The turn-off fluorescent sensors based on thioether-linked bisbenzamide for Fe3+ and Hg2+

Article abstract of DOI:10.1016/j.tet.2018.02.031

Some dibenzamide derivatives with a thioether linker were designed, synthesized and characterized. The specific responses to Hg²⁺ and Fe³⁺ were investigated by fluorescence. According to fluorescence titration, the Job plot, ¹H NMR, and ESI-mass analysis, the derivative with mono-hydroxyl substituent (1b) on the aromatic ring has high selectivity for Fe³⁺ ion with the formation of 1:1 1b-Fe³⁺ complexes. The specificity of 1c for Hg²⁺ could be switched by swapping the substituent from hydroxyl to amino, and a 1:2 (1c-Hg²⁺) complex was formed. Along with the obtained results, density functional theory (DFT) and natural bond orbital (NBO) analyses, Time-dependent (TD) DFT and natural transition orbital (NTO) analyses were employed to explore the geometric structures, properties and possible mechanisms.

Full text of DOI:10.1016/j.tet.2018.02.031

Accepted Manuscript
3+
The turn-off fluorescent sensors based on thioether-linked bisbenzamide for Fe
2+
and Hg
Hehong Lv, Ziqi Ren, Heng Liu, Ganbing Zhang, Hanping He, Xiuhua Zhang, Shengfu
Wang
PII:
S0040-4020(18)30165-0
10.1016/j.tet.2018.02.031
TET 29302
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 1 December 2017
Revised Date: 4 February 2018
Accepted Date: 14 February 2018
Please cite this article as: Lv H, Ren Z, Liu H, Zhang G, He H, Zhang X, Wang S, The turn-off
3+
2+
fluorescent sensors based on thioether-linked bisbenzamide for Fe and Hg , Tetrahedron (2018),
doi: 10.1016/j.tet.2018.02.031.
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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
The turn-off fluorescent sensors based on thioether-linked
bisbenzamide for Fe3+ and Hg2+
Hehong Lv b, Ziqi Ren b, Heng Liu a b, Ganbing Zhang* a b, Hanping He*a b, Xiuhua
Zhang a b, Shengfu Wang a b
a, Hubei Collaborative Innovation Center for Advanced Organic Chemical Materials;
Hubei University, Youyi Road 368, Wuchang, Wuhan, Hubei 430062, (PR China).
b, Ministry of Education Key Laboratory for the Synthesis and Application of Organic
Functional Molecules, College of Chemistry and Chemical Engineering, Hubei
University, Youyi Road 368, Wuchang, Wuhan, Hubei 430062, (PR China)
* Corresponding Authors: hehanping@hubu.edu.cn; gbzhang@hubu.edu.cn
Tel: +86-027-50865319; fax: +86-027-88663043
Abstract
Some dibenzamide derivatives with a thioether linker were designed, synthesized and
characterized. The specific responses to Hg2+ and Fe3+ were investigated by
1
fluorescence. According to fluorescence titration, the Job plot, H NMR, and
ESI-mass analysis, the derivative with mono- hydroxyl substituent (1b) on the
aromatic ring has high selectivity for Fe3+ ion with the formation of 1:1 1b-Fe3+
complexes. The specificity of 1c for Hg2+ could be switched by swapping the
substituent from hydroxyl to amino, and a 1:2 (1c-Hg2+) complex was formed. Along
with the obtained results, density functional theory (DFT) and natural bond orbital
(NBO) analyses were employed to explore the geometric structures, properties and
possible mechanisms.
Keywords: Fluorescent sensor; Metal ion; Fe3+; Hg2+; Bisbenzamide;
1

ACCEPTED MANUSCRIPT
1. Introduction
Fluorescent sensors for metal ions have been widely applied not only in
environmental monitoring but also in biological studies.1-3 Especially in recent years,
selective and sensitive detection of heavy and transition metal ions has been received
considerable attention because these metals caused adverse health and environmental
problems.4-9 Particularly, Hg (II) was regarded as one of the most toxic metal ions.
Due to its accumulative and highly toxic character, mercury can cause serious health
problems like prenatal brain damage, cognitive and motion disorder, minamata
diseases, etc.10-12 Fe (III) is an essential trace element in fundamental physiological
processes, being indispensable for all living systems. It performs the oxygen-carrying
capacity of heme as well as acts as a cofactor in many enzymatic reactions.13,14
However, its deficiency or overload has a toxic effect on living organisms and causes
diseases such as anemia and hemochromatosis.15,16 Because of these environmental
and health problems of Hg2+ and Fe3+, designing highly selective and sensitive
chemosensors for these two metals has been still a challenge.
Up to now, significant progress was achieved in the selective detection of Hg2+
and Fe3+.17-27 Most of them were based on different organic molecular
systems/materials
that
included
azo-derivatives17,25,20
,
crown-ethers24,26
,
naphthalimide19, anthraquinone21, rhodamine23,18, benzothiazole27, BODIPY25,26 and
so on. Among these ion recognition units, crown ethers-based chemosensors were
widely used due to their advantageous characteristics: the ability to coordinate the
cations of alkaline metals, high selectivity and accessibility. So far, Crown ether
derivatives incorporating a fluorescent moiety have been attractive tools for optical
sensing of all kinds of ions, such as Hg2+, Fe3+, Ag+, Pb2+.28 In 2008, Zhu et al.
designed a novel dye containing dithia-dioxa-monoaza crown ether moiety that can
perform highly sensitive detection of Hg2+ ion in the NIR region.29 In 2014, Sui et al.
presented a new Fe3+-recognizing cryptand with high selectivity, sensitivity, and
reversibility toward Fe3+ detection.30 However, for most fluorescent sensors of Hg2+
and Fe3+, a common limitation is that they are rather complicated, delayed response to
2

ACCEPTED MANUSCRIPT
the ion, and that their analytical results are easily influenced by coexisting ions.9
Additionally, most of sensors only worked well in the organic medium due to their
poor aqueous solubility. Therefore, development of fluorescent sensors with more
sensitivity, reliability, and aqueous medium solubility is in high demand for the
detection of Fe3+ and Hg2+.
In addition, podands linked different fluorescein are also used to identify various
kinds of metal. Chang group previously reported a series of copper-responsive
indicators containing both BODIPY and the double ethanedithiol units.31,32 Nolan and
co-workers synthesized a fluorescent sensor based on polioxo ethylene chain, which
was used with alkaline earth cations was examined.33 Inspired by these concepts, we
have focused our interest on designing simple molecules which could serve as
receptors to recognize Fe3+ based on a fluorescence ‘on-off’ mechanism. Herein we
synthesized some ethanedithiol derivatives 1a-g (Scheme 1) with two substituted
benzamide units for detecting Fe3+ and Hg2+ ions in aqueous solution. The chelating
groups like carbonyl and carboxamide had high binding affinity to transition metal
ions in comparison with alkali and alkaline earth metal ions. Furthermore, the
ethanedithiol linker is flexible, which is helpful to chelate Hg2+ and Fe3+ ions.34 In this
work, the changes of substituent groups on the benzene ring ortho of 1 significantly
altered the fluorescence response toward the metal ions. To our surprise, compared to
other six sensors, 1b with hydroxyl substituent showed remarkable selective response
toward Fe3+. 1c with amine enhanced the selectivity for Hg2+ ions over Fe3+ ions,
when the hydroxyl is replaced with amino.
2. Results and discussion
2.1 Synthesis
The compounds 1a-1g featured with the thioether linker and dibenzamide units.
Their synthetic pathways were similar (as shown in Scheme 1). Compound 2 was
readily obtained by the reaction of cysteamine hydrochloride and1,2-dibromoethane
in ethanol with a satisfied yield of 71%. Compound 2 was then reacted with different
substituents of benzoic acid derivatives to give 1a,1b,1c,1d,1e,1f,1g, respectively. The
3

