Micelle-induced versatile performance of amphiphilic intramolecular charge-transfer fluorescent molecular sensors

Article abstract of DOI:10.1002/chem.200700435

A series of amphiphilic intramolecular charge-transfer fluorescent molecular sensors AS1-3, equipped with a rod-shaped hydrophobic 2-phenylbenzoxazole fluorophore and a hydrophilic tetraamide Hg²⁺-ion receptor, have been prepared. These sensor

Full text of DOI:10.1002/chem.200700435

FULL PAPER
DOI: 10.1002/chem.200700435
Micelle-Induced Versatile Performance of Amphiphilic Intramolecular
Charge-Transfer Fluorescent Molecular Sensors
Jiaobing Wang,[a, b] Xuhong Qian,*[a, b] Junhong Qian,[a] and Yufang Xu[a]
Abstract: A series of amphiphilic intra-
molecular charge-transfer fluorescent
molecular sensors AS1–3, equipped
with a rod-shaped hydrophobic 2-phe-
nylbenzoxazole fluorophore and a hy-
drophilic tetraamide Hg2+-ion recep-
tor, have been prepared. These sensor
molecules could be incorporated into
the hydrophobic sodium dodecyl sul-
fate (SDS) micelle, which is confirmed
by the clear spectral blue shift and
emission enhancement observed at the
critical micelle concentration of SDS.
Systematic examination of the sensor–
Hg2+ complexation, by using both UV/
visible and fluorescence spectroscopy,
indicates that SDS significantly modu-
lates both the binding event and signal
transformation of these sensor mole-
cules. The potential advantages are
fourfold: 1) SDS substantially increases
the Hg2+-ion association constant and
results in an amplified sensitivity.
2) SDS initiates spectral features which
facilitate Hg2+-ion analysis, for exam-
ple, in addition to the strengthened
fluorescence of the free sensors AS1–3,
the original “on–off” response of AS2
toward the Hg2+ ion is transformed
into a self-calibrated two-wavelength
ratiometric signal, while for AS3, Hg2+
-ion complexation in the presence of
SDS results in a 180 nm blue shift,
which is preferred to the 51 nm spec-
tral shift obtained without SDS.
3) Thermoreversible tuning of the dy-
namic detection range is realized.
4) Highly specific Hg2+-ion identifica-
tion could be achieved by using the
SDS-induced
fingerprint
emission
(358 nm) of the AS2–Hg2+ complex.
Altogether, this work demonstrates a
convenient and powerful strategy that
remarkably elevates the performance
of
a
given fluorescent molecular
sensor. It also implies that for a specific
utilization, much attention should be
paid to the microenvironment in which
the sensor resides, as the behavior of
the sensor might be different from that
in the bulk solution.
Keywords: charge
transfer
·
fluorescence · mercury · micelles ·
sensors
Introduction
ing. It is generally accepted that a small fluorescent sensor
molecule (FSM), with its many inherent merits including
high sensitivity, high specificity, and real-time in situ re-
sponse, can provide a lot of crucial information for practical
applications. Moreover, deep insight into the recognition
and signal transduction processes at the molecular level will
provide indispensable knowledge for the understanding of
life phenomena.[2–4] Nonetheless, despite the fact that numer-
ous pieces of research aimed at the development of novel
FSMs have been performed and some basic principles gov-
erning the performance of a special FSM have been estab-
lished, it is still challenging to create FSMs with desired
properties. Under these circumstances, it will be invaluable
to find ways that can remarkably elevate the performance of
a given FSM.
Fluorescent molecular sensing has enjoyed intensive re-
search over the years,[1] which is mainly driven by the fact
that many fields, such as scientific research, industrial pro-
duction, and our daily lives, require efficient analyte report-
[a] Dr. J. Wang, Prof. Dr. X. Qian, Dr. J. Qian, Prof. Dr. Y. Xu
State Key Laboratory of Bioreactor Engineering and
Shanghai Key Laboratory of Chemical Biology
School of Pharmacy
East China University of Science and Technology
Shanghai 200237 (China)
E-mail: xhqian@ecust.edu.cn
[b] Dr. J. Wang, Prof. Dr. X. Qian
State Key Laboratory of Fine Chemicals
Dalian University of Technology
Dalian 116012 (China)
Surfactant, a versatile amphiphilic molecular star, finds
rather wide utilization in different fields. For instance, in the
textile industry, surfactant is used to prevent dye aggrega-
tion and to facilitate pigmentation, while in the mining pro-
cess surfactant can be exploited for the enrichment of the
Fax : (+86)21-6425-2603
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2007, 13, 7543 – 7552
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7543

sparsely distributed minerals. Although for different purpos-
es, most of these utilizations benefit from the advantage that
surfactant molecules can offer a special phase or microenvir-
onment, such as a micelle, distinct from the bulk solution.
These new microenvironments have been cleverly used for
self-assembled nanoreactors[5] or for the template synthesis
of novel mesoporous materials.[6] We reasoned that if a
tailor-made small FSM could share the virtues that surfac-
tants provide, then large-amplitude performance elevation
may be realized.[7] As a proof-of-principle demonstration of
this speculation, we report herein a series of amphiphilic in-
tramolecular charge-transfer (ICT) FSMs. We found that
1) enhanced sensitivity, 2) desirable spectral changes includ-
ing “on–off” to ratiometric
mance enhancement by virtue of the microenvironment of a
micelle, both inside and outside (Scheme 1).
Amphiphilic ICT sensors AS1–3 (for synthesis see
Scheme 2) are constructed by incorporating a hydrophilic
Scheme 1. Cartoon representation of the SDS micelle-assisted FSM per-
formance enhancement. Step A: micelle incorporation modulates the
ICT process. Step B: micelle functions as a “cation sponge”.
signal transformation, 3) a ther-
mocontrolled dynamic detec-
tion range, and 4) highly specif-
ic analyte identification by the
fingerprint emission of an ana-
lyte-sensor complex could be
achieved when sodium dodecyl
sulfate (SDS) was added up to
its critical micelle concentration
(CMC).
We chose the ICT FSMs[1a,b]
mainly based on the following
two considerations: 1) environ-
ment sensitivity and 2) compact
receptor–fluorophore integra-
tion. In detail, first, an ICT
FSM generally contains
a
“push–pull” p-electron system,
in which an electron-donating group (donor) is electronical-
ly conjugated with the electron-withdrawing group (ac-
ceptor). Upon photoexcitation, this kind of FSM will experi-
ence a strong ICT process from the donor to the acceptor,
which results in an enlarged excited-state dipole moment.
