Pyrazoline-based colorimetric and fluorescent probe for detection of sulphite

Article abstract of DOI:10.1039/C8NJ05017A

Two pyrazoline-based fluorescent and colorimetric probes have been synthesized and their photophysical properties have been investigated by means of electronic absorption and emission spectroscopy. The compounds differ from each other by the presence of a

Full text of DOI:10.1039/C8NJ05017A

NJC
View Article Online
View Journal | View Issue
PAPER
Pyrazoline-based colorimetric and fluorescent
probe for detection of sulphite†
Cite this: New J. Chem., 2019,
43, 874
a
a
b
b
Tomasz Uchacz, * Gabriela Jajko, Andrzej Danel,* Paweł Szlachcic and
a
Szczepan Zapotoczny
Two pyrazoline-based fluorescent and colorimetric probes have been synthesized and their photo-
physical properties have been investigated by means of electronic absorption and emission spectro-
scopy. The compounds differ from each other by the presence of a phenyl or thiophene end group
attached to the a,b-unsaturated ketone. The probes detect sulphite anions in the Michael addition
reaction to the a,b-unsaturated ketone with a high selectivity and good sensitivity (7.56 mM for phenyl
and 4.87 mM for the thiophene counterpart). Here, the optical response is based on the recovery of
triphenylpyrazoline fluorescence, which is largely blue-shifted as compared to weak charge transfer
Received 3rd October 2018,
Accepted 30th November 2018
emission of the sensors. This large hypsochromic shift of the emission maximum along with a strong
sulphite
DOI: 10.1039/c8nj05017a
fluorescence enhancement (up to I
0
/I = 43) can be an advantage in terms of the accurate
evaluation of the fluorescence intensity ratio. The pyrazoline-based sensor decorated with thiophene is
able to detect sulphite species in river water solutions with a good selectivity and sensitivity.
rsc.li/njc
foods, beverages and drugs because sulphating agents (sulphur
1
. Introduction
dioxide, metabisulphite, bisulphite and sulphite) are widely used
5
Sulphur dioxide is a vital component of acid rain and smog, as preservatives. In some cases it could be very dangerous for
which is one of the main problems faced by citizens in the big humans. Sulphite oxidase deficiency is a fatal disease that causes
metropolises. Acid rain contributes to destruction of monuments, neurological disorders, mental retardation, physical deformities,
6,7
while smog can affect the human respiratory system. In high doses degradation of the brain, and early death. Thus, the sulphite
gaseous sulphur dioxide is poisonous and there are records that it level in the body must normally be tightly maintained.
1
was used for deliberate poisoning of people. Prolonged exposure
In recent years, growing interest in the application of fluores-
to gaseous pollutants like SO , NO and CO can be a factor for cent sensors towards various analytes like cations, anions and
2
2
2,3
acute stroke death. One of the main sources of SO is combus- neutral molecules has been observed. Quite a lot of work was
2
tion of fossil fuels like coal and petroleum. On the other hand, it devoted to developing sensors for sulphite and bisulphite anions.
should be remembered that volcanoes can release a huge amount Zhou et al. synthesized a fluorescent turn-on probe based on
8
of that gas into the atmosphere as well. By way of illustration, in 4-hydrazinyl-1,8-napthalimide with a detection limit of 0.56 mM.
2
ꢀ
ꢀ
2014 the Holuhraun volcano (Iceland) emitted 1030 kilograms A large quantity of fluorescent probes for SO
3
and HSO
3
are
per second of sulphur dioxide during the first 3 months of the based on a coumarine system. For example, Li et al. applied
eruption, which obviously contributed to the deterioration of the 3-formyl-7-N,N-diethylcoumarine for bioimaging of hydrogen
4
health status of the Iceland population. SO is a gas which exists sulphite. The non-emissive aldehyde (F = 0.02) emitted green
2
in aqueous solution at neutral pH as an equilibrium between light (F = 0.43) upon nucleophilic addition of hydrogen
9
bisulphite and sulphite ions (NaHSO
Once SO
3
and Na
2
SO
3
, 1 : 1 M/M). sulphite. Due to our interest in strongly emissive 1H-pyrazolo-
1
0–13
2
is inhaled into the human body and dissolved in body [3,4-b]quinolines,
we decided to exploit another pyrazole
fluid, it will immediately induce its biological effects in the form of derivative, namely 1,3,5-triarylpyrazoline, as a potential fluo-
its derivatives. In addition, bisulphite/sulphite enters the body via rescent sensor for various analytes. This system was applied
mainly for the detection of various cations. For example, Wang
et al. synthesized 1-phenyl-3-(2-pyridyl)-5-phenylpyrazoline. The
spectroscopic studies showed that this compound is a typical
a
Faculty of Chemistry, Jagiellonian University, ul. Gronostajowa 2, 30-387 Krak o´ w,
Poland. E-mail: tomaszuchacz55@gmail.com
OFF–ON fluorescence sensor which exhibits high selectivity and
b
Institute of Chemistry, Department of Food Technology, Agricultural University,
14
affinity toward Zn(II) cations. A similar affinity was observed in
ul. Balicka 122, 31-149 Krak o´ w, Poland. E-mail: rrdanela@cyf-kr.edu.pl
15
the case of 1-(2-benzothizaole)-3-(2-thiophene)-2-pyrazoline. How-
ever, there are only a few examples of pyrazoline based sensors for
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8nj05017a
8
74 | New J. Chem., 2019, 43, 874--883
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
were carried out by means of a Bruker Avance III 600 spectro-
meter operating at 600 and 150 MHz, respectively, in CDCl
3
with TMS as an internal standard. Melting points were
determined on a Mel-Temp Apparatus II (capillary), and they
are uncorrected. FTIR spectra of the molecules were recorded
ꢀ
1
ꢀ1
in the range of 4000–500 cm at resolution of 4 cm using a
MATTSON 3000 FT-IR (Madison, Wisconsin, USA) spectro-
photometer. MS analysis was carried out by the ESI-MS-TOF
method.
