Ratiometric double channel borondipyrromethene based chemodosimeter for the selective detection of nerve agent mimics

Article abstract of DOI:10.1016/j.dyepig.2014.04.011

A new chromo-fluorogenic probe based on the borondipyrromethene dye has been synthesized. The dye has been attached to a sensing unit for the diethylcyanophosphonate and di-isopropylfluorophosphate detection. The new probe has been fully characterized, an

Full text of DOI:10.1016/j.dyepig.2014.04.011

Dyes and Pigments 108 (2014) 76e83
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Ratiometric double channel borondipyrromethene based
chemodosimeter for the selective detection of nerve agent mimics
*
R. Gotor, A.M. Costero , P. Gaviña, Salvador Gil
Centro de Reconocimiento Molecular y Desarrollo Tecnológico, Unidad Mixta Universidad Politécnica de Valencia e Universidad de Valencia and
Departamento de Química Orgánica, Universidad de Valencia, C/Dr. Moliner 50, 46100 Burjassot, Valencia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A new chromo-fluorogenic probe based on the borondipyrromethene dye has been synthesized. The dye
has been attached to a sensing unit for the diethylcyanophosphonate and di-isopropylfluorophosphate
detection. The new probe has been fully characterized, and its optical properties in front of these sim-
ulants have been evaluated. No interference from other organo-phosphorous or other common con-
taminants compounds has been observed in the detection conditions. A portable kit has been developed
and tested to demonstrate its practical application in real-time monitoring not only in solution but also
in gas phase.
Received 24 February 2014
Received in revised form
4 April 2014
Accepted 5 April 2014
Available online 25 April 2014
Keywords:
Nerve agents
Chemodosimeter
Gas
Ó 2014 Elsevier Ltd. All rights reserved.
BODIPY
Fluorescence
naked-eye
1. Introduction
within few minutes after intoxication. Therefore the need of
sensing these agents soon after any leakage is more than justified.
The international concern on the power of destruction of the
denominated chemical warfare has been raised in the last de-
cades. Threats of chemical warfare attacks are increasingly more
common at every war conflict, being used even as war propa-
ganda. The very well known nerve agents Sarin, Soman and
Tabun are popular among armies and terrorist groups [1]. Their
extreme toxicity, together with their low cost and ease of pro-
duction makes these compounds a big concern for governments
and civilians.
The detection and monitoring of these compounds has been
accomplished by means of different technologies: ion mobility
spectroscopy [4], mass spectrometry [5] and proton transfer mass
spectrometry [6] biosensors [7], electrochemical methods [8],
microcantilevers [9], photonic crystals [10], optical-fibre arrays
[11], and nanomaterials such as nanotubes or nanowires [12]; but
most of them have several drawbacks such as complexity, low
portability, high costs or the need of qualified personal to operate
the devices.
Due to the chemical nature of the nerve agents, i.e. organo-
phosphorous derivatives with a good leaving group (Scheme 1),
they irreversibly bind to a free serine residue of acetylcholines-
terase, inhibiting its enzymatic task [2]. After intoxication, acetyl-
choline is over-accumulated in the synaptic junctions of the nerves.
The disorder has immediate effects: muscle relaxation hindering,
hyperventilation and death in the cases of high exposure. Even if
some antidotes have proven to reduce the effects of the over-
accumulation of acetylcholine (atropine) or to reactivate the
enzymatic task of acetylcholinesterase [3], they urge to be applied
The use of chemodosimeters for in-field detection has recently
gained importance due to economical and practical factors
considering that the required instrumentation as well as their
synthesis has proved to be cost-effective and portable, providing
even semiquantitative analysis by the “naked-eye”.
Among the existing probes for nerve agents, the transduction
mechanism usually involves intramolecular charge transfer (ICT)
[13], photoinduced electron transfer (PET) [14a], or Förster
resonance electron transfer (FRET) processes [15], as well as plas-
mon band of gold nanoparticles [16], discolouration of coordination
complexes [17] or a mixture of them suited for colourimetric
arrays [18].
Within our nerve agents detection research, we have devel-
oped diverse sensing units that became part of chromophores
* Corresponding author.
E-mail address: ana.costero@uv.es (A.M. Costero).
http://dx.doi.org/10.1016/j.dyepig.2014.04.011
0143-7208/Ó 2014 Elsevier Ltd. All rights reserved.

R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
77
Scheme 2. Probe
1
and schematic representation of its reactivity towards some
organophosphorous compounds.
2. Results and discussion
2.1. Synthesis of the probe
Probe 1 was synthesized by means of a Sonogashira cross-
coupling between a bromo derivative of the meso-phenyl BODIPY
dye and the sensing unit, functionalized with an ethynyl moiety.
The synthesis of the sensing unit 4 began with the catalytic hy-
drogenation of 2-(2-nitrophenyl)ethanol in the presence of form-
aldehyde and Pd/C. This step was followed by aromatic bromination
of the resulting 2-(2-dimethylaminophenyl)ethanol with NBS cat-
alysed with ammonium acetate to form 2. A Sonogashira cross-
coupling reaction between 2 and trimethylsilylacetylene yielded
the compound 3, which was deprotected prior to use in a mixture of
K2CO3 in methanol to finally yield the sensing unit 4 (see
Scheme 3).
Bromination of dye 5 was performed using one equivalent of
NBS in a DMF/DCM solvent mixture, yielding 60% of dye 6. Dye 5
was previously synthesized from the condensation of 2,4-
dimethylpyrrole and benzaldehyde, followed by oxidation with
DDQ and reaction with BF3$Et2O as reported in the literature [20].
Finally, compound 4 was coupled with the 2-bromo BODIPY de-
rivative 6 in a Sonogashira cross-coupling reaction, using Pd(PPh3)4
as catalyst in toluene/Et3N mixtures.
