Copper ion-selective fluorescent sensor based on the inner filter effect using a spiropyran derivative

Article abstract of DOI:10.1021/ac051010r

A highly selective copper(II) ion fluorescent sensor has been designed based on the UV-visible absorption of a spiropyran derivative coupled with the use of a metal porphyrin operative on the fluorescence inner filter effect. Spiropyrans, which combine the characteristics of metal binding and signal transduction, have been widely utilized in cationic ion recognition by UV-visible spectroscopy. In the present work, the viability of converting the absorption signal of the spiropyran molecule into a fluorescence signal was explored. On account of overlap of the absorption band of the spiropyran (λ_abs = 547 nm) in the presence of copper ion with the Q-band of an added fluorophore, zinc meso-tetraphenylporphyrin (λ_abs = 556 nm), the effective light absorbed by the porphyrin and concomitantly the emitted light intensity vary as a result of varying absorption of the spiropyran via fluorescence inner filter effect. The metal binding characteristic of the spiropyran presents an excellent selectivity for copper ion in comparison with several other heavy or transition metal ions. Since the changes in the absorbance of the absorber translate into exponential changes in fluorescence of the fluorophore, the novelty of the present device is that the analytical signal is more sensitive over that of the absorptiometry or that of the fluorometry using one single dye. To realize a practical fluorescent sensor, both the absorber and fluorophore were immobilized in a plasticized poly(vinyl chloride) membrane, and the sensing characteristics of the membrane for copper ion were investigated. The sensor is useful for measuring Cu²⁺ at concentrations ranging from 7.5 × 10^-7 to 3.6 × 10 ^-5 M with a detection limit of 1.5 × 10^-7 M. The sensor is chemically reversible, the fluorescence was switched off by immersing the membrane in copper ion solution and switched on by washing it with EDTA solution.

Full text of DOI:10.1021/ac051010r

Anal. Chem. 2005, 77, 7294-7303
Copper Ion-Selective Fluorescent Sensor Based on
the Inner Filter Effect Using a Spiropyran
Derivative
Na Shao,† Ying Zhang,† SinMan Cheung,‡ RongHua Yang,*,† WingHong Chan,*,‡ Tain Mo,‡
KeAn Li,† and Feng Liu†
College of Chemistry and Molecular Engineering, Peking University, Beijing, 100871, China, and Department of Chemistry,
Hong Kong Baptist University, Kowloon Tong, Hong Kong, China
chemical sensors for several decades. Of the many different kinds
of optical sensors (luminescence, absorption, reflection, etc), a
fluorescence-based one is a very appealing motif for future
practical applications, thanks to its intrinsic sensitivity, its straight-
forward application to fiber optical-based detection, and its being
capable of affording high spatial resolution via microscopic
imaging.1,2 The most direct way to construct a fluorescent sensor
approach is to look for fluorescence sensory molecules instead
of absorbing ones. In general, a fluorescent sensory molecule
would involve the covalent linking of a receptor domain to a
fluorescent fragment (i.e., the signaling unit). The two components
are intramolecularly connected together such that the binding of
the target analyte causes significant changes to the photophysical
properties of the fluorescent fragment.3,4
Spiropyrans5 are an important class of photo- and thermochro-
mic compounds, which show reversible structural conversion upon
the actions of external optical, chemical, and thermal stimulation.6-9
Irradiation with ultraviolet light causes formation of an extended
π-conjugation open form (merocyanine form) by heterolytic
cleavage of the C (spiro)-O bond (Scheme 1). The merocyanines
may interact with their environment (i.e., solvent, matrix, etc.),
leading to different photochromic responses. Several years ago,
A highly selective copper(II) ion fluorescent sensor has
been designed based on the UV-visible absorption of a
spiropyran derivative coupled with the use of a metal
porphyrin operative on the fluorescence inner filter effect.
Spiropyrans, which combine the characteristics of metal
binding and signal transduction, have been widely utilized
in cationic ion recognition by UV-visible spectroscopy.
In the present work, the viability of converting the absorp-
tion signal of the spiropyran molecule into a fluorescence
signal was explored. On account of overlap of the absorp-
tion band of the spiropyran (λabs ) 547 nm) in the
presence of copper ion with the Q-band of an added
fluorophore, zinc meso-tetraphenylporphyrin (λabs ) 556
nm), the effective light absorbed by the porphyrin and
concomitantly the emitted light intensity vary as a result
of varying absorption of the spiropyran via fluorescence
inner filter effect. The metal binding characteristic of the
spiropyran presents an excellent selectivity for copper ion
in comparison with several other heavy or transition metal
ions. Since the changes in the absorbance of the absorber
translate into exponential changes in fluorescence of the
fluorophore, the novelty of the present device is that the
analytical signal is more sensitive over that of the absorp-
tiometry or that of the fluorometry using one single dye.
To realize a practical fluorescent sensor, both the absorber
and fluorophore were immobilized in a plasticized poly-
(vinyl chloride) membrane, and the sensing characteris-
tics of the membrane for copper ion were investigated.
The sensor is useful for measuring Cu2+ at concentrations
ranging from 7.5 × 10-7 to 3.6 × 10-5 M with a detection
limit of 1.5 × 10-7 M. The sensor is chemically reversible,
the fluorescence was switched off by immersing the
membrane in copper ion solution and switched on by
washing it with EDTA solution.
(1) For reviews, see: (a) Fluorescent Chemosensors for Ion and Molecule
Recognition; Czarnik, A. W., Ed.; ACS Symposium Series 538; American
Chemical Society: Washington, DC, 1992. (b) Chemosensors of Ion and
Molecule Recognition; Desvergne, J.-P., Czarnik, A. W., Eds.; NATO ASI
Series C Vol. 492; Kluwer Academic Press: Dordrecht, The Netherlands,
1997. (c) Wolfbeis, O. S.; Bohmer, M.; Durkop, A.; Enderlein, L.; Gruber,
M.; Klimant, I.; Krause, C.; Kurner, J.; Liebsch, G.; Lin, Z.; Oswald, B.; Wu,
M. In Fluorescence Spectroscopy, Imaging and Probes: New Tools in Chemical,
Physical and Life Sciences; Kraayenhof, R., Visser, A. J. W. G., Gerritsen, H.
C., Eds.; Springer: Berlin, 2002. (d) Handbook of Fluorescent Probes and
Research Chemicals, 9th ed.; Haugland, R. P. Ed.; Molecular Probes: Eugene,
OR, 2002.
(2) The biennial reviews on Molecular Fluorescence, Phosphorescence, and
Chemiluminescence Spectrometry and Fiber-Optic Chemical Sensors and
Biosensors; see, for examples: (a) Powe, A. M.; Fletcher, K. A.; St. Luce,
N. N.; Lowry, M.; Neal, S.; McCarroll, M. E.; Oldham, P. B.; McGown, L.
B.; Warner, I. M. Anal. Chem. 2004, 76, 4614-4634. (b) Wolfbeis, O. S.
Anal. Chem. 2004, 76, 3269-3283.
