Mercury detection in a microfluidic device by using a molecular sensor soluble in organoaqueous solvent

Article abstract of DOI:10.1039/c2pp25103e

A series of fluorescent sensor molecules based on a phosphane sulfide derivative that is soluble in an organoaqueous solvent were designed and synthesized. The structure of the fluorophore has been optimized in order to have the best compromise in terms of solubility and photophysical properties. The obtained properties are in full agreement with quantum chemical calculations. A fluorescent molecular sensor containing one polyoxoethylene group has been synthesized and an efficient quenching upon mercury complexation has been observed. Finally, this sensing molecule has been introduced in a microfluidic chip in which fluorescence detection has been integrated. An efficient fluorescence response was observed upon mercury addition. The Royal Society of Chemistry and Owner Societies 2012.

Full text of DOI:10.1039/c2pp25103e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Photochemical &
Photobiological Sciences
Cite this: Photochem. Photobiol. Sci., 2012, 11, 1737
www.rsc.org/pps
PAPER
Mercury detection in a microfluidic device by using a molecular sensor
soluble in organoaqueous solvent†
Djibril Faye, Haitao Zhang, Jean-Pierre Lefevre, Jérémy Bell, Jacques A. Delaire and Isabelle Leray*
Received 13th April 2012, Accepted 23rd July 2012
DOI: 10.1039/c2pp25103e
A series of fluorescent sensor molecules based on a phosphane sulfide derivative that is soluble in an
organoaqueous solvent were designed and synthesized. The structure of the fluorophore has been
optimized in order to have the best compromise in terms of solubility and photophysical properties. The
obtained properties are in full agreement with quantum chemical calculations. A fluorescent molecular
sensor containing one polyoxoethylene group has been synthesized and an efficient quenching upon
mercury complexation has been observed. Finally, this sensing molecule has been introduced in a
microfluidic chip in which fluorescence detection has been integrated. An efficient fluorescence response
was observed upon mercury addition.
the bisaryl thiophosphinoethane derivative DPPS1 was found to
Introduction
be very selective and sensitive for the Hg2+ ion. In order to
Heavy metal ions such as mercury are highly toxic pollutants to
the environment and are extremely dangerous for human health.
A wide variety of symptoms including digestive, cardiac, kidney
and especially neurological diseases result from a series of bio-
logical effects.1 Thus, the amount of mercury found in the
environment is the subject of several official reports. The World
Health Organization established in 2004 a guideline for drinking
water quality which included a mercury maximal value of
develop a portable fluorescent sensor based on a “lab on chip”
system, we have examined the possibility of using this family of
fluorescent sensors in a microfluidic device; however, the poor
solubility in water of DPPS1 limits its use in a “lab on chip”
system even in an organoaqueous solvent (acetonitrile/water
mixture). Thus, we have considered the design of new hydrophi-
lic fluorescent molecular sensors. This paper describes their syn-
thesis, their photophysical properties, their behavior in the
presence of Hg2+ ions, and finally their implementation in a
microfluidic device.
In our molecular design strategy, we first envisaged to syn-
thesize model compounds substituted with polyethylene glycol
in order to obtain the best solubility in organoaqueous solvents
and photophysical properties (PS-PEG, PS-3PEG) (Scheme 1).
With the incorporation of a fluorescent molecular sensor in the
microfluidic set-up in mind, the fluorophore must be excited by a
UV LED at 365 nm. We report afterwards the synthesis and the
photophysical properties of a chelator containing a thiophos-
1 μg L−1 2
.
Whereas a wide set of analytical methods including induc-
tively coupled plasma spectroscopy, atomic absorption, atomic
emission and electrochemical detection have been developed for
indoor determinations, the development of new, inexpensive and
real time monitoring methods for the detection of Hg2+ still
deserves interest. In this context, optical methods based on fluor-
escent molecular chemosensors have attracted ever increasing
attention.3 Separately, microfluidic systems are increasingly used
in the fields of analytical and biomedical sciences. Their main
advantages are their small size, the low consumption of reac-
tants, and their potential to develop portable devices. The combi-
nation of a specific fluorescent sensor and a microfluidic system
is a powerful approach to develop a selective sensor which has
been previously used for the detection of potassium4 and lead5
by our group. In our previous work, we have reported the syn-
thesis and the photophysical characterization of fluorescent mo-
lecular sensors based on phosphane oxide,6–9 phosphane
sulfide,10–12 and phosphane selenide13 derivatives. In particular,
phane group substituted with
a polyoxoethylene group
(DPPS-PEG). Finally the use of this fluorescence molecular
sensor in a microfluidic chip is described in order to make a por-
table sensor able to detect a low amount of mercury in water.