Scheme 1. Chemical structures of the nerve agents Sarin, Soman, and Tabun, their
simulants diisopropylfluorophosphate (DFP) and diethylcyanophosphonate (DCNP),
and other potential interferents.
such as stilbenes or triarylmethane dyes [13]. Other authors
opted for the fluorogenic sensing, employing pyrene or biphenyl
dyes [14].
The reputation of borondipyrromethene (BODIPY) dyes in
sensing and fluorescent labelling applications, energy transfer
cassettes or laser dyes among others has grown considerably [19]. It
is not surprising if one considers its outstanding optical properties,
such as high and environment-independent quantum yields, strong
extinction coefficients and photostability, together with the ease of
synthesis and functionalization (Table 1).
With the aim of developing a sensible, selective and high stable
chemical warfare chemodosimeter, we have decided to synthesize
the BODIPY based probe 1, (see Scheme 2) bearing a 2-(2-(dime-
thylamino)phenyl)ethanol sensing unit, which has already proved
to give good results regarding sensitivity and selectivity towards
nerve agent simulants [13a].
In the presence of nerve gases, the 2-(2-(dimethylamino)
phenyl)ethanol unit, reacts with the organophosphorous com-
pound through the hydroxyl group forming a phosphoester, which
in a second step is displaced by the nucleophilic attack of the ni-
trogen atom in an intramolecular cyclisation reaction, forming a
quaternary ammonium salt.
2.2. Spectroscopic properties
The absorption spectrum of probe 1 was recorded in acetoni-
trile (5 ꢀ 10ꢁ6 M, see Fig. 1). The probe showed two weak ab-
sorptions at ca. 415 and 310 nm, together with an intense broad
absorption band (ε ¼ 36,800 cmꢁ1
M
ꢁ1) in the visible region,
centred at 530 nm, corresponding to the S0 / S1 (p / p*) elec-
tronic transition.
This 33 nm red shift compared to that of parent BODIPY 5
labs ¼ 497 nm, lem ¼ 505 nm) follows the trend observed with
(
The sensing unit was decided to be electronically connected to
the BODIPY core in order to have a two optic channel chemo-
dosimeter. We expected that the energy of the extended p-system
could be modulated through an ICT process upon analyte detection,
which would turn into a modulation in the absorption and emis-
sion properties of the dye.
Table 1
Spectroscopic properties of probe 1 and 1 þ Hþ (5 ꢀ10ꢁ6 M in MeCN). lexc ¼ 480 nm.
Ff was measured Rhodamine 6G as standard [23].
Compound
labs (nm)
lem (nm)
Ff
s
1 (ps)
s2 (ps)
a
a
1
530
524
555
555
0.0006
0.639
1 þ Hþb
230
4198
a
Not measured due to low emission.
Proton source is TFA.
b
Scheme 3. Synthesis of the sensing unit 4.

78
R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
Fig. 1. Absorption and normalized emission spectra of probe
1 in acetonitrile
(5 ꢀ 10ꢁ6 M) in the absence (dotted line) and in the presence of 5 mM of TFA
Fig. 3. UVevis spectra of compound 1 (5 ꢀ 10ꢁ6 M in H2O/MeCN 3/1 v/v and 100 mM
MES pH 5.5) in the absence (dashed line) and presence (continuous line) of 10 mM
DCNP.
(
lexc ¼ 470 nm).
other 2,6-arylethynyl substituted BODIPYs [21]. The shift can be
ascribed mainly to the extension of the -system, together with the
p
6G (Ff ¼ 0.95 in ethanol) [23] as a standard). After protonation with
TFA, this band increases 1000-fold reaching a fluorescence quan-
tum yield of Ff ¼ 0.639. No shifts were observed for the emission
band upon addition of TFA. The large stokes shift of 1 and 1 þ Hþ if
compared to 5 (25 and 31 versus 8 nm) indicates the charge
redistribution after excitation.
formation of an internal charge transfer (ICT) system between the
dimethylaniline-ethynyl donor moiety and the BODIPY acceptor
core, lowering the HOMO-LUMO energy gap. It is worth to remark
that this class of donor-ethynyl substituted BODIPYs is scarce in the
literature [21,22], and most of the time consists in symmetrically
2,6-disubstituted BODIPYs forming
system.
a
donor-acceptor-donor
The low fluorescence intensity observed for 1 can be attributed
to a quenching process related to the excited ICT state. This state is
essentially non emissive. After protonation with TFA, the disabling
of the ICT process results in an increase of the fluorescence in-
tensity in the locally excited state (LE). The fluorescence decay of
1 þ Hþ is biexponential with short-lived and long-lived compo-
nents of s1 ¼ 0.23 ns and s2 ¼ 4.198 ns.
If the donor group is deactivated (i.e. protonation of the dime-
thylaniline ethynyl moiety with TFA) a 6 nm hypsochromic shift of
the main absorption band is observed together with a narrowing of
the band. Moreover, a slight shoulder can be then observed at ca.
495 nm, presumably corresponding to the vibronic (0 / 1) tran-
sition of the S0 / S1 electronic transition.
In order to take advantage of these optical features, and use
them as a signal transducer, the sensing unit was designed in such a
way that in the presence of diverse organophosphate compounds
its donor character irreversibly vanishes by the formation a qua-
ternary ammonium salt (see Scheme 2) with all the optical changes
that this implies.
The emission spectrum of 1 has a band peaking at 555 nm with
very low fluorescence quantum yield (Ff ¼ 0.006, using Rhodamine
2.3. Detection experiments in solution
Firstly, we explored the behaviour of probe 1 in the presence of
the organophosphorous compounds DCNP and DFP in acetonitrile.
To 5 ꢀ 10ꢁ6 M solutions of 1 in MeCN, different amounts of the
simulants, ranging from 5 mM to 12 mM, were added.
