1,8-Naphthyridine-Derived Ni2+/Cu2+-Selective fluorescent chemosensor with different charge transfer processses

Article abstract of DOI:10.1021/ic3018453

A highly fluorescent chemosensor based on 1,8-naphthyridine with high sensitivity and selectivity toward Ni²⁺/Cu²⁺ over other cations both in aqueous solution over a wide pH range (4-10) and in cellular environments was developed. Counteranions such as acetate, sulfate, nitrate, and perchlorate have no influence on the detection of such metal ions. Ethylenediamine showed high selectivity toward the in situ-prepared Cu ²⁺ complex over the Ni²⁺ complex, which can be applied to distinguish Ni²⁺ and Cu²⁺. The Ni²⁺-induced fluorescence on-off mechanism was revealed to be mediated by intramolecular charge transfer from the metal to the ligand, while that by Cu²⁺ involves intramolecular charge transfer from the ligand to the metal, as confirmed by picosecond time-resolved fluorescence spectroscopy and time-dependent density functional theory calculations.

Full text of DOI:10.1021/ic3018453

Article
pubs.acs.org/IC
2
+
2+
1
,8-Naphthyridine-Derived Ni /Cu -Selective Fluorescent
Chemosensor with Different Charge Transfer Processses
Zhanxian Li,* Wanying Zhao, Xiaoya Li, Yanyan Zhu, Chunmei Liu, Lina Wang, Mingming Yu,*
Liuhe Wei, Mingsheng Tang, and Hongyan Zhang*
,
†
†
†
†
†
†
,†
†
†
,‡
†
College of Chemistry and Molecular Engineering, Zhengzhou University, Zhengzhou 450001, China
‡
Key Laboratory of Photochemical Conversion and Optoelectronic Materials, Technical Institute of Physics and Chemistry, Chinese
Academy of Sciences, Beijing 100190, China
*
S Supporting Information
ABSTRACT: A highly fluorescent chemosensor based on 1,8-
2+
2+
naphthyridine with high sensitivity and selectivity toward Ni /Cu
over other cations both in aqueous solution over a wide pH range
(
4−10) and in cellular environments was developed. Counteranions
such as acetate, sulfate, nitrate, and perchlorate have no influence on
the detection of such metal ions. Ethylenediamine showed high
2
+
2+
selectivity toward the in situ-prepared Cu complex over the Ni
2+
2+
complex, which can be applied to distinguish Ni and Cu . The
Ni -induced fluorescence on−off mechanism was revealed to be
mediated by intramolecular charge transfer from the metal to the
2+
2
+
ligand, while that by Cu involves intramolecular charge transfer
from the ligand to the metal, as confirmed by picosecond time-
resolved fluorescence spectroscopy and time-dependent density functional theory calculations.
INTRODUCTION
be better explained by fluorescence decay processes as well as
■
2
+
quantum-chemical calculations. It is a pity that Ni sensing
mechanisms have never been reported and that there is only
Fluorescent chemosensors have several advantages over other
methods due to their sensitivity, specificity, and ability to be
used in real-time monitoring with fast response. Therefore,
research on metal-ion-selective fluorescent chemosensors has
2+
1
one example about sensing mechanisms of Cu with the above
8
two methods.
In this work, we synthesized a new 1,8-naphthyridine-derived
compound, 7-acetamidyl-4-methyl-2-(1H-benzo[d]imidazolyl)-
1,8-naphthyridine (A), which is highly fluorescent and
responsive to Ni2+ and Cu with excellent selectivity in the
presence of excess amounts of various competing metal ions.
We found that addition of ethylenediamine (EDA) causes a
attracted great attention from chemical scientists, and great
2
2+
achievements have been obtained. Ni is a significant
environmental pollutant yet an essential trace element in
biological systems. For example, loss of nickel homeostasis i3s
harmful to prokaryotic and eukaryotic organisms alike.
2+
2
+
Nevertheless, reports of Ni fluorescent probes are very
4
2+
rare. Copper also plays an important role in various biological
fluorescence revival of A only in the case of Cu , providing an
2+
2+
processes and is also a significant metal pollutant because of its
effective method for distinguishing between Ni and Cu . The
fluorescence quenching mechanism was studied by picosecond
time-resolved fluorescence (TRF) spectroscopy and density
functional theory (DFT) calculations. The application of A to
5
widespread use. For instance, exposure to a high level of
copper even for a short period of time can cause gastro-
intestinal disturbance, and long-time exposure can cause liver or
6
2+
kidney damage. Many fluorescent Cu chemosensors have
2+
2+
the detection of intracellular Ni and Cu was also achieved.
