A ratiometric TICT-type dual fluorescent sensor for an amino acid

Article abstract of DOI:10.1039/b924176k

A novel ratiometric fluorescent sensor 1 having the 4-(N,N-dimethylamino) benzoate (DMAB) group as a twisted intramolecular charge transfer (TICT) dual fluorescence site and the guanidinium group as a binding site for an amino acid at the 12- and 3-positions of cholic acid, respectively, was designed and synthesized. The sensor 1 showed characteristic dual fluorescence and it was shown that the guanidinium group in the proximity of the DMAB group influenced largely the TICT formation process in 1. Liquid/liquid extraction experiments demonstrated that the intensity ratio of the TICT fluorescence to the locally-excited (LE) fluorescence (I_TICT/I_LE) observed from the organic layer decreased with an increase in the concentration of N-acetyl-d-phenylalanine (AcPhe) in the water phase. Such fluorescence titration experiments afforded the apparent binding constant between 1 and AcPhe to be 7.0 × 10⁵ M^-1. Temperature dependence of the fluorescence spectrum of 1 in the absence and presence of AcPhe indicated that the primary origin of the change in the I_TICT/I_LE value upon recognition of the amino acid was the increase in the activation energy of the TICT formation process in 1. the Owner Societies.

Full text of DOI:10.1039/b924176k

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
A ratiometric TICT-type dual fluorescent sensor for an amino acid
Akitaka Ito, Shoji Ishizaka and Noboru Kitamura*
Received 17th November 2009, Accepted 13th March 2010
First published as an Advance Article on the web 21st April 2010
DOI: 10.1039/b924176k
A novel ratiometric fluorescent sensor 1 having the 4-(N,N-dimethylamino)benzoate (DMAB)
group as a twisted intramolecular charge transfer (TICT) dual fluorescence site and the
guanidinium group as a binding site for an amino acid at the 12- and 3-positions of cholic acid,
respectively, was designed and synthesized. The sensor 1 showed characteristic dual fluorescence
and it was shown that the guanidinium group in the proximity of the DMAB group influenced
largely the TICT formation process in 1. Liquid/liquid extraction experiments demonstrated
that the intensity ratio of the TICT fluorescence to the locally-excited (LE) fluorescence
(ITICT/ILE) observed from the organic layer decreased with an increase in the concentration
of N-acetyl-D-phenylalanine (AcPhe) in the water phase. Such fluorescence titration experiments
afforded the apparent binding constant between 1 and AcPhe to be 7.0 ꢀ 105 Mꢁ1. Temperature
dependence of the fluorescence spectrum of 1 in the absence and presence of AcPhe indicated that
the primary origin of the change in the ITICT/ILE value upon recognition of the amino acid was
the increase in the activation energy of the TICT formation process in 1.
a PET sensor should be considered to develop further sensitive
Introduction
detection of an ion/molecule recognition event in analytical
Molecular recognition phenomena play important roles in a
sciences. To avoid such disadvantages of a PET sensor,
variety of physical, chemical, and biological events. Examples
ratiometric fluorescence sensors have been also reported, by
in analytical sciences include the determination of the
which ion/molecular recognition can be evaluated by the
concentration of a target molecule/ion and separation of
fluorescence intensity ratio of those at two wavelengths: the
enantiomers. It has been also recognized generally that
ratio reports the concentration of a targeted ion/molecule
fluorescence detection enables one a very sensitive and non-
without effects of excitation light power and instrumental
destructive analysis of a target molecule or ion by a chemosensor.
conditions such as slit widths, fluctuation of light source
Therefore, the design and development of a fluorescent sensor
power, and so forth. As an example, Nishizawa et al. reported
is of primary importance in the fields of analytical chemistry,
a very sensitive detection of an anion on the basis of pyrene excimer
clinical biochemistry, medicinal science, and environment
formation, by which the monomer (unbound) and excimer (bound)
science. In practice, a variety of fluorescent receptors have
fluorescence intensity ratio can be employed as a quantitative
measure of the concentration of the targeted ion.10
been synthesized through supramolecular approaches in the
past decades.1–4 In addition to such advantages of fluorescence
4-(N,N-Dimethylamino)benzonitrile
(DMABN)
and
sensors, recent improvements of the instrumental sensitivities
and spatial/temporal resolutions in fluorescence spectroscopy
have led to advances in the research on fluorescent sensors. As
an example, photoinduced electron transfer (PET) is often
used as a key step in fluorescent sensing.5–9 A PET sensor is
based essentially on ON/OFF of PET fluorescence quenching
of the sensor by a targeted ion/molecule and the fluorescence
intensity at a certain wavelength is employed for quantitation
of the ion/molecule. It is certainly true that PET sensors are
capable of sensing a trace amount of a targeted ion/molecule.
However, such an approach is not necessarily convenient for
quantitative analysis since one should monitor the absolute
fluorescence intensity or quantum yield, which is very sensitive
to experimental conditions: excitation wavelength/power, a
photomultiplier voltage, excitation/detection slit width of an
apparatus, and so forth. Clearly, another approach other than
4-(N,N-dimethylamino)benzoate (DMAB) are known to show
characteristic dual fluorescence in polar solvents. Therefore, a
variety of studies on these molecules have been reported11–14
since the first report on the dual fluorescence from DMABN by
Lippert et al.15 Such dual fluorescence phenomena have been
reported to arise from the participation of two structurally-
different excited singlet states. While several models on the
dual fluorescence have been proposed,16 a ‘‘twisted intra-
molecular charge transfer (TICT)’’ model by Grabowski has been
widely accepted,17,18 Fig. 1. In the case of DMABN in the ground
state, N,N-dimethylamino and benzonitrile moieties are almost
planar leading to efficient electron conjugation between the two
moieties. The excited singlet state produced by light absorption is
called the locally excited (LE) state and the two moieties are
mutually planar, while solvent relaxation around the molecule
brings about concomitant rotation of the N,N-dimethyl-
amino group against the benzonitrile plane until it is twisted
about 901 and conjugation between the two parts is lost,
producing the TICT excited state. In the TICT excited state,
charge separation takes place between the N,N-dimethylamino
Department of Chemistry, Graduate School of Science,
Hokkaido University, Sapporo 060-0810, Japan.
E-mail: kitamura@sci.hokudai.ac.jp
ꢂc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 6641–6649 | 6641

View Online
Fig. 2 Schematic illustration of molecular recognition of an amino
acid by the guanidinium and crown-ether groups in S.
