2
82
J.-S. Wu et al. / Spectrochimica Acta Part A 67 (2007) 281–286
−
−
−
over other halide anions, i.e., F , Br , I . In this case, the emis-
ethanol/water = 5/1 (v/v) mixed solvent. The pH value of the
solution was modulated to neutral and scarlet precipitate (2.3 g)
was obtained in 84% yield. H NMR (400 MHz, DMSO) δ 9.90
(s, 2H, ArOH), 8.49 (d, 2H, ArH), 8.37 (d, 4H, ArH), 8.29 (d,
2H, ArH), 7.93 (s, 2H, ArH), 7.01 (d, 4H, ArH). MS (m/e): 364
ꢀ
sion of ligand 2,9-bis(4 -hydroxyphenyl)-phenanthroline was
1
revived upon addition of chloride anion. The combination of
both steric hindrance and ligand substitution is probably the
main reason for its high selectivity for chloride anion.
(347, 335, 333, 281, 273, 226, 158, 151, 94). Anal. Calcd. for
C24H N2O2: C 79.11%, H 4.43%, N 7.69%, found: C 79.18%,
H 4.48%, N 7.53%.
16
2
. Experimental
2
.1. Instruments and reagents
2.2.3. L/Cu(II)
A portion of L (364 mg, 1 mmol) and copper perchlorate
All fluorescence spectra in this work were recorded in
1
(370 mg, 1 mmol) were combined in CH3CN (20 mL), then the
solvent was evaporated under reduced pressure. The residue was
recrystalled in ethanol/acetonitrile = 10/1 (v/v) mixed solvent
Hitachi F-4500 fluorescence spectrometer. H NMR spectra
wererecordedat400 MHz, Bruker-400instrument. Massspectra
were recorded at Finnigan 4021C MS-spectrometer and Biflex
III MALDI-TOF MS-spectrometer. Elemental analysis were
recorded at Flash EA 1112 instrument.
1
and blue crystal was obtained in 27% yield. H NMR (400 MHz,
DMSO) δ 10.00 (s, 2H, ArOH), 9.09 (s, 2H, ArH), 8.68 (s,
H, ArH), 8.15 (d, 4H, ArH), 7.99 (s, 2H, ArH), 6.83 (d, 4H,
ArH), 5.91(s, 6H, CH3CN). MALDI-TOF: 708.2. Anal. Calcd.
for C28H22Cl2CuN4O10: C 47.44%, H 3.13%, N 7.90%, found:
C 47.67%, H 3.32%, N 7.73%.
2
Copper(II) perchlorate hydrate and anionic compounds tetra-
−
−
−
−
n-butyl ammonium salts of F , Cl , Br and I were all
purchased from Aldrich and used without further purification.
,10-phenanthroline was analytical grade from Beijing chemical
1
works and was recrystallized with ethanol once before use. Ace-
tonitrile was chromatographic grade and other reagents were all
analytical purity.
3
. Results and discussion
Ligand L and complex L/Cu(II), were synthesized according
to the route shown in Scheme 1. Ligand L is a strongly emissive
compound peaked at 402 nm, whereas complex L/Cu(II) is non-
emissive because of the photoinduced electron transfer (PET)
process [34,35]. In the presence of different anions, the fluores-
cence intensity of the solution in acetonitrile can be recovered to
2
2
.2. Synthesis
ꢀ
.2.1. 2,9-Bis(4 -methoxyphenyl)phenanthroline
Lithium (2.6 g, 0.375 mol) and ethyl ether (60 mL) were
combined in a 250 mL three-neck flask and aerated by nitrogen-
stream for 10 min. The mixed solution was stirred with refluxing
condition, then 4-bromoanisole (19 mL, 0.15 mol) was added
dropwise within an hour. Continuing to stir the reaction for
another 2 h, 4-bromophenoxy lithium was obtained. After that,
−
−
−
−
different extents as the following sequence: Cl > F > Br > I
Fig. 1A). Upon addition of 10 equivalents of chloride anion, the
(
fluorescence of complex L/Cu(II) at 402 nm was revived, and
the fluorescence intensity of this solution is about 70% of the
original intensity of ligand L. Whereas, at the same conditions,
1
,10-phenanthroline (4.0 g, 22 mmol) and toluene (50 mL) were
−
−
−
F , Br and I only induced 15, 12 and 10% enhancement,
respectively. This result indicates that L/Cu(II) may be used
as an efficient chemosensor for chloride-selective detection. To
confirm our observation, fluorescence titration of L/Cu(II) in
acetonitrile at different concentrations of (C4H9)4NCl was per-
formed (Fig. 1B). The fluorescence of the solution increases with
increasing concentrations of chloride anion regularly. Job’s plot
combined into another three-neck flask under nitrogen atmo-
sphere, then 4-bromophenoxy lithium solution prepared above
was added dropwise within half an hour. The reaction was stirred
for 48 h and then cooled in ice-bath. Water (40 mL) was added to
hydrolyze the reaction and the solution became yellow. Organic
product was extracted with dichloromethane and separated from
water layer. The organic phase was added to MnO2 (40 g) and
stirred for an hour, then MnO2 was filtrated and the solvent
was evaporated under reduced pressure. The crude product was
recrystallized by toluene/petroleum ether = 1/1 (v/v) and color-
less needle crystal (2.75 g) was obtained in 45.5% yield (mp:
−
shows that L/Cu(II) can bind Cl in 1:2 stoichiometery (inset in
−
Fig. 1B). The stability constant between L/Cu(II) and Cl was
determined by a nonlinear curve fitting procedure of the fluores-
cence titration data (Fig. 1B). The value of Kass was calculated
8
−2
to be 2.6 × 10 M based on 1:2 stoichiometery, whereas the
◦
1
1
8
82–184 C). H NMR (400 MHz, CDCl3) δ 8.45 (d, 2H, ArH),
stability constants for L/Cu(II) with other halides are all below
0 M at the same calculation model [36], indicating its good
selectivity for chloride anion over other halide anions, i.e., F ,
Br and I .
To look into the nature of recognition process, H NMR
.31 (d, 4H, ArH), 8.12 (d, 2H, ArH), 7.78 (s, 2H, ArH), 7.13
3
−2
1
(
d, 4H, ArH), 3.95 (s, 6H, OCH3).
−
−
−
ꢀ
1
2
.2.2. 2,9-Bis(4 -hydroxyphenyl)phenanthroline (L)
ꢀ
Pyridine (7 mL), 2,9-bis(4 -methoxyphenyl)phenanthroline
experiments were performed in DMSO-d because of its low
6
1
(
2.75 g, 7.0 mmol) and concentrated HCl (8 mL) were combined
solubility in acetonitrile. The H NMR spectra of L/Cu(II)
and reacted for 3 h at refluxing conditions under nitrogen atmo-
sphere. The reaction was cooled to room temperature and then
poured into 100 mL water. During this process, a lot of pre-
cipitate appeared. The residue was filtrated and dissolved in
(Fig. 2a) show dramatic changes upon addition of 10 equiva-
lents of (C4H9)4NCl in DMSO-d (Fig. 2b). Ho, Hm and –OH
6
proton signals show significant downfield shifts (ꢀδ = +0.22,
+0.26 and +0.33, respectively) and H4,7 and H3,8 proton signals