Spectroscopy and a theoretical study of colorimetric sensing of fluoride ions by salicylidene based Schiff base derivatives

Article abstract of DOI:10.1016/j.molstruc.2021.131132

Three colorimetric anion sensors based on salicylidene Schiff bases (salicylaldehyde-o-aminophenol [SA1], 3,5-dimethyl-salicylaldehyde-o-aminophenol [SA2], and 3,5-dichloro-salicylaldehyde-o-aminophenol [SA3] were comprehensively studied based on experimental methods combined with theoretical calculations. All derivatives showed high sensitivity for colorimetric detection of fluoride ions (F^?) with a binding stoichiometry of 1:1 in acetonitrile solutions. The color of the sensor solutions visibly changed from light yellow to orange red in the presence of F^?. From the experimental results, SA1 and SA2 showed higher potential as F^? sensors than SA3 due to their higher selectivity with a limit of detection (LOD) as low as 8 × 10^?5, 21 × 10^?5 and 7 × 10^?5 M, respectively. The optimized structure and electronic transitions were confirmed by DFT and TDDFT studies. In this study, F^? detection mechanism is proposed based on experimental and the density functional theory (DFT) calculation.

Full text of DOI:10.1016/j.molstruc.2021.131132

Journal of Molecular Structure 1245 (2021) 131132
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstr
Spectroscopy and a theoretical study of colorimetric sensing of
fluoride ions by salicylidene based Schiff base derivatives
Buchitar Nakwanicha,b, Amonchat Koonwonga,b, Anwaraporn Suramitrc, Panida Prompinitd,
Rungtiva.P. Poo-arporne, Supa Hannongbuaa,b, Songwut Suramitra,b,∗
a Department of Chemistry, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand
b Center for Advanced Studies in Nanotechnology for Chemical, Food and Agricultural Industries, KU Institute for Advanced Studies, Kasetsart University,
Bangkok 10900, Thailand
c Faculty of Science at Si Racha, Kasetsart University Si Racha Campus, Chonburi 20230, Thailand
d National Nanotechnology Center (NANOTEC), National Science and Technology Development Agency, Klong Luang, Pathumthani 12120, Thailand
e Bsiological Engineering Program, Faculty of Engineering, King Mongkut’s University of Technology Thonburi, Bangkok 10140, Thailand
a r t i c l e i n f o
a b s t r a c t
Article history:
Three colorimetric anion sensors based on salicylidene Schiff bases (salicylaldehyde-o-aminophenol [SA1],
3,5-dimethyl-salicylaldehyde-o-aminophenol [SA2], and 3,5-dichloro-salicylaldehyde-o-aminophenol
[SA3] were comprehensively studied based on experimental methods combined with theoretical calcu-
lations. All derivatives showed high sensitivity for colorimetric detection of fluoride ions (F−) with a
binding stoichiometry of 1:1 in acetonitrile solutions. The color of the sensor solutions visibly changed
from light yellow to orange red in the presence of F−. From the experimental results, SA1 and SA2
showed higher potential as F− sensors than SA3 due to their higher selectivity with a limit of detection
(LOD) as low as 8 × 10−5, 21 × 10−5 and 7 × 10−5 M, respectively. The optimized structure and electronic
transitions were confirmed by DFT and TDDFT studies. In this study, F− detection mechanism is proposed
based on experimental and the density functional theory (DFT) calculation.
Received 29 March 2021
Revised 12 July 2021
Accepted 15 July 2021
Available online 20 July 2021
Keywords:
Salicylidene Schiff base
Colorimetric sensing
Photophysical property
Density functional theory (DFT)
© 2021 Elsevier B.V. All rights reserved.
1. Introduction
tions [14], polymer-based interactions [15], quantum dots and gold
nanoparticles-based interactions [16], and mesoporous silica or sil-
Fluoride is a mineral present in water and soil and plays a
crucial role in biological and chemical applications [1-3]. Fluoride
is commonly added to toothpaste and drinking water at limited
concentrations recommended by the World Health Organization
(WHO) (≤ 2 mg/L) to reduce dental cavities [4, 5] and for clin-
ical treatment of osteoporosis [6]. High fluoride levels can cause
skeletal and dental fluorosis, resulting in bleaching or a brownish
discoloration of teeth and bones [7-8]. Furthermore, excess fluoride
exposure causes human health impacts such as decreased thyroid
function, Type II diabetes, attention deficit hyperactivity disorder
(ADHD), and neurotoxicity [4,5,9]. Fluoride can also be an environ-
mental pollutant [8,10]. Therefore, detection of fluoride levels in
drinking water, food, and the environment is crucial and necessary.
In recent years, high sensitivity and selectivity of colorimetric
sensors for fluoride ion (F−) detection have been developed based
on chemical interactions such as Lewis acids-based interactions
[11], hydrogen-bonding interactions [12,13], reaction-based interac-
ica particle-based interactions [17]. In many cases, the binding unit
is primarily composed of F− anion which interacts with hydrogen
donors, called chemosensors, through hydrogen bonding in a sens-
ing system. Various types of chemosensors have been introduced
for this purpose, including urea or thiourea groups, amides, por-
phyrin, and phenols [18]. Among these chemosensors, salicylidene
Schiff bases have received widespread interest due to their pho-
tochromic and thermochromic properties caused by tautomerism
of two different forms (enol and keto forms) [19-21]. These prop-
erties depend on the molecular planarity, nature of crystal pack-
ing, effect of substituent, and nature of the solvent [22], which can
be visible colorimetric chemosensors for anion detection via naked
eye.