Fig. 1 Energy diagram of TICT formation in DMABN.
backbone, respectively, as the binding units for the
carboxylate25,26 and ammonium groups6,27 in an amino acid,
respectively. As a fluorescence sensor unit, a DMAB group is
introduced at the 12-position of the cholic acid. Our molecular
modeling suggests that S will bind with an amino acid as
shown in Fig. 2, and we anticipate that this allows the changes
in the local polarity and/or steric hindrance around the
DMAB chromophore. Therefore, we expect that S can sense
molecular recognition of an amino acid sensitively and
quantitatively on the basis of the change in the TICT/LE
fluorescence intensity ratio.
and cyanophenyl or carbonylphenyl groups in DMABN or
DMAB, respectively. The LE and TICT excited states are in
equilibrium, and both LE and TICT fluorescence bands are
observed with the latter being observed at a longer wavelength as
compared with the former. Since the rotation of the N,N-
dimethylamino group and stabilization/charge separation of the
TICT state by polar solvents are necessary, the dual fluorescence
behavior is very sensitive to micropolarity and/or steric
hindrance around the molecule.19 Although TICT-type
compounds are very unique as a fluorescence probe for
microenvironments as mentioned above, the number of
reports on a TICT fluorescence sensor for molecular recognition
is still limited.20
In the present study, we focus our attention to reveal
experimentally the capability of the ratiometric fluorescence
sensing of an amino acid by the cholic acid having a DMAB
chromophore. In the first step of the study mentioned above,
therefore, we synthesized the sensor 1 without a crown-ether
group as shown in Scheme 2. We report here the synthesis of a
novel TICT-type fluorescent sensor, 1, and the ratiometric
sensing of an amino acid based on the change in the fluorescence
intensity ratio (ITICT/ILE) of the DMAB chromophore.
In the present study, we explored sensing of an amino acid
based on a TICT-type fluorescence probe. For this purpose,
we have designed the cholic acid derivative having a TICT
chromophore and the binding sites for an amino acid, S, in
Scheme 1. Cholic acid possesses three hydroxyl groups on its
rigid steroidal backbone and is widely used as a scaffold for
various sensors, since the hydroxyl group(s) can be replaced
synthetically by various functional groups.21–24 In the sensor
S, we have employed the guanidinium and crown-ether group
introduced at the 3- and 7-positions of the cholic acid
Experimental section
Synthesis of fluorescent sensor 1
Synthetic routes to the sensor 1 are shown in Scheme 2.
In the first step, the reactive hydroxyl groups at the 3- and
7-positions of cholic acid methyl ester (2) were protected by
acetyl groups, giving 3. A DMAB group was then introduced
to the 12-position of 3. After hydrolysis of the acetyl group at
the 3-position of 4 by hydrochloric acid, a Mitsunobu reaction
was employed to synthesize 6 with methanesulfonic acid as a
nucleophile and the derivative was converted to the azide
derivative, 7. The azido group in 7 was reduced by a zinc
powder, and then a guanidinium unit was introduced to 8 by
using 10 as a reagent,28 giving the sensor 1.
All of the chemicals used for the synthesis of 1, purchased
from Wako Pure Chemical Ind., Kanto Chemical Co. Inc.,
Tokyo Kasei Kogyo Co. Ltd., or Sigma-Aldrich Co., were
used as supplied. 4-(N,N-Dimethylamino)benzoic acid ethyl
ester (11) was purchased from Tokyo Kasei Kogyo Co. Ltd.
and purified by repeated recrystallizations from ethanol. For
fluorescence sensing of an amino acid by 1, we employed
water-to-1,2-dichloroethane extraction experiments as described
later in detail. For this purpose, we employed N-acetyl-
D-phenylalanine (AcPhe, Advanced ChemTech Co. Inc.,
Scheme 1 Structure of sensor S.
ꢂc
This journal is the Owner Societies 2010
6642 | Phys. Chem. Chem. Phys., 2010, 12, 6641–6649

View Online
over anhydrous MgSO4. After evaporation of the solvents
under reduced pressure, the colorless solids were recrystallized
from methanol, affording 3 as colorless needles (22.6 g, 75%).
Rf = 0.31 (SiO2, chloroform/ethyl acetate = 9/1); mp
183.0–186.0 1C;
n
max/cmꢁ1 3533 (O-H), 2945 (CH2),
1737 (CQO), 1249 (C-O); dH(CDCl3) 0.69 (3H, s, 18-CH3),
0.93 (3H, s, 19-CH3), 0.98 (3H, d, J = 5.9 Hz, 21-CH3), 2.03
(3H, s, 3a-CO2CH3), 2.07 (3H, s, 7a-CO2CH3), 3.67 (3H, s,
24-CO2CH3), 4.00 (1H, m, 12b-H), 4.59 (1H, m, 3b-H), 4.89
(1H, m, 7b-H).
Methyl
3a,7a-diacetoxy-12a-[4-(N,N-dimethylamino)-
benzoyloxy]-5b-cholan-24-oate (4). An oven-dried Schlenk tube
was evacuated and filled subsequently with Ar gas. These
procedures were conducted twice. Into the tube were added
3 (6.1 g, 0.012 mol), dry dichloromethane (15 ml), N-ethyl-
diisopropylamine (2.5 ml, 0.015 mol, 1.2 equiv.), and 4-(di-
methylamino)benzoyl chloride (5.3 g, 0.029 mol, 2.4 equiv.).
After stirring at room temperature for 10.5 h, the reaction
mixture was evaporated under reduced pressure and the crude
product was purified successively by column chromatography
(SiO2, chloroform/ethyl acetate = 9/1, v/v) and recrystallizations
from ethanol, yielding 4 as colorless needles (6.9 g, 88%).
Rf = 0.76 (SiO2, chloroform/ethyl acetate = 9/1, v/v); mp
198.6–201.8 1C; nmax/cmꢁ1: 2946 (CH2), 1728 (CQO), 1378
(Ar), 1276 (C-O); dH(CDCl3) 0.79 (3H, s, 18-CH3), 0.81
(3H, d, J = 6.5 Hz, 21-CH3), 0.94 (3H, s, 19-CH3), 1.96
(3H, s, 3a-CO2CH3), 2.12 (3H, s, 7a-CO2CH3), 3.08 (6H, s,
N-CH3), 3.62 (3H, s, 24-CO2CH3), 4.49 (1H, m, 3a-H), 4.96
(1H, m, 7a-H), 5.27 (1H, m, 12a-H), 6.71 (2H, d, J = 8.9 Hz,
Ar-H meta), 7.96 (2H, d, J = 9.5 Hz, Ar-H ortho); elemental
analysis calcd (%) for C38H55NO8: C69.80, H8.48, N2.14;
found: C69.65, H8.35, N2.04.
