Single molecular dual analyte thio-urea based probes for colorimetric Hg2+ and fluorometric AcO? detection and its’ application in bioimaging

Article abstract of DOI:10.1016/j.ica.2019.04.017

Two thio-urea based bifunctional probes (H₂L¹-S and H₂L²-S) have been designed and fabricated for the colorimetric and fluorometric detection of mercury(II) and acetate ions respectively in DMSO media. Both the

Full text of DOI:10.1016/j.ica.2019.04.017

Accepted Manuscript
Research paper
Single Molecular Dual Analyte Thio-Urea based Probes for Colorimetric Hg2+
and Fluorometric AcO − Detection and Its’Application in Bioimaging
Sumit Kumar Patra, Bhaskar Sen, Sanjoy Kumar Sheet, Bodhisattwa Banerjee,
Nirmalendu Saha, Snehadrinarayan Khatua
PII:
S0020-1693(18)31855-3
https://doi.org/10.1016/j.ica.2019.04.017
ICA 18862
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
14 December 2018
20 March 2019
10 April 2019
Please cite this article as: S. Kumar Patra, B. Sen, S. Kumar Sheet, B. Banerjee, N. Saha, S. Khatua, Single Molecular
Dual Analyte Thio-Urea based Probes for Colorimetric Hg2+ and Fluorometric AcO− Detection and Its’Application
in Bioimaging, Inorganica Chimica Acta (2019), doi: https://doi.org/10.1016/j.ica.2019.04.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Single Molecular Dual Analyte Thio-Urea based Probes
for Colorimetric Hg2+ and Fluorometric AcO− Detection
and Its’Application in Bioimaging
Sumit Kumar Patra,a Bhaskar Sen,a Sanjoy Kumar Sheet,a Bodhisattwa Banerjee,b
Nirmalendu Saha,b Snehadrinarayan Khatua* a
aCentre for Advanced Studies, Department of Chemistry, North Eastern Hill University,
Shillong, Meghalaya 793022, India. Email: snehadri@gmail.com; skhatua@nehu.ac.in
bDepartment of Zoology, North Eastern Hill University, Shillong, Meghalaya 793022, India.
Abstract
Two thio-urea based bifunctional probes (H2L1-S and H2L2-S) have been designed and
fabricated for the colorimetric and fluorometric detection of mercury(II) and acetate ions
respectively in DMSO media. Both the H2L1-S and H2L2-S were fully characterized by
various spectroscopic techniques and the solid-state structure of H2L1-S was determined via
single-crystal X-ray diffraction. The high sensitivity and selectivity of the probes for Hg2+
were achieved by the Hg2+ promoted desulfurization reactions utilizing the high thiophilic
nature of mercury and was monitored by UV–vis spectroscopy. Most interestingly the
desulfurized product was isolated and characterized by ESI-MS and single crystal X-ray
diffraction. The probes H2L1-S and H2L2-S were found to be highly selective towards acetate
ion (AcO-) in fluorescence channel. These probes showed a distinct fluorescence
enhancement only in the presence of AcO− at 493 nm due to the inhibition of photo induced
electron transfer (PET). Finally, the cytotoxicity study was carried out for biological
application. The probe showed minimum cytotoxicity and was suitable for intracellular
acetate ion imaging.
Keywords: Thio Urea; Bifunctional probe; Hg2+ detection; AcO- detection, Bioimaging
1

1. Introduction
The development of single molecular multifunctional probe is an expanding area of chemistry
because of their wide practical application in the environmental and medical fields [1-3].
Several strategies were already employed for multianalyte probes: (I) incorporation of several
binding sites, one for each analyte [4-6], (II) operating in single channel at different
experimental condition [7-9] and (III) single receptor in different channel interrogation,
ideally one channel per analyte [10-12]. In contrast to approaches (I) and (II), detection
through different channel interrogation which differently react with probe is more facile and
advantageous from a synthetic and atom-economic point of view. To date, there have been
several reports on single molecular bifunctional probes for multi cations [13-15] and multi
anions [16-18], but highly selective and sensitive detection for both cation and anion in two
different channels with different mode of action are rarely reported [19-20].
Among variety of heavy metal ions, mercury(II) ion (Hg2+) is one of the most toxic
metal ions which widely distributed in water and soil, especially those originated from
chemical industries [21-23]. Hg2+ metabolism by aquatic microbes produces methyl mercury,
which is a highly potent neurotoxin that leads to many serious human afflictions, such as
Minamata, edema and anemia, damage the central nervous system, even a very low
concentration [24-26]. Therefore, rapid and competent detection of Hg2+ has become very
vital to control its concentration level in environmental and biological samples to avoid its
direct impact in human health [27-29]. Recently, colorimetric methods for detection of Hg2+
are becoming a potential contender for replacing the traditional instrument-based methods
due to its operational simplicity, high selectivity, sensitivity, rapidity, low cost of equipment
and direct visual perception [30-32].
2

has long
been perceived as a crucial cellular molecule involved in several metabolic processes.
Currently,
many elegant techniques are available for qualitative and quantitative analysis of acetate ions,
however, these methods are less satisfactory for endogenous detection of acetate ion due to
tedious sample preparation procedures, sophisticated instruments and high maintenance
expenditure [38-39]. To realize in situ visualization of AcO− within a living cell, there is a
growing interest to develop small molecule based fluorescent probes in view of their high
sensitivity, good selectivity and real-time detection [40-42].
Among the variety of signal transductions, photo induced electron transfer (PET) has
been successfully employed in fluorescent sensor design for different analytes by many
groups [43-44]. Upon excitation of the fluorophore, PET occurs from electron rich group to
electron deficient module. The interaction of analytes with electron donating or electron
withdrawing group are preventing the PET by obstructed the electron transfer process
resulting switch-ON of the fluorescence [45].
For several decades thio-urea based receptors have been acclaimed for detection of
either F‒ and AcO‒ or Hg2+ ion [46-47]. As part of our ongoing research in small organic,
ruthenium(II) and organoiridium(III)-complex based multifunctional probes [48-52], herein
we present the synthesis of two simple thio-urea based bifunctional probes (H2L1-S, H2L2-S)
and their aptness for highly sensitive and selective detection of Hg2+ and AcO- ions by two
distinct detection approaches. These probes selectively detect Hg2+ in the UV−vis
spectroscopy through the colour change from yellow to colourless associated with a notable
3

