Identifying the Stoichiometry of Metal/Ligand Complex by Coupling Spectroscopy and Modelling: a Comprehensive Study on Two Fluorescent Molecules Specific to Lead

Article abstract of DOI:10.1007/s10895-019-02405-0

Two new chemosensors for lead (II) were synthesized based on 5-((anthracen-9-ylmethylene) amino)quinolin-10-ol (ANQ). ANQ was modified in the para position of the imine group via a methoxy link either with methylmethacrylate (ANQ-MMA) or styrene (ANQ-ST).

Full text of DOI:10.1007/s10895-019-02405-0

Journal of Fluorescence
https://doi.org/10.1007/s10895-019-02405-0
ORIGINAL ARTICLE
Identifying the Stoichiometry of Metal/Ligand Complex by Coupling
Spectroscopy and Modelling: a Comprehensive Study on Two
Fluorescent Molecules Specific to Lead
William René1,2,3 & Madjid Arab2 & Katri Laatikainen4 & Stéphane Mounier3 & Catherine Branger1 & Véronique Lenoble3
Received: 10 May 2019 /Accepted: 24 June 2019
#
Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
Two new chemosensors for lead (II) were synthesized based on 5-((anthracen-9-ylmethylene) amino)quinolin-10-ol (ANQ). ANQ
was modified in the para position of the imine group via a methoxy link either with methylmethacrylate (ANQ-MMA) or styrene
(ANQ-ST). Complexation of those molecules with Pb2+ was studied at room temperature using UV-Visible absorption and
fluorescence spectroscopies. Thanks to the UV-visible absorption spectroscopy, it appeared that ANQ-MMA formed 1:1 and 1:2
complexes with lead (II) and ANQ-STonly 1:1 complex. For both molecules, the fluorescence excitation-emission matrices (EEM)
signal intensity increased from 0 to 100 μmol.L−1 of lead (II) followed by a saturation for higher concentrations. The decomposition
of the obtained EEMs gave a set of empiric fluorescent components that have been directly linked to the distribution of lead
complexes obtained with the UV-visible absorption spectroscopy study. This correlation allowed to evidence metal/ligand complex
stoichiometry and emerge as a new method to identify empiric components. Moreover, the two ligands showed a promising
selectivity for Pb2+, turning them interesting probes for this hazardous metal.
.
.
.
.
Keywords Metal/ligand complex Spectroscopy Modelling Fluorescent molecules Lead
Introduction
presence in the environment is mainly due to human activities
such as fuel combustion and industrial processes [1–3]. This
trace metal can accumulate in water and soil organisms and
influence the global balances and food chains [4]. Taken up
from food, air or drinking water, lead causes inhibition of brain
development, kidney and physical growth impairments [5, 6].
The maximum concentration level in drinking water is set at
10 μg.L−1 [7]. The development of probes and sensors to detect
lead therefore appears to be necessary especially for its most
toxic species, the ionic lead (II) form. Fluorescence probes for
the detection of trace metal ions is getting attention due to easy
signal transduction and high sensitivity [8]. Fluorescent ligand
for lead (II) have been recently developed based on fluorescein
[9], rhodamine [10, 11], crown ether derivates [12, 13] or scaf-
fold molecules [14–16].
Lead is one of the most toxic metal for humankind and the
environment. Lead naturally exists in Earth’s crust but its
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s10895-019-02405-0) contains supplementary
material, which is available to authorized users.
* Catherine Branger
branger@univ-tln.fr
* Véronique Lenoble
lenoble@univ-tln.fr
1
MAPIEM Laboratory, University of Toulon, CS 60584,
Typical fluorescent chemosensors, also called
fluoroionophores, include a recognition site or chelating group
responsible for the efficiency of the binding and a fluorescent
part that converts the ion recognition into a fluorescent signal.
Upon ion complexation, their fluorescence response can be
either quenched or enhanced but the turn-on response is pref-
erable in order to avoid interferences with external factors that
can be observed in the case of turn-off response [17]. One of the
83041 Toulon, Cedex 9, France
2
IM2NP Laboratory, University of Toulon, CS 60584,
83041 Toulon, Cedex 9, France
3
MIO Laboratory, University of Toulon, CS 60584,
83041 Toulon, Cedex 9, France
4
Laboratory of Computational and Process Engineering,
Lappeenranta-Lahti University of Technology LUT, P.O. Box 20,
FI-53851 Lappeenranta, Finland

J Fluoresc
possible mechanism responsible for a turn-on fluorescence re-
sponse is the photoinduced electron transfer (PET) process
[18–20]. This process is based on a quenching of the fluores-
cence response of the fluorophore moiety by an electron trans-
fer from the ionophore. Upon ion binding, this electron transfer
is prevented and consequently fluorescence intensity is
enhanced.
1H, 13C and 1H - 13C heteronuclear single quantum coher-
ence (HSQC) nuclear magnetic resonance (NMR) measure-
ments were obtained on a Bruker 400 MHz Ultrashield
spectrometer.
Mass spectra were measured on a Bruker Esquire 6000
instrument.
Melting points were determined on a Bushi M-560
apparatus.
Anthracene is widely used as a fluorophore to design
chemosensors, owing to the ease to modify its structure [21].
A n a n d e t a l . s y n t h e s i z e d 5 - ( ( a n t h r a c e n - 9 -
ylmethylene)amino)quinoline-10-ol (ANQ), based on an an-
thracene platform, to detect lead (II) with a turn-on fluores-
cence response based on PET process [22]. In the present
study, ANQ was modified in the para position of the imine
group via a methoxy link to design two new fluorescent li-
gands for lead (II): (10-(8-hydroxyquinolin-5-
ylimino)methyl)anthracen-9-yl)methyl methacrylate (ANQ-
MMA) and ((10-((4-vinylbenzyloxy)methyl)anthracen-9-
yl)methyleneamino)quinoline-8-ol (ANQ-ST). The formation
of the complexes between lead (II) with ANQ-MMA and
ANQ-ST was first investigated using UV-visible absorption
spectroscopy and the distribution of the various complexes
(1:1 complex, 1:2 complex,) as well as free metal and
free ligand was calculated. In parallel, the fluorescence
response of the two ligands upon the addition of lead
(II) was analysed. Within this step, the recovered fluo-
rescent signal was decomposed into a combination of
fluorescent components. These components were then
correlated to the above-mentioned possible 1:1 and 1:2
complexes. Such innovative correlation allowed to asso-
ciate a modelled fluorescent signal with a complex of
known-stoichiometry, an approach that opens a range of pos-
sibilities in the study of the interactions between ligand and
metal ion.