Scheme 2 Synthetic route to sensor 1: (a) pyridine, Ac2O, toluene, rt,
24 h, (b) DMAB–Cl, i-Pr2NEt, CH2Cl2, rt, 10.5 h, (c) conc. HCl,
MeOH, rt, 6 h, (d) MsOH, TPP, DMAP, DEAD, THF, rt, 22 h,
(e) NaN3, DMPU, 60 1C, 7.5 h, (f) Zn, AcOH, rt, 1 h, (g) 10, NaHCO3,
Methyl 3a-azido-7a-acetoxy-12a-[4-(N,N-dimethylamino)-
benzoyloxy]-5b-cholan-24-oate (7). Into a suspension of 4
(1.00383 g, 1.54 mmol) in methanol (25 ml) was added 6 ml
of conc. HCl, resulting in a homogeneous solution. After
stirring at room temperature for 6 h, the mixture was poured
into water (30 ml) and the products were extracted with
dichloromethane (40 ml ꢀ 3). The combined organic layer
was washed successively with aqueous HCl (10%, 100 ml) and
NaHCO3 (5%, 100 ml) solutions, and then with distilled water
(100 ml ꢀ 3). The organic layer dried over anhydrous MgSO4
was evaporated under reduced pressure, giving crude 5 as
colorless solids, which were used for the following reaction
without further purification. Rf = 0.24 (SiO2, chloroform/
ethyl acetate = 9.5/0.5, v/v); dH(CDCl3) 0.79 (3H, s, 18-CH3),
0.83 (3H, d, J = 6.5 Hz, 21-CH3), 0.93 (3H, s, 19-CH3), 2.13
(3H, s, 7a-CO2CH3), 3.06 (6H, s, N-CH3), 3.39 (1H, m, 3b-H),
3.61 (3H, s, 24-CO2CH3), 4.95 (1H, m, 7b-H), 5.28 (1H, m,
12b-H), 6.68 (2H, d, J = 8.9 Hz, Ar-H meta), 7.93 (2H, d,
J = 9.2 Hz, Ar-H ortho).
DMF, 80 1C, 2 h, then 1 M NaOHaq, (h) 0.5 M HClaq
.
used as supplied) as an amino acid owing to the solubility
of the AcPhe:1 complex in 1,2-dichloroethane. Column chromato-
graphy was carried out by using Merck silica gel 60 (particle
size 0.063–0.200 mm) or Wakogel 50C18 (particle size
0.038–0.063 mm).
1H NMR spectra were recorded on a JEOL JME-EX270
FT-NMR system (270 MHz). The chemical shifts of the
spectra determined in CDCl3 were given in ppm with tetra-
methylsilane being an internal standard (0.00 ppm). IR spectra
were recorded on a Shimadzu FTIR-8300 spectrometer.
Methyl 3a,7a-diacetoxy-12a-hydroxy-5b-cholan-24-oate (3).
According to the previous report,29 pyridine (28.0 ml,
0.348 mol, 5.9 equiv.) and acetic anhydride (28.0 ml, 0.296 mol,
5.0 equiv.) were added to a suspension of cholic acid methyl
ester (2, 25.1 g, 0.0594 mol) in toluene (135 ml). Upon
prolonged stirring at room temperature, the reaction mixture
became gradually a homogeneous solution. After 24 h stirring,
the reaction mixture was diluted with dichloromethane
(100 ml), washed with distilled water (150 ml ꢀ 3), and dried
Methanesulfonic acid (0.25 ml, 3.9 mmol, 2.5 equiv.) was
added to a dry tetrahydrofuran solution (16 ml) of crude 5,
triphenylphosphine (1.28880 g, 4.91 mmol, 3.2 equiv.), and
4-dimethylaminopyridine (0.38171 g, 3.12 mmol, 2.0 equiv.)
under an argon gas atmosphere. A toluene solution of diethyl
azodicarboxylate (40%, 2.01761 g, 4.63 mmol, 3.0 equiv.) was
ꢂc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 6641–6649 | 6643

View Online
added dropwise to the reaction mixture over 10 min, resulting
in a turbid and viscous solution. The mixture was stirred at
room temperature for 22 h. The solvent was evaporated under
reduced pressure and the residues were purified by column
chromatography (SiO2, chloroform/ethyl acetate = 9.5/0.5, v/v),
yielding 6 as a colorless oil. Rf = 0.58 (SiO2, chloroform/
ethyl acetate = 9.5/0.5, v/v); dH(CDCl3) 0.80 (3H, s, 18-CH3),
0.83 (3H, d, J = 6.8 Hz, 21-CH3), 0.99 (3H, s, 19-CH3),
2.17 (3H, s, 7a-CO2CH3), 2.95 (3H, s, SO2-CH3), 3.07 (6H, s,
N-CH3), 3.61 (3H, s, 24-CO2CH3), 4.87 (1H, m, 3a-H), 4.99
(1H, m, 7b-H), 5.29 (1H, m, 12b-H), 6.68 (2H, d, J = 8.9 Hz,
Ar-H meta), 7.92 (2H, d, J = 9.5 Hz, Ar-H ortho).
stirred vigorously at 80 1C for 2 h and then allowed to cool to
room temperature. Dichloromethane (40 ml) was added to the
reaction mixture, and the organic layer was washed with an
aqueous NaOH (1 M, 40 ml ꢀ 2) solution and distilled water
(40 ml). Crude 9 was obtained as a dichloromethane solution.
The sample for spectroscopic measurements was obtained by
purification of the solution by column chromatography (Al2O3,
dichloromethane then dichloromethane/methanol = 9/1, v/v)
for two times, giving 9 as a colorless solid. Mp 200.0 1C (dec.);
n
max/cmꢁ1 2952 (CH2), 1731 (CQO), 1379 (Ar), 1278 (C-O);
dH(CDCl3) 0.78 (3H, s, 18-CH3), 0.82 (3H, d, J = 6.5 Hz,
21-CH3), 0.93 (3H, s, 19-CH3), 2.15 (3H, s, 7a-CO2CH3),
3.00 (6H, s, N-CH3), 2.80–3.18 (1H, m, 3b-H), 3.51 (4H, m,
CH2’s at the 4- and 5-positions of the imidazolinium ring),
3.61 (3H, s, 24-CO2CH3), 4.93 (1H, m, 7b-H), 5.29 (1H, m,
12b-H), 6.68 (2H, d, J = 9.0 Hz, Ar-H meta), 7.92 (2H, d,
J = 8.9 Hz, Ar-H ortho); elemental analysis calcd (%) for
C39H58N4O6 + 0.5H2CO3: C66.83, H8.38, N7.89; found:
C66.82, H8.05, N7.54.