blue shift. The very weakly emissive probes H2L1-S and H2L2-S selectively interact with
AcO− rapidly (within 1.5 mins) and emits strongly at 493 nm. Finally, the cytotoxicity and the
intracellular AcO- imaging in live cell are also discussed herein. To the best of our knowledge
there is no such report on a single molecular thio-urea based probe for detection of two
analytes, namely toxic Hg2+ and biologically important AcO- in two different spectroscopic
channels through different reaction mechanism.
2. Experimental
2.1. Material and physical measurements
All chemicals were used as received from commercial suppliers (Aldrich, Alfa Aesar and
1
Spectrochem India). H and 13C NMR spectra were measured on a Bruker Avance II
(400MHz) spectrometer and chemical shifts were expressed in ppm using using residual
protic solvent as the internal standard . ESI-MS was performed with a Waters ZQ-4000 and
QToF−Micro YA 263 mass spectrometer. Infrared spectra were recorded using a Perkin-
Elmer FT- IR spectrometer with KBr pellets in the range of 4000–400 cm−1. The X-ray data
was collected at 293 K with a Agilent Xcalibur (Eos, Gemini) diffractometer and Bruker
KAPPA APEX II CCD diffractometer using graphite-monochromated Mo-Kα radiation (λ =
0.71073 Å). Elemental analysis were done using the PerkinElmer 2500 series II elemental
analyzer. UV-visible and PL spectra were recorded on a Perkin-Elmer Lambda 25 UV–vis
spectrophotometer and Hitachi F-4500 fluorescence spectrophotometer with quartz cuvette
(path length = 1 cm).
2.2. Methodology for UV-vis experiments
In the UV-vis channel, the measurements were conducted as follows. Firstly, the stock
solutions of H2L1-S and H2L2-S were prepared at a concentration of 100 μM in DMSO. All
perchlorate salt of cation including Hg(ClO4)2 were used to prepare the stock solution in
water (10 mM). Before the measurements, the probe stock solution of H2L1-S and H2L2-S
4

were diluted to 20 M, and then the Hg2+ was added into the diluted probe solution and UV–
vis spectra were recorded at room temperature immediately. All spectra of H2L1-S and H2L2-
S were recorded in 30% and 5% aqueous DMSO solution respectively.
2.3. Methodology for emission experiments
A stock solution (1 mM) of the probe H2L1-S and H2L2-S were prepared in DMSO which
were diluted to 20 M or 50 M before data collection. DMSO solution of all anions
including tetra-n-butyl ammonium acetate (100 mM) were prepared. Samples for emission
measurement were contained in 1 cm × 1 cm quartz cuvettes (3.5 mL volume). All
spectroscopic measurements of H2L1-S and H2L2-S were performed in DMSO solution. An
excitation wavelength at 412 nm and 405 nm was used for probe H2L1-S and H2L2-S
respectively. Excitation and the emission slit were set to 10 mm/10 nm and the PMT voltage
at 700 V.
2.4. Calculation of excited states lifetimes
The luminescence lifetime of H2L1-S and H2L2-S before and after addition of AcO¯ were
carried out under ambient conditions using a time-correlated single-photon counting
(TCSPC) spectrometer [a picosecond diode laser (IBH, UK)] with the detection wavelength
at 493 nm for H2L1-S and H2L2-S both in the presence and absence of AcO¯ (ex = 412 nm
and 405 nm for H2L1-S and H2L2-S respectively). The fluorescence decays were monitored at
the corresponding emission maxima as observed in the steady state fluorescence
measurement. The fluorescence decay data were collected on a Hamamatsu MCP PMT
(R3809) and were analysed by using IBH DAS 6 software. Nano LED at 416 nm was used as
the excitation source. All the fluorescence decays were fitted with a biexponential function
considering a χ2 value close to 1, which is an indication of good fitting. Experimental time-
resolved luminescence data were calculated using the following multiexponential decay
equation
5

<>aii
Whereas, ai is the amplitude of the ith decay component (ai = αi/Σαi) and i is the excited state
luminescence life time of the ith component.
2.5. Determination of detection limit
The detection limit (DL) of the probes H2L1-S and H2L2-S was calculated based on the
fluorescence titration data and according to the equation DL = 3 × SD/slope, Where SD is the
standard deviation of the blank solution (obtained by the 10 consecutive scan of the blank
sample)
2.6. Quantum Yield Calculation
The fluorescence quantum yield was determined using fluorescein (ФR = 0.91 in 0.1 M
NaOH) as an optically matching standard. The quantum yield is calculated according to the
following equation:
ɸx = ɸs (Fx/Fs) {(1-10^-As)/ 1-10^-Ax)} (x/s) 2
Here, ФX and ФS are the fluorescence quantum yield of the probe and standard respectively.
FX and FS are the integrated area under the corrected emission spectra of the probe and the
standard, respectively. Ax and AS are the corresponding absorbance of probe and standard at
the excitation wavelength and  is the refractive index of the solvents.
2.7. Synthesis of 1-(4-methylbenzo[d]thiazol-2-yl)-3-(4-nitrophenyl) thiourea (H2L1-S)
The synthetic route of 1-(4-methylbenzo[d]thiazol-2-yl)-3-(4-nitrophenyl) thiourea (H2L1-S)
is depicted in Scheme 1. 2-amino-4-methylbenzothiazole (0.33 g, 2.00 mmol) and 4-
nitrophenyl isothiocyanate (0.40 g, 2.20 mmol) were dissolved in 15 ml of dry CH3CN. Then
the mixture was refluxed for 3 hours under N2. An off-white precipitate was obtained after 30
mins and the reaction was continued for 3 hours. The precipitate was collected by filtration
and washed thoroughly with CH3CN and diethyl ether and then dried under vacuum. Yield:
6