Such an excited state is very sensitive to the microenviron-
ment in which the sensor resides. For example, polar protic
solvent often results in a decreased fluorescence, which is
problematic for practical utilization. Thus, the SDS micelle,
with its decreased polarity inside,[8] should cause favorable
fluorescence enhancement for the incorporated ICT FSM.
Second, to effectively modulate the ICT process by analyte
complexation, some key atoms of the receptor are often
shared by the fluorophore. In this way, the receptor is direct-
ly infused into the fluorophore platform. Such a compact re-
ceptor–fluorophore integration may offer the receptor a
chance to make use of the microenvironment of the micelle
surface. Namely, micelle incorporation of the hydrophobic
fluorophore can largely drag the closely attached hydrophil-
ic receptor into the counterionic cloud, in which the local
concentration of cationic analyte is boosted by electrostatic
interaction.[9] As a result, the receptor catches the analyte
more efficiently. The above two points stress that an ICT
FSM has the potential to experience a remarkable perfor-
tetraamide Hg2+-ion receptor[10] into a hydrophobic 2-phe-
nylbenzoxazole fluorophore. Once the Hg2+ ion is caught by
the receptor, p-electron conjugation within the fluorophore
will be significantly perturbed and clear fluorescence signals
(fluorescence quenching is also possible because of the
heavy-atom effect) will be produced. The rod-shaped non-
ionic hydrophobic fluorophore would be compatible with
the hydrocarbon tail of SDS and facilitates its incorporation
into the SDS micelle. Moreover, different substituents on
the para position of the 2-phenyl ring are introduced to
modify the photophysical property and Hg2+ ion binding
strength of AS1–3. The detailed synthesis of these sensors is
provided in the Experimental Section.
Results and Discussion
UV/visible absorption and fluorescence spectra of free sen-
sors AS1–3: As shown in Figure 1a, in neutral water solution
AS1 shows a strong absorption band centered at 363 nm
(e=43400mꢀ1 cmꢀ1), whereas AS2 and AS3 both display
two major bands, similar to each other in terms of band
shape and molar absorption coefficient. The absorption of
AS3
(366 nm,
e=18300mꢀ1 cmꢀ1
;
256 nm,
e=
25400mꢀ1 cmꢀ1) is found at a longer wavelength than that of
7544
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7543 – 7552

Fluorescent Molecular Sensors
FULL PAPER
AS2
(340 nm,
e=
18000mꢀ1 cmꢀ1
;
248 nm, e=
22800mꢀ1 cmꢀ1). This is caused
by the electron-deficient cya-
nide group, which enhances p-
electron push–pull interactions.
Remarkably, the absorption of
AS1 is different from that of
AS2 and AS3. This observation
could be ascribed to the differ-
ent electronic configurations of
sensors AS1–3. In detail, AS1 is
no longer a simple sensor with
D–A (D: electron donor, 5,6-di-
amino nitrogen atoms; A: elec-
tron acceptor, oxazole ring)
constitution but
a
D–A–D1
one.[11,12] If the D–A constitu-
tion is authentic, the electron-
donating dimethylamino group
should just play the “minor
role” of tuning the electron
density of the electron-acceptor
Scheme 2. Synthesis of sensors AS1–3: a) NaNO2, CH3COOH, 258C, 15 min, 98%; b) Pd/C (5 wt%), cyclohex-
ene, THF, reflux, 2 h, 96%; c) Zn, CH3COOH, 258C, 5 min; d) aromatic aldehyde, CH2Cl2, 358C, 10 min;
e) BaMnO4, benzene, reflux, 20 min, ꢁ50%; f) 2-aminoethanol, CH3CN, reflux, 2 h, ꢁ84%.
moiety, as the cyanide group does on AS3, and a similar but
blue-shifted absorption spectrum for AS1 should have been
observed. However, in the D–A–D1 system, the dimethyla-
mino group acts as another electron donor D1 and results in
the absorption of the D1!A[13] electron transition overlap-
ping that of the D!A transition, thus giving the strong ab-
sorption band at 363 nm.
Sensors AS1–3 exhibit different emission colors (Fig-
ure 2b) ranging from blue (AS1, lem =436 nm), via light
green (AS2, lem =497 nm), to deep orange (AS3, lem
=
576 nm). Their fluorescence quantum yields (F) are deter-
mined to be 0.50, 0.48, and 0.067, respectively.[14] Clearly,
AS1 is the brightest sensor molecule due to its large molar
absorption coefficient and high fluorescence quantum yield.
The comparatively weak emission of AS3 is not surprising
considering its inherent, strong push–pull p-electron config-
uration because the intensive dipole moment, generated by
photoexcitation, will result in a higher electron density on
the oxazole nitrogen atom, which turns on new fluorescence
deactivation pathways, such as enhanced hydrogen bonding
with water molecules.[1a,b]
SDS effects on free sensors AS1–3: We next examined the
SDS-induced spectral changes of sensors AS1–3 (Figure 2)
in neutral aqueous solution. Generally speaking, 1) a clear
spectral blue shift and emission enhancement were observed
for all sensors and 2) steep transformations were found at
the CMC (8.2 mm)[8] of SDS (Figure 2, insets). When 12 mm
SDS[15] was present, fluorescence enhancement rates of 20
(F=0.60), 43 (F=0.69), and 176% (F=0.185), accompa-
nying emission blue shifts of 16, 20, and 18 nm, were ach-
ieved for sensors AS1–3, respectively. These results indicat-
ed that the hydrophobic fluorophore was successfully incor-
porated into the SDS micelle. In contrast to the emission
Figure 1. a) Absorption and b) normalized fluorescence spectra of sensors
AS1–3 (10 mm) in neutral water solution. Excitation of AS1–3 was at 359,
320, and 339 nm, respectively. FI: fluorescence intensity, c: AS1,
a: AS2, d: AS3.
Chem. Eur. J. 2007, 13, 7543 – 7552
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7545

X. Qian et al.
UV/visible responses on Hg2+-ion titration: The UV/visible
absorptions of AS1–3 all displayed clear spectral responses
upon Hg2+-ion[16] titration, indicating that complexation oc-
curred both in the absence and presence of SDS. Distinctly,
from these responses, three major features could be ab-
stracted (Table 1). First, the UV/visible responses of AS1
Table 1. Summarized maximum absorption wavelength [nm] and molar
absorption coefficient [mꢀ1 cmꢀ1] of AS1–3 (10 mm) in the absence or
presence of the Hg2+ ion (20 mm) and SDS (12 mm).