Sodium azide (0.30532 g, 4.70 mmol, 3.1 equiv.) was added
to
6 dissolved in 1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-
pyrimidinone (5 ml) and the mixture was stirred at 60 1C for
7.5 h with a CaCl2 tube being equipped to the flask. After the
reaction, the mixture was poured into dichloromethane
(40 ml). The organic layer was washed with distilled water
(150 ml ꢀ 3), dried over anhydrous MgSO4, and evaporated to
dryness under reduced pressure. Successive purification of the
crude product by column chromatography (SiO2, chloroform)
and recrystallizations from ethanol afforded 7 as colorless
needles (0.60443 g, 66%). Rf = 0.68 (SiO2, chloroform/ethyl
acetate = 9.5/0.5, v/v); mp 183.6 1C (dec.); nmax/cmꢁ1 2947
(CH2), 2085 (N3), 1697 (CQO), 1371 (Ar), 1275 (C-O);
dH(CDCl3) 0.79 (3H, s, 18-CH3), 0.85 (3H, d, J = 6.5 Hz,
21-CH3), 0.94 (3H, s, 19-CH3), 2.15 (3H, s, 7a-CO2CH3), 3.02
(1H, m, 3b-H), 3.06 (6H, s, N-CH3), 3.61 (3H, s, 24-CO2CH3),
4.97 (1H, m, 7b-H), 5.32 (1H, m, 12b), 6.69 (2H, d, J = 8.9 Hz,
Ar-H meta), 7.93 (2H, d, J = 8.9 Hz, Ar-H ortho); elemental
analysis calcd (%) for C36H52N4O6: C67.90, H8.23, N8.80;
found: C67.89, H8.21, N8.66.
A dichloromethane solution of 9 was treated with an
aqueous HCl (0.5 M, 40 ml ꢀ 2) solution and distilled water
(40 ml). After evaporation of the solution under reduced
pressure, crude sensor 1 was obtained as colorless solids
(0.19003 g, 67%). The sample for spectroscopic measurements
was obtained by purification of the solution by column
chromatography (50C18, methanol/water = 8/2, v/v) for
two times, giving 1 as a colorless solid. Mp 159.0 1C (dec.);
n
max/cmꢁ1 2947 (CH2), 1732 (CQO), 1377 (Ar), 1276 (C-O);
dH(CDCl3) 0.79 (3H, s, 18-CH3), 0.84 (3H, d, J = 6.5 Hz,
21-CH3), 0.94 (3H, s, 19-CH3), 2.30 (3H, s, 7a-CO2CH3),
3.06–3.12 (6H, m, N-CH3), 3.55–3.70 (1H, m, 3b-H),
3.59–3.63 (4H, m, CH2’s at the 4- and 5-positions of the
imidazolinium ring), 3.61 (3H, s, 24-CO2CH3), 4.94 (1H, m,
7b-H), 5.32 (1H, m, 12b-H), 6.34 (1H, m, NH at the 1- or
3-position of the imidazolinium ring), 6.66–7.10 (2H, m, Ar-H
meta), 7.96 (2H, d, J = 8.9 Hz, Ar-H ortho), 8.29 (1H, m, NH
at the 1- or 3-position of the imidazolinium ring).
Methyl 3a-(4,5-dihydro-1H-imidazol-2-yl)-7a-acetoxy-12a-
[4-(N,N-dimethylamino)benzoyloxy]-5b-cholan-24-oate hydro-
chloride (1). Acetic acid (15 ml) was added to a mixture of 7
(0.25314 g, 0.398 mmol) and a zinc powder (0.59286 g,
9.07 mmol, 23 equiv.), which was purified by the accepted
procedures.30 The mixture was protected from air moisture by
equipping a CaCl2 tube with the flask and stirred vigorously at
room temperature for 1 h. The zinc powder was removed by
filtration and washed with acetic acid (2 ml ꢀ 2). The filtrates
were evaporated under reduced pressure. Acetic acid contaminated
in the filtrate was removed by azeotropic distillation with
toluene under reduced pressure. The colorless oil was
dissolved in CH2Cl2 (40 ml), and the solution was washed
with an aqueous NaHCO3 solution (5%, 40 ml ꢀ 2), distilled
water (40 ml), and dried over anhydrous MgSO4. After
evaporation of the solvents under reduced pressure, crude 8
was obtained as a colorless solid. dH(CDCl3) 0.78 (3H, s,
18-CH3), 0.83 (3H, d, J = 6.5 Hz, 21-CH3), 0.93 (3H, s,
19-CH3), 2.12 (3H, s, 7a-CO2CH3), 2.44 (1H, m, 3b-H), 3.07
(6H, s, N-CH3), 3.61 (3H, s, 24-CO2CH3), 4.95 (1H, m, 7b-H),
5.28 (1H, m, 12b), 6.68 (2H, d, J = 7.3 Hz, Ar-H meta), 7.95
(2H, d, J = 7.3 Hz, Ar-H ortho).
Spectroscopic measurements
Absorption and fluorescence spectra were recorded on a
Hitachi UV-3300 spectrometer and a Hitachi F-4500 spectro-
fluorometer, respectively. The observed fluorescence spectrum
of 1 or 11 was separated into Gaussian functions and the
fluorescent quantum yields of the LE (FLE) and TICT (FTICT
)
bands were determined on the basis of the separated spectra by
using 9,10-diphenylanthracene in cyclohexane as a standard
(F = 0.91).31 Refractive index correction was made to
evaluate FLE and FTICT in a given solvent. A temperature
controlled study on the fluorescence spectrum of 1 was
conducted by using the spectrofluorometer mentioned above
and a temperature-controlled sample cell holder, whose
temperature was controlled and monitored by using a water
circulator (Fine, FR-007N) and a thermocouple (IWATSU,
SC-0107 and VOAC7522), respectively. All of the fluorescence
spectra reported in the present study were obtained by 300 nm
excitation.