However, development of sensors which exhibit high selectiv-
ity and sensitivity in the visible region of the spectrum remains
challenging because traditional methods to detect F− such as high-
performance electrochemical sensors are difficult and expensive
[23]. Here the researchers report on high selective and sensitive
optical chemosensors based on salicylidene Schiff bases for F− an-
ion detection. Three anion sensors, salicylaldehyde-o-aminophenol
∗
Corresponding author.
E-mail address: fsciswsm@ku.ac.th (S. Suramitr).
https://doi.org/10.1016/j.molstruc.2021.131132
0022-2860/© 2021 Elsevier B.V. All rights reserved.

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Fig. 1. Structure and synthetic route of salicylidene Schiff base derivatives (SA1, SA2 and SA3).
Table 1
(SA1), 3,5-dimethyl-salicylaldehyde-o-aminophenol (SA2), and 3,5-
dichloro-salicylaldehyde-o-aminophenol (SA3), are designed, syn-
thesized, and characterized using several methods. The sensor
molecules were modified by substitution of aromatic proton (H) (in
SA1) with electron donating groups (methyl for SA2) and electron
withdrawing group (chlorine for SA3) (Fig. 1). The photophysical
properties of the compounds were studied followed by an evalua-
tion of their fluoride sensing performance. The structural geome-
tries and proton transfers processes of all molecules were investi-
gated and F− detection mechanism was also proposed using quan-
tum chemical calculations. It shown that the developed salicyli-
dene Schiff base sensors exhibited high sensitivity and selectivity
which can clearly be seen with the naked eye when detecting flu-
oride anions.
Summary of absorption and emission spectral data and molar absorptivity of SA1,
SA2 and SA3 in different solvents.
Molar absorptivity
Compounds
Solvents
λabs (nm)
(cm−1 M−1
)
λem (nm)
SA1
Chloroform
Ethanol
354
350
348
366
353
341
367
358
357
8,585
525
519
529
564
555
559
553
549
554
11,675
11,602
8,177
Acetonitrile
Chloroform
Ethanol
SA2
SA3
5,101
Acetonitrile
Chloroform
Ethanol
10,340
16,628
20,135
27,893
Acetonitrile
Table 2
The association constants of SA1, SA2 and SA3 with F−
,
−
AcO−, H2PO4 in CH3CN as determined from UV-vis ab-
2. Experimental
sorption titrations.
Association constant (Ka in M−1
)
2.1. Materials and equipment
Sensors
F−
AcO−
H2PO4
−
All materials for synthesis were purchased from Sigma Aldrich
and used without further purification. Solvents and reagents used
in spectroscopic studies were of analytical grade. Nuclear Magnetic
Resonance analyses were recorded on an INNOVA UNITYVARIN
spectrometer operated at 400.00 MHz for 1H-NMR in deuterated
dimethyl sulfoxide-d6 (DMSO-d6) and tetramethylsilane (TMS) was
used as an internal reference. UV-vis absorption spectra were taken
using Jasco V-670 UV/VIS/NIR spectrophotometer in the wave-
length range of 200-700 nm. The experiments were carried out
using a 2 nm slit width in 1 cm quartz cells at room temper-
ature. In the titration experiments, the stock solutions of sensor
molecules (SA1, SA2 and SA3) (1.0 × 10−4 mol L−1) were prepared
in acetonitrile (CH3CN) solution. All anions (F−, Cl−, Br−, I−, AcO−
and H2PO4−) used in this study were prepared by dissolving the
anions, which were presented in the form of tetrabutylammonium
(TBA) salts in acetonitrile to reach a concentration of 0.004 mol
SA1
SA2
SA3
3.92 × 103
1.38 × 103
8.30 × 103
-
-
-
-
14.10 × 103
5.28 × 103
reaction was completed. After that, the synthesized compound was
washed with DI water.
Salicylaldehyde-o-aminophenol, (SA1) was obtained as an or-
ange powder with a yield of 82% and melting point in the range
of 181-182°C. 1H NMR (DMSO-d6, δppm): 13.77 s 1H (OH); 9.73
s 1H (OH); 8.96 s 1H (CH=N); 7.79-6.45 m 8H (Aromatic). UV-vis
(CH3CN) λmax: 348 nm.
3,5-dimethyl-salicylaldehyde-o-aminophenol, (SA2) was ob-
tained as a brown powder with a yield of 81% and melting point in
the range 164-165°C. 1H NMR (DMSO-d6, δppm): 13.89 s 1H (OH);
9.74 s 1H (OH); 8.87 s 1H (CH=N); 7.70-6.38 m 6H (Aromatic);
2.38-1.69 m 6H (CH). UV-vis (CH3CN) λmax: 341 nm.
3,5-dichloro-salicylaldehyde-o-aminophenol, (SA3) was ob-
tained as an orange powder with a yield of 75% and melting
point in the range 237-238°C. 1H NMR (DMSO-d6, δppm): 10.22
s 1H (OH); 9.08 s 1H (CH=N); 7.95-6.58 m 5H (Aromatic). UV-vis
(CH3CN) λmax: 359 nm.