UV-visible absorption spectra were obtained with a
Shimazu UV-2501 spectrometer.
Synthesis of (10-(8-Hydroxyquinolin-5-Ylimino)
Methyl)Anthracen-9-Yl)Methyl Methacrylate
(ANQ-MMA)
Step 3: Synthesis of (10-Formylanthracen-9-Yl)Methyl
Methacrylate
2.51 mL of freshly distilled methacryloyl chloride
( 2 6 m m o l ) w e r e a d d e d t o
a s o l u t i o n o f
10-(hydroxymethyl)anthracen-9-carbaldehyde (3.04 g,
13 mmol) and triethylamine (10.76 mL, 77 mmol) in
tetrahydrofuran cooled at 0 °C. The solution mixture
was stirred at 0 °C for 2 h and then at room tempera-
ture for 20 h. The solvent was distilled off under re-
duced pressure. The crude product was solubilized in
dichloromethane, washed five times with 100 mL of a satu-
rated potassium carbonate solution and then twice with
100 mL of distilled water. The product was finally dried over
magnesium sulphate to give 3.48 g of (10-formylanthracen-9-
yl)methyl methacrylate. (Fig. 1a) Yield: 89%. Melting point:
132.7 °C.
1H NMR (400 MHz, DMSO-d6, δ in ppm), δ: 11.45 (s, 1H,
CHO), 8.90 (dd, J = 7.6, 2.2 Hz, 2H, position 4), 8.54 (dd, J =
7.5, 2.1 Hz, 2H, position 7), 7.73 (m, 4H, position 5 and 6),
6.25 (s, 2H, CH2), 5.94 (s, 1H, position 13), 5.63 (s, 1H,
position 13), 1.83 (s, 3H, CH3). (Online Resource, Fig. A).
13C NMR (400 MHz, CDCl3, δ in ppm), δ: 194.05 (C10),
167.37 (C11), 135.93 (C2), 134.15 (C12), 131.14(C8),
130.66(C3), 128.39(C5), 127.74(C9), 126.82 (C6), 126.53
(C13), 124.92 (C7), 124.24 (C4), 58.94 (C1), 18.39 (C14).
(Online Resource, Fig. B).
Experimental
Reagents and Instruments
Anthraquinone, sodium hydride (60% in oil dispersion),
lithium bromide, triethylamine, 4-vinylbenzylchloride,
18-crown-6 ether and 5-amino-8-hydroxyquinoline
dihydrochloride were purchased from Sigma-Aldrich (re-
agent grade).
LC-MS: calculated: 304. Found: 305.13 ((10-
formylanthracen-9-yl)methyl methacrylate + H)+ (Online
Resource, Fig. C).
AgNO3, CaCO3, ZnSO4 and NaNO3 were purchased from
Fisher Scientific (Analytical grade), Al(NO3)3, CdSO4,
Co(NO3)2 and CuSO4 were from Merk (pro analysis grade),
Pb(NO3)2 and Fe(NO3)3 were purchased from Carlo Erba
(Analytical grade).
All other chemical reagents and solvents were purchased
from Acros Organics (reagent grade). Dry solvents were pur-
chased as extra dry grade (Acros Organics).
Step 4: Synthesis of (10-(8-Hydroxyquinolin-5-Ylimino)
Methyl)Anthracen-9-Yl)Methyl Methacrylate (ANQ-MMA)
100 mL of an ethanolic solution containing (10-
formylanthracen-9-yl)methyl-methacrylate (1.10 g,
3.62 mmol), 5-amino-8-hydroxyquinoline dihydrochloride
(0.84 g, 3.62 mmol) and 5 drops of triethylamine was refluxed

J Fluoresc
for 4 h. Then the solvent was removed under reduced pressure.
The crude product was washed with 100 mL of diethyl ether
then with 100 mL of a solution containing potassium carbon-
ate (0.55 g, 3.98 mmol) and extracted with dichloromethane.
After distillation under reduced pressure, 1.21 g of the product
ANQ-MMA is obtained. (Fig. 1b) Yield: 75%. Melting point:
220.0 °C.
(7.0 g, 42.7 mmol) in acetone under argon was refluxed for
20 h. After cooling, the mixture was poured into water
and extracted with dichloromethane. To neutralize the
excess of iodine, the mixture was washed twice with
100 mL of Na2SO3 saturated solution. Then, the organic
phase was washed twice with 100 mL of water and
dried over magnesium sulphate. After removal of the solvent
under reduced pressure, 7,9 g of product was obtained.
(Fig. 2a) Yield: 99%.
1H NMR (400 MHz, DMSO-d6, δ in ppm), δ: 10.03 (s, 1H,
OH), 9.96 (s, 1H, position 10), 8.95 (dd, J = 4.1, 1.6 Hz, 1H,
position 15), 8.80 (d, J = 9.4 Hz, 2H, position 4), 8.73 (dd, J =
8.5, 1.6 Hz, 1H, position 14), 8.54 (d, J = 8.2 Hz, 2H, position
7), 7.78 (d, J = 8.2 Hz, 1H, position 13), 7.67 (m, 5H, position
5, 6 and 19), 7.25(d, 1H, position 18), 6.30 (s, 2H, CH2), 5.96
(s, 1H, position 22), 5.65 (s, 1H, position 22), 1.86 (s, 3H,
CH3). (Online Resource, Fig. D).
1H NMR (400 MHz, CDCl3, δ in ppm), δ: 7.38 (s, 4H,
aromatic H), 6.74 (dd, J = 17.6, 10,7 Hz, 1H, position 6),
5.82 (dd, J = 17.6, 0.7 Hz, 1H, position 7), 5.32 (dd, 10.7,
0.7 Hz, 1H, position 7), 4.50 (s, 2H, position 1). (Online
Resource, Fig. G).