ACCEPTED MANUSCRIPT
1
13
chemical structures of compounds 2 and 1a-1g were characterized by H NMR, C
NMR and ESI-MS (Fig. S1–S24, Supplementary data).
Scheme 1 Synthesis route to compounds 1a-1g.
2.2 Spectral studies and specificity of 1a-1g to metal ions
At first, the binding of probes 1a-1g with metal ions was investigated by
fluorescence spectroscopic measurements in DMSO/H2O (5:95; v/v) mixed solution.
As shown in Fig. 1 and Fig. S25, the probes (10ꢀM) exhibited better fluorescence.
Particularly, the 10ꢀM free ligands 1b, 1c and 1d with electron-donating groups
exhibited higher fluorescence emission intensities on excitation at 297, 317 and
290nm, respectively. On the contrary, 1a, 1e, 1f and 1g with electron-accepting
groups showed lower fluorescence emission intensities. The interaction of all the
probes and various ions were investigated by fluorescence spectra. In the presence of
excess metal ions (100 ꢀM), it was found that the addition of Fe3+ into the receptors
solution resulted in a varying decrement of fluorescence for free receptors 1a-1g.
Especially for probe 1b, the fluorescence was almost quenched completely. The result
indicated the probe 1b exhibited an excellent selectivity for Fe3+. Similarly, probe 1c
had good selectivity for Hg2+ as shown in Fig. 1. Pleasantly, for other ions including
Na+, K+, Cu2+, Fe2+, Ba2+, Mg2+, Mn2+, Ca2+, Cr3+, Ni2+, Pb2+, Zn2+, Co2+ and Cd2+,
the receptors (1a-1g) showed no apparent fluorescence intensity alternation. In
addition, other +3 cations (such as Tm3+ꢀGd3+ꢀY3+ꢀSc3+ and La3+) didn’t interfere the
4

ACCEPTED MANUSCRIPT
fluorescent emission (Fig. S26).
Fig. 1.
Fluorescence emission spectra of 1b (A) and 1c (B) (10ꢀM) in the presence of 10.0 equiv. of different metal
ions (Na+,K+,Mn2+,Mg2+,Ca2+,Cr3+,Fe3+,Ni2+,Cu2+,Zn2+,Co2+,Ag+,Cd2+,Hg2+,Fe2+ and Pb2+) in DMSO/H2O (5:95,
v/v).
The detail fluorescence changes of the probes in the presence of Fe3+ and Hg2+
were analyzed as shown in Fig. 2. Here, the value of F/F0 was used to represent
fluorescence quenching degree, where F and F0 represent the fluorescence intensity in
the presence and absence of metal ions respectively. Although the fluorescence of
these probes was quenched some certain after addition of Fe3+, quenching of probe 1b
was most (F/F0 = 0.04791) compared to other six probes. When the interactions of
these probes and Hg2+ were analyzed, probe 1c with amine substituent exhibited most
binding to Hg2+ interestingly. The quenching was also studied using some different
excitation wavelengths (260, 280, 330, 350nm) again in a broad range. It was found
that the fluorescence was similarly quenched by the addition of Fe3+ or Hg2+ (as
shown in Fig. S27). The photos obviously showed the quenching (as shown in Fig.2B
and 2C). We didn’t find obvious color changes in 1b or 1c solution when metal ions
were added. The data of UV-vis spectrometry (Figure S28) also showed that no
significant changes were observed even in the presence of Fe3+ or Hg2+. A small
hyperchromicity was observed for 1b, the intensity of absorption peak was decreased
for 1c.
Further, the corresponding photoluminescence quantum yield of 1b and 1c was
investigated for the quenching. The fluorescence quantum yield of 1b decreased from
0.329 to 0.147 in the absence or presence of 10 eq. Fe3+. And the fluorescence
5

ACCEPTED MANUSCRIPT
quantum yield of 1c decreased from 0.951 to 0.29 in the absence or presence of 10 eq.
Hg2+. These results indicated that probe 1b with hydroxyl group had excess specificity
to Fe3+, and probe 1c with amine group showed very good specificity to Hg2+ over
other competitive metal ions. In addition, the fluorescence lifetime were 6.13 ns, 3.26
ns, 2.81 ns, and 3.53 ns for 1b, 1c, 1b–Fe3+ and 1c-Hg2+ complex, respectively. These
are very important for understand the emission nature of complexes.
Another issue was whether pH of test solution would affect the selectivity and
sensitivity of the sensors. The acid titration experiments were examined at a pH range
from 3.0 to 12.0 in DMSO-H2O (v/v = 5/95) solution. As displayed in Fig.S29, there
were no noticeable change in the fluorescence intensity for both free 1b and 1b–Fe3+
complex within the pH range 3.0–9.0. However, at basic pH (>9), a fluorescence
increase was found for 1b and 1b-Fe3+ complex. For free 1b, deprotonation of
phenolic-OH may break down hydrogen bonded chelation and could enhance
fluorescence emission. For 1b-Fe3+ complex, the high pH would cause precipitation
of ferric hydroxide. The fluorescence of 1c was stable over a wide range of pH values
(4.0-12.0). While the intensity of 1c-Hg2+complex gradually increased with increasing
pH value, the quenching was also obvious in wide range (pH: 4.0-10.0). Overall, both
1b and 1c showed a highly selective fluorescence “turn-off” response in a wild pH
range from 4.0 to 9.0.
Finally, it was found that the fluorescence signals became immediately weakened
and arrived a stable value within 1 min (As shown in Figure S30), following the
addition of 10 equiv. Fe3+ to 1b (10ꢀM) and 10 equiv. Hg2+ ion to 1c (10ꢀM). These
illustrated that the reactions of 1b with Fe3+ and 1c with Hg2+ ware completed rapidly.
6

ACCEPTED MANUSCRIPT
Fig. 2. (A) Selectivity of 1a, 1b, 1c, 1d, 1e, 1f and 1g fluorescence response for Fe3+ and Hg2+. F and F0 represent
the fluorescence intensity in the presence and absence of metal ions, respectively. (B) The sensor 1b under the UV
lamp (365 nm) in the presence of Fe3+. (C) The sensor 1c under the UV lamp (365 nm) in the presence of Hg2+.
2.3 Interference studies
The above spectrum studies indicated that Fe3+ and Hg2+ lead to selectively
fluorescence quenching of probe 1b and 1c, respectively. The interference must also
be considering due to that there were mixture ions normally in practical samples. Here
the interferences of other ions were further investigated, as shown in Fig. 3A. It was
found that the fluorescence intensity of 1b had no apparent changes in the presence of
other metal ions including Na+, K+, Mn2+, Mg2+, Ca2+, Cr3+, Ni2+, Cu2+, Zn2+, Co2+,
Ag+, Cd2+, Fe2+ and Pb2+. Importantly, the strong fluorescence quenching was
observed once there was Fe3+ in samples. Similarly, the fluorescence intensities of 1c
wasn’t almost affected except that Fe3+ resulted in some quenching. Existence of Hg2+
ions made the strong fluorescence quenching to be observed. The selectivity of 1c to
Hg2+ ions were almost not influenced by the presence of other competing metal ions
(Fig. 3B). Therefore, probes 1b and 1c was ability to be used as highly selective
sensors for detections of Fe3+ and Hg2+, respectively.
7