L
−1. Equilibrium constants of the sensors and anions were calcu-
lated following the method reported previously [24].
2.2. Synthesis of salicylidene Schiff base derivatives (SA1, SA2 and
SA3)
Sensor molecules were prepared using condensation of 2-
amino-phenol (5 mmol) with 0.5 ml of salicylaldehyde, 3,5-
dimethylsalicylaldehyde, and 3,5-dichlorosalicylaldehyde to pro-
duce SA1, SA2, and SA3, respectively. The chemicals were mixed
in 10 ml glycerol, stirred, and refluxed for 2 hours at 80°C. The re-
action was monitored using thin-layer chromatography (TLC). Dis-
appearance of starting compounds on the TLC indicated that the
2.3. Computational details
Theoretical calculations were performed with the Gaussian 09
package [25]. The geometries of the electronic ground state (S0) of
the three sensors (SA1, SA2 and SA3) are shown in Fig. 1 which
were calculated using the density functional theory (DFT) method.
2

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Fig. 2. UV-visible absorption spectra of (a) SA1, (b) SA2 and (c) SA3 in CH3CN (1 × 10−4 M) in the presence of various anions (4 × 10−3 M) (F−, Cl−, Br−, I−, AcO−, and
−
H2PO4
)
The long-range corrected version of B3LYP using the Coulomb-
attenuating method CAM-B3LYP functional, which contains 65%
long-range HF exchange [25,26] and the 6-31G(d,p) basis set was
chosen for all atoms. Frequency calculations were also conducted
at the same level to ensure that the optimized structure was the
minimum on the potential energy surface (PES). Absorption spectra
were also studied using the Time-dependent DFT method with the
6-311G(d,p) basis set and the Coulomb-attenuating method (CAM)
with a long-range corrected version of B3LYP (TD-CAM-B3LYP/6-
311G(d,p)) calculations under the S0 and S1 minimum structures
3

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Fig. 3. UV-visible absorption titration spectra of (a) SA1, (b) SA2 and (c) SA3 recorded in CH3CN (1 × 10−4 M) after addition of TBAF (0-8 equivalent).
[27-30]. The integral equation formalism variant of the polarizable
continuum model (PCM) [31] was employed using acetonitrile as
a solvent throughout the whole theoretical calculations. Further-
more, complexes consisting of the three sensors and F− ion were
optimized at the same theoretical levels as above with the consid-
eration of the BSSE correction using the counterpoise method [32].
Schiff base structures of all compounds. Photophysical properties
including absorption, emission, and molar absorptivity of the sali-
cylidene derivatives were investigated in three solvents with differ-
ent polarities, as shown in Table 1. The solvents were chloroform
(CHCl3, dielectric constant = 4.81), ethanol (C2H6O, dielectric con-
stant = 24.5), and acetonitrile (CH3CN, dielectric constant = 37.5).
The absorption spectra of the derivatives in all solvents showed
maximum wavelengths of absorption (λabs) in the range of 340-
370 nm. With increased solvent polarity, the λabs of all derivatives
slightly shifted to a longer wavelength (blue-shift). This is poten-
tially caused by increasing of the interaction stabilization between
hydroxyl group of the solvents and phenol group of the salicyli-
dene enabling high delocalization of electron cloud on the aro-
matic system.
3. Results and discussion
3.1. Photophysical properties
3.1.1. Absorption and fluorescence properties
Three salicylidene Schiff base derivatives, SA1, SA2, and SA3,
were synthesized followed by structural characterizations using
1H-NMR and FT-IR techniques. The results obtained from 1H-NMR
measurements in dimethyl sulfoxide (DMSO-d6) detailed in the
Supporting Information (SI) (Fig. S1) confirmed the formation of
For maximum wavelengths of emission (λem) (See in Support-
ing Information, Fig. S2) of all compounds, emission wavelengths
4

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Fig. 4. Partial 1H NMR spectra of SA1 (7 mM in DMSO-d6) with addition of different equivalent (0.0, 1.0 and 2.0) of TBAF.
Fig. 5. ESP molecular surfaces of the sensors (SA1, SA2 and SA3). Positive and negative ESP surfaces are represented in blue and red regions, respectively. Only positive
values presented on the ESP surfaces are shown.
at a large Stokes shift in a range of 525-560 nm were observed
when using non-polar solvent (CHCl3) and aprotic solvent (CH3CN).
The emission is normally caused by an excitation of enol isomer
to produce keto isomer, and this resulted in the presence of ESIPT
process in the system. In the case of using protic solvent (EtOH),
the short wavelength emission bands (< 450 nm) disappeared and
only the emission bands in a large Stokes shift range (519-555
nm) were observed. This phenomenon is caused by the presence
of intermolecular hydrogen bonding of the derivatives with solvent
molecules leading to stabilization of the solvated keto isomers and
prevention of the ESIPT process [30].
