Colorimetric anion chemosensors based on anthraquinone: Naked-eye detection of isomeric dicarboxylate and tricarboxylate anions

Article abstract of DOI:10.1039/b811172c

Three new colorimetric anion receptors 1-3 were synthesized and characterized. Among them, both 1 and 3 showed good sensitivity and selectivity for discrimination of maleate vs. fumarate or malate vs. tartrate by dramatic colour changes in DMSO or DMSO-H<

Full text of DOI:10.1039/b811172c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Colorimetric anion chemosensors based on anthraquinone: naked-eye
detection of isomeric dicarboxylate and tricarboxylate anionsw
Yi-Shan Lin, Guan-Min Tu, Chi-Yung Lin, Yen-Tzu Chang and Yao-Pin Yen*
Received (in Durham, UK) 1st July 2008, Accepted 11th December 2008
First published as an Advance Article on the web 3rd February 2009
DOI: 10.1039/b811172c
Three new colorimetric anion receptors 1–3 were synthesized and characterized. Among them,
both 1 and 3 showed good sensitivity and selectivity for discrimination of maleate vs. fumarate or
malate vs. tartrate by dramatic colour changes in DMSO or DMSO–H2O (95 : 5). Thus, both 1
and 3 can be used as optical chemosensors for recognition of maleate vs. fumarate or malate
vs. tartrate anion. Besides that, the receptor 1 has also a unique colour change for recognition of
either cis-aconitate or trans-aconitate in DMSO, accordingly it can be used for detection of each
isomer and discrimination of cis-aconitate from trans-aconitate anion.
Introduction
Anions, especially dicarboxylates or tricarboxylates, play an
important role in chemical and biochemical processes and
their recognition and sensing by artificial chemo-sensors has
been a focus of interest for chemists in the past decades.1 The
selective recognition of specific anionic species is generally
based on electrochemical, visible and optical response.2
Colour changes, as signaling an event detected by the naked
eye, are widely used owing to the low cost or the lack of
required equipment. The strategy to prepare colorimetric
anion sensors is the binding site-signaling unit approach in
Chart 1
which an appropriate chromophore is attached to a specific
of these anions. In fact, whereas fumarate is generated in the
anion receptor.3 These chromophores may contain electron-
Krebs cycle, maleate is a well known inhibitor of this cycle and
withdrawing groups that can enhance the acidity of the anion
binding subunit. Urea and thiourea subunits are currently
its implication in different kidney diseases has been widely
described.8 On the other hand, the cis-aconitate is generated in
the citric acid cycle, however, the trans-aconitate is an effective
used in the design of neutral receptors for anions, owing to
their ability to act as H-bond donors,4 and many ligands
inhibitor of both aconitase and fumarase.9 Moreover, the
containing either one or two of these groups have been
interest to selective differentiation between the structurally
reported to be excellent sensors for dicarboxylate anions.5
similar analytes, malate and tartrate, is due to the malate
During recent years, we have been studying the synthesis of
being an essential metabolic intermediate, the abnormal levels
colorimetric chemosensors for dicarboxylate anions and their
of its concentration being closely related to some diseases.10
possible application in sensing.6 Now we wish to report the
In this work, a chromogenic unit, an anthraquinone was
preparation of new chromogenic receptors 1–3 and their utility
chosen as a scaffold to link the recognition units (Scheme 1).
in the selective colorimetric discrimination between certain
The introduction of binding sites at 1- and 8-positions of
organic isomers (cis/trans dicarboxylates and cis/trans
1,8-diaminoanthraquinone through a propylene spacer would
tricarboxylates) and structurally similar dicarboxylates
form a flexible convergent binding site for a feasible com-
(malate/tartrate) (Chart 1). Differentiation of geometric
plexation with target species. The 4-nitrophenyl or 4-trifluoro-
isomers is, in general, a difficult task because of their rather
methylphenyl fragment which is linked to the thiourea moiety
similar chemical and physical properties. To the best of our
was chosen as a chromophore to provide spectral sensing
knowledge, only a few examples have been published.6,7
character upon complexation with anions. In spite of lacking
The interest in selective sensors that are able to distinguish
electronic conjugation between the thiourea and anthraqui-
maleate vs. fumarate is not only related to p-diastereoisomer
none moiety, the receptors 1–3 showed UV-vis spectral
recognition but is also due to the different biological behaviour
changes on complexation with anions.
Department of Applied Chemistry, Providence University,
Results and discussion
200 Chungchi Road, Sha-Lu, Taichung Hsien, 433, Taiwan.
E-mail: ypyen@pu.edu.tw; Fax: +886-4-2632-7554;
Preparation of receptors 1–3 is depicted in Scheme 1. A
Tel: +886-4-2632-8001 ext 15218
synthetic intermediate, 1,8-di-(3-aminopropylamino) anthra-
quinone (4) was prepared from 1,8-dichloroanthraquinone.11
w Electronic supplementary information (ESI) available: Further
experimental details. See DOI: 10.1039/b811172c
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
860 | New J. Chem., 2009, 33, 860–867

View Article Online
Scheme 1 Reagents and conditions: (i) 4-nitrophenyl isothiocyanate or 4-nitrophenyl isocyanate, THF, reflux, 18 h; (ii) 4-nitrophenyl
isothiocyanate, CHCl3, reflux, 46 h; (iii) 4-trifluoromethylphenyl isothiocyanate, THF, reflux, 30 h.
