Turn-off fluorene-based chemosensor switch to Fe3+: Spectroscopic study, merit parameters, theoretical calculations, and its application in Brazilian ethanol fuel

Article abstract of DOI:10.1016/j.ica.2021.120511

Development of a new chemosensor is challenging, because it has to deliver sensitiveness, selectiveness, reproducibility, robustness, and the least number of interfering analytes. In this sense coordination chemistry can offer an attractive approach to the problem. Diethyl 2-(9-fluorenyl) malonate (DEFM) was found to be an efficient fluorosensor for Fe³⁺ ions in ethanolic solutions, showing fluorescence quenching of 99% in the concentration range of 0.1–1.0 μmol L^?1, and most importantly, with no effect of interfering ions. Theoretical calculations supported photoelectron transfer (PET) mechanism to explain the fluorescence suppression observed. The method passed several analytical tests and showed the limit of detection and the limit of quantification at 15.56 and 51.85 nmol L^?1, respectively. The sensor passed a series of analytical tests and proved to be an attractive low-cost approach for Fe³⁺ quantification in ethanol fuel.

Full text of DOI:10.1016/j.ica.2021.120511

Inorganica Chimica Acta 526 (2021) 120511
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
3
+
Turn-off fluorene-based chemosensor switch to Fe : Spectroscopic study,
merit parameters, theoretical calculations, and its application in Brazilian
ethanol fuel
a
a
b
Fabiane dos Santos Carlos , Letícia Aparecida da Silva , Cristiano Zanlorenzi , F a´ bio Souza
Nunes
a,*
a
Grupo de Espectroscopia e Reatividade de Compostos de Coordenaç a˜ o, Departamento de Química, Universidade Federal do Paran a´ , Cx. Postal 19081, 81531ꢀ 980
Curitiba, PR, Brazil
b
Laborat ´o rio Polímeros Paulo Scarpa, Departamento de Química, Universidade Federal do Paran a´ , Cx. Postal 19081, 81531ꢀ 980 Curitiba, PR, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords:
Development of a new chemosensor is challenging, because it has to deliver sensitiveness, selectiveness,
reproducibility, robustness, and the least number of interfering analytes. In this sense coordination chemistry can
offer an attractive approach to the problem. Diethyl 2-(9-fluorenyl) malonate (DEFM) was found to be an effi-
Fluorescencent chemosensor
Molecular switch
3
+
Fe
DFT
3+
cient fluorosensor for Fe ions in ethanolic solutions, showing fluorescence quenching of 99% in the concen-
ꢀ
1
tration range of 0.1–1.0
μ
mol L , and most importantly, with no effect of interfering ions. Theoretical
calculations supported photoelectron transfer (PET) mechanism to explain the fluorescence suppression
observed. The method passed several analytical tests and showed the limit of detection and the limit of quan-
ꢀ
1
tification at 15.56 and 51.85 nmol L , respectively. The sensor passed a series of analytical tests and proved to
be an attractive low-cost approach for Fe3 quantification in ethanol fuel.
+
1
. Introduction
Iron is an essential element. According to the World Health Orga-
the levels of these ions within their maximum allowed limits [3].
Iron can be inserted in this organic matrix through the steps involved
in ethanol production or the storage phase [2–4]. Despite being found in
low concentrations, the presence of metal ions can compromise the good
performance of the engines, especially those powered by ethanol-
gasoline mixture, due to the formation of gums (insoluble solids) by
peroxidation processes of hydrocarbons present in gasoline; or oxidative
processes of engine components, leading to possible exhaust gas emis-
sions such as carbon monoxide, one of those responsible for the aggra-
vation of the greenhouse effect [5]. The selective and sensitive
identification of iron has been of interest to many researchers, and
several detections and quantification methods have been developed, for
example, voltammetric, fluorimetric, and ICP-OES measurements
[6–10]. Fluorimetric protocols have played an important role in the
detection of metal ions due to their simplicity and high sensitivity
ꢀ 1
nization, it is found in levels from 0.5 to 50 mg L at natural waters [1].
The minimum daily requirement for iron ranges from about 10 to 50
mg/day [1]. A healthy adult has 40 to 160 mg of iron in their blood,
indexes above or below this are a warning sign. Despite this importance,
excess or deficiency of iron in specific types of cells or organs can cause
serious health problems, such as anemia, cirrhosis, diabetes, chronic
fatigue, and osteoporosis.
Another interesting aspect about iron is related to the use of ethanol
as a fuel in automobiles because it is considered an undesired inorganic
contaminant [2]. From the point of view of the quality of fuel ethanol,
metal ions are considered inorganic contaminants in any concentration,
since metallic species can accelerate the corrosion of engines or promote
the formation of gums and sediments. However, there are specifications
on ions such as iron (5 mg / kg), copper (0.07 mg / kg), sodium (2 mg /
kg), chloride (1 mg / kg) and sulfate (4 mg / kg) that can be observed in
quality control resolutions for fuel ethanol from regulatory agencies
such as ANP (Brazilian agency), which makes it mandatory to determine
3
+
[11–16]. Fluorescent sensors for Fe have been continuously devised
[17–20] but, so far, easy-to-make chemosensors, with a low limit of
detection and extensive work range are limited, and hence are still an
analytic challenge.
Fluorene and its derivatives are rigid aromatic molecules that
*
Corresponding author.
E-mail address: fsnunes@ufpr.br (F. Souza Nunes).
https://doi.org/10.1016/j.ica.2021.120511
Received 29 March 2021; Received in revised form 30 June 2021; Accepted 1 July 2021
Available online 6 July 2021
0
020-1693/© 2021 Elsevier B.V. All rights reserved.

