Squaramide-based reagent for selective chromogenic sensing of Cu(II) through a zwitterion radical

Article abstract of DOI:10.1021/ol101514y

A minimalist squaramide-based chemodosimeter for Cu²⁺ is described. Upon selective chelation to 2, Cu²⁺ induces the formation of a highly colored zwitterionic radical, which is kinetically stable for hours. The presence of a radical is confirmed by EPR and ESI-MS. It is then possible to use reagent 2 for visual and selective sensing of Cu²⁺ at neutral pH.

Full text of DOI:10.1021/ol101514y

ORGANIC
LETTERS
2010
Vol. 12, No. 17
3840-3843
Squaramide-Based Reagent for Selective
Chromogenic Sensing of Cu(II) through
a Zwitterion Radical
Elena Sanna,† Luis Mart´ınez,† Carmen Rotger,† Salvador Blasco,‡
Jorge Gonza´lez,‡ Enrique Garc´ıa-Espan˜a,‡ and Antoni Costa*,†
Departament de Qu´ımica, UniVersitat de les Illes Balears,
07122 Palma de Mallorca, Spain, and Instituto de Ciencia Molecular (ICMOL),
UniVersidad de Valencia, C/catedra´tico Jose´ Beltra´n 2, 46098 Valencia, Spain
antoni.costa@uib.es
Received July 1, 2010
ABSTRACT
A minimalist squaramide-based chemodosimeter for Cu2+ is described. Upon selective chelation to 2, Cu2+ induces the formation of a highly
colored zwitterionic radical, which is kinetically stable for hours. The presence of a radical is confirmed by EPR and ESI-MS. It is then possible
to use reagent 2 for visual and selective sensing of Cu2+ at neutral pH.
Copper-induced inhibition of important metabolic processes
in plants1 and in humans indicates that upregulated levels
of copper are causative in several neurodegenerative diseases,
such as Alzheimer’s and Parkinson’s diseases, atheroscle-
rosis, and diabetes.2 Moreover, because of the widespread
use of copper for industrial, agricultural, and household uses,
episodes of Cu contamination of natural waters are relatively
frequent. Although organically bound copper appears to be
less toxic, free solvated Cu2+ is particularly damaging as it
catalyzes the formation of reactive organic species (ROS)
that cause damage to most biomolecules.3 A large number
of chemodosimeter4 and chemosensing5 devices for copper
have been proposed so far. However, it is still highly
advisible to develop new or improved methods for the
selective in situ evaluation of solvated Cu2+ in aqueous
environments. In this regard, visible detection of excess Cu2+
in untreated natural waters with a copper-selective chro-
mogenic sensor offers the possibility of detecting copper
contamination events in their early stages.
Properties required for an effective Cu2+ chemodosimeter
include selectivity over other metal ions, high sensitivity with
minimum manipulation of samples, and structural and
synthetic simplicity.
In this work, we present a minimalist turn-on copper
chemodosimeter featuring high selectivity and sensitivity for
Cu2+. The new radical-based5u reagent for Cu2+ combines a
Wurster-type amine, namely N,N-dimethylaminophenylenedi-
† Universitat de les Illes Balears.
‡ Instituto de Ciencia Molecular (ICMOL).
(1) Ku¨pper, H.; Kroneck, P. M. H. Met. Ions Biol. Syst. 2005, 44, 97.
(2) (a) Kim, B.-E.; Nevitt, T.; Thiele, D. J. Nature Chem. Biol. 2008, 4,
176. (b) Brewer, G. J. Curr. Opin. Chem. Biol. 2003, 7, 207.
(3) (a) Brewer, G. J. Chem. Res. Toxicol. 2010, 23, 319. (b) Gaggelli,
E.; Kozlowski, H.; Valensin, D.; Valensin, G. Chem. ReV. 2006, 106, 1995–
2044.
(4) For examples of chemodosimeters for Cu, see: (a) Dujols, V.; Ford,
F.; Czarnik, A. W. J. Am. Chem. Soc. 1997, 119, 7386. (b) Qi, X.; Jun,
E. J.; Xu, L.; Kim, S. J.; Hong, J. S. J.; Yoon, Y. J.; Yoon, J. J. Org.
Chem. 2006, 71, 2881. (c) Kim, Y. R.; Kim, H. J.; Kim, J. S.; Kim, H.