been reported and used with some success in biological
7
applications. However, some of them have shortcomings for
EXPERIMENTAL SECTION
practical applications, such as a low fluorescence quantum yield,
■
low water solubility, and a narrow pH range. Therefore, it is
Methods and Materials. All of the chemicals were purchased
from commercial suppliers and used without further purification. All of
the reactions were performed under an argon atmosphere using
solvents purified by standard methods. H and C NMR spectra were
recorded on a Bruker 400 NMR spectrometer. Chemical shifts are
reported in parts per million using tetramethylsilane (TMS) as the
2
+
2+
quite essential to design and synthesize Ni /Cu -selective
fluorescent chemosensors that have high fluorescence quantum
yields, excellent water solubility, and a wide pH range. To
develop a novel efficient chemosensor, it is necessary to
understand the sensing mechanism. It is well-known that
fluorescence is a property of electronically excited states and
that the lifetime of the fluorescence decay is a key factor for the
mechanism. Therefore, metal ion identification mechanisms can
1
13
Received: August 23, 2012
Published: October 30, 2012
©
2012 American Chemical Society
12444
dx.doi.org/10.1021/ic3018453 | Inorg. Chem. 2012, 51, 12444−12449

Inorganic Chemistry
Article
2
+
2+
internal standard. Mass spectra were obtained on a Micromass GCT-
MS mass spectrometer.
Scheme 2. Binding Modes of A−Ni and A−Cu Obtained
at the B3LYP/6-31G(d) Level
Absorption and Fluorescence Spectroscopy. All spectral
characterizations were carried out in HPLC-grade solvents at 20 °C
within a 10 mm quartz cell. UV−vis absorption spectra were measured
with a TU-1901 double-beam UV−vis spectrophotometer, and
fluorescence spectra were determined on a Hitachi F-4500
spectrometer. The fluorescence quantum yield (Φ ) was measured
fl
at room temperature with excitation at 350 nm (Xe lamp in the F-4500
spectrometer) using quinine bisulfate in 1 M H SO (Φ = 0.546) as
2
4
fl
9
the reference.
Time-Resolved Fluorescence Spectroscopy. TRF spectra were
measured on a LifeSpec picosecond TRF spectrometer (Edinburgh
−5
Instruments Ltd.). The sample concentration was set to 1.25 × 10
M.
DFT Calculations. The charge distributions of compounds A, B,
and C were calculated with the 6-31G** basis set using the natural
bond orbital (NBO) model. DFT calculations using the Becke three-
parameter exchange/Lee−Yang−Parr correlation hybrid functional
7
-Amidyl-2,4-dimethyl-1,8-naphthyridine. 7-Amidyl-2,4-dimethyl-
(
B3LYP) with 6-31G(d) and 6-31G**+LanLZDZ basis sets as
1
,8-naphthyridine was synthesized and purified from 2,6-diaminopyr-
idine according to a modified version of the literature procedure.
Yield: 35%. H NMR (400 MHz, DMSO-d , TMS): δ 8.02 (d, 1H, J
8.8 Hz), 6.88 (s, 1H), 6.76 (d, 1H, J = 8.8 Hz), 6.66 (s, 2H), 2.51 (s,
implemented in the Gaussian 09 suite of programs were carried out
10b,c
2
+
2+
for the geometry optimizations of A, A−Cu and A−Ni . The
energies and oscillator strengths of the 80−120 lowest-energy
electronic transitions, which involve calculations of the singlet
excited-state energies of the complexes, were obtained using time-
dependent DFT (TD-DFT) with the same basis sets.
1
6
H
=
3
1
13
H), 2.47 (s, 3H). C NMR (100 MHz, DMSO-d ): δ 160.82,
6 C
60.25, 156.86, 1, 145.02, 134.28, 118.95 114.47, 111.97, 25.27, 17.90.
-Acetamidyl-2,4-dimethyl-1,8-naphthyridine. 7-Acetamino-2-
7
Synthesis and Characterization. 7-Amidyl-2,4-dimethyl-1,8-
naphthyridine, 7-acetamidyl-2,4-dimethyl-1,8-naphthyridine, and 7-
acetamidyl-4-methyl-1,8-naphthyridine-2-aldehyde were synthesized
methyl-1,8-naphthyridine was synthesized from 7-amidyl-2,4-dimeth-
yl-1,8-naphthyridine according to a modified version of the literature
method.
10.95 (s, 1H) 8.48 (d, 1H, J = 8.8 Hz), 8.30 (d, 1H, J = 8.8 Hz), 7.24
(s, 1H), 2.61 (s, 3H), 2.51 (s, 3H), 2.17 (s, 3H). C NMR (100 MHz,
DMSO-d ): δ 170.51, 162.43, 155.03, 154.15, 145.83, 136.31, 122.15,
1
0d,e
1
Yield: 84%. H NMR (400 MHz, DMSO-d , TMS): δ
6 H
10
according to the reported procedures. According to modified
11
13
versions of the reported methods, 7-acetamidyl-4-methyl-2-(1H-
benzo[d]imidazolyl)-1,8-naphthyridine (A), 7-acetamidyl-4-methyl-2-
6
C
(
2
benzo[d]oxazolyl)-1,8-naphthyridine (B), and 7-acetamidyl-4-methyl-
-(benzo[d]thiazolyl)-1,8-naphthyridine (C) were synthesized
118.28, 113.72, 25.61, 24.62, 18.01.
7-Acetamidyl-4-methyl-1,8-naphthyridine-2-aldehyde. 7-Acet-
amidyl-2,4-dimethyl-1,8-naphthyridine (2 g, 0.0093 mol) and selenium
dioxide (1.3 g, 0.0117 mol) were added to 100 mL of 1,4-dioxane, and
the reaction mixture was refluxed for 4 h with stirring under nitrogen.