Immediately, the expected changes were observed in the two
optical channels: a gradual hypsochromic shift of the main ab-
sorption band, together with a strong increase of the emission band
intensity occurred until a maximum concentration of 3 mM and
12 mM for DCNP and DFP respectively (see Fig. 2 for DFP). The
colourimetric changes could be easily observed by the naked eye,
together with the increase in the fluorescence, which was easily
followed by exposing the samples under a hand-held UV lamp
(Fig. 3).
In order to achieve from lab to field chemodosimeter, this has to
be capable of detecting the analyte in more realistic samples. For
instance, one of the targets of this chemodosimeter could be the
detection of organophosphonates in contaminated water (e.g. city
water supply in conflict zones), hence the need of the sensor to
work in aqueous environment.
After testing several aqueous mixtures, we found our optimal
conditions by using a 3/1 v/v water/acetonitrile mixture. Addi-
tionally, 100 mM 2-(N-morpholino)ethanesulfonic acid (MES) was
employed to buffer the solution at pH 5.5, which guarantees the
absence of false positives due to proton traces (e.g. mineral acids
such as HF from the decomposition of DFP). Additionally, the buffer
Fig. 2. UVevis spectra (top) and emission spectra (bottom, lexc ¼ 470 nm) of com-
pound 1 (5 ꢀ 10ꢁ6 M in MeCN) titrated with DFP (from 5
mM to 3 mM). Insets:
Observable colour changes upon addition of the simulant.

R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
79
Fig. 5 shows the emission intensity changes of probe 1 at 555 nm
for an initial DCNP concentration of 7.2 mM. The reaction is rela-
tively fast, and maximum emission changes are observed after a
few minutes. The global rate constant for 1 in the presence of DCNP
was k ¼ 242 Mꢁ2
s
ꢁ1. The calculated half-life time of 1 in the
presence of i.e. 7.2 mM of DCNP (kobs ¼ 12.6 ꢀ 10ꢁ3
s
ꢁ1) was ca
1 min.
Additionally, kinetic experiments in pure acetonitrile were
attempted too. Unfortunately, the velocity of the reaction in this
solvent was too quick to be monitored by conventional methods.
2.5. Limits of detection and potential interferents
Fig. 4. Titration profiles of the emission intensity at 555 nm of 1 with DCNP and DFP in
acetonitrile and with DCNP in a water-MeCN mixture.
Detection limits for the reaction of the nerve agent simulants
DCNP and DFP with 1 in acetonitrile or buffered water-acetonitrile
mixture were evaluated. For this determination, increasing
amounts of the simulants were added to 3 mL solutions of 1 until
the emission values fulfilled the condition:
ensures a constant ionic strength aiding to the reproducibility of
the analysis. It is worth to remark the good solubility of 1 in this
medium, without the need of tailoring the BODIPY core.
After addition of 5 mM of DCNP, a 10-fold increase in the
emission intensity was observed, together with a hypsochromic
shift. The pH of the solution remained the same, ensuring that all
the achieved response was due to the selective detection of DCNP.
On the other hand, the addition of up to 10 mM of DFP to probe 1
made no changes in the optical properties of the aqueous mixture,
probably due to the higher lipophilic character of this simulant that
causes miscibility problems. This observation precluded the use of
DFP in aqueous systems.
Is ¼ Iblank þ 3 ꢀ sblank
where Is is the sample emission intensity at lem ¼ 560 nm after
2 min of addition of the simulant, Iblank is the emission intensity
mean of the blank and sblank is the standard deviation of the blank.
The achieved LODs were 2.7 and 1.8 ppm (v/v) for DCNP and DFP
(respectively) in acetonitrile and 14 ppm (v/v) for DCNP in the
buffered aqueous mixture [24].
Once more, we wanted to ensure that chemically related species
and other conventional compounds present in the fields of appli-
cation of this probe did not interfere rendering false positives or
disabling the probe. Thus, we tested the reactivity of 1 towards
other organophosphorous compounds (OP) (DECP, DPEP, DCTP,
DMTP and DOPP, see Scheme 1) and pesticides (1,1-dichloro-2,2-
bis(p-chlorophenyl)ethane (4,40-DDD) and 1,1-dichloro-2,2-bis(p-
chlorophenyl)ethylene (4,40-DDE)) besides gasoline and diesel.
The test was performed contaminating buffered aqueous solu-
tions of 1 (water/acetonitrile 3:1 v/v pH ¼ 5.5 MES 0.1 M) with ca.
300 ppm of the possible interferents. None of them was able to
yield a significant response on the sensor similar to that of the
DCNP. The highest response was observed for DCTP which
increased the fluorescence of 1 up to 7% of the maximum, probably
due to the similar chemical nature to that of DCNP (see Fig. 6).
Additionally, after each test was performed, the analyte DCNP
(300 ppm) was added to the contaminated aqueous mixture of 1,
reproducing an optical response similar to that of pure DCNP. These
results prove not only that 1 is selective to the nerve agent simu-
lants, but also that it is capable to detect them even inside a more
complex matrix (Schemes 3 and 4).
2.4. Kinetic studies
The kinetics of the reaction between the nerve agent simulant
DCNP and probe 1 in buffered water-acetonitrile mixtures was
evaluated. Different excess amounts of DCNP were added to four
different samples containing probe 1 (5 ꢀ 10ꢁ6 M in H2O/MeCN 3/1
v/v and 100 mM MES pH 5.5) in such a way that the final concen-
tration of DCNP was 1.8, 3.6, 5.4 and 7.2 mM. The samples were
placed in a multi-cell holder and the emission intensity at 555 nm
was constantly monitored vs. time. As the simulant is in 300e1500-
fold excess, the experiment fulfilled the conditions of the Ostwald
isolation method, ensuring pseudo first-order kinetic conditions.