13C NMR (400 MHz, CDCl3, δ in ppm), δ: 138.95 (C2),
137.32 (C5), 136.39 (C6), 129.13 (C3), 126.81 (C4), 114.58
(C7), 6.09 (C1). (Online Resource, Fig. H).
13C NMR (400 MHz, DMSO-d6, δ in ppm), δ: 167.05
(C20), 159.14 (C10), 153.24 (C17), 149.26 (C15), 139.58,
138.92, 136.23, 132.80, 130.70, 130.18, 130.01, 127.29,
126.79, 126.30, 125.26, 122.65, 115.34 (C18), 111.81 (C19),
59.39 (C1), 18.55 (C23). (Online Resource, Fig. E).
1H - 13C HSQC NMR (400 MHz, CDCl3): Expected
crosspeaks signals were observed. (Online Resource, Fig. I).
LC-MS: calculated: 446. Found: 447.14 (ANQ-MMA + H+)
(Online Resource, Fig. F).
Step 3′: Synthesis of 9-((4-Vinylbenzyloxy)Methyl)
Anthracene-10-Carbaldehyde
Two new chemosensors for lead (II) were synthesized
based on 5-((anthracen-9-ylmethylene) amino)quinolin-10-ol
(ANQ). ANQ was modified in the para position of the imine
group via a methoxy link either with methylmethacrylate
(ANQ-MMA) or styrene (ANQ-ST). Complexation of those
molecules with Pb2+ was studied at room temperature using
UV-Visible absorption and fluorescence spectroscopies.
Thanks to the UV-visible absorption spectroscopy, it appeared
that ANQ-MMA formed 1:1 and 1:2 complexes with lead (II)
and ANQ-ST only 1:1 complex. For both molecules, the
fluorescence excitation-emission matrices (EEM) signal
intensity increased from 0 to 100 μmol.L-1 of lead
(II) followed by a saturation for higher concentrations.
The decomposition of the obtained EEMs gave a set of
empiric fluorescent components that have been directly
linked to the distribution of lead complexes obtained
with the UV-visible absorption spectroscopy study.
This correlation allowed to evidence metal/ligand com-
plex stoichiometry and emerge as a new method to identify
empiric components. Moreover, the two ligands showed a
promising selectivity for Pb2+, turning them interesting probes
for this hazardous metal.
In a 250 mL flask, 287 mg of sodium hydride (7.2 mmol),
35 mg of 18-crown-6 ether and 10 mL of dry tetrahydrofuran
were added. The mixture was put under argon and cooled
down with a water/ice bath. Then 1. 00 g of
10-(hydroxymethyl)anthracen-9-carbaldehyde (4.23 mmol)
dissolved in 100 mL of dry tetrahydrofuran were slowly
added. After one hour of stirring at room temperature and
under argon, the mixture was cooled with a water/ice bath
and 1,76 g of 4-vinylbenzyliodide (7.2 mmol) diluted in
10 mL of dry tetrahydrofuran was added dropwise. The mix-
ture was stirred for 24 h under argon at room temperature, then
a few drops of water were added and the tetrahydrofu-
ran was distilled off under reduced pressure. The crude
product was extracted with dichloromethane and dried
over magnesium sulphate. The dichloromethane was dis-
tilled off under reduced pressure to give a residue which
was washed twice with 50 mL of cold hexane and pu-
rified on a silica gel column eluting with ethyl acetate–
cyclohexane (1:9, v/v) to obtain 490 mg of the product as a
yellow oil. (Fig. 2b) Yield: 33%.
1H NMR (400 MHz, CDCl3, δ in ppm), δ: 11.46 (s, 1H,
CHO), 8.87 (d, J = 8.8 Hz, 2H, position 13), 8.36 (d, J =
8.8 Hz, 2H, position 14), 7.61 (dtd, 4H, positions 4 and 7),
7.42 (dd, 4H, positions 5 and 6), 6.78 (dd, J = 17.6, 10.9 Hz,
1H, position 16), 5.82 (dd, J = 17.6, 0.8 Hz, 1H, CH2), 5.42 (s,
2H, position 1), 5.31 (dd, J = 10.9, 0.8 Hz, 1H, CH2), 4.73 (s,
2H, position 11). (Online Resource, Fig. J).
Synthesis of ((10-((4-Vinylbenzyloxy)Methyl)
Anthracene-9-Yl)Methyleneamino)Quinoline-8-Ol
(ANQ-ST)
Synthesis of 4-Vinylbenzyliodide
13C NMR (400 MHz, CDCl3, δ in ppm), δ: 193.96 (C10),
137.57 (C2), 137.45 (C15), 136.75 (C12), 136.60 (C16), 131.28
(C8), 130.53 (C9), 128.45 (C5), 128.33 (C6), 126.82 (C3),
This synthesis was inspired from Chalal et al. work [23]. A
mixture of 4-vinylbenzylchloride (5.0 g, 32.9 mmol) and NaI

J Fluoresc
126.46 (C7), 126.36 (C4), 125.23 (C14), 124.07 (C13), 114.18
(C17), 72.71 (C11), 64.05 (C1). (Online Resource, Fig. K).
1H - 13C HSQC NMR (400 MHz, CDCl3): Expected
crosspeaks signals were observed. (Online Resource,
Fig. L).
(C2), 137.35 (C21), 136.64 (C25), 133.48 (C13), 132.39
(C24), 130.90 (C9), 130.34 (C8), 129.69 (C3), 128.38
(C5), 126.67 (C6), 126.41 (C7), 126.22 (C4), 125.52
(C22), 125.19 (C12), 124.14 (C23), 122.00 (C14), 114.02
(C19), 113.94 (C26), 109.72 (C8), 72.38 (C20), 64.20
(C1). (Online Resource, Fig. N).