(3) For reviews of fluorescent molecule probes, see: (a) de Silva, A. P.; Nimal
Gunaratne, H. Q.; Gunnlaugsson, T.; Huxley, A. J. M.; McCoy, C. P.;
Rademacher, J. T.; Rice, T. E. Chem. Rev. 1997, 97, 1515-1566. (b) Valeur,
B.; Leray, I. Coord. Chem. Rev. 2000, 205, 3-40. (c) Prodi, L.; Bolletta, F.;
Montalti, M. Zaccheroni, N. Coord. Chem. Rev. 2000, 205, 59-83. (d) Beer,
P. D.; Gale, P. A. Angew. Chem., Int. Ed. 2001, 40, 487-516. (e) Pu, L.
Chem. Rev. 2004, 104, 1687-1716. (f) Gokel, G. W.; Leevy, W. M.; Weber,
M. E. Chem. Rev. 2004, 104, 2723-2750.
The development of optical sensing approaches for the detec-
tion of environmentally and biologically important species, such
as heavy metal ions, has been an important goal in the field of
* To whom correspondence should be addressed. E-mail: Yangrh@pku.edu.cn;
Whchan@hkbu.edu.hk. Fax: +86-10-62751708.
† Peking University.
‡ Hong Kong Baptist University.
7294 Analytical Chemistry, Vol. 77, No. 22, November 15, 2005
10.1021/ac051010r CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/13/2005

Scheme 1. Photoreversible Equilibria and Metal
Complexation of Spiropyran Derivates with
Divalent Metal Ions
reported so far have been mainly confined to UV-visible absorp-
tion spectroscopy.14-30 Less attention has been paid to lumines-
cence spectroscopy due to the common susceptibility of lower
quantum yields of such dyes,12,13 although the “signal communica-
tions” of absorption and fluorescence between a spiropyran
nucleus and a fluorophore have been studied based on intermo-
lecular energy transfer or photoinduced proton transfer.31,32
Another alternative way to convert absorbance signal into
fluorescence signal without covalent linking of a chromotropic
receptor and a fluorophore can be exploited by fluorescence inner
filter effects (IFEs). In the approach, two dyes are employed, one
absorbent, the other fluorescent. If the absorption spectrum of
the absorbing dye possesses a complementary overlap region with
the excitation or emission spectrum of the fluorescent dye, the
fluorescence emission of the fluorophore is thus modulated by
the absorber.33-36 The IFE is a source of errors in fluorometry,
but it can be useful for an optical chemical sensor by converting
the analytical absorption signals into fluorescence signals.37-40
These sensors did not require the establishing of any covalent
linking between the receptor and a fluorophore but utilize the
fluorophore and the receptor as such. Moreover, since the
changes in the absorbance of the absorber translate into expo-
nential changes in fluorescence of the fluorophore, an enhanced
sensitivity and decreased detection limits for the analytical method
is reasonable with respect to the absorbance values alone.37,38
In our preliminary experiments, we observed that the UV-
visible absorption property of a newly synthesized spiropyran
derivative, 1 (Chart 1), was modulated by copper ion. The metal-
free state of 1 is colorless and exhibits no absorption in the visible
Taylor and Phillips independently demonstrated that the negatively
charged phenolic oxygen atom in the merocyanine form of a
spiropyran nucleus could bind to a metal center in cooperation
with other chelating function attached at the 8′-position (Scheme
1).10,11 During the past decade, a number of such interesting
compounds have been synthesized for divalent metal ion
recognition.12-21 More sophisticated systems, with multiple metal
binding sites in the spiropyran-crown (or cryptand) ether
conjugates have also been developed for alkali metal ion
recognition.22-30 However, most of such metal binding events
(4) For selected examples of cationic ion fluorescent probes, see: (a) Nakatsuji,
Y.; Kita, K.; Inoue, H.; Zhang, W.; Kida, T.; Ikeda, I. J. Am. Chem. Soc. 2000,
122, 6307-6308. (b) Watanabe, S.; Ikishima, S.; Matsuo, T.; Yoshida, K. J.
Am. Chem. Soc. 2001, 123, 8402-8403. (c) Walkup, G. K.; Burdette, S. C.;
Lippard, S. J.; Tsien, R. Y. J. Am. Chem. Soc. 2000, 122, 5644-5645. (d)
Chen, C. T.; Huang, W. P. J. Am. Chem. Soc. 2002, 124, 6246-6247. (e)
Yang, R. H.; Chan, W. H.; Lee, A. W. M.; Xia, P. F.; Zhang, H. K.; Li, K. A.
J. Am. Chem. Soc. 2003, 125, 2884-2885. (f) Ono, A.; Togashi, H. Angew.
Chem., Int. Ed. 2004, 43, 4300-4302. (g) Huang, H. M.; Wang, K. M.;
Tan. W. H.; An. A. L.; Huang, S. S,; Zhai, Q.; Zhou, L. J.; Jin, Y. Angew.
Chem., Int. Ed. 2004, 43, 5635-5638. (h) Arunkmar, E.; Ajayaghosh, A.;
Daub, J. J. Am. Chem. Soc. 2005, 122, 3156-3164.
(22) Inouye, M.; Ueno, M.; Kitao, T.; Tsuchiya, K. J. Am. Chem. Soc. 1990, 112,
8977-8979.
(23) Kimura, K.; Yamashita, T.; Yokoyama, M. J. Chem. Soc., Chem. Commun.
1991, 147-148.
(24) Kimura, K.; Kaneshige, M.; Yamashita, T.; Yokoyama, M. J. Org. Chem.
1994, 59, 1251-1256.
(25) Inouye, M.; Noguchi, Y.; Isagawa, K. Angew. Chem., Int. Ed. Engl. 1994,
33, 1163-1166.
(5) The IUPAC nomenclature of spiropyran is 1′,3′-dihydro-1′,3′,3′-trimethyl-
spiro-[2H-1-benzopyran-2, 2′-indoline]
(26) Kimura, K.; Utsumi, T.; Teranishi, T.; Yokoyama, M.; Sakamoto, H.;
Okamoto, M.; Arakawa, R.; Moriguchi, H.; Miyaji, Y. Angew. Chem., Int.
Ed. Engl. 1997, 36, 2452-2454.
(6) Fischer, E.; Hirshberg, Y. J. Chem. Soc. 1952, 4522-4524.
(7) Bertelson, R. C. In Photochromism; Brown, G. H.; Ed.; Wiley-Interscience:
New York, 1971; pp 45-431, and references therein.
(27) Filley, J.; Ibrahim, M. A.; Nimlos, M. R.; Watt, A. S.; Blake, D. M. J.
(8) Guglielmetti, R. In Photochromoism: Molecules and Systems, Studies in
Organic Chemistry; Du¨rr, H., Bouas-Laurent, H. Eds.; Elsevier: Amsterdam,
1990; Chapter 8 and 23, and references therein.
(9) ] Berkovic, G.; Krongauz, V.; Weiss, V. Chem. Rev. 2000, 100, 1741-1753.
(10) Phillips, J.; Mueller, A.; Przystal, F. J. Am. Chem. Soc. 1965, 87, 4020.
(11) Taylor, L. D.; Nicholson, J.; Davis, R. B. Tetrahedron Lett. 1967, 8, 1585-
1588.
(12) Preigh, M. J.; Lin, F. T.; Ismail, K. Z.; Weber, S. G. J. Chem. Soc., Chem.
Commun. 1995, 2091-2092.
(13) Winkler, J. D.; Bowen, C. M.; Michelet, V. J. Am. Chem. Soc. 1998, 120,
3237-3242.
(14) Chibisov, A. K.; Gorner, H. Chem. Phys. 1998, 237, 425-442.
(15) Gorner, H.; Chibisov, A. K. J. Chem. Soc., Faraday Trans. 1998, 94, 2557-
2564.
(16) Evans, L.; Collins, G. E.; Shaffer, R. E.; Michelet, V.; Winkler, J. D. Anal.
Chem. 1999, 71, 5322-5327.
Photochem. Photobiol. A 1998, 117, 193-198.
