Cu2 fluorescent sensor based on mesoporous silica nanosphere

Article abstract of DOI:10.1016/j.dyepig.2012.01.004

Monodisperse mesoporous silica nanosphere (MSN) modified with anthracene derivative, 2-(3-(triethoxysilyl) propylamino)-N-(9-anthryl methyl) acetamide (SGAAn) has been fabricated as a fluorescent sensor (SGAAn-MSN) for the detection of metal ions, which shows an exclusive selectivity to Cu². Compared with SGAAn, SGAAn-MSN presents a higher sensitivity to Cu². The loading amount of SGAAn in MSN has a great influence on the detection sensitivity of SGAAn-MSN, which varies with the local concentration and accessibility of SGAAn in the pore channel of MSN. Interference studies indicate that the addition of other metal ions such as Ag⁺, Cd² and Pb² has a negligible effect on the selectivity of SGAAn-MSN to Cu². The possible mechanism for the sensing behavior of SGAAn-MSN to Cu² is proposed based on the relation between fluorescence quench degree and Cu² concentration.

Full text of DOI:10.1016/j.dyepig.2012.01.004

Dyes and Pigments 94 (2012) 239e246
Contents lists available at SciVerse ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Cu2þ fluorescent sensor based on mesoporous silica nanosphere
*
*
Deli Lu, Juying Lei, Zhidan Tian, Lingzhi Wang , Jinlong Zhang
Key Lab for Advanced Materials and Institute of Fine Chemicals, East China University of Science and Technology, 130 Meilong Road, Shanghai 200237, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Monodisperse mesoporous silica nanosphere (MSN) modified with anthracene derivative, 2-(3-(trie-
thoxysilyl) propylamino)-N-(9-anthryl methyl) acetamide (SGAAn) has been fabricated as a fluorescent
Received 8 November 2011
Received in revised form
6 January 2012
Accepted 10 January 2012
Available online 13 January 2012
sensor (SGAAn-MSN) for the detection of metal ions, which shows an exclusive selectivity to Cu2þ
.
Compared with SGAAn, SGAAn-MSN presents a higher sensitivity to Cu2þ. The loading amount of SGAAn
in MSN has a great influence on the detection sensitivity of SGAAn-MSN, which varies with the local
concentration and accessibility of SGAAn in the pore channel of MSN. Interference studies indicate that
the addition of other metal ions such as Agþ, Cd2þ and Pb2þ has a negligible effect on the selectivity of
SGAAn-MSN to Cu2þ. The possible mechanism for the sensing behavior of SGAAn-MSN to Cu2þ is
proposed based on the relation between fluorescence quench degree and Cu2þ concentration.
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Mesoporous silica nanosphere
Anthracene
Cu2þ
Sensitivity
Selectivity
Reversibility
1. Introduction
be bound by chelate groups composed of N, O and S [9]. Fluorescent
sensors generally show good selectivities to metal ions, which are
The pollution caused by heavy metal ions is extremely harmful
to the environment and human health [1]. Heavy metal ions such as
Pb2þ, Hg2þ and Agþ should not even exist in human beings at all,
which may cause great poisonous effect even with trace amount
[2,3]. Others like Zn2þ, Fe3þ and Cu2þ are involved in the bioactivity
of human beings, but may also cause toxicity when exceeding the
limit [4e6].
Heavy metal ions entering into the river and soil caused by the
exploitation and smelting may gradually accumulate in human
beings after they are involved in bioactivities, so, it is very impor-
tant to detect the trace of heavy metal ions. Currently, the
commonly used techniques for metal ions detection include atomic
absorption spectrophotometry, X-ray fluorescence spectropho-
tometry, inductively coupled plasma atomic emission spectrometry
and electrochemical method [7,8]. Compared with the above
methods, fluorescence analysis is more popular due to its real-time
operation, efficiency and convenience. Moreover, the precise
information including ions valence or molecule structure can also
be read out through the variation of fluorescence spectra.
A metal ion fluorescent sensor generally composes of a recog-
nition site designed to bind the target ion and a readout system,
which is often a fluorophore. It is known many metal ions tend to
determined by binding affinity and size matching between chelate
groups and metal ions [10]. However, the detection sensitivity of
traditional fluorescent sensor for metal ions is often not high
enough to satisfy the increasing application requirement. Currently,
detection of tiny metal ions often needs an enriching procedure,
which is troublesome and inefficient. With the development of
nanoscience, the application of nanomaterials for metal ions
detection becomes more and more interesting and has caused
tremendous attention since this kind of composite shows a high
sensitivity to biomolecules, poisonous gas and heavy metal ions
[11e20]. However, fluorescent sensor for the detection of tiny metal
ions is still in great need due to the less effective performance of
most sensors in the new emerging application fields such as ions
sensing and imaging in biosystem.
Mesoporous silica material reported in 1992 is a good solid base
for the fabrication of organiceinorganic hybrid materials by
introducing dye molecules into the pore framework or the pore
channel due to its abundant surface hydroxyl groups and open pore
system [21]. The hybrid composite may present novel and inter-
esting properties compared with the corresponding dye molecules.
Moreover, due to its nontoxicity and biocompatibility, composite
based on mesoporous silica has good potential applications in
biosystem. However, the failure to obtain discrete or non-
aggregated silica nanoparticles has obstructed its application in the
biosystem since monodisperse mesoporous silica nanomaterial is
a more suitable alternative used as in vivo fluorescence imaging
* Corresponding authors. Tel./fax: þ86 21 64252062.