Step 4′: Synthesis of ((10-((4-Vinylbenzyloxy)Methyl)
Anthracen-9-Yl)Methyleneamino)Quinoline-8-Ol (ANQ-ST)
LC-MS: calculated: 494. Found: 495.18 (ANQ-ST + H+)
(Online Resource, Fig. O).
1H -13C HSQC NMR (400 MHz, CDCl3): Expected
crosspeaks signals were observed. (Online Resource,
Fig. P).
100 mL of an ethanolic solution containing 9-((4-
vinylbenzyloxy)methyl)antharcene-10-carbaldehyde
(380 mg, 1.08 mmol), 5-amino-8-hydroxyquinoline
dihydrochloride (252 mg, 1.08 mmol) and 10 drops of
triethylamine was refluxed for 4 h. Then the solvent
was removed under reduced pressure and the crude
product was washed twice with 100 mL of diethyl ether.
After distillation under reduced pressure of the filtrate,
the product was chromatographed on silica gel 60 elut-
ing with ethyl acetate–cyclohexane (1:9, v/v).
Evaporation of the appropriate fractions gave 160 mg
of ANQ-ST, as brown powder. (Fig. 2c) Yield: 30%.
Melting point: 210 °C.
Lead (II) and ANQ-MMA or ANQ-ST Species
Distribution Modelling from UV-Visible Spectra
Lead (II) complexation by ANQ-MMA and ANQ-ST
was studied in acetone (80%) - water (20%) mixture
at room temperature. The experiments were performed
in 3 mL quartz Suprasil cells. The combined concentra-
tion of lead (II) (0.01–0.09 mmol.L−1) and ANQ-MMA
or ANQ-ST (0.09–0.01 mmol.L−1) was kept constant
(0.1 mmol.L−1), but the ratio ligand/lead was varied
from 0.1 to 10 using Pb(NO3)2 as lead source. For the
two molecules, 14 spectra were recorded using UV-vis
spectrometer.
At equilibrium, the distribution of the various species (free
metal, 1:1 complex, 1:2 complex and free ligand) was calcu-
lated using a commercial program (HypSpec) based on the
least-squares minimization scheme [24, 25]. Stability con-
stants, extinction coefficients and concentrations of all absorb-
ing components were simultaneously estimated. For
uncomplexed lead (II), the extinction coefficients were calcu-
lated from independent measurements.
1H NMR (400 MHz, CDCl3, δ in ppm), δ: 9.84 (s, 1H,
position 10, 8.89 (dd, J = 4.2, 1.6 Hz, 1H, position 15) 8.85
(dd, j = 8.5, 1.6 Hz, 1H, position 13), 8.78 (m, 2H, position
22), 8.44 (m, 2H, position 23), 7.60 (m, 4H, positions 4 and 7),
7.52 (dd, J = 8.5, 4.5 Hz, 1H, position 14), 7.44 (m, 5H, posi-
tions 5, 6 and 19), 7.31 (m, 1H, position 18), 6.77 (dd, J =
17.6, 10.9 Hz, 1H, position 25), 5.80 (dd, J = 17.6, 0.8 Hz, 1H,
CH2), 5.57 (s, 2H, position 1), 5.29 (dd, J = 10.9, 0.8 Hz, 1H,
CH2), 4.76 (s, 2H, position 20). (Online Resource, Fig. M).
13C NMR (400 MHz, CDCl3, δ in ppm), δ: 158.84 (C10),
151.33(C17), 148.59 (C15), 140.88 (C11), 138.33 (C16), 137.87
OH
a)
b)
OH
17
a)
b)
c)
N
O
18
19
16
12
15
14
17
N
10
9
18
19
16
12
15
14
7
7
4
7
4
O
6
5
6
5
8
3
8
3
6
5
10
9
11
13
11
N
13
7
4
7
4
4
3
4
3
2
1
N
10
9
6
8
3
8
3
6
5
10
9
2
1
7
4
7
4
O
5
6
8
3
8
3
6
5
7
4
7
4
2
1
13
14
11
6
5
8
3
8
3
6
5
I
5
14
15
12
13
2
1
O
16
17
2
1
O
11
13
22
23
20
12
14
O
23
24
21
22
O
20
22
25
26
21
23
O
Fig. 2 Indexation of the carbons observed on the 13C NMR spectrum of
a) 4-vinylbenzyliodide, b) 9-((4-vinylbenzyloxy)methyl)anthracene-10-
carbaldehyde and c) ANQ-ST
Fig. 1 Indexation of the carbons observed on the 13C NMR spectrum of
a) (10-formylanthracen-9-yl)methyl methacrylate and b) ANQ-MMA

J Fluoresc
Fluorescence Measurements
Excitation-Emission Matrix (EEM) of Fluorescence
Measurement
Lead Experiments
The EEMs were measured on a HITACHI F4500 spectroflu-
orimeter. The excitation wavelength ranged from 320 to
460 nm, with a step of 10 nm and an excitation slit of 1 nm.
The corresponding emission spectra were acquired from 350
to 550 nm with a scan speed of 2400 nm.min−1 and a slit of
1 nm. The photomultiplicator tension was fixed at 950 V and
the integration time set at 0.1 s. The extraction of the 5 nm
stepped emission was obtained by FL-Solution software.
All experiments and measurements were performed at room
temperature, in acetone (80%) - water (20%) mixture. AgNO3,
CaCO3, NaNO3, CdSO4, Co(NO3)2, CuSO4, Pb(NO3)2,
ZnSO4 Al(NO3)3, and Fe(NO3)3 were used as metal sources.
The fluorescent signal as a function of increasing lead con-
centration was studied. Solutions containing 50 μmol.L−1 of
the fluoroionophore ANQ-MMA or ANQ-ST and 0, 5,10, 30,
50, 100 and 300 μmol.L−1 of lead (II) were prepared and
analysed.
CP/PARAFAC Analysis
The fluorescence signal of each the two synthesized mole-
cules in presence of lead or other ions, taken separately, was
measured. Solutions containing 50 μmol.L−1 of the molecule
and 200 μmol.L−1 of lead (II) or 200 μmol.L−1 of silver (I),
sodium (I), calcium (II), cadmium (II), cobalt (II), copper (II),
zinc (II), aluminium (III) or iron (III) were prepared and
analysed for their fluorescence signal.