(28) Salhin, A. M. A.; Tanaka, M.; Kamada, K.; Ando, H.; Ikeda, T.; Shibutani,
Y.; Yajima, S.; Nakamura, M.; Kimura, K. Eur. J. Org. Chem. 2002, 655-
662.
(29) Ahmed, S. A.; Tanaka, M.; Ando, H.; Iwamoto, H.; Kimura.; K, Eur. J. Org.
Chem. 2003, 2437-2442.
(30) Sakamoto, H.; Takagaki, H.; Nakamura, M.; Kimura, K. Anal. Chem. 2005,
77, 1999-2006.
(31) (a) Bahr, J. L.; Kodis, G.; dp la Garza, L.; Lin, S.; Moore, A. L.; Moore, T.
A.; Gust, D. J. Am. Chem. Soc. 2001, 123, 7124-7133. (b) Guo, X. F.; Zhang,
D. Q.; Zhou, Y. C.; Zhu, D. B. J. Org. Chem. 2003, 68, 5681-5687.
(32) (a) Raymo, F. M.; Giordani, S. J. Am. Chem. Soc. 2002, 124, 2004-2007.
(b) Raymo, F. M.; Giordani, S. Org. Lett. 2001, 3, 3475-3478. (c) Raymo,
F. M.; Alvarado, R. J.; Giordani, S.; Cejas, M. A. J. Am. Chem. Soc. 2003,
125, 2361-2364. (d) Giordani, S.; Cejas, M. A.; Raymo, F. M. Tetrahedron
2004, 60, 10973-10981.
(17) Collins, G. E.; Choi, L. S.; Ewing, K. J.; Michelet, V.; Bowen, C. M.; Winkler,
(33) Parker, C. A.; Barnes, W. J. Analyst 1957, 82, 606-617.
(34) Parker, C. A.; Tees, W. T. Analyst 1962, 87, 83-111.
(35) Holland, J. F.; Teets, R. E.; Kelly, P. M.; Timnick, A. Anal. Chem. 1977,
49, 706-710.
J. D. Chem. Commun. 1999, 321-322.
(18) Nishikiori, H.; Sasai, R.; Arai, N.; Arai, N.; Takagi, K. Chem. Lett. 2000,
1142-1143.
(19) Leaustic, A.; Dupont, A.; Yu, P.; Clement, R. N. J. Chem. 2001, 25, 1297-
1301.
(20) Wojtyk, J. T. C.; Kazmaier, P. M.; Buncel, E. Chem. Mater. 2001, 13, 2547-
2551.
(21) Kopelman, R. A.; Snyder, S. M.; Frank, N. L. J. Am. Chem. Soc. 2003, 125,
13684-13685.
(36) Leese, R.; Wehry, E. L. Anal. Chem. 1978, 50, 1193-1197.
(37) Yuan, P.; Walt, D. R. Anal. Chem. 1987, 59, 2391-2394.
(38) Gabor, G.; Walt, D. R. Anal. Chem. 1991, 63, 793-796.
(39) He, H. R.; Li, H.; Mohr, G.; Kova´cs B.; Werner, T.; Wolfbeis, O. S. Anal.
Chem. 1993, 65, 123-127.
(40) Yang, X. H.; Wang, K. M. Guo, C. C. Anal. Chim. Acta 2000, 407, 45-52.
Analytical Chemistry, Vol. 77, No. 22, November 15, 2005 7295

Chart 1. Chemical Structures of the Absorber
and Fluorophores
there is still a need to develop a highly selective fluorescent
sensing method that allows real-time monitoring of the metal ion.
EXPERIMENTAL SECTION
Materials. Zntpp was synthesized and purified as reported
previously.50 For membrane preparation, PVC, potassium tetrakis-
(p-chlorophenyl)borate, bis(2-ethylhexyl) sebacate (BOS), and
tetrahydrofuran (THF) were purchased and used. All stock
solutions of metal ions were prepared from analytical grade nitrate
salts and were dissolved in doubly distilled water. The work
solutions of metals were obtained by series diluting the stock
solutions with 0.1 M Tris-HCl buffer (pH 6.98, I ) 0.1 (NaNO3)).
Other chemicals were of analytical reagent grade and used without
further purification.
1
Apparatus. H NMR and 13C NMR were measured with an
Invoa-400 (Invoa 400, 400 MHz for 1H and 100 MHz for 13C)
spectrometer with tetramethylsilane as the internal standard. J
values were given in hertz. Low-resolution mass spectra (MS) were
obtained at 50-70 eV by fast atomic bombardment on a Finnigan
MAT SSQ-710 mass spectrometer. High-resolution MS were
obtained on a Q-Star Pulsar I (Applied Biosystem/PE Sciex). UV-
visible measurements were performed on a Hitachi U-3010 UV/
Vis spectrophotometer (Kyoto, Japan). Fluorescence spectra were
conducted on a Hitachi F-4500 fluorescence spectrofluorometer
with an excitation slit width of 2.5 nm and an emission silt width
of 5.0 nm. Fluorescence decays were recorded using a FL920
fluorescence lifetime spectrometer. To perform the membrane
measurement, a self-made flow-through measuring cell51 was used.
The pH of the test solutions were measured with an electrode
connected to a PHS-3C pH meter (Shanghai, China) and adjusted
if necessary.
region; but the copper complex of 1 shows significant absorption
at 547 nm. On the other hand, we observed that the Q-band
absorption positions of a fluorophore, zinc meso-tetraphenylpor-
phyrin (Zntpp), are at 597, 556, and 518 nm. The main 556-nm
band of the porphyrin overlaps nicely with the absorption band
of the copper complex of 1. Thus, the effective intensity of the
excitation light beam of Zntpp would be decreased with increasing
absorbance of 1 if the two dyes coexist in a sensory device. This
experimental observation encouraged us to study the possibility
of detecting copper ion by Zntpp fluorescence spectroscopy. In
the present contribution, we report the IFE between Zntpp and 1
and establish a unique determination of copper ion via the
application of fluorescence sensing of the copper ion. To develop
a practical fluorescent sensor, both 1 and Zntpp were immobilized
in a plasticized poly(vinyl chloride) (PVC) membrane, and the
response characteristics of the sensing membrane to copper ion
were investigated. In contrast to the solution-phase measurement,
the polymeric membrane-based sensor possesses more favorable
properties with respect to practicability. For example, the mem-
brane-based sensor can be used routinely to achieve real-time and
continuous measurements, it also allows the regeneration of the
sensing reagent through simple washing procedure.
Synthesis of 1. The spiropyran derivative 1 was synthesized
by a modification of the procedure reported by Phillips as outlined
in Scheme 2.
(a) 5-tert-Butyl-2-hydroxylbenzaldehyde (4). To a solution
of 4-tert-butylphenol (3; 6.01 g, 40 mmol) in dry toluene (35 mL)
was added SnCl4 (1.07 g, 4.1 mmol) in dichloromethane (1.0 M,
4.1 mL). After completely mixing, Bu3N (3.91 mL, 16.4 mmol)
was added via syringe and allowed to stir for 20 min. To the
reaction mixture, (CH2O)n (2.43 g, 81 mmol) was added and
heated to reflux for 8 h. After cooling, the solution was poured
into a separating funnel and then extracted with diethyl ether (3
× 30 mL). The organic layer was washed with saturated NaCl
solution and dried with anhydrous Na2SO4. The solvent was
removed, and the product was purified by silica gel chromatog-
raphy using petroleum ether/ethyl acetate (9:1, v/v) as eluent to
The detection of copper ion is attracting continuous attention,
as copper is a significant metal pollutant due to its widespread
use, but it is also an essential trace element in biological systems.