Sensor
l
l
l
l SDS/
e
e
e
e SDS/
Free Hg2+ SDS Hg2+
Free
Hg2+ SDS
Hg2+
AS1
AS2
AS3
363 359
340 304
366 315
364 367
340 304
366 315
43400 42500 46000 49000
18000 24600 16300 28600
18300 27300 17800 32500
are weaker than those of AS2 and AS3. This could be
traced back to the different molecular constitutions of sen-
sors AS1–3. In detail, for sensors AS2 and AS3 with the D–
A structure, Hg2+-ion complexation will significantly de-
crease the electron density on electron donor D and result
in a depressed electron conjugation. Consequently, a distinct
blue shift is observed.[17] However, for AS1, Hg2+-ion com-
plexation is expected to exert two reverse effects on its D–
A–D1 p system: 1) For the D–A part, it will result in a re-
duced electron conjugation and promote a spectral shift to a
shorter wavelength. 2) For the D1–A part, reduction of the
electron density on the whole D–A moiety is expected to
enhance the push–pull interactions between dimethylamine
D1 and the oxazole A, which facilitates a spectral shift to a
longer wavelength. Thus, overall, these two effects might
counterbalance each other and produce a spectral change
less intensive than those of AS2 and AS3 (see reference [11]
and the Supporting Information). Second, in contrast to the
fluorescence spectra (see below), SDS incorporation did not
initiate distinct changes in the shape of the absorption spec-
tra during Hg2+-ion addition. Third, the titration curve was
steeper in the presence of SDS, thus giving a larger associa-
tion constant Ks (Table 2).
Table 2. Summarized Hg2+-ion association constants [mꢀ1] in the absence
or presence of 12 mm SDS.
Sensor 1Ks/105[a]
2Ks/105[a] 3Ks/105[b]
4Ks/105[b] 2Ks/1Ks 4Ks/3Ks
(no SDS)
(SDS)
(no SDS)
(SDS)
AS1
AS2
AS3
4.8
4.6
3.4
18.6
17.4
16.4
3.5
2.4
1.2
22.0
16.3
10.0
3.8
3.8
4.8
6.3
6.8
8.9
Figure 2. Fluorescence spectra of sensors (10 mm) a) AS1, b) AS2, and
c) AS3 (peak at 540 nm is caused by Rayleigh scattering) in neutral
water solution (pH 7.1, 258C) containing different concentrations of SDS.
Excitation of AS1–3 was at 354, 320, and 270 nm, respectively, which was
the isosbestic point on the absorption spectra. Inset: fluorescence intensi-
ty as a function of SDS concentration.
[a] Determined from the absorption spectra. [b] Determined from the
fluorescence spectra.
spectra, SDS only induced small spectral changes in the UV/
visible absorption of AS1–3 (see the Supporting Informa-
tion), indicating that the weaker dipole moment of the
ground state is less sensitive to the polarity of the environ-
ment.
Fluorescence responses on Hg2+-ion titration: Hg2+-ion ti-
tration of sensor AS1 (10 mm) gradually quenched its fluo-
rescence (Figure 3a).
A chelation-enhanced quenching
CHEQe (CHEQ=(I0ꢀI)/I0, I0 and I are equal to fluores-
cence intensity in the absence or presence of the Hg2+ ion
7546
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7543 – 7552

Fluorescent Molecular Sensors
FULL PAPER
Figure 4. Hg2+-ion-induced (two equivalents) fluorescence intensity
changes of sensor AS1 (0.5 mm) in the absence or presence of SDS
(12 mm), in neutral water solution (pH 7.1, 258C).
Figure 3. Hg2+ ion titration-induced fluorescence changes of sensor AS1
(10 mm) in the a) absence (lex =359 nm) or b) presence (lex =346 nm) of
SDS (12 mm), in neutral water solution (pH 7.1, 258C). Insets: fluores-
cence intensity as a function of Hg2+-ion concentration.
and the superscript “e” is equal to the presence of one
equivalent of the Hg2+ ion) of 61% was obtained. However,
in the presence of 12 mm SDS, CHEQe was increased to
79%. Hg2+-ion association[18] constants in the absence or
presence of SDS were determined to be 3.5”105 and 22.0”
105 mꢀ1, respectively. Stronger fluorescence of the free sensor
and a higher CHEQ value, together with the larger associa-
tion constant, in the presence of SDS facilitate Hg2+-ion de-
tection at lower concentrations. This was clearly demonstrat-
ed by Hg2+-ion titration using more dilute AS1 (0.5 mm). As
shown in Figure 4, micellization significantly increased the
CHEQ value from 15 to 84% in the presence of 1 mm Hg2+
ions.
Figure 5. Hg2+ ion titration-induced fluorescence changes of sensor AS2
(10 mm) in the a) absence (lex =320 nm) or b) presence (lex =320 nm) of
SDS (12 mm), in neutral water solution (pH 7.1, 258C). Inset: fluores-
cence intensity I497 or ratio I477/I358 as a function of Hg2+-ion concentra-
tion.
For sensor AS2, Hg2+-ion titration in the absence of SDS
exhibited similar spectral changes to those of AS1 (Fig-
ure 5a). Fluorescence at 497 nm gradually vanished with in-
creasing Hg2+-ion concentrations. A CHEQe value of 61%
was observed. The maximum emission shifted slightly from
497 to 483 nm, producing a clear isoemissive point at
418 nm. However, the peakless emission from 340 to 418 nm
was too weak to be used for ratiometric purposes. In sharp
contrast, in the presence of 12 mm SDS, besides the larger
CHEQe value of 82% at 477 nm (Figure 5b), a new well-
structured emission[19] centered at 358 nm formed and devel-
Chem. Eur. J. 2007, 13, 7543 – 7552
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7547

X. Qian et al.
oped with the gradual increasing of Hg2+-ion concentration.