ACCEPTED MANUSCRIPT
Fig. 3. (A) Metal ion selectivity profiles of 1b(10ꢀM): gray bars, fluorescence of 1b in the absence and the
presence of 10 equiv. of other competition metal ions; pink bars, fluorescence of 1b in the presence of 10 equiv. of
various metal ions, followed by 10 equiv. of Fe3+. (B) metal ion selectivity profiles of 1c (10ꢀM): black bars,
fluorescence of 1c in the absence and the presence of 10 equiv. of other competition metal ions; red bars,
fluorescence of 1c in the presence of 10 equiv. of various metal ions, followed by 10 equiv. of Hg2+.
Besides, the interferences of various anions on Fe3+-triggered and Hg2+-triggered
fluorescence quenching efficiency were also investigated respectively. As shown Fig.
4A, no obvious interferences were observed in presence of different anions including
Br-, Cl-, F-, CO3 , HCO3 , HSO3 , NO2 , NO3 , S2O3 , SO3 , SO42- and CH3COO-. For
2-
-
-
-
-
2-
2-
Hg2+-triggered fluorescence of 1c, it was clearly shown that the quenching efficiency
-
−
2-
2-
2-
was not affected by some anions such as F-, NO2 , NO3 , S2O3 , SO3 , SO4 and
CH3COO- (in Fig. 4B). However, the Hg2+-triggered quenching efficiency were
-
2-
2-
affected by Br-, Cl-, HSO3 , HCO3 and CO3 . These were probably attributed to the
2-
form of the tetrahedral anion HgX4 between HgX2 (X = C1, Br, or I) with excess
-
2-
halide35,36 and the form of precipitation between Hg2+ and HSO3 , HCO32- and CO3 .
8

ACCEPTED MANUSCRIPT
Fig. 4. (A) Fluorescence intensity of 1b (10 µM) upon addition of Fe3+ (100 µM) in different anions (100 µM). (B)
Fluorescence intensity of 1c (10 µM) upon addition of Hg2+ (100 µM) in different anions (100 µM).
2.4 Fluorescence titration with metal ions
To further understand the binding behavior of 1b and 1c to Fe3+ and Hg2+
respectively, the fluorescence titrations were investigated in DMSO/H2O (v/v = 5/95)
solution. As shown in Fig. 5A, the fluorescence intensity decreased sharply with
increasing concentration of Fe3+ ion ranging from 0 to 1.5 equiv. The relationship of
fluorescence quenching and concentration ratio ([Fe3+]/[1b]) was gave, which showed
that the quenching tended to stable after the ratio of [Fe3+]/[1b] was more than one.
These results indicated that the 1:1 metal-ligand complex was probably formed
between the probe 1b and Fe3+ in DMSO/H2O (5:95, v/v) solution. Importantly, a
good linear relationship (y = -0.01044 + 1.13487x) observed between the fluorescence
quenching efficiency and ratio of [Fe3+]/[1b] in the range of 0–0.5 (Fig. 5B).
Moreover, the limit of detection (LOD) was calculated to be 0.3 ꢀM according to the
reported method,37 which was below the upper limit of Fe3+ level (about 5.357ꢀM) in
drinking water.38 Similarly, the fluorescence intensity decreased obviously with the
gradual titration of Hg2+ ion. The fluorescence quenching tended to be relative stable
when the ratio of [Hg2+]/[1c] was about 2, which indicated that the 2:1 metal-ligand
complex was probably formed between the probe 1c and Hg2+ in DMSO/H2O (5:95,
v/v) solution (Fig. 6A). A linear relationship (y = 0.00115 + 0.30578x) was observed
between the fluorescence quenching efficiency and ratio of [Hg2+]/[1c] in the range of
0-1.0 (Fig. 6B), and the detection limit was 0.5ꢀM. Compared with previous methods
(Table S1), probe 1b and 1c had the lower detection limit to Fe3+ and Hg2+,
9

ACCEPTED MANUSCRIPT
respectively.
Fig. 5. (A) Fluorescence spectra of 1b (10ꢀM) upon the addition of different amounts of Fe3+ (0–1.5equiv) in
DMSO/H2O (5:95, v/v) solution, excitation and emission wavelengths were 297 and 426 nm, respectively. (B) The
ratio of integrated fluorescence (F/F0) of 1b as a function of Fe3+concentration. Inset: the linear relationship of 1b
between F0-F/F0 and Fe3+ concentration. F0 and F represent the fluorescence intensities at 426 nm in the absence
and presence of Fe3+.
Fig. 6. (A) Fluorescence spectra of 1c (10ꢀM) upon the addition of different amounts of Hg2+ (0–8equiv) in
DMSO/H2O (5:95, v/v) solution, excitation and emission wavelengths were 317 and 419 nm, respectively. (B) The
ratio of integrated fluorescence (F/F0) of 1c as a function of Hg2+concentration. Inset: the linear relationship of 1c
between F/F0 and Hg2+ concentration. F0 and F represent the fluorescence intensities at 419 nm in the absence and
presence of Hg2+.
2.5 Binding Stoichiometry
Based on the above fluorescence titration, the complexes with 1:1 1b-Fe3+ and
1:2 1c-Hg2+ were probably formed respectively in DMSO/H2O (5:95, v/v) solution.
Here, a Job’s Plot and ESI-MS were further constructed to determine the
stoichiometry of binding. As shown Fig. 7, Job’s plot obtained from the fluorescence
10

ACCEPTED MANUSCRIPT
measurements showed 1:1 stoichiometric ratio for 1b-Fe3+ complex, and 1:2
stoichiometric ratio 1c-Hg2+ complex.
Fig. 7. Job’s plots for the complexation of 1b with Fe3+ (A) and 1c with Hg2+ (B) in DMSO-H2O (v/v = 5/95)
solution.
The data were further confirmed by EIS-MS analysis. (As shown in Fig.
S31-S34 in supporting information). EIS-MS of 1b-Fe3+ complex showed that a peak
at m/z 474.1 that be assignable to [1b+Fe3+ -2H]+ and a peak at m/z 551.7 that be
assignable to [1b+Fe3++Cl-+Na++H2O-H]+ (Fig. S31 and Fig. S32), which indicated
that the 1:1 complex was formed. EIS-MS of complex 1c-Hg2+ exhibited a peak at
m/z =891.9, which corresponded to [1c+2Hg2++2Cl-+2H]+. The data were consistent
with the above fluorescence titration and Job’s Plot.
According to the above fluorescence titration and Job’s plot, association
constants were studied based on Benesi-Hildebrand equation (1),39 where F0 and F are
the fluorescence intensities of the ligand in the absence and presence of the metal ion,
respectively. Fmax is the fluorescent intensity at a complete interaction concentration
of the metal ion. Ka is the association constant, and n is binding stoichiometry ratio
for the ligand and the metal ion. The association constant could (Ka) be determined by
plotting 1/(F0-F) against 1/[M]n. And the association constants of 1b-Fe3+ and 1c-Hg2+
complexes were determined as 2.80 × 104 M-1 and 3.27 × 1010 M-2 respectively,
assuming 1:1 stoichiometry for 1b-Fe3+ and 1:2 stoichiometry for 1c-Hg2+ (Fig. 8A
and Fig. 8B).
ꢀ
ꢀ
ꢀ
=
+
ꢊ
ꢌꢍꢎꢏ
ꢅꢆ×(ꢇꢂꢃꢁꢈꢆꢉ)× ꢋ
(1)
ꢄ
ꢁ ꢃꢁ
ꢂ
ꢇꢂꢃꢁꢈꢆꢉ
11