E-mail addresses: jlzhang@ecust.edu.cn (L. Wang), wlz@ecust.edu.cn (J. Zhang).
0143-7208/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.dyepig.2012.01.004

240
D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
medium, efficient ions remover and drug releasing medium
[22e24], which may present more powerful characteristics and
potential application than other nano solid platform when it is used
as a fluorescent sensor.
After 1 h, the solvent was removed under vacuum. The resultant
residue was dissolved in DMF (10 mL), and sodium azide (777 mg,
11.95 mmol) was added. The reaction mixture was heated for 1 h at
50 ꢀC, cooled to room temperature and diluted with 50 mL water.
After extraction with EA, the organic layer was washed with brine,
dried over MgSO4 and concentrated under vacuum. Purification of
the residue by silica gel chromatography (EA/PE ¼ 1/19) gives 2 as
Cu2þ is an essential element for human beings and a disorder of
Cu2þ metabolism is closely associated with many severe neuro-
logical diseases such as Menkes and Wilson disease, Alzheimer’s
disease, familial amyotrophic lateral sclerosis and prion disease
[25]. The maximum permissible limit of copper ion discharged in
water for most of countries around the world is 3.0 ppm. However,
the pollution situation caused by copper ion is still very severe and
it has been intensively studied as the detection target for many
fluorescent sensors [26e28]. This paper presents a novel Cu2þ
fluorescent sensor based on mesoporous silica nanosphere by using
anthracene as the readout unit. We chose anthracene and intro-
duced it into the pore channel of mesoporous silica nanosphere
since this kind of polycyclic aromatic hydrocarbon has high fluo-
rescence quantum quantity and stable fluorescence signal, which is
an ideal report unit for fluorescent sensor. It is known Cu2þ tends to
interact with peptide and heterocyclic ring by coordinating with N,
O or S atom [29], so we modified anthracene with groups composed
of amine and amide before it is anchored on the pore surface. The
selectivity and sensitivity of this composite to Cu2þ was carefully
studied by fluorescence spectra. The sensing mechanism of this
nanocomposite was proposed based on the relation between the
fluorescence intensity and Cu2þ concentration.
a yellow solid (1.669 g, 97% yield). 1H NMR (400 MHz CDCl3)
d 5.33
(s, 2H), 7.50 (t, 2H, J ¼ 8 Hz), 7.59 (t, 2H, J ¼ 8 Hz), 8.06 (d, 2H,
J ¼ 8.4 Hz), 8.28 (d, 2H, J ¼ 8.8 Hz), 8.50 (s, 1H); 13C NMR (100 MHz
CDCl3)
d 46.39, 123.54, 125.23, 126.87, 129.02, 129.33, 130.74,
131.41; EI found 233.1 [M]ꢁþ calculated for C15H11N3: 233.1.
2.2.2. Synthesis of 9-aminomethylanthracene (3)
The suspension of compound 2 (0.4 g, 1.93 mmol) and 10% Pd/C
(204 mg) in EA (20 mL) sealed in a bottle was charged with N2 for 3
times and H2 for 4 times, and then put under H2 for 4 h. The
completion of the reaction was determined by thin layer chroma-
tography (TLC). The reaction was filtered and washed with EA. The
filtrate was concentrated to give product 3 as a pale yellow solid
(391 mg, 98% yield). 1H NMR (400 MHz CDCl3)
d 1.69 (s, 2H), 4.82 (s,
2H), 7.47 (t, 2H, J ¼ 7.4 Hz), 7.55 (t, 2H, J ¼ 8 Hz), 8.02 (d, 2H,
J ¼ 8.4 Hz), 8.33 (d, 2H, J ¼ 8.8 Hz); 8.40 (s, 1H); 13C NMR (100 MHz
CDCl3)
d 38.24, 123.72, 125.01, 126.19, 126.97, 129.33, 131.72; EI
found 207.1 [M]ꢁþ calculated for C15H13N:207.1.
2.2.3. Synthesis of N-(9-anthylmethyl)-2-chloro acetamide (4)
Compound 3 (414 mg, 2 mmol) and triethylamide (0.3 mL) were
dissolved in dichloromethane (20 mL) at 0 ꢀC. Then chloroacetyl
chloride (0.17 mL) was added. The mixture was warmed to room
temperature and allowed to react overnight. The solvent was
evaporated under vacuum. The crude product was further purified
by silica gel chromatography (PE/EA ¼ 3). A yellow solid was ob-
2. Experimentals
2.1. Materials
Chemicals for reactions were purchased from SigmaeAldrich
and are of A.R. grade. Solvents used in silica gel chromatography
including ethyl acetate (EA) and petroleum ether (PE) were
purchased from Sinopharm Chemical Reagent Co. Ltd. NMR solvent
was purchased from J&K Chemical Ltd. All starting materials and
solvents were used as received without further purification.
tained (416 mg, Yield: 74%). 1H NMR (400 MHz CDCl3)
d 4.10 (s, 2H),
5.49 (d, 2H, J ¼ 5.2 Hz), 6.72 (s, 1H), 7.52 (t, 2H, J ¼ 7.2 Hz), 7.60 (t,
2H, J ¼ 8.0 Hz), 8.06 (d, 2H, J ¼ 8.4 Hz), 8.27 (d, 2H, J ¼ 9.2 Hz), 8.51
(s, 1H); 13C NMR (100 MHz CDCl3)
d 36.45, 42.50, 123.50, 125.30,
127.00, 128.62, 129.36, 130.43, 131.49; ESI found 284.1 [MþH]þ
calculated for C17H14ClNO: 283.08.