The fluorescence intensity ratio I477/I358 changed from 27.4 to
1.2 when one equivalent of Hg2+ ion was added. Clearly,
SDS translated the original on–off response of AS2 toward
the Hg2+ ion into a ratiometric signal.[20] These impressive
spectral changes were preferred for practical utilization be-
cause the ratiometric signal was self-calibrated, which could
largely or entirely cancel out the influences of light-source
fluctuation and effective sensor concentration. Moreover,
SDS significantly strengthened AS2–Hg2+ complexation, as
could be seen from the enlarged association constant from
2.4”105 to 16.3”105 mꢀ1.
For sensor AS3, SDS initiated another attractive spectrum
change distinct from those of AS1 and AS2. AS3 was origi-
nally a ratiometric FSM. As shown in Figure 6a, in the ab-
sence of SDS, Hg2+-ion complexation shifted the emission
peak from 576 to 525 nm but no distinguishable separate
emission peaks, corresponding to free AS3 and the AS3–
Hg2+ complex, were observed during Hg2+-ion titration.
The fluorescence intensity ratio I576/I525 changed from 1.5 to
0.9 in the presence of one equivalent of Hg2+ ion. The quan-
tum yield decreased slightly from 0.067 to 0.062. However,
in sharp contrast, Hg2+-ion titration in the presence of SDS
resulted in well-resolved separate emission peaks centered
at 558 and 378 nm, respectively, corresponding to free AS3
and AS3–Hg2+ complexes (Figure 6b). Clear emission dis-
crimination was desirable for ratiometric fluorescence mo-
lecular sensors because a strong output signal—large ampli-
tude changes in the intensity ratio of two selected wave-
lengths—was expected to bring about enhanced sensitivity
for analyte detection. As can be seen from AS3, the fluores-
cence intensity ratio I558/I378 was changed from 22.3 to 0.31
in the presence of one equivalent of Hg2+ ion, which is
more favorable than the counterpart “1.5 to 0.9” response
obtained without SDS. The Hg2+-ion association constants
before and after the addition of SDS were determined to be
1.2”105 and 10.3”105 mꢀ1, respectively.
Discussion of the fluorescence responses of sensors AS1–3
toward the Hg2+ ion: Hg2+-ion titration of sensors AS1–3
produces versatile fluorescence responses. Typically, 1) Hg2+
-ion association constants are increased substantially upon
SDS addition and 2) both on–off and ratiometric signals are
observed. We tentatively analyze these results based on the
inherent electronic properties and the microenvironment
factors of sensors AS1–3.
In detail, different substitutions of sensors AS1–3 result in
different electron densities on the electron-donating 5,6-dia-
mine nitrogen atoms, which in turn modulate the Hg2+ ion
affinity of the tetraamide receptor.[21] For example, AS1,
armed with the electron-rich dimethylamine group and con-
taining the highest electron density on the 5,6-diamine nitro-
gen atoms, catches the Hg2+ ion more tightly and gives a
higher Ks value (Table 2). As a result, the association con-
stants decrease in line with AS1>AS2>AS3.
AS–Hg2+ complexation is notably enhanced by SDS, as
demonstrated by the large ratio of Ks/1Ks or Ks/3Ks. This
sensitivity amplification could be attributed to the effective
molarity enhancement of the Hg2+ ion around the negative-
ly charged micelle surface, which promotes complexa-
tion.[7b,c,9] We noticed that association constants determined
from the absorption spectra show some deviations from
those determined from the fluorescence spectra (Table 2).
This observation should be caused by the experimental un-
certainties.[22] The absorption (steady-state fluorescence)
spectra reflect the vertical transitions of a given ground (ex-
cited)-state ensemble to the corresponding Franck–Condon
excited (ground) state. During these transitions, only elec-
trons move and the nuclei are frozen. Consequently, com-
plexes do not associate or dissociate during these processes,
which means that absorption and steady-state fluorescence
titrations should yield identical Ks values.[23,24]
2
4
Figure 6. Hg2+ ion titration-induced fluorescence changes of sensor AS3
(10 mm) in the a) absence (lex =339 nm) or b) presence (lex =339 nm) of
SDS (12 mm), in neutral water solution (pH 7.1, 258C). Inset: fluores-
cence intensity ratio I576/I525 or I558/I378 as a function of Hg2+-ion concen-
tration.
No matter whether SDS is present, AS1 reports the Hg2+
ion all the time with an on–off response (Figure 4). This ob-
servation could be ascribed to the heavy-atom effect, as
7548
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7543 – 7552

Fluorescent Molecular Sensors
FULL PAPER
tighter AS1–Hg2+ complexation (compared with AS2 and
AS3) accelerates spin–orbit coupling and intersystem cross-
ing, which quench the fluorescence. For AS3, Hg2+-ion com-
plexation is not expected to introduce such a strong heavy-
atom effect as AS1, because the binding strength of AS3 is
significantly weaker than that of AS1. Thus, AS3–Hg2+ com-
plexation typically results in a wavelength shift instead of
fluorescence quenching (Figure 6). However, sensor AS2,
with its Hg2+-ion binding strength intermediate between
those of AS1 and AS3, exhibits the spectral characteristics
of both the other two sensors, namely, a clearly quenched
fluorescence (Figure 5a, as observed on AS1) together with
a distinct spectral shift (Figure 5b, as observed on AS3).
Cation decoordination, which is commonly observed for
metal-ion complexes of D–A-substituted fluoroionophores
in highly polar media (see the many works of Bernard
Valeur, Monique Martin, Rene Lapouyade, or Claude Rul-
liere), can only be operative in AS2 and AS3, but not in
Figure 7. Temperature-controlled sensitivity of AS1 (0.5 mm) toward the
Hg2+ ion in the absence or presence of SDS (12 mm), in neutral water so-
lution (pH 7.1). Inset: modulating the sensitivity by alternated heating
AS1. This explains why AS2 and AS3 show dramatic Hg2+
-
~
!
&
*
and cooling processes. : 258C, SDS; : 688C, SDS; : 688C; : 258C.
induced shifts in absorption (see the Supporting Information
Figures S7a and S8a) but only weak shifts in emission (Figur-
es 5a and 6a). In contrast to highly polar neat aqueous solu-
tion, when the sensors are immersed in the much less polar
micelles the ICT process in the excited complexes is much
less promoted and clearly does not lead to cation decoordi-
nation: the complexes show well-separated blue-shifted
spectra in both absorption and emission (see Figures S7b
and S8b in the Supporting Information and Figures 5b
and 6b). In addition, the real degree of quenching by the
Hg2+ ion for AS2 and AS3 can only be assessed when the
intrinsic fluorescence quantum yields of the parent com-
pounds, 2-phenylbenzo[d]oxazole and 4-(benzo[d]oxazol-2-
yl)benzonitrile, in the micellar medium are known.