Then, the fluorescent signal of a constant lead con-
centration in presence of interfering ions was studied. A
solution containing 50 μmol.L−1 fluoroionophore,
100 μmol.L−1 of lead and 100 μmol.L−1 of one inter-
fering ion was prepared. The interfering ion was chosen
among silver (I), sodium (I), calcium (II), cadmium (II),
cobalt (II), copper (II), zinc (II), aluminium (III) or iron
(III).
First, all the EEMs were cleaned from the diffusion signals:
Rayleigh by cutting the diffusion band (20 nm) and Raman
from first and second order by applying Zepp procedure [26].
Then, CP/PARAFAC algorithm was used [27, 28]. This algo-
rithm allows the decomposition of a dataset of matrices into a
set of fluorescent components, considering that all the consid-
ered EEM are constituted by a linear combination of the same
independent components. The correct number of components
needed to model the dataset is defined by evaluating the
CORCONDIA score. Users have to test a range of model,
i.e. number of components, to detect the best number of com-
ponents [29]. In this work, decomposition investigation
was done from two to five components and the higher
number of component giving a CORCONDIA test over
Step 1
O
O
I-
S+
NaH, DMSO
O
O
Step 2
LiBr, Ch3CN
O
OH
OH
N
Step 3'
Step 3
HO
N
Cl
N
Et
N
18-crow n-6 ether
3
O
H2C
CH3
I
O
N
O
Step 4
Step 4'
Et3N, EtOH
NH2 HCl
Et3N, EtOH
NH2 HCl
O
O
O
O
CH2
CH2
CH3
O
O
OH
CH3
OH
ANQ-MMA
ANQ-ST
Fig. 3 Synthesis route of ANQ-MMA and ANQ-ST

J Fluoresc
Fig. 4 Absorbance spectra of (a)
ANQ-MMA and (b) ANQ-ST
(50 μmol.L−1) in acetone-water
(4:1, v/v) solution and in presence
of 100 μmol.L−1 of Pb2+
a)
b)
0.2
0.15
0.1
0.4
0.3
0.2
0.1
0
ANQ-ST
ANQ-MMA
ANQ-ST + Pb2+
ANQ-MMA + Pb2+
0.05
0
330
430
530
630
330
430
530
630
Wavelength (nm)
Wavelength (nm)
60% was selected as the optimal component number
[30]. Then some components given by this decomposi-
tion were directly linked to the complexes metal-ligand evi-
denced by the UV-visible study.
v i n y l b e n z y l i o d i d e , w a s p r e p a r e d t o c o u p l e
10-(hydroxymethyl)anthracen-9-carbaldehyde by its hydroxyl
group with 4-vinylbenzyle before the final Schiff-base forma-
tion reaction. Molecular structures were characterized by 1H
1
NMR, 13C NMR, H-13C HSQC NMR and LC-MS tech-
niques (see Experimental section).
Results and Discussions
The photophysical properties of ANQ-MMA and ANQ-ST
were investigated. Both molecules showed a major absorption
band at 405 nm. Upon addition of Pb2+, the absorbance of
ANQ-MMA (Fig. 4a) and ANQ-ST (Fig. 4b) underwent a
red shift of 25 and 3 nm respectively. This result emphasized
a photophysical effect of lead (II) complexation by those
molecules.
Complex formation was studied by keeping the over-
all concentration of lead and fluoroionophore constant
while varying their molar ratio. The distribution of the
complexes formed between Pb2+ and the two
fluoroionophores was calculated using HypSpec, a com-
mercial program, based on the least squares minimiza-
tion method (Fig. 5) [24], using the following equilibri-
um equations:
Synthesis and Properties of Fluoroionophores
ANQ-MMA and ANQ-ST
The synthesis of ANQ-MMA and ANQ-ST took place in four
steps with an overall yield of 63% for ANQ-MMA and 10%
for ANQ-ST (Fig. 3). For both fluoroionophores, the first two
steps were inspired by the work of various authors [31–34].
They enable the functionalization of anthracene in the para
position of the future imine group. For ANQ-MMA, the third
step is an esterification reaction catalysed by a tertiary amine,
followed by a coupling between the intermediate compound
and the 5-amino-8-hydroxyquinoline dihydrochloride. In or-
der to synthesize ANQ-ST, the intermediate compound, 4-
Fig. 5 Metal and ligand species
distribution calculated from
UV-vis spectra using HypSpec
program [25] for lead and (a)
ANQ-MMA or (b) ANQ-ST.
Solvent: acetone-water (4:1, v/v)
a)
b)
100
100
Free
Pb(ANQ-MMA)
Free
ANQ-ST
ANQ-MMA
Pb(ANQ-ST)
80
60
40
20
0
80
60
40
20
0
Pb(ANQ-MMA)2
0
1
2
3
4
5
6
0
1
2
3
4
5
6
[Pb2+]/[ANQ-MMA]
[Pb2+]/[ANQ-ST]

J Fluoresc
Table 1 Stability constants of the complexes formed with lead for
ANQ-MMA and ANQ-ST
For ANQ-ST ligand, PARAFAC treatment gave optimum
results for two components (CORCONDIA = 94%).
CORCONDIA scores obtained for 2 to 5 components decom-
position are given in Table A (Online Resource). Component
1 showed a maximum located at λex/λem = 385–405/430–
440 nm. Component 2 showed a maximum located at λex/
Log (β1)
Log (β2)
ANQ-MMA
ANQ-ST
7.2
5.1
11.8
/
λ
em = 370–385/405–420 nm (Fig. 9a). Once those compo-
nents extracted, the contribution to fluorescence of each com-
ponent to each EEM was calculated (Fig. 9b). Each compo-
nent fluorescence contribution is proportional to their concen-
tration and to a quantum yield of fluorescence [35]. Coupling
PARAFAC decomposition with the species distribution (see
section 3.1), it seemed difficult to directly linked an empiric
component to species with known stoichiometry. Indeed, the
two empiric components given by the decomposition had the
same contribution evolution. Correlating components contri-
bution to Pb(ANQ-ST) complex concentration, both compo-
nents had a good correlation with the formed complex (Fig.