5

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
nm) in the presence of F−, while no apparent color change was ob-
served as treatment with the other anions (Cl−, Br−, I−, AcO−, and
H2PO4−). In the case of SA3, the solution color changed from light
yellow to orange red (420 nm) after treatment with three anions,
which were F−, AcO−, and H2PO4−. No noticeable change in color
was found when SA3 was treated with Cl−, Br−, and I−. These find-
ings indicate that all derivative compounds had colorimetric sensi-
tivity to F− in CH3CN solution. However, SA1 and SA2 could be po-
tentially utilized as sensors because they showed higher selectivity
toward the F− detection than SA3.
Binding affinities between the derivatives and F− ions were ex-
amined using titration experiments. The titrations were carried
out by adding tetrabutylammonium fluoride (TBAF) solution, which
contained F− ions, into the derivative solutions (1 × 10−4 M) pre-
pared in CH3CN and monitored using UV-visible spectroscopy. Ab-
sorption titration spectra of all compounds with increasing F− con-
centrations are shown in Fig. 3. Before F− addition, the spectra of
SA1, SA2, and SA3 showed λabs peaks at 348, 341 and 359 nm, re-
spectively, which corresponded to enol structure of the compounds
[30]. Obviously, weak absorption bands at 440 and 470 nm were
detected in SA1 and SA3 spectra, respectively, which are presum-
ably caused by the presence of both enol and keto forms in the so-
lutions. After increasing F− concentrations, new peaks at 406, 425,
and 420 nm for SA1, SA2, and SA3, respectively, were visibly de-
tected. The formation of these peaks is likely caused by either hy-
drogen bonding between F− and Ar-OH or deprotonation of Ar-OH
protons [33].
Binding stoichiometry of the SA1, SA2, and SA3 with F− can be
calculated using the Benesi-Hildebrand equation [34] as follows:
1/(A-A )=1/(A -A )[1/K[F−] +1]
∞
0
0
0
where, A0 is an absorbance of free sensor, A is an absorbance
∞
measured with excess amount of F−, A is an absorbance measured
with a specific amount of F−, Ka is the association constant (M−1),
and [F−]0 is a concentration of F− added (M).
The plot of 1/(A-A0) versus 1/[F−]0 of SA1, SA2, and SA3 showed
a linear relationship with R2 of 0.9855, 0.9912, and 0.9923, respec-
tively. This indicates that all sensors and fluoride ions formed a 1:1
binding stoichiometry. The association constants (Ka) between the
sensors and F− were calculated from the ratio of intercept/slope
and summarized in Table 2. The Ka values revealed that the flu-
oride affinities of all sensors are in an order of SA2 < SA1 <
SA3. This implies that proton substitution by electron withdrawing
group (Cl in SA3) in the sensor molecule could lead to higher fluo-
ride affinity than that by electron donating groups (CH3 for SA2)
and higher than the molecule without substitution (H for SA1).
Limit of detections (LOD) for all sensors could be calculated from
3σ/m, where σ is the standard deviation of the blank solution and
m is the slope of the calibration curve [35]. The calculated values
were 8 × 10−5, 21 × 10−5, and 7 × 10−5 M for SA1, SA2, and SA3,
respectively.
Fig. 6. Geometry structures of (a) SA1enol, (b) SA1keto and (c) SA1keto-F− complex.
To further understand the interaction between the fluoride ion
and the developed sensors, 1H NMR titration experiments were
conducted in DMSO-d6 solution, as shown in Fig. 4. Before treat-
ment SA1 with F−, the chemical shifts of OH protons were s 13.77
and s 9.73 ppm and of –HC=N proton combined with aromatic
protons was s 8.87 ppm. After addition of one equivalent of F−,
the OH peaks disappeared indicating the formation of hydrogen-
bond interactions between the SA1 and the fluoride ions. The re-
maining peaks presented at s 9.73 and s 8.96 ppm could be mainly
attributed to aromatic protons in the SA1 molecule. As increase F−
concentration to be 2 equivalent of F−, these peaks were slightly
shifted to larger value, which is likely caused by deprotonation of
the SA1 at high concentration of F−. Similar phenomenon was also
observed for SA2 and SA3. (See in Supporting Information, Fig. S1).
˚
Only the intramolecular H-bond distances (in A) are shown. The colors of various
atoms within the molecules are grey for carbon, white for hydrogen, red for oxygen,
blue for nitrogen, green for chlorine, and indigo blue for fluoride.
3.1.2. UV-Visible absorption of anion sensing
The interactions between the salicylidene Schiff base derivatives
(SA1, SA2, and SA3) and various types of anions (F−, Cl−, Br−, I−,
AcO−, and H2PO4−) were investigated in CH3CN solutions through
colorimetric analysis. UV-vis absorption spectra of the salicylidene
Schiff base derivatives solutions (1 × 10−4 M) after adding different
anions (4 × 10−3 M) were recorded after three minutes, as shown
in Fig. 2. It was found that the color of SA1 and SA2 solutions
changed from light yellow (340-350 nm) to orange red (400-425
6

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Scheme 1. Potential mechanism for F− detection of all sensors.