AdV. Mater. 2008, 20, 4428. (d) Xiang, Y.; Tong, A. Luminescence. 2008,
23, 28.
10.1021/ol101514y  2010 American Chemical Society
Published on Web 08/06/2010

amine (DMPD), as redox reporter with a squarate ligand for
Cu2+. Related Wurster-based derivatives have been described
as redox sensors for alkaline cations6 and for anions.7
The preparation of the sodium salt of amidosquaric acid
2 is outlined in Figure 1. Although structurally very simple,
acid with DMPD in refluxing water, followed by treatment
with aqueous NaOH. Compounds 3-5 were prepared for
comparison purposes by standard methods (see the Support-
ing Information).
The structure of this amidosquaric acid has been estab-
lished unambigously by X-ray difraction of single crystals
obtained from aqueous solutions of 1 by slow-driven
crystallization. The crystal structure confirms the zwitterionic
nature of 1 in the solid state and evidence the coplanarity of
the squarate-NH-aromatic system. Accordingly, the length
of the C1-O1 bond (1.251 Å), which is conjugated with
the aromatic NH, is larger than the lengths observed in
aliphatic secondary squaramides (Figure 1).8 The close bond
distances found for C2-O2 and C3-O3 (1.244 and 1.238
Å, respectively) also support the delocalization of the
negative charge in 1. The active reagent, the sodium salt 2,
could not be crystallized. Remarkably, 2 displays negative
solvatochromism, which is revealed by the hypsochromic
shift (∆λ ) -12 nm) of the most intense band from 330 nm
in MeCN to 318 nm in H2O. In principle, the negative
solvatochromism is a sign of a dipolar ground-state structure,
where the charge of the dipole heads decreases on excitation.
Taken together, these observations suggest the contribution
of polar and dipolar forms I-IV to the structure of 2 (Scheme
1).
Scheme 1. Resonance Structures of Amidosquarate Anion 2
Figure 1. Preparation of amidosquaric acid and ORTEP view of 1
showing its zwitterionic nature in the solid state. Some structural
information is provided: Bond distances (Å): C2-O2 ) 1.238(4);
C3-O3 ) 1.244(4); C1-O1 ) 1.251(4); C4-N1 ) 1.343(4). Two
molecules of disordered water have been omitted for clarity.
Displacement ellipsoids are drawn at the 50% probability level.
this reagent has not previously been described. It was
obtained in good yield via direct condensation of squaric
(5) For examples of chemosensors for Cu, see: (a) Gunnlaugsson, T.;
Leonard, J. P.; Murray, N. S. Org. Lett. 2004, 6, 1557. (b) Cheng, Y. F.;
Zhao, D. T.; Zhang, M.; Liu, Z. Q.; Zhou, Y. F.; Shu, T. M.; Li, F. Y.; Yi,
T.; Huang, C. H. Tetrahedron Lett. 2006, 47, 6413. (c) Huang, J.; Xu, Y.;
Qian, X. Dalton. Trans. 2009, 10, 1761. (d) Grandini, P.; Mancin, F.; Tecilla,
P.; Scrimin, P.; Tonelato, U. Angew. Chem., Int. Ed. 1999, 38, 3061. (e)
Kaur, S.; Kumar, S. Chem. Commun. 2002, 23, 2840. (f) Cao, Y. D.; Zheng,
Q. Y.; Chen, C. F.; Huang, Z. T. Tetrahedron Lett. 2003, 44, 4751. (g) Xu,
Z.; Qian, X.; Cui, J. Org. Lett. 2005, 7, 3029. (h) Kov´acs, J.; Ro¨dler, T.;
Mokhir, A. Angew. Chem., Int. Ed. 2006, 45, 7815. (i) Xiang, Y.; Tong,
A.; Jin, P.; Ju, Y. Org. Lett. 2006, 8, 2863. (j) Kumar, S.; Singh, P.; Kaur,
S. Tetrahedron 2007, 63, 11724. (k) Chen, Z.; Li, F.; Li, X.; Gao, Y.; Xu,
J.; Yang, H.; Zhou, Z.; Yi, T.; Huang, Ch. Chem.sEur. J. 2008, 14, 6892.
