Phenylethylidene-3,4-dihydro-1H-quinoxalin-2-ones: Promising building blocks for Cu2+ recognition

Article abstract of DOI:10.1016/j.tet.2011.06.047

A series of sulfonamido-substituted phenylethylidene-3,4-dihydro-1H- quinoxalin-2-one derivatives in which both of the fluorophore and ionophore are integrated into one structural unit, have been investigated. They all exhibit high selectivity toward Cu²⁺ in ethanol in the presence of other metallic ions (Zn²⁺, Mg²⁺, Co²⁺, Ni ²⁺, Mn²⁺, Ca²⁺, and Ag⁺), as well as fast, stable, and reversible binding, as is evidenced by the observation of a red shift in the UV-vis spectrum, 'ON-OFF' fluorescence response. In addition, titration and MALDI-TOF measurements indicated that a 1:1 (and possibly also 2:1 (organic ligand: Cu²⁺) complexes were formed, depending on the relative amount of Cu²⁺ added to the solution of the organic ligand. It was also found that the binding constant could be tuned by modifying the nature and position of the substituents attached to the central benzene ring in the quinoxalone derivative. In acetonitrile, unlike in ethanol, these ligands undergo oxidation-decomposition by Cu²⁺ and therefore, no UV-vVis absorption bands could be observed. However, due to color change (from yellow to transparent) they could be useful as dosimeters in this solvent.

Full text of DOI:10.1016/j.tet.2011.06.047

Tetrahedron 67 (2011) 6252e6258
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Phenylethylidene-3,4-dihydro-1H-quinoxalin-2-ones: promising building blocks
for Cu2þ recognition
,
b
,
b
a,
*
Efrat Korin a, Beny Cohen a , Cheng-Chu Zeng , Yi-Sheng Xu , James Y. Becker
*
*
a Department of Chemistry, Ben-Gurion University of the Negev, Beer Sheva 84105, Israel
b College of life Science and Bioengineering, Beijing University of Technology, Beijing 100124, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of sulfonamido-substituted phenylethylidene-3,4-dihydro-1H-quinoxalin-2-one derivatives in
which both of the fluorophore and ionophore are integrated into one structural unit, have been in-
vestigated. They all exhibit high selectivity toward Cu2þ in ethanol in the presence of other metallic ions
(Zn2þ, Mg2þ, Co2þ, Ni2þ, Mn2þ, Ca2þ, and Agþ), as well as fast, stable, and reversible binding, as is evi-
denced by the observation of a red shift in the UVevis spectrum, ‘ONeOFF’ fluorescence response. In
addition, titration and MALDI-TOF measurements indicated that a 1:1 (and possibly also 2:1 (organic
ligand: Cu2þ) complexes were formed, depending on the relative amount of Cu2þ added to the solution of
the organic ligand. It was also found that the binding constant could be tuned by modifying the nature
and position of the substituents attached to the central benzene ring in the quinoxalone derivative. In
acetonitrile, unlike in ethanol, these ligands undergo oxidation-decomposition by Cu2þ and therefore, no
UVevVis absorption bands could be observed. However, due to color change (from yellow to transparent)
they could be useful as dosimeters in this solvent.
Received 30 January 2011
Received in revised form 25 May 2011
Accepted 16 June 2011
Available online 22 June 2011
Keywords:
Quinoxalone derivatives
Cu2þ recognition
Fluorescence
Chemosensors
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
on several colorimetric and fluorescent chemosensors with azine or
2-aza-1,3-diene moieties as putative cation-binding sites, where
The development of synthetic optical sensors for essential heavy
metallic ions continues to attract a considerable research interest
due to their important role in living systems and their impact on
environmental toxicity.1,2 Based on the different pathways for ion
sequestration, these sensors can be divided into two categories:
‘fluorogenic’ chelating agents (e.g., 8-hydroxyquinoline3) and ‘flu-
oroionophores’ with separate fluorophore and ionophore units, in
which the fluorophore responses for signal transduction and the
ionophore for selective recognition of metal ions.
anthracene or pyrene serves as fluorophores for signal trans-
duction. The latter property is based on photo-induced intra-
molecular electron-transfer (PET) mechanism or fluorescence
quenching caused by the paramagnetic nature of Cu2þ
.
In parallel studies, diketo acid derivatives have been demon-
strated12e14 to be among the most promising HIV integrase in-
hibitors, which are able to selectively inhibit a strand transfer
reaction of integrase in vitro and in infected cells. The molecular
basis of inhibition of this type of compounds is the sequestration
of the critical metal cofactor(s) (Mg2þ is now generally accepted)
in the integrase active site. In this context, the binding of a di-
valent metal ion (Mg2þ) is a prerequisite for the HIV integrase
inhibition.
Base on the above mentioned information, recently we have car-
riedoutaprojecttodesignandsynthesizesulfonamido-, caffeoyl-and
galloyl-containing diketo acids, and quinoxaline-2-ones, as potential
HIV integrase inhibitors15,16 However, the latter derivatives did not
bind Mg2þ but showed selective response toward Cu2þ. In the present
work, we report new results of phenylethylidene-3,4-dihydro-1H-
quinoxalin-2-ones (involving integrated fluorophore and ionophore
units)aspromising buildingblocks forselective recognition of Cu2þ. It
is noteworthy that this system is unique because other known fluo-
rogenic chelating agents, such as 8-hydroxyquinoline3 have shown
low selectivity toward various divalent metal cations.