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
fluoresce in the ultra-violet region with high quantum yield and, owing
to their numerous applications, have been the topic of studies for a long
time [21–25]. However, fluorene-derivatives that act as selective fluo-
at 400 MHz, with CHCl
3
, and tetramethylsilane (TMS) as an internal
reference. Chemical shifts are expressed in δ (ppm). Electronic spectra in
the UV–Vis range (190–820 nm) were obtained on a diode array Hewlett
Packard 8452A spectrophotometer in ethanol solutions using a 1.0 cm
path length quartz cell.
rescent chemosensors for the detection of metal ions, including Fe3
+
,
together with statistical tests of the detection methods, are not often
reported [26–31]. In this context, this work presents the investigation of
a turn-off chemosensor based on a fluorescent fluorene derivative
Fluorescence measurements and fluorescence quantum yields were
recorded at room temperature with a 1.0 cm optical path quartz cuvette
using a Shimadzu RF5301-PC spectrofluorimeter. The equipment was
set in between 1.5 and 5.0 nm slit width for excitation and emission
(
diethyl 2-(9-fluorenyl)malonate, DEFM, see Scheme 1) for detection of
3
+
Fe . The structural simplicity and undemanding preparation of DEFM,
together with the fact that this molecule has never been reported as a
fluorescent sensor, are noteworthy.
ꢀ 1
spectra, employing a 600 nm min scan rate and maximum wave-
lengths of excitation and emission at 296 nm and 316 nm, respectively.
DFT was carried out with the B3LYP functional and 6–311 g(d,p)
basis to all atoms. The solvent was included (ethanol) in the calculation
using SCRF formalism. Calculations were done using the Gaussian 09
[42]. Orbitals surfaces and population were obtained by Avogadro [43]
and GaussSum 3.0 [44], respectively.
Density functional theory (DFT and TD-DFT) calculations were car-
ried out and afforded a plausible explanation for the sensing mechanism.
Also, wide analytical parameters such as a calibration curve, limits of
detection and quantification, as well as, repeatability, intermediate
precision, robustness, and recovery testes confirmed the good applica-
bility for this chemosensor for Fe3 , compared to similar systems of
chemosensors [32–40]. Moreover, works that employed wide analytical
tests are rare in literature, making this work more applicable in real
systems.
+
2.4. Sensitivity and selectivity measurements, and merit parameters
The selectivity was tested from a stock solution of DEFM (10 mmol
ꢀ 1
L
) in ethanol, various dilutions were made to attain the working so-
2
. Experimental section
lution concentrations. The following species were used to evaluate the
+ + + + +
4 3 4
selectivity of the DEFM chemosensor: NH , [N(CH ) ] Li , Na , K ,
2
+
+
2+
3+
3+
3+
3+
3+
2+
2+
2+
2+
2+
2+
2
.1. Chemicals and materials
Mg , Ca , Eu , Tb , Ru , Co , Ni , Cu , Zn , Hg , Cd ,
2
3+
Mn , Fe , Cr
and Al they were accessed from the respective
ꢀ
1
The chemicals are from Sigma-Aldrich and used as supplied. Before
chloride salts in 10 mmol L alcoholic stock solutions.
The sensitivity of DEFM for Fe3 ions was determined from an
analytical curve, which includes calculation of limit of detection (LOD)
and quantification (LOQ). Merit parameters such as repeatability, in-
termediate precision, robustness, and recovery were calculated, using
statistical approaches, to evaluate the analytical method, such as
Grubbs, Fisher, Cochran, and Durbin-Watson tests [45–48]. Ethanol was
employed as a solvent in all tests. Samples were measured in triplicate
and standard variations were calculated from the results.
+
being used, all glassware was calibrated, washed by soaking in freshly
prepared aqua regia (1HNO /3HCl (v/v)) and then with ultrapure
water, and finally dried in air.
3
2
.2. Preparations
Diethyl 2-(9-fluorenyl) malonate (DEFM) was prepared as described
elsewhere, Scheme 2 [41]. The yield was 62% (8.02 g). Elemental
ꢀ 1
analysis found (calculated) for DEFM, C20
4.10 (74.06); H% 6.27 (6.21); N% 0.10 (0.00). ESI-MS (positive mode)
H O , 324.14 g mol : C%
20 4
3
. Results and discussion
7
+
1
at m/z – found (calculated): 325.14 (325.15) (molecular íon-H ). The H
NMR, C NMR, DEPT-135, and ESI-MS (including simulated) spectra
are shown in the Supplementary Material, Figures SM1–SM3, with the
appropriate assignments.
1
3
3.1. Photophysics of the sensor DEFM
The photophysical and selectivity properties of sensor DEFM were
carried out from alcoholic solutions. Absorption, excitation, and emis-
sion spectra are in Fig. 1. UV–vis spectra showed intraligand
π
- * and n-
π
2
.3. Apparatus
π
* transitions in the
50–400 nm range, with maximum molar absorptivities of 1.21 ×
2
CHN analyses were carried out with a Perkin-Elmer 2400 analyzer.
5
ꢀ 1
ꢀ 1
3
ꢀ 1
ꢀ 1
1
0 L mol cm at 268 nm and of 8.65 × 10 L mol cm at 302 nm.
Mass spectra were measured in high-resolution ESI-MS on a microTOF
QII mass spectrometer (Bruker Daltonics, Billerica) from solutions in
When excited at 296 nm, which corresponds to the population of the first
excited state S
efficiency.
0
1
→ S , the sensor emitted at 316 nm, with the highest
1
13
ethanol. H and C NMR spectra were recorded on a Bruker Avance HD
3
3
.2. Fluorimetric evaluation towards cations
.2.1. Individual and collective selectivity
The majority of the tested cations did not change significantly the
fluorescence of sensor DEFM, except for Fe3 which caused a strong
suppression of 99% as observed in Fig. 2(top), suggesting the formation
of a complex formed in solution with high specificity for this metal ion
and low values of the standard deviation. Individual selectivity was
+
ꢀ 1
investigated upon quantitative addition of 100 mol L solutions of
μ
metal ions to equimolar solutions of the sensor DEFM.
In a collective selectivity test (Fig. 2 bottom), all cations used in the
individual selectivity tests were tested together as potential in-
3
+
terferences for the determination of Fe
.
It shows the high specificity of
the chemosensor towards Fe3 ions since the set of cations did not
+
3
+
interfere significantly with the suppression of the Fe -DEFM complex.