(l) Li, L.; Ji, S.; Liu, Y. Chin. J. Chem. 2008, 26, 979. (m) Shang, L.;
Dong, S. J. Mater. Chem. 2008, 18, 4636. (n) Varazo, K.; Xie, F.; Gulledge,
D.; Wang, Q. Tetrahedron Lett. 2008, 49, 5293. (o) Frigoli, M.; Ouadahi,
K.; Larpent, C. Chem.sEur. J. 2009, 15, 8319. (p) Ajayakumar, G.;
Sreenath, K.; Gopidas, K. R. Dalton. Trans. 2009, 1180. (q) Zeng, X.; Dong,
L.; Wu, C.; Mu, L.; Xue, S. F.; Tao, Z. Sens. Acta B 2009, 506. (r) Reynal,
A.; Etxebarria, J.; Nieto, N.; Serres, S.; Palomares, E.; Vidal-Ferran, A.
Eur. J. Inorg. Chem. 2010, 1360. (s) Xu, Z.; Pan, J.; Spring, D. R.; Cui, J.;
Yoon, J. Tetrahedron 2010, 66, 1678. (t) Wang, W.; Fu, A.; You, J.; Gao,
G.; Lan, J.; Chen, L. Tetrahedron 2010, 66, 3695. (u) Rodr´ıguez-Morgade,
M.; Planells, T.; Torres, T.; Ballester, P.; Palomares, E. J. Mat. Chem. 2008,
18, 176.
Compound 1, formally an amino acid, should have two
pKa’s. A potentiometric titration afforded a pKa of 5.21.
However, the protonation at pH < 2 could not be accurately
studied by potentiometry. NMR titrations do not suffer from
these limitations and revealed two protonation steps at 1.15
and 4.99 in water (Figures S1 and S2, Supporting Informa-
tion). These results confirm the strong acidic nature of
amidosquaric acid 1.9 Remarkably, the two pKa’s are,
roughly, 1-2 units lower than those reported for dialky-
laminoanilines related to DMPD at around 2.8 and 6.3,
respectively.10 As a consequence, the redox active monoan-
ion 2 is the prevalent species in aqueous solutions at pH
around 7 used in this study.
(8) (a) Rotger, M. C.; Pin˜a, M. N.; Frontera, A.; Martorell, G.; Ballester,
P.; Deya`, P. M.; Costa, A. J. Org. Chem. 2004, 69, 2302. (b) Rotger, C.;
Soberats, B.; Quin˜onero, D.; Frontera, A.; Ballester, P.; Benet-Buchholz,
J.; Deya`, P. M.; Costa, A. Eur. J. Org. Chem. 2008, 11, 1864.
(6) De Backer, M.; Hureau, M.; Depriester, M.; Deletoille, A.; Sargent,
A. L.; Forshee, P. B.; Sibert, J. W. J. Electroanal. Chem. 2008, 612, 97.
(7) Clare, J. P.; Statnikov, A.; Lynch, V.; Sargent, A. L.; Sibert, J. W.
J. Org. Chem. 2009, 74, 6637.
(9) Neuse, E. W.; Green, B. R. J. Org. Chem. 1974, 39, 3881.
Org. Lett., Vol. 12, No. 17, 2010
3841

The UV-vis spectra of 2 recorded in MeCN-H2O (1:1
v/v) at room temperature is very different from that of DMPD
in the same solvent mixture. The spectrum of 2 exhibits an
absorption band at λmax 325 nm which is related to the
presence of the amidosquarate moiety (Figure 2).11 When a
show partial recovery of the starting reagent 2 as evidenced
by the reappearance of a purple coloration upon addition of
additional solid CuCl2 to the above solution. It is interesting
to observe that the new band forms irrespective of the
-
counterion of the copper salts used (SO42-, AcO-, ClO4 ,
-
NO3 , Cl-) and that the addition of 8-hydroxyquinoleine,
an effective Cu2+ chelating agent, inhibited the whole
process.
