New fluorescent rhodamine hydrazone chemosensor for Cu(II) with high selectivity and sensitivity

Article abstract of DOI:10.1021/ol0610340

A new fluorescent probe, salicylaldehyde rhodamine B hydrazone (1), was synthesized and displayed selective Cu(II)-amplified absorbance and fluorescence emission above 500 nm in neutral buffered media. Upon the addition of Cu(II), the spirolactam ring of

Full text of DOI:10.1021/ol0610340

ORGANIC
LETTERS
2006
Vol. 8, No. 13
2863-2866
New Fluorescent Rhodamine Hydrazone
Chemosensor for Cu(II) with High
Selectivity and Sensitivity
Yu Xiang, Aijun Tong,* Peiyuan Jin, and Yong Ju
Department of Chemistry, Tsinghua UniVersity, Beijing 100084, P. R. China
tongaj@mail.tsinghua.edu.cn
Received April 28, 2006
ABSTRACT
A new fluorescent probe, salicylaldehyde rhodamine B hydrazone (1), was synthesized and displayed selective Cu(II)-amplified absorbance
and fluorescence emission above 500 nm in neutral buffered media. Upon the addition of Cu(II), the spirolactam ring of 1 was opened and a
1:1 metal−ligand complex was formed. The detection of Cu(II) by 1 at a lower micromolar level was successful even in buffered water.
The development of artificial receptors for the sensing and
recognition of environmentally and biologically important
ionic species, especially transition-metal ions, is currently
of great interest.1 Because copper is a widely used pollutant
and an essential trace element in biological systems, much
attention has been drawn to the design of fluorescent probes
for the detection of copper ions due to the nondestructive,
quick, and sensitive advantages of emission signals.2 Most
of the classic and early-reported cation sensors, however,
generally undergo fluorescence quenching upon the binding
of Cu(II),3 which is not as sensitive as a fluorescence
enhancement response; moreover, the selectivity for Cu(II)
over other ions, such as Fe(III) and Pb(II), is not very
satisfactory for some of the probes. To overcome these
disadvantages, fluoroionophores which show a selective
response to Cu(II) by a copper-amplified fluorescence
emission have been well developed in recent years.4 Among
these sensors, unfortunately, those that can be applied in
aqueous solutions at neutral pH are still rare mainly because
of the strong hydration ability of Cu(II) in water. Actually,
sensors of this kind are always considered to be much more
attractive and efficient in the respect that most copper-
containing samples are near-neutral aqueous systems.1,2
On the other hand, fluorophores of long-wavelength
emissions are often preferred to serve as the reporting group
for analyte to avoid the influence of background fluorescence
(1) For reviews, see: (a) 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, 1515. (b) Valeur, B.; Leray, I. Coord.
Chem. ReV. 2000, 205, 3. (c) Rurack, K. Spectrochim. Acta 2001, 57A,
2161. (d) Amendola, V.; Fabbrizzi, L.; Forti, F.; Licchelli, M.; Mangano,
C.; Pallavicini, P.; Poggi, A.; Sacchi, D.; Taglieti, A. Coord. Chem. ReV.
2006, 250, 273.
(4) (a) Wu, Q.; Anslyn, E. V. J. Am. Chem. Soc. 2004, 126, 14682. (b)
Gunnlaugsson, T.; Leonard, J. P.; Murray, N. S. Org. Lett. 2004, 6, 1557.
(c) Mokhir, A.; Kra¨mer, R. Chem. Commun. 2005, 2244. (d) Royzen, M.;
Dai, Z.; Canary, J. W. J. Am. Chem. Soc. 2005, 127, 1612. (e) Zheng, L.;
Miller, E. W.; Pralle, A.; Isacoff, E. Y.; Chang, C. J. J. Am. Chem. Soc.
2006, 128, 10.
(2) Kra¨mer, R. Angew. Chem., Int. Ed. 1998, 37, 772.
(3) (a) Breuer, W.; Epsztejn, S.; Millgram, P.; Cabantchik, I. Z. Am. J.
Physiol. 1995, 268, C1354. (b) McCall, K. A.; Fierke, C. A. Anal. Biochem.
2000, 284, 307. (c) Chavez-Crooker, P.; Garrido, N.; Ahearn, G. A. J. Exp.
Biol. 2001, 204, 1433.
10.1021/ol0610340 CCC: $33.50
© 2006 American Chemical Society
Published on Web 06/03/2006