Furthermore, notice that Fe3 produces a 99% quenching of fluores-
+
Scheme 1. Representation of the chemosensor DEFM.
cence of the sensor, while it was reduced to 86% in presence of 21 metal
2

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
Scheme 2. (Top) Representation of the PET effect that quenches the fluorescence of the sensor DEFM upon coordination with Fe3+, high spin, 2S+1 = 6. (Bottom)
Illustrative scheme of the PET redox process for the complex with 99% of quenching.
Fig. 1. Absorbance (blue line), excitation (red line) (λex = 296 nm) and
ꢀ
1
emission (black line) (λem = 316 nm) spectra of sensor DEFM in 100
μmol L
and ethanol. Excitation and emission slits are both 3 nm. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web
version of this article.)
cations. This outcome reflects multiple, simultaneous equilibria of
complex formation that are established in the presence of the different
cations used. Under this condition, the concentration of [Fe
+
(
DEFM)
2
Cl
2
]
in solution is reduced, which accounts for the increase of
the fluorescence, compared to that observed in the competition tests
Fig. 2 top), carried out in the presence of individual ions. Testing the
(
selectivity of DEFM in the presence of several different contaminants
offers the best simulation of a potential application of the sensor in real
samples, as opposed to the experiment in the presence of individual
cations.
Finally, in the collective selectivity test, upon adding DEFM and 21
3
+
metal cations (except Fe ), the fluorescence intensity goes down by
3
+
5
7%. While the metal cation (other than Fe ) that lowers the fluores-
3
+
cence intensity the most is Ru by ~ 30%. The reason for that behavior
is directly associated with the stability of the metal complexes formed in
solution, which is given by their specific formation constants. In the
presence of a mixture of metal ions, it is expected the formation of
multiple equilibria. The particular stability of each complex influences
the energy of the frontier orbitals, and consequently the ability to affect
the PET mechanism. Thus, the suppression of the luminescence intensity
is, indeed, expected to be different in the presence of a single metal ion
of specific nature than in the presence of different cations.
Fig. 2. (Top) Individual selectivity test: relative fluorescence of sensor DEFM in
ꢀ 1
the presence of equimolar concentrations (100 mol L ) of several cations in
μ
ethanolic solutions. Results are the average of triplicate measurements. Exci-
tation (at 296 nm) and emission (at 316 nm) slits of 3 and 5 nm, respectively.
(
(
Bottom) Collective selectivity test, in the presence of equimolar concentrations
ꢀ 1
100
μ
mol L ) of all cations and sensor DEFM, in ethanol. Excitation and
emission slits of 1.5 nm.
ratio as seen in Fig. 4(top). The high value of the formation constant
supports a strong affinity of the sensor to the Fe3 ion.
+
.2.2. Stoichiometry and stability of the Fe3+-DEFM complex
3
The stoichiometry was determined by Job’s plot for absorbance as
1
1
0
1
ꢀ F
=
+
(1)
seen in Fig. 3. The method showed a maximum absorption when the
n
F ꢀ F
0
(F
∞
ꢀ F
)K[Q]
(F
∞
0
)
molar fraction of Fe3 is close to 0.3, in accordance with the formation of
+
a complex in a 1:2 (metal:sensor) ratio.
where: F and F = fluorescence of the sensor in the absence and presence
o
3
+
Benesi-Hildebrand [49], equation (1), allowed access to the binding
constant for the complex, 1.5x106 L2 mol-2, for a 1:2 (metal:sensor)
of quencher; [Q] = quencher concentration of Fe ion; n = relation of
3
+
Fe ion and sensor (=0.5); K = binding constant of complex formed.
3

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
Interaction of Fe3 ion with the sensor DEFM was studied using
+
Stern-Volmer’s formalism [50] in accordance with a 1:2 (metal:sensor)
3
+
]
stoichiometry given by the equation (2). A plot of F-F
o
versus [Fe
2
ꢀ 7
exhibited a linear fit with an R of 0.9942, from 1.00 to 6.00 × 10 mol
ꢀ 1
3+
L
of Fe concentration. The linearity of the Stern-Volmer plot in-
dicates a dynamic quenching mechanism, however the proximity be-
6
2
ꢀ 2
6
2
tween the values of KSV and K at 1.0 × 10 L mol and 1.5 × 10 L
ꢀ 2
mol respectively, also suggests some static suppression during the
formation of the complex in the ground state.
F
F
0
n
=
1 + k
q
τ
f
[Q] = 1 + KSV [Q]
(2)
where: F
o
and F = fluorescence of the sensor in the absence and presence
3
+
of quencher, respectively; [Q] = quencher concentration of Fe ; n =
molar relation between Fe3 ion and sensor (=0.5); k
+
= bimolecular
q
quenching constant,
τ = lifetime of fluorescence in the absence of the
f
quencher; KSV = Stern-Volmer’s suppression constant, which measures
Fig. 3. Job’s plot of the variation of the absorption at 266 nm for sensor DEFM
the metal:sensor affinity.
3
+
ꢀ 1
and Fe in a total molar concentration of 10 mol L in ethanol.
μ
Fig. 4. (Top) Benesi-Hildebrand’s plot for calculation of binding constant (K). (Bottom) Stern-Volmer’s plot for calculation of suppression constant (KSV).
4

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
3
.2.3. Theoretical calculations
alternative mechanism for the quenching of fluorophores [54]. Inter-
3
+
To better understand the electronic processes in molecular systems,
estingly to note that for a ground state with 2S + 1 = 2 (Fe , low spin),
the excitation of at the same wavelength (296 nm) excites an electron
from the HOMO-12(-8.57 eV)(centered in the chloride ligand) to the
such as emission and absorption processes, theoretical calculations are a
valuable tool, that permits to access the energy diagram and their mo-
lecular orbitals involved in electronic transitions. The optimized ge-
ometries of the sensor and the Fe3 complex, named [FeCl
2
3+
LUMO(-3.57 eV)(which has significant contribution of the dz of Fe
,
+
+
2
(DEFM)
2
]
Fig. SM5). In that case, thus, the static quenching is associated with a
LMCT transition.
with octahedral configuration, are shown in Fig. 5 (the coordinates of
the optimized geometries are showed in Table SM1) along with the
energy diagram in Fig. 6, and the most important electronic transitions
are summarized in Table 1.