2.2. Photophysical and electrochemical measurements
Fluorescence quantum yield and fluorescence lifetime measure-
ments were carried on the deaerated solutions (by saturation with
argon) and the sample concentration was ca. 2.8 mM. UV-vis
electronic absorption spectra were recorded using a Varian Cary
Eclipse spectrometer and the emission spectra (with the correction
for spectral sensitivity) were measured by Fluorolog-tau3 equipped
with time-correlated single photon counting apparatus (diode laser
N-455, IBH-UK). Deconvolution analysis of the decay curves yielded
the corresponding fluorescence lifetimes. For this purpose, the
instrument-response function was recorded by using a light-
scattering Ludox solution. The fluorescence quantum yield
Fig. 1 Molecular structures of the investigated compounds.
thiol derivatives. For instance, Miao et al. synthesized a fluorescent measurements were estimated with reference to fluorescein
2
1
probe based on coumarine and pyrazoline to detect cysteine in in 0.1 M NaOH aqueous solution (F = 0.92).
living cells with an appreciable limit of detection (5.08 mM).
fl
1
6
Differential pulse voltammograms of the studied compounds
Another group investigated a pyrazoline-coumarine hybrid were measured with a PalmSens3 potentiostat. One platinum wire
1
7
combined with a 2,4-dinitrobenzenosulfonyl (DNBS) moiety.
(+ = 0.5 mm), platinum coil (+ = 3.0 mm) and gold wire (+ =
In this case, the emission is recovered in the presence of thiols 0.5 mm) were used as a counter, working and quasi-reference
due to the release of the fluorescence quencher – DNBS. To the electrode, respectively. The potential of the quasi-reference electrode
best of our knowledge pyrazoline based probes for the detec- was calibrated using the ferrocene/ferrocenium couple as an
tion of sulphite ions have not been investigated yet.
internal standard. The solution of the sensor (B0.5 mM) was
+
ꢀ
In this paper, two 1,3,5-triarylpyrazoline based sensors were prepared in a suitable electrolyte (B50 mM solution of Bu N ClO
4
4
presented (4a and 4b) as doubly activated Michael addition in DMF). Prior to the measurements the solutions were purged
type indicators for the colorimetric and ratiometric fluorescent with argon to remove residual oxygen.
detection of sulphite (see Fig. 1). The sensors exhibit high
2
ꢀ
2.3. Procedure of anion sensing
selectivity and sensitivity to SO
3
. Two different end groups
were introduced to control the kinetics of analyte addition and The absorption and fluorescence titration experiments were
the sensitivity to the sulphite anion. The advantage of the carried out on non-degassed samples by adding a suitable
reported probes is their high optical response and relatively amount of titrant to a quartz cuvette containing 2 ml of the
fast time response in the presence of small amounts of analyte dye at a concentration of B2.8 mM (1 : 1 deionised water or
(9–11 min in the presence of 45 equivalents) which is compar- Hepes buffer/dimethylformamide solution). Measurements
able with the majority of currently available Michael type were carried out in 1 cm cuvettes after 15 min stabiliza-
indicators (5–60 min in the presence of 10–500 equivalents of tion at room temperature. The excitation wavelengths for
1
8–20
sulphite).
In addition, the compound decorated with thio- the fluorescence measurements on the addition of sodium
phene was able to detect sulphite anions in river water samples. salt were chosen to agree with the isosbestic points of the
UV/vis absorption titration spectra: lex = 403 nm for 4a,
l
ex = 407 nm for 4b.
The Michael addition reaction kinetics was investigated in
2
. Experimental
2.1. Materials and apparatus
the presence of 45 equivalents of sulphite. The analysis was
All reagents were purchased from Sigma Aldrich or Alfa Aesar carried out by monitoring the fluorescence changes at 587 nm
and used without further purification. The applied dimethyl- (4a) and 605 nm (4b), which correspond to the maximum
formamide (DMF) was of spectroscopic grade. The salts of the emission band of the substrate. As the excitation
employed in stock solutions of sodium anions were: nitrate, wavelength, 479 nm for 4a and 496 nm (4b) was chosen. In
nitrite, bromide, chloride, fluoride, iodide, acetate, thiocyanate, order to avoid the oxygen-sulphite reaction, the measurements
sulphate, sulphite, glutathione, cysteine, homocysteine, hydro- were conducted on deaerated solutions (1 : 1 Hepes buffer
1
13
sulphide, bisulphite, and cyanide. H NMR and C NMR spectra (pH = 8)/DMF).
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019
New J. Chem., 2019, 43, 874--883 | 875

View Article Online
NJC
Paper
The determination of sulphite in river water samples was 1Hpyrazoline ring H4a), 3.93 (dd, J = 12.1, 17.3 Hz, 1Hpyrazoline ring H3),
1
3
carried out by the addition of an appropriate amount of 3.25 (dd, J = 17.3, 5.4 Hz, 1Hpyrazoline ring H4b). C NMR (150 Hz,
sulphite into a 1 : 1 Hepes buffer (pH = 8)/DMF mixture. The CDCl , d/ppm, Fig. S1, bottom, ESI†): 189.9, 155.5, 151.1, 147.9,
3
absorption and emission spectra were measured after 30 minutes 140.9, 137.1, 134.1, 132.5, 131.6, 129.9, 129.5, 129.0, 128.8,
of equilibration.