Plotting ln[(If ꢁ I)] versus time (in which, If is the final emission
intensity and I is the emission intensity at a given time), allowed us
the determination of the observed rate constants (kobs). From the
kobs values and the corresponding DCNP concentrations we were
able to evaluate the global constant rate (k) (Fig. 4).
2.6. Solid-liquid detection experiments
The sensing properties of 1 immobilized on a solid support were
also evaluated, since immobilization of chemodosimeters is a key
step to jump from the lab to the field in sensing applications. We
decided to immobilize 1 in a hydrophilic polyurethane based ma-
trix (Hydromed D4) due to its similarity to water environments.
This matrix was later applied as a coating over polyethylene strips.
After drying for 12 h, the solid material presented no observable
fluorescence under a hand-held UV365nm lamp. When the strips
were dipped into distilled water contaminated with DCNP (from 10
to 1000 ppmv), a strong enhancement of the fluorescence was
observed by the naked eye (see Fig. 7).
After the test was performed, washing the strips with distilled
water did not quench the fluorescence, indicating that the observed
changes in fluorescence were due to the irreversible cyclization of 1.
Fig. 5. Kinetic profile of the emission intensity at 555 nm of probe 1 after the addition
of 7.2 mM DCNP. Probe 1 is 5 ꢀ 10ꢁ6 M water/acetonitrile 3/1 v/v mixture (pH ¼ 5.5).
The inset figure shows the correlation between kobs and [DCNP]2.

80
R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
Fig. 6. Relative (to DCNP) emission intensity of 1 (measured at 560 nm) with 300 ppm
(v/v) of DCNP, pesticides, OP compounds, gasoline and diesel.
Scheme 4. Synthesis of the bromo-BODIPY precursor
coupling reaction to obtain probe 1.
6 and Sonogashira cross-
As an extra experiment, tap water samples were contaminated
with DCNP, and after dipping the solid material containing 1 in the
sample, the same strong luminescence was observed. In addition,
no luminescence was produced for non-contaminated tap-water,
indicating that alkaline ions and other substances present in this
liquid do not interfere with the measurement.
(2-(dimethylamino)phenyl)ethanol sensing unit using a BODIPY
core as chromophore and fluorophore. We have studied its spec-
troscopic properties and we have found that 1 has very low fluo-
rescence quantum yield, presumably due to a quenching by an ICT
process. The probe has been designed in such a way that in the
presence of the simulants, the ICT becomes cancelled. Thus, after
reaction with DCNP or DFP an increase of the fluorescence intensity,
along with a hypsochromic shift of the main absorption band is
observed. Compound 1 is able to detect DCNP in aqueous mixtures
and has proved to be selective towards nerve agent simulants in the
presence of other organophosphorous compounds, pesticides, and
other interferents. The probe can detect DCNP and DFP in solution
in the 1e20 ppm range.
Furthermore, we have demonstrated the solidegas detection
capabilities of 1 by adsorbing the probe on solid silica matrices, and
exposure to DCNP vapours reaching LODs of 0.2 ppm.
We believe that the good sensitivity and selectivity achieved by
this chemodosimeter together with the photostability ascribed to
the BODIPY dyes can be a great contribution for the production of
trustworthy and robust portable sensing devices.
2.7. Detection in the gas phase
The high volatility of the nerve agent gases makes absolutely
necessary for a chemodosimeter to be able to effectively perform a
fast but still trustworthy detection of these compounds in the gas
phase. Thus, as a simple experiment, we prepared a small assay kit
by adsorbing probe 1 over commercial silica plate strips containing
no fluorophore. The adsorption was performed by dipping the strip
during 5 s in a 1.5 ꢀ 10ꢁ3 M acetonitrile solution of the probe.
The sensing experiment was carried out by suspending the strip
in the middle of a 5 L glass round bottom flask, which was sealed
with a septum. The experimental atmosphere was common air, at a
relative humidity of 60e65%. The inner atmosphere was contami-
nated with DCNP by adding different amounts via syringe to the
bottom of the flask. Within 5 min, a lightening of the sample could
be observed by the naked eye, whereas the fluorescent changes
were evident under UV365nm light irradiation (see inset of Fig. 8).
The solid state fluorescence spectra of these strips were
measured prior and after exposure to the nerve gas simulant using
a 20ꢂ configuration to minimize the amount of scattered light. Prior
to the exposure, the probe showed a weak fluorescence, possibly
due to slight interaction with the acidic silica. After exposure to
2 ppm of DCNP, this band increased 16-fold, yielding a slightly
broadened spectrum which was 10 nm red shifted when compared
with the corresponding spectrum in solution (see Fig. 8).
4. Experimental section
4.1. General methods
Nerve agent simulants DCNP and DFP are very toxic com-
pounds, thus, special precautions need to be taken when
2.7.1. Limits of detection in gas phase and interferents
With this experimental device we were able to visually detect
within 1 min down to 0.2 ppm of DCNP (1 mL of DCNP spread in a 5 L
flask).
The strips were also tested versus different gases that could be
present in civilian or military setting. Thus, sensing strips con-
taining probe 1 were exposed ozone, gasoline and diesel vapours,
as well as ammonia, and exhaust pipe fumes. In all cases, negligible
changes were found for the emission intensity of the strips.
3. Conclusions
Fig. 7. Polyethylene strips coated with hydromed previously doped with compound 1.
The strips were dipped in distilled water contaminated with 0, 10, 100 and 1000 ppmv
of DCNP. Top values indicate the relative (to the rightmost strip) luminance of the
measured squared areas. Measurement was carried out by meaning the CIE luminance
of all the pixels inside the squared area with custom-made image processing software.