Experimental
General procedures
Acetonitrile (99.9%) from Aldrich (spectrometric grade) and
Millipore filtered water (conductivity < 6 × 10−8 Ω−1 cm−1 at
20 °C) were used for the fluorescence measurements either in a
cuvette or in the microcircuit. The different solutions were pre-
pared with perchloric acid (99.99%) at 70% in water. Mercury
PPSM, ENS-Cachan, 61 avenue du Président Wilson, 94235 Cachan
Cedec, France. E-mail: Isabelle.LERAY@ens-cachan.fr;
Fax: +33 1 47 40 24 54; Tel: +33 1 47 40 27 04
†This article is published as part of a themed issue in honour of Jean-
Pierre Desvergne on the occasion of his 65th birthday.
This journal is © The Royal Society of Chemistry and Owner Societies 2012
Photochem. Photobiol. Sci., 2012, 11, 1737–1743 | 1737

View Article Online
Scheme 1 Structure of the used molecules PS-3PEG, PS-PEG and DPPS-PEG.
perchlorate from Aldrich was of the highest quality available and
vacuum dried over P2O5 prior to use.
combined organic phases were dried (MgSO4) and concentrated
under reduced pressure. The resulting crude solid was purified
by chromatography (AcOEt then CH2Cl2/MeOH 9 : 1) to afford
the corresponding 3a as a yellow solid (170 mg, 57%).
1H NMR (400 MHz, CDCl3), δ (ppm): 3.00 (s, 1H), 3.37 (s,
9H), 3.53–3.54 (m, 6H), 3.69–3.72 (m, 6H), 3.72 (m, 6H), 4.41
(m, 6H), 6.72 (s, 2H).
Synthesis
Trimethyl ((3,4,5-tris(2-(2-methoxyethoxy)ethoxy)phenyl)-
ethynyl)silane 2a. A Schlenk flask was charged with compound
1a (450 mg, 0.88 mmol, 1 equiv.) in 10 mL of distilled piper-
idine. The mixture was degassed with argon. Trimethylsilylace-
tylene (130 mg, 1.3 mmol, 1.5 equiv.), CuI (8.3 mg,
0.044 mmol, 0.05 equiv.) and Pd(PPh3)4 (20 mg, 0.018 mmol,
0.02 equiv.) were then added and the mixture was stirred at
70 °C for 24 h, then cooled to room temperature. After evapor-
ation of the solvent, the residue was dissolved in CH2Cl2. The
organic phase was filtered over celite, washed with water
(150 mL) and dried over MgSO4. After removal of the solvent,
chromatography using silica gel with a gradient from 100%
cyclohexane to a 10 : 90 mixture cyclohexane/AcOEt gave the
desired compound 2a as a brown powder (345 mg, 74%).
1H NMR (400 MHz, CDCl3), δ (ppm): 0.19 (s, 9H); 3.33 (s,
9H); 3.48–3.51 (m, 6H); 3.63–3.67 (m, 6H); 3.78–3.82 (m, 6H);
4.09–4.12 (m, 6H); 6.65 (s, 2H).
13C NMR (100 MHz, CDCl3), δ (ppm): 58.2, 67.8,
68.3–71.4.
1-Ethynyl-4-(2-(2-(2-methoxyethoxy)ethoxy)ethoxy)benzene 3b.
Same procedure as for 3a starting from compound 2b to afford
the corresponding 3b as a white oil (940 mg, 61%).
1H NMR (400 MHz, CDCl3), δ (ppm): 2.28 (s, 1H), 3.37 (s,
3H), 3.54 (t, J = 4.6 Hz, 2H), 3.64–3.72 (m, 6H), 3.84 (t, J =
4.6 Hz, 2H), 4.13 (t, J = 5.0 Hz, 2H), 6.86 (d, J = 8.7 Hz, 2H),
7.42 (d, J = 8.7 Hz, 2H). 13C NMR (100 MHz, CDCl3), δ
(ppm): 59.2, 67.9, 69.9, 70.7–71, 72.1, 83.7, 114.2, 117.2,
132.7, 159.4.