2.2. Preparation
2.2.4. Synthesis of 2-(3-(triethoxysilyl) propylamino)-N-(9-
anthrylmethyl) acetamide(5, SGAAn)
The synthesis of target compound 5, 2-(3-(triethoxysilyl) pro-
pylamino)-N-(9-anthrylmethyl) acetamide (SGAAn) was achieved
by the route outlined in Scheme 1.
A mixture of 4 (309 mg, 1.40 mmol) and K2CO3 (261 mg,
1.91 mmol) in CH3CN (10 mL) was added to a solution of 4 (360 mg,
1.27 mmol) in CH3CN (20 mL) and the mixture was refluxed over-
night under N2 protected. The solvent was evaporated under
vacuum and the crude product was purified by silica gel
2.2.1. Synthesis of 9-azidomethylanthracene (2)
Compound 1 (1.54 g, 7.40 mmol) was dissolved in CH2Cl2
(30 mL), and then SOCl2 (810 m
L, 11.10 mmol) was added at 0 ꢀC.
Scheme 1. Synthetic route for SGAAn.

D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
241
chromatography (EA) to give 5 as yellow oil. 1H NMR (400 MHz
CDCl3)
solution of SGAAn-MSN (30% EtOH, 100 mL, 10ꢃ5 M SGAAn); (2)
excessive amount of Cu2þ (2 ꢂ 10ꢃ4 mol) was added to the solution
to adequately coordinate with SGAAn; (3) 5 ꢂ 10ꢃ4 mol EDTA-2Na
was added to the above complex solution to replace SGAAn. The
fluorescence intensities of the above solutions for each step were
recorded. SGAAn-MSN was recovered by centrifuging and washed
with a triseHCl (0.01 M) solution (water, pH ¼ 7.10) and distilled
water. The above process was totally repeated for 4 circles.
d
0.37 (t, 2H, J ¼ 8 Hz), 1.13 (t, 9H, J ¼ 7.2 Hz), 1.30e1.40 (m,
2H), 1.61 (s, 1H), 2.43 (t, 2H, J ¼ 7.2 Hz), 3.27 (s, 2H), 3.67 (q, 6H,
J ¼ 6.8 Hz), 5.42 (d, 2H, J ¼ 5.2 Hz), 7.47 (t, 3H, J ¼ 7.6 Hz), 7.55 (t, 2H,
J ¼ 8 Hz), 7.80 (d, 2H, J ¼ 8.4 Hz), 8.30 (d, 2H, J ¼ 8.4 Hz), 8.44 (s,1H);
13C NMR (100 MHz CDCl3)
d 7.61, 18.23, 23.13, 35.47, 152.28, 52.58,
58.29, 123.91, 125.16, 126.60, 128.00, 129.18, 130.37, 131.51, 171.61;
ESI found 469.1 [MþH]þ calculated for C26H36N2O4Si: 468.24.
2.2.5. Synthesis of SGAAn modified mesoporous silica nanosphere
(MSN)
2.3. Characterization
Mesoporous silica nanosphere (MSN) was prepared according to
a method reported in our previous publication [30]. The prepara-
tion of dye modified MSN was as follows: 100 mg dried MSN was
first suspended in 30 mL of anhydrous acetonitrile in a round
X-ray powder diffraction (XRD) patterns of all samples were
recorded on Rigaku D/MAX-2550 diffractometer using Cu K
ation on wavelength of 0.154 nm, operated at 40 kV and 100 mA.
UVevisible absorbance spectra were obtained with scan
a radi-
a
bottomed flask connected to
a Dean-stark apparatus under
UVevisible spectrophotometer (Varian, Cary 500). The fluores-
cence spectra of all samples were recorded with Cary Eclipse
fluorescence spectrophotometer. Scanning electron microscopy
(SEM) images were obtained with a JEOL JSM-6360LV microscope
at an accelerating voltage of 15 kV. Transmission electron micros-
copy (TEM) images were collected on a JEOL JEM 2010F, electron
microscope operated at an acceleration voltage of 200 kV. 1H and
13C NMR spectra were obtained on a Bruker AVANCE DMX500
spectrometer in CDCl3 with tetramethylsilane (TMS) as internal
standard. ICP-AES data were recorded on Varian-710-ES instru-
ment. Electron impact mass spectra were recorded on an Agilent
5973N MSD instrument, and ESI mass spectra were recorded on
Agilent 11100-Finnigan instrument.
nitrogen, and then 14 mg SGAAn was added. The mixture was
stirred for 24 h at 100 ꢀC. The SGAAn modified MSN was collected
by filtration and repeatedly washed with anhydrous acetonitrile,
dichloromethane and ethanol under ultrasonic condition. Unreac-
ted organic material was completely removed by monitoring the
absorbance of the washing liquid. The ultimate powder was dried
under vacuum and named as SGAAn-MSN. The actual loading
amount of SGAAn per gram of SGAAn-MSN was calculated to be
1.26 ꢂ 10ꢃ4 mol according to the UVevis absorption spectra of the
filtration eluent. The following study concerned with SGAAn-MSN
without specific explanation refers to sample loaded with
1.26 ꢂ 10ꢃ4 mol SGAAn per gram of SGAAn-MSN. Moreover,
another two samples loaded with 0.96 ꢂ 10ꢃ4 and 2.31 ꢂ 10ꢃ4 mol
SGAAn per gram of SGAAn-MSN were also prepared to study the
influence of SGAAn amount to the detection sensitivity.