Reactions of 4 with 4-nitrophenylisothiocyanate and 4-nitro-
phenyl isocyanate afforded the corresponding receptors 1 and
2, respectively, in moderate yields.12 Reaction of 4 with 0.8
equivalents of 4-nitrophenyl isothiocyanate in CHCl3 gave 5 in
43% yield. Subsequently, reaction of 5 with 4-trifluoromethyl-
Fig. 2 Effect of anions (as (C4H9)4N+ salt) on colour changes of 1 in
phenyl isothiocyanate afforded the receptor 3. All of these
1
compounds were characterized by H NMR, 13C NMR, IR
and HRMS.
DMSO after the addition of various anions: (a) 1 only; (b) 1 + 0.1 equiv.
of maleate; (c) 1 + 2.0 equiv. of maleate; (d) 1 + 2.0 equiv. of
fumarate.
Anion binding studies
(Fig. 1). Interestingly, the colour of the solution of receptor 1
was changed from its initial purple colour, via medium
The colorimetric selective sensing ability of the receptors 1–3
with maleate and fumarate anions in DMSO was monitored
violet–red, to red, visible to the naked eye (Fig. 2). A clear
by UV-Vis absorption and by the naked eye observation. The
isosbestic point at 386 nm indicates the presence of a complex
anions were added as tetrabutylammonium salts to the DMSO
in equilibrium with the free ligand. The changes in the
absorbance as a function of the concentration of maleate
solutions of the receptors 1–3 (5 Â 10À5 M). The receptor 1
itself displays two absorption bands at 353 and 554 nm,
added can be fitted to a 1 : 1 binding equilibrium model, with
the association constant given in Table 1.12 In fact, the 1 : 1
respectively, in DMSO. The interaction of receptor 1 with
maleate anion was investigated in detail through the UV-Vis
ligand-to-anion complex implicated that the two thiourea
spectroscopic titration, and spectral behaviours were observed
(Fig. 1). Upon addition of maleate to receptor 1 in DMSO, the
intensity of the absorption peak at 353 nm was gradually
functionalities of the receptor 1 can act as cooperative
binding sites.
The initial purple to medium violet–red colour change was
decreased, while the band at 493 nm evolved and reached its
observed at low maleate concentrations. This is suggestive of
limiting value after the addition of 2.0 equivalents of maleate
the formation of a charge-transfer complex of the anion
through a hydrogen bonding interaction with receptor 1.
At high guest concentrations, the medium violet–red colour
eventually changed to red. This may be ascribed to the
deprotonation of the N–H proton of the thiourea moiety of
the receptor 1, resulting in the formation of the monodepro-
tonated receptor L1À (1 = L1H). The new band that develops
at À493 nm pertains to the monodeprotonated receptor
L1 which was confirmed by the Brønsted acid–base reaction
of adding 1.0 equivalent of strong base [n-Bu4N]OH
(cf. Fig. SI-1, ESIw). The negative charge causes an increase
in the intensity of the electrical dipole along which the optical
transition takes place, which accounts for the substantial
red-shift of the band (140 nm). Such a deprotonation was
related to the acidity of the H-bond donor site.13 The process
of the combination of hydrogen bonding interaction and
deprotonation of receptor 1 with maleate was corroborated
by 1H NMR titration experiments carried out in DMSO-d6
Fig. 1 Family of spectra taken in the course of the titration of a
5 Â 10À5 M DMSO solution in 1 with a standard solution of maleate
(Fig. 3). It was found that the proton signal of N–H2
(d = 10.12 ppm) which is close to the 4-nitrophenyl group
(signals of N–H protons were assigned by referring to the 2D
at 25 1C. Titration profile (inset) indicates the formation of a 1 : 1
complex.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 860–867 | 861

View Article Online
Table 1 Association constants Ka/M of receptors 1, 2 and 3 with guest anionsa
Anion
1
Rb
2
Rb
3
Rb
Maleatec
(8.74 Æ 0.05) Â 103
(8.31 Æ 0.12) Â 102
(8.88 Æ 0.10) Â 103
(3.85 Æ 0.03) Â 103
(3.83 Æ 0.16) Â 103
(6.66 Æ 0.31) Â 102
0.9963
0.9966
0.9941
0.9955
0.9978
0.9939
(5.29 Æ 0.14) Â 102
(3.82 Æ 0.47) Â 102
(5.31 Æ 0.64) Â 102
(2.63 Æ 0.02) Â 102
(4.58 Æ 0.36) Â 102
(2.29 Æ 0.15) Â 102
0.9958
0.9935
0.9974
0.9913
0.9982
0.9911
(3.72 Æ 0.12) Â 103
(4.21 Æ 0.07) Â 102
(1.92 Æ 0.26) Â 103
(1.82 Æ 0.06) Â 103
(1.85 Æ 0.31) Â 103
(2.18 Æ 0.07) Â 102
0.9923
0.9979
0.9922
0.9939
0.9979
0.9946
Fumaratec
cis-Aconitatec
trans-Aconitatec
Maleatec
Tartratec
a
b
The data were calculated from UV-Visible titration in DMSO. The data values of R were obtained by the results of nonlinear curve
c
fitting. The anions were used as their tetrabutylammonium salts.