50% (0.320 g). M.P = 228-233 °C. Elemental analysis: calcd. (%) for C15H12N4O2S2 (MW =
344.41): C, 52.31; H, 3.51; N, 16.27; found: C, 52.35; H, 3.49; N, 16.28; FTIR in KBr disk
(νmax /cm-1): 2956, 1575, 1515, 1335, 1264, 1193, 849, 740, 693. ESI-MS (positive mode):
1
(m/z, %): 345.05 [(C15H12N4O2S2+H)+, 100%]; calcd. m/z, 345.04]; H-NMR (400 MHz,
DMSO-d6): (ppm) = 13.21 (s, 1H, Ha), 10.94 (s, 1H, Hb), 8.21 (d, J = 8.8 Hz, 2H, H2, H2 ),
́,
8.14 (d, J = 8.8 Hz, 2H, H3, H3 ), 7.73 (d, J = 7.2 Hz, 1H, H9), 7.23- 7.30 (m, 2H, H10,H11),
́
2.57 (S, 3H, H12); 13C-NMR (100 MHz, DMSO-d6): (ppm) = 145.8 (2C,C1, C5), 141.8 (1C,
C4), 128.1 (3C, C7, C10, C12), 124.6 (2C, C2, C2), 124.0 (1C, C11), 120.8 (2C, C3, C3), 120.1
́
́
(3C, C6, C8, C9), 17.9 (1C, C13).
2.8. Synthesis of 1-(1H-benzo[d]imidazol-2-yl)-3-(4-nitrophenyl) thiourea (H2L2-S)
Synthesis of 2-(4-naphthalen-1-yl-phenyl)-2H-[1,2,3] triazole-4-carbaldehyde (H2L2-S) is
shown in Scheme 1. A mixture of 2-amino benzimidazole (0.27 g, 2.00 mmol) and 4-
nitrophenyl isothiocyanate (0.40 g, 2.20 mmol) was refluxed in dry CH3CN under N2 for 3
hours. A greenish yellow precipitate was appeared after 30 mins and the reaction was
continued to 3 hours to complete the conversion. The precipitate was collected by filtration,
washed with CH3CN and diethyl ether thoroughly and dried under vacuum. Yield: 61%
(0.382 g). M.P = 217-222 °C. Elemental analysis: calcd. (%) for C14H11N5O2S (MW =
313.33): C 53.66, H 3.54, N 22.35; found: C 53.70, H 3.52, N 22.38; FTIR in KBr disk (νmax
/cm-1): 3259, 1621, 1582, 1547, 1453, 1323, 1295, 1246, 1112, 848, 747. ESI-MS; (positive
+
1
mode): (m/z, %): 314.07 [(C14H11N5O2S + H) , 100%]; calcd. m/z, 314.06; H-NMR (400
MHz, DMSO-d6): (ppm) = 12. 95 (s, 2H, Ha, Hb), 10.39 (s, 1H, Hc), 8.11-8.16 (m, 4H, H2,
H2 H3, H3), 7.52-7.55 (m, 2H, H10, H12), 7.24-7.27 (m, 2H, H9, H11);13C-NMR (100 MHz,
́
́
DMSO-d6): (ppm) = 153.0, 146.6, 140.9, 128.9, 124.5, 123.1, 121.1, 119.9, 111.9, 111.4.
2.9. Cytotoxicity Study.
7

The cytotoxicity of the probe H2L1-S on HEK293T cells was determined by MTT assay as
described by Mosmann, (1983) in a 96-well cell culture plate. The cells were seeded in a 96-
well plate containing 2 ml of DMEM (Dulbecco's Modified Eagle Medium) at a density of 4
× 103 cells/well and incubated at 37 °C, 5% CO2 incubator. Cells were treated with at
different concentration (5, 10, 25, 50, and 100 μM) at 70% confluency and incubated for 24
h. After incubation, media was replaced by same volume of serum free DMEM media and
MTT salt was added in medium to a final concentration of 0.5 mg/ml. The plate was
incubated for 4 hours at 37 °C until intracellular purple formazan crystals were visible under
microscope. After incubation, the media was discarded and 200 μl of DMSO was added in
well. The DMSO was mixed with purple formazan salt by shaking the plate gently. The
absorbance change was monitored at 570 nm using iMark™ (Bio-Rad, USA) microplate
absorbance reader. Readings were taken in triplicate and the % cell viability was calculated
for samples and controls based on the following formula:
% cell viability = {(Abstreated – Absblank) / (Abs of control – Abs of blank)} × 100
2.10. Cell Culture and Imaging.
Cells were seeded at normal confluency (~10%) in 12 well sterile culture plates on poly-l-
lysine coated coverslips containing 2 ml of DMEM (Dulbecco's Modified Eagle Medium)
medium and incubated inside a CO2 incubator at 37 °C with 5% CO2 supplemented with 1%
streptomycin penicillin and 10% FBS (Fetal Bovine Serum). When the cells were
approximately 70% confluent, fresh media was added in two successive wells. One well was
treated with 50 μM aqueous solution of H2L1-S (20% DMSO as co solvent) and incubated for
30 min. Afterward, 50 μM of AcOˉ was added and incubated for 10 min. The other well was
treated with only H2L1-S (50 μM) and incubated for 30 min. After incubation, cells were
washed with 500 μl of HBSS (Hank's Balanced Salt Solution) and fixed with chilled absolute
ethanol. Further, the cells were washed three times with HBSS (500 μl) to remove residual
8

ethanol. After washing, the cell containing coverslips were mounted on glass slides. Another
set of slides were processed in the same way except incubation with probes, which served as
negative controls. Imaging studies of probe H2L1-S were performed by conducting fixed-cells
imaging experiments by confocal laser scanning microscope (Leica, TCS SP5, Germany).
Cross-talk of fluorochromes was excluded by the use of the acousto optical tunable filter.
2.11. Theoretical study
The electronic properties of all probes and their final products were investigated by means of
density functional theory (DFT) calculations [53]. All calculations were performed using the
B3LYP exchange correlation functional, as implemented in the Gaussian 03(G03) program
package [54]. The 6–31G (d, p) basis set was assigned for all the elements.2.12.
Crystallographic studies
The X-ray data of H2L1-S and H2L1-O was collected at 293 K with Agilent Xcalibur (Eos,
Gemini) diffractometer and Bruker KAPPA APEX II CCD diffractometer using graphite-
monochromated Mo-Kα radiation (λ = 0.71073 Å). For H2L1-S the data was collected and
reduced in CrysAlis PRO software [55]. The absorption was corrected by SCALE3
ABSPACK multi-scan method in CrysAlisPro. The structures were solved by direct methods
using the program SHELXS-97 [56] and refined by full matrix least-squares calculations (F2)
by using the SHELXL-2014 [57] software. For H2L1-O the data integration and reduction
were processed with SAINT [58] software. A multi-scan absorption correction (SADABS)
[59] was applied to the collected reflections. The structures were solved by direct methods
using the program SIR92 [60] and refined by full matrix least-squares calculations (F2) by
using the SHELXL-2014 [57] software. For both compounds all non-H atoms were refined
anisotropically against F2 for all reflections. All hydrogen atoms except N-H were placed at
their calculated positions and refined isotropically. The N-H hydrogen was found on a
difference Fourier map and refined isotropically. Structure analysis was aided by use of the
9