We have synthesized and demonstrated the use of compound 1
as fluoro-chromogenic probe for the selective detection of the
nerve agent simulants DCNP and DFP. The probe 1 is based on a 2-

R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
81
4.4. Experimental procedures
4.4.1. Synthesis of 1
BODIPY 6 (80 mg, 0.2 mmol) was dissolved in a toluene/Et3N 2/1
v/v mixture (20 mL), under an argon atmosphere. The mixture was
sparged with argon for 10 min. Then, CuI (1.9 mg, 5 mol%) and
tetrakis(triphenylphosphine)palladium(0) (11.5 mg, 5 mol%) were
added, and the acetylene derivative 4 was added dissolved in the
minimum amount of previously degassed toluene. The reaction was
sparged with argon for 10 more min. Then, the mixture was heated
to 70 ꢂC and allowed to react for 24 h. After this time, solvents were
evaporated and the mixture was then dissolved in DCM. The
organic solvent was washed twice with water, and the aqueous
layer was extracted with DCM. The combined organic layers were
washed with aq NaCl (sat.) and dried with MgSO4. After solvent
evaporation, the remaining mixture was purified by silica column
chromatography using EtOAcehexane 2:8 as eluent to yield a dark
Fig. 8. Solid state fluorescence spectrum (lexc ¼ 480 nm) of compound 1 over silica
plates in the absence (dotted line) and in the presence (continuous line) of 2 ppm
atmosphere of DCNP (2 mL of DCNP spread in a 1 L flask). Additionally, dashed line
shows normalized spectra in of 1 þ DCNP in acetonitrile. The inset picture shows the
prepared silica strips of (from left to right) 1, 1 þ 2 ppm DCNP and both samples under
UV365nm light.
red solid (18 mg, 18%). 1H NMR (500 MHz, CD2Cl2)
d
7.57 (dd, J ¼ 5.1,
2.0 Hz, 3H), 7.38e7.34 (m, 3H), 7.31 (d, J ¼ 2.0 Hz, 1H), 7.16 (d,
J ¼ 8.2 Hz, 1H), 6.12 (s, 1H), 3.83 (t, J ¼ 5.8 Hz, 2H), 2.98 (t, J ¼ 5.9 Hz,
2H), 2.73 (s, 6H), 2.69 (s, 3H), 2.59 (s, 3H), 1.55 (s, 3H), 1.45 (s, 3H).
13C NMR (126 MHz, CD2Cl2)
d 157.58, 156.13, 152.60, 144.94, 142.49,
performing any experiment with them. Manipulations involving
DCNP or DFP were performed in a glove box with constant air
flow. The stream coming from the glove box exhaust was bubbled
in a saturated NaOH water solution placed inside the fume hood.
UVevis and fluorescence measurements were performed with
tight capped cuvettes previously prepared in the glove box. In
the cases were the experimental setup was too large to be fitted
in the glove box, the experiment was performed in the fume-
hood under special precautions.
All the synthetic manipulations were performed in a dry
argon atmosphere using standard techniques. Compound 5 was
prepared according to the procedures described in the literature
[20]. Tetrahydrofuran was distilled over Na prior to use. The other
materials were purchased and used as received. Silica gel 60 F254
(Merck) plates were used for TLC. 1H and 13C NMR spectra were
recorded using a Bruker DRX-500 spectrometer (500 MHz for 1H
and 126 MHz for 13C) and a Bruker Avance 400 MHz (400 MHz
for 1H and 100 MHz for 13C) with the deuterated solvent as the
lock and residual solvent as the internal reference. HRMS were
recorded using a Shimadzu QP5050A. Absorption spectra were
recorded with a Shimadzu UV-2101PC spectrophotometer. Fluo-
rescence spectra were carried out in a Varian Cary Eclipse
fluorimeter.
142.28, 135.78, 134.57, 133.64, 132.44, 130.37, 130.21, 129.26, 129.13,
127.93, 122.00, 119.86, 119.29, 115.14, 95.69, 81.15, 63.81, 44.58,
35.61, 14.48, 14.27, 13.26, 12.90. HR-MS: calcd for C31H33N3OF2B
[M þ H]þ: 512.2679, found: 512.2685.
4.4.2. Synthesis of compound 6
BODIPY 5 (1 g, 3.08 mmol) was dissolved in a DMF/DCM 2/1 v/v
mixture (150 mL). Then, N-bromosuccinimide (660 mg, 3.69 mmol)
dissolved in DCM (50 mL) was added slowly. After one hour, the
reaction mixture was washed three times with water, followed by a
washing step with brine. The organic phase was dried with MgSO4,
and after evaporation of the solvent, the red solid was purified with
silica gel column chromatography using hexaneeDCM (1:9 / 4:6)
as eluent. Yield: 60% (726 mg).
Rf ¼ 0.52 (EtOAc hexane 1:9). 1H NMR (500 MHz, CDCl3)
d 7.55e
7.51 (m, 3H), 7.31e7.27 (m, 2H), 6.06 (s, 1H), 2.62 (s, 3H), 2.60 (s,
3H), 1.41 (s, 3H), 1.39 (s, 3H). 13C NMR (126 MHz, CDCl3)
d 157.93,
151.46, 145.10, 141.84, 138.77, 134.67, 132.04, 129.82, 129.28, 129.25,
127.87, 122.16, 110.62, 14.77, 14.55, 13.47, 13.40. HR-MS: calcd for
C
19H17BBrF2N2 [M þ H]þ: 403.0787, found: 403.0791.