(<400 nm). The introduction of the rhodamine moiety to
construct probes of the “off-on” type is a reliable method
because of the well-known spirolactam (fluorescence “off”)
to ring-opened amide (fluorescence “on”) equilibrium (Figure
1) of rhodamine derivatives.6 We have successfully designed
range immediately as a result of the Cu(II)-induced ring
opening of the spirolactam form. We also found that a
solution of 10 µM 1 in 50% (v/v) buffered (10 mM Tris-
HCl, pH ) 7.0) water/CH3CN could display an obvious
purple color in the presence of Cu(II) at the micromolar level,
and other ions of our interest showed little interference
(Figure 2). These results suggested that 1 could serve as a
Figure 1. Chemical structures of compound 1, 1-Cu(II),5 and 2.
a fluorescence chemosensor for Fe(III) utilizing rhodamine
B as the fluorophore in our previous work.6d Herein, we
describe a new rhodamine-based chemosensor 1 (Figure 1),
which shows a reversible, selective, and sensitive fluores-
cence enhancement response to Cu(II) in neutral buffered
media.
Figure 2. Pictures of 10 µM 1 as a selective naked-eye chemosen-
sor for Cu(II) in 50% (v/v) water/CH3CN (10 mM Tris-HCl, pH
) 7.0). Top (from left to right): 0, 2.5, 5.0, 10.0 µM Cu(II). Bottom
(from left to right): 50 µM Fe(III), Fe(II), Zn(II), Pb(II), Hg(II).
Compound 1 was facilely synthesized from rhodamine B
by a two-step reaction (Supporting Information, S-Figure 1).
This Shiff base was stable in neutral water solutions for at
least 2 days. It was designed to chelate with metal ions via
its carbonyl O, imino N, and phenol O atoms.7 The
spirolactam moiety of the rhodamine group acted as a signal
switcher, which was envisioned to turn on when the cation
was bound. A solution of 1 in Tris-HCl buffer (5 mM, pH
) 7.0) or organic media is colorless and weakly fluorescent,
indicating that the spirolactam form of 1 exists predomi-
nantly. The characteristic peak of the 9-carbon of 1 near 66
ppm in the 13C NMR spectrum (Supporting Information,
S-Figure 2d) also supports this consideration.8 Besides, no
obvious characteristic color or fluorescence of rhodamine
could be observed for 1 between pH 5.0 and 10.0 (Supporting
Information, S-Figure 3), suggesting that 1 is insensitive to
pH and that the spirolactam form is still preferred in this
range. As we expected, addition of Cu(II) to a solution of 1
in either CH3CN or water caused a significant enhancement
of absorbance and fluorescence intensity in the 500-650 nm
“naked-eye” chemosensor selective for Cu(II) in neutral
buffered media.
The absorption spectra of 1 (10 µM) in 50% (v/v) water/
CH3CN (10 mM Tris-HCl buffer, pH ) 7.0) exhibited only
a very weak band above 500 nm, which was ascribed to the
trace ring-opened form of molecules of 1. Upon the addition
of up to 5 equiv of Cu(II), the absorbance was significantly
enhanced (> 500-fold) and a new peak at 558 nm was
observed (Figure 3a), suggesting the clear formation of the
ring-opened amide form of 1. Other metal ions had little
interference. Only Fe(II) displayed a 65-fold enhancement
at the same concentration, but a much longer response time
was required (Supporting Information, S-Figure 4a-d). The
nonlinear fitting of the titration curve assumed a 1:1
stoichiometry for the 1-Cu(II) complex (Figure 1) with an
association constant Ka value of far more than 106,9 showing
the high affinity of 1 to Cu(II). This binding mode was also
supported by the data of Job’s plots10 evaluated from the
absorption spectra of 1 and Cu(II) with a total concentration
of 3.3 µM (Supporting Information, S-Figure 5). A more
direct evidence was obtained by comparing the ESI mass
spectra of 1 and 1-Cu(II) (Figure 4). The unique peak at
m/z ) 622.5 (calcd ) 622.3) corresponding to [1 + Cu-
H]+ was clearly observed when 1.2 equiv of Cu(II) was
(5) The copper ion may be chelated by the counteranion or solvent
oxygen to satisfy the need for four-coordination. They are omitted for clarity.
(6) For rhodamine-based chemosensors for metal ions, see: (a) Dujols,
V.; Ford, F.; Czarnik, A. W. J. Am. Chem. Soc. 1997, 119, 7386. (b) Yang,
Y.-K.; Yook, K.-J.; Tae, J. J. Am. Chem. Soc. 2005, 127, 16760. (c) Kwon,
J. Y.; Jang, Y. J.; Lee, Y. J.; Kim, K. M.; Seo, M. S.; Nam, W.; Yoon, J.
J. Am. Chem. Soc. 2005, 127, 10107. (d) Xiang, Y.; Tong, A. Org. Lett.
2006, 8, 1549.
(7) Lee, P. F.; Yang, C.-T.; Fan, D.; Vittal, J. J.; Ranford, J. D.
Polyhedron 2003, 22, 2781.
(8) Anthoni, U.; Christophersen, C.; Nielsen, P.; Puschl, A.; Schaumburg,
K. Struct. Chem. 1995, 3, 161.
(9) (a) Connors, K. A. Binding Constants-The Measurement of Molecular
Complex Stability; John Wiley & Sons: New York, 1987; Chapter 4.
(10) Vosburgh, W. C.; Cooper, G. R. J. Am. Chem. Soc. 1941, 63, 437.
2864
Org. Lett., Vol. 8, No. 13, 2006