Whereas copper toxicity for humans is rather low compared with
other heavy metal ions, certain microorganisms are affected by
submicromolar concentrations of metallic materials. Fluorescent
sensing of Cu2+ could be utilized to clarify the physiological role
of the metal in vivo as well as to monitor its concentration in the
metal-contaminated sources. Although a large number of copper
fluorescent sensors have been designed in the past few years,41-49
(43) Rurack, K.; Kollmannsberger, M.; Resch-Genger, U.; Daub, J. J. Am. Chem.
Soc. 2000, 122, 968-969.
(44) Mayr, T.; Klimant, I.; Wolfbeis O. S.; Werner, T. Anal. Chim. Acta 2002,
462, 1-10.
(45) Zheng, Y. J.; Orbulescu, J.; Ji, X. J.; Andreopoulos, F. M.; Pham, S. M.;
Leblanc, R. M. J. Am. Chem. Soc. 2003, 125, 2680-2686.
(46) Yang, R. H.; Zhang, Y.; Li, K. A.; Liu, F.; Chang, W. H. Anal. Chim. Acta
2004, 525, 97-103.
(47) Wu, Q. Y.; Anslyn, E. V. J. Am. Chem. Soc. 2004, 126, 14682-14683.
(48) Gunnlaugsson, T.; Leonard, J. P.; Murray, N. S. Org. Lett. 2004, 6, 1557-
1560.
(49) Royzen, M.; Dai, Z. H.; Canary, J. W. J. Am. Chem. Soc. 2005, 127, 1612-
1613.
(41) Kra¨mer, R. Angew. Chem., Int. Ed. 1998, 37, 772-773.
(42) Grandini, P.; Mancin, F.; Tecilla, P.; Scrimin, P.; Tonellato, U. Angew. Chem.,
Int. Ed. 1999, 38, 3061-3064.
(50) Zhang, Y.; Yang, R. H.; Li, K. A.; Liu, F. Anal. Chem. 2004, 76, 7336-7345.
(51) Wang, K. M.; Seiler, K.; Rusterholz, B.; Simon, W. Analyst 1992, 117, 57-
61.
7296 Analytical Chemistry, Vol. 77, No. 22, November 15, 2005

Scheme 2. Synthesis Scheme for Spiropyran 1 and the Copper Complex
1
give 4 as yellow oil in 85% yield. H NMR (400 MHz, CDCl3): δ
10.89(s, 1H), 9.86 (s, 1H), 7.56 (d, J ) 8.8, 1H), 7.51 (d, J ) 2.4,
1H), 6.91 (dd, J ) 2, 1H), 1.32 (s, 9H). 13C NMR (100 MHz,
CDCl3): δ 196.6, 159.3, 142.5, 134.5, 129.6, 119.8, 117.0, 33.9, 31.0.
(b) 5-tert-Butyl-3-[(N,N-dimethylamino)methyl]-2-hy-
droxylbenzaldehyde (5).52 At 0 °C and under N2, dimethylamine
(0.7 mL, 13.8 mmol) was introduced into glacial acetic acid (5.0
mL) and stirred for 15 min. (CH2O)n (0.17 g, 5.5 mmol) was added
into the reaction mixture at room temperature in 15 min. After
completely mixing, 4 (2.4 g, 13.5 mmol) in glacial acetic acid (2.0
mL) was injected to the reaction flask, and the resultant mixture
was heated to reflux for 4 h. The reaction mixture was poured
into 30 mL of water, and the pH was adjusted to slightly basiv.
The mixture was extracted with dichloromethane (3 × 30 mL).
The combined organic layers was dried with anhydrous Na2SO4.
The solvent was removed, and the product was purified using silica
gel chromatography (ethyl acetate/petroleum ether, 20:80, v/v)
(s, 2H), 2.36 (s, 6H), 1.28 (s, 9H). 13C NMR (100 MHz, CDCl3):
δ 191.6, 159.5, 141.4, 132.6, 124.3, 123.4, 121.8, 61.1, 44.4, 33.8,
31.1. MS (M+ + 1): 236.
(c) N-Methyl-2, 3, 3-trimethylindolenine (7). Under N2, a
mixture of 2,3,3-trimethylindolenine (6) (0.79 g, 5.0 mmol) and
iodomethane (0.85 g, 0.61 mmol) was heated to 50 °C for 2 h.
After cooling the solution to room temperature, yellow crystals
were collected by filtration. The solid was washed with ethanol
and diethyl ether successively. The solvent was removed under
vacuum, and the residue was recrystalized from absolute ethanol
1
to give 6 in 70% yield. HNMR (400 MHz, DMSO): δ 7.92 (m,
1H), 7.83 (dd, J ) 2, 1H), 7.62 (dd, J ) 2.8, 2H), 2.97 (s, 3H), 2.77
(s, 3H), 1.53 (s, 6H). 13C NMR (100 MHz, DMSO): δ 196.2, 142.3,
141.8, 129.5, 129.0, 123.5, 115.4, 54.2, 35.0, 21.9, 14.5.
(d) Spiropyran 1.53 Compound 7 (0.6 g, 2.0 mmol) in absolute
ethanol (30 mL) was heated to reflux. Piperidine (0.3 mL, 3.0
mmol) was added to the solution. After the resulting mixture was
completely mixed, 5-tert-butyl-3-[(N,N-dimethylamino)methyl]-2-
hydroxylbenzaldehyde (5; 0.68 g, 2.0 mmol) in absolute ethanol
(9.0 mL) was introduced. The mixture was refluxed for 5 h. The
1
to give 5 in 54% yield. H NMR (400 MHz, CDCl3): δ 11.42 (s,
1H), 10.35 (s, 1H), 7.64 d, J ) 2.8, 1H), 7.32 (d, J ) 2.4, 1H), 3.67
(52) Larrow, J. F.; Jacobsen, E. N.; Cao, Y.; Hong, Y.; Nie, X.; Zepp, C. M. J. Org.
Chem. 1994, 59, 1939-1942.
(53) Przystal, F.; Phillips, J. P. J. Heterocycl. Chem. 1967, 4, 131-132.
Analytical Chemistry, Vol. 77, No. 22, November 15, 2005 7297

solvent was removed by evaporation under reduced pressure. The
crude residue was purified by silica gel column chromatography
using ethyl acetate/heptane (80:20, v/v) as the eluant to afford 1
as a pink solid in 89% yield.1H NMR (400 MHz, CDCl3): δ 7.16
(m, 2H), 7.06 (d, J ) 6.8, 1H), 6.95 (d, J ) 2.4, 1H), 6.81 (m, 2H),
6.46 (d, J ) 7.6, 1H), 5.66 (d, J ) 10, 1H), 3.20 (q, J ) 13.2, 2H),
2.66 (s, 3H), 2.07 (s, 6H), 1.29 (s, 12H), 1.18 (s, 3H). 13C NMR
(100 MHz, CDCl3): δ 149.8, 142.1, 136.9, 130.0, 127.9, 127.3, 122.3,
121.3, 118.8, 118.5, 117.7, 106.5, 56.2, 51.1, 45.3, 33.9, 31.5, 28.9,
25.7, 20.4. MS(M+): 390.