programs PLATON [61]. The cif files were deposited with the Cambridge Crystallographic
Data Centre and the following codes were allocated for (H2L1-S) and (H2L1-O) respectively:
CCDC-1539236 and CCDC-1539237. This data can be obtained free of charge via the
Internet: www.ccdc.cam.ac.uk/ data request/cif. Crystal data collection and refinement
details, for probe (H2L1-S) and (H2L1-O) are given in Table 1 and selected bond lengths and
angles are given Table S2 and Table S3.
Table 1. Crystal data and structure refinement for H2L1-S and H2L1-O
H2L1-S
H2L1-O
C17H18N4O4S2
Empirical formula
Formula weight
Temperature
C17H18N4O3S3
422.53
406.47
293(2) K
0.71073 Å
Monoclinic
P21/c
293(2) K
Wavelength
0.71073 Å
Monoclinic
P21/c
Crystal system
Space group
Unit cell dimensions
a = 10.1731(8) Å
b = 22.6224(19) Å
c = 8.6431(5) Å
= 90°
a = 9.9940(11) Å
b = 9.8485(10) Å
c = 19.010(2) Å
= 90°

= 99.600(6) °
 = 90°
= 96.451(4) °
 = 90°


1961.3(3) Å3 ,4
,
3
Volume, Z
1859.3(4) Å 4
1.431 Mg/m3
0.404 mm-1
880
1.452 Mg/m3
0.318 mm-1
848
Density (calcd.)
Absorption coefficient
F (000)
Ɵ
range for data 3.38 to 26.37°.
2.804 to 25.071°.
collection
Reflections collected
Independent reflections
Refinement method
7598
32484
3999 [R(int) = 0.0260]
Full-matrix least-squares on F2
3296 [R(int) = 0.0814]
Full-matrix least-squares on
F2
Data
/
restraints / 3999 / 0 / 255
3296 / 0 / 255
parameters
Goodness-of-fit on F2
1.028
1.085
Final R indices [I>2σ(I)] R1 = 0.0525, wR2 = 0.1027
R1 = 0.0489, wR2 = 0.1097
R1 = 0.0647, wR2 = 0.1178
R indices (all data)
R1= 0.0800, wR2 = 0.1141
10

3. Result and Discussion
3.1 Synthesis and characterization
The benzothiazole and benzimidazole attached thiourea based probes H2L1-S and H2L2-S
were synthesized by refluxing 4-nitrophenyl isothiocyanate with 2-amino-4-
methylbenzothiazole and 2-aminobenzoimidazole in CH3CN under N2 atmosphere for 3 hours
respectively (Scheme 1). The probes, H2L1-S and H2L2-S were fully characterized by
13
elemental analysis, 1D (1H, C) and 2D NMR spectroscopy, ESI-MS spectrometry (Figure
S1- S12). The solid-state structure of H2L1-S was determined by the single crystal X-ray
diffraction (Figure 1). Photophysical properties were determined by UV−vis and PL
spectroscopy. NMR spectra of H2L1-S and H2L2-S were recorded in DMSO-d6 at room
temperature, clearly showed all expected resonance. All the proton and carbon signals of
1
H2L1-S and H2L2-S are fully assigned with the help of H-1H COSY, 1H-13C HSQC and 1H-
13C HMBC NMR spectra. The ESI-MS spectra of H2L1-S and H2L2-S exhibited peaks at m/z
=345.05 (calcd. 345.04) and 314.07 (calcd. 314.06) respectively, which are assigned to singly
positive charged species.
Scheme 1. Synthesis of probe, H2L1-S and H2L2-S.
11

3.2 Crystal structure description of compound H2L1-S.
The diffraction quality yellow, plate like single crystals of the compound H2L1-S was
obtained from diethyl ether diffusion in DMSO after two weeks. The compound was
crystallized in monoclinic space group P21/c. The asymmetric unit contains the compound
H2L1-S and a DMSO molecule as a solvent of crystallization (Figure 1a). Analysis of the
crystal packing in H2L1-S reveals that the compound forms noncovalent 2D zig-zag sheet
(Figure 1b and 1c) which is held together by weak C−H···O, N−H···O and C−H···S
hydrogen bonding interactions. The oxygen centres (O1 and O2) of nitro group form H-bond
with C(4)-H of benzothiazole and C(17)-H of solvent DMSO molecule respectively. Further,
the S2 and N(2)-H of thio urea form hydrogen bond with C(5)-H of neighbouring probe
molecule and O3 of solvent DMSO. The packing structures indicate that the solvent DMSO
molecule plays an important role in formation of the zig-zag 2D sheet.
(a)
(b)
(c)
12

Figure 1. (a) ORTEP diagram of H2L1-S with 30 % ellipsoid probability. (b) View of the
hydrogen bonded 2D zig-zag sheet along crystallographic c-axis. (c) Space filling model of the
zig-zag sheet.
3.3 Detection of Hg2+ in UV-vis spectroscopy
3.3.1 Selectivity and Sensitivity of H2L1-S and H2L2-S to Hg2+
Both the probes, H2L1-S and H2L2-S were used for the detection of various cations like Na+,
K+, Mg2+, Ca2+, Fe2+, Co2+, Ni2+, Zn2+, Ag+, Cd2+, Pb2+, Al3+ and Hg2+ by UV−vis
spectroscopy in aqueous DMSO solution. For H2L1-S all data was collected in 30% aqueous
DMSO solution, since, up to 30% aqueous DMSO solution notable change in UV-Vis
spectrum and by naked eye was clearly observed in presence of only 1 equiv. of Hg2+(Figure
S13). Whereas, in the case of H2L2-S,with increase in the water fraction above 5% no notable
colour change was observed in the presence of Hg2+. Upon the addition of various cations to
the 30% aqueous DMSO solution of H2L1-S (20 μM), the n- band at 370 nm ( = 25200
Lmol-1cm-1) was blue-shifted to 340 nm ( = 21750 Lmol-1cm-1) and a prominent colour
change from yellow to colourless can be readily observed by the naked eye only in the
presence of Hg2+ (Figure 2a). Likewise, H2L2-S (20 M) in the presence of Hg2+ shows a
distinguished colour change from yellow to colourless and a notable blue shift (~8 nm) in 5
% aqueous DMSO solution (Figure 2b). In contrast, no significant shift of max and the color
were observed in the presence of other cations. The addition of increasing amounts of Hg2+
(0-1.0 equiv.) to H2L1-S (20 μM) in 30% aqueous DMSO solution at room temperature
reveals a gradual blue shift of the absorption band from 370 nm to 340 nm (Figure 3a).
Whereas, during the UV-vis titration of H2L2-S (20 μM) with increasing amount of Hg2+ (0-
25 equiv.), a gradual blue shift of the absorption band from 375 ( = 23000 Lmol-1cm-1) to
13