According to the ligand field applied by the ligands, the Fe3 com-
plex can present spin multiplicities of doublet and sextet (ground state),
and quartet in excited states, depending on the different possibilities of
matching the 5 electrons in the d orbitals. The theoretical calculations
performed (under the unrestricted mode) investigated the absorption
profile (Fig. SM4), from the first 150 transitions, simulating the three
possibilities of spin state, and showed that the electronic energy of the
3
3
.3. Evaluation of the analytical method
+
.3.1. Calibration curve and the sensibility of sensor
Limit of detection (LOD) (3 s/l, where: s = standard deviation; I =
slope of the line) and limit of quantification (LOQ = 10 s/l) were
calculated from the analytical curve shown in Fig. 7. LOD and LOQ were
ꢀ 1
found at 15.56 and 51.85 nmol L , respectively, within linearity of
ꢀ 1
2
0
.1–1.0
μ
mol L (R = 0.997, least squares method). The quality of the
3+
correlation between the concentration of the Fe ion and the intensity
of emission were checked by several statistical tests (Grubb’s, Fisher’s,
Cochran’s, Durbin-Watson’s test) described in the Supplementary sec-
tion of the manuscript (Tables SM3-SM6, equations SM1-SM4). The tests
were approved and recommended by regulatory agencies [45–48,58].
ꢀ
1
sextet state is about 85.02 kJmol less than the first quartet state, and
ꢀ
1
8
2.50 kJmol
more stable than the doublet state. This way, the
following discussion about the observed suppression phenomena will be
referred to the sextet state.
+
2 2
Also, it was investigated the stability of complex [FeCl (DEFM) ]
based on the free energy calculations obtained from optimized geome-
tries of reactants and products, and the results revealed a free energy of
3
.3.2. Repeatability and intermediate precision tests
Results of the repeatability and intermediate precision tests
ꢀ 1
+
–
1239 kJ mol for [FeCl
It should be pointed out the good correspondence between the
2 2
(DEFM) ] .
[
45–48,58] are summarized in Table SM7 for the lower level of con-
3+
centration for detection of Fe
ions with precision and accuracy,
calculated and the experimental absorption spectra (Table 1) of the free
sensor (Fig. 1) and its complex (Fig. SM4) that show mainly intraligand
transitions. An inspection of the molecular orbital diagram of DEFM in
Fig. 6 shows that
–1
therefore using the limit of quantification (~51.85 nmol L ), using ten
replicate measurements, Fig. 8(left). The intermediate precision tests
out the accordance of the results from the same laboratory (repeatability
and analyst 1) but performed by a different analyst (analyst 2). All the
samples showed a HorRat ratio (RSD/RSD Horwitz) lower than 1, and
thus approved in the test. The Fisher’s test also gives that Fcalc (analyst 1;
analyst 2) = 0.52, which is lower than Fcritical = 4.03, hence supporting
that the analytical method is reproducible and repeatable.
HOMO(-6.23 eV) and LUMO + 3(-021 eV) are mainly centered on the
fluorene unit, thus the excitation at 296 nm, through a
π
- * transition is
π
followed by the reverse process, which accounts for the fluorescence of
+
2 2
the free sensor. In contrast, excitation of the complex [FeCl (DEFM) ]
at 296 nm is associated with the transition HOMO–12(–9.05 eV) →
LUMO + 3(-1.59 eV) (the first has major contributions are from malo-
nate and the chloride ions, while the second is centered in the malonate
unit). Therefore, after the photo-excitation of the complex at 296 nm, an
electron from HOMO(-6.93 eV) can be transferred to HOMO-12(–9.05
eV), through the photoinduced electron transfer (PET) mechanism
3
.3.3. Robustness of sensor DEFM
The robustness was evaluated by Youden’s test [45–48,58] from
fractional factorial planning using 8 different combinations of seven
nominal experimental conditions, duplicate measurements (see
Tables SM8-SM10). So each condition was varied 4 times in four
different combinations, and with that, it was possible to calculate the
error of the measurements in the various conditions (equation (3)).
These results are gathered within Pareto’s graph in Fig. 8(right); as well
as the calculation of the significance line, which corresponds to the
maximum deviation allowed (equation (4)).
[
51,52] (See Scheme 3), drastically decreasing the fluorescence in-
tensity observed, also known as an ON–OFF fluorescent switch [53–57].
The complete set of orbitals for both DEFM and its Fe3 can be found in
+
the Supplementary Material Section, Table SM2)
Thus, coordination of the sensor with Fe3 inhibits the back electron
transfer observed in the free ligand from LUMO + 3 to HOMO orbitals,
thus turning off the fluorescence [53]. Besides this static quenching,
+
The chart shows the quality deviation for each condition within
Youden’s test, and showed no relevant significance, since they are below
the maximum deviation calculated at 0.400 thus indicating the high
robustness of the method.
3
+
Fe can provoke a nonradiative deactivation of the sensor, as suggested
by the Stern-Volmer analysis (Fig. 4), and has been proposed as an
+
Fig. 5. Optimized geometries for DEFM chemosensor and their complex formed in solution, [FeCl
2
(DEFM)
2
] .
5

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
+
2 2
Fig. 6. Frontier molecular orbitals of the chemosensor DEFM and [FeCl (DEFM) ] .
√
̅̅̅̅̅̅̅̅̅̅̅̅
Table 1
2
4
∗ s
n
+
Significance line = tcritical
∗
(4)
TD-DFT electronic transitions in DEFM and [FeCl
2
(DEFM)
2
] .
Major contribution
N
Compound
Wavelength / nm
experimental)
Osc.
(
Strength
where: tcritical = t student (2.36); s = standard deviation; N = 16.