interaction, followed by the deprotonation at high concentra-
tion of the guest. To further provide support for the supposi-
tion of the deprotonation, the intermolecular N–HÁ Á ÁO
hydrogen bonded distances were calculated at the
HF/6-31G(d) level using ab initio calculations (Fig. 4). Four
protons of thioureas are directed toward anion ligands but
each hydrogen-bond distance is different as shown in Table 2
of ESI.w Among them, only the proton (H2) which is
connected to the 4-nitrophenyl group has a much shorter
distance to the carboxylic group than a typical hydrogen-bond
distance, which ranges between 1.86 and 2.16 A.16
In contrast, a similar experiment with fumarate anion was
Fig. 3 1H NMR (400 MHz) spectra of sensor 1 (10 mM) in DMSO-d6
carried out and no significant changes in spectra were observed
in the UV-Vis absorption. In this process, no noticeable colour
change was observed and the solution remained a purple
colour (Fig. 2). This indicates that the receptor 1 is weakly
binding or not interacting significantly with fumarate in this
solvent medium. The interaction of receptor 1 with fumarate
upon addition of various quantities of maleate: (a)
(b) 0.1 equiv.; (c) 1.0 equiv.
0 equiv.;
NOESY spectrum of 1) (cf. Fig. SI-2, ESIw), broadened
significantly early in the titration and underwent downfield
shifts with increasing maleate concentration. The N–H2 peak
disappeared after addition of 1.0 equivalent of maleate, whilst
a new signal was observed at d = 20.16 ppm. This confirms the
formation of a [HM]À (M = maleate anion) species.14 The
monodeprotonation is also signaled by the significant upfield
shift of the protons of the phenyl group.15 Such an effect
derives from the through-bond propagation onto the phenyl
framework of the electronic charge generated on N–H depro-
tonation. In addition, the other signal of thiourea (N–H1,
d = 8.42 ppm) was also found to disappear when 1.0
equivalent of maleate was added. The results implied that at
low concentration of maleate, the initial complex is formed by
the receptor 1 with maleate anion through hydrogen bonding
1
was corroborated by H NMR titration experiments. It was
found that when receptor 1 formed a complex with fumarate,
the proton signal of the N–H2 underwent downfield shifts with
increasing fumarate concentration from 10.13 to 11.89 ppm
and the chemical shift of N–H1 changed from 8.42 to
10.32 ppm (Fig. 5). The relatively small downfield shifts
indicate that the complex is formed solely through weak
hydrogen bonding interactions and is inconsistent with a
deprotonation process between fumarate and the receptor.
This rationale was also supported by the ab initio calculations
(see Table 2, ESIw). The comparison of the UV-Vis absorption
spectra of the complex 1 upon addition of either maleate or
fumarate is shown in the ESI (Fig. SI-3, ESIw).
Fig. 4 Optimized geometries from ab initio HF/6-31G(D) calculations.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
862 | New J. Chem., 2009, 33, 860–867

View Article Online
Fig. 5 1H NMR (400 MHz) spectra of sensor 1 (10 mM) in DMSO-d6
upon addition of various quantities of fumarate: (a) 0 equiv.;
(b) 0.5 equiv.; (c) 1.0 equiv.
Fig. 7 UV-Vis spectra change of 1 in DMSO (5 Â 10À5 M) after the
addition of 2.0 equivalents of anions: (a) 1 only; (b) 1 + cis-aconitate;
(c) 1 + trans-aconitate.
Fig. 8 Effect of anions (as (C4H9)4N+ salts) on colour changes of 1 in
DMSO after the addition of 2.0 equivalents of anions: (a) 1 only;
(b) 1 + cis-aconitate; (c) 1 + trans-aconitate.
Fig. 6 Possible binding model of 1 with maleate anion.
Apparently, the receptor 1 has a unique colour change
and higher sensitivity and selectivity recognition for maleate
than fumarate, and can act as an optical chemosensor for
recognition of maleate vs. fumarate. The different sensitivity
and selectivity with maleate and fumarate can be related to the
receptor stereochemistry that gives rise to different geometries
depending on the anion stereochemistry. Thus, the maleate
anion with its cis configuration perfectly fits into the complex
inducing a conformation change in the receptor. By contrast,
the fumarate anion with a trans disposition of carboxylate
moiety which does not induce changes in the ligand conforma-
tion results in a negligible change of the UV-Vis absorption.
The proposed conformational structure for the complex
formed between receptor 1 and maleate dianion is shown in
Fig. 6. Besides that, the basicity of the anion may also play an
important role for recognition. Since the maleate dianion
Fig. 9 Possible binding model of 1 with cis-aconitate anion.
vs. trans-aconitate. The selectivity of 1 for recognition of these
anions can also be rationalized on the basis of the receptor
geometry that gives rise to different geometries depending on
the anion stereochemistry and the basicity of the anionic
species (pKa values of the cis-aconitic and the trans-aconitic
3
(pKa = 6.23 in H2O, 18.8 in DMSO) is more basic than the
fumarate dianion (pKa = 4.46 in H2O, 11.0 in DMSO,
are 6.21 and 6.16 in H2O, respectively).17 The cis-aconitate can
fit more perfectly into the complex inducing a conformation
change in the receptor. The additional carboxylate group can
enhance the deprotonation effect of N–H2 and resulted in a
higher absorbance in the UV-Vis spectra. The proposed
conformational structure for the complex formed between
receptor 1 and cis-aconitate is shown in Fig. 9. The dramatic
colour changes can be attributed to the occurrence of the
combination of hydrogen bonding interaction and deprotona-
tion of the thiourea proton with the anion, similar to that
mentioned above. The monodeprotonated receptor is responsible
for the absorption at 489 nm. This was similarly confirmed
2
2
respectively),17 deprotonation of N–H2 will occur preferen-
tially for the maleate anion.