3.2. Theoretical calculation
SA1keto and SA1keto-F− complexes. For the SA1-enol and SA1-
keto, it shown that the O-H stretching vibrations of the SA1-enol
appeared at 3760 cm−1 and 3322 cm−1, while SA1keto’s O–H and
To investigate the active site of SA1, SA2, and SA3 molecules for
F− detection, theoretical calculations were conducted. Electrostatic
potential (ESP) analysis was undertaken using the Gauss View 5.0
program. The ESP molecular surfaces of the sensor molecules are
presented in Fig. 5. For all sensing compounds, it is revealed that
the H atom of the OH moieties gave a positive ESP surface, while
the aromatic unit showed a negative ESP surface. This resulted in
the preference of F− anions to close to the positively charged H
atoms rather than to the negatively charged aromatic units. In con-
clusion, detection of the sensor compounds can be performed via
interaction between F− anion and H atom of the OH moieties of
the sensors.
Structures of the SAenol, SAketo, and SAketo-F− intermediate and
the deprotonated form were optimized separately (Fig. 6) in DFT
formalism with 6-311 G(d,p) basis set and the CAM-B3LYP func-
tional. At each optimized structure, lowest few electronic transi-
tions of the molecules (SAenol, SAketo, and SAketo-F−) were exam-
ined with TDDFT formalism. All calculations were carried out in
acetonitrile as solvent, treated in Onsager’s SCRF formalism, with
polarisable continuum (PCM) approximation. A single imaginary
frequency confirmed the obtained structure as the desired transi-
tion state.
N–H stretching vibrations appeared at 3777 cm−1 and 2493 cm−1
,
respectively. Compared to SA-keto+F− complex, the N–H stretching
vibration of complex appeared at 3260 cm−1, indicating large red-
shifts. All calculation in Fig. 8 indicated that there were strong in-
teractions in the O•••H–X (X = N and F) hydrogen bond and there
were large red-shifts for H–X stretching vibrational frequencies in
the SA-enol, SA-keto and SA-keto+F− complexes.
The interaction energies with BSSE corrections for com-
plexes (SA1-keto+F−, SA2-keto+F− and SA3-keto+F−) are listed in
Table 3. All the values of the adsorption energies being negative, it
is implied that the adsorption structures are stable, and the pro-
cess is exothermic. It is clear that the BSSE-corrected interaction
energy (ꢀEBSSE) of complex SA1-keto+F−, SA2-keto+F− and SA3-
keto+F− are favorable for the distinct selectivity of F−. These calcu-
lation results are in good agreement with the reported experimen-
tal observations that intermolecular proton transfer (IPT) between
chemosensor substrate SA1, SA2 and SA3. Thus, one can conclude
that the host chemosensors have a much stronger affinity to F− ion
through intermolecular proton transfer, which leads to the forma-
tion of chemosensor anions by F−.
To simulate the absorption spectra of the molecules, vertical ex-
citation energies of the SAenol, SAketo, and SAketo-F− for SA1, SA2,
and SA3 were calculated based on the time-dependent density
functional theory (TD-DFT) method using their ground-state opti-
mized structures. The calculated electronic transitions energies and
corresponding transition oscillator strengths for transition from the
ground state (S0) to the excited state (S1) of the sensor molecules
as well as the experimental UV-vis absorption data are summa-
rized in Table 4.
The optimized structure of the SAenol, SAketo, and SAketo-F− form
confirmed the mechanism of fluoride binding (Scheme 1) through
hydrogen bond formation between the O–H and N–H with F−.
Some geometrical parameters were shown in Fig. 6. The lengths
of O–H of SA1enol, SA1keto, and SA1keto-F− were increased to 0.991
˚
˚
˚
A, 1.561 A and 1.998 A, respectively, while the N–H lengths were
˚
˚
˚
decreased to 1.741 A, 1.063 A and 1.034 A, respectively. Moreover,
this study revealed that the NH–O and N–HO hydrogen bonds in
Figs. 6 and S2 are involved in NH and OH tautomers from which
it could be inferred that F− binding occurs through the SAketo-F−
form, followed by deprotonation of the ligand, formation of hydro-
gen bond between deprotonated oxygen, and -NH and release of
HF. In addition to, the frequency calculations were performed on
each optimized structure. Similar phenomenon was also observed
for SA2 and SA3 (See in Supporting Information, Fig. S3).
From the DFT/TDDFT studies, it was observed that the theoreti-
cal absorption peak (λabs) of SA1, SA2, and SA3 at 321, 327, and 329
nm of the enol form were caused due to H→L transition which
was compared with experimental values (λexpt) at 334, 343, and
349 nm, respectively. The theoretical absorption peaks of the keto
form, at 368, 382, and 383 nm may be compared with the exper-
imental values at 380, 398, and 400 nm, respectively. It is found
that all the calculated wavelengths (λcal) showed a slight blue shift
compared to that obtained from the experiments with an under-
estimated error of ~3-7%, which was in an acceptable range. The
absorption of the probable intermediate (SAketo-F−) were obtained
theoretically at 378, 386, and 395 nm which is very close to the ex-
perimental value of keto at 380, 398, and 400 nm, respectively. The
Frequency calculations were carried out on each optimized
structure, and their IR spectra were discussed. Fig. 7 showed the
simulated infrared (IR) spectra for the SA-enol, SA-keto and SA-
keto+F−, where the intensity was plotted against the harmonic vi-
brational frequencies. For the complexity of vibrational modes, it
was difficult to attribute all bands, so we have only analyzed some
H–X (X = O and N) vibrational frequencies [36-38] in the SA1enol,
7

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Fig. 7. IR spectra of SA1enol, SA1keto and SA1keto+F− complexes.