128.4, 128.2, 126.3, 125.5, 122.3, 118.8, 113.2, 103.0, 63.3,
ꢀ
1
The limit of detection (LOD) was determined by the analysis 43.6. IR (KBr): nmax 2206 (CRN), 1652 (CQO) cm . MS (ESI,
of the fluorescence intensity changes at 460 nm (for both m/z (%), Fig. S2, ESI†): 476(100) [M + Na ], 360(13), 297(63),
+
studied compounds), induced by the presence of analyte. The 178(9), 116(9), 105(33). HRMS (ESI): m/z calc. for C31
H
H
23
N
N
3
3
O:
O:
+
detection limit was calculated using the following equation:
453.1841; found 476.1732 [M + Na] . Anal. calcd for C31
23
C, 82.10; H, 5.11; N, 9.26. Found: C, 81.60; H, 5.05; N, 9.09.
(E)-3-[4-(3,5-Diphenyl-3,4-dihydropyrazol-2-yl)phenyl]-2-
thiophene-2-carbonyl)prop-2-enenitrile (4b). This compound was
3
:3 ꢁ s
LOD ¼
(1)
b
(
where: s – standard deviation of low concentration, b – slope prepared in the same way as 4a i.e. from aldehyde 4 (160 mg,
of calibration line.
0.5 mmol) and 3-oxo-3-(thiophen-2-yl)propanenitrile (115 mg,
.75 mmol). Red crystals, 200 mg, 90%, mp. 246.8 1C (toluene).
0
1
2.4. Synthesis of compounds 4b and 4a
3
H NMR (600 Hz, CDCl , d/ppm, Fig. S3, top, ESI†): 8.25 (dd, J =
The presented pyrazoline derivatives (4) were prepared accord- 3.9, 1.0 Hz, 1Hthiophene ring H3), 8.18 (s, 1H, Hvinyl), 7.97 (d, J = 9.2 Hz,
ing to the synthetic procedure illustrated in Scheme 1. The 2H), 7.78–7.76 (m, 2H), 7.69 (dd, J = 5.0, 1.0 Hz, 1Hthiophene ring H5),
7
.46–7.40 (m, 3H), 7.38–7.35 (m, 2H), 7.32–7.30 (m, 1H), 7.27–7.25
starting 1,3,5-triphenylpyrazoline 2 was synthesized from chalcone
reacted with phenylhydrazine hydrochloride in sodium acetate/
(
8
(
m, 2H), 7.16 (dd, J = 5.0, 3.9 Hz, 1Hthiophene ring H4), 7.13 (d, J =
.9 Hz, 2H), 5.45 (dd, J = 12.0, 5.3 Hz, 1Hpyrazoline ring H4a), 3.94
dd, J = 17.4, 12.1 Hz, 1Hpyrazoline ring H3), 3.26 (dd, J = 17.4,
1
22
acetic acid solution under ultrasound irradiation. Pyrazoline 2
23
was formylated with POCl
3
/DMF yielding aldehyde 3. Arylaceto-
1
3
nitriles (3-oxo-3-phenylpropanenitrile and 3-oxo-3-(thiophen-2-yl)- 5.4 Hz, 1Hpyrazoline ring H4b). C NMR (150 Hz, CDCl
3
, d/ppm,
propanenitrile) were prepared according to the literature protocol Fig. S3, bottom, ESI†): 178.9, 155.8, 151.3 148.0, 143.1, 141.0,
from acetonitrile and corresponding ester: methyl benzoate or 134.6, 134.4, 133.8, 131.8, 130.0, 129.7, 128.9, 128.5, 128.4,
24
thiophene-2-carboxylic acid methyl ester, respectively. Finally, 126.5, 125.7, 122.7, 119.8, 113.4, 101.0, 63.5, 43.7. IR (KBr):
ꢀ
1
nmax 2203 (CRN), 1605 (CQO) cm . MS (ESI, m/z (%), Fig. S4,
aldehyde 3 and the corresponding nitrile were reacted together
in the presence of piperidine under ultrasound irradiation.
The reaction was finished within two hours.
+
ESI†): 482(100) [M + Na ], 297(21). HRMS (ESI): m/z calc. for
C H N OS: 459.1405; found 482.1298 [M + Na] . Anal. calcd
+
29 21
3
(E)-2-Benzoyl-3-[4-(3,5-diphenyl-3,4-dihydropyrazol-2-yl)phenyl]-
21 3
for C29H N OS: C, 75.79; H, 4.61; N, 9.14; S 6.98. Found: C,
prop-2-enenitrile (4a). Aldehyde 3 (163 mg, 0.5 mmol) and 3-oxo-3- 75.86; H, 4.62; N, 8.92; S. 6.85.
phenylpropanenitrile (108 mg, 0.75 mmol) were dissolved in 96%
2
.5. DFT calculations
ethanol (10 mL) in a 25 mL round-bottomed flask. To the mixture
drops of piperidine were added and the reaction mixture was Gaussian09 was used for DFT (density functional theory)
2
5,26
3
2
7
irradiated in the water bath of a laboratory ultrasonic cleaner calculations. The ground-state structures of molecules were
for two hours. After finishing, the precipitate was filtered off optimized using the B3LYP method
2
8,29
with the cc-pVDZ
and crystallized from toluene. standard basis set. The optimized structures of the sensors
Yellow crystals, 200 mg, 81%, mp. 178–180 1C (toluene). (and adducts with sulphite) can be found in Fig. S5–S8, ESI.†
1
H NMR (600 Hz, CDCl , d/ppm, Fig. S1, top, ESI†): 7.96 The real energy minima of the optimized structures have been
3
(s, 1Hvinyl); 7.93 (d, J = 9.2 Hz, 2H), 7.82–7.81 (m, 2H), 7.77– checked by calculation of the Hessian matrix and verification
7.75 (m, 2H), 7.58–7.55 (m, 1H), 7.48–7.46 (m, 2H), 7.42–7.40 of the absence of imaginary frequencies. With the aid of the
(
(
m, 3H), 7.37–7.34 (m, 2H), 7.31–7.28 (m, 1H), 7.26–7.24 TD-B3LYP/cc-pVDZ method the vertical excited-state energies
m, 2H), 7.12 (d, J = 8.2 Hz, 2H), 5.44 (dd, J = 12.1, 5.4 Hz, of the dyes and analyte-dye products were estimated as well.