ACCEPTED MANUSCRIPT
Fig. 8. Benesi–Hildebrand plots assuming 1:1 stoichiometry from Fluorometric titration data of receptor 1b (10
ꢀM) with Fe3+ (A) and 1:2 stoichiometry for receptor 1c with Hg2+ (B).
2.6 The reversibility of fluorescence quenching
In order to observe whether the spectra of probes 1b and 1c could be regenerated
upon the addition of cation-chelating agents, the reversibility of Fe3+ -triggered and
Hg2+-triggered quenching was performed with EDTA titration method.40 As shown in
Fig. 9, the fluorescence of probes 1b and 1c were not affected by the presence of
EDTA. Fe3+-triggered fluorescence quenching at 426 nm was recovered to 60%
instantaneously after the addition of EDTA. Interestingly, further addition of 10 eq. of
Fe3+ to the mixture solution again resulted in almost completely fluorescence
quenching of probe 1b. These indicated that the fluorescence probe 1b detecting Fe3+
ion was reversible. The phenomenon was also observed for Hg2+-triggered
fluorescence system (Fig. 9B).
Fig. 9. (A) Reversible fluorescence spectral response of 1b to Fe3+. (B) Reversible fluorescence spectral response
of 1c to Hg2+.
2.7 Possible recognition pattern
1
To further elucidate the sensing mechanism, H NMR, FTIR spectrum and
1
computational studies were investigated. H NMR spectra of the probes 1b and 1c
12

ACCEPTED MANUSCRIPT
were recorded in the absence and presence of metal ion (Note: the same D2O was
1
added when H NMR of control sample was tested, because the solvent for the Fe3+
salt was heavy water). As shown in Fig. 10, the proton in phenolic-OH (H1) was at
down-shifted from δ 12.48 to δ 12.50 ppm, and the proton in the methylene of -NCH2-
(H5) shifted downfield from 8.94 ppm to 9.06 ppm upon the addition of 1.0
equivalent of Fe(NO3)3to the 1b solution. Additionally, the aryl protons (H2, H3, H4)
also moved slightly to downfield, and the chemical shifts of the resonance peaks
around δ 2.70–2.80 ppm (-SCH2CH2S-) shifted upfield by around 0.03 ppm. As
shown in the IR spectrum (Fig. S34), the phenolic-OH absorption bands exhibited a
significant blue-shift, blue-shift was also observed for the amide C–N from 1337 cm−1
to 1359 cm−1, and signals in the region of 1350–1400 cm−1 assigned to -SCH2 moved
to high frequency. These results indicated that the O atom in phenolic-OH, N atom
and S atom might be coordinated to Fe3+ ion.
Fig.10. Partial 1H NMR spectra of 1b (7.5 mM) in DMSO-d6: (a) free 1b; (b) 1b in the presence of Fe(NO3)3 (1.0
equiv.).
1
Fig. 11. Partial H NMR spectra of 1c (7.5 mM) in DMSO-d6: (a) free 1c; (b) 1c in the presence of HgCl2
(10equiv.).
13

ACCEPTED MANUSCRIPT
Upon addition of 10 equiv. of HgCl2 into 1c solution, all protons in 1c were
shifted to downfield (Fig. 11). The aryl protons (H1, H2, H3, H4) were shifted
obviously to downfield about 0.05 ppm. The aliphatic protons (H8) in the thioether
group (-SCH2CH2S-) displayed considerable downfield shifts from 2.77 ppm to 3.15
ppm, which probably due to the interaction of S with Hg2+. The other the aliphatic
protons H6 and H7 were shifted to downfield about 0.16, 0.23 ppm respectively. In
the IR spectrum (Fig. S35, S36), the characteristic absorption peak of –NH2 (3469
cm−1) on the benzene ring was changed in the presence of Hg2+. These indicated that
the N atom in amine and amide and S atom might be coordinated to Hg2+ ion.
According to 1H NMR, the Job plot, fluorescence titration and ESI-mass analysis,
we proposed possible structures of 1:1 complex of 1b and Fe3+ and 1:2 complex of 1c
and Hg2+. The computational study was conducted by using the density functional
theory (DFT) combing natural bond orbital (NBO) analyses, and time-dependent (TD)
DFT methods combining natural transition orbital (NTO) analyses in order to get a
deep insight into the mechanism of the ‘turn off’ system for sensor 1b and 1c.
375
2.
2
.
0
6
3
side view
1.958
top view
8
5
9
.
1
1b
L.Fe3+
1.012
1.850
1.525
1.840
2
3
9
.
2
1.842
1.846
1.527
1.013
1.019
1.014
1c
2+
1c.Hg2
Fig.12. B3LYP and SMD salvation model optimized geometric structures for 1b (1b=H2L), L.Fe3+, 1c and
2+
1c.Hg2 in aqueous solution, where the bond lengths are in the unit of Å. For clarity, both side views and top
14

ACCEPTED MANUSCRIPT
views for both 1b and L.Fe3+ are shown.
Table 1 Second Order Perturbation Theory Analysis of Fock Matrix in NBO Basis.
E(2) a)
E(j)-E(i)b)
a.u.
F(i,j)c)
a.u.
Donor NBO (i)
Acceptor NBO (j)
kcal/mol
L.Fe3+ (1b = H2L)
LP (1) N8
LP*(4) Fe51
LP*(7) Fe51
LP*(5) Fe51
LP*(5) Fe51
LP*(4) Fe51
LP*(7) Fe51
LP*(5) Fe51
LP*(6) Fe51
LP*(5) Fe51
LP*(6) Fe51
LP*(7) Fe51
LP*(7) Fe51
72.79
24.77
14.08
12.67
63.17
25.65
54.89
55.86
48.09
66.75
14.13
14.03
0.32
0.59
0.64
0.65
0.24
0.59
0.61
0.73
0.66
0.77
0.88
0.88
0.137
0.115
0.089
0.085
0.085
0.116
0.165
0.184
0.160
0.206
0.102
0.102
LP (1) N8
LP (1) S11
LP (1) S14
LP (1) N17
LP (1) N17
LP (2) O19
LP (2) O19
LP (2) O28
LP (3) O28
BD (1) N8-H33
BD (1) N17-H46
2+
1c.Hg2
LP (2) S 9
LP*(6) Hg26
LP*(6) Hg26
LP*(8) Hg26
LP*(1) Hg26
BD*(1) N29-H56
LP*(6) Hg27
LP*(6) Hg27
LP*(7) Hg27
BD*(1) N25-H54
20.33
21.33
1.02
0.20
0.26
0.83
0.77
0.60
0.20
0.26
0.77
0.60
0.059
0.068
0.026
0.012
0.028
0.059
0.067
0.011
0.028
LP (1) N29
LP (1) O30
BD (1) N29-H56
LP*(6) Hg 26
LP (2) S22
0.24
0.24
20.56
20.81
0.20
LP (1) N25
BD (1) N25-H54
LP*(6) Hg27
0.23
a) E(2) denotes second order perturbation energy of hyperconjugative interactions between donor and acceptor i and j NBO
orbitals; b) Energy difference between i and j NBO orbitals; c) The Fock matrix element i and j NBO orbitals
Available calculated results may support the energy and the charge transfer
model for explaining the mechanism of the fluorescence quenching. As shown in Fig.
12 and Table S2, which displayed the optimized geometric structures for 1b
(1b=H2L), L.Fe3+, 1c and 1c.Hg22+, the Fe3+ ion was chelated through six
coordination sites on L in 1b ( N atoms of two imino groups, O atoms of two phenol
hydroxyls, and two S atoms), and each Hg2+ may be coordinated to 1c through three
sites (N atom of amino group, O atom of carbonyl group, and S atoms), where the
lone pairs of the coordination atoms transferred to the metals. These binding
situations were confirmed by larger second-order perturbation energy E(2) (which
15