The mixture was filtered while hot, and the crude product was
obtained by concentration in vacuum. The final product was obtained
by column chromatography over silica gel using 50:1 dichloro-
2
+
2+
(
Scheme 1). The binding modes of A−Ni and A−Cu are
illustrated in Scheme 2.
Scheme 1. Synthetic Pathways to Compounds A, B, and C
1
methane/ethanol as the eluent (0.86 g, 43% yield). H NMR (400
MHz, DMSO-d , TMS): δ 11.19 (s, 1H), 10.05 (s, 1H), 8.63 (d, 1H,
6
H
J = 9.2 Hz), 8.48 (d, 1H, J = 9.2 Hz), 7.80 (s, 1H), 2.75 (s, 3H), 2.21
(
s, 3H). 13C NMR (100 MHz, DMSO-d ): δ 194.57, 170.85, 155.51,
6 C
1
54.92, 154.48, 148.85, 136.84, 122.70, 117.30, 116.89, 24.70, 18.42.
-Acetamidyl-4-methyl-2-(1H-benzo[d]imidazolyl)-1,8-naphthyr-
idine (A). 7-Acetamidyl-4-methyl-1,8-naphthyridine-2-aldehyde
0.2130 g, 0.93 mmol) was added to an ethanol solution of
phenylenediamine (0.1807 g, 1.66 mmol) and NaHSO (0.2244 g,
7
(
3
2
.16 mmol). The above mixture was refluxed for 1.5 h under an argon
atmosphere, and the crude product was obtained under vacuum. The
final product was obtained by column chromatography (200−300
mesh, 60:1 ethyl acetate/CH OH) (0.2318 g, 78.6% yield).
3
Decomposition temperature: 230 °C. Characterization of A: HRMS
+
(
EI) m/z: calcd for C H N O [M + H] , 318.1277; found, 318.1354.
18
15
5
1
H NMR (400 MHz, DMSO-d , TMS): δH 13.15 (s, 1H), 10.76 (s,
6
1
H), 8.57 (d, 1H), 8.36 (d, 1H), 8.31 (s, 1H), 7.67 (d, 2H), 7.26 (m,
2
1
1
H), 2.75 (s, 3H), 2.22 (s, 3H). 13C NMR (100 MHz, DMSO-d ): δ
6
C
70.69, 154.98, 154.84, 154.73, 151.46, 150.98, 147.73, 136.60, 136.58,
20.74, 119.42, 115.31, 24.78, 18.40.
7
-Acetamidyl-4-methyl-2-(benzo[d]oxazolyl)-1,8-naphthyridine
(
B). 2-Aminophenol (0.1726 g, 1.58 mmol) and 7-acetamidyl-4-
methyl-1,8-naphthyridine-2-aldehyde (0.3496 g, 1.53 mmol) were
added to a 25 mL two-neck flask containing xylenes (15 mL; mixture
of m-, o-, and p-xylene). After the mixture was allowed to react for 0.5
h at 120 °C, 2,2,6,6-tetramethylpiperidine N-oxyl (TEMPO) (4.3 mg)
was added. The solution was then stirred for 8 h at 120 °C, and the
crude product was obtained under vacuum. The final product was
obtained by column chromatography (200−300 mesh, 5:3 ethyl
1
2445
dx.doi.org/10.1021/ic3018453 | Inorg. Chem. 2012, 51, 12444−12449

Inorganic Chemistry
Article
acetate/light petroleum) (0.3388 g, 71% yield). Mp = 206−208 °C.
changes were observed in the emission spectrum in the
+
2+
+
2+
3+
2+
3+
2+
2+
2+
Characterization of B: HRMS (EI) m/z: calcd for C H N O [M] ,
1
8
14
4
2
presence of Mg , K , Ca , Cr , Mn , Fe , Zn , Cd , Hg
1
3
9
7
18.1117; found, 318.1122. H NMR (400 MHz, CDCl , TMS): δ
2+
3
H
or Pb (light bars in Figure 2 and Figure S5). The left panel of
.32 (s, 1H), 8.63 (d, 1H), 8.43 (d, 1H), 8.32 (s, 1H), 7.86 (d, 1H),
.70 (d, 1H), 7.44 (m, 2H), 2.81 (s, 3H), 2.37 (s, 3H). C NMR (100
13
MHz, CDCl ): δ 169.99, 161.32, 154.45, 151.36, 148.39, 147.32,
3
C
1
1
41.78, 135.63, 126.55, 125.05, 121.46, 120.73, 120.73, 120.58, 116.13,
11.51.