5-((4-Iodophenyl)ethynyl)-1,2,3-tris(2-(2-methoxyethoxy)ethoxy)-
benzene 4a. A Schlenk flask was charged with compound 3a
(160 mg, 0.29 mmol, 1 equiv.), Pd(PPh3)4 (9.9 mg, 0.009 mmol,
0.03 equiv.), CuI (3.3 mg, 0.017 mmol, 0.06 equiv.) and
1,4-diiodobenzene (380 mg, 1.15 mmol, 4 equiv.) in a mixture
of anhydrous toluene (5 mL) and distilled triethylamine (1 mL).
The mixture was exhaustively degassed with argon and stirred at
80 °C overnight. The organic phase was filtered over celite,
washed with water (20 mL) and dried over MgSO4. After
removal of the solvent, chromatography using silica gel with a
gradient from AcOEt to a 90 : 10 mixture AcOEt/EtOH gave the
desired compound 4a as a brown solid (110 mg, 58%).
1H NMR (400 MHz, CDCl3), δ (ppm): 3.38 (s, 9H), 3.5–3.6
(m, 6H), 3.7–3.73 (m, 6H), 3.83–3.88 (s, 6H), 4.16–4.21 (m,
6H), 6.76 (s, 2H), 7.23 (d, J = 7.9 Hz, 2H), 7.69 (d, J = 8.7 Hz,
2H).
13C NMR (100 MHz, CDCl3), δ (ppm): 0.0 58.2, 67.7,
68.2–71.4, 110.1, 151.91.
((4-(2-(2-(2-Methoxyethoxy)ethoxy)ethoxy)phenyl)ethynyl)
trimethylsilane 2b. The same procedure as for 2a starting from
1-bromo-4-(2-(2-(2-methoxyethoxy)ethoxy)ethoxy)benzene to
give the desired compound 2b as a white oil (1.98 g, 82%).
1H NMR (400 MHz, CDCl3), δ (ppm): 0.23 (s, 9H), 3.37 (s,
3H), 3.55 (t, J = 4.1 Hz, 2H), 3.6–3.71 (m, 6H), 3.84 (t, J =
5.5 Hz, 2H), 4.11 (t, J = 5.5 Hz, 2H), 6.82 (d, J = 8.7 Hz, 2H),
7.39 (d, J = 8.7 Hz, 2H).
13C NMR (100 MHz, CDCl3), δ (ppm): 19.4, 59.2, 67.9,
69.9, 70.7–70.9, 72.1, 83.8, 114.3, 117.2, 132.7, 159.4.
13C NMR (100 MHz, CDCl3), δ (ppm): 58.2, 67.8,
68.2–71.4, 93.3–93.4, 99.6, 110.1, 128.5, 137.6, 151.9.
5-Ethynyl-1,2,3-tris(2-(2-methoxyethoxy)ethoxy)benzene 3a.
A solution of 2a (345 mg, 0.65 mmol, 1 equiv.) and K2CO3
(360 mg, 2.6 mmol, 4 equiv.) in a mixture of CH2Cl2 (4 mL)
and MeOH (6 mL) was stirred at room temperature for 4 h and
quenched with water and then extracted with ethyl acetate. The
1-Iodo-4-((4-(2-(2-(2-methoxyethoxy)ethoxy)ethoxy)phenyl)-
ethynyl)benzene 4b. Same procedure as for 4a to give the
1738 | Photochem. Photobiol. Sci., 2012, 11, 1737–1743
This journal is © The Royal Society of Chemistry and Owner Societies 2012

View Article Online
desired compound 4b as a brown solid (680 mg, 50%). T: 72 °C.
1H NMR (400 MHz, CDCl3), δ (ppm): 3.38 (s, 3H), 3.55 (t,
J = 4.6 Hz, 2H), 3.66–3.74 (m, 6H), 3.87 (t, J = 5.0 Hz, 2H),
4.15 (t, J = 5.0 Hz, 2H), 6.88 (d, J = 7.8 Hz, 2H), 7.21 (d, J =
8.2 Hz, 2H), 7.43 (d, J = 7.8 Hz, 2H), 7.66 (d, J = 8.2 Hz, 2H).
13C NMR (100 MHz, CDCl3), δ (ppm): 59.1, 67.5, 69.4,
70.6–71.1, 72.1, 93, 99.7, 115.4, 122.60, 133.0–133.1, 137.5,
159.2.