4,5-Dihydro-1H-imidazolium 2-sulfonate (10, 0.22392 g,
1.49 mmol, 3.7 equiv.) was added to a suspension of 8 and
NaHCO3 (0.42567 g, 5.07 mmol, 13 equiv.) in N,N-dimethyl-
formamide (4 ml) at 80 1C under stirring. The mixture was
Fluorescence decay measurements were conducted by using
a time-correlated single-photon counting system. Optical
ꢂc
This journal is the Owner Societies 2010
6644 | Phys. Chem. Chem. Phys., 2010, 12, 6641–6649

View Online
parametric amplification (Coherent, model 9400) of the output
pulses from a mode-locked Ti:sapphire laser (Coherent, Mira
model 900-F) gave 300 nm laser pulses (repetition rate, 100 kHz;
fwhm, 200 fs autocorrelation trace) as an excitation light
source. By using a Soleil-Babinet compensator (Melles Griot),
the polarized direction of the excitation laser beam was set at a
magic angle (35.31) or vertical direction for fluorescence decay
measurements. Fluorescence from the sample was detected by
using a microchannel-plate photomultiplier (Hamamatsu,
R3809U-50) equipped with a monochromator (Jobin Ybon,
H-20) and analyzed by a single-photon counting module
(Edinburgh Instruments, SPC-300). The excitation pulse
profile was measured at 300 nm by using a dye-free polystyrene
resin as a scattering material. Decay profiles were analyzed by
using an iterative nonlinear least-squares deconvolution
method. Since the fluorescence from 1 or 11 was not quenched
by O2 owing to very short fluorescence lifetimes of the
derivatives, we employed aerated samples for all spectroscopic
measurements.
Results and discussion
General features of the spectroscopic and photophysical
properties of 1 and 11
Fig. 3 shows the absorption and fluorescence spectra of 1 (a)
and 11 (b) in various solvents. Both 1 and 11 showed p–p*
absorption transitions at around 310 nm. The close similarities
of the absorption band between 1 and 11 demonstrate that the
absorption transition of 1 in this wavelength region is ascribed
safely to that of the 4-(N,N-dimethylamino)benzoate group in
the compound. The fluorescence spectra of 1 and 11 exhibited
two peaks as seen in Fig. 3, where the fluorescence intensity of
1 or 11 in a given solvent was normalized to that at the shorter-
wavelength fluorescence maximum wavelength: 340–350 nm.
In acetonitrile (dielectric constant (e) = 37.5), 11 showed
fluorescence bands at around 350 and 480 nm (shown by the
black curve in Fig. 3(b)). The longer-wavelength fluorescence
maximum of 11 shifted gradually to the shorter wavelength
with a decrease in a solvent polarity and, in toluene (e = 2.379,
shown in the purple curve in Fig. 3), the longer-wavelength
fluorescence band was observed as the fluorescence shoulder of
the shorter-wavelength band. Furthermore, the shorter wave-
length shift of the longer-wavelength fluorescence band with a
decrease in the e value accompanied an increase in the
fluorescence intensity ratio of the shorter-wavelength band
to the longer-wavelength band. Such spectroscopic characteristics
of 11 have been already reported and explained by the TICT
model in Fig. 1, with the shorter- and longer-wavelength
fluorescence bands being assigned to the LE and TICT
fluorescence, respectively.32
Fig. 3 Absorption (broken line) and fluorescence (solid line) spectra
of 1 (a) and 11 (b) in acetonitrile (black), dichloromethane (red),
tetrahydrofuran (green), ethyl acetate (blue), diethyl ether (light blue),
and toluene (purple). The fluorescence intensity in a given solvent is
normalized to that at the LE fluorescence maximum wavelength.
fluorescence intensity ratio (ITICT/ILE) observed for 1 in a
given solvent was much smaller than the corresponding value
of 11. In acetonitrile, as an example, the ITICT/ILE value
observed for 1 was 0.125, while that of 11 was 7.79. It is worth
emphasizing, furthermore, that, although 11 showed a decrease
in the ITICT/ILE value with a decrease in solvent polarity,
1 showed an opposite trend to that observed for 11: an
increase in the ITICT/ILE value. These results demonstrate
explicitly that the peripheral structures around the DMAB
group in 1 influence considerably the TICT phenomena. Before
discussing fluorescence sensing behaviors of 1 toward an amino
acid, therefore, the excited state characteristics of 1 are worth
comparing with those of 11.
The CT nature of the excited states of both 1 and 11 can be
checked on the basis of the solvent polarity dependence of the
Stokes shift (Dn) of the compound, defined by the difference
between the absorption (nabs) and fluorescence (nf) peak
energies (Dn = nabs ꢁ nf). According to Lippert and Mataga,
Dn should correlate linearly with a solvent polarity parameter
(Df 0) given in eqn (1),33–35
Since the absorption transition at around 310 nm in 1 is
localized to the DMAB group as discussed above, the
compound should exhibit dual fluorescence similar to 11. In
practice, 1 in acetonitrile showed the LE and TICT fluorescence
at around 350 and 480 nm, respectively, as seen in Fig. 3(a). A
decrease in a solvent polarity from acetonitrile to toluene gave
rise to the shorter wavelength shift of the TICT band of 1,
similar to that observed for 11. Nevertheless, the TICT/LE
ꢀ
ꢁ
e ꢁ 1
1
2
n2 ꢁ 1
Df 0 ¼
ꢁ
ð1Þ
2e þ 1
2n2 þ 1
where n represents the refractive index of a solvent. The first
and second terms in eqn (1) imply the contributions of the
ꢂc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 6641–6649 | 6645

View Online
restricted as compared with 11 owing to the presence of the
guanidinium group in 1. Furthermore, we have confirmed that
the photophysical parameters of 4 and 9 (see also Scheme 2)
are very similar to those of 11. The photophysical characteristics
of 1 are thus not ascribed to the bulkiness of the guanidinium
group, but to the cationic charge of the guanidinium group in
1. When an amino acid binds to the guanidinium group in 1,
therefore, the fluorescence characteristics of the compound
should vary, by which we can sense the amino acid by
fluorescence spectroscopy.