7
-Acetamidyl-4-methyl-2-(benzo[d]thiazolyl)-1,8-naphthyridine
(C). The synthetic procedure for C was similar to that for B, and the
yield was 67%. Decomposition temperature: 223 °C. Characterization
+
of C: HRMS (EI) m/z: calcd for C H N OS [M] , 334.0888; found,
3
18
14
4
1
34.0893. H NMR (400 MHz, CDCl , TMS): δ 9.62 (s, 1H), 8.62
3 H
(
(
1
1
d, 1H), 8.42 (d, 1H), 8.36 (s, 1H), 8.12 (d, 1H), 8.00 (d, 1H), 7.51
m, 2H), 2.80 (s, 3H), 2.32 (s, 3H). 13C NMR (100 MHz, CDCl ): δC
3
69.94, 154.39, 154.33, 154.24, 153.95, 141.14, 136.70, 135.93, 126.41,
26.13, 123.84, 122.20, 121.65, 118.64, 115.62, 25.00, 18.40.
RESULTS AND DISCUSSION
Absorption and Fluorescence Properties of Com-
pounds A, B, and C. The absorption spectra of compounds A,
■
Figure 2. Fluorescence responses of A [1.25 × 10− M in 3:1 (v/v)
5
0
.05 M HEPES (pH 7.4)/C H OH] upon addition of different metal
2 5
salts (3 equiv of salt ion relative to A) (light bars) and fluorescence
2+
changes of the mixture of A and Ni after addition of an excess of the
indicated metal ions (12 equiv relative to A) (dark bars). The
excitation wavelength was 370 nm. For the light gray bars, I and I
0
represent the emission intensities at 433 nm in the fluorescence
spectra of compound A before and after addition of the metal ion to
the solution of A; for the dark bars, I represents the emission intensity
0
at 433 nm in the fluorescence spectrum of compound A, and I
represents the intensity in the fluorescence spectrum of the mixture of
A and Ni2 after addition of an excess of the metal ion. The metal salts
used were MgCl , KCl, CaCl , CrCl , MnCl , FeCl , NiCl , CuCl ,
+
2
2
3
2
3
2
2
ZnCl , CdCl , HgCl , and PbCl .
2
2
2
2
electron-pushing benzo[d]oxazolyl unit.
2
+
2+
Effect of Ni /Cu on the Fluorescence Spectra of A.
2
+
2+
−5
Addition of 2 equiv of NiCl /CuCl (Ni /Cu , 2.5 × 10 M)
2
2
to an aqueous solution of A [3:1 (v/v) 0.05 M HEPES (pH
7
5
.4)/C H OH] produced a new absorption band in the 400−
2
5
00 nm range and a 10 nm red shift of the absorption
maximum (Figure 1 left and Figure S3), while the fluorescence
Figure 3. (left) Plot of emission intensity of compound A at 432 nm
2
+
(
black ■) vs the reciprocal of the concentration of added Ni . (right)
−5
Variation of the fluorescence spectrum of A [1.25 × 10 M in 3:1 (v/
v) 0.05 M HEPES (pH 7.4)/C H OH] with (red ●) and without 1.92
2
5
equiv of Ni2 ion (black ■) as a function of pH. The fluorescence
+
intensity was measured at 432 nm.
Figure 1. (left) Absorption and (right) emission spectral changes for
compound A in aqueous solution [1.25 × 10−5 M in 3:1 (v/v) 0.05 M
HEPES (pH 7.4)/C H OH] upon addition of NiCl . In (b), the
2
5
2
excitation wavelength was 370 nm.
Figure 3 demonstrates the relationship between the concen-
tration of Ni2+ and the emission intensity at the special
wavelength. The detection limit of the fluorescence response of
was almost completely quenched (Φ changed from 0.252 to
fl
2+
2+
2+
−6
2+
0
.089 and 0.060 with I/I = 0.044 and 0.039 for Ni and Cu ,
the sensor to Ni was 6.42 × 10 M (Figure 3 left). For Cu ,
6
0
−
respectively) with no change in the emission wavelength
the detection limit was 7.09 × 10 M (Figure S6), which is
2+
2+
2+
(
Figure 1 right and Figure S4), indicating efficient Ni /Cu -
satisfactory for the detection of Cu in drinking water within
selective on−off behavior.
U.S. EPA limit (≈20 μM). Job’s plots indicated that A chelates
2+
2+
2+
2+
Selectivity Study of the Detection of Ni /Cu with A.
Ni /Cu with 2:1 stoichiometry, and the binding constants of
2
+
2+
9
Upon the addition of equal amounts of various metal ions (2.5
A with Ni and Cu were calculated to be 2.84 × 10 and 5.74
−
5
2+
2+
7
×
10 M), only Ni and Cu decreased the emission of A;
× 10 , respectively (Figures S7 and S8). We also found a
2+
2+
under identical conditions, nearly no fluorescence intensity
remarkable selectivity of A for Ni /Cu ion over various other
1
2446
dx.doi.org/10.1021/ic3018453 | Inorg. Chem. 2012, 51, 12444−12449

Inorganic Chemistry
Article
metal ions, which turned out to be applicable in environmental
technology (dark bars in Figure 2 and Figure S5).