367 nm ( = 18150 Lmol-1cm-1) is observed (Figure 3b). In addition, a new band is appeared
slowly at 270 nm ( = 34750 Lmol-1cm-1).
(a)
(b)
Figure 2. UV-vis spectra of (a) H2L1-S (b) H2L2-S (20 μM) upon addition of various metal
ions (Hg2+, Al3+, Na+, K+, Mg2+, Ca2+, Mn2+, Co2+, Ni2+, Cu2+, Zn2+, Pb2+, Ag+, and Cd2+,) in
30% and 5% aqueous DMSO solution. (Inset) Naked eye visible color change of in addition
of Hg2+ in H2L1-S.
(a)
(b)
Figure 3. UV−vis titration of (a) H2L1-S (20 μM) (b) H2L2-S (20 μM) with increasing
amount of Hg2+ solution (0-1 equiv. and 0-25 equiv.) in 30% and 5% aqueous DMSO
solution respectively.
A competition experiments with other metal ions were carried out in UV-vis
spectroscopy to check for possible interferences in Hg2+ detection. Addition of other metal
ions in excess (5.0 equiv.) did not influence the UV-vis response of probe H2L1-S (Figure 4a)
14

and H2L2-S (Figure 4b) for Hg2+detection, indicating the high selectivity and sensitivity
towards Hg2+.
(a)
(b)
Figure 4. The UV response of (a) H2L1-S (20 μM) (b) H2L2-S (20 μM) in the presence of
various competing cations (5 equiv. and 125 equiv.) without and with Hg2+ (1 equiv. and 25
equiv.) in 30% and 5 % aqueous DMSO solution at 370 nm and 375 nm respectively.
The time course for the Hg2+ detection by H2L1-S and H2L2-S in 30% and 5 %
aqueous DMSO solution at room temperature was monitored. An immediate absorbance and
respective colour change was observed for H2L1-S (<40 s) (Figure 5a), while for H2L2-S ~2.5
mins after Hg2+ addition was essential for complete reaction (Figure 5b).
(a)
(b)
Figure 5. Time course of the UV response of (a) H2L1-S and (b) H2L2-S upon addition of 1
equiv. and 25 equiv. of Hg2+ at 370 and 375 nm respectively.
15

3.3.2 The plausible mechanism of Hg2+ detection and supporting evidences
Similar to reported S-based Hg2+probes [62], herein, we assume that the Hg2+ selective
signalling by H2L1-S and H2L2-S is due to the Hg2+-promoted desulfurization of thiourea
derivative (Scheme 2). According to previous reports [62], due to the interaction of Hg2+ (soft
acid) with sulphur center (soft base) of thiourea, the thio urea carbon (C=S) in H2L1-S and
H2L2-S will be more electrophilic and attacked by water more easily, leading to the formation
of 1-(4-methylbenzo[d]thiazol-2-yl)-3-(4-nitrophenyl) urea (H2L1-O) and 1-(1H-
benzo[d]imidazol-2-yl)-3-(4-nitrophenyl) urea (H2L2-O). In H2L1-S, the Hg2+ assisted
desulfurization reaction is more facile because of the presence of two ‘S’ atoms which assist
the chelation with Hg2+. Consequently, only 1 equiv. of Hg2+ is required for desulfurization of
H2L1-S and the reaction completed within 40 seconds. Whereas, in H2L2-S, the
desulfurization reaction is comparatively slow (2.5 minute) and require ~25 equiv. of Hg2+
since the probe contain only one ‘S’ atom and the coordination with Hg2+ is weaker.
To verify the mode of interaction, we performed a reaction of H2L1-S and H2L2-S
with HgClO4·6H2O in 5 % aqueous DMSO solution and characterized the reaction mixture
by ESI-MS. In ESI-MS spectra, a peak at m/z = 345.05 (calcd. m/z = 345.04) for compound
H2L1-S entirely vanished with a generation of new peak at m/z = 329.06 attributed to
desulfurized product, H2L1-O (calcd. m/z =329.06) (Figure S14). Similarly, for H2L2-S a
new peak at 298.08 (calcd. m/z = 298.09), corresponding to [C14H11N5O3 + H]+ confirm the
formation of desulfurized product H2L2-O (Figure S15). Finally, the isolated desulfurized
product, H2L1-O was structurally characterized by single crystal X-ray diffraction. (Figure 6).
The diffraction quality crystal was obtained from ether diffusion in DMSO solution. The
H2L1-O was crystallized in monoclinic P21/c space group. The asymmetric unit contains the
compound H2L1-O and a DMSO molecule as a solvent of crystallization. In the compound,
the solvent DMSO molecule forms H-bond with the N-H hydrogens of the urea (Figure 6b).
16

The crystal packing analysis shows the formation of supramolecular 3D network in H2L1-O
through the intermolecular C−H···O, efficient - interaction (Figure 6c).
Scheme 2. Hg2+-selective signalling mechanism of thiourea based chemodosimeter H2L1-S
and H2L2-S
(a)
(b)
(c)
Figure 6. (a) Crystal structure of the compound H2L1-O with 30 % ellipsoid probability. (b)
View of the intermolecular hydrogen bonding interaction in the compound with solvent DMSO
molecule. (c) View of the supramolecular 3D network in compound H2L1-O.
17