3. Results and discussion
3.1. Mesoporous structure
2.2.6. Relations between pH value and fluorescence of SGAAn-MSN
The pH value of SGAAn-MSN solution was adjusted using 2 M
NaOH aqueous solution. The specific operation was as follows: (1)
Preparation of a EtOHeWater solution of SGAAn-MSN with pH ¼ 3
Fig. 1 shows the SEM and TEM images of mesoporous silica
nanoparticles. It is obvious that these particles are spherical and
highly dispersive. The approximately average particle size is about
125 nm as seen from TEM by measuring over 100 particles. The
pore channel is aligned from the center to the fringe, which is an
ideal structure for the access and dispersion of guest molecules. As
seen from Fig. 2(a), sample without dye modification shows a clear
(30% EtOH, 100 mL, 10ꢃ5
M for SGAAn); (2) 100 mL
Cu2þ
(2 ꢂ 10ꢃ2 M) was added to the SGAAn-MSN solution and the
fluorescent intensity was recorded; (3) The pH value of the above
complex solution was adjusted from 3 to 12 using 2 M NaOH
solution and the fluorescent intensity change was recorded.
peak at 2
q
¼ 1.53, which further verifies the mesoporous charac-
teristic of MSN. However, it is obvious that the introduction of dye
molecules into the pore channel makes some deterioration of pore
ordering since the peak intensity is largely decreased, which is
further approved by N2 adsorptionedesorption experiment.
Fig. 2(b) shows the N2 adsorptionedesorption isotherms and
2.2.7. Reversibility of SGAAn-MSN as Cu2þ fluorescent sensor
The chelate agent used in this experiment was a solution of
EDTAe2Na (0.1 M). The whole process for the investigation of
reversibility was as follows: (1) Preparation of a EtOHeWater
Fig. 1. SEM (a) and HRTEM (b) images of MSN.

242
D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
Fig. 2. XRD patterns (a) and N2 absorption/desorption isotherms (distribution of pore
size, insert) of MSN before (black) and after (gray) loaded with dye molecules.
distribution of pore size (inset b) of samples before and after dye
loading, which both exhibit a type IV isotherm typical for meso-
porous material. It can be seen that the introduction of dye mole-
cules into MSN decreases the N2 adsorption ability of MSN. The BET
surface area decreases from 631 to 497 m2 ꢃ1, the pore volume
g
calculated by BJH method decreased from 0.526 to 0.472 cc gꢃ1 and
the pore size calculated by BJH method show small decrease from
2.50 to 2.49 nm.
3.2. Optical properties
Fig. 3(a) shows the absorption spectrum of SGAAn and solid
diffuse reflection spectrum of SGAAn-MSN. It is obvious that both
SGAAn and SGAAn-MSN show a maximum absorption peak at
about 254 nm. Therefore, the incorporation of SGAAn in the pore
channel does not change its absorption property, which indicates
the preserving of molecule structure of anthracene after reaction
with surface silanol groups. The average dye molecule numbers
doped into each MSN particle is calculated as follows: First, the
standard curve of absorbance versus concentration for SGAAn was
plotted; then the loading amount of SGAAn molecules per gram of
MSN was calculated from the maximum absorption intensity of
filtrated SGAAn solution; Subsequently, the numbers of per gram of
MSN was determined by drying and weighing a certain volume of
nanoparticle solution. Ultimately, the dye molecule numbers per
nanoparticle is calculated to be about 2000. The local concentration
of dye molecules in one silica nanosphere is calculated to be about
3 ꢂ 10ꢃ3 M. The fluorescence spectra of SGAAn and SGAAn-MSN
solution are shown in Fig. 3(b). The total dye concentration for
SGAAn and SGAAn-MSN in the solution using 30% EtOH and 70%
water as solvents is fixed as 10ꢃ5 M. EtOH in SGAAn solution is used
to relieve the hydrolysis of alkoxyl silane groups of SGAAn during
fluorescence analysis, and to keep the same with SGAAn, EtOH was
also added into the SGAAn-MSN solution although the alkoxyl
silane groups has already reacted with the silanol groups on the
pore surface. As seen from Fig. 3(b), SGAAn-MSN shows similar
fluorescence spectrum with SGAAn but has lower peak intensity.
Nevertheless, it seems that SGAAn-MSN still preserves observable
fluorescence intensity although SGAAn molecules anchored in MSN
has a high local concentration (3 ꢂ 10ꢃ3 M) in the pore channel.
Generally, dye molecules in the homogeneous solution system will
encounter severe self-quench in the high concentration state.
Moreover, it is known the packing of dye molecules in the solid base
will also cause the fluorescence quench. The relieved fluorescence
quench of SGAAn-MSN can be explained as follows: MSN has
abundant pore channel and surface silanol groups. SGAAn mole-
cules can be highly dispersed into the pore channel of MSN and
covalently fixed in the different location of pore surface by reaction
Fig. 3. (a) Absorption and solid diffuse reflection spectra of SGAAn and SGAAn-MSN
(inset) and (b) Fluorescence spectra of SGAAn (Dash) and SGAAn-MSN (solid). The
solution composes of EtOH and water with a ratio of 3:7.
with silanol groups, which means the mobility and rotation of
SGAAn is restricted in a fixed area. Therefore, the molecule collision
can be effectively avoided even dye molecules are densely located
in MSN, and the generally observed severe fluorescence quench in
dye solution with high concentration can be much reduced.