The strong selectivity of receptor 1 for maleate compared to
that of fumarate prompted us to investigate the geometric
tricarboxylates, cis-aconitate and trans-aconitate anions.
A similar phenomenon of UV-Vis absorption is observed in
Fig. 7. Spectrum (a) was measured in the absence of anion. As
shown in spectrum (b) and (c), the CT absorption bands
appear at 489 nm and the solution colour changes from purple
to red for cis-aconitate and from purple to medium violet–red
for trans-aconitate, respectively (Fig. 8). It is apparent that
receptor 1 has a selective colour change for cis-aconitate
1
by H NMR titration experiments with cis-aconitate in which
the peak of [HM]À (M = cis-aconitate anion) appeared
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 860–867 | 863

View Article Online
at 20.20 ppm (Fig. SI-4, ESIw). However, when trans-aconitate
was titrated against receptor 1 significant broadening of N–H1
and N–H2 peaks were observed and indeed after the addition
of 2.0 equivalents the resonances assigned to both N–H1 and
N–H2 completely disappeared but no [HM]À (M = trans-
aconitate anion) peak was observed in the range of 17–21 ppm
(cf. SI-4). This result implied that the receptor 1 may strongly
bind with trans-aconitate through multiple hydrogen bonds.
To further discriminate whether the process is a hydrogen
bond complexation or a proton transfer of a Brønsted-type
acid–base equilibrium, a 1H NMR dilution experiment of a
1 : 1 mixture of 1 and trans-aconitate was performed (cf. SI-5).
If the process is a hydrogen bonding interaction, the equili-
brium is shifted to the dissociation direction; therefore,
spectral shifts should be observed by dilution. On the other
hand, if the process is a proton transfer, the equilibrium is
independent of the concentration of host and guest because
the equilibrium constant is dimensionless.18 The CH protons
of the phenyl groups showed a shift to the direction of
dissociation by dilution. Therefore, this process is clearly
hydrogen bond complexation rather than proton transfer.
Judging from the UV-Vis titrations, the Job plot method
showed the formation of a 1 : 1 stoichiometry complex of 1
with either cis-aconitate or trans-aconitate (Fig. SI-6, ESIw).
The association constants were calculated and listed in Table 1.
To further elucidate the interaction between cis-aconitate
or trans-aconitate with receptor 1, ab initio calculations
of the [1Ácis-aconitate] complex and the [1Átrans-aconitate]
complex were undertaken. These calculations clearly shows
only the proton (H2) of the cis-aconitate has a much shorter
distance to the carboxylic group than a typical H-bond
distance (see Table 2, ESIw).
Fig. 11 Effect of anions (as (C4H9)4N+ salts) on colour changes of 1
in DMSO after the addition of 2.0 equivalents of anions: (a) 1 only;
(b) 1 + malate; (c) 1 + tartrate.
exhibits negligible perturbation upon addition of 2.0 equiva-
lents of tartrate and the solution still retains a purple colour
(Fig. 11). Apparently, the receptor 1 has a unique colour
change and higher selectivity for malate than tartrate. That
the receptor 1 preferentially binds malate over tartrate anion is
also interpreted in terms of the receptor geometry and the
higher basicity of the malate (pKa values of malic acid and
2
tartaric acid are 5.20 and 4.40 in H2O, respectively).19 The
tartrate with an additional hydroxyl group that would reduce
the binding ability between the carboxyl group and the N–H
of the thiourea moiety, will result in weak or insignificant
interaction with the receptor. The proposed conformational
structure for the complex formed between receptor 1 and
tartrate is shown in Fig. 12. From 1H NMR titration of
receptor 1 with malate, the signal of N–H2 disappeared and
the N–H1 signal underwent a downfield shift from 8.42 to
11.35 ppm (Fig. 13). However, no [HM]À (M = malate anion)
peak was observed. To discriminate whether the process is a
hydrogen bond complexation or a proton transfer, a 1H NMR
dilution experiment of a 1 : 1 mixture of 1 and malate was
performed (Fig. SI-7, ESIw). The CH protons of the phenyl
groups also showed a shift to the direction of dissociation
by dilution. Therefore, the process is demonstrated to be
hydrogen bond complexation.18
In order to investigate whether the similar structural
dicarboxylates can be differentiated by colour change, the
binding of receptor 1 with malate and tartrate were studied.
Consistent with the above results of the titration of maleate
and fumarate, the titration of 1 with malate and tartrate gave
similar phenomena in their UV-Vis absorption spectra
(Fig. 10). As shown in spectrum (b) the CT absorption band
appears at 492 nm and the solution colour changes from
purple to purple–red (Fig. 11). On the contrary, spectrum (c)
On the other hand, for titration with tartrate anion, the
chemical shift of the N–H1 changed from 8.42 to 11.05 ppm
and the signal of the N–H2 changed from 10.12 to 13.28 ppm,
respectively (Fig. 14). The relatively small downfield shifts
indicate the formation of a complex through intermolecular
hydrogen bonds between the receptor 1 and tartrate anion.
The Job plots show each anion and receptor 1 formed the
complexes in 1 : 1 binding stoichiometry (Fig. SI-8, ESIw). The
binding constants were determined and listed in Table 1.
