCCH3)2·4H2O, KOOCCH3, NaOOCCH3·3H2O,
LiCl, Ca(OOCCH3)2·H2O, Fe(NO3)3·9H2O, Cu(OOCCH3)2·H2O, and
NiCl2·6H2O were used as received from Aldrich.
sTPE
ð2Þ
ð3Þ
s ¼
FF
TPE ccal ncal
S
sTPE ¼ scal
c
n Scal
1314
www.chemphyschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2012, 13, 1313 – 1324

Chromophores Containing Hydroxyl and Imino Groups
in which s is the TPA cross-section, sTPE is the two-photon excited
cross-section, c is the concentration of reference (ccal) and sample
molecules, n is the refractive index of the solvent, and S is the two-
photon up-conversion fluorescence intensity, cal represents a refer-
ence.
500 MHz, TMS): d=1.09 (t, J=7.0, Hz, 6H; -CH3), 3.35 (t, J=7.0 Hz,
4H; -CH2-), 6.65 (d, J=8.5 Hz, 2H; Ar-H), 6.95 (d, J=8.0 Hz, 2H;
-CH=CH-), 6.97 (d, J=7.5 Hz, 1H; Ar-H), 7.13 (d, J=16.5 Hz, 1H;
Ar-H), 7.40 (m, 5H; Ar-H), 7.59 (d, J=8.0 Hz, 2H; Ar-H), 7.63 (t, J=
3.75 Hz, 1H; Ar-H), 8.99 (s, 1H; -CH=N-), 13.22 ppm (s, 1H; Ar-OH);
13C NMR([D6]DMSO, 125 MHz): d=13.0, 44.1, 111.9, 117.0, 119.6,
119.9, 122.3, 122.6, 127.2, 128.4, 129.6, 133.0, 133.6, 137.6, 146.3,
147.7, 160.8, 162.6 ppm; elemental analysis calcd (%) for
C25H26N2O: C 81.05, H 7.07, N 7.56; found: C 80.98, H 7.13, N 7.48.
Calculation of the Binding Constants
The binding constant, K, was determined by the fluorescence spec-
tra changes using the Benesi–Hildebrand equation,[25] which could
describe 1:1 complex as shown in Equation (4):
4-[(4’-N,N-Diethylaminostilbene-4-ylimino)methyl]phenol (C2): A
similar experimental procedure to that used for C1 gave C2 as a
1
brown solid (62.8%). M.p. 218–2198C; H NMR([D6]DMSO, 500 MHz,
F0
F0
F0
1 1
¼
þ
ð4Þ
TMS): d=1.10 (t, J=7.0 Hz, 6H; -CH3), 3.34 (t, J=6.75 Hz, 4H;
-CH2-), 3.37 (s, 1H; Ar-OH), 6.65 (d, J=9.0 Hz, 2H; Ar-H), 6.87 (d, J=
8.5 Hz, 2H; Ar-H), 6.94 (d, J=16.5 Hz, 1H; -CH=CH-), 7.08 (d, J=
16.5 Hz, 1H; -CH=CH-), 7.21 (d, J=8.5 Hz, 2H; Ar-H), 7.39 (d, J=
9.0 Hz, 2H; Ar-H), 7.53 (d, J=8.5 Hz, 2H; Ar-H), 7.77 (d, J=8.5 Hz,
2H; Ar-H), 8.50 ppm (s, 1H; -CH=N-); 13C NMR([D6]DMSO,
125 MHz): d=12.9, 44.1, 111.9, 116.2, 121.9, 123.0, 124.4, 127.0,
127.9, 128.3, 28.6, 131.1, 135.9, 147.5, 150.4, 159.5, 161.2 ppm; ele-
mental analysis calcd (%) for C25H26N2O: C 81.05, H 7.07, N 7.56;
found: C 81.11, H 7.13, N 7.47.
F0ꢀF F0ꢀFcomplex F0ꢀFcomplex
K
½Mꢂ
in which K is the binding constant, F0 is the integrated fluorescence
intensity of free ligand, F is the observed integrated fluorescence
intensity, Fcomplex is the emission of the ligand–metal complex, and
[M] is the concentration of added metal. The binding constants are
given by the ratio intercept/slope from the plot of F0/F0ꢀF versus
1/[M].
Titration Procedures
p-Methoxy-p’-aminostilbene: The title compound was prepared
as a brown yellow solid by a similar procedure to that used for p-
N,N-diethylamino-p’-aminostilbene (59%). M.p. 158.2–159.68C;
1H NMR (CDCl3, 500 MHz, TMS): d=3.80 (s, 3H; -OCH3), 6.96 (d, J=
9.0 Hz, 2H; Ar-H),7.26 (d, J=16.5 Hz, 1H; Ar-CH=CH-Ar), 7.48 (d,
J=16.5 Hz, 1H; Ar-CH=CH-Ar), 7.62 (d, J=8.5 Hz, 2H; Ar-H), 7.81
(d, J=9.0 Hz, 2H; Ar-H), 8.21 ppm (d, J=8.5 Hz, 2H; Ar-H); elemen-
tal analysis calcd (%) for C15H15NO: C 79.97, H 6.71, N 7.10; found:
C 79.76, H 6.54, N 7.32.
Metal ion stock solution (a few mL, 5ꢁ10ꢀ3 to 1 molLꢀ1) was added
to the corresponding ligand solution (3 mL, 1ꢁ10ꢀ5 molLꢀ1) in hy-
drophilic solvents. The total volume added was less than 10 mL.
The blank experiments were performed by using the same proce-
dure without metal ions to check the effects of additional water
on the optical properties. Upon each addition, the solution was
stirred rapidly for 30 min to reach equilibrium, and UV/Vis spectra
and fluorescence spectra were subsequently recorded. No clear in-
fluence on either the UV/Vis spectra or the emission spectra was
found for the additional water.