Fluorescence sensing of amino acid by 1
A 1,2-dichloroethane solution of 1 (4.80 ꢀ 10ꢁ6 M) and an
aqueous N-acetyl-D-phenylalanine solution (AcPhe, 0–1.56 ꢀ
10ꢁ4 M) were mixed vigorously and, after the distribution
equilibrium, the two phases were separated. The absorption
and fluorescence spectra of 1 in the organic layer before and
after liquid/liquid extraction were measured to evaluate the
sensing ability of 1 toward AcPhe. The results are shown in
Fig. 5, where the fluorescence intensity in Fig. 5(b) at a given
AcPhe concentration is normalized to those predicted from the
incident photon number absorbed by 1 at the excitation
wavelength (300 nm). With an increase in the concentration
of AcPhe in the aqueous phase, the absorbance of 1 in the oil
phase at around 310 nm increased (Fig. 5(a)), demonstrating
an increase in the water-to-oil distribution coefficient of 1 in
the presence of AcPhe. As seen in Fig. 5(b), furthermore, this
accompanied the increase and decrease in the LE (350 nm) and
TICT (440 nm) fluorescence intensities, respectively, and the
fluorescence intensity ratio, R = I440/I350, decreased with an
increase in the AcPhe concentration. In practice, the I440/I350
ratio decreased very sharply with an increase in the concentration
of AcPhe as shown in Fig. 6. To reveal the fluorescence sensing
ability of 1 toward AcPhe, we conducted the following kinetic
analysis.
Fig. 4 Solvent polarity dependences of the Stokes shifts (nabs ꢁ nf)
observed for the LE (square) and TICT (circle) fluorescence for
1 (filled) and 11 (blank).
dipole orientation and electronic polarizability of a solvent
molecule, respectively, to solvent-induced stabilization of the
ground- and excited-states of a molecule concerned.
Fig. 4 shows the relationships between Dn and Df0 for both
LE (square) and TICT (circle) fluorescence of 1 (filled) and 11
(opened). It is clear from the figure that each plot shows a
good linear correlation between these parameters and the
slope value of the Dn (TICT) vs. Df0 plot for 1 or 11 is much
larger than the corresponding value of the Dn (LE) vs. Df0 plot,
which supports the assignment of the shorter and longer
wavelength fluorescence as LE and TICT fluorescence,
respectively. Fig. 4 also demonstrates that the slope value of
the Dn (TICT) vs. Df0 plot is almost comparable between 1 and
11 with the slope values being 17100 and 20700 cmꢁ1
,
respectively. Therefore, it is concluded that the electronic
structures of the TICT excited state produced via the LE state
are very similar for both 1 and 11. This can be also confirmed
by the photophysical parameters of the compounds that are
summarized in Table 1. The fluorescence lifetime (t), quantum
yield (F), and the fluorescence (radiative) rate constant (kr) of
the TICT excited state of 1 are very similar to the corresponding
values of 11: t = 3.5–3.8 ns, F = 0.025–0.032, and
kr = (0.71–0.84) ꢀ 107 sꢁ1. The results agree completely with
the conclusion described above: the electronic structures of
the TICT excited state are very similar for both 1 and 11.
Contrarily, the F and kr values of the LE state of 1 were larger
than the corresponding values of 11. It is worth pointing out
that the t value and the nonradiative decay rate constant (knr)
of the LE state of 1 are almost comparable with those of 11.
Therefore, these results indicate that TICT formation in 1 is
Under the assumption of 1 : 1 binding between 1 as a host
(H) and AcPhe as a guest (G), the binding constant, K, can be
given by eqn (2),
½HGꢃ
K ¼
ð2Þ
½Hꢃ½Gꢃ
where [H], [G], and [HG] are the concentrations of H, G, and
an H–G complex, respectively. When the fluorescence intensity
ratio of R = I440/I350 in the H–G complex is proportional to
the mole fraction of 1, the observed R value at a given [G]
(Robs) in the organic phase is given by eqn (3),
(
R
1 ꢁ R0
1
Robs ¼ R0 þ
½Hꢃ0 þ ½Gꢃ0 þ
2½Hꢃ0
Kass
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
)
ꢀ
ꢁ
2ꢁ4½Hꢃ0½Gꢃ0
ð3Þ
1
Table 1 Spectroscopic properties of 1 and 11 in ethyl acetate
ꢁ
½Hꢃ0 þ ½Gꢃ0 þ
Kass
1
11
LE
TICT
LE
TICT
where R0 and RN are the R values in the absence and presence
of an infinite amount of AcPhe, respectively. [H]0 and [G]0 are
the concentrations of 1 in the organic phase and AcPhe in the
aqueous phase before liquid/liquid extraction, respectively.
The data in Fig. 6 were then fitted by eqn (3) as the results
t/ns
1.6
0.037
2.3
3.5
1.6
0.0031
0.20
3.8
F
0.025
0.71
—
0.032
0.84
—
kr/107 sꢁ1
knr/107 sꢁ1
60
64
ꢂc
This journal is the Owner Societies 2010
6646 | Phys. Chem. Chem. Phys., 2010, 12, 6641–6649

View Online
1 and AcPhe to be 7.0 ꢀ 105 Mꢁ1 (log Kass = 5.8). The value is
sufficiently large for sensitive analysis of the amino acid
through H–G complexation. The high ionization coefficient
of AcPhe in a water phase (pKa = 3.7)36 and high distribution
coefficient of the host–guest complex to the organic phase
enabled such sensitive analysis.
Photophysical origin of fluorescence sensing of amino acid
The decrease in the I440/I350 value in the presence of AcPhe in
Fig. 5 and 6 demonstrates that binding of the amino acid with
the guanidinium group in 1 influences largely the TICT
formation process as expected from the results by the
molecular modeling, Fig. 2. This should reflect on the activation
barrier for the TICT formation process. Therefore, we studied
a
temperature (T) dependence of the TICT formation
efficiency for the 1,2-dichloroethane (DCE) solution of 1 after
the distribution equilibrium between the organic phase (DCE)
and an aqueous AcPhe solution, together with that in the absence
of the amino acid in the aqueous phase. Since a variation of T
brings about both TICT and LE fluorescence shifts, we employed
the fluorescence quantum yield ratio, FTICT/FLE, for the
following discussion: see also Experimental section.
Fig. 7 shows the fluorescence spectra of 1 in the organic
layer after liquid/liquid extraction in the absence (a) and
presence of AcPhe (b) as a function of T (22–65 1C). In both
cases, T elevation gave rise to a shorter wavelength shift of the
TICT fluorescence band and the increase in the FTICT/FLE
value. The shorter wavelength shifts of the TICT fluorescence
of 1 in both absence and presence of the amino acid upon
T elevation can be readily understood by the decrease in the
medium polarity. For further detailed discussion, the relation-
ships between the FTICT/FLE value vs. Tꢁ1 of 1 in the absence
and presence of AcPhe are shown in Fig. 8. It is very clear that
the FTICT/FLE value increases linearly with increasing T. It is
worth pointing out that the slope value of the plot in Fig. 8 is
larger in the presence of the amino acid (1530 K) as compared
with that in the absence of the amino acid (1280 K),
demonstrating the large influence of amino acid binding with
the guanidinium group in 1 to the TICT formation process.