ACCEPTED MANUSCRIPT
means the energy of hyperconjugation interaction between donor NBO i and acceptor
NBO j) between lone pair NBOs of each relevant S, N and O atoms and the
unoccupied orbital of the metals than those between other sites in ligand and metals
(shown in Table 1). Although weaker interactions existed between metals and ligand
through other sites, the donation-back donation interaction mode was still clear. For
2+
example, the interaction energy E(2) between N-H bond of NH2 and Hg2+ in 1c.Hg2
was just ca 0.24 kcal/mol, the donation of electron in each N-H bonding orbital to the
virtual orbital of Hg2+ and back-donation of electron in Hg2+ to the anti-bonding
orbital of N-H resulted in the weakening of the N-H bonds, which was in accordance
2+
with the elongation of the bond length from 0.1014 nm in 1c to 0.1019 nm in 1c.Hg2
,
and the decrease of the vibrational frequency from 3536.54 cm-1 in 1c to 3512.26 cm-1
in 1c.Hg22+. Besides, optimized geometries and natural bond orbital (NBO) analyses
showed that the interaction between N atom of imino group and metal ion in L.Fe3+
was fairly strong but very small in 1c.Hg22+. In addition, as shown in Table S2, on
coordination of Fe3+ and Hg2+, both the HOMO-LUMO gaps of two complexes
L.Fe3+ and 1c.Hg22+ became lower as compared to free 1b and 1c, respectively. From
orbital interaction viewpoint, the smaller H-L gap was, the easier the electron was
excited. This can explain what the electron density distribution of 1b and 1c were
significantly influenced on complexation with Fe3+ and Hg2+, respectively, due to the
occurrence of the possible charge transfer processes between the ligands and the metal
cations. The charge transfer could be further confirmed by the calculated natural
electron density population. From Table S3 we could find the decrease of the electron
2+
densities in all H atoms of 1c.Hg2 from those of 1c, due to the charge transfer
1
through complexation. This was in good agreement with the above-mentioned H
NMR spectroscopy observations.
16

ACCEPTED MANUSCRIPT
Hole
Electron
44.87%
NTO111
NTO112
NTO113
S0
S1
1b(1b=H2L)
4.8977 eV, 253.15 nm
f=0.0175
44.68%
NTO110
55.43%
S0
S24
NTO117
NTO116
NTO118
NTO119
L.Fe3+
4.1699 eV, 297.33 nm
f=0.0256
41.72%
Fig. 13. The Natural transition orbital (NTO) pairs that mainly contribute to the intense singlet excitation states in
UV-vis absorption for 1b and its complex to Fe3+ in aqueous solution calculated by using TD-CAM-B3LYP and
SMD salvation model, where f is oscillator strength
Electron
Hole
48.7%
NTO111
NTO112
S0
S1
1c
4.5976 eV, 269.67 nm
f=0.2615
44.95%
NTO110
NTO113
S0
S1
2+
94.95%
4.1201 eV, 300.93 nm
f=0.3206
1c.Hg2
NTO121
NTO122
Fig. 14. The Natural transition orbital (NTO) pairs that mainly contribute to the intense singlet excitation states in
UV-vis absorption for 1c and its complex 1c. Hg22+ in aqueous solution calculated by using TD-CAM-B3LYP and
SMD salvation model, where f is oscillator strength
Furthermore, the energy and the charge transfer assumption could also be
confirmed by TDDFT calculated results and the natural transition orbital (NTO)
17

ACCEPTED MANUSCRIPT
distribution. From TD-CAM-B3LYP calculated results, each s0→s1 excitation for 1b,
1c and 1c.Hg22+, and s0→s24 excitation for L.Fe3+ could be assigned for the intense
absorption of the corresponding molecules, respectively. Especially 297.33 nm of
2+
s0→s24 excitation for L.Fe3+ and 300.93 nm of s0→s1 excitation for 1c.Hg2 agree
well with the experimental absorption of 298 and 313 nm (Fig. S28), respectively.
While combining the observed short fluorescence lifetime (this may rule out the
possibility of the intersystem crossing), the s1→s0 emission could be assigned for
experimental fluorescence emission for relevant molecules. As shown in Fig. 13, 14
and Fig. S37, where the optical excitations and emissions were mainly contributed by
the transitions from the occupied (hole) NTOs to the unoccupied (electron) NTOs.
Whatever s0→s1 and s0→s24 excitations or s1→s0 emission processes presented charge
transfer character to some extent. Individually, the excitations of the four molecules
could be depicted as follows. 1b had the s0→s1 excitation energy of 4.8977 eV
(253.15 nm) and the oscillator strength of 0.0175, which was mainly combined with
two transitions between hole-electron NTO pairs. Since two NTO pairs predominantly
localized on the central area of phenyl moiety, thus these transitions may be attributed
as local excitation (LE) and intraligand charge transfer (ILCT). And L.Fe3+ had the
s0→s24 excitation energy of 4.1699 eV (297.33 nm) and the oscillator strength of
0.0256. This excitation was also mainly attributed to the mixing of two transitions. In
one transition, the hole NTO spread over the whole ligand backbone moiety, but the
partner electron NTO predominantly localized on the partial area of ligand and the
metal, thus this transition may be attributed as the mixing of ILCT and ligand to metal
charge transfer (LMCT). From NTO distribution, another transition within this
excitation may be assigned as the mixing of ILCT and metal to ligand charge transfer
(MLCT). Then the s0→s1 excitation of 1c (4.5976 eV, 269.67 nm, f=0.1680) was
mainly combined with two NTO pairs transitions with ILCT and LE character.
2+
Moreover, the s0→s1 excitation of 1c.Hg2 (4.1201 eV, 300.93 nm, f=0.3206) was
mainly contributed with one NTO pairs transition with the mixing of ILCT and
LMCT obviously. Besides, the s1→s0 emission of 1c (3.525 eV, 351.98 nm, f=0.2589)
was predominantly LE and ILCT transition.
In a summary, on the one hand, it revealed that the present computational results
along with the experimental data were efficient enough for exploring the interaction
pattern. On the other hand, it confirmed that the fluorescence quenching in the
presence of Fe3+ and Hg2+ may be due to the intramolecular charge transfer.
18

ACCEPTED MANUSCRIPT
2.8 Practical application
To assess the practicability of 1b and 1c to real samples, they were used as the
fluorescent sensors for the determination of Fe3+ and Hg2+ in drinking water. The
detection results were displayed in Table 2. It was found the detected concentrations
were close to the added concentration of ions. The results exhibited satisfactory
recoveries and very low the relative standard deviation (RSD) values for both Fe3+ and
Hg2+. These results suggested that 1b and 1c could satisfactorily detect Fe3+ and Hg2+
in water samples, respectively.
Table 2 Determination of Fe3+ and Hg2+ in water samples
Sample
Added
(ꢀM)
0.100
0.200
0.300
0.200
0.400
0.800
Found
(ꢀM)
0.109
0.193
0.280
0.199
0.400
0.790
RSD.
(%, n=3)
2.2
0.71
1.0
0.22
2.2
0.70
Recovery
(%)
109
96.5
93.3
99.5
100
Fe3+
Hg2+
98.8
3. Conclusion
In summary, we have successfully designed and synthesized some thioether-linked
bisbenzamide derivatives and switched their selectivity for metal ions by introducing
different substituent groups into the benzene ring. The experiment data displayed the
o-hydroxyl-attached probe 1b exhibited selective ‘turn-off’ fluorescence response to
Fe3+, and the o-amino-attached probe 1c selectively recognized Hg2+ by fluorescence
quenching quickly. The binding stoichiometry was determined to be 1:1 for 1b-Fe3+,
1:2 for 1c-Hg2+ based on Job’s plot, fluorescent titration and ESI-MS. The detection
limits of 1b for Fe3+ and 1c for Hg2+ were calculated as 0.3 ꢀM and 0.5 ꢀM by
fluorescence titration, respectively. Moreover, combining SMD salvation model,
theoretical calculations with B3LYP and NBO analyses, TD-CAM-B3LYP and NTO
analyses, provided valuable confirmation to the experimental observation and rational
recognition pattern, which supported the intramolecular charge transfer model for the
mechanism of the fluorescence quenching. Both probes 1b and 1c were successfully
19