Scheme 3. Cartoon representation of the reversible TCS process.
ature is low, free SDS molecules assemble into well-shaped
micelles which entrap AS1 sensors and enhance their sensi-
tivity. However, elevation of temperature increases[25] the
CMC of SDS and disrupts the host micelle. As a result, the
entrapped sensor molecule is again liberated into the bulk
solution, in which higher concentrations of Hg2+ ions are
needed to induce distinct fluorescence changes. We notice
that elevation of temperature alone can also modulate the
sensitivity.[26] On the other hand, in the absence of SDS,
only a onefold enlargement of the detection window from
1.0–5.0 to 1.0–10.0 mm (the detection window is approxi-
mately estimated from the titration curve) is achieved, in-
stead of the more than one-ordered magnification from
0.05–0.9 to 1.0–10.0 mm. Moreover, this sensitivity modula-
tion is thermoreversible. As demonstrated by the inset of
Figure 7, in the presence of two equivalents of Hg2+ ions,
the SDS–AS1 sensing system provides two sets of fluores-
cence signals, corresponding to different sensitivities, when
the sample is subject to alternated heating and cooling pro-
cesses.
Thermocontrolled sensitivity: Normally, a given FSM only
has one dynamic detection window in which variation of the
analyte concentration can be traced by its corresponding
fluorescence signal. Analyte concentrations below or
beyond this dynamic range (DR) are the blind spots which
the sensor fails to report.[1a] Thus, it is desirable to widen the
DR of a sensing system so as to monitor analyte at different
concentrations, providing both qualitative and quantitative
data.
The SDS–AS sensing system displays a thermocontrolled
sensitivity (TCS), which could be adjusted to meet different
applications. We demonstrate the principle of TCS by using
AS1. As seen from Figure 7, taking advantage of the steep
titration curve, sensor AS1 (0.5 mm, 258C) in the presence of
SDS is good at indicating the Hg2+ ion from 0.05 to 0.9 mm.
However, it is incapable of indicating Hg2+ ions beyond
1 mm, because the titration curve has leveled off. In sharp
Hg2+-ion identification by SDS-induced fingerprint emis-
sion: In the absence of SDS, AS1–3 displayed high Hg2+-ion
selectivity consistent with our polyamide FSMs reported
previously.[10] As demonstrated by sensor AS2, among the
detected alkali-metal, alkaline-earth, transition-metal, and
heavy-metal ions, Cu2+ and Co2+ induced a moderate fluo-
contrast, when the sample is heated to 688C, higher Hg2+
-
ion concentrations, ranging from 1.0 to 10.0 mm, can find
their corresponding fluorescence signals on the titration
curve. This large-amplitude sensitivity modulation could be
attributed to a temperature-induced micelle assembly and
disassembly process (Scheme 3). In detail, when the temper-
rescence quenching (CHEQe, ꢁ20%) with
a binding
strength (for Cu2+
,
3Ks =3.2”4mꢀ1, ꢂ2%) significantly
Chem. Eur. J. 2007, 13, 7543 – 7552
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7549

X. Qian et al.
presence of other competing metal ions, such as Cu2+. As
seen from Figure 8, when Hg2+ ions were added to the
water solution containing AS2 and Cu2+ ions (AS2/Hg2+
/
Cu2+ 1:1:1), clear emission enhancement at 358 nm was ob-
served. An emission ratio I477/I358 of 0.054 was obtained, dis-
tinct from the counterpart 1.2 obtained in the “pure” AS2–
Hg2+ sensing system (Figure 5b). Similarly, in the SDS–AS3
sensing system, the short-wavelength emission at 378 nm
could also be used as the fingerprint emission of the AS3–
Hg2+ complex. Unfortunately, it was difficult to discriminate
the Hg2+ ion from the competing metal ions in the SDS–
AS1 sensing system (see the Supporting Information).
Conclusion
A promising and convenient fluorescent molecular sensing
strategy, demonstrated by the combination of anionic surfac-
tant SDS and amphiphilic ICT Hg2+-ion sensors AS1–3, is
presented. This strategy displays several advantages includ-
ing strengthened analyte–sensor complexation, an optimized
report signal, and a thermocontrolled dynamic detection
range. The success of this strategy is based on a dual “lock–
key” mechanism: the first lock–key refers to the common
analyte–receptor binding event, while the second lock–key
refers to programming the whole sensor molecule to adapt a
special microenvironment, such as the micelle, so as to fully
exert its potential. We are confident that this strategy could
be further extended to a wide range of sensing systems, as
different microplatforms, ranging from micelle and vesicle[5]
to molecular capsule[28] and dendrimer,[29] can provide mi-
croenvironments with varied polarity, rigidity, and special
functional groups, which could all possibly be used to modu-
late the performance of a specifically fabricated fluorescent
molecular sensor.
Figure 8. Cu2+ ion titration-induced fluorescence changes of sensor AS2
(10 mm) in the presence (lex =320 nm) of SDS (12 mm), in neutral water
solution (pH 7.1, 258C). The bold line indicates the Hg2+-ion identifica-
tion by the fingerprint emission around 358 nm in the presence of one
equivalent of Cu2+ ions.
weaker than that of Hg2+. The other detected metal ions did
not induce any detectable fluorescence response (see the
Supporting Information).