In order to gain a clear picture of how the urea unit can
affect the binding property of 1, a UV-Vis study was con-
ducted on the control compound 2. Upon progressive increase
Fig. 10 UV-Vis spectra change of 1 in DMSO (5 Â 10À5 M) after the
addition of 2.0 equivalents of anions: (a) 1 only; (b) 1 + malate;
(c) 1 + tartrate.
Fig. 12 Possible binding model of 1 with tartrate anion.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
864 | New J. Chem., 2009, 33, 860–867

View Article Online
tricarboxylate anions (cis-aconitate and trans-aconitate) were
also studied. Unfortunately, no distinct colour change was
observed (Fig. SI-11, ESIw).
To explore potential and analytical applications of the
sensor 1 for anions tested, UV-Vis titrations were carried
out in DMSO–H2O (95 : 5) mixtures (Fig. SI-12, ESIw). As
we know, protic solvent such as H2O and ethanol would
compete with the anion for binding sites of the host and,
therefore, could disturb the hydrogen-bonding interactions
between host and guest. However, in the case of the sensor
1, the distinct colour changes for the recognition of maleate
and malate anion in DMSO–water (95 : 5) solution were also
observed (Fig. SI-13, ESIw). The results revealed that the
spectrum changed similarly as that of dry DMSO solution.
Likewise, there was significant increase in the intensity of the
band at 493 nm upon addition of maleate or malate, which
corresponded to light to dark colour changes. This result
indicated that the addition of small amounts of water had
little effect on maleate or malate anion sensing property of 1.
On the other hand, a distinct colour change was also observed
for the recognition of tricarboxylate anions (cis-aconitate
and trans-aconitate) with sensor 1 in DMSO–water (97 : 3)
solution. Thus, the addition of water to DMSO up to 3% did
not interfere in the anion sensing (Fig. SI-13, ESIw). This
probably is due to that receptor 1 bears two thiourea groups
which have been proven to be good hydrogen bonding accep-
tors that could make strong hydrogen bonds with anions,
which water molecules could not break.21 Additionally,
DMSO is a very good hydrogen-bond accepting solvent
and in this situation it can interact strongly with the water
molecules at the concentration utilized. Consequently, the
possibility of the interaction of anions with the sensor was
improved greatly.
Fig. 13 1H NMR (400 MHz) spectra of sensor 1 (10 mM) in
DMSO-d6 upon addition of various quantities of malate: (a) 0 equiv.;
(b) 1.0 equiv.
Fig. 14 1H NMR (400 MHz) spectra of sensor 1 (10 mM) in
DMSO-d6 upon addition of various quantities of tartrate: (a) 0 equiv.;
(b) 1.0 equiv.
of the concentrations of different anions such as maleate,
cis-aconitate, malate and other isomeric anions, no significant
change in the UV-Vis spectra or colour changes were
observed. This weak binding could be explained by the
strongly electronegative oxygen atom of the urea subunit
which poorly contributes to the charge transfer transition.
Thiourea is a much stronger protonic acid than urea
(pKa = 21.1 and 26.9, respectively, in DMSO).20
Furthermore, in order to check how the different substituent
on the one of the two thiourea groups would influence the
anion binding and sensing property, a 4-nitrophenyl group
Conclusion
was replaced by
a weaker electron-withdrawing group,
4-trifluoromethylphenyl group, in receptor 3. The UV-Vis
absorption spectral profiles of 3 with addition of either
maleate or malate are similar to those of 1 with these two
anions. It appeared that the receptor 3 has the same propen-
sity; combination of hydrogen bonding interaction and
deprotonation of the N–H2 group occurred upon addition
In conclusion, the new colorimetric anion receptors 1–3 were
synthesized. Among them, both 1 and 3 show good sensitivity
and selectivity for discrimination of maleate vs. fumarate or
malate vs. tartrate by dramatic colour changes in DMSO or
DMSO–H2O (95 : 5). Thus, both 1 and 3 can be used as optical
chemosensors for recognition of maleate vs. fumarate or
malate vs. tartrate anion. Besides that, the receptor 1 has also
a unique colour change for recognition of either cis-aconitate
or trans-aconitate in DMSO or DMSO–H2O (97 : 3),
accordingly it can be used for detection of either cis-aconitate
or trans-aconitate anion. To the best of our knowledge, this is
the first synthetic host that can selectively recognize either
cis-aconitate or trans-aconitate anion by colour changes.
Future work will continue to develop the practical colori-
metric sensors for recognition of chiral dicarboxylates.
of the increasing concentration of maleate whereas
a
strong hydrogen bonding complexation occurred upon addi-
tion of the increasing concentration of malate anion.
This is similarly proved by 1H NMR titration experiments
(Fig. SI-9, ESIw).