13C NMR (400 MHz, CDCl3), δ (ppm): 38.8, 59.4, 67.5,
69.7, 70.6–70.7, 72.0, 114.9, 131.8–133.1.
31P NMR (162 MHz, CDCl3, δ (ppm): 44.7 (s, PvS).
Molecular modelling
All calculations were performed with the Gaussian 03 program14
on a Nec TX7 with 32 processors (Itanium 2 of the MESO
centre of the ENS Cachan). All the geometry optimizations were
performed considering a water cavity using (IEFPCM) level and
the hybrid density functional B3LYP potential with the 6-31G
basis set, as implemented in the Gaussian03 software package.
Orbital surfaces have been generated with the cubgen module of
Gaussian03 and visualized with GaussView 3.0 of Gaussian Inc.
PS-3PEG. A Schlenk flask was charged with compound 5
(5.6 mg, 0.015 mmol,
1 equiv.), Pd(PPh3)4 (1.4 mg,
0.0012 mmol, 0.08 equiv.), CuI (1.2 mg, 0.006 mmol, 0.04
equiv.), and compound 4a (50 mg, 0.075 mmol, 5 equiv.) in a
mixture of anhydrous toluene (1 mL) and distilled triethylamine
(0.2 mL). The mixture was exhaustively degassed with argon
and stirred at 80 °C for 34 h then cooled to room temperature.
The organic phase was filtered over celite, washed with water
(10 mL) and dried over MgSO4. After removal of the solvent,
chromatography using silica gel with a gradient from cyclo-
hexane to a 60 : 40 mixture cylohexane/AcOEt gave the desired
compound PS-3PEG as a brown solid (5 mg, 17%).
2.2. Spectroscopic measurements
UV/Vis absorption spectra were recorded on a Varian Cary5E
spectrophotometer. Corrected emission spectra were obtained on
a Spex Fluorolog 1681 spectrofluorometer. The thermodynamic
constants were determined using the Specfit Global Analysis
System V3.0 for a 32-bit Windows system, which provided the
complexation constants with a good accuracy. This software uses
singular value decomposition and non-linear regression model-
ing by the Levenberg–Marquardt method.15 The fluorescence
quantum yields were determined using quinine sulfate dihydrate
in sulfuric acid (0.5 N) as a standard (ΦF = 0.546).16
1H NMR (400 MHz, CDCl3), δ (ppm): 3.38 (s, 27H),
3.55–3.56 (m, 18H), 3.71–3.72 (m, 18H), 3.73–3.86 (m, 18H),
4.17–4.18 (m, 18H), 6.76 (s, 6H), 7.26 (d, J = 8.2 Hz, 6H),
7.46–7.47 (m, 6H), 7.67–7.69 (m, 12H).
13C NMR (100 MHz, CDCl3), δ (ppm): 59.2, 68.9, 69.7,
70.8, 72.1, 90.3, 94.2, 111.2, 118.3, 122.8, 128.5, 132.1, 137.6,
139.5.
31P NMR (162 MHz, CDCl3, δ (ppm): 43.9 (s, PvS).
2.3. Microfluidic chip and fluorescence detection
PS-PEG. Same procedure as PS-3PEG starting from com-
pound 4b to give the desired compound PS-PEG as a brown
solid (27 mg, 56%). T: 101 °C.
The device, shown in Scheme 2, is based on our previously
described set-up.5 It is essentially made of a Y-shape micro-
channel moulded in PDMS and fixed on a glass substrate. In
order to ensure a complete complexation reaction,17 we have
used a 1.3 m long microchannel. The excitation of the fluor-
escent molecules was achieved by two light emitting diodes (UV
LED365-10 Roitner, 365 nm) whose power supply, modulated
with a square signal (77 Hz), allowed heterodyne detection of
fluorescence. Fluorescent light was collected through a bundle of
7 optical fibres (400/440 μm) and focused through a high pass
filter (OG515) at the entrance of a PM tube (Hamamatsu R928).