ACCEPTED MANUSCRIPT
applied to the determination of Fe3+ or Hg2+ in drinking water. All these facts
indicated that probe 1b and 1c could be served as simple, rapid, sensitive and
selective chemosensors for Fe3+ and Hg2+ recognition, respectively. The presence
sensing systems have great prospective in biomedical and environment detection.
4. Experimental
4.1 Reagents and materials
Unless stated otherwise, all analytical grad chemicals and solvents used in this
paper were purchased from commercial vendors. Dimethyl formamide (DMF) was
dried over 4 Å molecular sieves. Dichloromethane (DCM) was distilled from calcium
hydride. Others were used directly without further purification. Ultrapure water was
used to prepare stock solution (0.01 M) of NaNO3, KNO3, Cu(NO3)2, Fe(NO3)2,
Fe(NO3)3, MgNO3, Ca(NO3)3, CrNO3)3, Ni(NO3)2, Zn(NO3)2, Co(NO3)2, HgCl2,
Pb(NO3)2, Cd(NO3)2 and AgNO3. Stock solutions (0.01M) of anions in water were
prepared from NaCl, NaF, NaBr, NaNO3, NaNO2, NaHCO3, Na2CO3, Na2SO4,
Na2SO3, NaHSO3, Na2S2O3 and CH3COONa. The stock solution of sensors was
prepared by dissolving compounds 1a-g in dimethyl sulfoxide (DMSO) and was used
to prepare the DMSO aqueous solution.
4.2 Instrumentation
Melting points were determined with an X6 melting apparatus without correction.
IR spectra in KBr were recorded on a Germany Bruker corporation VECTOR22
1
Fourier transform infrared spectrometer. H NMR (600 MHz) and 13C NMR (150
MHz) spectra were measured on a Varian Unity INOVA 600 NMR magnetic
resonance spectrometer (TMS as internal standard). Mass spectrometry was recorded
with a Finnigan LCQ mass spectrometer and an Agilent 1200 LC/MSD mass
spectrometer. UV–Vis spectra were collected from a Shimadzu UV-2700
spectrophotometer at room temperature. The fluorescence spectra were obtained with
RF-5301(PC) S spectrometer with a 1cm standard quartz cell. Fluorescence lifetimes
were measured on Edinburgh FLS980 fluorescence spectrophotometer.
4.3 Synthesis and characterization of receptors
20

ACCEPTED MANUSCRIPT
The compound 2 was prepared according to the reported procedure.41,42 Sodium
ethylate (3.13g, 46.1mmol) was dissolved in dry ethanol (40mL) and then cooled to
15°C. The cool solution was added to 2-aminoethanethiol hydrochloride (2.61g, 23.0
mmol). After the mixture was stirred for 15 min under nitrogen atmosphere, 1,
2-dibromoethane (1.0mL, 11.5mmol) was added and stirred for 4 h at 40°C.The
mixture was filtered and evaporated under reduced pressure. The yellowish mass was
dissolved in sodium hydroxide solution (5.0 g in 15mL water), and the resulting
solution was kept in a refrigerator overnight. The solution was extracted with
1
dichloromethane. The extract was evaporated to give 1.48 g (71%) of 2. H NMR
(600 MHz, CDCl3) δ 2.89 (t, J = 6.3 Hz, 2H), 2.74 (s, 2H), 2.66 (t, J = 6.3 Hz, 2H);
13C NMR (147 MHz, CD3OD): δ 41.53, 35.45, 32.71; LC-MS (ESI) m/z calculated
+
for [M+H]+ C6H16N2S2 181.0800. Found 181.0810.
Synthesis of receptors 1a-1g
The synthesis procedure ii was performed according to the literature method.43,44 To
an ice-cooled solution of the acid (2.5mmol) in dry CH2Cl2 (30mL), HOBt (0.40 g,
3mmol) and EDC (0.58 g, 3mmol) were added. The reaction mixture was stirred for
30 min, then compound 2 (0.18 g, 1mmol) in DCM (2mL) was added slowly. The
resulting mixture was stirred at RT for 18h, and was monitored by TLC. After
completion of the reaction, the mixture was extracted with CH2CCl2 (30mL), washed
with aqueous NaHCO3 (30mL) and H2O (30mL). The organic layer was dried over
Na2SO4 and removed by rotary evaporation. The extract was evaporated and the crude
product was purified by column chromatography (DCM/ Methanol, 100:1 to 10:1) to
afford pure product (1a; 1b; 1c; 1e). The receptors 1d, 1f, 1g were prepared by the
similar methodology as the above,45,46 but HOBt and EDC were replaced with HBTU
(1.42 g, 3.75mmol) and DIPEA (653ꢀL,3.8 mmol).
1a:0.25 g, yield 63.3%; Rf = 0.7 (CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C; 1H
NMR (600 MHz, DMSO-d6) δ12.48 (s, J = 6.24 Hz, 2H), 2.73 (s, 4H), 2.66 (t, J =
6.24 Hz, 4H); 13C NMR (150 MHz, CDCl3) : δ 167.54, 134.08, 131.38, 128.37,
+
126.79, 39.19, 31.33, 31.14; LC-MS (ESI) m/z calculated for [M+H]+ C20H24N2O2S2
389.1310. Found 389.1340.
21