Optode Membrane Fabrication. A typical membrane cocktail
consisted of 3.6 mg (10.0 µmol) of 1, 0.35 mg (0.5 µmol) of Zntpp,
50 mg of PVC, and 100 mg of BOS. These compounds were
dissolved in 2.0 mL of fresh distilled THF to make a clear and
homogeneous solution. A 0.1-mL aliqout of this solution was
applied to the surface of a circular 25-mm-diameter quartz plate,
which was mounted on a rotating (rotating frequency 600 rpm)
aluminum alloy rod.51 After a spinning time of only 5 s, a thin
film was coated onto the plate. From the amount of materials
employed, we estimate the thickness of the membrane to be 2-4
µm.
Optical Measurements. The absorption and fluorescence
titrations of metal ions in ethanol solution were run by directly
adding small aliquots (typically 20 µL) of the metal stock solutions
to 1.5 mL of the ethanol solution containing 5.0 × 10-5 M 1 and
2.5 × 10-6 M Zntpp in a quartz sample cuvette (1 cm × 1 cm).
After being mixed thoroughly and allowed to equilibrate for 5 min,
the absorption or fluorescence spectra were then recorded at room
temperature.
To assess the fluorescence responses of the optode membrane
to copper ion, the quartz plate with sensing membrane was first
fitted onto the front side of the measuring cell. A 25-mm-diameter
black PVC plate without sensing membrane was then fitted onto
the other side to complete the cell. The cell was introduced into
the spectrofluorometer at the appropriate position54 to measure
the fluorescence emission without interference from the excitation
source. About 2.0 mL of sample solution was introduced, and the
fluorescence spectra of the optode membrane were recorded with
an excitation wavelength of 556 nm and an emission wavelength
of 603 nm. Before each measurement, the membrane was
conditioned in plain buffer until the fluorescence intensity was
stabilized.
great affinity toward the d- and f-elements.55,56 Additionally, as a
practical sensory material, the metal binding interaction of the
spiropyran should be chemically reversible rather than photo-
chromically reversible. It was demonstrated that, for a given
medium, the position of equilibrium of the closed form and open
form of a spiropyran nucleus usually depends on the nature of
the substituent group in 6′-position. A strong electron-withdrawing
group in 6′-position produces a stable open form, while an electron-
donating group in the position leads to a more photostationary
closed state.11,57,58 Therefore, a tert-butyl group was introduced in
the 6′-position, superseding the nitro group in the position to
obtain a stable optical signal without photochromic behaviors
triggered by irradiating with ultraviolet light or visible light. As
an additional merit in the sensor design, the introducing of an
alkylated chain in the molecular skeleton ensures its high
hydrophobicity of the material, which meets the basic condition
for a component in a polymeric film.
To exploit the IFE for fluorescence response to the absorption
change of 1, a set of fluorophores were then tested. Although
many organic fluorescent dyes such as fluorescein (λex/λem ) 496/
520 nm), acridine orange (λex/λem ) 490/530 nm), rhodamine 6G
(λex/λem ) 530/590 nm), and methylrhodamine-5-iso-thiocyanate
(λex/λem ) 541/572 nm) exhibit an excellent excitation or emission
spectrum overlapping with the absorption maximum of the copper
complex of 1, they cannot function as a fluorescent indicator in
the present system, since the fluorescence of such dyes is
seriously quenched by copper ion. Thus, it is impractical to use
any of them to signal fluorescently the metal binding event via
IFE. To meet our need, we decided to use a metallorganic
fluorophore, Zntpp, as the fluorescent reporter. Although the
maximal emission position of Zntpp locates at 603 nm, the Q-band
absorption positions of the porphyrin are at 597, 556, and 518 nm,
the main 556-nm band overlaps well with the absorption band of
the copper complex of 1. Therefore, if the two dyes are present
simultaneously in this approach, the effective intensity of the
excitation light beam of Zntpp at 556 nm will be decreased with
increasing absorbance of 1 as a result in response to increasing
copper ion concentration. Further, we selected Zntpp as the
fluorophore because of its fluorescence feature free from the
influence of external factors such as pH, metal ions, and anions
(see below).
Binding Interactions of 1 and Metal Ions. Prior to applica-
tion in fluorescence sensing of copper ion, the binding interaction
of 1 with divalent metal ions was first studied by UV-visible
spectroscopy. Figure 1 shows the absorption spectra of 1 in
ethanol. The free state of 1 is colorless (5.0 × 10-5 M). and the
absorption spectrum displays an absorption maximum at 314 nm.
No significant absorption is observed in the visible region, which
indicates the nearly complete absence of the merocyanine
component. Figure 1 also shows the absorption spectral changes
RESULTS AND DISCUSSION
Choices of Absorber and Fluorophore. As the starting point,
the performance of IFE requires the presence of two dyes, one
acting as analyte-sensitive absorber and the other as the analyte-
independent fluorophore whose the excitation or emission inten-
sity is modulated by varying the absorption of the absorber. The
requisite copper-sensing absorber, spirospyran 1, was synthesized
in a three-step convergent sequence shown in Scheme 2. To
achieve strong and highly specific binding to metal ion, the ion
receptor domain of the spiropyran nucleus was formed by
introducing a dimethylaminomethyl moiety in the 8′-position of
the molecule via the Mannich reaction. The bidentate N,O-
chelating binding site present in the molecule would generate
of 1 in the presence of 5.0 × 10-5 M Hg2+, Cd2+, Zn2+, and Cu2+
,
respectively. The absorption spectrum of 1 was hardly influenced
(55) Barthram, A. M.; Ward, M. D.; Gessi, A.; Armaroli, N.; Flamigni, F.;
Barigelletti, F. J. New Chem. 1998, 913-917.
(56) Tong, H.; Wang, L. X.; Ting, X. B.; Wang, F. Macromolecules 2002, 35,
7169-7171.
(57) Keum, S.-R.; Lee, K.-B.; Kazmaier, P. M.; Buncel, E. Tetrahedron Lett. 1994,
35, 1015-1018.
(54) Yang, R. H.; Wang, K. M.; Xiao, D.; Yang, X. H.; Li, H. M. Anal. Chim. Acta
2000, 404, 205-211.
(58) Swansburg, S.; Bucel, E.; Lemieux, R. P. J. Am. Chem. Soc. 2000, 122,
6594-6600.
7298 Analytical Chemistry, Vol. 77, No. 22, November 15, 2005

between Cu2+ and 1, one can describe the equilibrium as follows:
K
2SP + Cu2+
y
z Cu(SP)2
(1)
where SP and Cu(SP)2 denotes 1 and its copper complex,
respectively. The corresponding association constant, K, can be
expressed as follows:
[Cu(SP)2]
[SP]2[Cu2+
K )
(2)
]
According to the derivation reported elsewhere,60 a response
function for Cu2+ is given below following the mass law:
Figure 1. UV-visible absorption spectra of 1 (5.0 × 10-5 M) in
ethanol solution in the absence and presence of 5.0 × 10-5 M Cd2+
,
Hg2+, Zn2+, and Cu2+, respectively.
R2
1
)
(3)
2KCT[Cu2 +
]
1 - R
where CT denotes the total concentration of 1 in the system. The
R term in eq 3 can be defined as the ratio of the concentration of
uncomplexated 1 to the total concentration of 1 present in the
system.
R ) [C]/CT
(4)
R can be determined from the absorbance changes in the presence
of different concentrations of Cu2+
:
Figure 2. Job plot for determining the stoichiometry of 1 and Cu2+
.
The total concentration of 1 and copper was 1.0 × 10-4 M. Molar
R ) (A - A0)/(A1 - A0)
(5)
fraction was given by [Cu2+]/([Cu2+] + [1]).
where A1 and A0 are the limiting absorbance values for R ) 1 (in
the absence of Cu2+) and R ) 0 (1 is completely complexed with
Cu2+), respectively. The term R can also be named relative
absorption signal. It is apparent from eq 3 that the response
parameter (R) has a distinct functional relationship with the
concentration of copper ion and the association constant, K.