Figure 4. ESI mass spectra (positive) of 1 (10 µM) in the absence
and presence of Cu(II) (1.2 equiv). Right: calculated (top) and
observed (bottom) isotopic patterns for the [Cu(1-H)]+ cation,
indicating the formation of a 1:1 metal-ligand complex.
The high enhancement factor (>500) of absorbance of 1
upon the binding of Cu(II) and the large association constant
(Ka > 106) of the 1-Cu(II) complex implied the possibility
of copper sensing at a very low concentration level by 1.
Indeed, the absorbance of 1 (10 µM) at 558 nm was found
to increase linearly with the concentration of Cu(II) in the
range of 25 nM-3.3 µM (Supporting Information, S-Figure
7a-d). However, signals acquired from absorption spectra
are generally not as efficient as those evaluated from
fluorescence spectra. The fluorescence titration12 was also
carried out using 10 µM 1 in 50% (v/v) buffered water/CH3-
CN at pH 7.0. Although the emission spectra of 1 underwent
a red shift from 576 to 583 nm upon the addition of 5 equiv
of Cu(II), the fluorescence intensity at the new peak (583
nm) had only a 3.8-fold enhancement (Figure 3b), which
was much weaker compared to that of absorbance. Neverthe-
less, the selective sensing of Cu(II) at the 100 nM level was
still available when using 1.0 µM 1 under the same
conditions (Supporting Information, S-Figure 8).
To get a practical application view, the fluorescence
sensing behavior of 1 for Cu(II) in buffered water solution
(with no more than 2% CH3CN) was also investigated. The
titration was performed using 10 µM solutions of 1 in Tris-
HCl (5 mM, pH ) 7.0) aqueous buffer (Figure 5). A 9.4-
fold enhancement of fluorescence and a continuous red shift
of the emission peak from 573 to 585 nm were observed
upon the addition of 20 equiv of Cu(II) compared to that of
only 1 in solution. The nonlinear fitting of the titration curve
and the data of Job’s plots from absorption spectra (Sup-
porting Information, S-Figures 9 and 10) also assumed a 1:1
stoichiometry for the 1-Cu(II) complex with an association
constant around Ka ) 69 110. The response was proved to
be revisable, and the selectivity of 1 to Cu(II) over other
cations was still satisfactory. We found that alkali and
alkaline earth metal ions had hardly any effect on the
Figure 3. (a) Absorption spectra of 1 (10 µM) in Tris-HCl (10
mM, pH ) 7.0) buffer containing 50% (v/v) water/CH3CN in the
presence of different amounts of Cu(II). Inset: absorbance at 558
nm as a function of Cu(II) concentration, indicating a 1:1 metal-
ligand ratio. (b) Fluorescence spectra of 1 (10 µM) under the same
conditions. Excitation was performed at 520 nm. Inset: fluorescence
enhancement factor (F - F0)/F0 at 583 nm as a function of Cu(II)
concentration.
added to 1, whereas 1 without Cu(II) exhibited peaks only
at m/z ) 561.4 (calcd ) 561.3) and 583.3 (calcd ) 583.3)
which corresponded to [1 + H]+ and [1 + Na]+, respectively.
To achieve this 1:1 stoichiometry, carbonyl O, imino N, and
phenol O atoms of 1 are the most possible binding sites for
Cu(II).7 In fact, the phenol group of 1 which undergoes
deprotonation during the binding plays an indispensable role
in the affinity of 1 to Cu(II) because the contrast compound
2 (no phenol group compared with 1, Figure 1) displayed
nearly no response to Cu(II) in either absorption or ESI mass
spectra when treated with excess (10 equiv) Cu(II) (Sup-
porting Information, S-Figure 6a-c). It was also confirmed
that the response of 1 to Cu(II) was reversible rather than a
cation-catalyzed reaction:11 (i) the color and fluorescence of
1-Cu(II) disappeared instantly upon the addition of EDTA,
whereas excess Cu(II) would recover the signal; (ii) the TLC
results and the mass spectra data of a buffered solution
containing 10 µM 1 and 12 µM Cu(II) after standing at room
temperature for 2 days indicated no other chemical species
except for 1 and 1-Cu(II).
(12) All the fluoresence spectra were measured by excitation at 520 nm
to obtain a full view of emissions from 530 to 700 nm.
(11) For sensors of this type, see ref 6a,b.
Org. Lett., Vol. 8, No. 13, 2006
2865