2
+
2+
Effect of Ni /Cu on the Fluorescence Spectra of B
and C. To understand the crucial role of the 1H-benzo[d]-
imidazolyl unit, which behaves as an additional binding site for
2+
2+
the Ni /Cu ions with the aid of the naphthyridine group, we
synthesized B and C bearing benzo[d]oxazolyl and benzo[d]-
thiazolyl units, respectively, on the 7-acetamidyl-4-methyl-1,8-
naphthyridine. Subsequently, B and C were tested for
2
+
2+
fluorescence changes upon the addition of Ni and Cu
ions as well. As shown in Figure 4 and Figures S9−S12, unlike
Figure 5. (top) Structures of A, B and C, including the labeling of the
N atoms. (bottom) Table summarizing the charge densities of selected
nitrogen and oxygen atoms.
Table 1. Fluorescence Decay Time Constants of A in the
2
+
Presence of Ni Ions
Figure 4. Relative fluorescence responses of A, B, and C [1.25 × 10−5
equiv of Ni2+
A1
τ1 (ns)
A2
τ2 (ns)
χ2
M in 3:1 (v/v) 0.05 M HEPES (pH 7.4)/C H OH] at 433, 400, and
2
5
2+
2+
4
08 nm, respectively, without Ni /Cu (red bars) and after addition
0
14%
16%
14%
13%
1.12
0.92
0.81
0.79
86%
84%
86%
87%
2.30
2.20
2.17
2.18
1.059
1.010
0.971
1.131
of 3 equiv of NiCl (green bars) or CuCl (blue bars). Excitation was
0.5
1
2
2
at 370 nm.
2
A, neither B nor C showed any distinct absorption or
2+
fluorescence spectral changes upon the addition of Ni or
the solution, the lifetime was almost unchanged under the
experimental conditions (Table 1 or Table S1, respectively).
From the combined observations of the absorption and
emission titration experiments, a static quenching mechanism
can be deduced. The fluorophore A ligates Ni or Cu ,
forming a nonfluorescent complex in the ground state. The
fluorescence quenching of A may result from excitated-state
energy transfer or charge transfer.
2+
Cu ions. This strongly supports the conclusion that having a
nitrogen atom at the 1-position in the substituent at the 2-
position of the 7-acetamidyl-4-methyl-1,8-naphthyridine core
2
+
2+
2+
2+
plays an important role in the Ni /Cu complexation.
Computational Study of the Coordinating Abilities of
A, B, and C. To interpret further the influence of the 1H-
benzo[d]imidazolyl unit in the reaction of A with metal ions,
theoretical calculations were performed on compounds A, B,
and C. From Figure 5 and Figures S13−S15, it is obvious that
the negative charge density of N4 (−0.412) in the 1H-
benzo[d]imidazolyl unit is higher than those of the benzo[d]-
oxazolyl (−0.389) and benzo[d]thiazolyl (−0.364) units,
indicating a stronger coordination ability for N4 in the 1H-
benzo[d]imidazolyl unit in A than for the corresponding units
TD-DFT Calculations of Absorption Properties. To
2
+
interpret further the absorption properties of A and the A−Ni
2
+
and A−Cu complexes, TD-DFT calculations were performed
on this system. In compound A, the highest occupied molecular
orbital (HOMO) is mainly located on the 1H-benzo[d]-
imidazolyl unit, and the lowest unoccupied molecular orbital
(LUMO) is on the naphthyridine group (Figure S24). In A−
2
+
2+
2+
2+
in B and C. Considering the Ni /Cu -induced absorption and
emission spectral changes for A, B, and C and the similar
negative charge densities of N1, N2, N3 and O1 in A, B, and C,
it is easy to conclude that the coordinating atoms should be N3
and N4 (Scheme 2).
Ni , the HOMO and LUMO are located on Ni and ligand A,
2
+
respectively (Figure S25). In A−Cu , both the HOMO and
LUMO are mainly located on A (Figure S26). As shown in
Table S2, the lowest-energy transition of A comes from the
HOMO−1 → LUMO and HOMO → LUMO orbital
transitions. As shown in Table 2, the lowest-energy transition
Study of the Fluorescence Decay Behavior. As is well-
2
+
known, heavy metal ions tend to quench the luminescence
of A−Ni comes from the HOMO−4 → LUMO, HOMO−5
8
,12
through electron- and/or energy-transfer processes. To gain
→ LUMO, HOMO → LUMO+12, HOMO−4 → LUMO+1,
2
+
2+
2+
insight into the quenching behavior in the A−Ni and A−Cu
and HOMO−10 → LUMO orbital transitions; for A−Cu , the
complexes, we measured picosecond TRF spectra. The
lowest-energy transition comes from the HOMO →
LUMO+12 and HOMO−5 → LUMO orbital transitions.
The calculated absorption-peak positions are in good agree-
ment with the experimental results (Figures S27−S29), which
proves that intramolecular charge-transfer processes occur in
2+
2+
fluorescence decay behavior in the presence of Ni /Cu is
shown in Figures S16 to S23, and the double-exponential fit
results are summarized in Table 1 and Table S1.
2
+
2+
In the absence of Ni /Cu , the fluorescence decayed in a
2
+
2+
double-exponential manner with time constants of 1.12 and
the excited states of A, A−Ni , and A−Cu .