The voltage signal was routed to a lock-in amplifier (Signal
Recovery 7265 DSP) and acquired from the lock-in on a PC
computer via an RS232 bus; it was in the order of one hundred
millivolts and far from a signal-to-noise ratio of one. When
water was circulating through the chip, there was no detectable
fluorescence and the background signal due to the scattering of
the excitation light was less than 5 mV, to be compared with the
minimum signal due to fluorescence which was 100 mV.
1H NMR (400 MHz, CDCl3), δ (ppm): 3.37 (s, 9H),
3.54–3.55 (m, 6H), 3.64–3.73 (m, 18H), 3.85–3.89 (m, 6H),
4.14 (t, J = 5.0 Hz, 6H), 6.9 (d, J = 9.16 Hz, 6H), 7.44 (d, J =
8.7 Hz, 6H), 7.48–7.50 (m, 6H, H-10), 7.66–7.71 (m, 18H).
13C NMR (400 MHz, CDCl3), δ (ppm): 59, 70.5, 72.9, 104.6,
115.1, 131.8–138.1, 138.1.
31P NMR (162 MHz, CDCl3), δ (ppm): 43.21 (s, PvS).
DPPS-PEG. A Schlenk flask was charged with 1,2-bis-
(bis(4-ethynylphenyl)phosphorothioyl)ethane
6
(50 mg,
0.089 mmol, 1 equiv.), Pd(PPh3)4 (16 mg, 0.0014 mmol, 0.16
equiv.), CuI (3 mg, 0.0014 mmol, 0.16 equiv.) and 4b (271 mg,
0.58 mmol 6.5 equiv.) in a mixture of anhydrous toluene (2 mL)
and triethylamine (0.4 mL). The mixture was degassed
with argon and stirred at 80 °C overnight then cooled to
room temperature. The organic phase was filtered over celite,
washed with water (10 mL) and dried over MgSO4. After
removal of the solvent, chromatography using silica gel with a
gradient from AcOEt to a 95 : 5 mixture AcOEt/EtOH gave the
desired compound DPPS-PEG as a brown solid (78 mg, 45%).
T: 125 °C
Results and discussion
Synthesis and photophysical studies of the model compounds
PS-PEG and PS-3PEG
1H NMR (400 MHz, CDCl3), δ (ppm): 3.38 (s, 12H),
3.64–3.70 (m, 16H, H-5, H-4), 3.73 (t, J = 5.0 Hz, 8H), 3.87 (t,
3J2–1 = 4.6 Hz, 8H,), 4.15 (t, J = 4.2 Hz, 8H), 6.89 (d, J = 8.4
Hz, 8H), 7.44 (d, J = 8.7 Hz, 8H), 7.46–7.49 (m, 16H),
7.57–7.60 (m, 8H), 7.77–7.81 (m, 8H).
The phosphane sulfide derivatives substituted with one
(PS-PEG) or three (PS-3PEG) polyoxoethylene groups were
synthesized from the bromo-derivatives 1a18 or 1b19 with the
This journal is © The Royal Society of Chemistry and Owner Societies 2012
Photochem. Photobiol. Sci., 2012, 11, 1737–1743 | 1739

View Article Online
Scheme 2 Left: schematic diagram of the microfluidic device for mercury detection; right: photograph of the microchip.
Fig. 1 Absorption and corrected normalized emission spectra of
PS-3PEG in H2O pH 4 and PS-PEG in CH3CN/H2O 60 : 40, v/v, pH 4.
phosphane sulfide derivative PS-3PEG in 17% isolated yield
and PS-PEG in 56% yield.
In order to evaluate the more appropriate structure for the
detection of mercury in water using a microfluidic device, we
studied the photophysical properties of PS-3PEG and PS-PEG.