According to the kinetic model by Grabowski et al.,37,38 the
Fig. 5 Absorption (a) and fluorescence (b) spectra of the 1,2-dichloro-
ethane layer after liquid/liquid extraction. [1]0 = 4.80 ꢀ 10ꢁ6 M in
1,2-dichloroethane. [AcPhe]0 = 0, 1.56, 7.80, 15.6, 78.0, and 156 ꢀ 10ꢁ6 M
in water.
F
TICT/FLE value should exhibit a bell-shaped T dependence,
in which the slope of the FTICT/FLE value vs. Tꢁ1 plot becomes
positive or negative above or below a certain critical temperature
(Tc), respectively, owing to the participation of the LE
state–TICT state equilibrium, Fig. 9. Since the TICT state
formation process from the LE state is dominant below Tc
(i.e., above Tcꢁ1), the slope of the FTICT/FLE vs. Tꢁ1 becomes
negative and the slope value corresponds to the activation
energy (Ea) of the TICT formation process: ka in Fig. 9 and
eqn (4).
ln(FTICT/FLE) = ꢁEa/RT + const.
(4)
Fig. 6 Fluorescence (I440/I350) titration curve upon addition of AcPhe
Above Tc, contrarily, since the backward reaction from the
TICT state to the LE state contributes largely to the LE
state–TICT state equilibrium, the slope of the FTICT/FLE vs.
Tꢁ1 plot should be positive with the slope value being
expressed by eqn (5),
in an aqueous layer. The data were taken from Fig. 5.
were shown by the solid curve. It is very clear that the observed
AcPhe concentration dependence of I440/I350 is reproduced
very well by eqn (3) (correlation coefficient = 0.981) and the
fitting gives the apparent binding constant Kass between sensor
ln(FTICT/FLE) = ꢁDH/RT + const.
(5)
ꢂc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 6641–6649 | 6647

View Online
Fig. 9 Kinetic scheme of the TICT phenomena.
Tc value was not observed upto 70 1C, we could not evaluate
the DH value. Nonetheless, the slope values in Fig. 8 demonstrate
that the activation energies of the forward TICT state
formation process in the absence and presence of AcPhe are
10.6 and 12.7 kJ molꢁ1, respectively. The data indicate that the
increase in the activation energy of the forward reaction (Ea)
results in the decrease in the FTICT/FLE value upon molecular
recognition of AcPhe. As a result, the change in the I440/I350 or
FTICT/FLE value upon addition of AcPhe can be explained by
molecular recognition of AcPhe by the guanidinium group in 1.
Conclusions
Fig. 7 Temperature dependences (22–65 1C) of fluorescence spectra
of 1 (4.03 ꢀ 10ꢁ5 M) in the organic layer after liquid/liquid extraction
in the absence (a) and presence (b) of AcPhe (5.12 ꢀ 10ꢁ4 M) in the
water phase.
We synthesized a novel TICT-type fluorescent sensor for an
amino acid derivative on the basis of a DMAB chromophore,
and studied its spectroscopic and sensing properties. Sensor 1
showed TICT-type dual fluorescence, however the sensor
exhibited an unusual solvent polarity dependence of the
TICT/LE fluorescence intensity ratio, different from that of
a model compound, 11. The cationic charged guanidinium
group in the proximity of the DMAB group played an
essential role in the characteristic solvent effects of 1.
The fluorescence intensity ratio (I440/I350) decreased with an
increase in the amino acid (AcPhe) concentration in the
aqueous phase before liquid/liquid extraction. The apparent
binding constant between 1 and AcPhe was determined to be
log Kass = 5.8 by fluorescence intensity titration experiments.
The Kass value was sufficiently large for molecular recognition
of an amino acid. The temperature dependence of the
fluorescence spectrum demonstrated that the change in the
TICT formation efficiency of 1 upon recognition of AcPhe
could be due to the increase in the activation energy of the
TICT formation process. The novel fluorescent sensor 1 could
sense an amino acid sensitively and quantitatively on the basis
of the change in the fluorescence intensity ratio (ITICT/ILE) of
DMAB chromophore by using liquid/liquid extraction.
It is worth noting that analogous experiments with those of
D-type AcPhe were also conducted for L-type AcPhe and the
Fig. 8 Temperature dependence of FTICT/FLE of 1 in the absence
(open box) and presence of AcPhe (closed box). The solid lines show
the best fits by least-squares method.
K
ass value was evaluated to be 7.9 ꢀ 105 Mꢁ1 (log Kass = 5.9),
where DH corresponds to the standard enthalpy of the TICT
state formation process. In the present experiments, since the
whose value was almost the same as that determined for
D-type AcPhe: log Kass = 5.8. Since the sensor 1 does not possess
ꢂc
This journal is the Owner Societies 2010
6648 | Phys. Chem. Chem. Phys., 2010, 12, 6641–6649

View Online
the crown-ether group as a sensing unit for the ammonium
group in an amino acid, the compound cannot discriminate
chirality of the amino acid. Since the present study
demonstrated that the sensor 1 could act certainly as a
ratiometric fluorescence sensor toward an amino acid, we
expect that the sensor S having the crown-ether group shown
in Scheme 1 would sense chirality of a natural amino acid
(without N-acetyl group). Furthermore, although enzyme
fluorescence assay has been also employed to sense chirality
of an amino acid, it is required to follow the time dependence
of the fluorescence intensity during the enzymatic reaction.39
Since ratiometric fluorescence sensing is more straightforward
for analyzing a targeted ion/molecule, we believe that the
11 W. Schuddeboom, S. A. Jonker, J. M. Warman, U. Leinhos,
W. Kuhnle and K. A. Zachariasse, J. Phys. Chem., 1992, 96(26),
10809–10819.
12 S. Cogan, S. Zilberg and Y. Haas, J. Am. Chem. Soc., 2006,
128(10), 3335–3345.
13 Y.-P. Sun, T. L. Bowen and C. E. Bunker, J. Phys. Chem., 1994,
98(48), 12486–12494.
14 C.-H. Zhang, Z.-B. Chen and Y.-B. Jiang, Spectrochim. Acta, Part
A, 2004, 60(12), 2729–2732.