Synthesis of C3 and C4: Compounds C3 and C4 were obtained by
using a similar procedure to that for C1 and C2. C3: Pale-yellow
1
solid; yield: 55.1%; m.p. 208–2108C; H NMR ([D6]DMSO, 500 MHz,
TMS): d=3.79 (s, 3H; Ar-OCH3), 7.03 (m, 4H; Ar-H), 7.15 (d, J=
15.0 Hz, 2H; -CH=CH-), 7.24 (s, 1H; Ar-H), 7.50 (m, 3H; Ar-H), 7.78
(m, 4H; Ar-H), 9.02 (s, 1H; -CH=N-), 13.18 ppm (s, 1H; Ar-OH); ele-
mental analysis calcd (%) for C22H19NO2: C 80.22, H 5.81, N 4.25;
found: C 80.31, H 5.74, N 4.31. C4: Yellow solid; yield: 54.3%; m.p.
252–2548C; 1H NMR ([D6]DMSO, 500 MHz, TMS): d=3.78 (s, 3H;
-OCH3), 6.89 (d, J=8.5 Hz, 2H; Ar-H), 6.95 (d, J=8.5 Hz, 2H; Ar-H),
7.11 (d, J=16.5 Hz, 1H; Ar-CH=CH-Ar), 7.18 (d, J=16.5 Hz, 1H; Ar-
CH=CH-Ar), 7.23 (d, J=8 Hz, 2H; Ar-H), 7.54 (d, J=8.5 Hz, 2H; Ar-
H), 7.58 (d, J=8.5 Hz, 2H; Ar-H), 7.78 (d, J=8.5 Hz, 2H; Ar-H), 8.51
(s, 1H; -CH=N-), 10.13 ppm (s, 1H; -OH); 13C NMR ([D6]DMSO,
125 MHz): d=55.6, 114.7, 116.6, 121.9, 126.2, 127.5, 127.8, 128.1,
128.2, 130.3, 131.1, 135.3, 151.1, 159.4, 159.8, 161.1 ppm; elemental
analysis calcd (%) for C22H19NO2: C 80.22, H 5.81, N 4.25; found: C
80.27, H 5.73, N 4.18.
Synthesis
p-N,N-Diethylamino-p’-aminostilbene: p-N,N-Diethylamino-p’-ni-
trostilbene, which was prepared according to
a well-known
method,[26] was dissolved in anhydrous ethanol. Under an argon at-
mosphere, stannous chloride dehydrate was added slowly into the
solution in ethanol. The mixture was heated at reflux for 24 h and
then concentrated in vacuum. The solid was dissolved in an aque-
ous solution of NaOH (3 molLꢀ1) and the pH was adjusted to 8–9
with dilute HCl. A brown solid was obtained after filtration and
was then was washed with distilled water three times. Further pu-
rification by column chromatography using cyclohexane and di-
chloromethane (v/v 1:1) as the eluent gave p-N,N-diethylamino-p’-
1
aminostilbene as a dark red solid (63%). M.p. 95.6–96.88C; H NMR
(CDCl3, 500 MHz, TMS): d=1.17 (t, J=7.0 Hz, 6H; -(CH3)2), 3.37 (m,
4H; -CH2-), 6.66 (d, J=8.5 Hz, 4H; Ar-H), 6.82 (d, J=10.0 Hz, 2H;
Ar-CH=CH-Ar), 7.29 (d, J=8.5 Hz, 2H; Ar-H), 7.34 ppm (d, J=
8.5 Hz, 2H; Ar-H); elemental analysis calcd (%) for C18H22N2: C
81.16, H 8.32, N 10.52; found: C 81.32, H 8.23, N 10.63.
2. Results and Discussion
2.1. ESIPT under One-Photon Excitation
2.1.1. Absorption Spectroscopy
2-[(4’-N,N-Diethylaminostilbene-4-ylimino)methyl]phenol (C1): p-
N,N-Diethylamino-p’-aminostilbene (0.300 g, 1.127 mmol) was dis-
solved in absolute ethanol (30 mL). o-Hydroxybenzaldehyde
(0.206 g, 0.173 mmol) was added to the mixture slowly under an
argon atmosphere. The solution was heated at reflux for 3 h. After
cooling, a yellow solid was precipitated and was filtered and
washed with anhydrous alcohol three times. The product was ob-
tained as a red solid after being recrystallization from ethanol
twice (0.257 g 61.5%). M.p. 192–1938C; 1H NMR ([D6]DMSO,
Typical actual and normalized absorption spectra of these de-
rivatives in benzene and THF are presented in Figure 1. The ab-
sorption maxima and the molar extinction coefficients of the
compounds in various solvents are shown in Table 1. The ab-
sorption maxima of C1 are clearly redshifted with respect to
those of C2 (ca. 15 nm, Table 1). Similar phenomena were ob-
ChemPhysChem 2012, 13, 1313 – 1324
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1315

F. Gao, H. Li et al.
Figure 1. UV/Vis absorption spectra of derivatives C1–C4 in benzene (a) and THF (b); c(C1)=c(C2)=c(C3)=c(C4)=1ꢁ10ꢀ5 molLꢀ1
.
served for C3 and C4 (ca. 13 nm, Table 1). This reflects the in-
ternal hydrogen-bonding effect on the absorption spectra of
C1 and C3. As shown in Scheme 1, compounds C1 and C3 are
characterized by internal hydrogen bonds, whereas there are
no internal hydrogen bonds in C2 and C4. The maximal ab-
sorption wavelengths of C1 and C2 are redshifted compared
with those of C3 and C4, respectively (ca. 30 nm, Table 1),
whereas the absorption maxima of all compounds are less sen-
sitive to solvent polarity (Figure 1 and Table 1). These results
indicate that the absorption could be ascribed to a p!p*
transition of the entire molecule with weak intramolecular
charge transfer in the ground state.[27]
nate from similar ground-state precursors. On the other hand,
the second emission bands of C1 and C3 redshift greatly with
respect to the first ones. The emission maxima and the fluores-
cence quantum yields of the compounds in various solvents
are also presented in Table 1. The emission maxima of C2 and
C4 show small Stokes shifts (ca. 30 nm, Table 1) and the first
maximal emission peaks of C1 and C3 also exhibit small Stokes
shifts (ca. 30 nm, Table 1), whereas in contrast the second max-
imal fluorescence peaks of C1 and C3 show quite large Stokes
shifts (ca. 140 nm, Table 1).