The absorption and emission spectra of these compounds are dis-
played in Fig. 1 and their photophysical characteristics are sum-
marized in Table 1. The fluorophores PS-3PEG and PS-PEG
show an intense absorption band in the near-UV region and the
absorption maximum of PS-3PEG is blue-shifted in comparison
with PS-PEG, which means that the electron-donating character
of the phenoxy group is reduced in the case of PS-3PEG. The
fluorescence quantum yields of these two compounds are equiv-
alent, and as observed in absorption, the emission spectrum of
PS-3PEG is blue shifted compared to PS-PEG. Such shifts are
Scheme 3 Synthesis of PS-3PEG and PS-PEG.
synthetic strategy shown in Scheme 3. Sonogashira coupling
between 1a or 1b with trimethysilane acetylene affords com-
pounds 2a in 74% yield or 2b in 82% yield. Theses compounds
can be deprotected under basic conditions to obtain the alkyne
derivatives 3a or 3b. Another coupling with diiodobenzene leads
to the formation of 4a in 58% yield and 4b in 50% yield. Sono-
gashira coupling between the aryliodides 4a or 4b and com-
pound 5 was performed in the presence of PdCl2, (PPh3)2 and
CuI in diisopropylamine/toluene at 80 °C and led to the
1740 | Photochem. Photobiol. Sci., 2012, 11, 1737–1743
This journal is © The Royal Society of Chemistry and Owner Societies 2012

View Article Online
Table 1 Photophysical properties of PS-3PEG and PS-PEG
λabs (nm)
λem (nm)
ΦF
PS-3PEG
PS-PEG
297
336
319
343
0.56
0.43
Scheme 4 Calculated geometry (a) and frontier orbitals (b) of the
fluorophore of PS-PEG.
consistent with the quantum chemical calculations made by
DFT.
The geometry optimizations of PS-PEG and PS-3PEG lead
to two linear structures for the π-conjugated system. However, in
the case of PS-3PEG, the polyoxoethylene group in the para
position is out of the plane because of steric effects. The highest
occupied molecular orbital (HOMO) is always a π orbital
centred on the phenoxy group, which reflects its electron-
donating ability. The lowest unoccupied orbital (LUMO) is
found at a higher energy and the transition between HOMO and
LUMO orbitals illustrates the efficient charge transfer from the
phenoxy group to the phosphane sulfide derivative. For
PS-3PEG, the donor effect of the two polyoxoethylene groups
in the meta positions modifies the distribution of the HOMO
which reduces the charge transfer character of the molecule in
agreement with the measured absorption spectrum (Schemes
4 and 5).
Scheme 5 Calculated geometry (a) and frontier orbitals (b) of the
fluorophore of PS-3PEG.
Thus, in order to make a fluorescent sensor which is suitable
for incorporation into a microfluidic device, we have focused our
effort on the synthesis of a chelate containing only one poly-
oxoethylene chain.
Scheme 6 Synthesis of DPPS-PEG.
complexation studies). Furthermore, since this compound does
not contain any pH sensitive function, no effect was observed
upon changing the pH of the solution. The evolution of the emis-
sion spectra of DPPS-PEG upon complexation with Hg2+ is dis-
played in Fig. 2. The inset in Fig. 2 shows a decrease of the
fluorescence intensity at 444 nm upon progressive addition of
Hg2+. The kinetics of the complexation have been studied and an
equilibration time of 250 s is needed to obtain a plateau.
Mercury complexation induces a 4 nm bathochromic shift of the
absorption spectra, because of an enhancement of the electron-
withdrawing character of the phosphane sulfide and oxide
Synthesis and photophysical studies of the fluorescent molecular
sensor
Scheme 6 outlines the synthesis of the fluorescent sensor
DPPS-PEG. Sonogashira coupling between the chelate 610 and
the iodide derivative 4b was then performed and afforded the
corresponding DPPS-PEG in 45% yield. Complexation studies
of mercury with DPPS-PEG were performed in a mixture of
CH3CN/H2O 6 : 4 at pH 4.0 with practical applications in mind.
The choice of pH 4 was made in order to prevent precipitation of
the mercury cation (especially for the stock solution for the
This journal is © The Royal Society of Chemistry and Owner Societies 2012
Photochem. Photobiol. Sci., 2012, 11, 1737–1743 | 1741

View Article Online
Fig. 2 Corrected emission spectra of DPPS-PEG (1.2 × 10−6 M) in
the presence of Hg2+ in CH3CN/H2O 6 : 4 at pH 4.0, λexc = 365 nm.
Fig. 3 Calibration curves for the fluorimetric determination of mercury
in the microfluidic device ([DPPS-PEG] = 5 × 10−7 mol L−1, solvent
CH3CN/H2O 80 : 20, v/v, pH 4). Inset: Fluorescence response upon
cation concentration.