All these observations support the involvement of an initial
adduct between amidosquarate anion 2 and solvated Cu2+,
which then undergoes metal-chelation-controlled electron
transfer with formation of a highly colored zwitterionic
radical 2a. This view is consistent with previous experimental
work on the oxidation of DMPD with several oxidants13 and
also of related Wurster-type amines by Cu(ClO4)2 leading
to the formation of cation radicals in MeCN but not in H2O
or MeOH, due to mistmatch of redox potentials in these
solvents.14
The zwitterionic radical can evolve by two routes (Scheme
2). One by degradative radical coupling that leads to
Figure 2
.
UV/vis spectra of a solution of 2 (5.0 × 10-5 M) in
Scheme 2. Proposed Formation and Degradation Pathways of
Zwitterionic Radical Obtained from 2 [Cu2+(solv)] Stands for
MeCN-H2O (1:1 v/v) (red line) and spectral changes observed upon
addition of 4 equiv of CuCl2 (blue line). The corresponding spectra
of DMPD before (orange dotted line) and after (green dotted line)
the addition of Cu2+, are also shown for comparison.
Cu2+ Solvates of Variable Composition [Cu(H2O)n(MeCN)6-n
]
2+
with n ) 1 - 6, Assuming that CuCl2 Is Fully Ionized15
solution of 2 (5 × 10-5 M) at pH 6.8-7.0 (Tris buffer) is
treated with a solution containing excess Cu2+ (4 equiv),
changes in the UV-vis spectrum are apparent within
seconds. The initially colorless solution becomes deep purple
due to the appearance of a very strong absorption band
around 536 nm (Figure 2). This abrupt change does not occur
when the amidosquarate anion is blocked as in the ami-
dosquarate ester 4 or in absence of the redox portion as in
model compounds 3 and 5. Remarkably, a solution of DMPD
in the same solvent mixture gives comparatively weak bands
at 515 and 553 nm, respectively (Figure 2).
The absorption band around 536 nm is solvent- and time-
dependent. Therefore, while addition of excess Cu2+(4 equiv)
to 2 (5.0 × 10-5 M) in pure MeCN triggers an intense band
at 573 nm with concomitant disappearance of the ami-
dosquarate absorption at 330 nm, in water or MeOH the
solution remains almost colorless with changes in the UV
absorption region of 2 which are assigned to complexation
of Cu2+ to 2.12 On the other hand, the intensity of the long
wavelength band in MeCN or MeCN-H2O mixtures decays
with time until it reaches a minimum value after 2 h (Figure
S3, Supporting Information). In this case, the UV-vis spectra
polymerization. The other by disproportionation16 and sub-
sequent hydrolysis to p-benzoquinone.17 As a matter of fact,
benzoquinone is the main product isolated after hydrolyis
and workup of crude mixtures of 2 with Cu2+.
(10) Ernst, S.; Baltruschat, H.; Ho¨nes, J.; Lungu, M. Electrochim. Acta
2009, 54, 3129.
(11) Tietze, L. F.; Schroeter, C.; Gabius, S.; Brinck, U.; Goerlach-Graw,
A.; Gabius, H. J. Bioconjugate Chem. 1991, 2, 148.
(13) Fogliano, V.; Verde, V.; Randazzo, G.; Ritieni, A. J. Agric. Food
Chem. 1999, 47, 1035.
(12) The formation of CuL2 (log ꢀ≈ 6) have been detected by
potentiometric as well as spectrophotometric titrations; see the Supporting
Information, Table S1b. See also: Piggot, P. M. T.; Hall, L. A.; White,
A. J. P.; Williams, D. J.; Thompsons, L. K. Inorg. Chem. 2004, 43, 1167.
(14) Sumalekshmy, S.; Gopidas, K. R. Chem. Phys. Lett. 2005, 413,
294.
(15) Khan, M. A.; Schwing-Weill, M. J. Inorg. Chem. 1976, 15, 2202.