Among the heavy metallic ions, Cu2þ is important because it is
the third most abundant (after Fe3þ and Zn2þ) among essential
heavy metals in the human body and it is important in various
physiological processes. In addition, Cu2þ is a significant environ-
mental pollutant. For these reasons, the past few years have wit-
nessed a number of reports on the design and synthesis of optical or
fluorescent sensors for the detection of Cu2þ ions.4e9 Gunnlaugsson
et al.10 designed an azobenzene-based colorimetric chemosensor,
which could detect Cu2þ under physiological conditions and give
rise to a major color change from red to yellow. Molina11 reported
* Corresponding authors. Tel.: þ972 8 6461197; fax: þ972 8 6472943 (J.Y.B.); tel.:
þ972 8 646 1639; fax: þ972 8 6472943 (B.C.); tel.: þ86 10 67396211; fax: þ86 10
67392001 (C.-C.Z.); e-mail addresses: bcohen@bgu.ac.il (B. Cohen), zengcc@bju-
t.edu.cn (C.-C. Zeng), becker@bgu.ac.il (J.Y. Becker).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.06.047

E. Korin et al. / Tetrahedron 67 (2011) 6252e6258
6253
2. Results and discussion
been used in order to prove the specific binding of Cu2þ and the lack
of binding of other cations). Clearly no significant change in the
The structures of quinoxaline-2-one derivatives 3 and 4 are
shown in Scheme 1. These compounds were synthesized in >90%
yield according to our previous report.15 A mixture of 1 or 2 and o-
phenylenediamine in water was heated to reflux for 2e4 h and the
desired 3 or 4 was isolated after cooling, filtering, and washing with
ether. The products are yellowish solids and exhibit strong fluo-
rescence under UVevis excitation at 365 nm.
spectra takes place when more than 100 equiv of Zn2þ, Mg2þ, Co2þ
,
Ni2þ, Mn2þ, Ca2þ, and Agþ were added. However, when Cu(ClO4)2
was added, each of the three absorption bands at 392, 419, and
442 nm undergoes a red shift by 26e28 nm, with concomitant
appearance of three new absorption bands at 421, 445, and 470 nm
(it is noteworthy that under these conditions, Cu(ClO4)2 has no
absorption above 320 nm). Moreover, upon addition of EDTA as
a strong complexing agent, the original absorption bands (at 392,
419, and 442 nm) of 4b re-appeared, indicating the existence of
a reversible process.
The absorptions and ion binding properties of other derivatives of
types 3 and 4 exhibit similar patterns to those of 4b. Their absorption
maxima and wavelength changes in the presence of Cu2þ are sum-
marized in Table 1. The red shift observed after the addition of
Cu(ClO4)2 to a solution of the organic ligand can be attributed to
deprotonation of one of the secondary amine moieties in the qui-
noxalone ring. Accordingly, the UVevis spectrum of free 3b
(2.4ꢀ10ꢁ5 M) in the presence of NaOH (5ꢀ10ꢁ3 M) also exhibits a red
shift (although smaller, w14 nm) compared to the spectrum without
the base, emphasizing the effectof deprotonation (see Supplementary
data). To confirm that the red shift observed upon addition of NaOH
was not due to Naþ ions, NaClO4 (1.4ꢀ10ꢁ2 M) was added to 3b
(2.4ꢀ10ꢁ5 M) and no change in the spectrum was observed.
Scheme 1. Synthesis of compounds 3 and 4.
2.1. UVevis spectra of 3 and 4 in ethanol
The absorption spectra of 3 and 4 in ethanol solutions were
investigated and a typical spectrum of compound 4b is shown in
Fig. 1. In the absence of metallic ions, all these compounds display
a shoulder band at 392 nm and two intense absorption bands
centered at about 418 and 441 nm in the visible region with ex-
Table 1
Change of absorption peaks (lmaxꢂ0.5 nm) of compounds 3 and 4 in the presence of
tinction coefficients of 3.1ꢀ104 and 2.8x104 Lmolꢁ1 cmꢁ1
, re-
Cu2þ in EtOH
spectively. The intense and low-energy absorptions of compounds
‘Host’ l1max
l2max
l3max
‘Host’ l1max
l2max
l3max
3 and 4 in the visible region suggest a
pep transition (K band)
*
3a
3b
3c
3d
3e
395a (416)b 416 (442) 442 (470) 4a
393 (419) 418 (441) 440 (465) 4b
394 (418) 418 (440) 439 (466) 4c
395 (416) 416 (443) 442 (467) 4d
394 (420) 418 (444) 444 (472)
392 (421) 419 (445) 442 (470)
396 (422) 420 (444) 441 (471)
395 (422) 419 (445) 441 (471)
395 (421) 418 (445) 444 (472)
taking place in these conjugated systems, which has also been
supported by their X-ray structures. For example, in compounds
3a,17 4b,15 and 4d,18 the two nitrogen atoms in the quinoxalone ring
are of sp2 hybridization, resulting in a planar arrangement of the
quinoxalone ring. Moreover, an intramolecular NeH/O hydrogen-
bond is present and allows the quinoxalone ring and the 2-oxo-
ethylidene unit to be in the same plane.