9c). In other terms, it appeared that PARAFAC decomposition
was not able to identify and separate real substances. It was
probably because of the too close fluorescence ranges of the
real species. Indeed, a small 3 nm red-shift was observed on
the absorbance spectra when lead (II) was added to ANQ-ST
(Fig. 4) meaning that free ANQ-ST and complexed ANQ-ST
have probably close fluorescence ranges.
½MLŠ
M þ L ⟺ ML with KML
ML þ L ⟺ ML2 with KML
¼
½MŠ:½LŠ
½ML2Š
¼
2
½MLŠ:½LŠ
½LŠ ¼ ½LŠ þ ½MLŠ þ 2½ML2Š
0
½MŠ ¼ ½MŠ þ ½MLŠ þ ½ML2Š
0
The amount of each complex is given as a fraction of the
total distribution and stability constants in Table 1. According
to the distribution diagrams, for ANQ-MMA, 1:1 and 1:2
complexes coexist up to a ratio of 1 between Pb2+ and
ANQ-MMA, then only the 1:1 remains in solution. For
ANQ-ST, whatever the ratio used, only the 1:1 complex is
formed in solution. Stability constants values proved that the
formation of all the complexes is quantitative with a higher
Log (β1) value for ANQ-MMA.
For ANQ-MMA ligand, PARAFAC treatment decomposi-
tion gave an optimum result for three components
(CORCONDIA = 77%). CORCONDIA scores obtained for
2 to 5 components decomposition are given in Table A.
Component 1 presented an excitation wavelength maximum
at 365–385 nm and emission maximum at 420–440 nm.
Component 2 presented an excitation wavelength maximum
at 385–400 nm and emission maximum at 495–505 nm.
Fluorescent Detection of Pb2+
The fluorescence of the molecules was monitored upon the
addition of lead (II). The EEMs showed a massive peak locat-
ed at (λex/λem) = 388/425 nm for ANQ-MMA (Fig. 6) and at
(λex/λem) = 380/420 nm for ANQ-ST (Fig. 7). For both mole-
cules, the relative peak intensity of the EEMs underwent an
important increase from 0 to 100 μmol.L−1 of lead (II) follow-
ed by a saturation for higher concentrations (Fig. 8).
Component 3 showed two maxima: one located at λex/λem
=
410/450 nm and the other one located at λex/λem = 345–350/
a)
c)
g)
d)
b)
f)
e)
Fig. 6 EEMs obtained for acetone-water (4:1, v/v) solutions containing 50 μmol.L−1 of ANQ-MMA and (a) 0, (b) 5, (c) 10, (d) 30, (e) 50, (f) 100 or (g)
300 μmol.L−1 of Pb2+

J Fluoresc
Fig. 7 EEMs obtained for acetone-water (4:1, v/v) mixtures solutions containing 50 μmol.L−1 of ANQ-STand (a) 0, (b) 5, (c) 10, (d) 30, (e) 50, (f) 100
or (g) 300 μmol.L−1 of Pb2+
445–455 nm (Fig. 10a). The contribution of each component
to each EEM measurements is depicted in Fig. 10b. Coupling
PARAFAC decomposition with the species distribution (see
section 3.1), it was possible to discriminate the 3 PARAFAC
components, compared to species of known stoichiometry.
Looking at the components contribution, it was clear that none
of them presented the variation corresponding to Pb(ANQ-
MMA)2 (Fig. 5). Furthermore, no correlation could be pointed
out for this complex (Fig. 10c). On another hand, components
1 and 3 could be linked to Pb(ANQ-MMA), looking at their
correlation (Fig. 10d). Yet, two arguments clearly led to the
conclusion that component 3 is directly linked to the predom-
inant complex Pb(ANQ-MMA). First, component 3 concen-
tration is increasing starting from zero, unlike component 1.
Then, the comparison of the maximum excitation wavelengths
observed in these components EEM (Fig. 10a) with the 25 nm
red-shift observed on absorbance spectra when lead (II) was
added to ANQ-MMA (Fig. 1) indicated that component 3 was
more likely to be associated to Pb(ANQ-MMA) complex. For
ANQ-MMA, it was therefore possible to link the fluorescent
modelled component to an existing complex and to further
demonstrate its stoichiometry.
Competition Analysis
The fluorescence properties of ANQ-MMA and ANQ-ST
were measured in presence of different metal ions. Results
showed that both fluoroionophores could almost selectively
recognize Pb2+ via fluorescence “off-on” responses. Indeed,
an exaltation of the fluorescence intensity on the EEMs was
observed only when Pb2+ was added to ANQ-MMA or ANQ-
ST. For ANQ-MMA, a fluorescent signal was also recorded in
Fig. 8 Relative fluorescence
intensity of acetone/water
a)
b)
120
250
(4:1, v/v) solutions containing
50 μmol.L−1 of (a) ANQ-MMA
or (b) ANQ-ST upon the
addition of lead ion
225
200
175
150
125
100
115
110
105
100
0
50
100
150
200
250
300
0
50
100
150
200
250
300
[Pb2+] (μmol.L-1)
[Pb2+] (μmol.L-1)

J Fluoresc
Fig. 9 (a) Component 1 and 2
obtained with PARAFAC
a)
decomposition with an optimum
CORCONDIA score, (b) their
contribution to fluorescence for
ANQ-ST lead range study and (c)
their contribution correlated to
Pb(ANQ-ST) concentration as
calculated in Fig. 2
b) 650
600
Component 1
Component 2
550
500
450
400
350
0
1
2
3
4
5
6
[Pb2+]/[ANQ-ST]
600
500
400
300
c)
R² = 0.9764
Composant 1
Composant 2
R² = 0.9753
8.00E-06
1.20E-05
1.60E-05
[Pb(ANQ-ST)] (mol.L-1)
presence of Na+ (Online Resource, Fig. Q) and for ANQ-STin
presence of Co2+ (Online Resource, Fig. R) but the extent of
the measured fluorescence was clearly below that obtained in
presence of Pb2+.