Scheme 1 Synthesis route leading to (E)-2-benzoyl-3-[4-(3,5-diphenyl-3,4-dihydropyrazol-2-yl)phenyl]prop-2-enenitrile (4a) and (E)-3-[4-(3,5-
diphenyl-3,4-dihydropyrazol-2-yl)phenyl]-2-(thiophene-2-carbonyl)prop-2-enenitrile (4b).
8
76 | New J. Chem., 2019, 43, 874--883
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
3
0,31
The Polarisable Continuum Model (PCM)
simulate the influence of DMF polarity.
was employed to that all of the above-mentioned processes may contribute to the
deactivation pathway of the excited state of the molecules
studied. However, a definite answer to the question requires
a deep analysis of photophysical processes (e.g. solvatochromism)
which will be carried out in our laboratory with a subsequent
report in due course.
3. Results and discussion
3
.1. Photophysical and electrochemical properties of 4a and 4b
Room temperature absorption and emission spectra of 4a and (Table 1), one can notice that the reduction potential of 4b is
b in DMF are presented in Fig. 2 and some basic photo- slightly decreased as compared to 4a. This is somewhat
Analyzing the results from electrochemical measurements
4
physical and electrochemical parameters are listed in Table 1. expected if we consider the fact that the acceptor properties
Both compounds are characterized by a strong absorption of thiophene are better than that of benzene (ꢀ1.84 V and
ꢀ
1
ꢀ1
(
4
4
e
abs = 45 000 M cm ) covering the range from 550 nm to ꢀ2.56 V – estimated by addition of energy gap to the oxidation
32
00 nm. The maximum of absorption for 4a was located at potential of thiophene or benzene, respectively ). Surprisingly,
69 nm, while for 4b it was 10 nm bathochromically shifted. it does not improve the exothermicity of the photoinduced
The same trend was observed in fluorescence but the shift electron transfer process because the thiophene simulta-
was slightly larger (18 nm). Both compounds displayed weak neously deteriorates the oxidation properties of the probe.
fluorescence and short fluorescence lifetimes. The calculated
To apply the studied compounds in environmental or
constant rates of nonradiative deactivation were found to be medical studies, the impact of water on the absorption and
two orders of magnitude larger than radiative ones. A closer emission spectra of the dyes was also explored. In our further
look at the molecular structure of the investigated molecules experiments, we decided to use a 1 : 1 water/DMF solvent
shows that this effect may be associated with free motions mixture as a good compromise between the solubility of the
(rotations) of the ethylenecyanocarbonyl group, trans–cis photo- sensor and analyte, in particular, sulphite anions. In both
isomerisation or a radiationless back electron transfer process. cases, the maximum of absorption is red-shifted by about
However, a Gaussian-like, broad absorption band and a large 12 nm (emission spectrum: 12 nm for 4a and 4b, respectively)
Stokes-shift indicate the existence of the charge transfer emitting as compared to DMF solution. This bathochromic shift may be
state. In addition, the process is feasible from the thermodynamic explained in terms of the energy stabilization of the molecule
point of view. The calculated Gibbs energy of intramolecular charge induced by highly polar solvent (water). Moreover, the changes
transfer (ICT) is strongly negative (see Table 1). Therefore, it seems in the colour of absorption/emission were accompanied by
a fluorescence intensity reduction (F = 0.006 and F = 0.007
fl
fl
for 4b and 4a, respectively).
3.2. Anion recognition
Probes contained an a,b-unsaturated carbonyl system usually
detect anions in the Michael addition of a nucleophile to
the activated double bond system. The presented probes are
composed of a triphenylpyrazoline fluorophore and an ethylene-
cyanocarbonyl group ended with thiophene or phenyl. It is
expected that anions can be added to the b-position of the
doubly activated Michael acceptor to generate the stabilized
anionic species. In order to explore the selectivity of the sensors
to sulphite, absorption and emission spectra were recorded in
the presence of 45 equivalents of different anions: nitrate,
nitrite, bromide, chloride, fluoride, iodide, acetate, thiocyanate,
sulphate, sulphite, glutathione, cysteine, homocysteine, hydro-
Fig. 2 Electronic absorption and emission spectra of 4a (blue line) and 4b sulphide, bisulphite and cyanide. As shown in Fig. 3, the
(
red line) recorded in DMF solution.
investigated probes are sensitive to sulphite. Upon addition of
Table 1 Photophysical and electrochemical parameters of the investigated dyes dissolved in DMF. The Gibbs energy of the photoinduced electron
transfer process was calculated according to the following equation: DGet = Eox ꢀ Ered ꢀ E00 + C; C – the Coulomb term was taken as ꢀ0.056 eV. Redox
potentials were obtained from the position of the first oxidation/reduction peak and the DPV voltammograms can be found in the ESI (Fig. S9 and S10).
l
abs corresponds to the maximum of the lowest energy absorption band, E00 – energy gap between the HOMO and LUMO energy levels estimated from
the onset of the lowest energy absorption band
ꢀ
1
ꢀ1
7
ꢀ1
Compound
labs [nm]
e
abs [M cm
]
lflu [nm]
F
flu
t
flu [ns]
k
nr [10 s
]
Eox [V]
Ered [V]
E00 [eV]
DGet [eV]
4
4
a
b
469
479
47 090
43 496
575
593
0.014
0.016
0.18
0.20
7.78
8.00
0.679
0.713
ꢀ1.349
ꢀ1.273
2.353
2.275
ꢀ0.381
ꢀ0.345
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019
New J. Chem., 2019, 43, 874--883 | 877

View Article Online
NJC
Paper
Fig. 3 Absorption spectra of 4a (a) and 4b (b) in the presence of 45 equivalents of different anions. The bottom picture illustrates the colour change of
b upon addition of the anions under study.