Inset: Calibration curve as a function of mercury concentration, λem
=
444 nm.
groups due to interaction of their sulfur atoms with the Hg2+.
A strong decrease in fluorescence is observed upon mercury
complexation which can be attributed to an electron transfer
from the excited fluorophore to the complexed mercury cation,
as previously shown for DPPES.11 Global analysis of the evol-
ution of the emission spectra upon mercury binding (by means
of the SPECFIT program) discloses the formation of complexes
of different stoichiometries (ML 1 : 1 and M2L 2 : 1); the appar-
ent stability constants are equal to log β11 = 6.1 0.3 and log
β12 = 12.5 0.3 respectively. The observed stoichiometries and
the stability constants are in agreement with the results pre-
viously reported on the compound DPPES.11 According to the
determined calibration curve, a detection limit of 63 nM of Hg2+
was found. We examined the selectivity of DPPS-PEG for Hg2+
with respect to other cations such as K+, Na+ Ca2+, Mg2+, Cu2+,
Ag+ and Cd2+. Since an efficient quenching is observed upon
addition of 10 equiv., this effect is negligible upon addition of
100 equiv. of interfering cation and the quenching is less
than 12%.
consisting of bas-relief herringbone grooves on the microchannel
is not needed. In order to confirm that efficient mixing occurs in
the microfluidic chip, we compared the obtained fluorescent
signal to one of our previously developed fluorescent sensors,
Calix-DANS4,17 obtained by mixing outside (before the intro-
duction of the two branches of the Y type microchannel) or
inside the microchip. The fluorescence signals recorded at the
output of the microchannel were the same, which indicates good
mixing.
A calibration plot for the determination of mercury in the
microfluidic device shown above (Fig. 3) was then constructed.
A solution of DPPS-PEG (5 × 10−7 mol L−1) in CH3CN/H2O
80 : 20 (v/v) containing HClO4 at measured pH = 4 was intro-
duced in one channel at a flow rate of 0.25 mL h−1, and solu-
tions of mercury perchlorate at concentrations between 0 and
10−5 mol L−1 in CH3CN/H2O (80 : 20, v/v) at the same pH were
introduced in the second channel at the same flow rate.
The DPPS-PEG fluorescence signals recorded for increasing
concentrations of Hg2+ are depicted in Fig. 3. As observed in
solution, a fluorescence quenching is observed upon complexa-
tion with mercury. The inset of Fig. 3 shows the calibration plot,
i.e. the plot of fluorescence intensity versus mercury concen-
tration in the input solution. According to this calibration curve,
the detection limit was 10 nM.
Fluorimetric detection in the microfluidic device
We then examined the possibility of introducing our fluorescent
sensing molecule into a PDMS microchip. In order to avoid any
specific adsorption of the fluorescent sensor on the walls of the
PDMS channel, the following conditions were chosen: mixture
CH3CN/H2O 80 : 20 with Triton X100 (1.25 × 10−4 mol L−1).
Since we have established that the rate of the complexation is
relatively slow (250 s for the used concentrations), a long micro-
channel (1.3 m) was designed and manufactured in PDMS. The
residence time has been estimated to be 420 s, so it is sufficient
to allow a complete complexation reaction. Inside this long
microreactor, efficient mixing between the solution and the fluor-
escent molecule occurs and the use of a passive mixer20
Conclusions
We have described the synthesis, and the photophysical and
complexing properties of fluorescent molecular sensors based on
phosphorus ligands which are soluble in organoaqueous sol-
vents. The photophysical studies revealed interesting properties,
particularly that the position of the absorption band is dependent
on the number of alkoxy groups present on the phenyl moiety.
An efficient fluorescence quenching has been observed in
1742 | Photochem. Photobiol. Sci., 2012, 11, 1737–1743
This journal is © The Royal Society of Chemistry and Owner Societies 2012

View Article Online
8 M. H. Ha-Thi, V. Souchon, A. Hamdi, R. Metivier, V. Alain,
K. Nakatani, P. G. Lacroix, J. P. Genet, V. Michelet and I. Leray, Chem.–
Eur. J., 2006, 12, 9056–9065.
9 R. Metivier, R. Amengual, I. Leray, V. Michelet and J. P. Genet, Org.
Lett., 2004, 6, 739–742.
acetonitrile/water media. Furthermore we have shown that these
molecules can be incorporated into a microfluidic device in
order to make a portable fluorescent sensor.