3.3.3 Theoretical Study
To gain insight into the electronic transitions and relation between structural and molecular
spectral changes of H2L1-S and H2L2-S upon Hg2+ addition, the geometry optimization and
time-dependent DFT (TD-DFT) calculations were carried out in DMSO. The transitions in
the UV−visible region between 280 and 540 nm are assigned to various electronic transitions.
Selected calculated molecular orbitals (MOs) and geometry optimized structures of H2L1-S,
H2L2-S, H2L1-O and H2L2-O are shown in Figure 7. The final products, H2L1-O (E =
−49508.78 eV) and H2L2-O (E = −39111.45 eV) are more stable than H2L1-S (E =
−49505.24eV) and H2L2-S (E = −39106.83 eV) and the stability is gained as 3.5 eV and 4.6
eV free energy is released, respectively. In Table S1 (see SI), the computed vertical excitation
energies and composition of the related transitions assigned to the experimental UV−vis
spectra and theoretical UV−vis spectra (Figure S16) are displayed. The TD-DFT calculations
for H2L1-S indicate that the experimental transition band at ∼370 nm arises from a strong
transition characterized as HOMO→ LUMO (f = 0.12, max = 391.84 nm). In contrast, for
H2L1-O, the band centered at 340 nm is due to the HOMO−2 →LUMO (f = 0.29, max = 334
nm) transition. In H2L2-S, experimental transition band at 375 nm is due to the transitions
from HOMO−1 →LUMO and HOMO−2 → LUMO (f = 0.10, max= 376.48 nm) whereas, in
H2L2-O the band at 270 nm is due to the HOMO →LUMO+2 (f = 0.45,max = 263.04 nm)
transition.
18

Figure 7. View of the frontier molecular orbitals (MOs) of H2L1-S, H2L2-S, H2L1-O and
H2L2-O [isovalue = 0.03].
3.4 Acetate (AcOˉ) Detection in Fluorescence Spectroscopy
3.4.1 Fluorescence spectral changes of probes in the presence of AcOˉ
The aptitude of H2L1-S and H2L2-S to serve as probes for different anions (100.0
equiv.) was tested in fluorescence spectroscopy in DMSO at room temperature. In the
fluorescence channel, probes H2L1-S and H2L2-S display a very weak fluorescence band at
465 (Ф = 0.008) and 460 nm (Ф = 0.012), respectively (λex = 412 and 405 nm) in DMSO at
25 °C. Both the probes, H2L1-S and H2L2-S show maximum fluorescence enhancement only
in the presence of AcOˉ at 493 nm (Ф = 0.095 and 0.083 respectively) and greenish blue
fluorescence (Figure 8a and 8b inset) while other anions such as Fˉ, Clˉ, Brˉ, Iˉ, H2PO4ˉ,
2
2
HSO4ˉ, PPi, CNˉ, N3ˉ, S2O3 ˉ, SO4 ˉcould only give negligible fluorescence increase (Figure
8a, b). During fluorescence titration in DMSO, ~10-fold (Figure 9a) fluorescence intensity
enhancement is observed for H2L1-S in the presence of 100 equiv. of AcOˉ (for H2L2-S, ~4-
fold; Figure 9b). The nonlinear curve fitting analysis from the fluorescence titration data of
H2L1-S and H2L2-S with AcO– shows good fit with the 1:1 model (R = 0.9936 and 0.98775
19

respectively). (Figure 10a, 10b). Moreover, the Job plot analysis also showed the inflection
point at 0.5, signifying a 1:1 stoichiometry between probe H2L1-S and H2L2-S toward AcOˉ.
(Figure S17, S18). Under well-established sensing conditions, the detection limit of H2L1-S
and H2L2-S was calculated to be as low as 0.47 μM and 0.15 μM respectively (Figure S19,
S20) from fluorescence titration data in DMSO. This specifies that these probes can be used
to detect very low concentration of AcO-.
(a)
(b)
Figure 8. (a) Fluorescence spectra of (a) H2L1-S (50 μM, λex = 412 nm, λem = 493 nm) (b)
H2L2-S (50 μM, λex = 405 nm, λem = 493 nm) with various anions, (100 equiv). (Inset) Naked
eye fluorescence colour change under UV light.
The fluorescence lifetime (τ) of H2L1-S (0.04 ns) and H2L2-S (0.33 ns), measured in DMSO,
were increased to 2.89 and 2.79 ns respectively in the presence of 100 equiv. of AcO – (Figure
11 a, b). Also, a time-dependent fluorescence response studies were carried out for both the
probes H2L1-S and H2L2-S with 100 equiv. of AcOˉ over 20 mins. As shown in Figure 12 (a
and b), a steady fluorescence intensity enhancement is observed during the first 1.5 min
before it reaches to saturation.
20

(a)
(b)
Figure 9. Fluorescence titration of (a) H2L1-S (50 μM) (b) H2L2-S (50 μM) in DMSO with
increasing amount of AcOˉ solution (0-100 equiv.).
(a)
(b)
Figure 10. Non-Linear curve fitting for the 1:1 model from the PL titration of (a) H2L1-S
(b) H2L2-S with AcO– Binding (λem = 493 nm).
21

(a)
(b)
Figure 11. Time resolved fluorescence spectra of (a) H2L1-S (50 μM) (b) H2L2-S (50 μM)
before and after addition of AcO– in DMSO.
(a)
(b)
Figure 12. Time course of the fluorescence response of (a) H2L1-S (b) H2L2-S with 100
equiv. of AcOˉ in DMSO at 493 nm.
The PL response of H2L1-S and H2L2-S toward AcOˉ (100 equiv) the presence of
other competitive anions and mercury (500 equiv.) were measured to examine the possible
interference. The tested anions and Hg2+ did not interfere in AcOˉ detection in fluorescence
spectroscopy channel. (Figure S21, S22)
3.4.2 Plausible mechanism and ESI–MS analysis:
The proposed mechanism for the AcOˉ detection by the probes H2L1-S and H2L2-S is
described in Scheme 3. As reported earlier [46], here we assumed that the deprotonation of
H2L1-S and H2L2-S by the AcO‒ ion increased the electron density in the p-nitrobenzene
22