The results indicate that the second emission bands of C1
and C3 could be assigned to ESIPT. The ESIPT emission intensi-
ty of C1 and C3 is much lower than the normal emission inten-
sity in benzene. The intensity ratios of the emission bands,
IE*/IN* (wherein IE* is the representative of the intensity of ESIPT
emission intensity; IN* refers to the normal emission intensity),
are approximate 0.25 for C1 and 0.1 for C3 in benzene. The
fact that the normal emission prevails indicates the existence
of an appreciable energy barrier associated with ESIPT; there-
fore, the S1!So decay of enol emission becomes a competitive
pathway,[7] which is in accordance with our theoretical calcula-
tion.[15] On the other hand, IE*/IN* ratios for C1 and C3 are small-
er in THF than those in benzene (C1: 0.17, C3: 0.07 in THF).
The ESIPT emission of C1 and C3 diminishes further or even
2.2. One-Photon Fluorescence Spectroscopy
Compounds C1 and C3 could undergo ESIPT due to an intra-
molecular hydrogen-bonding effect, whereas C2 and C4 could
not due to a lack of internal hydrogen bonds. Figure 2 shows
that C1 and C3 exhibit well-separated dual emission bands in
benzene and THF, whereas C2 and C4 display only single emis-
sion bands. The first emission bands of C1 and C3 are almost
identical to the emission bands of C2 and C4, respectively.
They are effectively the same as the S0!S1 absorption spectral
feature, which demonstrates that both emission bands origi-
1316
www.chemphyschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2012, 13, 1313 – 1324

Chromophores Containing Hydroxyl and Imino Groups
C1 could be inhibited in methanol. This means that excited-
state deprotonation instead of ESIPT occurs in protic solvents
for C1.[30] On the other hand, the more positively charged hy-
drogen atom of the hydroxyl group in C1 could be favorable
for internal proton transfer in the excited state in aprotic sol-
vents, and thus, the IE*/IN* ratio of C1 is larger than that of C3.
It is accepted generally that ESIPT could diminish the emis-
sion; hence, compounds C2 and C4 have larger fluorescence
quantum yields than C1 and C3, respectively (Table 1). On the
other hand, the emission of all compounds is low in various
solvents, even if C2 and C4 do not undergo ESIPT. It was dem-
onstrated that efficient conical intersections in the twisted ge-
ometries could give rise to dramatic nonradiative deactivation
of the excited singlet state to the ground state of different iso-
mers for some Schiff base compounds.[31] A large dihedral
angle exists between the imino group and C=C band of these
analogues (C1: 30.3788, C2:35.5078, C3: 31.7808, C4: 36.7918).
This indicates that conical intersections as an efficient internal
conversion decay pathway could play a dominant role on the
deactivation of singlet states of these compounds. Conse-
quently, the emission of C1–C4 is reduced remarkably. In par-
ticular, such as in strong, polar solvents, non-luminescent
decay of twisted intramolecular charge transfer could further
quench the emission greatly.[32]
Table 1. The maximum absorption wavelengths, molar extinction coeffi-
cients, emission maxima, and fluorescence quantum yields of C1–C4 in
various solvents.[a]
Solvents
C1
C2
C3
C4
la,max [nm] 405.6
390.4
0.442
428
0.0291
388
372.8
0.397
423, 527 422
0.0109
370.4
0.439
360.0
0.433
e
0.448
lf,max [nm] 426, 542
0.0224
la,max [nm] 403.2
0.435
benzene
F
0.0162
358.4
0.439
e
0.458
1,4-dioxane
THF
lf,max [nm] 405,425, 543 406,429 423, 523 423
F
0.0249
la,max [nm] 403.2
0.450
lf,max [nm] 427, 551
0.0304
la,max [nm] 402.4
0.443
lf,max [nm] 423, 581
0.0122
la,max [nm] 400.8
0.446
lf,max [nm] 434
0.0278
386.4
0.461
430
0.0672
388
0.478
431
0.00895 0.00791
0.00988
371.2
0.351
0.0128
359.2
0.293
e
426, 524 421
F
0.0143
367.2
0.473
434
0.0329
356
e
0.441
434
0.0107
354.4
0.448
438
acetone
acetonitrile
DMF
F
387.2
0.437
432
365.6
0.463
438
e
F
0.00393
0.00600 0.00675
0.00886
363.1
0343
440
la,max [nm] 408.3
e
394.7
0.424
438
373.3
0.257
438
0.380
lf,max [nm] 434
F
0.0634
0.0172
379.3
0.277
430
0.0183
367.4
0.335
435, 523 429
0.00301 0.00709
0.0149
342.5
0.374
la,max [nm] 399.8
e
0.446
lf,max [nm] 434
0.0172
methanol
2.3 ESIPT under Two-Photon Excitation
F
0.0872
We utilized a Ti:sapphire femtosecond laser to determine the
TPA-induced emission of the compounds. Compounds C1 and
C2 exhibit remarkable TPA emissions in benzene, and C3 and
C4 display clear TPA emissions in THF under 740 nm laser exci-
tation. As seen in Figure 3a, compound C1 exhibits dual TPA
emission bands in benzene, whereas C2 shows a single TPA
emission band. Furthermore, the maximal emission wavelength
of the first TPA emission band of C1 is almost identical to that
of C2. The maximal emission wavelength of the second TPA
emission band of C1 is much larger than that of the first emis-
sion band. Furthermore, the maximal emission wavelength of
the second TPA emission band of C1 is close to that of the
one-photon ESIPT emission of C1. As a result, the first TPA
emission band of C1 could be normal TPA emission and the
second TPA emission band of C1 could be assigned to the
ESIPT TPA emission. The C3/C4 pair exhibits similar contrasting
TPA optical properties to those of C1/C2 (as shown in Fig-
ure 3b). The experimental results suggest that C1 and C3
could undergo ESIPT under two-photon excitation, whereas C2
and C4 do not process ESIPTs under the same conditions.