2
.30 ns, indicating that the fluorescence spectra consist of two
TD-DFT Calculations of the Fluorescence Quenching
Mechanism. To elaborate upon the fluorescence quenching
2
+
2+
contributions from free A. When Ni or Cu was titrated into
1
2447
dx.doi.org/10.1021/ic3018453 | Inorg. Chem. 2012, 51, 12444−12449

Inorganic Chemistry
Article
Table 2. Contributions of Various Orbital Transitions to the
Lowest-Energy Transitions in A−Ni and A−Cu
the LMCT contribution is ∼49.5%, energy transfer may partly
contribute to the nonradiative deactivation of the excited state.
Study of the Reversibility of the Binding of Chemo-
2
+
2+
oscillator strength
2+
2+
sensor A to Ni and Cu . To examine the reversibility of
2
+
2+
2+
2+
orbital transition
A−Ni
A−Cu
the binding of chemosensor A to Ni and Cu , an ethanol
2
+
HOMO−10 → LUMO
HOMO−5 → LUMO
HOMO−4 → LUMO
HOMO → LUMO+12
HOMO−4 → LUMO+1
0.1313
0.2237
0.1009
0.1436
0.2893
solution containing 64 equiv of EDA was added to the A−Ni
2+
0.4151
0.4949
and A−Cu solutions, respectively. No enhanced emission
2+
intensity was observed when EDA was added to the A−Ni
solution (Figure 7 left). This indicates that EDA is not able to
mechanism, we also carried out TD-DFT calculations (Figure
, Figures S25 and S26, and Table 2). It is noted that the
6
Figure 7. (left) Relative fluorescence responses of the complexes A−
2
+
2+
−5
Ni and A−Cu [1.25 × 10 M in 3:1 (v/v) 0.05 M HEPES (pH
7
.4)/C H OH] before (light bars) and after (dark bars) the addition
2 5
of EDA. Excitation was at 370 nm, and emission was monitored at 432
nm. (right) Fluorescence emission spectra of A [1.25 × 10− M in 3:1
5
(
v/v) 0.05 M HEPES (pH 7.4)/C H OH] upon addition of different
2 5
−5
nickel salts (2.4 × 10 M) with excitation at 370 nm.
remove Ni2+ ions from the A−Ni complex and hence cannot
2+
restore the emission of A. However, when EDA was added to
2
+
the A−Cu solution, fluorescence signals identical to those of
A were restored (Figure 7 left and Figure S30), demonstrating
that Cu2+ is removed from the A−Cu complex by EDA. That
2+
2+
is, the binding of A and Cu is really chemically reversible. In
other words, EDA showed high selectivity toward the in situ-
prepared Cu2+ complex over the Ni complex, which can be
2+
2
+
2+
applied to distinguish Ni and Cu .
2
+
2+
Counterion Effect on the Ni /Cu -Selective Proper-
ties of A. Experiments to explore the counterion effect on the
2
+
2+
Ni /Cu -selective properties of A were also performed
Figure 7 right and Figure S31). Acetate, sulfate, nitrate, and
2
+
2+
Figure 6. Frontier molecular orbitals of A−Ni and A−Cu relevant
to the fluorescence quenching.
(
perchlorate counteranions had similar influence as chloride
when it comes to Ni2+ and Cu salts, demonstrating that the
anions do not coordinate the metal ions while compound A
2+
fluorescence quenching by Ni2+ can be rationalized in terms of
the occupancy of the frontier orbitals. The HOMO →
LUMO+12 excitation was found to be related to metal-to-
ligand charge transfer (MLCT), and its contribution to the
lowest-energy excitation was 14.4%. Such excitation corre-
sponds to charge transfer from the excited Ni center to the
naphthyridine and 1H-benzo[d]imidazolyl moieties and thus
provides a pathway for nonradiative deactivation of the excited
state. On the basis of the calculations, the MLCT contribution
is ∼14.4%. Thus, considering the results of the TRF
experiments, energy transfer may also contribute to the
nonradiative deactivation of the excited state to a significant
extent. As for the fluorescence quenching by Cu , different
calculation results were obtained (Figure 6 and Figure S26).
Although the HOMO → LUMO+12 excitations were also
found to be relevant for the fluorescence quenching by Cu ,
such excitation corresponds to ligand-to-metal charge transfer
reacts with Ni or Cu . This indicates that compound A as a
fluorescent sensor has a much wider application range in
sensing Ni2+ and Cu ions.
2+
2+
2+
pH Range of Application of A toward Ni and Cu2+.
For both environmental and biological applications of the
fluorescent sensor, it would be much better if the sensing works
over a wide range of pH. The right panel of Figure 3 shows that
2+
2
+
2
+
in aqueous solution the suitable pH range for Ni
determination is 4−10, where the fluorescence on−off behavior
2+
can be operated by Ni binding. Consequently, our present
2+
Ni -selective receptor would be an ideal fluorometric chemo-
2+
sensor for monitoring Ni . Similar results were obtained for
2+
2+
Cu (Figure S32).
2
+
2+
In Situ Imaging and Intracellular Ni /Cu Response.