ACCEPTED MANUSCRIPT
1b: 0.24 g, yield 62%; Rf = 0.6 (CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C;
1H NMR (600 MHz, DMSO-d6) δ 12.48 (s, 2H), 8.95 (t, J = 5.6 Hz, 2H), 7.82 (d, J =
7.9 Hz, 2H), 7.39 (t, J = 8.1 Hz, 2H), 6.93 – 6.85 (m, 4H), 3.48 (q, J = 6.6 Hz, 4H),
2.78 (s, 4H), 2.74 (t, J = 7.1 Hz, 4H); 13C NMR (150 MHz, DMSO-d6) : δ 169.47,
160.56, 134.34, 128.43, 119.25, 118.03, 115.89, 40.61, 40.47, 40.32, 40.18, 40.04,
+
31.73, 30.87; LC-MS (ESI) m/z calculated for [M+H]+ C20H24N2O4S2 421.1201.
Found 421.1208.
1c: 0.25 g, yield 58%; Rf = 0.72 (CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C;
1H NMR (600 MHz, DMSO-d6) δ 8.35 (t, J = 5.8 Hz, 2H), 7.45 (d, J = 8.1 Hz, 2H),
7.13 (t, J = 7.6 Hz, 2H), 6.68 (d, J = 8.2 Hz, 2H), 6.50 (t, J = 7.5 Hz, 2H), 6.41 (s, 5H),
3.38 (q, J = 6.6 Hz, 4H), 2.77 (s, 4H), 2.70 (t, J = 7.2 Hz, 4H);13C NMR (150 MHz,
DMSO-d6) : δ 168.94, 168.83, 149.62, 149.59, 131.71, 131.66, 128.00, 127.95, 116.34,
114.52, 114.42, 39.93, 38.95, 38.82, 38.38, 37.20, 31.04, 30.42. LC-MS (ESI) m/z
calculated for [M+H]+ C20H26N4O2S2 419.1526. Found 419.1530.
+
1d: 0.25 g, yield 63%; Rf = 0.8(CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C; 1H
NMR (600 MHz, CDCl3) δ 8.33 (s, 2H), 8.19 (d, J = 7.6 Hz, 2H), 7.45 (d, J = 8.4 Hz,
2H), 7.07 (dt, J = 11.0, 5.6 Hz, 2H), 6.97 (dd, J = 8.2, 3.7 Hz, 2H), 3.98 (d, J = 3.5 Hz,
6H), 3.83-3.66 (m, 4H), 2.99-2.77 (m, 8H). 13C NMR (150 MHz, CDCl3) : δ 165.37,
157.53 , 132.87, 132.16 , 121.23 , 111.31 , 55.99 , 39.05 , 38.33 , 38.00 ,
+
31.82.LC-MS (ESI) m/z calculated for [M+H]+ C22H28N2O4S2 449.1550. Found
449.1549.
1e: 0.25 g, yield 63%; Rf = 0.5(CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C; 1H
NMR (584 MHz, DMSO-d6) δ 8.45 (t, J = 5.9 Hz, 2H), 7.75 (d, J = 7.6 Hz, 2H), 7.55
(t, J = 7.5 Hz, 2H), 7.50 (t, J = 7.6 Hz, 3H), 7.40 (d, J = 7.4 Hz, 2H), 3.36 (d, J = 7.0
Hz, 4H), 2.78 (s, 4H), 2.70 (t, J = 7.4 Hz, 4H);13C NMR (150MHz, DMSO-d6) : δ
169.22, 168.57, 138.97, 131.81, 131.40, 129.84, 129.82, 128.18, 40.59, 31.73, 30.68.
+
LC-MS (ESI) m/z calculated for [M+H]+ C22H24N2O6S2 477.1155. Found 477.1160.
1f: 0.25 g, yield 63%; Rf = 0.65 (CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C;
1H NMR (600 MHz, DMSO-d6) δ 8.58 (d, J = 5.9 Hz, 2H), 7.65 (dd, J = 8.0, 3.9 Hz,
2H), 7.39 (ddd, J = 25.7, 13.4, 6.6 Hz, 4H), 3.40 (q, J = 6.4 Hz, 4H), 2.80 (d, J = 4.3
22

ACCEPTED MANUSCRIPT
Hz, 4H), 2.72 (t, J = 6.8 Hz,4H).13C NMR (150 MHz, DMSO-d6): δ 167.94, 167.86,
139.64, 139.60, 134.34, 133.33, 131.50, 129.38, 128.33, 128.17, 119.54, 40.59, 31.72,
+
30.93. LC-MS (ESI) m/z calculated for [M+H]+ C20H22Br2N2O2S2 546.9540. Found
546.9542.
1g: 0.25 g, yield 63 %; Rf = 0.55 (CH2Cl2/ Methanol = 10 : 1) ; m.p. 143-144°C; 1H
NMR (600 MHz, DMSO-d6) δ 8.44 (s, 2H),7.63 (t, J = 7.7 Hz, 2H), 7.53 (q, J = 7.2
Hz,2H), 7.28 (q, J = 8.2, 7.2 Hz, 2H), 3.43 (q, J = 6.8 Hz, 4H), 2.78 (s, 4H), 2.72 (t, J
= 7.3 Hz, 4H).13C NMR (150 MHz, CDCl3) : δ 163.28, 163.26, 161.24, 159.55,
133.16, 133.10, 131.68, 124.58, 124.56, 120.81, 120.72, 115.92, 115.75, 76.88, 76.66,
39.16, 38.62, 37.42, 31.59, 31.33. LC-MS (ESI) m/z calculated for [M+H]+
+
C20H22F2N2O2S2 425.1130. Found 425.1135.
4.4 Calculation of the relative fluorescence quantum yield
Fluorescence quantum yield was determined by using quinine sulphate (Φs =0.546 in
0.1 M H2SO4) as a fluorescence standard. The quantum yield was calculated using
the following equation:
Φx=Φs(AsFx/AxFs)(nx/ns)2
Where Φx is the fluorescence quantum yield, A is the absorbance at the excitation
wavelength, F is the integrated area under the corrected emission curve, and n is the
refractive index of the medium. Subscripts S and X refer to the standard and to the
unknown, respectively.
4.5 Computational methodology
Density functional theory (DFT) method with B3LYP functional47,48 had been
utilized to optimize geometric structures of molecules 1b (1b=H2L), 1c and their
2+
complexes L.Fe3+, and 1c.Hg2 in the lowest singlet spin state s0, then vibrational
analyses had been done by frequency calculations to verify that the geometries
obtained are minima or not and obtain vibrational spectra. To get more detailed
information on the chemical bonds and bonding interaction within all four molecules,
natural bonding orbital (NBO) calculations49,50 were carried out using B3LYP method
on the optimized geometries. Based on the optimized geometries, vertical electronic
excitation energies and absorption spectra were calculated with time-dependent (TD)
23

ACCEPTED MANUSCRIPT
DFT, TD-CAM-B3LYP method.51 The geometries for the first excited singlet states s1
of 1b and 1c were optimized with TD-CAM-B3LYP method to obtain their
fluorescence emission energies and spectra. In order to analyze the nature of
absorption and emission, natural transition orbital (NTO) analyses were performed
based on the TDDFT calculations and the calculated transition density matrices52. In
all calculations, the salvation effect was considered with SMD salvation model53 and
water media; the 6-31G (d, p) basis set for the atoms C, N, O, S and H, and relativistic
pseudo-potential LanL2dz basis set for Fe and Hg were selected. All calculations
were performed with Gaussian 09 program54.
Acknowledgments
This work was supported by the National Natural Science Foundation of China
(Grant 21575035), and Foreign Science and Technology Cooperation Fund of Hubei
province, China (2015BHE025).
Appendix A. Supplementary data
The following is the supplementary data related to this article:
Supplementary data associated with this article (1H, 13C NMR and MS spectra).
References
1.
Boonkitpatarakul, K.; Wang, J.; Niamnont, N.; Liu, B.; McDonald, L.; Pang, Y.; Sukwattanasinitt,
M. ACS Sensors 2016, 1, 144-150.
2.
3.
Han, X.; Wang, D.-E.; Chen, S.; Zhang, L.; Guo, Y.; Wang, J. Anal. Methods 2015, 7, 4231-4236.
Yang, M.; Meng, W.; Ding, Q.; Su, N.; Liu, X.; Zhang, M.; Yang, B. New J. Chem. 2015, 39,
4790-4795.
4.
5.
Ding, J.; Li, H.; Wang, C.; Yang, J.; Xie, Y.; Peng, Q.; Li, Q.; Li, Z. ACS Appl Mater Interfaces 2015,
7, 11369-76.
Kaushik, R.; Ghosh, A.; Singh, A.; Gupta, P.; Mittal, A.; Jose, D. A. ACS Sensors 2016, 1,
1265-1271.
6.
7.
Liu, T.; Wan, X.; Yao, Y. Sensors and Actuators B: Chemical 2018, 254, 1094-1100.
Wang, Z.; Yang, J.; Yang, Y.; Fang, H.; Xu, X.; Rui, J.; Su, F.; Xu, H.; Wang, S. RSC Adv. 2017, 7,
33263-33272.
8.
Wu, W.; Chen, A.; Tong, L.; Qing, Z.; Langone, K. P.; Bernier, W. E.; Jones, W. E., Jr. ACS Sens
2017.
9.
Wu, X.; Zhang, S.; Niu, Q.; Li, T. Tetrahedron Letters 2016, 57, 3407-3411.
Foster, K. L.; Stern, G. A.; Pazerniuk, M. A.; Hickie, B.; Walkusz, W.; Wang, F.; Macdonald, R. W.
Environ Sci Technol 2012, 46, 12952-9.
10.
11.
Huang, X.; Li, M.; Friedli, H. R.; Song, Y.; Chang, D.; Zhu, L. Environ Sci Technol 2011, 45,
9442-8.
24