DEFM
272 (302)
0.42
0.21
HOMO → LUMO
HOMO → LUMO +
3
3
.3.4. Reversibility test
2
17 (290)
To study the reversibility of the sensor DEFM towards Fe3 ions in
+
1
93 (266)
0.75
0.09
0.10
0.27
0.66
HOMO-1 → LUMO
ethanolic solutions, the fluorescence intensity of DEFM after successive
+
3
3+
addition of Fe ions and EDTA was measured out (Fig. 9). The addition
+
[
FeCl
2
(DEFM)
2
]
347 (320)
HOMO-13 →
LUMO
of one equivalent of Fe3 to a solution of DEFM, resulted in a significant
+
2
2
2
86 (290)
61 (302)
67 (266)
HOMO-12 →
LUMO + 3
HOMO-13 →
LUMO + 4
HOMO → LUMO +
5
decrease of fluorescence intensity, because of the established equilib-
+
rium in favor of the formation of the [FeCl
2
(DEFM)
2
]
complex. After
the addition of one equivalent of EDTA to the system, the fluorescence of
3+
the solution increases due to the binding of EDTA to the free Fe , while
3
+
DEFM was released to the solution. The alternate addition of Fe and
EDTA to DEFM solution produced an on/off fluorescence response for at
least two cycles, illustrating thus, the satisfactory reversibility of sensor
Considering that out of the 8 combinations performed for each
3+
DEFM with Fe ions.
investigated parameter, four were made with the nominal condition and
four with the varied condition, then it was possible to apply the Fisher’s
test which compares the variances of both sets of data (equation SM2).
For all nominal condition the result the value of Fcalc (Table SM11) was
less than Fcritical = 15.44, considering a 95% of confidence level, hence
supporting that the analytical method is robust.
3
.3.5. Practical application
The best performance of the DEFM sensor was observed when it is
dissolved in ethanol. Therefore, recovery tests were carried out using
ethanol fuel for automobiles from a gas station as a matrix, since it is
3+
known that it is contaminated with undesired Fe ions. The samples
were prepared as follows: To a 5.0 mL volumetric flask were added 50
μ
L
A + B + C + D E + F + G + H
Effect of the tem perature =
-
(3)
ꢀ 1
of the stock solution of the sensor DEFM in ethanol (10 mmol L ) and
4
4
–
1
5
0
μ
L of a 500
μ
mol L stock solution of ferric chloride in ethanol (10
ꢀ 1
μ
mol L ). The final volume was completed to 5.0 mL with the matrix,
6

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
Scheme 3. Chemical equation for the preparation of DEFM.
and all measurements were triplicated. Results are gathered in Table 2,
where satisfactory recovery values were observed in the range of
sensor binding constant. Moreover, it does not suffer interference from
other ions and exhibits one of the best linear ranges. Also, it is the only
study that presents merit parameters, such as repeatability, intermediate
precision, robustness, and recovery that allows a thorough evaluation of
the analytical usage of the DEFM sensor towards metal ions sensing.
4
6–547% [45–48,58].
3
.4. Comparison with similar sensors
4
. Conclusions
Table 3 shows analytical data from some recently published sensors
for Fe3 . In all samples, the sensors are active in organic/inorganic
media or a mixture of solvents. Among those examples, DEFM has the
lowest limit of detection, low relative error, and the highest metal-to-
+
A new sensor for selective and specific detection of Fe3 was tested
+
out for measurements in ethanolic solutions. Complex formation with
7

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
Fig. 7. Analytical curve for sensor DEFM. Results are the average of triplicate
measurements. Excitation and emission slits are 3 and 5 nm, respectively.
Fig. 9. Reversible fluorescence changes of sensor DEFM (100
μ
M; emission in
3
+
3
16 nm) in ethanol solution after successive addition of Fe (0.5
μ
M) and
Fe3 led to a strong fluorescent quenching (99%), good linearity, and
+
EDTA (0.5
μ
M). Excitation and emission slits are both 1.5 nm.
limit of detection of
ꢀ
1
1
5.56 nmol L . Job’s and Benesi-Hildebrand’s method relationships
suggested a 1:2(metal:sensor) stoichiometry, with a fairly high associ-
Table 2
6
2
ꢀ 2
ation constant (K = 1.5x10 L mol ) reported for similar systems, thus
3+
Recovery values (%) from solutions of DEFM and Fe in ethanol fuel.
proving the strong sensor:analyte affinity. Besides, Stern-Volmer’s
3
+
ꢀ 1
)
MATRIX
[Fe ] (nmol L
RECOVERY (%)
6
2
ꢀ 2
constant (KSV = 1.0x10
L mol ) suggests a dynamic quenching,
ETHANOL FUEL
100
45.99 ± 0.15
96.60 ± 0.32
547.23 ± 0.88
though the proximity of K and KSV suggests some static suppression due
to the complex formed in solution. Theoretical calculations suggested
that the photoelectron transfer (PET) mechanism can explain the fluo-
rescent suppression observed, a process which is associated with an
intraligand charge transfer (ICT) and a LMCT, respectively, for the
5
1
00
000
CRediT authorship contribution statement
3
+
sextet, and dublet ground states of the Fe complex. The fluorimetric
method for detection and quantification of Fe3 was further studied to
check its quality and analytical applicability. The excellence of the
analytical curve was confirmed through the application of a series of
statistical tests. Furthermore, the practicability of the method passed the
repeatability as well as intermediate precision, robustness, and recovery
tests from fuel ethanol as a matrix. Finally, the sensor showed no com-
mon interfering ions, thus showed that the sensor DEFM can be suc-
+
Fabiane Santos Carlos: Methodology, Investigation, Formal anal-
ysis, Visualization, Supervision. Letícia Aparecida Silva: Investigation,
Validation. Cristiano Zanlorenzi: Software, Investigation, Visualiza-
tion. F a´ bio Souza Nunes: Conceptualization, Resources, Writing - re-
view & editing, Funding acquisition.
Declaration of Competing Interest
cessfully used to determine Fe3 ions.
+
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
3
+
Fig. 8. (left) Repeatability and intermediate precision tests of Fe ions. Colors refer to two different analysts and error bars apply to ten replicates. Slit is 3 nm.