3.3. Influence of pH value to the fluorescence of SGAAn-MSN
solution
Fig. 4 shows the relation between pH value and the fluorescence
intensity of SGAAn-MSN (10ꢃ5 M) in the presence of Cu2þ
(2 ꢂ 10ꢃ5 M). It can be found from Fig. 4 that the fluorescence
intensity of SGAAn-MSN has different responsive behaviors in
different pH ranges, which first slowly increases with the
increasing pH value from 3 to 6, especially in the range of 5e6, then
almost remains unchanged in the range of 6e9, and finally abruptly
increases with the increasing pH value from 9 to 12. The small
increase of the fluorescence intensity with the increasing pH value
from 3 to 6 should be attributed to the deprotonation of imine by
reducing the electron transfer fluorophore to protonated N. The
significant increase of the fluorescence intensity with the
increasing pH value from 9 to 11 should be attributed to the dis-
placing of Cu2þ from the complex, leading to the recovery of the
fluorescence of SGAAn. From Fig. 4 we can conclude that the
coordination between SGAAn-MSN and Cu2þ is stable in the pH
range of 7e9. Therefore, 0.05 M of HEPES buffer solution (pH ¼ 7)
was selected for the subsequent studies.

D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
243
size matching state between metal ions and the recognition site of
sensor molecule as well as the quenching ability of ions. It is well
known that ions such as Zn2þ, Cu2þ, Pb2þ and Agþ have high
thermodynamic affinity for chelate ligand composed of N, O and S
and fast metal-to-ligand binding kinetics. However, as seen from
Fig. 5, only Cu2þ cause obvious fluorescence quench of anthracene.
It is because Cu2þ with d9 electronic configuration has good ability
to accept electron. The radiationless deactivation of the excited
state of the fluorophore by Cu2þ can take place by means of either
an energy transfer mechanism occurring by electron exchange with
an unpaired d orbital of Cu2þ or an anthracene-to-Cu2þ electron
transfer process [9]. Therefore, the effective fluorescence quench
can occur when Cu2þ coordinates with the recognition site of
sensor molecule. Different from Cu2þ, ions such as Zn2þ and Cd2þ
with d10 electronic configuration and no unpaired electron has poor
ability to accept electron from anthracene, leading to the negligible
influence to the fluorescence. Ions such as Agþ also have no influ-
ence on the fluorescence of anthracene, which may be due to the
inefficient coordination between Agþ with large size and chelate
group. Except for Cu2þ, Pb2þ shows more obvious influence to the
fluorescence of anthracene than other metal ions, which should be
attributed to the heavy atom effect. The interference of different
metal ions to Cu2þ was also studied as seen from Fig. 6. It is obvious
that the addition of other metal ions in the presence of Cu2þ has
a negligible influence on the fluorescence intensity of SGAAn and
SGAAn-MSN, which indicates the sensor presented here has high
anti-interference ability.
Fig. 4. Relation between pH value and fluorescence intensity of SGAAn-MSN (10ꢃ5 M)
in the presence of Cu2þ (2 ꢂ 10ꢃ5 M).
3.4. Selectivity
Fig. 5 shows the fluorescence spectra of SGAAn and SGAAn-MSN
with dye concentration of 10ꢃ5 M in the presence of various metal
ions. It can be found that the fluorescence intensities of SGAAn and
SGAAn-MSN considerably decrease in the presence of Cu2þ with
10ꢃ4 M. However, the addition of other metal ions such as Zn2þ
,
3.5. Sensitivity
Fe3þ, Pb2þ and Agþ with the same concentration does not trigger
obvious fluorescence change of the sensor, which well illustrates
the selectivity of the above sensors to Cu2þ. Generally, the fluo-
rescence quench efficiency is determined by the coordinating and
The contact-time between Cu2þ and SGAAn-MSN is very short,
which is only about 4e5 s. Here, the detection sensitivities of
SGAAn and SGAAn-MSN were mainly evaluated by using 10%
decrease of the initial fluorescence intensity of SGAAn and SGAAn-
MSN in the presence of Cu2þ. As shown in Fig. 7, the obvious 10%
Fig. 5. Fluorescence spectra of SGAAn (a) and SGAAn-MSN (b) in the presence of
various metal ions (10ꢃ4 M) excited with 254 nm. 0.05 M HEPES buffer solution was
used to keep pH at 7.0. The solution composes of EtOH and water with a ratio of 3:7.
Fig. 6. Fluorescence intensity of SGAAn (a) and SGAAn-MSN (b) (10ꢃ5 M) in the
presence of various metal ions (10ꢃ4 M) and Cu2þ (10ꢃ4 M).

244
D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
molecule is also disadvantageous to the sensitivity since the local
concentration of dye molecules around Cu2þ in the pore channel is
low. Furthermore, to study the removing ability of SGAAn-MSN for
Cu2þ, we evaluated the removing efficiency of sample b for Cu2þ by
ICP-AES. A Cu2þ solution of 3 ꢂ 10ꢃ6 M (50 mL, 0.19 ppm) was
treated with 40 mg of sample b (1 ꢂ10ꢃ4 M SGAAn) for 1 h and then
the nanoparticle was filtrated. The residual Cu2þ in the filtrate is
0.02 ppm determined by ICP-AES, which means that about 90% of
Cu2þ in aqueous solution is adsorbed by SGAAn-MSN. Therefore,
SGAAn-MSN is a promising functional material for the use in the
removal of Cu2þ from polluted water.