4.4.3. Synthesis of compound 2
2-(2-Nitrophenyl)ethanol (10 g, 61 mmol), formaldehyde (37%,
11.24 mL, 75.2 mmol), ethanol (200 mL), and Pd/C (500 mg, 10%)
were placed under an H2 atmosphere at 60 PSI until the uptake of
hydrogen ceased. After filtration through celite, the solvent was
evaporated, and the residue was dissolved in EtOAc, and washed
twice with water, and aq NaCl (sat.). The organic phase was dried
using MgSO4, and the solvent was evaporated to give 2-(2-(N,N-
dimethylamino)phenyl)ethanol as crude oil. The oil (10.12 g,
61 mmol) and ammonium acetate (470 mg, 6.1 mmol) were dis-
solved in acetonitrile (300 mL) in a round bottom flask. Then, the
mixture was cooled down to 0 ꢂC, and a solution of N-bromo-
succinimide (10.85 g, 61 mmol) in acetonitrile (10 mL) was added
dropwise. After 1 h, the solvent was evaporated and the mixture
dissolved in EtOAc. The organic solvent was washed twice with 10%
Na2CO3 and sat. aq NaCl. The organic phase was dried with MgSO4
and evaporated. The remaining oil was purified by silica column
chromatography using EtOAcehexane 4:6 as eluent to yield
brownish oil (10.61 g, 71% overall yield). Rf ¼ 0.77 (EtOAc: hexane
4.2. Preparation of the hydrogel coated polyethylene films
Polyurethane based hydrogel (Hydromed D4Ò) (4.0 g) was
mixed by 1 h rotation with EtOH (55 mL) and water (4.8 mL). To
2.5 mL of the viscous solution, an acetonitrile solution of 1 (1.5 mM,
500 mL) was added, and the resulting mixture was again rotated for
an additional hour.
Transparent polyethylene strips (45 ꢀ 6 ꢀ 0.5 mm) were then
dipped once in the solutions. The excess of hydrogel solution was
removed and the strips were allowed to dry at room temperature
for 2 h.
4.3. Preparation of the silica doped strips
Commercial silica gel over aluminium foil without fluorescent
indicator (Sigma Aldrich) were cutted in 45 ꢀ 6 mm strips and then
dipped for 10 s in a 1.5 mM solution of 1. The strip was allowed to
dry at room temperature for 30 min. Additionally, part of the silica
was removed from the aluminium plate in such a way that the
active area shaped a 20 ꢀ 6 mm rectangular surface.
6:4). 1H NMR (500 MHz, CDCl3)
d
7.24 (dd, J ¼ 8.5, 2.4 Hz, 1H), 7.21
(d, J ¼ 2.3 Hz, 1H), 6.97 (d, J ¼ 8.5 Hz, 1H), 3.75 (t, J ¼ 5.7 Hz, 2H),
2.87 (t, J ¼ 5.7 Hz, 2H), 2.59 (s, 6H). 13C NMR (125 MHz, CDCl3)

82
R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
Structure-activity relationship and efficacy in the treatment of poisoning with
d
151.36, 138.22, 133.64, 130.35, 121.77, 117.61, 77.00, 63.86, 44.81,
organophosphorus compounds. Curr Med Chem 2009;16:2177e88.
[4] Roscioli KM, Lamabadusuriya MR, Harden CS, Midey AJ, Wu C, Siems WF, et al.
Structure selective ion molecule interactions (SSIMI) in ion mobility spec-
trometry. Int J Ion Mobil Spectrom; 2013:1e11;
35.61. HR-MS: calcd for C10H14BrNO [M þ H]þ: 244.0337, found:
244.0341.
ꢀ
Maziejuk M, Ceremuga M, Szyposzynska M, Sikora T. Effect of temperature on
4.4.4. Synthesis of compound 3
separation of Sarin (GB) ions in differential mobility spectrometry. Open J Phys
Chem 2013;3:170e6.
Compound 2 (4.15 g, 17.0 mmol) was dissolved in a toluene/Et3N
2:1 v/v mixture (75 mL) in a three-neck round bottom flask. The
inner atmosphere was evacuated and replaced with argon. This
operation was repeated two more times. Then, the mixture was
sparged with argon for 10 min. After this time, tetrakis(-
triphenylphosphine)palladium(0) (982 mg, 5 mol%) and copper(I)
iodide (162 mg, 5 mol%) were added to the reaction mixture with
the help of toluene (10 mL). After 10 more min of sparging with
argon, trimethylsilaneacetylene (2.0 g, 20.4 mmol) dissolved in
degassed toluene (20 mL) was added and the temperature was
raised to 50 ꢂC. The mixture was allowed to react for 72 h. After the
reaction cooled down, DCM was added to the mixture and then it
was washed three times with 10% aq NH4Cl. The aqueous layers
were reunited, and extracted with DCM. The combined organic
layers were washed with brine, dried with MgSO4 and evaporated.
The residue was purified with silica gel column chromatography
using EtOAcehexane 2:8 as eluent, which yielded 3 as yellowish oil
(3.07 g, 69%). Rf ¼ 0.41 (EtOAcehexane 4:6). 1H NMR (500 MHz,
[5] Seto Y, Kanamori-Kataoka M, Tsuge K, Ohsawa I, Iura K, Itoi T. Sensitive
monitoring of volatile chemical warfare agents in air by atmospheric pressure
chemical ionization mass spectrometry with counter-flow introduction. Anal
Chem 2013;85:2659e66.
[6] Ringer JM. Detection of nerve agents using proton transfer reaction mass
spectrometry with ammonia as reagent gas. Eur J Mass Spectrom 2013;19:
175e85.
[7] Pohanka M. Cholinesterases in biorecognition and biosensors construction: a
review. Anal Lett 2013;46:1849e68;
Pohanka M, Adam V, Kizek R. An acetylcholinesterase-based chronoampero-
metric biosensor for fast and reliable assay of nerve agents. Sensors 2013;13:
11498e506;
Dhull V, Gahlaut A, Dilbaghi N, Hooda V. Acetylcholinesterase biosensors for
electrochemical detection of organophosphorus compounds: a review. Bio-
chem Res Int 2013;2013:731501.