Consistent with the result of receptor 1, titration of 3 with
either fumarate or tartrate anion also gave no apparent colour
change. The colour of the solution still remains the original
purple (Fig. SI-10, ESIw). The weak hydrogen-bonding
between 3 and fumarate or tartrate was corroborated by the
1H NMR titration experiments and the hydrogen-bonded
distances were also reflected in the ab initio calculations
(Table 3, ESIw). Due to the fact that receptor 3 has the unique
colour change and higher selectivity for maleate than fumarate
anion or malate than tartrate, it can be used as an optical
chemosensor for recognition of maleate vs. fumarate or
malate vs. tartrate. The binding of receptor 3 with isomeric
Experimental
General
The chemical reagents were purchased from Acros or Aldrich
Corporation and utilized as received, unless indicated
otherwise. All solvents were purified by standard procedures.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 860–867 | 865

View Article Online
Melting points were measured on
a
Yanaco MP-S3
The resulting mixture was stirred and heated to reflux
for 18 h. After cooling to room temperature, the solution
was concentrated in vacuum. The crude product was washed
with CH2Cl2 several times to afford pure 1. Yield 54%, mp
202–203 1C. 1H NMR (600 MHz, DMSO-d6): d 1.94–2.04
(m, 4H), 3.36–3.46 (m, 4H), 3.60–3.72 (m, 4H), 7.21 (d, 2H, J =
8.4 Hz), 7.38 (d, 2H, J = 7.2 Hz), 7.58 (dd, 2H, J = 7.8 Hz, J =
8.4 Hz), 7.79 (m, 4H), 8.13 (m, 4H), 8.42 (br s, 2H), 9.63
(br s, 2H), 10.12 (br s, 2H). 13C NMR (DMSO-d6): d 27.7,
40.0, 41.8, 113.4, 114.3, 118.4, 120.3, 124.4, 133.5, 134.5, 141.7,
146.3, 150.7, 180.1, 183.4, 187.9. FT-IR (KBr): 3278, 2925,
2853, 1614, 1511, 1322, 1214 cmÀ1. UV-Vis (DMSO): 353 nm
melting-point apparatus. The infrared spectra were performed
on a Perkin Elmer System 2000 FT-IR spectrophotometer.
UV–Vis spectra were measured on a Cary 300 spectrometer.
All NMR spectra were measured on a Bruker spectrometer at
400 (1H) and 100 MHz (13C) and Varian Unity Inova-600
spectrometers at 600 (1H) and 150 MHz (13C) with DMSO-d6
as solvent. High-resolution mass spectra were measured with a
Finnigan/Thermo Quest MAT 95XL instrument.
Preparation of tetrabutylammonium salts
To a stirred solution of a di- or tricarboxylic acid (2.5 or 1.0 mmol)
in dry methanol (5 mL), 2.0 or 3.0 equiv. of a 1.0 M solution
of tetrabutylammonium hydroxide in methanol (5 mL) was
added. The resulting mixture was stirred for 2 h at room
temperature. The solvent was evaporated in vacuo and dried
over P2O5. The resulting tetrabutylammonium salt was stored
under anhydrous condition before use.
(e = 35 454), 554 (14 188). HRMS (FAB) calc. for
C34H32O6N8S2 [M + 1] 713.1890; found 713.1968.
Synthesis of 1,8-bis(4-nitrophenylurelyenepropeneamino)-
anthraquinone (2). A similar procedure to synthesis of 1 was
carried out using 4-nitrophenyl isocyanate in place of 4-nitro-
1
phenyl isothiocyanate. Yield 43%, mp 251–252 1C. H NMR
Synthesis of 1,8-di(3-aminopropylamino)anthraquinone (4).
A solution of 1,8-dichloroanthraquinone (1.00 g, 3.60 mmol),
K2CO3 (1.49 g), 1,3-diaminopropane (1.03 g, 13.64 mmol) and
tetrabutylammonium hydrogensulfate (20 mg) (catalyst) in
CH3CN (30 mL) was stirred and heated to reflux for 12 h,
the solid residue was filtered off. After cooling, the reaction
mixture was treated with crushed ice. The resulting precipitate
was collected by filtration, washed well with water and further
purified by recrystallization from ethyl acetate–n-hexane to
afford the product as red needles. Yield: 1.01 g (80%), mp
162–163 1C. 1H NMR (CDCl3): d 1.94–1.99 (m, 4H), 2.82–2.84
(m, 4H), 2.97–2.99 (m, 4H), 3.42–3.46 (m, 4H), 7.07–7.10
(m, 2H), 7.48–7.51 (m, 2H), 7.57–7.60 (m, 2H), 9.67 (br s, 2H).
HRMS (FAB) calc. for C20H24O2N4 [M + 1] 353.1980; found
353.1974.
(600 MHz, DMSO-d6): d 1.81–1.90 (m, 4H), 3.23–3.31
(m, 4H), 3.34-3.43 (m, 4H), 6.61 (br s, 2H), 7.17 (d, 2H, J =
8.4 Hz), 7.35 (d, 2H, J = 7.2 Hz), 7.55 (dd, 2H, J = 7.8 Hz,
J = 8.4 Hz), 7.59 (m, 4H), 8.08 (m, 4H), 9.27 (br s, 2 H), 9.60
(br s, 2H). 13C NMR (DMSO-d6): d 29.0, 37.2, 40.0, 113.4,
114.2, 116.7, 118.2, 125.0, 133.5, 134.4, 140.3, 147.2, 150.7,
154.5, 183.4, 187.9. FT-IR (KBr): 3304, 3088, 2919, 2853,
1660, 1614, 1562, 1491, 1393, 1322, 1230 cmÀ1. UV-Vis
(DMSO): 344 nm (e = 38 460), 552 (11 484). HRMS (FAB)
calc. for C34H32O8N8 [M + 1] 681.2346; found 681.6412.
Synthesis of 1-(4-nitrophenylthiourelyenepropeneamino)-8-(4-
trifluoromethylphenylthiourelyenepropeneamino)anthraquinone (3).
To a stirred solution of 5 (0.26 g, 0.48 mmol) in THF (80 mL),
4-trifluoromethylphenyl isothiocyanate (0.15 g, 0.72 mmol)
in THF (40 mL) was added at room temperature. The
resulting mixture was stirred and heated to reflux for 30 h.