MMA and ANQ-ST can act as selective probes for fluores-
cence detection of Pb2+.
To determine the effect of other metal ions on the selectiv-
ity of ANQ-MMA and ANQ-ST for Pb2+, competition exper-
iments were carried out by measuring the fluorescence behav-
iour of the molecules in presence of Pb2+ ions and another
metal ion (Ag+, Na+, Ca2+, Cd2+, Co2+, Cu2+, Zn2+, Al3+ or
Fe3+) (Online Resource, Fig. S and Fig. T). As seen in Fig. 11,
for both molecules, all solutions containing an interfering ion,
except for Fe3+, showed very weak variation of fluorescence
intensity compared to the fluorescence intensity of the solu-
tion containing lead (II). These results demonstrate that ANQ-
Conclusion
Two fluorescent ligands specific to lead (II) were synthesized:
ANQ-MMA and ANQ-ST. Their turn-on fluorescence re-
sponse upon the addition of Pb2+ was studied. The decompo-
sition of the fluorescence excitation-emission matrices
allowed to give the number of fluorescent components in-
volved in the fluorescent signal. The coupling of those results
with the complex distribution of ANQ-MMA and ANQ-ST
with lead (II) established by a UV-visible study allowed to link

J Fluoresc
Fig. 10 (a) Component 1, 2 and 3
obtained with PARAFAC
a)
decomposition with an optimum
CORCONDIA score and (b) their
contribution to fluorescence for
ANQ-MMA lead range study and
their contribution correlated to (c)
Pb(ANQ-MMA)2 and (d)
Pb(ANQ-MMA) concentration as
calculated in Fig. 2
40
30
20
10
0
b)
Composant 1
Composant 2
Composant 3
0
1
2
3
4
5
6
[Pb2+]/[ANQ-MMA]
d)
c) 40
40
Composant 1
Composant 2
Composant 3
Composant 1
Composant 2
Composant 3
35
30
25
20
15
10
5
35
30
25
20
15
10
5
R² = 0.9006
R² = 0.8666
R² = 0.6655
R² = 0.6387
R² = 0.9047
R² = 0.8689
0
0
5.E-06
6.E-06
7.E-06
8.E-06
9.E-06
4.E-06
5.E-06
[Pb(ANQ-MMA)](mol.L-1)
6.E-06
7.E-06
8.E-06
[Pb(ANQ-MMA)2] (mol.L-1)
a modelled fluorescent component to a real complex of a
known stoichiometry. Comparing fluorescence decomposi-
tion and UV-visible study appeared to be a new prom-
ising way to identify species. Moreover, the two
synthesized ligands showed interesting properties which
could turn them as interesting probes, specific for lead
(II). As a matter of fact, in addition to their turn-on
fluorescence responses upon the addition of lead (II),
50 µmol.L-1 ANQ-MMA + 100 µmol.L-1 Pb2+
50 µmol.L-1 ANQ-ST + 100 µmol.L-1 Pb2+
a)
50 µmol.L-1 ANQ-MMA + 100 µmol.L-1 Pb2+ + 100 µmol.L-1 metal ions
50 µmol.L-1 ANQ-ST + 100 µmol.L-1 Pb2+ + 100 µmol.L-1 metal ions
b)
125
100
75
50
25
0
300
250
200
150
100
50
0
Pb2+
Cu2+ Zn2+
Pb2+
Na+
Pb2+
Cd2+
Pb2+
Co2+
Pb2+
Pb2+
Al3+
Pb2+
Pb2+
Ag+
Pb2+
Ca2+
Pb2+
Fe3+
Pb2+
Cu2+
Pb2+
Na+
Pb2+
Cd2+
Pb2+
Co2+
Pb2+
Zn2+
Pb2+
Al3+
Pb2+
Pb2+
Ag+
Pb2+
Ca2+
Pb2+
Fe3+
Fig. 11 Competition analysis of (a) ANQ-MMA and (b) ANQ-ST (50 μmol.L-1) in acetone-water (4:1, v/v) at room temperature

J Fluoresc
16. Goswami S, Chakrabarty R (Jul. 2010) Highly Selective
Colorimetric Fluorescent Sensor for Pb2+. Eur J Org Chem
2010(20):3791–3795
the performed competition analysis demonstrated their
selectivity.
17. Valeur B, Berberan-Santos MN (2012) Molecular Fluorescence:
Principles and Applications. John Wiley & Sons
18. Valeur B, Leray I Design principles of fluorescent molecular sen-
sors for cation recognition. Coord Chem Rev 205(1):3–40 août
2000
19. Liu Z, He W, Guo Z (Jan. 2013) Metal coordination in
photoluminescent sensing. Chem Soc Rev 42(4):1568–1600
20. de Silva AP et al (Aug. 1997) Signaling Recognition Events with
Fluorescent Sensors and Switches. Chem Rev 97(5):1515–1566
21. Jun EJ, Swamy KMK, Bang H, Kim S-J, Yoon J (May 2006)
Anthracene derivatives bearing thiourea group as fluoride selective
fluorescent and colorimetric chemosensors. Tetrahedron Lett
47(18):3103–3106
22. Anand T, Sivaraman G, Mahesh A, Chellappa D (Jan. 2015)
Aminoquinoline based highly sensitive fluorescent sensor for
lead(II) and aluminum(III) and its application in live cell imaging.
Anal Chim Acta 853:596–601
Acknowledgements This work is part of the PREVENT program fi-
nanced by the University of Toulon, Toulon-Provence-Méditerranée and
the Conseil Départemental du Var, France. The authors acknowledge
financial support from the Regional Council of Provence Alpes Côte
d’Azur (France) and Academy of Finland.