3
+
–1
[
Fe ] = LOQ. [DEFM] = 10
μ
mol L ; ethanol medium. (right) Pareto’s graph, showing the changes in the measurement of experimental conditions for chemosensor
◦
DEFM. Conditions are: 1) temperature during measurement (room temperature and 10 C), 2) reagent supplier (SIGMA and MERCK), 3) time between sample
◦
preparation and measurement (2 and 1.5 h), 4) solvent supplier (DINAMICA and BIOTEC), 5) sample storage before measurements (room temperature and 15 C), 6)
purge of the sample with inert gas (yes and no), 7) cleaning the cuvette between replicates (2xethanol + 1xacetone; versus only 1xacetone). The concentration of
3
+
-4
-6
ꢀ 1
DEFM and Fe ions were constant at 1.00x10 and 1.00x10 mol L , respectively.
8

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
Table 3
[13] M.C. Nunes, F.S. Carlos, O. Fuganti, F.S. Nunes, A Facile Preparation of a New
Water-Soluble Acridine Derivative and Application as a Turn-off Fluorescence
Chemosensor for Selective Detection of Hg2+, J. Fluoresc. 30 (2020) 235–247.
Comparison between some fluorescent sensors for Fe3+. LOD = limit of detec-
tion. MP = merit parameters.
Sensor
[
14] M.C. Nunes, F.S. Carlos, O. Fuganti, D.D.M. Galindo, L.D. Boni, G. Abate, F.
S. Nunes, Turn-on fluorescence study of a highly selective acridine-based
chemosensor for Zn2+ in aqueous solutions, Inorg. Chim. Acta 499 (2020)
LOD
Linear Range
Interfering
Ions
Medium
ethanol
logK
6.0
MP
Yes
ꢀ
1
(μ
mol L
)
1
19191–119200.
DEFM
15.56
nmol
0.1–1.0
none
none
none
[15] F.S. Carlos, C. Zanlorenzi, L.A. Silva, F.S. Nunes, A novel macrocycle acridine-
based fluorescent chemosensor for selective detection of Cd2+ in Brazilian
sugarcane spirit and tobacco cigarette smoke extract, Inorg. Chim. Acta 508 (2020)
119634–119641.
ꢀ
1
L
[
[
28]
34]
2.22
200–1000
H
2
O/tris-
–
–
No
No
[
16] F.S. Carlos, M.C. Nunes, B.A. Iglesias, P.J. Zambiazi, M. H o¨ rner, G.A. Moraes, G.
S. Machado, R.R. Ribeiro, F.S. Nunes, Fluorenyl-Schiff-base ligands and their
dicopper(II) complexes. Synthesis, structural and spectroscopic characterization
and DNA binding assays, Polyhedron 144 (2018) 18–29.
μ
L
–
mol
HCl
ꢀ
1
–
H
2
O /DMF
5:5 v/v)
THF
O/
(
[
[
17] W.E. Jones, L.-J. Fan, A Highly Selective and Sensitive Inorganic/Organic Hybrid
Polymer Fluorescence “Turn-on” Chemosensory System for Iron Cations, J. Am.
Chem. Soc. 128 (2006) 6784–6785.
2
+
[
[
32]
35]
–
0.29
–
Cu
–
4.8
No
No
3
+
3+
20–100
Al , Cr
H
2
μ
mol
MeOH(1:1
v/v)
18] V. Safarifard, Y.D. Farahani, Highly selective detection of Fe3+, Cd2+ and CH2Cl2
based on a fluorescent Zn-MOF with azine-decorated pores, J. Solid State Chem.
275 (2019) 131–140.
ꢀ
1
L
[
[
19]
18]
275
0.1–50
40–160
none
H
2
O
–
–
No
No
nmol
[19] Z. Mu, J. Hua, S. Feng, Y. Yang, A ratiometric fluorescence and light scattering
sensing platformbased on Cu-doped carbon dots for tryptophan and Fe(III),
Spectrochim. Acta A 219 (2019) 248–256.
ꢀ
1
L
2
+
+
2+
0.5
Ni , Co ,
DMF
2
μ
mol
Mn
[20] S. Ghosh, R. Manna, S. Dey, Epoxy-based polymer incorporating 1-naphthylamine
and sebacic acid moieties: A selective fluorescent sensor for ferric ions, J. Mol.
Struct. 1180 (2019) 406–410.
ꢀ
1
L
[
[
[
21] Rodrigues, P. C.; Fontes, B. D.; Torres, B. B. M.; Sousa, W. S.; Faria, G. C.; Balogh, D.
T.; Faria, R. M.; Akcelrud, L. Synthesis of a PPV-fluorene derivative: Applications in
luminescent devices. J. Appl. Polym. Sci. 2015, 42579-42586.
the work reported in this paper.
22] X. Cao, A. Gao, N. Zhao, F. Yuan, C. Liu, R. Li, Surfaces wettability and morphology
modulation in a fluorenederivative self-assembly system, Appl. Surf. Sci. 368
Acknowledgements
(
2016) 97–103.
23] K. Shanmugasundaram, M.S. Subeesh, C.D. Sunesh, R.K. Chitumalla, J. Jang,
Y. Choe, Synthesis and photophysical characterization of an ionic fluorene
derivative for blue light-emitting electrochemical cells, Org. Electron. 24 (2015)
Brazilian Research Council CNPq supported this work. F.S.N. and F.
S.C. thank CNPq and CAPES for research fellowships. We gratefully
thank Professor Leni Campos Akcelrud for allowing access to the spec-
trofluorimeter at Paulo Scarpa Laboratory, UFPR.
2
97–302.
[
[
24] K. Baumg ¨a rtner, F. Rominger, M. Mastalerz, An Oxidative Macrobicyclic Ring
Opening of a Triptycene to a Highly Functionalized Fluorene Derivative, J. Org.
Chem. 80 (2015) 8881–8886.