ACCEPTED MANUSCRIPT
12.
Lambert, K. F.; Evers, D. C.; Warner, K. A.; King, S. L.; Selin, N. E. Environ Res 2012, 119,
132-42.
13.
14.
15.
Carter, K. P.; Young, A. M.; Palmer, A. E. Chem Rev 2014, 114, 4564-601.
Chen, X.; Pradhan, T.; Wang, F.; Kim, J. S.; Yoon, J. Chem Rev 2012, 112, 1910-56.
Comin-Colet, J.; Lainscak, M.; Dickstein, K.; Filippatos, G. S.; Johnson, P.; Luscher, T. F.; Mori, C.;
Willenheimer, R.; Ponikowski, P.; Anker, S. D. Eur Heart J 2013, 34, 30-8.
Zheng, M.; Tan, H.; Xie, Z.; Zhang, L.; Jing, X.; Sun, Z. ACS Appl Mater Interfaces 2013, 5,
1078-83.
16.
17.
18.
Bozkurt, E.; Gul, H. I. Sensors and Actuators B: Chemical 2017.
Liu, Y.; Shen, R.; Ru, J.; Yao, X.; Yang, Y.; Liu, H.; Tang, X.; Bai, D.; Zhang, G.; Liu, W. RSC Adv.
2016, 6, 111754-111759.
19.
20.
21.
Shu, W.; Yan, L.; Liu, J.; Wang, Z.; Zhang, S.; Tang, C.; Liu, C.; Zhu, B.; Du, B. Industrial &
Engineering Chemistry Research 2015, 54, 8056-8062.
Yang, X.; Liu, X.; Li, Y.; Wu, F.; Mao, J.; Yuan, Y.; Cui, Y.; Sun, G.; Zhang, G. Biosens Bioelectron
2016, 80, 288-93.
Ahmad Wani, M.; Thakur, N.; Pandey, M. D.; Pandey, R. New Journal of Chemistry 2017, 41,
10000-10008.
22.
23.
Hirayama, T.; Nagasawa, H. Journal of Clinical Biochemistry and Nutrition 2017, 60, 39-48.
Jiang, C.; Wang, M.; Wang, Y.; Tang, X.; Zhang, Y.; Zhang, H.; Ma, L.; Wang, J. Tetrahedron
Letters 2017, 58, 2560-2565.
24.
Ju, H.; Chang, D. J.; Kim, S.; Ryu, H.; Lee, E.; Park, I. H.; Jung, J. H.; Ikeda, M.; Habata, Y.; Lee, S.
S. Inorg Chem 2016, 55, 7448-56.
25.
26.
27.
28.
29.
Maity, A.; Sil, A.; Nad, S.; Patra, S. K. Sensors and Actuators B: Chemical 2018, 255, 299-308.
Wu, B.; Xu, L.; Wang, S.; Wang, Y.; Zhang, W. Polym. Chem. 2015, 6, 4279-4289.
Xu, D.; Tang, L.; Tian, M.; He, P.; Yan, X. Tetrahedron Letters 2017, 58, 3654-3657.
Li, J.; Yim, D.; Jang, W. D.; Yoon, J. Chem Soc Rev 2017, 46, 2437-2458.
Zhu, M.; Yuan, M.; Liu, X.; Xu, J.; Lv, J.; Huang, C.; Liu, H.; Li, Y.; Wang, S.; Zhu, D. Organic
Letters 2008, 10, 1481-1484.
30.
31.
Sui, B.; Tang, S.; Liu, T.; Kim, B.; Belfield, K. D. ACS Appl Mater Interfaces 2014, 6, 18408-12.
Dodani, S. C.; Leary, S. C.; Cobine, P. A.; Winge, D. R.; Chang, C. J. J Am Chem Soc 2011, 133,
8606-16.
32.
Domaille, D. W.; Zeng, L.; Chang, C. J. Journal of the American Chemical Society 2010, 132,
1194-1195.
33.
34.
35.
36.
Xu, C.; Liu, W.; Qin, W. J Phys Chem A 2011, 115, 4288-98.
Zhao, B.; Xu, Y.; Fang, Y.; Wang, L.; Deng, Q. Tetrahedron Letters 2015, 56, 2460-2465.
Du, J.; Fan, J.; Peng, X.; Li, H.; Wang, J.; Sun, S. J Fluoresc 2008, 18, 919-24.
Lee, S. J.; Jung, J. H.; Seo, J.; Yoon, I.; Park, K.-M.; Lindoy, L. F.; Lee, S. S. Organic Letters 2006,
8, 1641-1643.
37.
38.
39.
Liao, G.; Zheng, C.; Xue, D.; Fan, C.; Liu, G.; Pu, S. RSC Adv. 2016, 6, 34748-34753.
Bozkurt, E.; Arık, M.; Onganer, Y. Sensors and Actuators B: Chemical 2015, 221, 136-147.
Erdemir, S.; Yuksekogul, M.; Karakurt, S.; Kocyigit, O. Sensors and Actuators B: Chemical 2017,
241, 230-238.
40.
41.
Roy, N.; Nath, S.; Dutta, A.; Mondal, P.; Paul, P. C.; Singh, T. S. RSC Adv. 2016, 6, 63837-63847.
Choudhury, S. B.; Ray, D.; Chakravorty, A. Inorganic Chemistry 1991, 30, 4354-4360.
25

ACCEPTED MANUSCRIPT
42.
43.
Nutthanara, P.; Ngeontae, W.; Imyim, A.; Kreethadumrongdat, T. Journal of Applied Polymer
Science 2009, n/a-n/a.
Musiol, R.; Tabak, D.; Niedbala, H.; Podeszwa, B.; Jampilek, J.; Kralova, K.; Dohnal, J.; Finster, J.;
Mencel, A.; Polanski, J. Bioorg Med Chem 2008, 16, 4490-9.
44.
45.
Taj, R.; Sorensen, J. L. Tetrahedron Letters 2015, 56, 7108-7111.
Bajaj, M.; Mamidyala, S. K.; Zuegg, J.; Begg, S. L.; Ween, M. P.; Luo, Z.; Huang, J. X.; McEwan,
A. G.; Kobe, B.; Paton, J. C.; McDevitt, C. A.; Cooper, M. A. ACS Chem Biol 2015, 10, 1511-20.
Peng, X.; He, H.; Xia, J.; Lou, Z.; Chang, G.; Zhang, X.; Wang, S. Tetrahedron Letters 2014, 55,
3541-3544.
46.
47.
48.
49.
Becke, A. D. Physical Review A 1988, 38, 3098-3100.
Lee, C.; Yang, W.; Parr, R. G. Physical Review B 1988, 37, 785-789.
Glendening, E. D.; Landis, C. R.; Weinhold, F. Wiley Interdisciplinary Reviews: Computational
Molecular Science 2012, 2, 1-42.
50.
51.
52.
53.
54
Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chemical Reviews 1988, 88, 899-926.
Yanai, T.; Tew, D. P.; Handy, N. C. Chemical Physics Letters 2004, 393, 51-57.
Martin, R. L. J. Chem. Phys., 2003, 118, 4775-4777.
Marenich, A. V. ; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B, 2009,113, 6378-
Gaussian 09, Revision E.01, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.;
Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Vreven, T. ; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J. ; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam,
J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J. and Fox, D. J. Gaussian, Inc.,
Wallingford CT, 2013.
26