The maximal TPA ESIPT emission wavelengths of C1 and C3
are almost identical to the maximal one-photon ESIPT emission
wavelength (one-photon ESIPT C1: 542 nm in benzene, C3:
527 nm in THF; two-photon ESIPT: C1: 544 nm in benzene, C3:
ca. 531 nm in THF), which indicates that ESIPT could occur
through the same or similar excited states for one- and two-
photon process. However, the IE*/IN* ratio in the two-photon
process is much larger than that in the one-photon emission
process (C1: ca. 1.40 in benzene; C3: ca. 1.0 in THF). At pres-
ent, we do not fully understand why ESIPT becomes predomi-
[a] e: 1ꢁ105 Lmolꢀ1 cmꢀ1
, la,max: the maximal absorption wavelength,
lf,max: the emission maxima. The measurement errors of fluorescence
quantum yields do not exceed ꢃ10%.
completely disappears in strong, aprotic polar solvents, such as
acetone and acetonitrile (see the Supporting Information). This
could be caused by twisted intramolecular charge transfer,
which could be the major competitive deactivation approach
in the excited singlet state of the enol forms of C1 and C3 in
strong, polar solvents; therefore, the ESIPT emission could be
restricted.[28] Although photoinduced cis–trans isomerization of
stilbene and its derivatives is often a rapid process (several
hundred fs to ps),[29] it is still much slower than the ESIPT pro-
cess (less than one hundred fs).[1c)] Hence, cis–trans isomeriza-
tion of C1 and C3 may not be a competitive photochemical
decay channel to ESIPT. The above results demonstrate that C1
and C3 could undergo ESIPT under one-photon excitation via
a
reasonable six-membered ring in the excited states
(Scheme 1), while it is hard to process ESIPTs for C2 and C4.
It is interesting that C1 does not exhibit ESIPT emission in
protic methanol solvent, whereas C3 does present an ESIPT
emission in methanol (see the Supporting Information). Calcu-
lations show that the hydrogen atom of the hydroxyl group in
C1 has a more positive charge than that in C3, which could be
caused by larger intramolecular charge transfer in C1 due to a
stronger electron-donating effect of the diethylamino group,
inducing a greater hydrogen-bond effect with protic methanol.
Consequently, internal proton transfer in the excited state of
ChemPhysChem 2012, 13, 1313 – 1324
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1317

F. Gao, H. Li et al.
Figure 2. Actual and normalized fluorescence spectra of derivatives C1–C4 in benzene (a) and THF (b) under the same excitation condition;
c(C1)=c(C2)=c(C3)=c(C4)=1ꢁ10ꢀ5 molLꢀ1
.
nant in the two-photon process. It is generally accepted that
the two-photon excitation process of a compound is different
from one-photo excitation.[33] During two-photon excitation,
the excited molecule could reach the excited state S2 or Sn first
and then decay to S1 by a radiationless transition and vibration
relaxation, alternatively the excited molecule could reach S1
first through excitation by one photon. Therefore, because C1
and C3 are excited by two photons, we suppose that photo-
tautomerization could take place in higher excited states of
enols of C1 and C3,[34] and thus, the energy barrier could be re-
duced and easily overcome. Furthermore, a femtosecond laser
and the E–K phototautomerization process have wonderful
timescale matches. As a result, two-photon excitation could be
favorable for ESIPT.
displays larger TPA cross-sections than C3 under various near-
IR laser frequencies (Table 2).
2.4. Calculations
Calculations demonstrate that C1 and C3 have a high possibili-
ty of undergoing ESIPT;[15] this was verified by the above exper-
imental observations. We performed further various calcula-
tions, including bond lengths, energies, dipole moments, and
frontier orbitals, to give an in-depth analysis of why C2 and C4
could not undergo ESIPT. We expected that, unlike C1 and C3,
the transition states from E to K were not possible for the
ground and excited states of C2 and C4, which indicated that
proton transfer occurred less in the ground and excited states
of C2 and C4.
As in low polarity solvents, the maximal (normal and ESIPT)
emission maxima of C1 redshift in comparison with those of
C3 under one- and two-photon excitation, as observed in the
absorption spectra, which indicates that phototautomers E and
K of C1 could have larger dipole moment changes between
the excited and ground states than those of C3. The phototau-
tomers E and K of C1 exhibit larger dipole moment differences
between the excited and ground states than those of C3,[15]
while HOMO–LUMO gaps of E and K of C1 are lower than
those of C3.[15] This could be the fundamental reason why C1
2.5. Bond Length
The internal hydrogen-bonding effect is dependent on bond
length. The most important bond lengths of the E and K forms
of the compounds associated with ESIPT in S0 and S1 are
shown in Table 3. The data show that the HꢀN distances in the
E forms of C2 and C4 are much longer than C1 and C3 in S0
and S1, even up to about 6.7 ꢂ. The HꢀO distances in the K
1318
www.chemphyschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2012, 13, 1313 – 1324

Chromophores Containing Hydroxyl and Imino Groups
Figure 3. Actual and normalized TPA emission spectra of C1–C4 in benzene (a) and THF (b) under 740 nm laser excitation;
c(C1)=c(C2)=c(C3)=c(C4)=5ꢁ10ꢀ4 molLꢀ1
.
C2 and C4 in S0 and S1. Furthermore, owing to a weak stretch-
ing effect of N for H, the HꢀO distances in the E forms of C2
and C4 are shorter than those of C1 and C3 in S0 and S1. The
data in Table 3 also show that the HꢀN distances in the K form
of C2 and C4 are shorter than those of C1 and C3 in S0 and S1
due to the lack of a strong pulling effect of O for H. This
means that it is hard for H to leave O in the E forms and to
leave N in the K forms of C2 and C4. The bond length suggests
that it is hard for the ground and excited states of C2 and C4
to undergo internal proton transfer.
Table 2. Two-photon cross-sections of C1–C4 under various laser fre-
quencies.
Laser frequencies
[nm]
s[a] [GM]*
C1
C2
C3
C4
700
720
740
760
780
800
820
840
860
880
163
74
64
67
72
107
106
104
60
54
29
22
22
19
22
25
22
11
13
–
15
6
–
–
5
10
5
–
2
–
–
–
–
12
13
6
–
–
59
–
[a] Two-photon cross-section (1GM=10ꢀ50 cmsꢀ1 photonꢀ1); the errors do
not exceed 15%; C1 and C2 were recorded in benzene, whereas C3 and
C4 were recorded in THF.
2.6. Energies
As shown in Table 4, compounds C2 and C4 have approxi-
mately twice the energy differences between E and K in
ground state (DE) than those of C1 and C3. Furthermore, the
energy differences in the excited state are much larger than
those of C1 and C3 (>20 times). This indicates that the occur-
rence of internal proton transfer in the ground and excited
states of C2 and C4 is much less likely than that of C1 and C3,
particularly in the excited state. This could be the fundamental
reason that C2 and C4 do not show ESIPT.
forms of C2 and C4 are also much longer in S0 and S1 (ca.