absorption and fluorescence signals.5c,d In the case of 1,
binding of Cu(II) does open the spirolactam ring, but at the
same time, the fluorescence of the ring-opened amide form
is probably partially quenched by Cu(II). The quenching
mechanism may be similar to that of some Cu(II) probes
displaying fluorescence quenching for the paramagnetic
nature of the copper ion.1,13 From a sensitivity viewpoint, it
is preferable to inhibit this quenching effect to generate a
more notable fluorescence enhancement. Chemical modifica-
tion on the phenol group may be a good improvement,14 but
the affinity of 1 to Cu(II) could probably decrease for the
important role of the phenol moiety. Works related to this
topic are now under our investigation.
In conclusion, we have synthesized a new rhodamine-
based fluoroionophore 1, which displayed a reversible
absorption and fluorescence enhancement response to Cu(II)
via a 1:1 binding mode. Its selectivity toward Cu(II) is very
high because little interference was observed for other
commonly coexistent metal ions. Furthermore, the sensitivity
of 1 for Cu(II) can be lower than 25 nM in 50% (v/v)
buffered water/CH3CN by the absorption spectra method (still
at the 0.1 µM level for the fluorescence method under this
condition). Even in neutral buffered aqueous solutions, the
fluorescent sensing of Cu(II) at a lower micromolar level
was successful.
Figure 5. Fluorescence spectra of 1 (10 µM) in Tris-HCl (5 mM,
pH ) 7.0) aqueous buffer solutions in the presence of different
amounts of Cu(II). Excitation was performed at 520 nm. Inset:
fluorescence enhancement factor at 585 nm as a function of Cu(II)
concentration.
fluorescence spectra of 1 (10 µM) in buffered water even at
the millimolar level. Other transition-metal ions, such as
Fe(III), Co(II), Ni(II), Zn(II), Mn(II), Ag(I), Cd(II), and
Hg(II), gave a much weaker response compared to Cu(II) at
the same concentration (100 µM) (Supporting Information,
S-Figure 11), and the fluorescence signal of 1-Cu(II) in the
presence or absence of these contrast ions also exhibited only
a mild difference (Supporting Information, S-Figure 12).
Furthermore, the detection of Cu(II) from 0.25 to 2.0 µM
was successful when utilizing 1.0 µM 1 in 1 mM Tris-HCl
aqueous buffers at pH 7.0 (Supporting Information, S-Figure
13), indicating the high sensitivity of 1 for Cu(II) even in
aqueous media.
Acknowledgment. The authors are grateful for the
financial support from the National Natural Science Founda-
tion of China, No. 20375021.
Supporting Information Available: Experimental pro-
cedures, characterization data for the compounds described,
and selected spectroscopic data of 1 and 2. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL0610340
With careful investigation, one may note that when Cu(II)
was added to 1, the enhancement of absorbance was much
more significant than that of fluorescence intensity. However,
the ring opening of the spirolactam form of rhodamine
derivatives generally results in comparable amplifications of
(13) (a) Varnes, A. W.; Dodson, R. B.; Wehry, E. L. J. Am. Chem. Soc.
1972, 94, 946. (b) Torrado, A.; Walkup, G. K.; Imperiali, B. J. Am. Chem.
Soc. 1998, 120, 609. (c) Li, Y.; Yang, C. M. Chem. Commun. 2003, 2884.
(14) Wen, Z.-C.; Yang, R.; He, H.; Jiang, Y.-B. Chem. Commun. 2006,
106.
2866
Org. Lett., Vol. 8, No. 13, 2006