395 (d)
417 (444) 445 (470) 4e
a
Absorption wavelength (lmax) of free compounds (throughout the Table).
Absorption wavelength (lmax) of compounds after addition of Cu2þ(in paren-
b
theses, throughout the Table).
Additional proof about the origin of the red shift stems from the
effect of acid addition to the complex solution, which was in-
vestigated by both UVevis and EPR measurements. In each tech-
nique the addition of HClO4 to a solution of 3b containing Cu2þ
resulted in complex dissociation. Whereas in UVevis measure-
ments the spectrum was shifted back to the absorption bands of
free 3b, in the EPR spectrum, the signal of the complex was
replaced by the signal of free Cu2þ (see Supplementary data).
It is well known that fluorescence emission spectroscopy is
more sensitive than UVevis spectroscopy toward small changes,
which may affect the electronic properties of molecular receptors.
Therefore, fluorescence spectra have also been measured in this
study to assess the selectivity of quinoxalone derivatives toward
metal ions. Fig. 2(a) shows the fluorescence response of 3b in
ethanol upon addition of different metal cations. It has been ob-
served that the emission spectra (excitation at 418 nm) of free
quinoxalone 3b in ethanol involve two bands, at 474 and 495 nm.
Most of these fluorescence bands were quenched when Cu2þ ions
were added (see a detailed explanation at Section 2.4.2) but could
be recovered upon addition of EDTA, meaning the process is re-
versible. This behavior exhibits a pronounced and selective ‘ONe
OFF’ type of fluorescence response toward Cu2þ. However, no dis-
tinct responses were observed upon addition of other metal cations
(Zn2þ, Mg2þ, Co2þ, Ni2þ, Mn2þ, Ca2þ, Agþ). By the way, it is note-
worthy that 4b has shown a similar fluorescence quenching be-
havior (Fig. 2(b)).
Fig. 1. UVevis absorption spectra of 4b in the absence and presence of 100 equiv of
metal ions in ethanol; [4b]¼1.2ꢀ10ꢁ5 mol/L.
2.2. The effect of different metal ions on the UVevis and
fluorescence spectra of 3 and 4 in ethanol
Fig. 1 presents the absorption spectra of 4b in the presence of
100 equiv of metallic ions in ethanol (large excess of cations have

6254
E. Korin et al. / Tetrahedron 67 (2011) 6252e6258
experiments were carried out using compounds 3b and 4b as
examples.
2.4.1. UVevis titration experiments. Fig. 3(a) presents the absorp-
tion spectra of 3b in ethanol as a function of added Cu(ClO4)2. Upon
its addition, the absorption intensities of the bands centered at
about 393, 418, and 440 nm decrease continuously with concomi-
tant appearance (actually, a red shift occurs) of three new bands at
about 419, 441, and 465 nm (due to a red shift) and with increase in
their intensities. A similar behavior was also observed by 4b
[Fig. 3(b)].
Fig. 2. (a) Fluorescent emission spectra of 3b (1.2ꢀ10ꢁ5 M) in the absence and pres-
ence of Cu2þ Zn2þ Mg2þ Co2þ Ni2þ Mn2þ Ca2þ Agþ (100 equiv) in ethanol
, , , , , , ,
(
lex¼418 nm). (b) Relative fluorescence intensities of 3b (gray) and 4b (white)
(1.2ꢀ10ꢁ5 M) in the presence of 100 equiv of Zn2þ, Mg2þ, Co2þ, Ni2þ, Mn2þ, Ca2þ, Agþ
in ethanol (lex¼418 nm, lem¼475 nm, and lex¼419 nm, lem¼475 nm respectively).
2.3. The effect of co-existing cations on the response of Cu2D
in ethanol
Competition experiments of added Cu2þ to 4b solutions in the
presence of one of different metal ions (Zn2þ, Mg2þ, Ni2þ, Mn2þ
,
Ca2þ, and Agþ) indicate that the recognition of Cu2þ by 4b was not
significantly affected by these co-existing cations, because no
change was observed in the red shifted absorption band (except for
a slight change in its intensity; see Supplementary data). Therefore,
the red shift absorption peak of the complex at 470 nm could be
used to monitor Cu2þ concentration in ethanol. Similarly, the
fluorescence quenching of either 4b or 3b in the presence of Cu2þ
was not significantly affected by the presence of a competing
cation.
Fig. 3. UVevis absorption changes of 12
increasing amount of Cu2þ in ethanol (0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 34.5, 42, 49.5, 57,
64, 72, 79, 87, 117, 147, 177 M). Arrows indicate the absorbencies that increased (up)
mM of 3b (a) and 4b (b) upon addition of
m
and decreased (down) during the titration experiments.
Fig. 4 shows a plot of the absorbance (470 nm) profile of 4b
versus the ratio of [Cu2þ]/[4b]. It appears that the binding stoichi-
ometry depends on the total amount of Cu2þ added to the organic
ligand solution. The plot involves an initial linear increase in ab-
sorbance until w0.5 equiv of Cu2þ was added and then, before
a plateau was achieved, a second turning point was observed
around w1.1 equiv of Cu2þ was added. A similar behavior was also
observed by the absorbance (465 nm) profile of 3b.