In the presence of SDS, highly specific Hg2+-ion identifi-
cation could be achieved despite the fact that the binding
strengths of AS1–3 toward Cu2+ and Co2+ were also signifi-
cantly enhanced. As seen from AS2 (Figure 8), Cu2+ ion ti-
tration in the presence of SDS resulted in a rather high
CHEQe value of 97%,[27] but without introducing any clear
emission change around 358 nm, as observed for the Hg2+
ion (Figure 5b). Similar spectral changes were also observed
for Co2+ with a CHEQe value of 89% (see the Supporting
Information). Thus, in the SDS–AS2 sensing system, the
short-wavelength emission around 358 nm could be assigned
as the fingerprint emission of the AS2–Hg2+ complex, which
could be used for highly specific Hg2+-ion identification
(Figure 9). Moreover, the Hg2+ ion could be indicated in the
Experimental Section
General methods: All the reagents and solvents were of the highest com-
mercial quality available and were used without purification. 1H NMR
spectra were recorded in CDCl3 or D2O at 258C on a Bruker AV-400
spectrometer. Mass spectra were recorded on a HP 1100 LC-MS spec-
trometer. Melting points were determined by using an X-6 micro-melting
point apparatus and were uncorrected. The pH values were measured
with a Sartorius basic pH-Meter PB-20. Absorption spectra were deter-
mined on a PGENERAL TU-1901 UV/Vis Spectrophotometer. Fluores-
cence spectroscopic studies were performed with a Hitachi F-4500
Compound 2: Compound 1 (2.23 g, 4 mmol), prepared by using the previ-
ously reported method,[10b] was dissolved in dichloromethane (10 mL),
then acetic acid (100 mL) was added. NaNO2 (0.69 g, 10 mmol) was
added slowly to this solution over 2 min. The mixture was stirred vigo-
rously at room temperature for 15 min, then poured into water (400 mL)
and extracted twice with dichloromethane (200 mL). The combined or-
ganic phase was dried over sodium sulfate and evaporated to dryness.
The residue was purified by flash chromatography using dichloromethane
as the eluent, affording 2 (2.3 g, 3.92 mmol, 98%) as a yellow oil.
1H NMR (400 MHz, CDCl3, 258C): d=7.81 (s, 1H), 7.49 (d, J=7.4 Hz,
2H), 7.39 (t, J=7.4 Hz, 2H), 7.32 (t, J=7.4 Hz, 1H), 6.67 (s, 1H), 5.16 (s,
Figure 9. Metal-ion-induced (10 mm) fluorescence responses (at 358 nm)
of sensor AS2 (10 mm) in the presence of SDS (12 mm), in neutral water
solution (pH 7.1, 258C).
7550
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7543 – 7552

Fluorescent Molecular Sensors
FULL PAPER
2H), 4.40 (s, 4H), 4.15 (s, 4H), 4.14–4.10 (m, 8H), 1.24–1.20 ppm (m,
12H).
129.0, 127.0, 125.7, 111.1, 103.6, 60.6, 55.9, 55.8, 41.3, 41.2 ppm; HRMS
(ES+): m/z: calcd: 630.2888 [M+H+]; found: 630.2880.
Sensor AS3: Sensor AS3 was obtained as a light-yellow solid (79%).
M.p. 148.3–149.58C; 1H NMR (400 MHz, D2O, 25 8C): d=7.82 (d, J=
8.4 Hz, 2H), 7.50 (d, J=8.4 Hz, 2H), 7.31 (s, 1H), 7.29 (s, 1H), 4.20 (s,
4H), 4.12 (s, 4H), 3.55–3.50 (m, 8H), 3.30–3.25 ppm (m, 8H); 13C NMR
(100 MHz, D2O, 25 8C): d=172.5, 160.8, 147.1, 142.4, 141.0, 136.2, 132.6,
129.4, 126.8, 118.6, 112.9, 111.7, 103.8, 59.9, 55.7, 41.3, 41.2 ppm; HRMS
(ES+): m/z: calcd [M+Na+]: 677.2659; found: 677.2657.
Compound 3: Compound 2 (1 g, 1.7 mmol) was dissolved in dry THF
(100 mL) containing Pd/C (1 g, 5%) and cyclohexene (20 mL), and the
mixture was stirred and refluxed under nitrogen for 2 h. Then the Pd/C
was removed by filtration and the filtrate was concentrated under
vacuum. The product was purified by flash chromatography using
hexane/ethyl acetate (7:3) as the eluent, affording 3 (0.81 g, 1.63 mmol,
96%) as a red oil. 1H NMR (400 MHz, CDCl3, 258C): d=10.72 (s, 1H),
7.83 (s, 1H), 6.55 (s, 1H), 4.47 (s, 4H), 4.20–4.11 (overlapped, 12H),
1.27–1.23 ppm (m, 12H).
Compound 4: Compound 3 (0.6 g, 1.21 mmol) was dissolved in dichloro-
methane (5 mL) and acetic acid (50 mL). Zn powder (4 g, 62.5 mmol)
was added to this solution and the mixture was stirred at room tempera-
ture for 5 min. The unreacted Zn was removed by filtration and the fil-
trate was poured into dichloromethane (200 mL). The organic solution
was extracted three times with water (200 mL) to remove the acetic acid
and dried over sodium sulfate for 5 min. Note: product 4 was not charac-
terized because of its easy oxidation and was used directly in the follow-
ing reaction.
Acknowledgements
This work was supported by the National Natural Science Foundation of
China and the National Key Project for Basic Research
(2003CB 114400). We thank the referees for their smart and pertinent
comments; some discussions originate from their comments. We also
thank Dr. A. Wu and Dr. K. Bathula for valuable discussions.
Compound 5: 4-Dimethylaminobenzaldehyde (0.3 g, 2 mmol) was added
[1] For a recent book and reviews on fluorescent molecular sensors see:
a) B. Valeur, Molecular Fluorescence: Principles and Applications,
Wiley-VCH, Weinheim, 2001; b) A. P. de Silva, H. Q. N. Gunaratne,
T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher,
T. E. Rice, Chem. Rev. 1997, 97, 1515; c) D. T. McQuade, A. E.
Pullen, T. M. Swager, Chem. Rev. 2000, 100, 2537; d) R. Martínez-
MµÇez, F. Sancenón, Chem. Rev. 2003, 103, 4419; e) L. Pu, Chem.
Rev. 2004, 104, 1687.
[2] a) K. Komatsu, K. Kikuchi, H. Kojima, Y. Urano, T. Nagano, J. Am.
Chem. Soc. 2005, 127, 10197; b) E. Sasaki, H. Kojima, H. Nishimat-
su, Y. Urano, K. Kikuchi, Y. Hirata, T. Nagano, J. Am. Chem. Soc.
2005, 127, 3284.
[3] C. J. Chang, J. Jaworski, E. M. Nolan, M. Sheng, S. J. Lippard, Proc.
Natl. Acad. Sci. USA 2004, 101, 1129.
[4] L. Yang, R. McRae, M. M. Henary, R. Patel, B. Lai, S. Vogt, C. J.
Fahrni, Proc. Natl. Acad. Sci. USA 2005, 102, 11179.
[5] D. M. Vriezema, M. C. Aragon›s, J. A. A. W. Elemans, J. J. L. M.