It is essential for biosensing molecules to monitor guest species
2+
13
2+
selectively in living cells for biological applications. The Ni /
2
+
Cu sensitivity of A was examined in living cells by using
confocal microscopy. Human lung adenocarcinoma epithelial
cells (A549) were selected for our experiments. Fluorescence
images were recorded with excitation at 408 nm by a diode
(
LMCT) from the excited naphthyridine and 1H-benzo[d]-
imidazolyl moieties to the Cu center and thus provides a
pathway for nonradiative deactivation of the excited state. Since
2
+
1
2448
dx.doi.org/10.1021/ic3018453 | Inorg. Chem. 2012, 51, 12444−12449

Inorganic Chemistry
Article
laser, spinhole aperture, 100% gain of detector, and an oil
objective with 60× magnification and 1.40 NA. The qualitative
in vitro results are shown in Figure 8 and Figure S33. After
REFERENCES
■
(
1) (a) Valeur, B.; Leray, I. Coord. Chem. Rev. 2000, 205, 3−40.
b) Pawley, J. B. Handbook of Biological Confocal Microscopy; Plenum
(
Press: New York, 1995. (c) Lichtman, J. W.; Conchello, J. A. Nat.
Methods 2005, 2, 910−919.
(2) (a) Xu, Z.; Yoon, J.; Spring, D. R. Chem. Soc. Rev. 2010, 39,
1
996−2006. (b) Kikuchi, K. Chem. Soc. Rev. 2010, 39, 2048−2053.
(c) Wu, J.; Liu, W.; Ge, J.; Zhang, H.; Wang, P. Chem. Soc. Rev. 2011,
4
6
(
0, 3483−3495. (d) Quang, D. T.; Kim, J. S. Chem. Rev. 2010, 110,
280−6301.
3) Nickel and Its Surprising Impact in Nature; Sigel, A., Sigel, H.,
Sigel, R. K. O., Eds.; Metal Ions in Life Sciences, Vol. 2; Wiley:
Chichester, U.K., 2007.
(4) (a) Banerjee, A.; Sahana, A.; Guha, S.; Lohar, S.; Hauli, I.;
Mukhopadhyay, S. K.; Matalobos, J. S.; Das, D. Inorg. Chem. 2012, 51,
Figure 8. In situ imaging and intracellular Ni2+ response of A-labeled
living cells at 37 °C. (a) A549 cells incubated with 0.01 μmol of
5
699−5704. (b) Li, G.; Fang, H.; Cai, Y.; Zhou, Z.; Thallapally, P. K.;
compound A and 1 mL of PBS for 30 min at 37 °C. (b) A549 cells
from (a) 10 min after treatment with 0.03 μmol of Ni2+ in aqueous
Tian, J. Inorg. Chem. 2010, 49, 7241−7243. (c) Dodani, S. C.; He, Q.;
Chang, C. J. J. Am. Chem. Soc. 2009, 131, 18020−18021.
solution. (c) Bright-field image of living A549 cells.
(
5) (a) Rurack, K. Spectrochim. Acta, Part A 2001, 57, 2161−2195.
(b) Kramer, R. Angew. Chem., Int. Ed. 1998, 37, 772−773.
(c) Muthaup, G.; Schlicksupp, A.; Hess, L.; Beher, D.; Ruppert, T.;
A549 cells were incubated with 0.01 μmol of A and 1 mL of
phosphate-buffered saline (PBS) for 30 min at 37 °C, very
strong fluorescence was imaged (Figure 8a and Figure S33a).
These A-containing A549 cells were then incubated with 0.03
Masters, C. L.; Beyreuther, K. Science 1996, 271, 1406−1409.
d) Løvstad, R. A. BioMetals 2004, 17, 111−113. (e) Barnham, K.
J.; Masters, C. L.; Bush, A. I. Nat. Rev. Drug Discovery 2004, 3, 205−
14.
6) Georgopoulos, P. G.; Roy, A.; Yonone-Lioy, M. J.; Opiekun, R.
E.; Lioy, P. J. J. Toxicol. Environ. Health, Part B 2001, 4, 341−394.
7) (a) Li, Z.; Zhang, L.; Wang, L.; Guo, Y.; Cai, L.; Yu, M.; Wei, L.
(
2
(
2
+
2+
μmol of Ni /Cu for another 10 min. No obvious
fluorescence was imaged (Figure 8b and Figure S33b) because
2
+
2+
of the quenching of the emission of A by Ni /Cu .
(
Chem. Commun. 2011, 47, 5798−5800. (b) Lau, Y. H.; Price, J. R.;
CONCLUSIONS
A highly fluorescent 1,8-naphthyridine-derived chemosensor
with high sensitivity and selectivity toward Ni /Cu over
other cations both in aqueous solution over a wide pH range
4−10) and in a cellular environment was developed.
Counteranions such as acetate, sulfate, nitrate, and perchlorate
have no influence on the detection of such metal ions. EDA
showed high selectivity toward the in situ-prepared Cu
Todd, M. H.; Rutledge, P. J. Chem.Eur. J. 2011, 17, 2850−2858.