After cooling to room temperature, the solution was concen-
trated in vacuum. The crude product was washed with
CH2Cl2 several times to afford pure 3. Yield 41%, mp
200–201 1C. 1H NMR (400 MHz, DMSO-d6): d 1.87–2.13
(m, 4H), 3.38–3.48 (m, 4H), 3.52–3.79 (m, 4H), 7.13–7.25
(m, 4H), 7.36–7.41 (m, 2H), 7.52–7.65 (m, 4H), 7.69 (br s,
1H), 7.75–7.81 (m, 2H), 7.95 (br s, 1H), 8.08–8.18 (m, 2H),
8.53 (br s, 1H), 9.59 (br s, 1H), 9.65 (br s, 1H), 10.30 (br s, 1H).
13C NMR (DMSO-d6): d 22.9, 25.3, 27.9, 28.9, 36.6, 42.0, 67.2,
113.5, 113.6, 113.7, 114.4, 114.5, 118.6, 120.4, 124.7, 133.8,
134.7, 150.9, 169.4, 183.6, 188.1. FT-IR (KBr): 3291, 2931,
2864, 2348, 1614, 1512, 1327, 1303, 1215 cmÀ1. UV-Vis
(DMSO): 345 nm (e = 19 089), 360 (18 316), 559 (12 648).
HRMS (FAB) calc. for C35H32O4N7F3S2 [M + 1] 735.1936;
found 735.2004.
Synthesis of 1-(3-aminopropylamino)-8-(4-nitrophenylthio-
urelyenepropeneamino)anthraquinone (5). To a stirred solution
of 4 (0.30 g, 0.85 mmol) in CHCl3 (50 mL), 4-nitrophenyl
isothiocyanate (0.12 g, 0.67 mmol) in CHCl3 (30 mL) was
added at room temperature. The resulting mixture was stirred
and heated to reflux for 46 h. After cooling to room tempera-
ture, the solution was concentrated in vacuum. The crude
product was washed with CHCl3 several times to afford pure 5.
Yield: 0.42 g (43%), mp 187–189 1C. 1H NMR (400 MHz,
DMSO-d6): d 1.83–2.08 (m, 4H), 3.33 (br s, 2H), 3.39–3.48
(m, 4H), 3.52–3.75 (m, 4H), 7.20 (d, 2H, J = 8.4 Hz), 7.36
(d, 2H, J = 7.2 Hz), 7.59 (dd, 2H, J = 7.6 Hz, J = 8.0 Hz),
7.79 (d, 2H, J = 9.2 Hz), 8.12 (d, 2H, J = 8.8 Hz), 8.52
(br s, 1H), 9.62 (br s, 2H), 10.24 (br s, 1H). 13C NMR (DMSO-d6):
d 27.9, 31.5, 40.6, 42.0, 113.6, 114.5, 118.6, 120.5, 124.7, 133.8,
134.7, 146.6, 150.9, 180.3, 183.7, 188.1. FT-IR (KBr): 3446,
2929, 2867, 2346, 1741, 1614, 1511, 1326, 1217 cmÀ1. UV-Vis
(DMSO): 333 nm (e = 19 661), 543 (14 425). HRMS
(FAB) calc. for C27H28O4N6S [M + 1] 533.1895; found
533.1964.
Acknowledgements
Synthesis of 1,8-bis(4-nitrophenylthiourelyenepropeneamino)-
anthraquinone (1). To a stirred solution of 4 (0.20 g, 0.34 mmol)
in THF (40 mL), 4-nitrophenyl isothiocyanate (0.18 g, 1.03 mmol)
in THF (40 mL) was added at room temperature.
We thank the National Science Council of the Republic of
China for financial support (NSC 96-2113-M-126-001-MY2)
and the National Center for High-Performance Computing
for computing time and facilities.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
866 | New J. Chem., 2009, 33, 860–867

View Article Online
Am. J. Physiol., 1976, 231, 1010–1017; (c) S. EiamOng, M. Spohn,
N. A. Kurtzman and S. Sabatini, Kidney, 1995, 48, 1542–1548.
9 J. E. Kata, S. D. Dumlao, J. I. Wasserman, M. G. Lansdown,
M. E. Jung, K. F. Faull and S. Clarke, Biochemistry, 2004, 43,
5976–5986.
References
1 (a) J. L. Pierre and P. Baret, Bull. Soc. Chim. Fr., 1983, 367–380;
(b) S. Shinoda, M. Tadokoro, H. Tsukube and R. Arakawa, Chem.
Commun., 1998, 181–182.
2 (a) Anion Receptor Chemistry, eds. J. L. Sessler, P. A. Gale and W.-S.
Cho, RSC Publishing, Cambridge, 2006; (b) Supramolecular Chemistry
of Anions, eds. E. Bianchi, K. Bowman-James and E. Garcia-Espana,
Wiley-VCH, New York, 1997; (c) R. Martinez-Manez and
F. Sancenon, Chem. Rev., 2003, 103, 4419–4476.
10 (a) A. S-Y. Lo, C.-T. Liew, S.-M. Nagi, S. K.-W. Tsui, K.-P. Fung,
C.-Y. Lee and M. M.-Y. Waye, J. Cell. Biochem., 2005, 94,
763–773; (b) P. Bubber, V. Haroutunian, G. Fisch, J. P. Blass
and G. E. Gibson, Ann. Neurol., 2005, 57, 695–703;
(c) B. A. Barshop, Mitochondrion, 2004, 4, 521–527.