6.2 ꢂ) than those of C1 and C3. It is well known that hydrogen
bonds are short distance effects in essence; hence, intramolec-
ular hydrogen-bonding forces between H and N in the E form
and between H and O in the K form should be quite weak for
ChemPhysChem 2012, 13, 1313 – 1324
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1319

F. Gao, H. Li et al.
than those of C1 and C3 in the
ground and excited states (DmE–K
in the ground state: C1: 1.875,
C2: 9.844, C3: 1.877, C4: 8.265;
in the excited state for C1: 1.174,
C2: 9.252, C3: 1.635, C4: 8.702).
This means that if internal
proton transfer occurs in the
ground and excited states, com-
pounds C2 and C4 would have
to undergo much larger charge
transfer between E and K than
that in C1 and C3, which in turn
indicates that the possibility of
Table 3. The most important bond lengths associated with ESIPT obtained at DFT and HF (in bracket) level and
excited state at CIS level.
Structural
Ground State
Excited State
parameters
E
K
E*
K*
ROꢀH
RHꢀN
ROꢀN
ROꢀH
RHꢀN
ROꢀN
ROꢀH
RHꢀN
ROꢀN
ROꢀH
RHꢀN
ROꢀN
0.9988 (0.9548)
1.7171 (1.8876)
2.6202 (2.7116)
0.9665 (0.9429)
6.7039 (6.6815)
6.4396 (6.3949)
0.9982 (0.9546)
1.7203 (1.8887)
2.6218 (2.7121)
0.9665 (0.9429)
6.7053 (6.6817)
6.4398 (6.3947)
1.6306 (1.8678)
1.0542 (1.0056)
2.5565 (2.6569)
6.2667 (6.2455)
1.0108 (0.9928)
6.4433 (6.3834)
1.6287 (1.8666)
1.0544 (1.0055)
2.5550 (2.6561)
6.2651 (6.2433)
1.0107 (0.9928)
6.4420 (6.3823)
0.9566
1.8543
2.6912
0.9430
6.6657
6.3861
0.9570
1.8500
2.6872
0.9431
6.6633
6.3826
1.7925
1.0129
2.6365
6.1935
0.9948
6.3893
1.8078
1.0118
2.6461
6.1816
0.9952
6.3869
C1[15]
C2
C3[15]
C4
undergoing
internal
proton
transfer would be negligible for
C2 and C4.
Table 4. Calculated relative energy difference between E and K for the ground and excited states and all the
energies of E as a reference point at the corresponding methods (EE =0).
The electron distribution of
the frontier orbitals reflects the
electron transition characteris-
tics. Figure 4 shows that the
HOMO and LUMO of C2 and C4
are characterized with excellent
p-type symmetry. Furthermore,
the electron density distribution
of the frontier orbitals of C2 and
States
Method
DE [kcalmolꢀ1
]
C1[15]
C2
C3[15]
C4
DFT//DFT
DFT//HF
TD-DFT//CIS
4.703
5.431
0.246
8.364
9.093
5.883
4.762
5.436
0.143
8.508
9.220
6.758
ground
excited
2.7. Dipole Moment and Frontier Orbitals
C4 suggests that charge transfer does occur from the electron
donor part to the imino moiety during the HOMO!LUMO
transition and C2 exhibits a larger charge transfer than C4.
As seen from Table 5, the dipole moment differences between
the phototautomers E and K of C2 and C4 are much greater
Figure 4. Calculated qualitative shapes of the frontier molecular orbits of C2 and C4 in the ground and excited states.
1320
www.chemphyschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2012, 13, 1313 – 1324

Chromophores Containing Hydroxyl and Imino Groups
2.8. Sensing Properties
Table 5. The dipole moment of each phototautomer in the ground and
excited states.
Owing to the presence of adjacent heteroatoms, some ESIPT
compounds could be used as efficient chemosensors for spe-
cific metal ions through the inhibition of ESIPT. We performed
sensing response experiments with these chromophores for
metal ions in DMF. As expected, compounds C2 and C4 do not
exhibit clear responses due to their structural characteristics,
States
Phototautomers
Compounds
C2
C3[15]
C1[15]
C4
mE [D]
mK [D]
DmE–K [D]
mE* [D]
mK* [D]
5.446
7.321
1.875
6.370
7.544
1.174
1.784
3.882
5.559
1.677
4.072
5.707
1.635
1.192
9.457
8.265
0.983
9.685
8.702
ground
excited
11.628
9.844
3.466
12.718
9.252
DmE*–K* [D]
Hence, the HOMO!LUMO transition could be ascribed to p!
p* excitation with internal charge-transfer character.
The calculated absorption and normal fluorescence maxima
of C2 and C4, which equal the corresponding frontier orbital
transition energy difference (namely, singlet–singlet transition
energies, DE), are shown in Table 6. Herein, the values of the
absorption refer to the vertical transition from S0 states to the
Franck–Condon S1 states, while the values from S1 states to the
corresponding Franck–Condon S0 states are assigned to emis-
sions that are redshifted with respect to the absorption. The
H!L transitions of C2 and C4 show the largest f values for ab-
sorption (f: C2: 0.9506, C4: 1.0919) and fluorescence spectros-
copy (f : C2: 1.8163, C4: 1.7476) and the biggest weighting
(C2: 91%, C4: 87% for the absorption spectroscopy, C2: 78%,
C4: 76% the fluorescence spectroscopy). This further demon-
strates that the absorption spectra of C2 and C4 are character-
ized by p!p* excitations. The calculated absorption maxima
of C2 and C4 are in good agreement with experimental meas-
urements; this suggests the weak dependence of the absorp-
tion maxima on the polarity of the media. Furthermore, the
calculated maximal absorption maxima of C2 and C4 are blue-
shifted approximately 10 nm with respect to those of C1 and
C3, as also determined experimentally. Tables 5 and 6 show
that C2 exhibits larger dipole moment changes between the
excited state and the ground state than C4 (E!E*), and small-
er HOMO–LUMO gaps, which could cause larger absorption
and emission maxima and higher TPA cross-sections for C2
(Table 2).