Fig. 8. IR spectra of SA1, SA2 and SA3 and F− complexes.
Table 3
The interaction energies with BSSE corrections ꢀEBSSE (in kcal/mol) of complexes, bond
-H/N -H hydrogen bonds (in A) of SAketo-F− interaction at the CAM-
˚
length of
O
B3LYP/6-311G(d,p) level.
˚
˚
Complexes
SA1keto-F−
SA2keto-F−
SA3keto-F−
O-H bond length (A)
N-H bond length (A)
ꢀEBSSE (kcal/mol)
-59.67
1.998
2.001
2.052
1.034
1.033
1.039
-58.69
-56.62
8

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
Table 4
Excitation energies (Eex), maximum absorption wavelengths (λmax) and oscillator strengths (f) of enol
(SAenol) and keto (SAketo) form, and intermediate (SA1keto-F−) calculated in acetonitrile medium by the
TDDFT method.
States
SA1
Transition
Eex (eV)
λcal (nm)
f
Transition composition
λExpt (nm)
SAenol
SAketo
SAketo-F−
SA2
S0→S1
S0→S1
S0→S1
3.87
3.36
3.27
321
368
378
0.563
0.395
0.498
HOMO→LUMO (91%)
HOMO→LUMO (97%)
HOMO→LUMO (91%)
334
380
-
SAenol
SAketo
SAketo-F−
SA3
S0→S1
S0→S1
S0→S1
3.79
3.24
3.21
327
382
386
0.502
0.365
0.486
HOMO→LUMO (94%)
HOMO→LUMO (98%)
HOMO→LUMO (95%)
343
398
-
SAenol
SAketo
SAketo-F−
S0→S1
S0→S1
S0→S1
3.77
3.23
3.14
329
383
395
0.521
0.392
0.508
HOMO→LUMO (89%)
HOMO→LUMO (97%)
HOMO→LUMO (89%)
349
400
-
Fig. 9. Molecular orbital characteristics of SA1, SA2 and SA3 in keto-form.
theoretical values confirmed the probable deprotonation via hydro-
gen bonding.
of all compounds. It was also evident that the atomic orbitals of
the N atom in the sensor molecules took more proportion in LUMO
than in HOMO. This led to an increased electron density of the N6
upon excitation to the S1 state. Consequently, the N–H bond length
in the sensor was shortened in the S1 state resulting in the occur-
rence of ESPT process in each sensor molecule.
The calculated results showed that the S0→S1 states of all sen-
sor molecules were mainly derived from an electron excitation
from the highest occupied molecular orbital (HOMO) to the lowest
unoccupied molecular orbital (LUMO) (by ≥ 89%). This indicated
that all molecules exhibited a π- π∗ type character. From molec-
ular orbital studies of all the sensor molecules (in Fig. 9), it is re-
vealed that both of HOMO and LUMO were mainly contributed by
phenol and imine. This confirmed the presence of intramolecular
charge transfer character of the transitions from HOMO to LUMO
4. Conclusions
In summary, the present paper has reported anion chemosen-
sors based on salicylidene Schiff bases, salicylaldehyde-o-
9

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
aminophenol (SA1), 3,5-dimethyl-salicylaldehyde-o-aminophenol
(SA2), and 3,5-dichloro-salicylaldehyde-o-aminophenol (SA3), for
colorimetric detection of fluoride ions. The sensing mechanisms
were also investigated based on experimental methods combined
with theoretical calculations. The salicylidene derivatives showed
maximum absorption (λabs) and emission (λem) at wavelengths
ranges of 341-357 nm and 407-427 nm, respectively. All com-
pounds showed high sensitivity in colorimetric detection of F−
in CH3CN solutions with a binding stoichiometry of one F− ion
per sensor molecule. Although SA3 showed maximum F− binding
[4] H.S. Horowitz, The 2001 CDC recommendations for using fluoride to prevent
and control dental caries in the United States, J. Public Health Dent. 63 (1)
(2003) 3–8.
[5] R.J. Carton, Review of the 2006 United States National Research Council report:
fluoride in drinking water, Fluoride 39 (3) (2006) 163–172.
[6] K.L. Kirk, Biochemistry of inorganic fluoride, in: Biochemistry of the Elemental
Halogens and Inorganic Halides, Springer, 1991, pp. 19–68.
[7] V. Kumar, M. Kaushik, A. Srivastava, A. Pratap, V. Thiruvenkatam, T.G. Row,
Thiourea based novel chromogenic sensor for selective detection of fluoride
and cyanide anions in organic and aqueous media, Anal. Chim. Acta 663 (1)
(2010) 77–84.
[8] P. Singh, M. Barjatiya, S. Dhing, R. Bhatnagar, S. Kothari, V. Dhar, Evidence sug-
gesting that high intake of fluoride provokes nephrolithiasis in tribal popula-
tions, Urol. Res. 29 (4) (2001) 238–244.
affinity (8.30 × 103
M
−1) and highest stability in keto-form, it
[9] A.J. Malin, C. Till, Exposure to fluoridated water and attention deficit hyper-
activity disorder prevalence among children and adolescents in the United
States: an ecological association, J. Environ. Health 14 (1) (2015) 17.