3842
Org. Lett., Vol. 12, No. 17, 2010

Since the redox chromogenic event is easily visible in
H2O-MeCN mixtures, the sensitivity and selectivity of
compound 2 for Cu2+ was evaluated by mixing aqueous
samples of Cu2+ with 2 (1.0 × 10-4 M) dissolved in MeCN
and diluting with water to a constant 1:1 v/v mixture of both
solvents (Figure 3). Under these conditions, the lower limit
for detection of Cu2+ is 1.0 × 10-6 M and the upper lineal
limit is 1.0 equiv (1.0 × 10-4 M).
The chromogenic response of 2 upon exposure of 2 (5.0
× 10-5 M, Tris buffer, pH 7) to different metals (2.0 × 10-4
M) are depicted in Figure 4. Remarkably, at neutral pH,
Figure 3. Absorption spectra of 2 upon addition of Cu2+ (CuCl2)
until a final concentration of 1.0, 2.5, 5.0, 7.5, 10.0, 25.0, 50.0,
75.0, and 100.0 × 10-6 M, from bottom to top, respectively. All
spectra are taken 2 min after mixing the two components. (Inset)
Lineal response of 2 (1.0 × 10-4 M, Tris buffer, pH 7) with
Figure 4
.
Visual response of 2 (5.0 × 10-5 M, Tris buffer, pH
6.9) upon adddition of different metal ions: Ca2+, Mg2+, Cr3+, Mn2+
,
increasing concentration of Cu2+
.
Fe3+, Co2+, Ni2+, Cu2+, Zn2+, Cd2+, Hg2+, Al3+ (2.0 × 10-4 M).
The picture was taken 2 min after the addition of Cu2+
.
To check the formation of a radical, X-band EPR spectra
of solutions containing anion 2 and Cu2+ were recorded in
acetonitrile at -79 °C (Figure S4, Supporting Information).
The initial EPR spectrum shows a signal with g∞ ) 2.00
and g| ) 2.09 that can be attributed to a Cu2+ ion with axial
geometry. This signal elapses with time appearing a more
isotropic new signal with g ) 2.01 that can be unambigously
ascribed to a radical species.
Further evidence obtained by electrospray ionization (ESI-
MS) supports a redox reaction of Cu2+.18 In the presence of
CuCl2, formation of different mono- but also of dinuclear
complexes of copper was evident in the mass spectra of
solutions of CuCl2 with 2 in H2O, MeCN, or mixtures of
both solvents (Figure S4, Supporting Information). Com-
parison of the measured isotope pattern with the calculated
ones supports the composition of the base peak at m/z 355.16,
which is assigned to [Cu2d(H2O)5]+ and two prominent
fragments at m/z 253.20 and 255.23 to [2b + Na]+ and [2c
+ Na]+, respectively, among others.
compound 2 displays very good selectivity for Cu2+ over
other potentially competing metals. The pale yellowish
tonality observed for Fe3+ is the natural color of a 20 µM
solution of FeCl3·6H2O in MeCN-H2O (50:50 v/v) used in
the above experiment (Figure 4).
In summary, we set out to design a minimalist sensor for
Cu2+ that works on the principle of chelation-induced free-
radical formation. The chromogenic and selective response
of compound 2 at neutral pH is very well suited for the redox
chemodosimetric evaluation of nonorganically bound Cu2+
in natural waters.
Acknowledgment. This work is supported by a MEC
grant (ref CTQ2008-00841/BQU) and a CAIB grant (ref
PCTIB-2005GC3-08) contract. We thank Dr. F. Lloret for
his help with the EPR measurements.
Supporting Information Available: Full experimental
details and X-ray crystallographic data (CIF) and additional
spectral data. This material is available free of charge via
the Internet at http://pubs.acs.org.
(16) Nickel, U.; Peris, C. V.; Ramminger, U. J. Phys. Chem. A 2002,
106, 3773.
(17) Modestov, A. D.; Gun, J.; Savotine, I.; Lev, O. J. Electroanal.
Chem. 2004, 565, 7.
(18) Tintaru, A.; Charles, L.; Milko, P.; Roithova´, J.; Schro¨der, D. J.
Phys. Org. Chem. 2008, 22, 229.
OL101514Y
Org. Lett., Vol. 12, No. 17, 2010
3843