25] S.A. Kurhuzenkau, A.W. Woodward, S. Yao, K.D. Belfield, Y.O. Shaydyuk, C. Sissa,
M.V. Bondar, A. Painelli, Ultrafast spectroscopy, superluminescence and
theoretical modeling for a two-photon absorbing fluorene derivative, Phys. Chem.
Chem. Phys. 18 (2016) 1239–1246.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.ica.2021.120511.
[
26] J.L. Pablos, S. Ibeas, A. Mu n˜ oz, F. Serna, F.C. García, J.M. García, Solid polymer
and metallogel networks based on a fluorene derivative as fluorescent and
colourimetric chemosensors for Hg(II), React. Funct. Polym. 79 (2014) 14–23.
References
[27] Q. Li, Y. Jia, Z. Feng, Liu; F., A highly sensitive and selective fluorescent probe
without quencher for detection of Pb2+ ions based on aggregation-caused
quenching phenomenon, RSC Adv. 8 (2018) 38929–38934.
[
[
1] World Health Organization (WHO). Guidelines for Drinking-water Quality
incorporating the 1st and 2nd addenda, 3rd ed.; WHO: Geneva, 2008.
2] R.M. Takeuchi, A.L. Santos, N.R. Stradioto, P.M. Padilha, Copper determination in
ethanol fuel by differential pulse anodic stripping voltammetry at a solid paraffin-
based carbon paste electrode modified with 2-aminothiazole organofunctionalized
silica, Talanta 71 (2007) 771–777.
[
[
[
28] L. Li, Q. Chen, Z. Niu, X. Zhou, T. Yanga, W. Huang, Lanthanide metal–organic
frameworks assembled from a fluorene-based ligand: selective sensing of Pb2+ and
Fe3+ ions, J. Mater. Chem. C 4 (2016) 1900–1905.
29] A. Hens, K.K. Rajak, Selective fluorometric detection of F-1 and Zn(II) ions by a N,
O coordinating sensor and naked eye detection of Cu(II) ions in mixed-aqueous
solution, RSC Adv. 5 (2015) 44764–44777.
[
[
3] Ag ˆe ncia Nacional Do Petr o´ leo (ANP). Especificaç o˜ es do etanol anidro combustível
30] K. Belfield, M.V. Bondar, A. Frazer, A.R. Morales, O.D. Kachkovsky, I.A. Mikhailov,
A.E. Masunov, O.V. Przhonska, Fluorene-Based Metal-Ion Sensing Probe with High
Sensitivity to Zn2+ and Efficient Two-Photon Absorption, J. Phys. Chem. A 114
(
EAC) e do etanol hidratado combustível (EHC); Brasil, 2015.
4] A.L. Santos, R.M. Takeuchi, N.R. Stradioto, L. Angnes, R.A.A. Mu n˜ oz,
Electrochemical Determination of Inorganic Contaminants in Automotive Fuels,
Electroanalysis 24 (2012) 1681–1691.
(
2010) 9313–9321.
[
[
[
[
[
31] J. Cao, Y. Li, J. Feng, 9-Benzylidene-9H-fluorene Derivatives Linked to Monoaza-
[
[
5] R.C.C. Pereira, V.M.D. Pasa, Effect of Alcohol and Copper Content on the Stability
of Automotive Gasoline, Energy Fuels 19 (2005) 426–432.
1
1
5-crown-5: Synthesis and Metal Ion Sensing, Chin. J. Chem. Phys. 30 (2012)
571–1574.
6] J. Hou, Z. Sun, Z. Yang, L. Zhao, Y. Zhang, Y. Li, L. Ding, Microwave-assisted
fabrication of multicolor photoluminescent carbon dots as a ratiometric
fluorescence sensor for iron ions, New J. Chem. 43 (2019) 853–861.
7] M. Ak, R. Ayranci, An Electrochemical Sensor Platform for Sensitive Detection of
Iron (III) Ions Based on Pyrene-Substituted Poly(2,5-dithienylpyrrole),
J. Electrochem. Soc. 166 (2019) B291–B296.
32] P. Li, Y. Zhao, L. Yao, H. Nie, M. Zhang, A simple, selective, fluorescent iron(III)
sensing material based on peripheral carbazole, Sensor Actuat. B-Chem. 191
(
2014) 332–336.
[
[
[
33] C.-S. Wu, Y.-J. Lin, Y. Chen, A fluorene-based material containing triple azacrown
ether groups: synthesis, characterization and application in chemosensors and
electroluminescent devices, Org. Biomol. Chem. 12 (2014) 1419–1429.
34] C. Yi, B. Song, W. Tian, X. Cui, Q. Qi, W. Jiang, Z. Qi, Y. Sun, Fluorescent sensor of
fluorene derivatives having phosphonic acid as a fluorogenic ionophore: synthesis
and static quenched properties for Fe(III), Tetrahedron Lett. 55 (2014) 5119–5123.
35] R. Alam, R. Bhowmick, A.S.M. Islam, A. Katarkar, K. Chaudhurib, M. Ali,
A rhodamine based fluorescent trivalent sensor (Fe3+, Al3+, Cr3+) with potential
applications for live cell imaging and combinational logic circuits and memory
devices, New J. Chem. 41 (2017) 8359–8369.
8] A.L. Saber, A.M. Hameed, A.A. Saygal, H. Alessa, A. Alharbi, Iron-selective Poly
(
Vinyl Chloride) Membrane Electrode Based on Norfloxacin as a Neutral Carrier,
Int. J. Electrochem. Sci. 13 (2018) 10076–10087.
9] Z.-A. Zonga, C.-B. Fana, C.-F. Bia, X. Zhanga, R. Luoa, X.-M. Mengc, F. Jing, Y.-
H. Fana, Highly selective functional luminescent sensor toward Cr(VI)/Fe(III) ion
and nitrobenzene based on metal–organic frameworks: Synthesis, structures, and
properties, J. Solid State Chem. 270 (2019) 651–665.