[10] Y. Michigami, Y. Kuroda, K. Ueda, Y. Yamamoto, Determination of urinary fluo-
ride by ion chromatography, Anal. Chim. Acta 274 (2) (1993) 299–302.
[11] M. Melaimi, F.P. Gabbaï, A heteronuclear bidentate Lewis acid as a phospho-
rescent fluoride sensor, J. Am. Chem. Soc. 127 (27) (2005) 9680–9681.
[12] Z.J. Chen, L.M. Wang, G. Zou, L. Zhang, G.J. Zhang, X.F. Cai, et al., Colorimetric
and ratiometric fluorescent chemosensor for fluoride ion based on perylene
diimide derivatives, Dyes Pigm. 94 (3) (2012) 410–415.
provided lowest selectivity in F− detection. Therefore, SA1 and
SA2 are selected as promising F− sensors due to their high selec-
tivity with a low detection limit (LOD). Theoretical calculations
confirmed the presence of intramolecular charge transfer character
of the transitions from HOMO to LUMO of all compounds. Based
on DFT and ESP calculations, it is revealed that F− detection could
occur through the interaction between F− anion and H atom of
the OH moieties of the sensor molecules.
[13] M. Lee, S. Jo, D. Lee, Z. Xu, J. Yoon, A new naphthalimide derivative as a se-
lective fluorescent and colorimetric sensor for fluoride, cyanide and CO2, Dyes
Pigments 120 (2015) 288–292.
Author statement
[14] J.M. Eun, B. Roy, A.K. Han, Turn-on“ fluorescent sensing with ”reactive" probes,
Chem. Commun. 47 (27) (2011) 7583–7601.
Buchitar Nakwanich: Conceptualization, Methodology, Data cu-
ration, Writing - original draft. Amonchat Koonwong: Data cura-
tion, Writing- Original draft preparation, Investigation
[15] T.H. Kim, T.M. Swager,
A fluorescent self-amplifying wavelength-responsive
sensory polymer for fluoride ions, Angew. Chem. Int. Ed. 42 (39) (2003)
4803–4806.
[16] A.M. Alkilany, S.E. Lohse, C.J. Murphy, The gold standard: gold nanoparticle
libraries to understand the nano-bio interface, Accounts Chem. Res. 46 (3)
(2013) 650–661.
[17] Y. Zhou, J.F. Zhang, J. Yoon, Fluorescence and colorimetric chemosensors for
fluoride-ion detection, Chem. Rev. 114 (10) (2014) 5511–5571.
[18] C.B. Da Silva, T. Kroetz, F.S. Santos, F.S. Rodembusch, Synthesis and photophysi-
cal characterization of proton transfer-based thiourea derivatives: Potential ap-
plication as colorimetric naked-eye chemosensor for fluoride detection in so-
lution, J. Braz. Chem. Soc. 28 (9) (2017) 1830–1841.
Anwaraporn Suramitr: Writing- Reviewing. Panida Prompinit:
Investigation. Rungtiva P. Poo-arporn: Investigation. Supa Hannong-
bua: Software, Investigation. Songwut Suramitr: Supervision, Writ-
ing - original draft, Writing - review & editing.
Declaration of Competing Interest
[19] E. Hadjoudis, I.M. Mavridis, Photochromism and thermochromism of Schiff
bases in the solid state: structural aspects, Chem. Soc. Rev. 33 (9) (2004)
579–588.
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
[20] G.Y. Li, D. Liu, H. Zhang, W.W. Li, F. Wang, Y.H. Liang, TDDFT study on the
sensing mechanism of a fluorescent sensor for fluoride anion: inhibition of the
ESPT process, Spectrochimica Acta Part A: Mol. Biomol. Spectrosc. 149 (2015)
17–22.
Acknowledgement
[21] Y.M. Hijji, B. Barare, A.P. Kennedy, R. Butcher, Synthesis and photophysical
characterization of a Schiff base as anion sensor, Sens. Actuat. B: Chem. 136
(2) (2009) 297–302.
This work is supported by Thailand Research Fund for basic
research Grant, Generous supply of computing time and research
facilities by the National Research University Project of Thailand
(NRU), the Center of Nanoscience Kasetsart University, Laboratory
of Computational and Applied Chemistry (LCAC), the Computing
Center of Kasetsart University. I would like to grateful thank to
my advisor, Assistant Professor Songwut Suramitr for inspiration
this work and valuable suggestion to solve many unexpected prob-
lems. This research is funded by Kasetsart University through the
Graduate School Fellowship Program. This work would not succeed
without the funding support including Office of the Higher Edu-
cation Commission, National Nanotechnology Center (NANOTEC),
Kasetsart University Research and Development Institute (KURDI)
were financial supporting during my research.
[22] S. Maity, M. Shyamal, P. Mazumdar, G.P. Sahoo, R. Maity, G. Salgado-Morán,
ꢀ
et al., Solvatochromism and turn-off fluorescence sensing property of N, N -bis
(3-pentyl) perylene-3, 4, 9, 10-bis (dicarboximide) towards nitroaromatics
and photophysical study of its microstructures, J. Mol. Liq. 224 (2016) 255–264.