For comparison, it is noteworthy that quite recently new dif-
ferent fluorescent sensing probes for Cu2þ that contain a carbonyl
group at the
a position of a pyridyl moiety as the metal binding site
were reported.19,20 Mashraqui20 described an ‘ONeOFF’ sensor that
exhibits a dramatic red shift in the absorption spectra. However,
while other cations, such as Liþ, Mg2þ, Ca2þ, and Cd2þ did not cause
any red shift, the presence of Ni2þ, Co2þ, and Zn2þ did cause a red
Additional examination of the binding stoichiometry in [Cu2þ]/
[3b] was done by using MALDI-TOF (as was previously done
by others21) mass spectrometry, and EPR measurements (see
Supplementary data). The MALDI-TOF mass spectrum obtained for
3b in ethanol in the presence of Cu2þ contains signals with m/z of
515.859, 537.845, 700.940, and 1024.865 that could be assigned
to the formation of [3bþCu(II)]þ, [3bþCu(II)ꢁHþNa]þ, [3bþCu(II)þ
4EtOHþH]þ, and [2(3b)þCu(II)þ3H2O]þ complexes, respectively.
The isotopic analysis of these suggested complexes also support
these assumptions. Actually, these observations support the
shift, although much smaller than the one observed with Cu2þ
.
2.4. Titration experiments and binding stoichiometries
Compounds 3 and 4 differ only in the position of the substituted
sulfonamide groups, one at the meta and the other at the para-
position. To further investigate the binding affinity of these com-
pounds to Cu2þ and to compare the influence of the position of the
sulfonamide moiety on the binding ability in ethanol, Cu2þ titration
existence of 1:1 and 2:1 binding ratio between 3b and Cu2þ
,

E. Korin et al. / Tetrahedron 67 (2011) 6252e6258
6255
equation3,22 and found to be (1.15ꢂ0.07)ꢀ105 Mꢁ1. Similar changes
in the absorption spectra were also observed for 4b upon addition
of Cu2þ. For a 1:1 stoichiometry of the 4b complex, Ks was esti-
mated as (3.33ꢂ0.07)ꢀ105 Mꢁ1
.
A non-linear regression analysis (by ReactLab Equilibria pro-
gram) of the same stoichiometry for 3b and 4b afforded binding
constants values of (2.56ꢂ0.05)ꢀ105 Mꢁ1 and be (6.93ꢂ0.01)ꢀ
105 Mꢁ1, respectively.
Consequently, it appears that there is an influence of the posi-
tion of the substituted sulfonamide moiety on the binding constant,
as is reflected by its enhancement when the substituent is located
at the para-position at the 2-oxo-ethylidene side chain compared
with the one at the meta-position. This is in agreement with the
expected electron density donating effect (by resonance) of the
sulfonamide substituent being greater at the para-position.
It is noteworthy that in a previous work16 on caffeoyl- or galloyl-
containing quinoxalone derivatives linked by an arylamido group,
a ratio of 2:1 chelating mode of quinoxalone derivatives to Cu2þ was
found. This indicates that also the nature of the substituent plays an
important role and certainly influences the metaleligand ratio.
Fig. 4. A plot of absorption of compound 4b in ethanol versus the ratio [Cu2þ]/[4b] at
470 nm.
respectively. In addition, EPR titration measurements were carried
out by adding increasing amounts of Cu2þ. The EPR measurements
also support the existence of two different complexes, one pre-
dominates at low concentrations of Cu2þ (<0.8 equiv) and the other
at higher ones. It is important to note that when more than 1 equiv
of Cu2þ was added to 3b there was no evidence of a formation of
new complex (e.g., with a 1:2 ratio).
Based on the above-mentioned findings, a step-by-step process is
suggested (Fig. 5) where first, at low concentrations of Cu2þ
(<0.5 equiv), a 2:1 complex is formed and later, at higher concentra-
tions of Cu2þ (>0.5 equiv) is converted to two possible 1:1 complexes
[(A) and (B)] that may contain also ethanol molecules as ligands.
In all the above possible complexes we have assumed that Cu2þ
binding induced deprotonation of a secondary NH moiety. The
structure of compounds 3 and 4 contain two possible binding sites
to which Cu2þ can bind to form a 1:1 complex with either a six- or
four-membered ring [Fig. 5, (a) and (b), respectively]. It is note-
worthy that preliminary computations (unpublished, by Zeng’s
group) of the binding energy of each of the optimized geometries of
the suggested complexes (using Gaussian 03) indicate that the 1:1
complex, with a six-membered ring, is the more stable one (by
w17 kcal/mol).
2.4.2. Fluorescence titration experiments. Fluorescence titration
experiments were also performed for 3b and 4b. Fig. 6 shows that
upon exciting at 441 nm, the fluorescence intensity of 3b, as an
example, decreased continuously upon addition of Cu2þ, with no
significant change in the position of the emission maxima. Since
Cu2þ is a well-known fluorescence quencher via energy or electron-
transfer processes,23,24 its addition can explain the fluorescence
quenching mechanism for types 3 and 4 compounds.
2.5. Evaluation of Cu2D concentrations in ethanol
In order to examine the possibility that the absorbance spectra
of either 3b (at 465 nm) or 4b (at 470 nm) can be used to quantify
the concentration of Cu2þ, calibration curves in ethanol were
obtained by plotting absorbance as a function of concentration
(0.24e1.08 m
M) of added Cu2þ. Good linear correlations with
R2¼0.961 and 0.995, respectively, were obtained. Then samples of
either 3b or 4b (12 mM) containing 8.95 m
M of Cu2þ and all other
competing cations (8.95 m ,
M of each of Zn2þ, Mg2þ, Co2þ, Ni2þ
Mn2þ, Ca2þ, Agþ) were prepared and absorbencies were measured.