[8] Ge X, Tao Y, Zhang A, Lin Y, Du D. Electrochemical detection of dual exposure
biomarkers of organophosphorus agents based on reactivation of inhibited
cholinesterase. Anal Chem 2013;85:9686e91;
Tan HY, Loke WK, Nguyen NT, Tan SN, Tay NB, Wang W, et al. Lab-on-a-chip
for rapid electrochemical detection of nerve agent Sarin. Biomed Micro-
devices; 2013:1e7;
Sahin A, Dooley K, Cropek DM, West AC, Banta S. A dual enzyme electro-
chemical assay for the detection of organophosphorus compounds using
organophosphorus hydrolase and horseradish peroxidase. Sens Actuators B
Chem 2011;158:353e60;
Delalande M, Clavaguera S, Toure M, Carella A, Lenfant S, Deresmes D, et al.
Chemical functionalization of electrodes for detection of gaseous nerve agents
with carbon nanotube field-effect transistors. Chem Commun 2011;47:6048e
50.
CDCl3)
J ¼ 8.2 Hz,1H), 3.83 (t, J ¼ 5.6 Hz, 2H), 2.97 (t, J ¼ 5.7 Hz, 2H), 2.72 (s,
6H), 0.24 (s, 9H). 13C NMR (126 MHz, CDCl3)
151.92, 135.70,134.76,
d
7.33 (dd, J ¼ 8.3, 2.0 Hz, 1H), 7.28 (d, J ¼ 2.0 Hz, 1H), 7.09 (d,
d
131.37, 119.78, 117.08, 104.70, 93.85, 64.18, 44.96, 35.87, 0.00. HR-
MS: calcd for [M þ H]þ C15H24NOSi: 262.1621, found: 262.1617.
[9] Yang Y, Ji H, Thundat T. Nerve agents detection using a Cu2þ/L-cysteine
bilayer-coated microcantilever. J Am Chem Soc 2003;125:1124e5;
Zhao Q, Zhu Q, Shih WY, Shih WH. Array adsorbent-coated lead zirconate
titanate (PZT)/stainless steel cantilevers for dimethyl methylphosphonate
(DMMP) detection. Sens Actuators B Chem 2006;117:74e9.
[10] Liu F, Huang S, Xue F, Wang Y, Meng Z, Xue M. Detection of organophosphorus
4.4.5. Synthesis of compound 4
Compound 3 (254 mg, 0.97 mmol) was dissolved in MeOH
(10 mL). Then, K2CO3 (1.34 g, 9.7 mmol) was added to the mixture.
After 30 min, thin layer chromatography revealed complete
deprotection. Then, water was added, and a solution of 0.1 M HCl
was slowly added until neutralization. Then, the mixture was
extracted three times with DCM, the organic layers were combined
and washed with brine, dried with MgSO4 and evaporated. The
product (169 mg, 92%) was used without further purification.
compounds using
a molecularly imprinted photonic cristal. Biosens Bio-
electron 2012;32:273e7;
He W, Liu Z, Du X, Jiang Y, Xiao D. Analytical application of poly{methyl[3-(2-
hydroxy-3,4-difluoro)phenyl]propyl siloxane} as a QCM coating for DMMP
detection. Talanta 2008;76:698e702.
[11] Aernecke MJ, Walt DR. Optical-fiber arrays for vapor sensing. Sens Actuators B
Chem 2009;142:464e9.
[12] Wang F, Gu H, Swager TM. Carbon nanotube/polytiophene chemiresistive
sensors for chemical warfare agents. J Am Chem Soc 2008;130:5392e3;
Claveguera S, Raoul N, Carella A, Delalande M, Celle C, Simonato J-P. Devel-
opment of an autonomous detector for sensing of nerve agents based on
functionalized silicon nanowire field-effect transistors. Talanta 2011;85:
2542e5;
Rf ¼ 0.34 (EtOAcehexane 4:6). 1H NMR (500 MHz, CDCl3)
d 7.38
(dd, J ¼ 8.2, 2.1 Hz, 1H), 7.32 (d, J ¼ 2.0 Hz, 1H), 7.13 (d, J ¼ 8.2 Hz,
1H), 3.86 (t, J ¼ 5.7 Hz, 2H), 3.05 (s, 1H), 2.99 (t, J ¼ 5.7 Hz, 2H), 2.73
(s, 6H). 13C NMR (126 MHz, CDCl3)
d 152.76, 135.89, 134.81, 131.55,
Claveguera S, Carella A, Caillier L, Celle C, Pécaut J, Lenfant S, et al. Sub-ppm
detection of nerve agents using chemically functionalized silicon nanoribbond
field-effect transistors. Angew Chem Int Ed 2010;49:4063e6;
Kwon OS, Park SJ, Lee JS, Park E, Kim T, Park H-W, et al. Multidimensional
conducting polymer nanotubes for ultrasensitive chemical nerve agent
sensing. Nano Lett 2012;12:2797e802;
119.93, 118.38, 83.37, 76.74, 64.17, 44.83, 35.83. HR-MS: calcd for
C
12H16NO [M þ H]þ: 190.1226, found: 190.1227.
Acknowledgements
Wei L, Shi D, Ye P, Dai Z, Chen H, Chen C, et al. Hole doping and surface
functionalization of single-walled carbon nanotube chemiresistive sensors for
ultrasensitive and highly selective organophosphor vapor detection. Nano-
technology 2011;22:425501e7.
We thank the Spanish Government and the European FEDER
funds (project MAT2012-38429-C04-02 and 01). SCSIE and ICMOL
(Universidad de Valencia) are gratefully acknowledged for all the
equipment employed. R.G. is grateful to the Spanish Ministry of
Education for its FPU grant. Knut Rurack is gratefully acknowledged
for the life-time measurements performed at his lab.