[23] R. Hein, P.D. Beer, J.J. Davis, Electrochemical anion sensing: Supramolecular ap-
proaches, Chem. Rev. 120 (2020) 1888–1935.
[24] C. Qinghui, D.A. Medvetz, P. Yi, A polymeric colorimetric sensor with excit-
ed-state intramolecular proton transfer for anionic species, Chem. Mater. 19
(26) (2007) 6421–6429.
[25] Gaussian09 RA. 1, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, GA Petersson et al.,
Gaussian. Inc, Wallingford CT. 2009.
[26] Y. Tawada, T. Tsuneda, S. Yanagisawa, T. Yanai, K. Hirao,
A long-range-cor-
rected time-dependent density functional theory, J. Chem. Phys. 120 (2004)
8425–8433.
[27] T. Yanai, D.P. Tew, N.C. Handy, A new hybrid exchange–correlation functional
using the Coulomb-attenuating method (CAM-B3LYP), Chem. Phys. Lett. 393
(2004) 51–57.
[28] W.J. Hehre, R. Ditchfield, J.A. Pople, Self-consistent molecular orbital methods.
XII. Further extensions of gaussian-type basis sets for use in molecular orbital
studies of organic molecules, J. Chem. Phys. 56 (5) (1972) 2257–2261.
[29] C. Van Caillie, R.D. Amos, Geometric derivatives of density functional theory
excitation energies using gradient-corrected functionals, Chem. Phys. Lett. 317
(1) (2000) 159–164.
Supplementary materials
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.molstruc.2021.131132.
[30] N. Klinhom, N. Saengsuwan, S. Sriyab, P. Prompinit, S. Hannongbua, S. Suramitr,
Photophysical properties for excited-state intramolecular proton transfer
(ESIPT) reaction of N-salicylidene-o-aminophenol: experimental and DFT
References
based approaches, Spectrochim. Acta
359–366.
A Mol. Biomol. Spectrosc. 206 (2019)
[1] J. Du, M. Hu, J. Fan, X. Peng, Fluorescent chemodosimeters using “mild” chem-
ical events for the detection of small anions and cations in biological and en-
vironmental media, Chem. Soc. Rev. 41 (12) (2012) 4511–4535.
[31] A. Pedone, Role of solvent on charge transfer in 7-aminocoumarin dyes: new
hints from TD-CAM-B3LYP and state specific PCM calculations, J. Chem. Theory
Comput. 9 (9) (2013) 4087–4096.
[2] V. Amendola, D. Esteban-Gómez, L. Fabbrizzi, M. Licchelli, What anions do to
N− H-containing receptors, Acc. Chem. Res. 39 (5) (2006) 343–353.
[3] Z.Q. Liu, M. Shi, F.Y. Li, Q. Fang, Z.H. Chen, T. Yi, et al., Highly selective two-pho-
ton chemosensors for fluoride derived from organic boranes, Org. Lett. 7 (24)
(2005) 5481–5484.
[32] S. Simon, M. Duran, J.J. Dannenberg, How does basis set superposition error
change the potential surfaces for hydrogen-bonded dimers? J. Chem. Phys. 105
(1996) 11024–11031.
10

B. Nakwanich, A. Koonwong, A. Suramitr et al.
Journal of Molecular Structure 1245 (2021) 131132
[33] A. Costela, I. Garcıa-Moreno, R. Mallavia, F. Amat-Guerri, J. Barroso,
R. Sastre, Proton-transfer lasers based on solid copolymers of modified
[36] Y.L. Leng, J.H. Zhang, Q. Li, Y.M. Zhang, Q. Lin, H. Yao, et al., An easy prepared
dual-channel chemosensor for selective and instant detection of fluoride based
on double Schiff-base, Spectrochim. Acta A Mol. Biomol. Spectrosc. 167 (2016)
116–121.
ꢀ
2-(2 -hydroxyphenyl) benzimidazoles with methacrylate monomers, Opt. Com-
mun. 152 (1-3) (1998) 89–95.
[34] Q. Lin, B. Sun, Q.P. Yang, Y.P. Fu, X. Zhu, T.B. Wei, et al., Double metal
[37] S. Sriyab, M.P. Gleeson, S. Hannongbua, S. Suramitr, Photophysical properties
and computational investigation on substituent effects on the structural and
electronic properties of 3, 6-di (thiophene-2-yl)-carbazole-based derivatives, J.
Mol. Struct. 1125 (2016) 532–539.
ions competitively control the guest-sensing process:
a facile approach to
stimuli-responsive supramolecular gels, Chem. Eur. J. 20 (36) (2014) 11457–
11462.
[35] J. Wu, G. Lai, Z. Li, Y. Lu, T. Leng, Y. Shen, et al., Novel 2, 1, 3-benzothiadiazole
derivatives used as selective fluorescent and colorimetric sensors for fluoride
ion, Dyes Pigm. 124 (2016) 268–276.
[38] Y.X. Tan, Z.J. Zhang, Y. Liu, J.X. Yu, X.M. Zhu, D.Z. Kuang, et al., Synthesis, crystal
structure and biological activity of the Schiff base organotin (IV) complexes
based on salicylaldehyde-o-aminophenol, J. Mol. Struct. 1149 (2017) 874–881.
11