The experimental results obtained for both concentrations of Cu2þ
and absorbencies of its complex with the host are presented in
Table 2. They suggest that both complexing agents (3b and 4b) have
a potential ability to determine the concentration of Cu2þ in
For simplicity, we have estimated (at 418 nm) only the binding
constant (Ks) for the 1:1 stoichiometry between Cu2þ ion and 3b.
The determination of Ks was based on the HildebrandeBenesi
Fig. 5. Proposed stepwise binding processes between Cu2þ and 3b.

6256
E. Korin et al. / Tetrahedron 67 (2011) 6252e6258
Fig. 6. Fluorescent spectra changes of 3b (1.2ꢀ10ꢁ5 M) in ethanol upon addition of
increasing amounts of Cu2þ (from top to bottom: 0, 3, 6, 12, 18, 27
[Excitation at lex¼418 nm provided a similar emission spectrum].
mM); lex¼441 nm.
ethanol in the presence of various competing cations. It is note-
worthy that similar results were obtained either upon addition of
a mixture of competing cations to a solution of the host and Cu2þ or
addition of Cu2þ to a pre-prepared mixture of the host and com-
peting cations.
Fig. 7. (a) UVevis absorption spectra of 3b in the presence of 100 equiv of metal ions in
acetonitrile; [3b]¼1.2ꢀ10ꢁ5 mol/L; (b) The arrow indicates a decrease in absorbance
Table 2
Determination of [Cu2þ] in EtOH in the presence of competing cations
with increasing amounts of Cu2þ (0, 1.5, 3, 4.5, 6, 7.5, 9, 10.5, 12, 13.5,15, 18
mM).
[Cu2þ] (
m
M)a
[Cu2þ
Founda,b
]
Expectedb
absorbance
of complex
Measured
(m
M)
absorbance of
the complexa
3b in EtOH
4b in EtOH
8.95ꢂ0.42
9.30ꢂ0.24
8.75ꢂ0.52
0.198ꢂ0.020
0.274ꢂ0.005
0.203ꢂ0.004
0.270ꢂ0.009
in acetonitrile became colorless. Addition of Cu2þ to a solution of
3b in the presence of other metal ions (Zn2þ, Ni2þ, Mn2þ, Ca2þ
,
In the presence of 8.95 m
M of each of the Zn2þ, Mg2þ, Co2þ, Ni2þ, Mn2þ, Ca2þ, Agþ
a
Agþ) also caused the disappearance of the light yellow color of 3b,
resulting in a dramatic decrease in the absorbance intensity. It
should be pointed out that such a color change has been found to
be irreversible because it could not be recovered upon addition of
excess of EDTA. In contrast to what was observed in ethanol, this
behavior could be attributed to the decomposition of 3b in the
presence of Cu2þ in acetonitrile because Cu2þ is known to oxidize
the nitrogen moiety in similar compounds.25 The reason for the
different behavior in acetonitrile and ethanol stems from the fact
competing cations.
b
Extracted from calibration curves of absorbance versus [Cu2þ] in the presence of
3b or 4b.
2.6. UVevis spectra of 3 and 4 in acetonitrile
The absorption spectra of compounds 3 and 4 in acetonitrile
have also been investigated. Similar to the outcome observed in
ethanol, there are also two intense bands in the visible region for all
compounds studied. For example, 3b exhibits two bands centered
at about 416 and 440 nm with extinction coefficients of 3.3ꢀ104
and 2.8ꢀ104 Lmolꢁ1 cmꢁ1, respectively (Fig. 7). Compared with
ethanol, the wavelengths’ maxima in acetonitrile are also quite
similar. For example, compounds 3b and 4b show lmax¼418 and
419 nm, respectively, in ethanol, and 416 and 417 nm, in
acetonitrile.
that acetonitrile is known to stabilize reduced Cu1þ 26
Since this
.
property causes a color change upon addition of Cu2þ, this type of
compounds can also be used as ‘naked-eye’ Cu2þ dosimeters in
acetonitrile. To support the suggested oxidation-decomposition of
3b by Cu2þ in acetonitrile, EPR measurements were done at room
temperature to a sample of 3b in acetonitrile after adding 2 equiv
of Cu2þ. The observed EPR signal was that of free Cu2þ with no
indication of any complex being formed. However, the intensity of
the signal was about half of the intensity obtained for 2 equiv of
free Cu2þ, probably due to partial oxidation of Cu2þ to Cu1þ (the
latter has no EPR signal). Moreover kinetic measurement of the
sample showed that the signal for Cu2þ gradually diminished and
disappeared completely within w100 min (see Supplementary
data).