4
sulphite to the solution of 4a (4b), the absorbance at 481 nm sulphite species (Fig. 3 and Fig. S11, ESI†). Therefore, a further
491 nm) was gradually decreased with a new band appearing at analysis of the photophysical data involves only the interactions
70 nm. Simultaneously, a distinct colour change from orange to between the studied probes and sulphite anion.
(
3
colourless was clearly observed (see Fig. 3 and Fig. S11, ESI†)
indicating that the intermolecular charge transfer process is
turned off due to the Michael addition of sulphite.
3
SO3
.3. Fluorescence response of compounds 4a and 4b towards
2
ꢀ
The newly formed band (at 370 nm) may be assigned to Free compound 4a, excited at the isosbestic point (407 nm),
3
3
the excitation of 1,3,5-triphenylpyrazoline (parent molecule)
displays two main bands locating at 462 nm (some traces of the
which is now electronically decoupled from the end group: band) and 587 nm, which can be attributed to the emission
phenyl or thiophene. In order to support this hypothesis, DFT from the locally excited state (LE) and intramolecular charge
calculations (in DMF) were performed and the values of vertical transfer state (ICT), respectively. The presence of sulphite leads
energy excitations are as follows: 520 nm (385 nm) for 4b to the disappearance of the ICT band and the appearance of a
(
4b + sulphite), 506 nm (386 nm) for 4a (4a + sulphite). new band at 475 nm, which can be attributed to the emission of
33
Certainly, the obtained values are slightly underestimated; 1,3,5-triphenylpyrazoline. The analysis of fluorescence changes
nevertheless, they reflect a general trend in hypsochromism at 475 nm showed that the observed emission enhancement is
sulphite
after the binding process of sulphite.
significant (R4a = I
0
/I = 43) (Fig. 4). Again, the experiment
It is worth adding that at pH = 7.2 (pH of dionized water shows that the Michael addition reaction interrupts the
ꢀ
used) both of the studied probes can also interact with HSO3
,
p-conjugation leading to the recovery of triphenylpyrazoline
but the photophysical changes were less dramatic than in the emission. This statement is additionally supported by the
presence of a sulphite anion. Hence, the observed differences studies of fluorescence excitation spectra of the substrate and
in the absorption spectra of the sulphite and bisulphite adduct product (Fig. S12 and S13, ESI†). The obtained spectra match
may point to a better selectivity of the investigated probes to well with the absorption spectra confirming that the emission
8
78 | New J. Chem., 2019, 43, 874--883
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
Fig. 4 Fluorescence changes induced by the Michael addition of sulphite anions. Top corresponds to 4a and bottom to 4b. Orange and red lines
correspond to the HEPES/DMF mixture and black to water/DMF solution. Right black bars describe the enhancement of fluorescence induced by the
presence of different anions recorded in water/DMF solution. Orange and red bars relate to the competition between sulphite and other anions.
at 475 nm originates from the locally excited state of the parent It is noteworthy that both values are lower as compared to
molecule.
1,3,5-triphenylpyrazoline dissolved in protic methanol – F
=
fl
34
It is worth noting that the difference in emission maxima 0.38. The quenching of pyrazolines’ native fluorescence may
(
ICT and LE) before and after sulphite addition is very large be caused by photoinduced electron transfer occurring between
Dl = 112 nm which results in well separated fluorescence bands. the electronically decoupled thiophene/phenyl and 1,3,5-triphenyl-
In addition, due to the low intensity of the ICT emission band, the pirazoline part. Obviously, the quenching process is more efficient
sensor can be considered as an OFF–ON fluorescence probe. These in the 4b-sulphite adduct due to the presence of a better electron
photophysical features can be an advantage in terms of accurate donor/acceptor group – thiophene.
measurements of the intensities of the two emission peaks and, in
Fig. 4 (orange and red bars) presents also the influence of
consequence, a good evaluation of the fluorescence intensity ratio. competitive anions on the selectivity of probes 4a and 4b for the
The photophysical behaviour of 4b is very similar to that detection of sulphite anions. Our studies revealed no effect of
observed for 4a. The product of Michael addition is characterized the second anion except for glutathione for which the fluorescence
by blue-shifted emission with a maximum at 478 nm. Interestingly, signal is reduced by 20% for 4a (35% for 4b). The analysis of the
the emission shift is higher (Dl = 127 nm, see Fig. 4) as compared literature shows that glutathione can be converted to glutathione-
35
to probe 4a confirming that the electron withdrawing effect is S-sulphonate in a large excess of sulphite concentration. Conse-
more pronounced in 4b because of the presence of a slightly better quently, it is very likely that some part of sulphite may be bound to
acceptor. What distinguishes the investigated probes is their the glutathione resulting in reduced fluorescence signal.
fluorescence intensity ratio in the presence and absence of
2
ꢀ
3.4. The effect of pH on SO
3
recognition
sulphite. 4b is characterized by lower maximal intensity and
sulphite
reduced fluorescence enhancement R4b = I
0
/I = 15 as The impact of pH on fluorescence spectra was studied in buffer
compared to 4a. The determined fluorescence quantum yields mixed solutions (DMF/HEPES 1 : 1) and the results are presented
for 4a-sulphite and 4b-sulphite equal 0.28 and 0.15, respectively. in Fig. 5. The fluorescence intensity monitored at 475 nm slowly
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019
New J. Chem., 2019, 43, 874--883 | 879

View Article Online
NJC
Paper
Scheme 2 A mechanism of sulphite-4b interaction.
was repeated with a model molecule (much better soluble in
1
D O/DMSO-d mixture). The H NMR spectra of the model and
2
6
model-sulphite adduct are compared in Fig. S20 (ESI†). Addition-
13
ally, the C NMR spectrum of the model is depicted in Fig. S21
2ꢀ
2ꢀ
Fig. 5 Effect of pH on the fluorescence of the sulphite-4a product and
sulphite free 4a probe (inset) monitored at 475 nm.