Fig. 7. Fluorescence spectra of SGAAn (a) and SGAAn-MSN (b) with different amount of
Cu2þ (a: from the top of 4 ꢂ 10ꢃ6 to the bottom of 10ꢃ3 M; b: from the top of 5 ꢂ 10ꢃ8
to the bottom of 10ꢃ4 M).
3.6. Detection reversibility of SGAAn-MSN
To evaluate the sensing reversibility of SGAAn-MSN to Cu2þ, the
fluorescence changing of SGAAn-MSN coordinating with Cu2þ was
repeatedly studied for 4 cycles by using EDTA-2Na (10ꢃ3 M) as the
chelate agent and the results are shown in Fig. 9. As seen from Fig. 9,
the fluorescence of SGAAn-MSN can almost be totally recovered
after the addition of EDTA-2Na, which shows good response to the
further addition of Cu2þ for each cycle. The above result means
SGAAn-MSN possesses an excellent reusability and stability for the
decrease of the fluorescence intensity is observed when the
concentration of Cu2þ reaches to 10ꢃ6 M for SGAAn and 5 ꢂ 10ꢃ8 M
for SGAAn-MSN, which indicates the sensitivity of SGAAn-MSN is
significantly improved compared with SGAAn. The detection limit
of SGAAn-MSN was calculated according to equation (1), where
ꢀ
xmin is the minimal analytical signal that can be detected, x0 is the
average fluorescence intensity of blank SGAAn-MSN solution under
the same condition, K is the confidence level and S0 is the standard
sensing of Cu2þ
.
ꢀ
deviation of blank. Here, x0 and S0 are calculated to be 497 and 4.5,
respectively. xmin is calculated to be 483.5 by setting the K value as
generally used 3, which corresponds to 2 ꢂ 10ꢃ8 M Cu2þ. Therefore,
the detection limit of SGAAn-MSN for Cu2þ is 2 ꢂ 10ꢃ8 M.
3.7. Mechanism
To investigate the coordination information between Cu2þ and
SGAAn, the titration curve of SGAAn (10ꢃ5 M, Fig. 10) in the pres-
ence of various amount of Cu2þ was plotted. According to the
inflection point, SGAAn molecules and Cu2þ form a 1:1 complex.
The fluorescence quenched by further addition of Cu2þ above
10ꢃ5 M becomes moderate, which should be caused by molecule
collision between fluorophore and Cu2þ. Differently, quench caused
by the complex formation is named as static quench, which is
more effective and can be described as SterneVolmer equation,
I0/I ¼ 1 þ KSV [Q], where [Q] is the Cu2þ concentration, KSV is the
SterneVolmer quenching constant, and I and I0 are fluorescence
intensities of SGAAn in the presence and absence of Cu2þ, respec-
tively. As shown in the insert of Fig. 10, a good linearity is observed
when the concentration of Cu2þ is below 10ꢃ5 M and K value is
ꢀ
xmin ¼ x0 þ KS0
(1)
The sensitivity of SGAAn-MSN loaded with different amount of
SGAAn to Cu2þ was also compared. As seen from Fig. 8(a), the actual
loading amounts of SGAAn per gram of MSN for samples a, b and c
is 0.96, 1.26 and 2.31 ꢂ 10ꢃ4 mol/g, respectively. It is obvious that
sample b shows the best sensitivity. It indicates that too much or
too less loading amount leads to the decrease of sensitivity, which
can be explained as follows: Dye molecules introduced into the
pore channel may lead to some deterioration of mesoporous
structure and the obstruction of pore channel, which is approved by
N2 absorptionedesorption isotherms. Fig. 8(b) shows the N2
absorptionedesorption isotherms of MSN with various loading
amounts of SGAAn. It can be seen that the N2 adsorption ability of
SGAAn-MSN decreases with the increasing loading amounts of dye
molecules because the relative pressure becomes lower. The BET
surface area decreases from 497, 432 to 368 m2 gꢃ1 and the aver-
agepore size calculated by BJH method decreases from 2.49, 2.11 to
1.76 nm for sample a, b and c. Cu2þ can not well interact with dye
molecules if the pore channel is severely obstructed when too
much dye molecules are introduced. However, too less dye
found to be 1.6 ꢂ 105 Mꢃ1
.
Since Cu2þ and SGAAn form a 1:1 complex and Cu2þ desires
a square planar geometry when it is coordinated, we infer that the
Fig. 9. Reversibility of the sensing response of SGAAn-MSN (10ꢃ6 mol SGAAn) to Cu2þ
(2 ꢂ 10ꢃ4 mol). I1 w I5 corresponds to the fluorescence intensity of SGAAn-MSN after
the displacing of SGAAn from the Cu2þ complex by EDTAe2Na (5 ꢂ 10ꢃ4 mol). I1 w I4
0
0
Fig. 8. Sensitivities and N2 adsorption/desorption isotherms of SGAAn-MSN loaded
with different amount of anthracene.
corresponds to the fluorescence intensity of SGAAn-MSN coordinated with Cu2þ
.