2.7. The effect of metal ions on the UVevis and fluorescence
spectra of 3 and 4 in acetonitrile
2.7.1. UVevis measurements. Fig. 7(a) describes a typical and
representative example of UVevis spectra for compounds 3 and 4
in acetonitrile, in the presence of different metal ions. Similar to
what has already been observed in ethanol, no obvious change
In order to prove that the organic ligands decompose in ace-
tonitrile, MALDI-TOF measurements of 3b were conducted upon
addition of 2 equiv Cu2þ. In comparison with the outcome from
ethanol, the results in acetonitrile reveal that a new peak was
observed with m/z of 905.003. This value suggests a dimer
[(3b)2eH ] formation, another evidence for the decomposition of
the ligand.
was detected in the spectra upon addition of metal cations (Zn2þ
,
Ni2þ, Mn2þ, Ca2þ, Agþ) to the ligand solution. However, when Cu2þ
was added (Fig. 7(b)), both bands in the visible region disappeared
and no distinct new absorption was detected (in the range of
320e800 nm). Also, consistently, the yellowish solutions of 3 or 4

E. Korin et al. / Tetrahedron 67 (2011) 6252e6258
6257
2.7.2. Fluorescence measurements. The fluorescent behavior of
compounds 3b and 4b was also investigated in acetonitrile. As
shown in Fig. 8, 3b exhibits two intense bands centered at 474 and
495 nm and both totally disappear upon addition of Cu2þ. It is
plausible that this phenomenon occurs due to the decomposition of
the organic ligand by its oxidation with Cu2þ. In the case of other
cations (after addition of 100 equiv of Zn2þ, Ni2þ, Mn2þ, Ca2þ or
Agþ), only a slight enhancement of the fluorescence intensity was
observed.
In acetonitrile, 3b and 4b behave as dosimeters due to a signif-
icant color change originates from oxidationedecomposition of the
ligand by Cu2þ. Noticeably, this behavior has been found to be ex-
clusive to Cu2þ and not to other metal cations.
Finally, these ligands can also be used as a promising building
block for the construction of more sophisticated functional che-
mosensors or chemical switches.
4. Experimental section
4.1. General
The synthesis and characterization of compounds 3 and 4 was
described previously.15
Absorption spectra were acquired with an Agilent (HP) 8453
UVevis Diode Array spectrophotometer (single beam) with wave-
length accuracy <ꢂ0.5 nm and photometric accuracy <ꢂ0.005 AU.
Fluorescence excitation spectra were acquired with a JASCO
spectrofluorometer (Model FP-6500, Jasco Intern. Co., Japan)
equipped with a 150 W xenon lamp. Both excitation and emission
slit widths were 3 nm. For fluorescence measurements of 3b and 4b
in ethanol, excitations wavelength at 418 and 419 nm were used,
respectively, which were changed to 416 and 417 nm in acetonitrile,
respectively. Mass spectra were obtained with MALDI-TOF spec-
trometer reflex IV (Bruker Daltonic, Germany). EPR spectra were
recorded on Bruker EMX-220 X-band (nw96 Hz) EPR Spectrometer.
All measurements were conducted at 25 ꢃC. In the UVevis
measurements the corresponding solvent (ethanol absolute or
acetonitrile) was used as blank.
4.2. Reagents and general procedure
The solutions of the metal ions were prepared from Cu2þ, Zn2þ
,
Naþ as perchlorate salts and Mg2þ, Co2þ, Ni2þ, Mn2þ, Ca2þ, Agþ as
nitrate salts. (Caution: perchlorate salts are potentially explosive).
All of the reagents were purchased from commercial suppliers
(Aldrich; Merck). For testing the effect of pH 70% HClO4 (Mal-
linckrodt chemicals) was used and base solution was made from
NaOH pellets (Frutarom). The solvents (ethanol or acetonitrile)
were obtained from Bio Lab. Israel. Except for solvent that were
used in MALDI-TOF measurements all of the chemicals used in this
work were of analytical grade and were used without further pu-
rification. In MALDI-TOF measurements HPLC graded solvent were
used without further purification.
Fig. 8. (a) Fluorescent emission spectra of 3b (1.2ꢀ10ꢁ5 M) in the absence and pres-
ence of Cu2þ, Ni2þ, Ca2þ, Zn2þ, Agþ, Mn2þ (100 equiv) in acetonitrile; (b) Relative
fluorescence intensities of 3b (gray) and 4b (white) (1.2ꢀ10ꢁ5 M) in the presence of
Cu2þ, Ni2þ, Ca2þ, Zn2þ, Agþ, Mn2þ (100 equiv) in acetonitrile (lex¼416 nm, lem¼470 nm
for 3b and lex¼417 nm, lem¼475 nm for 4b).
Competition experiments were also carried out by adding Cu2þ
4.2.1. Preparation of sample solutions for the evaluation of selectivity
to specific metallic cation. Stock solutions of 1.2ꢀ10ꢁ1 M of Cu2þ
,
to 4b solutions containing an additional metal ion (Zn2þ, Ni2þ
,
Zn2þ, Mg2þ, Co2þ, Ni2þ, Mn2þ, Ca2þ, Agþ were prepared in ethanol
and acetonitrile. Stock solutions of hosts (1.2ꢀ10ꢁ5 M) 3b and 4b in
ethanol and acetonitrile were also prepared. Test samples were
Mn2þ, Ca2þ or Agþ). The results indicate that also in acetonitrile, the
recognition of Cu2þ by 4b was not significantly affected by either
one of the co-existing cations.
prepared by adding 40
4 mL of a host stock solution. For the competition experiment test
solutions were prepared by adding 40 L of the competing cations
and 40
L of Cu2þ to 4 mL of the host stock solution.