15 E. Lippert, W. Luder and H. Boos, in Advances in Molecular
¨
Spectroscopy, ed. A. Mangini, Pergamon Press, Oxford, 1962.
16 A. B. J. Parusel, G. Kohler and S. Grimme, J. Phys. Chem. A,
¨
1998, 102(31), 6297–6306.
17 Z. R. Grabowski, K. Rotkiewicz and W. Rettig, Chem. Rev., 2003,
103(10), 3899–4031.
18 K. Rotkiewicz, K. H. Grellmann and Z. R. Grabowski, Chem.
Phys. Lett., 1973, 19(3), 315–318.
19 S. I. Druzhinin, N. P. Ernsting, S. A. Kovalenko, L. P. Lustres,
T. A. Senyushkina and K. A. Zachariasse, J. Phys. Chem. A, 2006,
110(9), 2955–2969.
20 K. Kobiro and Y. Inoue, J. Am. Chem. Soc., 2003, 125(2), 421–427.
21 A. P. Davis and J.-B. Joos, Coord. Chem. Rev., 2003, 240(1–2),
143–156.
present experimental approach would provide
a more
convenient analytical tool for quantitation of an amino acid
over enzymatic fluorescence assay. Synthetic experiments of
the sensor S are now in progress in our research group and we
will report the results in near future.
22 L. J. Lawless, A. G. Blackburn, A. J. Ayling, M. N. Perez-Payan and
´ ´
A. P. Davis, J. Chem. Soc., Perkin Trans. 1, 2001, (11), 1329–1341.
23 S.-Y. Liu, L. Fang, Y.-B. He, W.-H. Chan, K.-T. Yeung,
Y.-K. Cheng and R.-H. Yang, Org. Lett., 2005, 7(26), 5825–5828.
Acknowledgements
24 L. Siracusa, F. M. Hurley, S. Dresen, L. J. Lawless, M. N. Pe
Payan and A. P. Davis, Org. Lett., 2002, 4(26), 4639–4642.
25 B. Baragana, A. G. Blackburn, P. Breccia, A. P. Davis, J. de
Mendoza, J. M. Padron-Carrillo, P. Prados, J. Riedner and
´
rez-
The authors thank Prof. Y. Sasaki and Prof. M. Kato at
Department of Chemistry, Graduate School of Science,
1
Hokkaido University for H NMR measurements. This work
´
´
was partly supported by the Global COE Program (Project
No. B01: Catalysis as the Basis for Innovation in Materials
Science) from the Ministry of Education, Culture, Sports,
Science and Technology, Japan.
J. G. de Vries, Chem.–Eur. J., 2002, 8(13), 2931–2936.
26 D. M. Kneeland, K. Ariga, V. M. Lynch, C.-Y. Huang and
E. V. Anslyn, J. Am. Chem. Soc., 1993, 115(22), 10042–10055.
27 P. Breccia, M. V. Gool, R. Prez-Fernndez, S. Martn-Santamara,
F. Gago, P. Prados and J. de Mendoza, J. Am. Chem. Soc., 2003,
125(27), 8270–8284.
28 M.-S. Muche and M. W. Gobel, Angew. Chem., 1996, 108(18),
¨
2263–2265 (Angew. Chem., Int. Ed. Engl., 1996, 35(18), 2126–2129).
29 L. F. Fieser and S. Rajagopalan, J. Am. Chem. Soc., 1950, 72(12),
5530–5536.
30 S. Paul and M. Gupta, Tetrahedron Lett., 2004, 45(48), 8825–8829.
31 S. R. Meech and D. Philips, J. Photochem., 1983, 23(2), 193–217.
32 V. Papper, V. Kharlanov and W. Retting, Phys. Chem. Chem.
Phys., 2002, 4(10), 1752–1759.
References
1 T. W. Bell and N. M. Hext, Chem. Soc. Rev., 2004, 33(9), 589–598.
2 P. B. Palde, P. C. Gareiss and B. L. Miller, J. Am. Chem. Soc.,
2008, 130(29), 9566–9573.
3 X. Zhang, Y. Xiao and X. Qian, Angew. Chem., Int. Ed., 2008,
47(42), 8025–8029.
4 W. Yang, L. Lin and B. Wang, Tetrahedron Lett., 2005, 46(46),
7981–7984.
5 T. Kalai and K. Hideg, Tetrahedron, 2006, 62(44), 10352–10360.
´
33 E. Lippert, Ber. Bunsen-Ges. Phys. Chem., 1957, 61(8), 962–975.
34 N. Mataga, Y. Kaifu and M. Koizumi, Bull. Chem. Soc. Jpn., 1956,
29(4), 465–470.
6 S. K. Kim, M. Y. Bang, S.-H. Lee, K. Nakamura, S.-W. Cho and
J. Yoon, J. Inclusion Phenom. Macrocyclic Chem., 2002, 43(1–2),
71–75.
35 N. Mataga, Introduction of Photochemistry, Kyoritsu Shuppan,
Inc., Tokyo, 1975.
36 G. Kortum, W. Vogel and K. Andrussow, IUPAC Dissociation
¨
7 S. K. Kim, N. J. Singh, J. Kwon, I.-C. Hwang, S. J. Park,
K. S. Kim and J. Yoon, Tetrahedron, 2006, 62(25), 6065–6072.
8 A. J. Bryan, A. P. de Silva, S. A. de Silva, R. A. D. D. Rupasinghe
and K. R. A. S. Sandanayake, Biosensors, 1989, 4(3), 169–179.
9 J. H. Kim, A.-R. Hwang and S.-K. Chang, Tetrahedron Lett.,
2004, 45(41), 7557–7561.
Constants of Organic Acids in Aqueous Solution, Butterworths,
London, 1961.
37 K. Rotkiewics, K. H. Grellmann and Z. R. Grabowski, Chem.
Phys. Lett., 1973, 19(3), 315–318.
38 Z. R. Grabowski, K. Rotkiewicz, A. Siemiarczuk, D. J. Cowley
and W. Baumann, Nouv. J. Chim., 1979, 3(7), 443–454.
39 W. M. Nau, G. Ghale, A. Hennig, H. Bakirci and D. M. Bailey,
J. Am. Chem. Soc., 2009, 131(32), 11558–11570.
10 S. Nishizawa, Y. Kato and N. Teramae, J. Am. Chem. Soc., 1999,
121(40), 9463–9464.
ꢂc
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 6641–6649 | 6649