References
1. Kim HN, Ren WX, Kim JS, Yoon J (Mar. 2012) Fluorescent and
colorimetric sensors for detection of lead, cadmium, and mercury
ions. Chem Soc Rev 41(8):3210–3244
23. Chalal M, Vervandier-Fasseur D, Meunier P, Cattey H, Hierso J-C
(May 2012) Syntheses of polyfunctionalized resveratrol derivatives
using Wittig and Heck protocols. Tetrahedron 68(20):3899–3907
24. Laatikainen M et al (Jul. 2014) Complexation of Nickel with 2-
(Aminomethyl)pyridine at High Zinc Concentrations or in a
Nonaqueous Solvent Mixture. J Chem Eng Data 59(7):2207–2214
25. Gans P, Sabatini A, Vacca A (1985) SUPERQUAD: an improved
general program for computation of formation constants from po-
tentiometric data. J Chem Soc Dalton Trans 0(6):1195–1200
26. Zepp RG, Sheldon WM, Moran MA (Oct. 2004) Dissolved organic
fluorophores in southeastern US coastal waters: correction method
for eliminating Rayleigh and Raman scattering peaks in excitation–
emission matrices. Mar Chem 89(1):15–36
2. Chen C-T, Huang W-P (Jun. 2002) A Highly Selective Fluorescent
Chemosensor for Lead Ions. J Am Chem Soc 124(22):6246–6247
3. Flegal AR, Smith DR (Jun. 1992) Current needs for increased ac-
curacy and precision in measurements of low levels of lead in
blood. Environ Res 58(1):125–133
4. Wierzbicka M, Antosiewicz D (Jan. 1993) How lead can easily
enter the food chain — a study of plant roots. Sci Total Environ
134:423–429
5. Wang L, Jin Y, Deng J, Chen G (Nov. 2011) Gold nanorods-based
FRET assay for sensitive detection of Pb2+ using 8-17DNAzyme.
Analyst 136(24):5169–5174
6. Wei Y, Liu R, Wang Y, Zhao Y, Cai Z, Gao X (Mar. 2013) Hairpin
oligonucleotides anchored terbium ion: a fluorescent probe to specif-
ically detect lead(II) at sub-nM levels. Analyst 138(8):2302–2307
7. The European Union (1998) Council Directive 98/83/EC of 3
November 1998 on the quality of water intended for human con-
sumption. Off J Eur Communities, vol L 330:32–54, Dec
8. Y. Chen and K. Wang, Azacrown[N,S,O]-modified porphyrin sen-
sor for detection of Ag+, Pb2+, and Cu2+. Photochem Photobiol
Sci, vol. 12, no. 11, pp. 2001–2007, Oct. 2013.
9. Zheng H, Zhan X-Q, Bian Q-N, Zhang X-J (2012) Advances in
modifying fluorescein and rhodamine fluorophores as fluorescent
chemosensors. Chem Commun 49(5):429–447, Dec
10. A. Pal, B. Bag, M. Thirunavoukkarasu, S. Pattanaik, and B. K.
Mishra, Solvent mediated tuning of selectivity in a rhodamine
based probe and bioimaging for Pb(II) detection in plant tissues.
RSC Adv, vol. 3, no. 40, pp. 18263–18266, Sep. 2013.
27. Bro R (Oct. 1997) PARAFAC. Tutorial and applications. Chemom
Intell Lab Syst 38(2):149–171
28. Stedmon CA, Markager S (Mar. 2005) Resolving the variability in dis-
solved organic matter fluorescence in a temperate estuary and its catch-
ment using PARAFAC analysis. Limnol Oceanogr 50(2):686–697
29. Bro R, Kiers HAL (Jun. 2003) A new efficient method for deter-
mining the number of components in PARAFAC models. J
Chemom 17(5):274–286
30. Mounier S, Zhao H, Garnier C, Redon R (Oct. 2011) Copper
complexing properties of dissolved organic matter: PARAFAC
treatment of fluorescence quenching. Biogeochemistry 106(1):
107–116
31. Mahmoodi NO, Mirkhaef S, Ghavidast A (2015) Synthesis of an-
thracene derivatives of 1,3-diazabicyclo[3.1.0]hex-3-ene. J Mol
Struct 1081:248–253, février
11. Sunnapu O, Kotla NG, Maddiboyina B, Singaravadivel S,
Sivaraman G (2015) A rhodamine based ‘turn-on’ fluorescent
probe for Pb(II) and live cell imaging. RSC Adv 6(1):656–660, Dec
12. Xia W-S, Schmehl RH, Li C-J, Mague JT, Luo C-P, Guldi DM
(Jan. 2002) Chemosensors for Lead(II) and Alkali Metal Ions
Based on Self-Assembling Fluorescence Enhancement (SAFE). J
Phys Chem B 106(4):833–843
32. Yu M, He F, Tang Y, Wang S, Li Y, Zhu D (Jun. 2007) Non-Ionic
Water-Soluble Crown-Ether-Substituted Polyfluorene as
Fluorescent Probe for Lead Ion Assays. Macromol Rapid
Commun 28(12):1333–1338
33. Varazo K, Xie F, Gulledge D, Wang Q (Sep. 2008) Synthesis of
triazolyl anthracene as a selective fluorescent chemosensor for the
Cu(II) ion. Tetrahedron Lett 49(36):5293–5296
13. Park J, Kim Y (Jun. 2012) A colorimetric probe for the selective
naked-eye detection of Pb(II) ions in aqueous media. Analyst
137(14):3246–3248
14. ChenY,JiangJ(Jun.2012)Porphyrin-basedmulti-signalchemosensors
for Pb2+ and Cu2+. Org Biomol Chem 10(24):4782–4787
15. Pandey R, Gupta RK, Shahid M, Maiti B, Misra A, Pandey DS
(Jan. 2012) Synthesis and Characterization of Electroactive
Ferrocene Derivatives: Ferrocenylimidazoquinazoline as a
Multichannel Chemosensor Selectively for Hg2+ and Pb2+ Ions
in an Aqueous Environment. Inorg Chem 51(1):298–311
34. Lin Y-I, Lang SA, Seifert CM, Child RG, Morton GO, Fabio PF
Aldehyde Syntheses. Study of the preparation of 9,10-
anthracenedicarboxaldehyde. J Organomet Chem 44(25):4701–
4703 décembre 1979
35. Ryan DK, Weber JH (May 1982) Fluorescence quenching titration
for determination of complexing capacities and stability constants
of fulvic acid. Anal Chem 54(6):986–990
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.