Equation 3 provides the basis for the determination of K value.
The experimental data were fitted to eq 3 by adjusting the K value.
Figure 3 shows the fitted curve to incorporate the experimental
data for copper ion, which refers to log K ) 6.8.
To study the binding affinity of 1 to other divalent metal ions,
1 was subjected to different concentrations of alkaline-earth and
heavy metals, separately (see Supporting Information). Additional
spectral properties of the metal complexes of 1 are summarized
in Table 1. The very large extinction coefficient of the copper
complex can contribute to the susceptibleness of the fully occupied
d-orbit of Cu2+. The response selectivity factors of 1 for different
metal ions were evaluated by comparing the binding affinity of 1
to metal ion and the absorption enhancement of 1 by metal ion
(Kꢀj),61 where K is the association constant of 1 with metal ion
and ꢀ is the extinction coefficient of the metal complex. The
response to copper ion was used as the standard. On the basis of
by addition of Hg2+ or Cd2+, separately, while a small enhancement
in the absorbance was observed upon the addition of Zn2+. In
contrast, when Cu2+ was added to the ethanol solution of 1, a
pink solution appeared with a strong absorption peak at 547 nm.
The absorbance of 1 at 547 nm is increased to 77.2-fold that of
the original value by Cu2+ (5.0 × 10-5 M), but it is only 2.34-fold
by Cd2+, 4.16-fold by Hg2+, and 7.84-fold by Zn2+
.
UV-visible spectroscopy was followed as aliquots of copper-
(II) nitrate were added to the solutions of 1. Dramatic increase
in the absorbance of 1 at 547 nm was evident with increasing
concentration of copper ion up to ∼0.5 equiv relative to the host
concentration. To determine the stoichiometry of the copper-
ligand complex, Job’s method for the absorbance was applied,
keeping the sum of the initial concentration of copper ion and 1
at 1.0 × 10-4 M, and the molar ratio of copper ion changing from
0 to 1.59,4h The absorbances of 1 in the absence (A0) and presence
(A) of copper ion were determined, respectively. A plot of (A -
A0)/A0 versus the molar fraction of copper ion is provided in
Figure 2. It shows that the (A - A0)/A0 value goes through a
maximum at a molar fraction of ∼0.33, indicating a 1:2 stoichi-
ometry of the Cu2+ to 1 in the complex.
The association constant can be determined from the absorp-
tion titration data. If a 1:2 metal-ligand complex is formed
(60) Yang, R. H.; Li, K. A.; Wang, K. M.; Zhao, F. L.; Li, N.; Liu, F. Anal. Chem.
2003, 75, 612-621.
(61) Zhao, J. Z.; Fyles, T. M.; James, T. D. Angew. Chem., Int. Ed. 2004, 43,
3461-3464.
(59) Job, P. Ann. Chim. 1928, 9, 113-116.
Analytical Chemistry, Vol. 77, No. 22, November 15, 2005 7299

Figure 4. Kinetics of 1 reaction with different concentrations of Cu2+
.
Figure 3. Response parameter (R) as a function of logarithm of
Cu2+ concentration at pH 6.98. The curve fitting the experimental
points (b) was calculated from eq 3.
Absorbance was recorded at 547 nm in 5.0 × 10-5 M 1 in ethanol
solution.
Table 1. Absorption Properties of 1 and Its Metal
Complexes in the Visible Region, Association
Constants, K, and Response Selectivity of 1 to Metal
Ions
λmax
(nm)
ꢀ, 103
response
species
(M-1‚cm-1
)
K (M-2
)
selectivitya
1
552
547
560
554
560
552
572
574
556
562
554
0.44
31.2
0.50
0.57
0.87
2.85
0.69
3.04
1.62
0.53
0.72
1 + Cu2+
1 + Mg2+
1 + Ca2+
1 + Cd2+
1 + Co2+
1 + Ni2+
1 + Zn2+
1 + Hg2+
1 + Mn2+
1 + Pb2+
6.31 × 106
3.42 × 102
2.65 × 102
1.12 × 104
3.02 × 105
5.36 × 104
6.67 × 104
1.90 × 103
8.56 × 102
1.08 × 103
1
9.66 × 10-7
8.53 × 10-7
5.50 × 10-4
4.86 × 10-4
2.09 × 10-6
1.15 × 10-4
1.74 × 10-5
2.56 × 10-6
4.39 × 10-6
Figure 5. Absorption spectra of 1 in the absence (a) and presence
of 5.0 × 10-5 M copper ion (b), and fluorescence excitation (λem
603 nm (c)) and emission spectra (λex )556 nm (d)) of Zntpp.
)
a Response selectivity, (Kiꢀ ) /(KCu
ꢀCu).
i
shows the merocyanine copper complex is reasonably photostable.
Irradiating the colored solution with visible light did not liberate
metal ion with regeneration change of the absorbance. We
envisioned that the 6-tert-butylphenolate merocyanine was a better
ligand for the metal ion than 6-nitrophenolate formation. The
6-nitro group withdraws electron density from the phenolate
oxygen, decreasing the density of the ligand and thereby facilitat-
ing reversible binding.14 The liberation of copper ion from the
complex could be readily achieved by addition of EDTA to the
complexed solution, demonstrating that the binding is readily
chemically reversible.
Fluorescence Inner Filter Effect between 1 and Zntpp in
the Presence of Copper Ion. Figure 5 shows the absorption
spectra of 1 and the fluorescence spectra of Zntpp at different
experimental conditions. In the present approach, the states of
the metal-free and metal-bound of 1 show different absorption
behaviors, the copper complex of 1 is highly absorbing, which
absorbs in the visible region of 450-600 nm, but the metal-free
form is not as shown in curves a and b. The curves c and d are
the fluorescence excitation and emission spectra of Zntpp. The
excitation maximum of Zntpp in the Q-band region is at 556 nm,
which has an excellent overlap with the absorption spectrum of
the copper complex of 1. The two bands of the emission spectrum
at 603 and 654 nm are assigned to 0-0 and 1-0 vibronic
transitions of Zntpp.62,63 Since the absorbance of both dyes follows
the Beer-Lambert law, the relative quantities of the light absorbed
at 556 nm by each of them is determined by the ratio of the
the spectroscopic data in Table 1, one can easily recognize that
remarkably high selectivity of 1 for Cu2+ is realized.
A series of 6′-nitro substituent spiropyran derivatives have been
widely used to investigate the photoinduced metal binding
interaction.9 The 6′-nitro group withdraws electron density from
the phenolate oxygen, which decreases the density of phenolate
oxygen and thereby facilitates formation of the merocyanine form
to bind almost all divalent metal ions. On the contrary, the 6′-tert-
butyl group (electron-donating group) in 1 tends to produce a
photostationary spiropyran state with a negligible merocyanine
component, which will reduce its metal binding ability. This
seemingly nonactivated spiropyran has rather limited use in
photochromic application, but it could be exploited as a practical
sensor with respect to its metal binding selectivity and optical
signal stability. We reasoned that the strong binding affinity of 1
toward Cu2+ over other divalent metal ions may contribute to the
particularly high thermodynamic affinity of copper for N,O-chelate
ligands and the fast metal-to-ligand binding kinetics.41,13
In view of photoreversible structural conversion of the mero-
cyanine metal complex, the real-time records of the metalation
degree of 1 were carried out in ethanol using the 547-nm band
absorption as a function of time. As shown in Figure 4, the
formation of the metal complex proceeds rapidly at first, followed
by more gradual increase in the absorption. In addition, the
reaction time increases with Cu2+ concentrations. Figure 4 also
7300 Analytical Chemistry, Vol. 77, No. 22, November 15, 2005

extinction coefficients at that wavelength. Thus, it is possible that
the efficient light absorbed at 556 nm by Zntpp and, concomitantly,
the emitted light intensity will be reduced with increasing
absorbance of 1 in response to increasing copper ion concentra-
tion. Fluorescence spectroscopy was followed as aliquots of
copper(II) nitrate were added to the ethanol solutions containing
1 and Zntpp. As expected from the original design, both the
fluorescence excitation and emission intensities of Zntpp remark-
ably decrease by adding increasing concentrations of copper ion
to the 1 + Zntpp ethanol solution without changes of the spectral
feature (see Supporting Information). This spectral change
constitutes the basis for the fluorescent detection of copper ion
based on IFE.