10 M. H. Ha-Thi, M. Penhoat, V. Michelet and I. Leray, Org. Lett., 2007, 9,
1133–1136.
11 M. H. Ha-Thi, M. Penhoat, V. Michelet and I. Leray, Org. Biomol.
Chem., 2009, 7, 1665–1673.
12 M. H. Ha-Thi, V. Souchon, M. Penhoat, F. Miomandre, J. P. Genet,
I. Leray and V. Michelet, Lett. Org. Chem., 2007, 4, 185–188.
13 I. Samb, J. Bell, P. Y. Toullec, V. Michelet and I. Leray, Org. Lett., 2011,
13, 1182–1185.
Acknowledgements
This work was supported by the National Research Agency
program CP2D “FLUOSENSIL” and by the European STREP
grant “microfluid”.
14 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery Jr., T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma,
G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. G. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
GAUSSIAN 03 (Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
15 H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbühler, Talanta,
1985, 32, 95–101.
Notes and references
1 (a) T. C. Hutchinson and K. M. Meena, Lead, Mercury, Cadmium
and Arsenic in the Environment, JohnWiley, New York, 1987;
(b) G. H. Scoullos, M. J. Vonkeman, L. Thorton and Z. Makuch,
Mercury, Cadmium, Lead: Handbook for Sustainable Heavy Metals
Policy and Regulation (Environment and Policy, V. 31), Kluwer Aca-
demic, Norwell, MA, 2001; (c) C. J. Koester and A. Moulik, Anal.
Chem., 2005, 77, 3737; (d) S. D. Richardson and T. A. Temes, Anal.
Chem., 2005, 77, 3807–3838.
2 Guidelines for Drinking-Water Quality, 3rd edn, World Health Organiz-
ation, Geneva, 2004, p. 188.
3 Chemosensors of Ion and Molecule Recognition, ed. J.-P. Desvergne and
A. W. Czarnik, NATO ASI series, Kluwer Academic Publishers,
Dordrecht, 1997; A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson,
A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem.
Rev., 1997, 97, 1515–1566; B. Valeur and I. Leray, Coord. Chem. Rev.,
2000, 205, 3–40.
4 E. Destandau, J. P. Lefevre, A. C. F. Eddine, S. Desportes, M. C. Jullien,
R. Hierle, I. Leray, B. Valeur and J. A. Delaire, Anal. Bioanal. Chem.,
2007, 387, 2627–2632.
16 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991.
17 T. Wu, I. Leray, V. Genot, J. P. Lefevre, A. Korovitch, N. T. Ha-Duong,
J. M. E. Chahine and J. A. Delaire, ChemPhysChem, 2010, 11,
3355–3362.
18 T. Sakurai, K. Shi, H. Sato, K. Tashiro, A. Osuka, A. Saeki, S. Seki,
S. Tagawa, S. Sasaki, H. Masunaga, K. Osaka, M. Takata and T. Aida,
J. Am. Chem. Soc., 2008, 130, 13812–13813.
19 M. R. Vallee, P. Majkut, I. Wilkening, C. Weise, G. Muller and
C. P. Hackenberger, Org. Lett., 2011, 13, 5440–5443.
5 L. Y. Zhao, T. Wu, J. P. Lefevre, I. Leray and J. A. Delaire, Lab Chip,
2009, 9, 2818–2823.
6 M. H. Ha-Thi, M. Penhoat, D. Drouin, M. Blanchard-Desce, V. Michelet
and I. Leray, Chem.–Eur. J., 2008, 14, 5941–5950.
7 V. A.-R. V. Alain-Rizzo, D. Drouin-Kucma, C. Rouxel, I. Samb, J. Bell,
P. Y. Toullec, V. Michelet, I. Leray and M. Blanchard-Desce, Chem.–
Asian J., 2011, 6, 1080–1091.
20 A. D. Stroock, S. K. W. Dertinger, A. Adjari, I. Mezic, H. A. Stone and
G. M. Whitesides, Science, 2002, 295, 647–651.
This journal is © The Royal Society of Chemistry and Owner Societies 2012
Photochem. Photobiol. Sci., 2012, 11, 1737–1743 | 1743