■
(c) Zhao, Y.; Zhang, X.; Han, Z.; Qiao, L.; Li, C.; Jian, L.; Shen, G.;
2
+
2+
Yu, R. Anal. Chem. 2009, 81, 7022−7029. (d) Domaille, D. W.; Zeng,
L.; Chang, C. J. J. Am. Chem. Soc. 2010, 132, 1194−1195. (e) Kumar,
M.; Dhir, A.; Bhalla, V. Org. Lett. 2009, 11, 2567−2569. (f) Jung, H. S.;
Park, M.; Han, D. Y.; Kim, E.; Lee, C.; Ham, S.; Kim, J. S. Org. Lett.
(
2
009, 11, 3378−3381. (g) Kim, H. J.; Hong, J.; Hong, A.; Ham, S.;
Lee, J. H.; Kim, J. S. Org. Lett. 2008, 10, 1963−1966. (h) Zhou, Y.;
Wang, F.; Kim, Y.; Kim, S.; Yoon, J. Org. Lett. 2009, 11, 4442−4445.
(i) Yu, M.; Shi, M.; Chen, Z.; Li, F.; Li, X.; Gao, Y.; Xu, J.; Yang, H.;
Zhou, Z.; Yi, T.; Huang, C. Chem.Eur. J. 2008, 14, 6892−6900.
(j) Zhao, Q.; Li, F.; Huang, C. Chem. Soc. Rev. 2010, 39, 3007−3030.
2
+
2
+
complex over the Ni complex, which can be applied to
distinguish Ni2 and Cu . The fluorescence on−off mechanism
was revealed to be mediated by intramolecular charge transfer,
as confirmed by picosecond time-resolved fluorescence spec-
troscopy and time-dependent density functional theory
calculations.
+
2+
(
5
k) Yu, M.; Li, Z.; Wei, L.; Wei, D.; Tang, M. Org. Lett. 2008, 10,
115−5118.
8) Jung, H. S.; Kwon, P. S.; Lee, J. W.; Kim, J. I.; Hong, C. S.; Kim, J.
W.; Yan, S.; Lee, J. Y.; Lee, J. H.; Joo, T.; Kim, J. S. J. Am. Chem. Soc.
009, 131, 2008−2012.
(9) Demas, J. N.; Grosby, G. A. J. Phys. Chem. 1971, 75, 991−1024.
10) (a) Li, Z.; Yu, M.; Zhang, L.; Yu, M.; Liu, J.; Wei, L.; Zhang, H.
(
2
ASSOCIATED CONTENT
Supporting Information
Absorbance and fluorescence spectra of A, B, and C in their
metal-free and metal-bound forms, Job’s plot, TRF spectra, and
calculation details for A in the presence of Ni /Cu . This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
*
S
(
Chem. Commun. 2010, 46, 7169−7171. (b) Henry, R. A.; Hammond,
P. R. J. Heterocycl. Chem. 1977, 14, 1109−1114. (c) Fan, N. Y. Organic
Synthesis; Beijing University of Technology Press: Beijing, China,
2
+
2+
1
1
995; p 183. (d) Kelly, T. R.; Zhao, C.; Bridger, G. J. J. Am. Chem. Soc.
989, 111, 3744−3745. (e) Kelly, T. R.; Bridger, G. J.; Zhao, C. J. Am.
Chem. Soc. 1990, 112, 8024−8034. (f) Li, Z.; Fu, W.; Yu, M.; Zhao, X.;
Chen, Y. Dyes Pigm. 2007, 75, 516−520.
AUTHOR INFORMATION
Corresponding Author
E-mail: lizx@zzu.edu.cn (Z.L.), yumm@zzu.edu.cn (M.Y.),
zhanghongyan@mail.ipc.ac.cn (H.Z.). Tel: +86-371-67781205.
Fax: +86-731-67781205.
■
*
(
11) (a) Chen, Y. X.; Qian, L. F.; Zhang, W.; Han, B. Angew. Chem.,
Int. Ed. 2008, 47, 9330−9333. (b) Liu, Z.; Zhang, C.; Li, Y.; Wu, Z.;
Qian, F.; Yang, X.; He, W.; Gao, X.; Guo, Z. Org. Lett. 2009, 11, 795−
798.
(12) (a) Kavallieratos, K.; Rosenberg, J. M.; Chen, W. Z.; Ren, T. J.
Am. Chem. Soc. 2005, 127, 6514−6515. (b) Zheng, Y.; Orbulescu, J.; Ji,
X.; Andreopoulos, F. M.; Pham, S. M.; Leblanc, R. M. J. Am. Chem.
Soc. 2003, 125, 2680−2686.
Notes
The authors declare no competing financial interest.
(13) (a) Chang, M. C. Y.; Pralle, A.; Isacoff, E. Y.; Chang, C. J. J. Am.
ACKNOWLEDGMENTS
■
Chem. Soc. 2004, 126, 15392−15393. (b) Peng, X.; Du, J.; Fan, J.;
Wang, J.; Wu, Y.; Zhao, J.; Sun, S.; Xu, T. J. Am. Chem. Soc. 2007, 129,
1500−1501.
This work was supported by the National Natural Science
Foundation of China (50903075, 60978034, and 50873093).
We are grateful to Professor Chenjie Fang at Capital Medical
University for her great help with our manuscript.
1
2449
dx.doi.org/10.1021/ic3018453 | Inorg. Chem. 2012, 51, 12444−12449