(
ESI†). Finally, the formation of the 4b-SO
3
(or 4a-SO
3
) adduct
was confirmed by HRMS measurements where a peak at m/z =
ꢀ
+
5
40.1057 (580.7468) corresponds to [4b-SO
3
]
3
([4a-SO + 2Na] )
(
see Fig. S22 and S23, ESI†). The adduct was received in two
but systematically increased over pH 5.0–10.00 (see inset
Fig. S14 and S15, ESI†), but the fluorescence still had a dual
character. This shows that the probes rather do not undergo
separate and independent reactions: the studied probes and
sulphite or bisulphite. In both cases, the same product was
obtained which was corroborated by the same MS peak position
corresponding only to the adduct with sulphite. Based on the
spectroscopic measurements, the following mechanism of the
Michael reaction between the investigated probes and sulphite
ꢀ
ꢀ
Michael reaction with OH and a higher concentration of OH
presumably leads to new charge relocation within the molecule
which may promote an increased population of the locally
LE CT
LE CT
excited state (I4b/I4b = 10, I4a/I4b = 2.73 while in water/DMF
the ratio amounts to 1 in both cases). Our experiment also
revealed that the best optical signal of product to optical signal
of substrate ratio was at pH = 8 for both studied probes. Similar
to water/DMF solution, the enhancement of fluorescence induced
by a Michael addition product is larger for 4a than 4b.
Finally, it is noteworthy that pH has a negligible impact on
the absorption of the free probes (see the Charts S16 and S17
in the ESI†) and, consequently, the observed effect arises from
nucleophilic addition of sulphite.
(
or bisulfite) was proposed and presented in Scheme 2.
3
.6. Limit of detection in DMF/buffer solution (pH = 8)
The detection limit was established by taking advantage of the
fluorescence intensity changes induced by the presence of different
analyte concentrations. The obtained calibration curves with fitted
parameters are presented in Fig. 6 (inset) and the sulphite
3.5. Sensing mechanism
In order to confirm the nucleophilic addition by sulphite to the
1
double bond system, the measurement of H NMR spectra
6 2
of the product (4a-sulphite adduct) in DMSO-d /D O was performed
(
1 mg, 2.2 mmol of 4a was dissolved in 0.3 ml of DMSO-d and a
6
solution of 50 mg, 400 mmol of anhydrous Na SO in 0.95 ml of
2
3
1
D O was added). The H NMR spectrum of the substrate exhibits a
2
characteristic singlet signal at 7.96 corresponding to the vinyl
proton which disappeared due to the nucleophilic addition of
sulphite to the vinyl linkage (Fig. S18, ESI†). Additionally, a new
signal appears around 5.25 ppm, which corresponds to the proton
linked to the carbon atom with an attached sulphite group.
This large downfield shift (from 7.96 to 5.24 ppm) is mostly caused
2
3
by the change of hybridization of the carbon atom from sp to sp .
As the solubility of the 4a-sulphite adduct was very low, the quality
of the H NMR spectrum is also diminished. To obtain more clear
Fig. 6 The kinetics of the analyte addition monitored at 587 nm (4a) and
1
6
05 nm (4b) in the presence of 45 equivalents of sulphite. Inset: The
results, a model compound was prepared (Fig. S19 and the
calibration curves prepared on the basis of fluorescence intensity changes
following synthetic procedure, ESI†). The addition of sulphite recorded at 475 nm (4a) and 478 nm (4b).
8
80 | New J. Chem., 2019, 43, 874--883
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
concentration dependent fluorescence spectra are gathered in Table 2 Determination of sulphite in Vistula river water samples
Fig. S24 (ESI†). Using the value of standard deviation of the
intercept (0.48 (4a), and 0.089 (4b)) and the slope of the
calibration line depicted in Fig. 6, the detection limit of
sulphite was determined as: 7.56 mM for 4a, and 4.87 mM for
32ꢀ
32ꢀ
Sample
SO
added (mM)
SO
found (mM)
Recovery (%)
4
b
a
9.69
7.2
19.8
8.67
16.4
31.5
90
95
159.3
1
4
4b. The yielded values are satisfactory, especially in terms
of WHO sulphite standard in water based flavoured drinks
3
6
3.8. Real sample applications
(maximum permitted level 1420 mM) which is significantly
higher than our detection limit. In the European Union, the
maximum permitted sulphite level in final food products is
The practical utility of the sensors used in this study has been
2ꢀ
tested in the determination of SO3 in Vistula river water
samples. The Vistula water sample was found to be free from
sulphite, so the sample was prepared by adding a known
amount of sulphite to the sample and compared with the same
amount dissolved in deionised water. The amounts of added
salt and recovery yields are listed in Table 2.
As it was shown in Table 2, compound 4b was able to
measure the concentration of spiked sulphite with good recovery,
suggesting that the probe can be applied for detecting sulphite in
real samples. In contrast, 4a cannot be used in environmental
applications. The compound 4a turned out to be not selective
enough in river water sample solutions. The found amount of
sulphite was twice as high as the real one. It is very likely that,
in the river sample solution, 4a may interact with another
species for which compound 4b is resistant.
ꢀ
1 37
2
mg kg
.
The obtained detection limit is compared to
3
8–42
recently investigated probes (Table S1 in the ESI†).