moiety. Probe H2L1-S and H2L2-S shows very weak fluorescence band due to the photo-
induced electron transfer (PET) from benzothiazole/benzimidazole to p-nitrobenzene moiety.
While, due to the formation of deprotonated species, [L1-S]2- and [HL2-S]- after to the
interaction of AcO- with H2L1-S and H2L2-S, the electron density in p-nitrobenzene moiety
is slightly increased and photo induced electron transfer process (PET) is inhibited,
consequently, the probes emit strongly from the benzothiazole/benzimidazole based locally
excited (LE) state. All the fluorescence AcO- sensing data were recorded in pure DMSO
solution because in the aqueous solution the water molecule converts the deprotonated
species to probe H2L1-S and H2L2-S and reinstates the PET from
benzothiazole/benzimidazole moiety to p-nitrobenzene (Figure S23, S24). The formation of
[L1-S]2- and [HL2-S]- was confirmed by ESI-MS analysis as the peak at m/z 397.07 (calcd. =
397.05) (Figure S25) and 312.08 (calcd. = 312.06) (Figure S26) corresponding to
[C15H12N4O2S2 -2H + CH3OH + Na+] ‒ and [C14H11N5O2S -H]‒ was clearly observed.
Scheme 3. Plausible mechanism for fluorometric acetate ion detection.
3.5 Live Cell Imaging and Cytotoxicity Study:
The biological usefulness of probe H2L1-S was assessed by screening its cytotoxicity in
HEK293T cell for 24 h. A concentration dependent cytotoxicity of probe H2L1-S was
observed (Figure S27). More than 70% cell viability was observed at 50 μM (working
concentration) and 58% viability at 100 μM (Maximum concentration) of probe H2L1-S after
23

24 h of treatment. This data implies the less toxicity of probe H2L1-S over a broad
concentration range.
On the basis of the cytotoxicity results, we investigated the ability of H2L1-S to detect
AcOˉ in biological systems by staining of living cells. As shown in Figure 13, incubation of
HEK293T cells with H2L1-S (50 μM) for 30 min failed to exhibit any fluorescence in the
green channel. Interestingly, when the probe H2L1-S loaded cells were subsequently exposed
to 50μM AcOˉ, a bluish green fluorescence was displayed. These results indicate that probe
H2L1-S has the potential to detect AcOˉ levels in living cells.
Figure 13. Bright field, fluorescence, and overlay images of (a-c) HEK293T cells treated
with only H2L1-S (50 μM), (d−f) HEK293T cells treated with H2L1-S (50μM) and AcOˉ (100
equiv).
4. Conclusions
We have developed two new thio-urea based bifunctional probes, H2L1-S and H2L2-S and
characterized them by several spectroscopies as well as single crystal X-ray diffraction. We
have shown that both the probes selectively detected Hg2+ ion through the colour change
from yellow to colourless associated with a notable blue shift through the formation of Hg2+
24

ion-promoted desulfurized product. This transformation was confirmed by ESI–MS, and also
supported by TDDFT calculation. Most interestingly, the desulfurized product H2L1-O was
separated and characterized by single crystal X-ray diffraction. The probes can quantify Hg2+
at the micromolar level even in the presence of other cations which is demonstrated by
interference study. In the fluorescence channel, the H2L1-S and H2L2-S acted as turn-on
fluorescent probe for AcO- ion in DMSO solution through the inhibition of PET. The
calculated detection limits suggested that the probes can detect AcO- in sub-micromolar level.
The fluorescence time course indicated that the deprotonation reaction of AcO- with the
probes was completed within ∼1.5 min. The cytotoxicity and live cell imaging study using
HEK293T cells with the probe revealed that the probe has low cytotoxicity, good cell
membrane permeability and is capable for exogenous AcO- imaging.
Acknowledgements
We thank DST SERB (SB/FT/CS/115/2012) for the financial assistance and
Sophisticated Analytical and Instrumentation Facility (SAIF), North Eastern Hill
University for NMR and MS data. S.P., B.S and S.K.S. thank North Eastern Hill
University, CSIR (SRF) and UGC respectively for their research fellowship. Mr. D.
Koner, Dept. of Zoology, NEHU is gratefully acknowledged for his help in cell
imaging study.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/.
25

References
1. K. Chen, Q. Shu, M. Schmittel, Design Strategies for Lab-on-a molecule Probes and
Orthogonal Sensing, Chem. Soc. Rev. 44 (2015) 136−160.
2. S. K. Sheet, B. Sen, R. Thounaojam, K. Aguan, S. Khatua, Highly Selective Light-up
Al3+ Sensing by a Coumarin Based Schiff Base Probe: Subsequent Phosphate
Sensing DNA Binding and Live Cell Imaging. J. Photochem. Photobiol. A: Chem
332, (2017), 101−111.
3. R. Singh, G. Das, Fluorogenic detection of Hg2+ and Ag+ ions via two
mechanistically discrete signal genres: A paradigm of differentially responsive metal
ion sensing, Sensors and Actuators B. 258 (2018) 478–483.
4. L. Yuan, W. Lin, Y. Xie, B. Chen, S. Zhu, Single Fluorescent Probe Responds to
H2O2, NO, and H2O2/NO with Three Different Sets of Fluorescence Signals, J. Am.
Chem. Soc. 134 (2012) 1305−1315
5. P.G. Sutariya, N. R. Modi, A. Pandya, B. K. Joshi, K. V. Joshi, S. K. Menon, An ICT
based ‘‘turn on/off’’ quinoline armed calix [4] arene fluoroionophore: its sensing
efficiency towards fluoride from waste water and Zn2+ from blood serum, Analyst,
137 (2012) 5491–5494.
6. Y. Peng, Y.M. Dong, M. Dong, Y.W. Wang, A Selective, Sensitive, Colorimetric,
and Fluorescence Probe for Relay Recognition of Fluoride and Cu(II) Ions with
“Off−On−Off” Switching in Ethanol−Water Solution, J. Org. Chem. 77 (2012)
9072–9080.
7. L. N. Neupane, J.-Y. Park, J. H. Park, K.-H. Lee, Turn-On Fluorescent Chemosensor
Based on an Amino Acid for Pb(II) and Hg(II) Ions in Aqueous Solutions and Role
of Tryptophan for Sensing. Org. Lett. 15 (2013) 254−257.
26