[
10] A.-Z. Hossein, Z.-K. Monireh, A turn-on/off fluorescent sensor based on nano-
structured Mg-Al layered double hydroxide intercalated with salicylic acid for
monitoring of ferric ion in human serum samples, Anal. Chim. Acta 1061 (2019)
[
[
36] P. Li, Y. Wang, A new fluorescent sensor containing glutamic acid for Fe3+ and its
resulting complex as a secondary sensor for PPi in purely aqueous solution, New J.
Chem. 41 (2017) 4234–4240.
1
52–160.
37] C.-J. Hua, H. Zheng, K. Zhang, M. Xin, J.-R. Gao, Y.-J. Li, A novel turn off
fluorescent sensor for Fe(III) and pH environment based on coumarin derivatives:
the fluorescence characteristics and theoretical study, Tetrahedron 72 (2016)
[
[
11] F.S. Carlos, M.C. Nunes, L.D. Boni, G.S. Machado, F.S. Nunes, A novel fluorene-
derivative Schiff-base fluorescent sensor for copper(II) in organic media,
J. Photochem. Photobiol. A 348 (2017) 41–46.
8
365–8372.
12] F.S. Carlos, C. Zanlorenzi, R.F. Monteiro, L.A. Silva, F.S. Nunes, A highly selective
acridine-based fluorescent probe for detection of Al(III) in alcoholic beverage
samples, Spectrochim. Acta A 231 (2020) 118–124.
[
38] L.-Q. Li, Z.-H. Liu, A Colorimetric and Fluorescent Sensor for Iron Recognition
Based on Rhodamine Derivative, J. Fluoresc. 27 (2017) 427–431.
9

F. dos Santos Carlos et al.
Inorganica Chimica Acta 526 (2021) 120511
[
[
39] S. Hua, S. Zhang, C. Gao, C. Xu, Q. Gao, A new selective fluorescent sensor for Fe3+
based on a pyrazoline derivative, Spectrochim. Acta A 113 (2013) 325–331.
40] M. Zhang, Y. Gao, M. Li, M. Yu, M. Li, L. Li, M. Zhu, J. Zhang, T. Yia, C. Huanga,
A selective turn-on fluorescent sensor for Fe(III) and application to bioimaging,
Tetrahedron Lett. 48 (2007) 3709–3712.
[48] D.N. Rutledge, A.S. Barros, Durbin-Watson statistic as a morphological estimator of
information content, Anal. Chim. Acta 454 (2) (2002) 277–295.
[49] H.A. Benesi, J.A. Hildebrand, A Spectrophotometric Investigation of the Interaction
of Iodine with Aromatic Hydrocarbons, J. Am. Chem. Soc. 71 (1949) 2703–2704.
[50] D.M. Roundhill (Ed.), Photochemistry and Photophysics of Metal Complexes,
Springer US, Boston, MA, 1994.
[51] B. Valeur, Molecular Fluorescence: Principles and Applications, Wiley-VCH,
Weinheim, 2001.
[52] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd ed., Cambridge
University Press, Cambridge, 1999.
[
[
[
[
41] Q.-H. Luo, L.-J. Jiang, C.-Y. Duan, M.-C. Shen, H.-W. Hu, Y.-J. Liu, A new
dioxotetraamine ligand appended with fluorenyl and its copper(II) complex.
Synthesis, crystal structure and solution behavior, Inorg. Chim. Acta 295 (1999)
4
8–55.
42] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
GAUSSIAN (2009).
[53] A. Shylaja, S.S. Roja, R.V. Priya, R.R. Kumar, Four-component Domino Synthesis of
Pyrazolo[3,4-h]quinoline-3-carbonitriles: “Turn-Off” Fluorescent Chemosensor for
3
+
Fe Ions, J. Org. Chem. 83 (2018) 14084–14090.
3
43] Marcus D Hanwell, Donald E Curtis, David C Lonie, Tim Vandermeersch,
Eva Zurek, Geoffrey R Hutchison, An advanced semantic chemical editor,
visualization, and analysis platform, J. Cheminformatics 4 (1) (2012), https://doi.
org/10.1186/1758-2946-4-17.
44] N.M. O’boyle, A.L. Tenderholt, K.M. Langner, cclib: a library for package-
independent computational chemistry algorithm, J. Comput. Chem. 29 (2008)
[54] R. Azadbakht, N. Chehri, A new fluorescent macrocyclic nano-chemosensor for Fe
+
-
and I in aqueous solution, New J. Chem. 42 (2018) 17690–17699.
[55] F.N. Moghadam, M. Amirnasr, K. Eskandari, S. Meghdadi, A new disulfide Schiff
base as versatile “OFF-ON-OFF” fluorescent-colorimetric chemosensor for
-
3+
sequential detection of CN and Fe ions: Combined experimental and theoretical
studies, New J. Chem. 43 (2019) 13536–13544.
[56] S. Manigandan, A. Muthusamy, R. Nandhakumar, C.I. David, Recognition of Fe3
by a new azine-based fluorescent “turn-off” chemosensor and its binding mode
analysis using DFT, J. Mol. Struct. 1208 (2020), 127834.
+
8
39–845.
[
[
[
45] S.L.R. Ellison, V.J. Barwick, T.J.D. Farrant, Practical Statistics for the Analytical
Scientist - A Bench Guide, 2nd ed.;, Royal Society of Chemistry:, Cambridge, 2009.
46] J.N. Miller, J.C. Miller, Statistics and Chemometrics for Analytical Chemistry 5th
ed., 2005. Edinburgh, Pearson - Prentice Hall Ptr.
[57] V. Dangi, M. Baral, B.K. Kanungo, Photophysical Studies of a Catechol Based
Polyfuncional Dipodal Chelator: Aplicaton ofr Optical Proble for Selective
Detection of Fe(III), J. Fluoresc. 30 (2020) 1131–1149.
47] D.B. Hibbert, J.J. Gooding, Data analysis for chemistry - an introductory guide for
students and laboratory scientists, Oxford University Press: Printed in the United
States of America, 2006.
[58] Anvisa (Ag ˆe ncia Nacional de Vigil aˆ ncia Sanit a´ ria). Guia para tratamento estatístico
da validaç ˜a o analítica, 2018.
1
0