[13] Costero AM, Parra M, Gil S, Gotor R, Mancini PM, Martínez-Máñez R, et al.
Chromo-fluorogenic detection of nerve-agent mimics using triggered cycli-
zation reactions in push-pull dyes. Chem Asian J 2010;5:1573e85;
Costero AM, Parra M, Gil S, Gotor R, Mancini PM, Martínez-Máñez R, et al.
Selective detection of nerve agent simulants by using triarylmethanol-based
chromogenic chemodosimeters. Eur J Org Chem 2012;26:4937e46;
Ajami D, Rebek Jr J. Chemical approaches for detection and destruction of
nerve agents. Org Biomol Chem 2013;11:3936e42.
References
[14] Dale TJ, Rebek J. Fluorescent sensors for organophosphorus nerve agent
mimics. J Am Chem Soc 2006;128:4500e1;
[1] Okumura T, Takasu N, Ishimatsu S, Miyanokim S, Mitsuhashi A, Kumada K,
et al. Report on 640 victims of the Tokyo subway Sarin attack. Ann Emerg Med
1996;28:129e35.
[2] Abu-Qare AW, Abou-Donia MB. Sarin: health effects, metabolism, and
methods of analysis. Food Chem Toxicol 2002;40:1327;
Zhang S, Timothy MS. Fluorescent detection of chemical warfare agents:
functional group specific ratiometric chemosensors. J Am Chem Soc 2003;125:
3420e1.
[15] Xuan W, Cao Y, Zhou J, Wang W. A FRET-based ratiometric fluorescent and
colorimetric probe for the facile detection of organophosphonate nerve agent
mimic DCP. Chem Commun 2013;49:10474e6.
[16] Martí A, Costero AM, Gaviña P, Gil S, Parra M, Brotons-Gisbert M, et al.
Functionalized gold nanoparticles as an approach to the direct colorimetric
detection of DCNP nerve agent simulant. Eur J Org Chem 2013;22:4770e9;
Wang J, Gu J, Leszczynski J. Phosphonylation mechanisms of Sarin and
acetylcholinesterase: a model DFT study 2006;110:7567e73.
[3] Sidell FR, Groff WA. The reactivatibility of cholinesterase inhibited by VX and
Sarin in man. Toxicol Appl Pharmcol 1974;27:241e52;
ꢀ
Jokanovic M, Prostran M. Pyridinium oximes as cholinesterase reactivators.

R. Gotor et al. / Dyes and Pigments 108 (2014) 76e83
83
Liu D, Chen W, Wei J, Li X, Wang Z, Jiang X. A highly sensitive, dual-readout
assay based on gold nanoparticles for organophosphorus and carbamate
pesticides. Anal Chem 2012;84:4185e91;
Lee JS, Shin DH, Jun J, Jang J. Multidimensional polypyrrole/iron oxyhydroxide
hybrid nanoparticles for chemical nerve gas agent sensing. ACS Nano 2013;7:
10139e47;
Sun J, Guo L, Bao Y, Xie J. A simple, label-free AuNPs-based colorimetric ul-
trasensitive detection of nerve agents and highly toxic organophosphate
pesticide. Biosens Bioelectron 2011;28:152e7.
[21] Wagner S, Brödner K, Coombs BA, Bunz UHF. Pyridine-substituted BODIPY as
fluorescent probe for Hg2þ. Eur J Org Chem 2012;11:2237e42;
Wang Y, Zhang D, Zhou H, Ding J, Chen Q, Xiao Y, et al. Nonlinear optical
properties and ultrafast dynamics of three novel boradiazaindacene de-
rivatives. J Appl Phys 2010;108:033520;
He H, Ng DKP. A ratiometric near-infrared pH-responsive fluorescent dye
based on distyryl BODIPY. Org Biomol Chem 2011;9:2610e3;
Bonardi L, Ulrich G, Ziessel R. Tailoring the properties of bor-
onꢁdipyrromethene dyes with acetylenic functions at the 2,6,8 and 4-B
substitution positions. Org Lett 2008;10:2183e6.
[17] Ordronneau L, Carella A, Pohanka M, Simonato J-P. Chromogenic detection of
Sarin by discolouring decomplexation of a metal coordination complex. Chem
Commun 2013;49:8946e8.
[18] Chulvi K, Gaviña P, Costero AM, Parra M, Gil S, Gotor R, et al. Discrimination of
nerve gases mimics and other organophosphorous derivatives in gas phase
using a colorimetric probe array. Chem Commun 2012;48:10105e7.
[19] Loudet A, Burgess K. BODIPY dyes and their derivatives: syntheses and
spectroscopic properties. Chem Rev 2007;107:4891e932.
[20] Kollmannsberger M, Rurack K, Resch-Genger U, Daub J. Ultrafast charge
Transfer in amino-substituted boron dipyrromethene dyes and its inhibition
by cation complexation: a new design concept for highly sensitive fluorescent
probes. J Phys Chem A 1998;102:10211e20.
[22] Niua S, Ulrich G, Retailleau P, Ziessel R. BODIPY-bridged pushepull chromo-
phores: optical and electrochemical properties. Tetrahedron Lett 2011;52:
4848e53;
Bonnier C, Machin DD, Abdi O, Koivisto BD. Manipulating non-innocent
p-
spacers: the challenges of using 2,6-disubstituted BODIPY cores within
donoreacceptor light-harvesting motifs. Org Biomol Chem 2013;11:3756e60.
[23] Kubin RF, Fletcher AN. Fluorescence quantum yields of some rhodamine dyes.
J Lumin 1982;27:455e62.
[24] The US Enviromental Protection Agency (EPA) states that the Sarin’s LD50 for
humans averaging 70 kg is 12 mg.