Figure 5. UV/Vis (a) and fluorescence spectra (b) of C3 (1ꢁ10ꢀ5 molLꢀ1) in
DMF at room temperature as a function of the addition of various aqueous
stock solutions of Zn(OAc)2 (5ꢁ10ꢀ3 to 1 molLꢀ1).
whereas, as shown in Figure 5,
Table 6. Calculated energy difference of frontier orbital transitions (in eV), absorption and fluorescence peak
wavelengths (in nm), oscillator strengths (f), and the weighting of the most important microstates (H: HOMO,
L: LUMO) in the anterior excited states for C2 and C4 at the TDDFT//HF and TDDFT//CIS levels.
compound C3 shows a remark-
able sensing response to Zn2+
ions in DMF that is much better
than that of C1. Continuous ti-
tration with an aqueous solution
of Zn2+ ions in DMF causes a
gradual redshift and a single iso-
sbestic point in the absorption
spectra of C3. Figure 5 also
shows a gradual increase, which
finally reaches an approximate
constant, of the emission intensi-
ty with a slight redshift as the
aqueous solution of Zn2+ ions is
Spectra
DE [eV]
l [nm]
f
Transition
1
2
3
1
2
3
1
2
3
1
2
3
3.10
3.90
3.97
3.40
4.15
4.23
2.55
3.39
3.71
2.71
3.50
3.70
399.7
317.8
312.2
365.2
298.9
293.2
485.4
366.2
334.5
458.2
353.9
335.3
0.9506
0.8651
0.0649
1.0919
0.6223
0.0423
1.8163
0.4246
0.0443
1.7476
0.2244
0.0917
H-0!L+0: 91%
C2
C4
C2
C4
H-0!L+1: 47%; H-1!L+0: 45%
H-0!L+1: 37%; H-1!L+0: 33%
H-0!L+0: 87%
absorption
H-0!L+1: 62%; H-1!L+0: 31%
H-2!L+0: 36%; H-1!L+0: 33%
H-0!L+0: 78%
H-1!L:+0: 55%; H-0!L+1: 40%
H-3!L+0: 38%; H-2!L+0: 21%
H-0!L+0: 76%
normal fluorescence
H-1!L:+0: 54%; H-0!L+1: 44%
H-2!L+0: 60%; H-3!L+0: 12%
ChemPhysChem 2012, 13, 1313 – 1324
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1321

F. Gao, H. Li et al.
added. These phenomena indicate an equilibrium of zinc com-
plex. Compared with C3, the absorption and emission spectra
of C1 are much less sensitive to Zn2+ concentration (see the
Supporting Information). This could be ascribed to stronger in-
ternal hydrogen bonding in C1, so that the hydrogen atom of
the OH group in C1 does not leave as easily to allow the for-
fluorescence intensity curve of C3 against the reciprocal of the
Zn2+ and Fe3+ concentration (Figure 7), the Benesi–Hildebrand
plot, gives linear fit characteristics for 1:1 complexation behav-
ior from which the association constants are estimated to be
0.5ꢁ104 (Zn2+) and 0.45ꢁ104 Lmolꢀ1 (Fe3+).
mation of a complex with Zn2+
.
We investigated further the sensing nature of C3 for other
metal ions. Figure 6 shows the fluorescence response and the
photoluminescence of C3 to metal ions at equilibrium. Com-
Figure 7. Benesi–Hildebrand plot of the fluorescence intensity of C3 against
the reciprocal of the Zn2+ (a) and Fe3+ (b) concentrations.
Figure 6. a) Sensing nature of C3 to various metal ions. The concentration of
Zn2+ ions added is 5ꢁ10ꢀ5 molLꢀ1; the concentrations of other metal ions
are 5ꢁ10ꢀ4 molLꢀ1. All surveys were performed at equilibrium. b) Photolumi-
nescence of C3 in DMF containing metal ions under excitation at 365 nm;
from left to right: K+, Ca2+, Na+, Mg2+, Li+, Zn2+, Cu2+, Ni2+, Fe3+, and
blank solution.
We also performed a survey of the C3 sensing response to
various metal ions in other hydrophilic solvents. Compound C3
displays a much weaker sensing response to metal ions in THF,
acetonitrile, and DMSO (see the Supporting Information), while
in methanol it shows some sensitivity to Zn2+ ions (Figure 8).
Furthermore, the ESIPT emission band of C3 disappeared with
the addition of Zn2+ ions in methanol, which demonstrated
that the adjacent O and N heteroatoms participated in the for-
mation of the complex (Figure 8).
pound C3 could recognize Zn2+ ions to bring about a large
fluorescence enhancement and was sensitive to Fe3+ ions with
a remarkable decrease in emission intensity. Only slight or neg-
ligible changes were observed for the addition of other aque-
ous solutions of metal ions, including significant physiological
metal ions (such as K+, Na+, Mg2+, Ca2+) and heavy metal ions
(such as Fe3+, Cu2+ and Ni2+ ions) in DMF (see the Supporting
Information), except for Zn2+ and Fe3+ ions. The emission in-
tensity of C3 in DMF reduced gradually with continuous titra-
tion of an aqueous solution of Fe3+ ions, while the optical den-
sity in the absorption spectra increased gradually (see the Sup-
porting Information).
3. Conclusions
We fully investigated the linear absorption, emission, and two-
photo properties of new chromophores containing hydroxyl
and imino groups, with the hydroxyl group being in the ortho
or para positions of the aromatic ring. One- and two-photon
optical properties of these compounds had a strong relation-
ship with their molecular structures. There was strong evidence
that C1 and C3 were able to undergo ESIPT under one- and
two-photon excitation, whereas C2 and C4 could not. Further-
more, ESIPT becomes competitive under two-photon excita-
Consequently, compound C3 shows unique sensing respons-
es to Zn2+ and Fe3+ ions; this demonstrates that C3 could be
an effective chemosensor for Zn2+ and Fe3+ ions. Fitting of the
1322
www.chemphyschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2012, 13, 1313 – 1324

Chromophores Containing Hydroxyl and Imino Groups
Figure 8. Sensing nature of C3 to various metal ions. The concentration of Zn2+ ions added is 5ꢁ10ꢀ5 molLꢀ1; the concentrations of other metal ions are
5ꢁ10ꢀ4 molLꢀ1. The structure of the complex is also shown; L is solvent or water.