3.7. Response time in DMF/buffer solution (pH = 8)
In order to evaluate the kinetics of the analyte binding, time-
dependent fluorescence intensity changes were monitored at
587 nm (4a) and 605 nm (4b) in the presence of 45 equivalents
of sulphite (Fig. 6). It is worth emphasizing that the application
of smaller amounts of equivalents is also possible, however,
the responsiveness is respectively slower. Assuming that the
applied analyte is in excess, the Michael reaction addition
follows a pseudo-first order kinetics. The time-dependent
changes of the substrate concentration can be described by
the expression:
Additionally, the practical application of the studied probes
for sulphite detection was demonstrated by means of a paper
test strip system and shown in Fig. 7. After dipping the
probe-stained paper into the solution of sulphite, the paper
discoloured (photographs at the top of Fig. 7) and the emission
ꢀ
0
k[B ]t
[A] = [A
0
]e
(2)
where A refers to the investigated probe, and B defines the
analyte; [A] – the concentration of A at time t; [A ] – the initial
concentration of A; [B ] – the initial concentration of B; k is the
0
0
changed from orange or yellow to blue (photographs at the bottom
ꢀ
pseudo-1st-order reaction rate constant.
of Fig. 7). Although HSO
3
also caused a colour change, it could be
The fitting of eqn (2) to the fluorescence decay of the
substrate yielded a reaction rate constant of: 4b + sulphite:
easily distinguished by the naked eye. This experiment is in line
ꢀ1
ꢀ1
ꢀ1 ꢀ1
k4b = 20.7 M
s
, 4a + sulphite: k4a = 17.4 M
s . The
experiment showed that the Michael addition reaction of
sulphite is within 9 min (4b) and 11 min. (4a), which indicates
that the probes have rapid detection ability to sulphite.
Finally, it should be emphasized that the sulphite detection
limit of the studied probes is comparable to that obtained for a
recently published pyrene based sensor, whereas the kinetics
of sulphite addition is much faster (20 min in the presence of
2
0
500 equivalents).
Summing up the photophysical behaviour of the investi-
gated probes, one can conclude that the introduction of a
thiophene end group improved the LOD by 1.5 times and
accelerated the kinetics of Michael addition by 1.2 times as
compared to the probe ended with phenyl. This may be a
consequence of larger polarization of the double bond induced
by the presence of a better electron acceptor (thiophene group)
which, in turn, allows faster nucleophilic attack. Although the
fluorescence enhancement recorded for 4b-sulphite is smaller
than for 4a, it seems, however, that the optical signal is more
stable in the presence of competitive anions (see Fig. 4). Never-
theless, in both cases glutathione may interfere with fluores-
cence readings and result in artificial decreases of sulphite
concentration in a real sample.
Fig. 7 Photographs of the test paper exposed to various species in the
absence (top) and presence (bottom) of UV light (365 nm); from the left:
ꢀ
2ꢀ
HSO
3
, SO
3
, probe alone, and paper without the probe.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019
New J. Chem., 2019, 43, 874--883 | 881

View Article Online
NJC
Paper
with the previous ones (pH dependent effect of sulphite–probe
interaction and selectivity studies) confirming an increased sensi-
tivity of the probes to sulphite compared to the bisulphite anion.
3 N. Sang, Y. Yun, H. Li, L. Hou, M. Han and G. Li, Toxicol. Sci,
2010, 114, 226–236.
4 P.-J. Gauthier, O. Sigmarsson, M. Gouhier, B. Haddadi and
S. Moune, J. Geophys. Res.: Solid Earth, 2016, 121, 1610–1630.
5
6
7
E. Renc u¨ zo ˘g ullari, H. B. ˙I la, A. Kayraldiz and M. Topaktas-,
Mutat. Res., 2001, 490, 107–112.
S. H. Mudd, F. Irreverre and L. Laster, Science, 1967, 156,
4
. Conclusions
In summary, we have demonstrated two OFF–ON ratiometric
pyrazoline-based sensors which are able to detect sulphite
anions. The compounds exhibit weak charge transfer fluores-
cence which is blocked after the sulphite attachment to the
a,b-unsaturated ketone part. As a result, a blue-shift of the
absorption and emission spectra of the probes was observed,
corresponding to an obvious colour change from orange to
colourless which can be noticed by the naked eye. The detection
limit of the studied indicators is comparable to that of a
recently published pyrene based sensor, while the response
speed is twice as fast at a smaller concentration of sulphite.
These two parameters can be easily tuned by the introduction
of a better acceptor (end group). Here, a simple replacement of
phenyl to thiophene improved the response time by 1.2 times
and the LOD by 1.5 times. We believe that further increase
of the electron acceptor abilities of the end group will lead to
higher polarization of the double bond system and faster
optical response. Finally, our studies revealed that only 4b
can detect sulphite anions with a good selectivity and sensitivity
in environmental samples.
1599–1602.
W.-H. Tan, F. S. Eichler, S. Hoda, M. S. Lee, H. Baris,
C. A. Hanley, P. E. Grant, K. S. Krishnamoorthy and V. E.
Shih, Pediatrics, 2005, 116, 757–766.
8
9
C. Wang, S. Feng, L. Wu, S. Yan, C. Zhong, P. Guo,
R. Huang, X. Weng and X. Zhou, Sens. Actuators, B, 2014,
1
90, 792–799.
X. Cheng, H. Jia, J. Feng, J. Qin and Z. Li, J. Mater. Chem. B,
013, 1, 4110–4114.
2
1
1
1
1
1
0 K. Rurack, A. Danel, K. Rotkiewicz, D. Grabka, M. Spieles
and W. Rettig, Org. Lett., 2002, 4, 4647–4650.
1 M. Mac, T. Uchacz, A. Danel and H. Musiolik, J. Fluoresc.,
2013, 23, 1207–1215.