The observed decrease in fluorescence intensity of the Zntpp
may also be attributed to the possibility that copper ion directly
quenches the Zntpp fluorescence. To clarify this issue, the effect
of copper ion on the fluorescence spectra of Zntpp in the absence
of 1 was also studied. The fluorescence signal of Zntpp without
1 hardly changes when copper ion concentration increases, which
demonstrates that the absorber, 1, is the key player to regulate
the fluorescence intensity of the fluorophore, Zntpp.
Figure 6. Effects of different composition ratios of 1 and Zntpp on
the fluorescence responses to Cu2+. The excitation was at 556 nm,
and emission was recorded at 603 nm.
Excited-state fluorescence energy transfer has been studied
between covalently linked spiropyran and fluorophore.31 To test
whether there is energy transfer between 1 and Zntpp in the
present approach, decay curve measurements of 1 + Zntpp in
ethanol were performed using a multichannel analyzer operating
with 512 channels at 20 °C.64 In the absence of copper ion, global
analysis of the resulting data as a single-exponential decay gave
a satisfactory fit (ø2 ) 1.218) with a lifetime of 2.18 ns, which is
nearly a lifetime for Zntpp in ethanol,63,65,66 indicating that 1 has
no significant effect on the lifetime of the porphyrin. When copper
ion was added to the solution of 1 and Zntpp, the decay curve
was slightly complicated and could be fitted as monoexponential
decay with two lifetimes (ø2 ) 1.194). The lifetime of the major
component (97.3%) was 2.07 ns, while that of the other component
(2.7%) was 5.7 ns. The almost no lifetime change of the major
component indicates that there is no significant excited-state
interaction between 1 and Zntpp in the presence of copper, further
demonstrating that the fluorescence change of Zntpp by copper
is due to the simple absorption of the excitation light by the
absorber.
Fluorescent Sensing of Copper Ion. The fluorescence
response of 1 + Zntpp to copper ion is strongly dependent on
the relative amount of the fluorophore and absorber. This can be
seen in the variation of the response characteristics (working
range and response sensitivity) due to the different amount ratios
of fluorophore and absorber. In Figure 6, the fluorescence
quenching efficiencies (F0/F) of Zntpp by copper ion for three
composition ratios are plotted as functions of the concentration
of Cu2+ at pH 6.98, where F0 and F are the fluorescence intensities
Figure 7. Fluorescence intensity-pH profiles for titrations of the
ethanol solution containing 5.0 × 10-5 M 1 and 2.5 × 10-6 M Zntpp
in the absence (a) and presence of 4.5 × 10-6 M (b) and 1.8 × 10-5
M (c) copper ion. The excitation was at 556 nm, and emission was
recorded at 603 nm.
of Zntpp at λex/em ) 556/603 nm in the absence and presence of
Cu2+. Obviously, when the molar ratio of 1 to Zntpp reaches 20:
1, the best response emerged in terms of the sensitivity and
efficient detection range for Cu2+. Lower concentration of Zntpp
in the system results in the weak blank fluorescence signal value,
which will reduce the dynamic working range of the measurement;
while at relatively higher fluorophore concentration, the fluores-
cence of Zntpp could not be quenched completely even if a high
concentration of copper ion was added to the solution. The
response slope, dynamic working range, and detection limit of
the system could be tuned by controlling the relative amounts of
the absorber and fluorophore. This is one of the advantages that
the IFE method has over the fluorometry based on a single
fluorophore. The flexibility of component controlling allows us to
monitor the metal ion concentration with appropriate adjustment
of the amount of the absorber, so that the sample response falls
within the most sensitive response region.
The ligation interaction of 1 with copper ion was influenced
by the acidity of the medium. Figure 7 depicts the pH dependence
of fluorescence changes of Zntpp in the absence and presence of
copper ion. The titration experiments were carried out by adding
standard NaOH or H2SO4 solution to a solution containing 5.0 ×
10-5 M 1, 2.5 × 10-6 M Zntpp as well as copper ion. It is obvious
that the fluorescence signal of Zntpp in the absence of copper
(62) Gust, D.; Moore, T. A.; Moore, A. L.; Macpherson, A. N.; Lopez, A.;
DeGraziano, J. M.; Gouni, I.; Bittersmann, E.; Seely, G. R.; Ronald, F. G.;
Ma, X. C.; Demanche, L. J.; Hung, S.-C.; Luttrull, D. K.; Lee, S.-J.; Kerrigan,
P. K. J. Am. Chem. Soc. 1993, 115, 11141-11152.
(63) Yu, H.-Z.; Baskin, J. S.; Zewail, A. H. J. Phys. Chem. A 2002, 106, 9845-
9854.
(64) The decay curves are shown in Supporting Information.
(65) Cho, H. S.; Rhee, H.; Song, J. K.; Min, C.-K.; Takase, M.; Aratani, N.; Cho,
S.; Osuka, A.; Joo, T.; Kim, D. J. Am. Chem. Soc. 2003, 125, 5849-5860.
(66) Mataga, N.; Shibata, Y.; Chosrowjan, H.; Yoshida, N.; Osuka, A. J. Phys.
Chem. B 2000, 104, 4001-4003.
Analytical Chemistry, Vol. 77, No. 22, November 15, 2005 7301

Figure 9. Fluorescence quenching of Zntpp by increasing concen-
trations of Cd2+, Hg2+, Zn2+, and Cu2+, separately, as well as a
Figure 8. Time history of the sensing membrane responses to
different concentrations of Cu2+. Fluorescence intensity was recorded
at 603 nm with an excitation wavelength of 556 nm.
mixture of 1.0 × 10-3 M Mg2+ and Ca2+ and 5.0 × 10-5 M Hg2+
,
Cd2+, and Zn2+ with increasing concentrations of Cu2+. The ordinate
shows the fluorescence quenching (F/F0). The excitation was at 556
nm, and emission was recorded at 603 nm.
ion is independent of pH in the range of 3.0-11.0 (curve a). Curves
b and c show the pH dependences of the Zntpp fluorescence
response to 4.5 × 10-6 and 1.8 × 10-5 M copper ion, respectively.
The efficient pH range for copper ion detection is 5.0-9.0. At a
pH value below 5.0, 1 converted to its protonated merocyanine
form (HME+), which reduces its ligation with copper ion; while
at relatively higher pH, copper ion may be complexed by OH-,
which in turn reduces its complex with 1.