Cornelissen, A. E. Rowan, R. J. M. Nolte, Chem. Rev. 2005, 105,
1445.
[6] Y. Wan, H. Yang, D. Zhao, Acc. Chem. Res. 2006, 39, 423.
[7] Very recently, surfactants have been exploited in fluorescent molec-
ular sensing: a) K. Niikura, E. V. Anslyn, J. Org. Chem. 2003, 68,
10156; b) A. Mallick, M. C. Mandal, B. Haldar, A. Chakrabarty, P.
Das, N. Chattopadhyay, J. Am. Chem. Soc. 2006, 128, 3126; c) Y.
Zhao, Z. Zhong, Org. Lett. 2006, 8, 4715.
to the above dichloromethane solution (70 mL) of compound
4
(ꢁ0.4 mmol). The solvent was evaporated under vacuum to obtain a
brown oil and dry benzene (50 mL) was added. The solution was refluxed
under
a nitrogen atmosphere for 20 min. After this time, BaMnO4
(0.31 g, 1.2 mmol) was added after slight cooling and the solution was re-
fluxed again for another 20 min, cooled, and concentrated under vacuum.
The residue was purified by flash chromatography by using hexane/ethyl
acetate (7:3) as the eluent, affording 5 (0.12 g, 0.19 mmol, 48%) as a
light-yellow solid. M.p. 78.2–79.68C; 1H NMR (400 MHz, CDCl3, 258C):
d=8.03 (d, J=8.8 Hz, 1H), 7.39 (s, 1H), 7.28 (s, 1H), 6.76 (d, J=8.8 Hz,
2H), 4.37 (s, 4H), 4.31 (s, 4H), 4.14–4.09 (m, 8H), 3.06 (s, 6H), 1.19 ppm
(t, J=5.6 Hz, 12H); 13C NMR (100 MHz, CDCl3, 258C): d=170.9, 164.0,
152.1, 147.0, 139.7, 139.6, 138.2, 128.7, 114.7, 111.6, 111.4, 103.3, 60.6,
60.5, 53.2, 40.1, 14.2, 14.1 ppm.
Compound 6: Compound 6 was obtained as a light-yellow solid (52%).
M.p. 73.1–74.88C; 1H NMR (400 MHz, CDCl3, 258C): d=8.18–8.16 (m,
2H), 7.50–7.49 (m, 3H), 7.47 (s, 1H), 7.34 (s, 1H), 4.39 (s, 4H), 4.33 (s,
4H), 4.15–4.09 (m, 8H), 1.22 ppm (t, J=5.6 Hz, 12H); 13C NMR
(100 MHz, CDCl3, 258C): d=170.8, 170.7, 162.5, 147.3, 140.9, 140.1,
137.7, 131.0, 128.8, 127.4, 127.2, 112.2, 103.6, 60.6, 60.5, 53.2, 53.1, 14.2,
14.1 ppm.
Compound 7: Compound 7 was obtained as a yellow solid (57%). M.p.
89.6–90.58C; 1H NMR (400 MHz, CDCl3, 258C): d=8.26 (d, J=8.2 Hz,
2H), 7.78 (d, J=8.2 Hz, 2H), 7.50 (s, 1H), 7.34 (s, 1H), 4.41 (s, 4H), 4.34
(s, 4H), 4.15–4.10 (m, 8H), 1.21 ppm (t, J=7.0 Hz, 12H); 13C NMR
(100 MHz, CDCl3, 258C): d=171.3, 161.2, 148.2, 142.7, 141.2, 138.1,
133.3, 132.0, 128.1, 118.9, 114.7, 113.3, 104.2, 61.4, 61.3, 53.7, 14.8 ppm.
[8] For microscopic polarity of the SDS micelle see: K. Kano, Y. Ueno,
S. Hashimoto, J. Phys. Chem. 1985, 89, 3161.
[9] S. Uchiyama, G. D. McClean, K. Iwai, A. P. de Silva, J. Am. Chem.
Soc. 2005, 127, 8920.
Sensor AS1: Compound 5 (80 mg, 0.13 mmol) was dissolved in acetoni-
trile (10 mL) and 2-aminoethanol (10 mL) was added. The solution was
refluxed under nitrogen for 2 h, then cooled and concentrated under
vacuum to remove the acetonitrile and 2-aminoethanol separately. The
product was purified by flash chromatography using methanol/ammonia/
dichloromethane (20:2:100) as the eluent, affording AS1 (72 mg,
0.11 mmol, 85%) as a light-yellow solid. M.p. 135.5–136.98C; 1H NMR
(400 MHz, D2O, 25 8C): d=7.60 (d, J=8.8 Hz, 2H), 7.27 (s, 1H), 7.23 (s,
1H), 6.48 (brs, 2H), 4.09 (s, 4H), 4.05 (s, 4H), 3.57–3.52 (m, 8H), 3.28–
3.24 (m, 8H), 2.63 ppm (s, 6H); 13C NMR (100 MHz, D2O, 25 8C): d=
172.5, 164.6, 152.5, 146.8, 140.3, 140.2, 137.1, 128.6, 112.6, 111.7, 110.5,
103.5, 59.9, 56.3, 56.2, 41.3, 41.2, 39.2 ppm; HRMS (ES+): m/z: calcd:
673.3310 [M+H+]; found: 673.3315.
[10] We recently demonstrated that polyamide receptors could selective-
ly bind Hg2+ ions in neutral aqueous solution: a) J. Wang, X. Qian,
Chem. Commun. 2006, 109; b) J. Wang, X. Qian, J. Cui, J. Org.
Chem. 2006, 71, 4308; c) J. Wang, X. Qian, Org. Lett. 2006, 8, 3721.
[11] A closely related FSM with typical D–A–D1 constitution using the
benzothiazole fluorophore has been reported: K. Rurack, A. Kovalꢁ-
chuck, J. L. Bricks, J. L. Slominskii, J. Am. Chem. Soc. 2001, 123,
6205.
[12] Like the parent compound 2-phenylbenzoxazole, in AS1–3 the 2-
phenyl ring should be electronically conjugated and coplanar with
the benzoxazole moiety (energy-minimized molecular structure cal-
culated with the Hyperchem software). This notion is supported by
the 1H NMR spectroscopic studies: Hg2+-ion complexation results
in a significant downfield shift of the aromatic protons of the 2-
phenyl ring (see the Supporting Information): J. C. del Valle, M.