2 M. Mac, T. Uchacz, T. Wr o´ bel, A. Danel and E. Kulig,
J. Fluoresc., 2010, 20, 525–532.
3 M. Mac, T. Uchacz, A. Danel, K. Danel, P. Kolek and E. Kulig,
J. Fluoresc., 2011, 21, 375–383.
4 P. Wang, N. Onozawa-Komatsuzaki, Y. Himeda, H. Sugihara,
H. Arakawa and K. Kasuga, Tetrahedron Lett., 2001, 42, 9199–9201.
5 H.-B. Shi, S.-J. Ji and B. Bian, Dyes Pigm., 2007, 73, 394–396.
6 X. Dai, T. Zhang, Y.-Z. Liu, T. Yan, Y. Li, J.-Y. Miao and B.-X.
Zhao, Sens. Actuators, B, 2015, 207, 872–877.
1
1
Conflicts of interest
1
1
1
2
2
2
2
2
7 X. Dai, T. Zhang, J.-Y. Miao and B.-X. Zhao, Sens. Actuators,
B, 2016, 223, 274–279.
8 T. Haiyu, Q. Junhong, S. Qian, B. Hongyan and Z. Weibing,
Anal. Chim. Acta, 2013, 788, 165–170.
There are no conflicts to declare.
Acknowledgements
9 Y.-Q. Sun, P. Wang, J. Liu, J. Zhang and W. Guo, Analyst,
2
012, 137, 3430–3433.
0 G. Xu, H. Wu, X. Liu, R. Feng and Z. Liu, Dyes Pigm., 2015,
20, 322–327.
1 J. Shen and R. D. Snook, Chem. Phys. Lett., 1989, 155,
83–586.
2 J.-T. Li, X.-H. Zhang and Z.-P. Lin, Beilstein J. Org. Chem.,
007, 3, art. No13.
3 L. A. Kutulya, A. E. Shevchenko and Y. N. Surov, Chem.
Heterocycl. Compd., 1975, 11, 215–217.
4 K. Drauz, A. Kleeman and E. Wolfheuss, US Pat., 4.716.237,
This work was performed with equipment purchased thanks to
the financial support of the European Regional Development
Fund in the framework of the Polish Innovation Economy
Operational Program (contract no. POIG.02.01.00-12-023/08).
Quantum chemical calculations were carried out in the Academic
Computer Center ‘‘Cyfronet’’ in Krak o´ w, Poland. This research was
supported in part by PL-Grid Infrastructure. Chemicals for the
purpose of the research were financed by the Ministry of
Science and Higher Education of the Republic of Poland
1
5
2
(DS 3711/ICh/2017). The authors would also like to acknowl-
1987.
edge financial support of TEAM programme (grant number:
TEAM/2016-1/9) of the Foundation for Polish Science co-financed
by the European Union under the European Regional Develop-
ment Fund.
2
2
5 W. Kohn and L. J. Sham, Phys. Rev., 1965, 140, A1133–A1138.
6 R. G. Parr and W. Yang, Density-Functional Theory of
Atoms and Molecules, Oxford University Press, Oxford, 1988
(and references therein).
2
7 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
References
1
C. Ribbe, Napoleon’s Crimes: A Blueprint for Hitler, Oneworld
Publications, 2008.
2
Y.-C. Hong, J.-T. Lee, H. Kim, E.-H. Ha, J. Schwartz and D. C.
Christiani, Environ. Health Perspect., 2002, 110, 187–191.
8
82 | New J. Chem., 2019, 43, 874--883
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
Paper
NJC
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, 33 M. R. V. Sahyun, G. P. Crooks and D. K. Sharma, Proc. SPIE,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, 1991, 1436, 125–133.
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, 34 K. Rurack, J. L. Bricks, B. Schulz, M. Maus, G. Reck and
N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, U. Resch-Genger, J. Phys. Chem. A, 2000, 104, 6171–6188.
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. 35 D. A. Keller and D. B. Menzel, Chem.-Biol. Interact., 1989, 70,
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, 145–156.
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, 36 National coordinated survey of sulphite levels in sausages,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D.
cordial and dried fruit, http://www.foodstandards.gov.au/
science/surveillance/documents/Sulphites%20survey.pdf.
¨
Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian 09, Revision A.1, Gaussian, Inc., Wallingford, 37 Commission Regulation (EU) no. 1130/2011, 11 November 2011.
CT, 2009.
38 S. Samanta, P. Dey, A. Ramesh and G. Das, Chem. Commun.,
2016, 52, 10381–10384.
2
2
3
8 A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38,
3
098–3100.
39 Q. Sun, W. B. Zhang and J. H. Qian, Talanta, 2017, 162,
107–113.
40 J. B. Chao, Y. H. Liu and Y. Zhang, Spectrochim. Acta, Part A,
2015, 146, 33–37.
9 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785–789.
0 S. Miertu ˇs , E. Scrocco and J. Tomasi, Chem. Phys., 1981, 55,
1
17–129.
41 M. G o´ mez, E. G. Perez, V. Arancibia, C. Iribarren, C. Bravo-
D ´ı az, O. Garc ´ı a-Beltr ´a n and M. E. Aliaga, Sens. Actuators, B,
2017, 238, 578–587.
3
3
1 S. Miertu ˇs and J. Tomasi, Chem. Phys., 1982, 65, 239–245.
2 M. Montalti, A. Credi, L. Prodi and M. T. Gandolfi, Handbook
of Photochemistry, Taylor & Francis Group, LLC, 3rd edn, 42 G. Wang, H. Chen, X. L. Chen and Y. M. Xie, RSC Adv., 2016,
006. 6, 18662–18666.
2
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019
New J. Chem., 2019, 43, 874--883 | 883