With the optimum conditions (pH 6.98, nZntpp:nSP ) 1:20), the
sensing characteristics of a PVC membrane containing 1 and
Zntpp were studied. Figure 8 shows the typical response of the
sensing membrane to Cu2+, as obtained after equilibration with
Tris-HCl buffer solutions containing different concentrations of
Cu2+. The fluorescence intensity was recorded after equilibration
with different concentrations of Cu2+ with excitation at 556 nm
and emission at 603 nm. The values of the fluorescence intensity
of the optode membrane decrease considerably as the concentra-
tion of copper ion increases. This illustrates that the optode
membrane can be used for assay of copper ion in aqueous sample
solution. The optode membrane exhibited the highest sensitivity
to Cu2+ in the 7.5 × 10-7-3.6 × 10-5 M range. The limit of
detection, which was determined by alternately exposing the
optode membrane 3 times to plain buffer and low concentration
of Cu2+ and calculating the Cu2+ concentration corresponding to
signal changes that are 6 times the standard deviation of the plain
buffer signal, was found to be 1.5 × 10-7 M.
total signal change to occur), was within 6 min. The liberation of
copper ion from the membrane could be achieved by contacting
it with Tris-HCl buffer containing 0.001 M EDTA, whereas the
time for the reverse response was in the range of 3-5 min over
the entire concentration range. The response time of such an
optode membrane is mainly governed by the diffusion process in
the bulk of the membrane and particularly the rate of complex
formation. Shaking the cell can shorten the response time by ∼2
min. Dynamic flow of the sample solution can also shorten the
response time.
From the fluorescence signal values of 1.2 × 10-5 M copper
ion taken every 15 min over a period of 6 h, a mean fluorescence
intensity value of 105.76 and a standard deviation of 3.93 were
obtained. The stability of the sensor at short times is reasonable.
The optode membrane is mainly subject to the photochemical
reaction and the leaching of 1 on which the lifetime of the optode
depends. The fluorescence response value of the optode mem-
brane decreased ∼8% after a continuous 100 measurements. In
addition, the leaching of plasticizer also affects the response time.
It would take 3-5-fold more time to reach t95 than that of the
freshly prepared membrane if the membrane is stored in buffer
solution for longer than two weeks. Nevertheless, the lifetime of
the optode membrane is acceptable for analytical application.
The fluorescence response of the sensor to copper ion presents
excellent selectivity in comparison with other heavy metal ions.
Zn2+, Hg2+, and Co2+ showed relatively weak effects on the Zntpp
fluorescence, while Mg2+, Ca2+, Mn2+, Pb2+, and Ni2+ could not
induce any change in the Zntpp fluorescence even with a high
concentration of the metal ion present in the sample solution. To
explore further the utility of the membrane as a copper-selective
fluorescent sensor, competition experiments were also conducted
in which the membrane was first immersed in a mixture of other
metal ions, and Cu2+ was then added to the mixture with
increasing concentrations. Figure 9 illustrates the changes of
Zntpp fluorescence intensity upon additions of different concentra-
The reproducibility and reversibility of the optode membrane
were evaluated by repetitively exposing it to 1.2 × 10-5 M copper
solution and Tris-HCl buffer solution (pH 6.98). The mean
fluorescence intensity values with their confidence intervals were
found to be 107.63 ( 2.86 (n ) 5, 1.2 × 10-5 M copper ion) and
197.7 ( 3.41 (n ) 5, Tris-HCl buffer solution). The response time
of the sensor depends on the thickness of the membrane, the
flow rate of the sample solution, and the change of the concentra-
tion of copper ion. When the thickness of the membrane reaches
the order of millimeters, there was no significant fluorescence
response observed even when the membrane was immersed in
copper ion solution for several hours. To prepare very thin
membranes, the spin-on technique was used, which allows the
production of very thin, homogeneous, and reproducible PVC-
based membranes. The forward response time (going from lower
to higher copper ion concentration), t95 (time needed for 95% of
tions metal ions as well as a mixture containing 1.0 × 10-3
M
Mg2+ and Ca2+ and 5.0 × 10-5 M Hg2+, Cd2+, and Zn2+ with
increasing amounts of copper ion. No significant variations in the
fluorescence intensity were found by comparison with those
without other metal ions rather than Cu2+
.
7302 Analytical Chemistry, Vol. 77, No. 22, November 15, 2005

ficient of the copper complex of 1 as well as the relative excess
of absorber over the fluorophore. As a result, even small changes
in the copper ion concentration cause a substantial change in the
effective intensity of excitation light beam of the fluorophore, and
thus in its fluorescence emission intensity.
CONCLUSION
The work described here demonstrates for the first time the
feasibility of exploiting the absorption-based spiropyran derivative
as a fluorescent sensor approach to detect metal ion via the inner
filter effect. Compared with classical fluorescent probes for Cu2+
detection in which the donor atoms are part of the fluorophore
π-system, the spatial separation of chelating group and fluorophore
offers considerable flexibility in design. The selectivity is governed
by the binding affinity of the absorber with the metal ion, while
the sensitivity and dynamic working range of the fluorescence
response is dependent on the extinction coefficient of the metal
complex of the absorber as well as the relative amount of the
absorber and fluorophore. The binding of 1 with copper ion
presents an excellent selectivity in comparison with several other
transition metal ions. Moreover, because the changes in the
absorption of the absorber translate into exponential changes in
fluorescence, this scheme offers an attractive alternative to
absorptiometric as well as fluorometric technique using one single
dye with respect to the response sensitivity. The sensing approach
presented here seems to be a very flexible and general and can
be applied to any absorption-based sensor by selecting a suitable
fluorophore.
Figure 10. Comparison of absorptiometric and fluorometric tech-
niques for copper ion detection.
Comparison of Fluorometric and Absorptiometric Tech-
niques. To compare the analytical characteristics of fluorometry
and absorptiometry in the present approach, in Figure 10, the
signal changes of absorbance and fluorescence are plotted as a
function of the logarithm of copper ion concentration. Noted that,
in general, the changes in absorbance are positive and those of
fluorescence are negative as expected. Since 1 absorbs the light
at the excitation position of Zntpp, it controls the amount of light
available to excite the porphyrin. In the absence of copper ion, 1
hardly absorbs light at 556 nm, the light available to excite Zntpp
is thus nearly the same as the initial excitation light. As the
concentration of copper ion increases, 1 absorbs more light at
556 nm, so there is less light available to excite the porphyrin,
which causes a decrease in the fluorescence emission.
From Figure 10, it is clear that the dynamic range of
fluorometry shifts to a low copper ion concentration range with
respect to that of absorptiometry. At low concentration of copper
ion, there is no real change in the absorbance of 1 while there is
clearly a significant decrease in the fluorescence intensity of Zntpp
upon increasing copper ion concentration. It probably can be even
improved by further optimizing the amount ratio of the absorber
and the fluorophore. The improvement in the limit of detection
and dynamic response range caused by IFEs can be explained
by the intrinsic relationship between fluorescence and absorbance;
i.e., the changes in the absorption of the absorber translate into
exponential changes in fluorescence. In addition, we attribute the
enhanced sensitivity of our sensor to the large extinction coef-
ACKNOWLEDGMENT
Financial support from the National Science Foundation of
China (Grant 20475005) is gratefully acknowledged.
SUPPORTING INFORMATION AVAILABLE
Absorption and fluorescence response curves of 1 + Zntpp to
metal ions, fluorescence decay curves, and HNMR spectra of 4,
5, 7, and 1. This material is available free of charge via the
Internet at http://pubs.acs.org.
Received for review June 8, 2005. Accepted September 5,
2005.
AC051010R
Analytical Chemistry, Vol. 77, No. 22, November 15, 2005 7303