D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
245
Fig. 10. Titration curve and SterneVolmer plot (inset) of SGAAn (10ꢃ5 M) treated with
Cu2þ
.
two nitrogen atoms on SGAAn, together with silanol oxygen on the
pore surface of MSN provide ligands to coordinate with Cu2þ
(Fig. 11), leading to the fluorescence quench of SGAAn-MSN. The
SterneVolmer plot for SGAAn-MSN in the presence of various
amount of Cu2þ is shown in Fig. 12. As seen from Fig. 12(a), the
addition of tiny amount of Cu2þ into SGAAn-MSN (1.26 ꢂ 10ꢃ4 mol/
g) solution containing 10ꢃ5 M SGAAn leads to much more obvious
fluorescence quench than SGAAn solution with the same concen-
tration. I0/I linearly increases with the increasing Cu2þ concentra-
tion in the range of 0e10ꢃ7 M and the corresponding KSV value is
1.24 ꢂ 107 Mꢃ1 (Fig. 12a (inset)). The improved quench constant can
be attributed to the following reasons: (1) The opening pore
channel and negatively charged pore surface of MSN make Cu2þ
more accessible to the dye molecules; (2) The dye molecules
anchored in the pore channel are fixed and well dispersed. The local
concentration of dye molecules in one mesoporous silica nano-
particle is much higher than the total dye concentration in solution.
This arrangement avoids the molecule collision between dye
molecules and improves the contact possibility between dye and
metal ions. When Cu2þ enters into the pore channel, it is sur-
rounded by anthracene in a restricted space of nanoporous channel,
leading to more effective energy or electron transfer. Therefore, the
fluorescence is more efficiently quenched even the total concen-
tration of anthracene in solution is low. For the homogeneous
SGAAn solution, unless the quencher concentration is very high,
significant quenching requires fast diffusion to bring molecules into
contact. The increasing of dye concentration may also lead to the
situation that dozens of times of SGAAn molecules center around
Cu2þ in a nanometer-sized circle region when tiny amount of Cu2þ
is added into the solution. However, an over high concentration of
dye molecules often results in a severe self-quench of fluorescence
Fig. 12. (a) SterneVolmer plot for SGAAn-MSN; (b) modified SterneVolmer plots for
two populations of fluorophores in SGAAn-MSN.
since dye molecules are free in solution and these molecules will
inevitably encounter frequent intermolecular collisions.
Moreover, the SterneVolmer plot of the above SGAAn-MSN
solution shows an unusually severe deviation from linearity
toward the x-axis when only half of fluorescence intensity is
quenched. Generally, a linear SterneVolmer plot is indicative of
a single class of fluorophore, all equally accessible to quencher. If
two fluorophore populations are present and one class is not
accessible to quencher, the SterneVolmer plots will deviate from
linearity toward the x-axis. This result is frequently found for the
quenching of tryptophan fluorescence in proteins by polar or
charged quenchers [9]. These molecules do not readily penetrate
the hydrophobic interior of proteins, and only those tryptophan
residues on the surface of the protein are quenched. For the dye
modified mesoporous structure, the dye anchored in the pore
channel may lead to the pore narrowing and even obstruction.
Therefore, dye molecules located in the interior center of meso-
porous sphere become inaccessible, which means two groups of
dye molecules divided into the accessible and the inaccessible lead
to the deviation of SterneVolmer plot to x-axis.
I0
I0 ꢃ I
1
1
¼
þ
(2)
faKa½Qꢄ fa
The accessible dye molecules in the pore channel can be calcu-
lated according to equation (2) [31], where Ka is the SterneVolmer
quenching constant of the accessible fraction, fa is the fraction of
the initial fluorescence that is accessible to quencher and [Q] is the
concentration of quencher. A plot of I0/(I0 ꢃ I) versus 1/[Q] yields 1/fa
as the intercept. As seen from Fig. 12(b), the fa value of the above
sampleis about0.6 calculatedfromtheintercept, which meansmore
than half of dye molecules are accessible to Cu2þ.
Fig. 11. Binding between Cu2þ and ligands and the turn-off fluorescence response of
SGAAn-MSN.

246
D. Lu et al. / Dyes and Pigments 94 (2012) 239e246
[12] Virji S, Kojima R, Fowler JD, Villanueva JG, Kaner RB, Weiller BH. Polyaniline
4. Conclusions
nanofiber composites with amines: novel materials for phosgene detection.
Nano Res 2009;2:135e42.
[13] Leng B, Jiang JB, Tian H. A mesoporous silica supported Hg2þ chemodosimeter.
AIChE J 2010;56:2957e64.
Monodisperse mesoporous silica nanoparticles modified with
anthracene derivative present a good selectivity and a high sensi-
tivity to Cu2þ. The presence of other metal ions such as Zn2þ, Cd2þ
,
[14] Mu LX, Shi WS, Chang JC, Lee ST. Silicon nanowires-based fluorescence sensor
for Cu (II). Nano Lett 2008;8:104e9.
Fe3þ, Pb2þ and Agþ has little influence on the selectivity of Cu2þ.
The sensitivity of this nanocomposite to Cu2þ is determined by the
local concentration and accessibility of dye molecules in the pore
channel. The improved sensitivity of the nanocomposite is due to
the high local concentration of dye molecules in the pore channel
with less self fluorescence quench caused by molecule collision.
[15] Chen QG, Zhou TY, He CY, Jiang YQ, Chen X. An in situ applicable colorimetric
Cu2þ sensor using quantum dot quenching. Anal Methods 2011;3:1471e4.
[16] Wang HG, Sun L, Li YP, Fei XL, Sun MD, Zhang CQ, et al. Layer-by-layer
assembled Fe3O4@C@CdTe core/shell microspheres as separable luminescent
probe for sensitive sensing of Cu2þ ions. Langmuir 2011;27:11609e15.
[17] Lee MH, Lee SJ, Jung JH, Kim JS. Luminophore-immobilized mesoporous silica
for selective Hg2þ sensing. Tetrahedron 2007;63:12087e92.