Kasha, J. Catalµn, J. Phys. Chem. A 1997, 101, 3260.
AS2 and AS3 were prepared similarly to AS1.
Sensor AS2: Sensor AS2 was obtained as a light-yellow solid (87%).
M.p. 123.7–124.88C; 1H NMR (400 MHz, D2O, 25 8C): d=7.94 (d, J=
8.4 Hz, 2H), 7.48–7.41 (m, 3H), 7.34 (s, 1H), 7.32 (s, 1H), 4.20 (s, 4H),
4.13 (s, 4H), 3.58–3.53 (m, 8H), 3.33–3.29 ppm (m, 8H); 13C NMR
(100 MHz, D2O, 25 8C): d=172.5, 163.5, 147.1, 141.1, 140.1, 136.5, 131.8,
[13] A D1!A transition is expected to absorb light around 350 nm,
judged from the absorption of the parent chromophore 2-(4’-N,N-di-
methylamino)phenylbenzoxazole: E. M. Vernigor, E. A. Lukyanets,
Chem. Eur. J. 2007, 13, 7543 – 7552
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7551

X. Qian et al.
L. P. Savvina, V. K. Shalaev, Chem. Heterocycl. Compd. 1993, 29,
208.
[14] Determined relative to quinine sulfate: S. R. Meech, D. C. Phillips,
J. Photochem. 1983, 23, 193.
[15] See reference [8]; provided that sensor incorporation does not
change significantly the aggregation number (AN=62), the micelle
concentration, was calculated to be about 6”10ꢀ5 m in the presence
of 12 mm SDS.
2-phenylbenzoxazole fluorophore (see references [12] and [17]).
Moreover, in a less polar acetonitrile solution, Hg2+-ion titration re-
sults in emission spectra similar to that in the presence of SDS (for
example, structured fluorescence band at 355 nm for AS2, see the
Supporting Information).
[20] Very recently, in a Triton X-100 micelle sensing system, the fluores-
cence signal could by modulated by changing the lipophilicity of the
receptor see: P. Pallavicini, Y. A. Diaz-Fernandez, F. Foti, C. Manga-
no, S. Patroni, Chem. Eur. J. 2007, 13, 178.
[16] For Hg2+-ion FSMs reported in the last two years, see: a) I.-B. Kim,
U. H. F. Bunz, J. Am. Chem. Soc. 2006, 128, 2818; b) Y. Zhao, Z.
Zhong, J. Am. Chem. Soc. 2006, 128, 9988; c) S.-K. Ko, Y.-K. Yang,
J. Tae, I. Shin, J. Am. Chem. Soc. 2006, 128, 14150; d) A. Coskun,
E. U. Akkaya, J. Am. Chem. Soc. 2006, 128, 14474; e) S. Tatay, P.
GaviÇa, E. Coronado, E. Palomares, Org. Lett. 2006, 8, 3857;
f) E. M. Nolan, M. E. Racine, S. J. Lippard, Inorg. Chem. 2006, 45,
2742; g) S. Yoon, A. E. Albers, A. P. Wong, C. J. Chang, J. Am.
Chem. Soc. 2005, 127, 16030; h) A. Caballero, R. Martínez, V. Llo-
veras, I. Ratera, J. Vidal-Gancedo, K. Wurst, A. Tµrraga, P. Molina,
J. Veciana, J. Am. Chem. Soc. 2005, 127, 15666; i) E. Coronado, J. R.
Galµn-Mascarüs, C. Martí-Gastaldo, E. Palomares, J. R. Durrant, R.
Vilar, M. Gratzel, Md. K. Nazeeruddin, J. Am. Chem. Soc. 2005,
127, 12351; j) J. V. Ros-Lis, M. D. Marcos, R. Martínez-MµÇez, K.
Rurack, J. Soto, Angew. Chem. 2005, 117, 4479; Angew. Chem. Int.
Ed. 2005, 44, 4405.
[21] a) S. Fery-Forgues, J. Bourson, L. Dallery, B. Valeur, New J. Chem.
1990, 14, 617; b) K. Rurack, M. Sczepan, M. Spieles, U. Resch-
Genger, W. Rettig, Chem. Phys. Lett. 2000, 320, 87; c) K. Rurack,
J. L. Bricks, G. Reck, R. Radeglia, U. Resch-Genger, J. Phys. Chem.
A 2000, 104, 3087.
1
3
2
4
[22] The uncertainty of Ks and Ks is about 20% and that of Ks and Ks
is about 40%.
[23] J. Bourson, J. Pouget, B. Valeur, J. Phys. Chem. 1993, 97, 4552.
[24] It is generally agreed that association constants larger than 4”
105 mꢀ1 cannot be reliably determined by conventional optical spec-
troscopic methods. The above discussion was initiated by the refer-
ee.
[25] B. Jçnnson, B. Lindman, K. Holmberg, B. Kronberg, Surfactants and
Polymers in Aqueous Solutions, Wiley, Chichester, 1998.
[26] The Hg2+-ion association constant was determined to be 4.8”
104 Mꢀ1 ꢂ10% at 688C.
[17] The sharp and structured absorption spectra of Hg2+-ion complexes
of AS2 and AS3 resembled closely, in terms of spectral shape and
absorbance, that of the unsubstituted 2-phenylbenzoxazole chromo-
phore (see the Supporting Information): D. G. Ott, F. N. Hayes, E.
Hansbury, V. N. Kerr, J. Am. Chem. Soc. 1957, 79, 5448.
[27] Cu2+ ion titration also initiated a distinct band shift (from 340 to
304 nm), similar to that of the Hg2+ ion, on the UV/visible absorp-
tion of AS2; see the Supporting Information.
[28] M. M. Conn, J. Rebek, Jr., Chem. Rev. 1997, 97, 1647.
[29] F. Zeng, S. C. Zimmerman, Chem. Rev. 1997, 97, 1681.
[18] A Jobꢁs plot indicated that AS1–3 formed a 1:1 complex with the
Hg2+ ion, consistent with our previously reported tetraamide ICT
Hg2+-ion-sensor molecule (see reference [10b]).
[19] Hg2+-ion complexation depresses the ICT emission and the struc-
tured fluorescence bands resemble closely that of the unsubstituted
Received: March 20, 2007
Published online: June 21, 2007
7552
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7543 – 7552