Keywords: charge transfer
photochemistry · sensors · synthesis design
· density functional theory ·
tion. Calculations demonstrated that E forms of C1 and C2 had
larger dipole moment differences between the ground and ex-
cited states than those of C3 and C4 due to greater internal
charge-transfer character from the N,N-diethylamino group. As
a result, TPA cross-sections of C1 and C2 were larger than
those of C3 and C4. The bond lengths, energies, and dipole
moments of phototautomers of C2 and C4 in S0 and S1 were
not favorable for ESIPT. Compound C3 in DMF displayed excel-
lent sensing properties for Zn2+ and Fe3+ ions with a large en-
hancement or significant decrease in fluorescence emission, re-
spectively. Disappearance of the ESIPT emission band of C3 in
methanol demonstrated that the hydroxyl and imino groups
played key roles in the formation of complexes with the metal
ions. The results presented herein will benefit the development
of new larger ESIPT molecules.
[1] a) W. T. Chuang, C. C. Hsieh, C. H. Lai, C. H. Lai, C. W. Shih, K. Y. Chen,
W. Y. Hung, Y. H. Hsu, P. T. Chou, J. Org. Chem. 2011, 76, 8189–8202;
b) C. C. Hsieh, P. T. Chou, C. W. Shih, W. T. Chuang, M. W. Chung, J. Lee, T.
Joo, J. Am. Chem. Soc. 2011, 133, 2932–2943; c) Y. M. Cheng, S. C. Pu,
C. J. Hsu, C. H. Lai, P. T. Chou, ChemPhysChem 2006, 7, 1372–1381; d) Y.
Qian, M. M. Cai, L. H. Xie, G. Q. Yang, S. K. Wu, W. Huang, ChemPhys-
Chem 2011, 12, 397–404.
[2] a) M. J. Paterson, M. A. Robb, L. Blancafort, A. D. Debellis, J. Phys. Chem.
A 2005, 109, 7527–7537; b) S. R. Grando, C. M. Pessoa, M. R. Gallas,
T. M. H. Costa, F. S. Rodembusch, E. V. Benvenutti, Langmuir 2009, 25,
13219–13223.
[3] a) P. S. Sardar, S. Samata, S. S. Maity, S. Dasgupta, S. Ghosh, J. Phys.
Chem. B 2008, 112, 3451–3461; b) M. Barbatti, A. J. A. Aquino, H. Lisch-
ka, C. Schriever, S. Lochbrunner, E. Riedle, Phys. Chem. Chem. Phys.
2009, 11, 1406–1415.
[4] a) A. O. Doroshenko, E. A. Posokhov, A. A. Verezubova, L. M. Ptyagina,
V. T. Skripkina, V. M. Shershukov, Photochem. Photobiol. Sci. 2002, 1, 92–
99; b) J. D. Coe, B. G. Levine, T. J. Martinez, J. Phys. Chem. A 2007, 111,
11302–11310; c) M. J. Paterson, M. A. Robb, L. Blancafort, A. D. DeBellis,
J. Am. Chem. Soc. 2004, 126, 2912–2922.
[5] K. M. Solntsev, E. N. Sullivan, L. M. Tolbert, S. Ashkenazi, P. Leiderman, D.
Huppert, J. Am. Chem. Soc. 2004, 126, 12701–12708.
[6] P. Toele, H. Zhang, M. Glasbeek, J. Phys. Chem. A 2002, 106, 3651–3658.
[7] W. Yu, C. Cheng, Y. Cheng, P. Wu, Y. Song, Y. Chi, P. Chou, J. Am. Chem.
Soc. 2003, 125, 10800–10801.
Acknowledgements
We thank encouragement from the National Science Foundations
of China and the Key Laboratory of Functional Crystals and Laser
Technology, Technical Institute of Physics and Chemistry, Chinese
Academy of Sciences. We appreciate financial support from the
Fundamental Research Funds for the Central Universities (no.
CDJZR10220006).
[8] C. Chen, C. Lin, C. Hsieh, C. Lai, G. Lee, C. Wang, P. Chou, J. Phys. Chem.
A 2009, 113, 205–214.
ChemPhysChem 2012, 13, 1313 – 1324
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
1323

F. Gao, H. Li et al.
[9] a) K. Chen, C. Hsieh, Y. Cheng, C. Lai, P. Chou, Chem. Commun. 2006,
4395–4397; b) K. Chen, Y. Cheng, C. Lai, C. Hsu, M. Ho, G. Lee, P. Chou,
J. Am. Chem. Soc. 2007, 129, 4534–4525.
[10] E. L. Roberts, J. Dey, I. M. Warner, J. Phys. Chem. A 1997, 101, 5296–
5301.
[24] M. A. Albota, C. Xu, W. W. Webb, Appl. Opt. 1998, 37, 7352–7356.
[25] a) H. A. Benesi, J. H. Hildebrand, J. Am. Chem. Soc. 1949, 71, 2703–2707;
b) S. Fery-Forgues, M. T. Le Bris, J. P. Guette, B. Valeur, J. Phys. Chem.
1988, 92, 6233–6237; c) C. F. Chow, M. H. W. Lam, W. Y. Wong, Inorg.
Chem. 2004, 43, 8387–8393.
[11] a) S. J. Lim, J. Seo, S. Y. Park, J. Am. Chem. Soc. 2006, 128, 14542–14547;
b) Y. Wu, X. Peng, J. Fan, S. Gao, M. Tian, J. Zhao, S. Sun, J. Org. Chem.
2007, 72, 62–70; c) X. Peng, Y. Wu, J. Fan, M. Tian, K. Han, J. Org. Chem.
2005, 70, 10524–10531; d) N. Kaur, N. Singh, D. Cairns, J. F. Callan, Org.
Lett. 2009, 11, 2229–2232; e) D. Svechkarev, B. Dereka, A. Doroshenko,
J. Phys. Chem. A 2011, 115, 4223–4230.
[12] a) F. Shen, H. Xia, C. Zhang, D. Lin, L. He, Y. Ma, J. Phys. Chem. B 2004,
108, 1014–1019; b) J. E. Kwon, S. Y. Park, Adv. Mater. 2011, 23, 3615–
3642; c) S. H. Kim, S. Park, J. E. Kwon, S. Y. Park, Adv. Funct. Mater. 2011,
21, 644–651.