8. S. K. Sheet, B. Sen, R. Thounaojam, K. Aguan, S. Khatua, Ruthenium (II) Complex-
Based Luminescent Bifunctional Probe for Ag+ and Phosphate Ions: Ag+Assisted
Detection and Imaging of rRNA, Inorg. Chem. 56 (2017) 1249−1263.
9. S. Sakunkaewkasem, A. Petdum, W. Panchan, J. Sirirak, A. Charoenpanich, T.
Sooksimuang, N. Wanichacheva, Dual-Analyte Fluorescent Sensor Based on
[5]Helicene Derivative with Super Large Stokes Shift for the Selective
Determinations of Cu2+ or Zn2+ in Buffer Solutions and Its Application in a Living
Cell, ACS Sens. 3 (2018) 1016−1023.
10. M, Schmittel, H.W. Lin, Quadruple-Channel Sensing: A Molecular Sensor with a
Single Type of Receptor Site for Selective and Quantitative Multi-Ion Analysis,
Angew. Chem. Int. Ed. 46 (2007) 893 –896.
11. K. Chen, M. Schmittel, An iridium (III)-based lab-on-a-molecule for cysteine/
homocysteine and tryptophan using triple-channel interrogation, Analyst. 138 (2013)
6742−6745.
12. M. Wang, F. Yan, Y, Zou, L. Chen, N. Yang, X. Zhou, Recognition of Cu2+ and Hg2+
in physiological conditions by a new rhodamine based dual channel fluorescent
probe, Sens. Actuators, B. 192 (2014) 512– 521.
13. Y. Li, J. Wu, X. Jin, J. Wang, S. Han, W. Wu, J. Xu, W. Liu, X. Yao, Y. Tang, A
bimodal multianalyte simple molecule chemosensor for Mg2+, Zn2+, and Co2+, Dalton
Trans., 43 (2014) 1881–1887.
14. S.M. Hwang, C. Kim, Fluorescent detection of Zn2+ and Cu2+ by a phenanthrene-
based multifunctional chemosensor that acts as a basic pH indicator, Inorg. Chim.
Acta, 482 (2018) 375–383.
27

15. B. X. Shen, Y. Qian, Building Rhodamine-BODIPY fluorescent platform using Click
reaction: Naked-eye visible and multi-channel chemodosimeter for detection of Fe3+
and Hg2+, Sens. Actuators, B. 260 (2018) 666–675.
16. A. K. Mahapatra, R. Maji, K. Maiti, S. S. Adhikari, C. D. Mukhopadhyay, D.
Mandal, Ratiometric sensing of fluoride and acetate anions based on a BODIPY-
azaindole platform and its application to living cell imaging, Analyst, 139 (2014)
309.
17. S. Samanta, P. Dey, A. Ramesh, G. Das, A Solo Fluorogenic Probe for the Real-Time
2-
2-
-
Sensing of SO3 and SO4 /HSO4 in Aqueous Medium and Live Cells by Distinct
Turn-On Emission Signals, Chem. Commun., 52 (2016) 10381.
18. S. K. Patra, S. K. Sheet, B. Sen, K. Aguan, D. R. Roy, S. Khatua, Highly Sensitive
Bifunctional Probe for Colorimetric Cyanide and Fluorometric H2S Detection and
Bioimaging: Spontaneous Resolution, Aggregation, and Multicolor Fluorescence of
Bisulfide Adduct. J. Org. Chem. 82 (2017) 10234−10246.
19. T. ꢀbalos, D. Jimꢁnez, M. Moragues, S. Royo, R. Martıꢂ ez-Maꢂ ̃ez, F. Sancenoꢂ, J.
Soto, A. M. Costero, M. Parra, S. Gil, Multi-channel receptors based on thiopyrylium
functionalised with macrocyclic receptors for the recognition of transition metal
cations and anions, Dalton Trans. 39 (2010) 3449−3459.
20. J. Y. Yun, J. B. Chae, M. Kim, M. H. Lim, C. Kim, A multiple target chemosensor
for the sequential fluorescence detection of Zn2+ and S2− and the colorimetric
detection of Fe3+/2+ in aqueous media and living cells, Photochem. Photobiol. Sci.,
(2019) DOI. 10.1039/C8PP00408K.
21. E. M. Nolan, S. J. Lippard, Tools and Tactics for the Optical Detection of Mercuric
Ion, Chem. Rev. 108 (2008) 3443–3480.
28

22. H Ozay, R Kagit, M Yildirim, S Yeşilot, O Ozay, Novel hexapodal triazole linked to
a cyclophosphazene core rhodamine-based chemosensor for selective determination
of Hg2+ ions, J Fluoresc 24 (2014) 1593–1601.
23. P. Samanta, A. V. Desai, S. Sharma, P. Chandra, S.K. Ghosh, Selective Recognition
of Hg(II) ion in Water by a Functionalized Metal-Organic Framework (MOF) Based
Chemodosimeter, Inorg. Chem. 57 (2018) 2360-2364.
24. S. Ando, K. Koide, Development and Applications of Fluorogenic Probes for
Mercury (II) Based on Vinyl Ether Oxymercuration, J. Am. Chem. Soc. 133 (2011)
2556–2566.
25. G. Chen, Z. Guo, G. Zeng, L. Tang, Fluorescent and colorimetric sensors for
environmental mercury detection, Analyst. 140 (2015) 5400–5443.
26. S. Manna, P. Karmakar, K. Maiti, S. S. Ali, D. Mandal, A. K. Mahapatra, A reactive
primary fluorescence switch-on sensor for Hg2+ and the generated fluorophore as
secondary recognition receptor toward Cu2+ in aqueous acetonitrile solution, J.
Photochem. Photobiol. A: Chem, 343 (2017) 7–16.
27. S. Khatua, M. Schmittel, A Single Molecular Light-up Sensor for Quantification of
Hg2+ and Ag+ in Aqueous Medium: High Selectivity toward Hg2+ over Ag+ in a
Mixture, Org. Lett. 15 (2013) 4422–4425.
28. K.G. Harshaa, E. Appalanaidu, N. R. Chereddy, T.R. Baggi, V. J. Rao, Pyrene
tethered imidazole derivative for the qualitative and quantitative detection of mercury
present in various matrices, Sens. Actuators, B 256 (2018) 528–534.
29. S. O. Tümay, A. Uslu, H. A. Alidağı, H. H. Kazan, C. Bayraktar, T. Yolaçan, M.
Durmuş, S. Yeşilot, A systematic series of fluorescence chemosensors with multiple
binding sites for Hg(II) based on pyrenyl-functionalized cyclotriphosphazenes and
their application in live cell imaging. New J. Chem., 42 (2018) 14219-14228
29