mL of each of their mixture metal stock to
3. Conclusions
m
m
In conclusion, a series of quinoxalin-2-one derivatives (3 and 4)
that contain both fluorophore and ionophore integrated into one
structural unit has been investigated for their affinity to bind var-
ious metal cations. It was found that in ethanol 3b and 4b show
a selective red-shift in the UV-vis absorption spectra and a selective
fluorescence quenching behavior toward Cu2þ, as well as fast and
4.2.2. Preparation of stock solutions and samples for the titration
experiments. Stock solutions of 3ꢀ10ꢁ3 M of Cu(ClO4)2 in ethanol or
acetonitrile were prepared. To 4 mL of compound 3b or 4b
(1.2ꢀ10ꢁ5 M) stock solution was added 0e80
mL of Cu(ClO4)2
solution.
reversible binding. However, no significant response to Zn2þ, Mg2þ
,
Co2þ, Ni2þ, Mn2þ, Ca2þ, and Agþ was observed. Therefore, they
could be utilized as molecular chemosensors for Cu2þ in ethanol. In
addition, it was found that the position at which the sulfonamide
substituent is linked to the central benzene ring affects the binding
4.2.3. Evaluation of the effect of added base/acid on the absorption of
ligands and complexes. Stock solutions of 2.4ꢀ10ꢁ5 M of 3b. To
Stock solutions of 2.4ꢀ10ꢁ5 M of 3b in ethanol and of 2.5ꢀ10ꢁ1
M
NaOH in tri-distilled water were prepared. To test the effect of base
strength toward Cu2þ
.
on host in ethanol, 80 mL of NaOH were added to a solution of 4mL

6258
E. Korin et al. / Tetrahedron 67 (2011) 6252e6258
of 3b. To test the acidity effect on copper complex, 80
m
L of 70%
L of
Project (No. 7112008), and Beijing City Education Committee
(KM201010005009). The authors gratefully acknowledge Dr. A.I.
Shames for performing and analyzing all EPR spectra in this work.
E.K., B.C., and J.Y.B. are thankful to Mrs. E. Solomon for technical
assistance.
HClO4 were added to a solution of 4mL of 3b and 80
m
Cu2þ(1.2ꢀ10-1 M). (Caution: HClO4 is a strong oxidant and explo-
sive). To test the effect of base on host (3b) in ethanol, 80 mL of 70%
HClO4 was added to a solution of 4 mL of 3b (Caution: HClO4 is
a strong oxidant).
4.2.4. Preparation of samples for the evaluation the effect competing
cations. Calibration curves were obtained by plotting the absorbance
(of 3b at 465 nm and 4b at 470 nm) as a function of added Cu2þ. Test
solutions were prepared byadding 3ꢀ10ꢁ3 M of Cu(ClO4)2 to thehost
Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2011.06.047. These data in-
clude MOL files and InChiKeys of the most important compounds
described in this article.
solution. The range ofconcentrationsof addedCu2þ was2.25e16
mM.
To test the potential of these hosts to quantify the concentrations of
Cu2þ, a stock solution of a mixture of 3ꢀ10-3 M of Zn2þ, Mg2þ, Co2þ
,
Ni2þ, Mn2þ, Ca2þ and Agþ was prepared in ethanol. Then six re-
peating samples were prepared for each host by adding 15 mL of the
L of 3ꢀ10-3
M
References and notes
stock solution of the seven competing cations and 15
m
of Cu2þ to a 5 mL of the host stock solution. The concentration of Cu2þ
was calculated by using the suitable calibration curve.
1. de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley, A. J. M.; McCoy, C.
P.; Rademacher, J. T.; Rice, T. E. Chem. Rev. 1997, 97, 1515e1566.
2. Amendola, V.; Fabbrizzi, L.; Foti, F.; Licchelli, M.; Mangano, C.; Pallavicini, P.;
Poggi, A.; Sacchi, D.; Taglietti, A. Coord. Chem. Rev. 2006, 250, 273e299.
3. Valeur, B. Molecular Fluorescence, Principles and Applications; Wiley-VCH:
Weinheim, 2001.
4. Singhal, N. K.; Ramanujam, B.; Mariappanadar, V.; Rao, C. P. Org. Lett. 2006, 8,
3525e3528.
5. Qi, X.; Jun, E. J.; Xu, L.; Kim, S.-J.; Joong Hong, J. S.; Yoon, Y. J.; Yoon, J. J. Org.
4.2.5. Preparation of samples for MALDI-TOF measurements. Stock
solutions (1 mM) of 3b were prepared in either absolute ethanol
(dehydrated) or acetonitrile (HPLC graded). Also solutions of
Cu(ClO4)2 in ethanol or acetonitrile (0.1 M) were prepared. For
measurements in ethanol three samples were prepared by adding
0.5, 1 or 2 equiv of Cu2þ to 2 mL of 3b in ethanol. For samples in
ethanol 2,5-dihydroxybenzoic acid matrix (DHB) (10 mg/mL in
90:10 v/v of deionized water/ethanol) was used. For measurements
conducted in acetonitrile the sample was prepared by adding
2 equiv of Cu2þ to 2 mL of 3b in acetonitrile. The matrix for the
sample in acetonitrile was prepared by dissolving DHB matrix
(20 mg/mL in 2:1 v/v aqua solution of 0.1% trifluoroacetic acid in
deionized water/acetonitrile). In both solvents the MALDI samples
were prepared by mixing 1 mL of the matrix with 1 mL of the
Chem. 2006, 71, 2881e2884.