[18] Barja BC, Sara E, Bari M, Marchi C, Iglesias FL, Bernardi M. Luminescent Eu(III)
hybrid sensors for in situ copper detection. Sensors Actuators B 2011;158:
214e22.
[19] Song YJ, Qu KG, Xu C, Ren JS, Qu XG. Visual and quantitative detection of
copper ions using magnetic silica nanoparticles clicked on multiwalled carbon
nanotubes. Chem Commun 2010;46:6572e4.
[20] Meng QT, Zhang XL, He C, He GJ, Zhou P, Duan CY. Multifunctional meso-
porous silica material used for detection and adsorption of Cu2þ in aqueous
solution and biological applications in vitro and in vivo. Adv Funct Mater
2010;20:1903e9.
Acknowledgments
This work has been supported by National Nature Science
Foundation of China (21007016 and 20977030), the Project of
International Cooperation of the Ministry of Science and Tech-
nology of China (2011DFA50530), Science and Technology
Commission of Shanghai Municipality (11142201100) and the
Fundamental Research Funds for the Central Universities.
[21] Kresge CT, Leonowicz ME, Roth WJ, Vartuli JC, Beck JS. Ordered mesoporous
molecular sieves synthesized by a liquid-crystal template mechanism. Nature
1992;359:710e2.
[22] Rosenholm JM, Meinander A, Peuhu E, Niemi R, Eriksson JE, Sahlgren C, et al.
Targeting of porous hybrid silica nanoparticles to cancer cells. ACS Nano 2009;
3:197e206.
References
[23] Song CX, Zhang XL, Jia CY, Zhou P, Quan X, Duan CY. A hybrid mesoporous
material functionalized by 1,8-naphthalimide-base receptor and the applica-
tion as chemosensor and absorbent for Hg2þ in water. Talanta 2010;81:
643e9.
[24] Zhou J, Wu W, Caruntu D, Yu MH, Martin A, Chen JF, et al. Synthesis of porous
magnetic hollow silica nanospheres for nanomedicine application. J Phys
Chem C 2007;111:17473e7.
[1] Hosono T, Su CC, Siringan F, Onodera SC. Effects of environmental regulations
on heavy metal pollution decline in core sediments from Manila Bay. Marine
Pollut Bull 2010;60:780e5.
[2] Nriagu JO. A silent epidemic of environmental metal poisoning. Environ Pollut
1988;50:139e61.
[3] Clarkson TW, Magos L, Myers GJ. The toxicology of mercury e current expo-
sures and clinical manifestations. N Engl J Med 2003;349:1731e7.
[4] Berg JM, Shi Y. The galvanization of biology: a growing appreciation for the
roles of Zinc. Science 1996;271:1081e5.
[25] Barnham KJ, Masters CL, Bush AI. Neurodegenerative diseases and oxidative
stress. Nat Rev Drug Discov 2004;3:205e14.
[26] Xu ZH, Zhang LK, Guo R, Xiang TC, Wu CZ, Zheng Z, et al. A highly sensitive
and selective colorimetric and offeon fluorescent chemosensor for Cu2þ based
on rhodamine B derivative. Sensors Actuators B 2011;156:546e52.
[27] Mu HL, Gong R, Ma Q, Sun YM, Fu EQ. A novel colorimetric and fluorescent
chemosensor: synthesis and selective detection for Cu2þ and Hg2þ. Tetrahe-
dron Lett 2007;30:5525e9.
[28] El-Safty SA, Ismail AA, Shahat A. Optical supermicrosensor responses for
simple recognition and sensitive removal of Cu (II) Ion target. Talanta 2011;
83:1341e51.
[5] Waggoner DJ, Bartnikas TB, Gitlin JD. The role of copper in neurodegenerative
disease. Neurobiol Dis 1999;6:221e30.
[6] Weizman H, Ardon O, Mester B, Libman J, Dwir O, Hadar Y, et al. Fluo-
rescently-labeled ferrichrome analogs as probes for receptor-mediated
microbial iron uptake. J Am Chem Soc 1996;118:12368e75.
[7] Manjusha R, Dash K, Karunasagar D. UV-photolysis assisted digestion of food
samples for the determination of selenium by electrothermal atomic
absorption spectrometry. Food Chem 2007;105:260e5.
[8] He Q, Chang XJ, Huang XP, Hu Z. Silica gel surface-imprinted solid-phase
extraction of Zr (IV) from aqueous solutions. Microchim Act 2008;160:147e52.
[9] Fabbrizzi L, Licchelli M, Pallavicini P. Transition metals as switches. Acc Chem
Res 1999;32:846e53.
[29] Torrado A, Walkup GK, Imperiali B. Exploiting polypeptide motifs for the
design of selective Cu (II) ion chemosensors. J Am Chem Soc 1998;120:
609e10.
[30] Wang JG, Xiao Q, Zhou HJ, Sun PC, Ding DT, Chen TH. Anionic surfactant-
templated mesoporous silica (AMS) nano-spheres with radially oriented
mesopores. J Colloid Interf Sci 2008;323:332e7.
[10] Kim JS, Quang DT. Calixarene-derived fluorescent probes. Chem Rev 2007;
107:3780e99.
[11] Lee HY, Son HJ, Lim JM, Oh JM, Kang DM, Han WS, et al. BODIPY-functionalized
[31] Lakowicz JR. Quenching of fluorescence. Principles of fluorescence Spectros-
copy. 3nd ed. New York: Springer; 2006. pp. 288e289.
gold nanoparticles as
a selective fluoro-chromogenic chemosensor for
imaging Cu2þ in living cells. Analyst 2010;135:2022e7.