[26] G. Cavallini, E. Massarani, US Patent 2, 878, 291, 1959.
[27] a) T. Stalin, N. Rajendiran, Chem. Phys. 2006, 322, 311–322; b) A. Chakra-
borty, S. Kar, D. N. Nath, N. Guchhait, J. Phys. Chem. A 2006, 110, 12089–
12095.
[28] Y. L. Wang, G. S. Wu, Acta Phys-Chin. Sin. 2008, 24, 552–560.
[29] a) R. M. Hochstrasser, Pure Appl. Chem. 1980, 52, 2683–2691; b) L.
Nikowa, D. Schwarzer, J. Troe, J. Schroeder, J. Chem. Phys. 1992, 97,
4827–4835; c) S. H. Courtney, S. K. Kim, S. Canonica, G. R. Fleming, J.
Chem. Soc. Faraday Trans. 2 1986, 82, 2065–2072; d) M. Sumitani, K.
Yoshihara, Bull. Chem. Soc. Jpn. 1982, 55, 85–89.
[13] a) Y. Qian, S. Li, G. Zhang, S. Wang, H. Xu, C. Li, Y. Li, G. Yang, J. Phys.
Chem. B 2007, 111, 5861–5858; b) D. W. Chang, S. Kim, S. Y. Park, Macro-
molecules 2000, 33, 7223–7225; c) S. Kim, S. Y. Park, I. Yoshida, H. Kawai,
T. Nagamura, J. Phys. Chem. B 2002, 106, 9291–9294; d) Y. Tian, C. Chen,
C. Yang, A. Young, S. Jang, W. Chen, A. Jen, Chem. Mater. 2008, 20,
1977–1987.
[14] a) S. K. Sahoo, R. K. Bera, M. Baral, B. K. Kanungo, J. Photochem. Photo-
biol. A 2007, 188, 298–310; b) M. Ziꢃłek, J. Kubicki, A. Maciejewski, R.
Naskrecki, A. Grabowska, Chem. Phys. Lett. 2003, 369, 80–89; c) P. Fita,
E. Luzina, T. Dziembowska, D. Kopec, P. Piatkowski, C. Radzewicz, A. Gra-
bowska, Chem. Phys. Lett. 2005, 416, 305–310.
[30] a) P. T. Chou, C. H. Huang, S. C. Pu, Y. M. Cheng, Y. H. Liu, Y. Wang, C. T.
Chen, J. Phys. Chem. A 2004, 108, 6452–6454; b) M. Mukhopadhyay, A.
Mandal, R. Misra, D. Banerjee, S. P. Bhattacharyya, S. Mukherjee, J. Phys.
Chem. B 2009, 113, 567–573; c) M. Lukeman, P. Wan, Chem. Commun.
2001, 1004–1005.
[31] a) J. M. Ortiz-Sꢄnchez, R. Gelabert, M. Moreno, J. M. Lluch, J. Chem. Phys.
2008, 129, 214–308; b) M. Ziꢃłek, M. Gil, J. A. Organero, A. Douhal,
Phys. Chem. Chem. Phys. 2010, 12, 2107–2115.
[32] a) A. Sanchez-Galvez, P. Hunt, M. A. Robb, M. Olivucci, T. Vreven, H. B.
Schlegel, J. Am. Chem. Soc. 2000, 122, 2911–2924; b) C. Potter, R.
Brown, F. Vollmer, W. Rettig, J. Chem. Soc. Faraday Trans. 1994, 90, 59–
67; c) S. Olsen, S. C. Smith, J. Am. Chem. Soc. 2007, 129, 2054–2065.
[33] a) C. V. Bindhu, S. S. Harilal, V. P. N. Nampoori, C. P. G. Vallabhan, J. Phys.
D 1999, 32, 407–411; b) D. S. Larsen, E. Papagiannakis, I. H. M. V. Stok-
kum, M. Vengris, J. T. M. Kennis, R. V. Grondelle, Chem. Phys. Lett. 2003,
381, 733–742; c) K. D. Belfield, A. R. Morales, B. S. Kang, J. M. Hales, D. J.
Hagan, E. W. V. Stryland, V. M. Chapela, J. Percino, Chem. Mater. 2004,
16, 4634–4641; d) F. Graf, A. Penzkofer, Opt. Quantum Electron. 1985,
17, 53–68; e) T. Lenzer, F. Ehlers, M, Scholz, R. Oswald, K. Oum, Phys.
Chem. Chem. Phys. 2010, 12, 8832–8839.
[15] Q. Wang, F. Gao, H. R. Li, S. T. Zhang, Chin. J. Chem. 2010, 28, 901–910.
[16] Purification of Laboratory Chemicals, D. D. Perrin, W. L. F. Armarego, Per-
gamon, New York, 1996.
[17] M. Z. Zgierski, A. Grabowska, J. Chem. Phys. 2000, 112, 6329–69367.
[18] P. G. Yi, Y. H. Liang, C. Z. Cao, Chem. Phys. 2005, 315, 297–302.
[19] Y. H. Liang, P. G. Yi, Chem. Phys. Lett. 2007, 438, 173–177.
[20] Z. Yang, S. Yang, J. Zhang, J. Phys. Chem. A 2007, 111, 6354–6360.
[21] a) G. Markus, S. Monika, L. Vladimir, G. Nikolai, E. Alexander, R. G. Ute,
Anal. Chem. 2009, 81, 6285–6294; b) M. Fischer, J. Georges, Chem. Phys.
Lett. 1996, 260, 115–118; c) D. F. Eaton, Pure Appl. Chem. 1988, 60,
1107–1114.
[34] V. I. Tomin, R. Jaworski, J. Mol. Struct. 2009, 924–926, 461–465.
[22] a) M. Maus, W. Reitigg, D. Bonafoux, R. Lapouyade, J. Phys. Chem. A
1999, 103, 3388–3401; b) M. Lukeman, D. Veale, P. Wan, V. Ranjit, N. Mu-
nasinghe, J. E. Corrie, Can. J. Chem. 2004, 82, 240–253.
Received: November 5, 2011
[23] C. Xu, W. W. Webb, J. Opt. Soc. Am. B 1996, 13, 481–491.
Published online on February 16, 2012
1324
www.chemphyschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2012, 13, 1313 – 1324