3 (a) C. Suksai and T. Tuntulani, Chem. Soc. Rev., 2003, 32,
192–202; (b) X. H. Qian and F. Liu, Tetrahedron Lett., 2003, 44,
795–799; (c) P. D. Beer and E. J. Hayes, Coord. Chem. Rev., 2003,
240, 167–189; (d) P. A. Gale, Chem. Rev., 2003, 240, 191–221;
(e) D. W. Yoon, H. Hwang and C. H. Lee, Angew. Chem., Int. Ed.,
2002, 41, 1757–1759; (f) J. Ren, Q. Wang, D. Qu, X. Zhao and
H. Tian, Chem. Lett., 2004, 33, 974–975; (g) B. Liu and H. Tian,
Chem. Lett., 2005, 34, 686–687; (h) V. Amendola, D. Esteban-
Gomez, L. Fabbrizzi and M. Licchelli, Acc. Chem. Res., 2006, 39,
343–353.
11 H.-S. Huang, H.-F. Chiu, W.-C. Lu and C.-L. Yuan, Chem.
Pharm. Bull., 2005, 53, 1136–1139.
12 K. A. Connors, Binding Constants. The Measurement of Molecular
Complex Stability, John Wiley & Sons, New York, 1987.
13 S. Gronert, J. Am. Chem. Soc., 1993, 115, 10258–10266.
14 (a) L. J. Altman, D. Laungani, G. Gunnarsson, H. Wennerstrom
and S. Forsen, J. Am. Chem. Soc., 1978, 100, 8264–8266;
(b) P. Schah-Mohammedi, I. G. Shenderovich, C. Detering,
H.-H. Limbach, P. M. Tolstoy, S. N. Smirnov, G. S. Denisov
and N. S. Golubev, J. Am. Chem. Soc., 2000, 122, 12878–12879.
15 M. Boiocchi, L. Del Boca, D. Esteban-Gomez, L. Fabbrizzi,
M. Licchelli and E. Monzani, J. Am. Chem. Soc., 2004, 126,
16507–16514.
4 (a) P. A. Gale, Coord. Chem. Rev., 2001, 213, 79–128;
(b) P. D. Beer and P. A. Gale, Angew. Chem., Int. Ed., 2001, 40,
486–516.
5 (a) Z. Y. Zeng, Y. B. He, L. H. Wei, J. L. Wu and L. Z. Weng, Can.
J. Chem., 2004, 82, 454–460; (b) D. A. Jose, D. K. Kumar,
B. Ganguly and A. Das, Org. Lett., 2004, 6, 3445–3448;
(c) J. L. Wu, Y. B. He, L. H. Wei, L. Z. Meng and T. X. Yang,
Aust. J. Chem., 2005, 58, 53–57; (d) E. J. Cho, B. J. Ryu, Y. J. Lee
and K. C. Nam, Org. Lett., 2005, 7, 2607–2609.
16 Y.-J. Kim, H. Kwak, S. J. Lee, J. S. Lee, H. J. Kwon, S. H. Nam,
K. Lee and C. Kim, Tetrahedron, 2006, 62, 9635–9640.
17 (a) L. Pfendt, B. Drazic, G. Popovic, B. Drakulic, Z. Vitnik and
I. Juranic, J. Chem. Res. (M), 2003, 501–514; (b) P. J. Choi,
K. A. Petterson and J. D. Roberts, J. Phys. Org. Chem., 2002, 15,
278–286.
6 (a) Y.-P. Yen and K.-W. Ho, Tetrahedron Lett., 2006, 47,
1193–1196; (b) Y.-P. Yen and K.-W. Ho, Tetrahedron Lett.,
2006, 47, 7357–7361; (c) Y.-P. Tseng, G.-M. Tu, C.-H. Lin,
C.-T. Chang, C.-Y. Lin and Y.-P. Yen, Org. Biomol. Chem.,
2007, 5, 3592–3598.
7 (a) F. Sancenon, R. Martinez-Manez, M. A. Miranda, M.-J. Segui
and J. Soto, Angew. Chem., Int. Ed., 2003, 42, 647–650;
(b) A. M. Costero, M. Colera, P. Gavina and S. Gil, Chem.
Commun., 2006, 761–763.
18 (a) S.-I. Kondo, T. Harada, R. Tanaka and M. Unno, Org. Lett.,
2006, 8, 4621–4624; (b) C. S. Wilcox, in Frontiers in Supramolecular
Organic Chemistry and Photochemistry, eds. H.-J. Schnerider and
H. Durr, VCH, Weinheim, 1991, pp. 123–143.
19 R. M. C. Dawson, Data for Biochemical Research, Clarendon
Press, Oxford, 1959.
20 (a) F. G. Bordwell, Acc. Chem. Res., 1988, 21, 456–463;
(b) D. Esteban-Gomez, L. Fabbrizzi, M. Licchelli and
E. Monzani, Org. Biomol. Chem., 2005, 3, 1495–1500.
21 T. Gunnlaugsson, P. E. Kruger, P. Jensen, J. Tierney, H. D. P. Ali
and G. M. Hussey, J. Org. Chem., 2005, 70, 10875–10878.
8 (a) J. Tepinski, D. Pawlowska and S. Angielski, Acta Biochim. Pol.,
1984, 31, 229–240; (b) A. Gougoux, G. Lemieux and N. Lavoie,
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
New J. Chem., 2009, 33, 860–867 | 867