6. Xu, Z.; Xiao, Y.; Qian, X.; Cui, J.; Cui, D. Org. Lett. 2005, 7, 889e892.
7. Park, S. M.; Kim, M. H.; Choe, J.-I.; No, K. T.; Chang, S.-K. J. Org. Chem. 2007, 72,
3550e3553.
8. Xiang, Y.; Tong, A.; Jin, P.; Ju, Y. Org. Lett. 2006, 8, 2863e2866.
9. Zhang, X.; Shiraishi, Y.; Hirai, T. Org. Lett. 2007, 9, 5039e5042.
10. Gunnlaugsson, T.; Leonard, J. P.; Murray, N. S. Org. Lett. 2004, 6, 1557e1560.
11. Martinez, R.; Zapata, F.; Caballero, A.; Espinosa, A.; Tarraga, A.; Molina, P. Org.
Lett. 2006, 8, 3235e3238.
12. Hazuda, D. J.; Felock, P. Science 2000, 287, 646e650.
13. Hazuda, D. J.; Young, S. D.; Cuare, J. P.; Anthony, N. J.; Gomez, R. P.; Wai, J. S.;
Vacca, J. P.; Handt, L.; Motzel, S. L.; Klein, H. J.; Dornadula, G.; Danovich, R. M.;
Witmer, M. V.; Wilson, K. A.; Tussey, L.; Schleif, W. A.; Gabryelski, L. S.; Jin, L.;
Miller, M. D.; Casimiro, D. R. Science 2004, 305, 528e532.
14. Zhuang, L.; Wai, J. S.; Embrey, M. W.; Fisher, T. E.; Egbertson, M. S.; Payne, L. S.;
Guare, J. P.; Vacca, J. P.; Hazuda, D. J.; Felock, P. J.; Wolfe, A. L.; Stillmock, K. A.;
Witmer, M. V.; Moyer, G.; Schleif, W. A.; Gabryelski, L. J.; Leonard, Y. M.; Lynch,
J. J.; Michelson, S. R.; Young, S. D. J. Med. Chem. 2003, 46, 453e456.
15. Zeng, C.-C.; Li, X.-M.; Yan, H.; Zhong, R.-G. Chin. J. Chem. 2007, 25,
1174e1182.
samples containing 3b in the presence of Cu2þ. Then 1
mL of these
final samples were loaded onto target plate allowed to dry at room
temperature and then analyzed.
4.2.6. EPR measurements. Stock solutions of 4ꢀ10ꢁ2 M Cu(ClO4)2
were prepared in ethanol and acetonitrile. For the measurements
conducted in ethanol titration sample were prepared by adding
16. Xu, Y. S.; Zeng, C.-C.; Li, X. M.; Zhong, R.-G.; Zeng, Y. Chin. J. Chem. 2006, 24,
1086e1094.
15e88 mL of Cu(ClO4)2 solution to 4 mL of compound 3b (1 mM). To
17. Li, X.-M.; Zeng, C.-C.; Niu, L.-T.; Yan, H.; Zheng, D. W.; Zhong, R.-G. Chem. Res.
Chin. Univ. 2006, 22, 747e752.
18. Li, X.-M.; Zeng, C.-C.; Xu, Y.-S.; Yan, H.; Zheng, D. W.; Zhong, R.-G. J. Chem.
Crystallogr. 2006, 36, 357e363.
19. Lin, W.-Y.; Yuan, L.; Cai, X.-W.; Tan, W.; Feng, Y.-M. Eur. J. Org. Chem. 2008, 29,
4981e4987.
20. Mashraqui, S. H.; Khan, T.; Sundaram, S.; Ghadigaonkar, S. Tetrahedron Lett.
2008, 49, 3739e3743.
carry out a qualitative test of the effect of acid on the reversibility of
the complex formation, three drops of 70% HClO4 were added to
a sample of 3b containing 40
sion). Then the signal was re-measured. For the measurements in
acetonitrile the sample was prepared by adding 100 L of Cu(ClO4)2
m
L of Cu2þ (see Section 2.2 for discus-
m
solution to 2 mL of compound 3b (1 mM) and then the signal was re-
measured several times after that during the next 180 min. Also the
signal of free 2 mM Cu2þ in acetonitrile was measured.
21. Singhal, N. K.; Mitra, A.; Rajsekhar, G.; Shaikh, M. M.; Kumar, S.; Guionneau, P.;
Rao, C. P. Dalton Trans. 2009, 39, 8432e8442.
22. Connors, K. A. Binding ConstantsdThe Measurement of Molecular Complex
Stability; John Wiley: New York, NY, 1987.
23. Varnes, A. W.; Dodson, R. B.; Wehry, E. L. J. Am. Chem. Soc. 1972, 94,
946e950.
24. Kemlo, J. A.; Shepherd, T. M. Chem. Phys. Lett. 1977, 47, 158e162.
25. Sumalekshmy, S.; Gopidas, K. R. Chem. Phys. Lett. 2005, 413, 294e299.
26. Kurnia, K.; Giles, D. E.; May, P. M.; Singh, P.; Hefter, G. T. Talanta 1996, 43,
2045e2051 and references therein.
Acknowledgements
This work was supported by grants from the National Basic
Research Program of China (No. 2009CB930200), Beijing Novel