Food Chemistry 420 (2023) 136138

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Analytical Methods

Diphenyl imidazole-based fluorescent chemosensor for Al3+ and its Al3+

complex toward water detection in food products
Yucun Liu a, Ning Li a, Yongling Zhang c, Yuan Wang b, *
a
College of Chemistry, Jilin Normal University, Siping 136000, China
b
Beijing National Laboratory for Molecular Science, CAS Key Laboratory of Colloid Interface and Chemical Thermodynamics, Institute of Chemistry, Chinese Academy of
Sciences, Beijing 100190, China
c
College of Information Technology, Jilin Normal University, Siping 136000, China

A R T I C L E I N F O A B S T R A C T

Keywords: A sophisticated fluorescent chemosensor, 2-(4-nitrophenyl)-4,5-diphenyl-1H-imidazole (NPDI), was designed

Diphenyl imidazole and synthesized through a one-step condensation reaction. NPDI exhibited a fluorescence enhancement response
Fluorescence toward Al3+, accompanied by significant emission color change without interference from other tested metal
Chemosensor
ions. The binding stoichiometry and mechanism was corroborated using various techniques. The limit of
Al3+
Water detection
detection (LOD) for Al3+ could reach 7.25 × 10− 8 mol/L and the binding constant was found to be 1.47 × 105 L/
mol. Furthermore, the in-situ formed NPDI⋅Al complex functioned a secondary chemosensor for water by
quenching effect. The fluorescence quenching mechanism could be attributed to hydrogen bonding interaction of
nitro substituent with water. The LOD was calculated to be 0.012 %, indicating NPDI⋅Al heightened sensitivity to
water. Additionally, NPDI⋅Al complex was employed for the moisture detection in the surroundings. Finally, the
practical application of NPDI⋅Al complex had been successfully used in the determination of water content in
food products.

1. Introduction widely employed in fields such as aluminum-based pharmaceuticals,

food processing, paper industry, medicines, ceramics, and cosmetics
In recent years, fluorescent chemosensors have garnered significant (Kumar et al., 2019; Song et al., 2022; Wang et al., 2021). Consequently,
interest compared to traditional analytical techniques, owing to their aluminum levels in soil and domestic water are continuously increasing,
merits of simplicity, naked-eye recognition, high sensitivity and real facilitating its entry into human body. It is well known that the excessive
time monitoring (Lan et al., 2018; Lan et al., 2017; Sivaraman et al., accumulate of Al3+ can be toxic and is associated with various diseases,
2018; Yuan et al., 2013). Given the advantages of fluorescent chemo­ for example, Alzheimer’s, Parkinson’s, anemia (Perl & Brody, 1980).
sensors, researchers worldwide have expressed considerable interest in Undoubtedly, the detection of Al3+ concentration level in environmental
developing chemosensors for the detection of ions and molecules. While and biological systems is of paramount importance. Recently, various
many of these chemosensors exhibit exceptional performance in sensing analytical methods have been employed for the detection of Al3+, such
single analytes, the development of chemosensors capable of continu­ as inductively coupled plasma mass spectrometry (Sanz-Medel et al.,
ously detecting dual-analytes has received less attention. Nevertheless, 2002), atomic absorption spectrometry (Cassella et al., 2005), electro­
due to the virtues of dual-analyte responsive single chemosensors, such chemical methods (Wang et al., 2011), and so on. However, these
as analytical time reduction (Tang et al., 2018), the construction of such methods suffer from various limitations due to complicated operation,
chemosensors presents a formidable challenge. time consuming and high cost. In contrast, many Al3+ fluorescent che­
Numerous efficient fluorescent chemosensors have been developed mosensors based on complexation mechanisms have been reported (Dey
for the determination of metal cation concentrations in different bio­ et al., 2019; Erdemir & Kocyigit, 2018; Mal et al., 2018). Nevertheless,
logical and environmental samples via enhancing, quenching or ratio­ detecting Al3+ remains challenging compared with other metal ions,
metric shifting of their fluorescence emission maxima (Goshisht et al., because of the lack of spectral characteristics of Al3+ (Wang et al., 2017).
2022; Zhou et al., 2014). Among all cationic elements, aluminum is Moreover, some chemosensors only stay in the phase of sensing Al3+,

* Corresponding author.
E-mail address: [email protected] (Y. Wang).

https://doi.org/10.1016/j.foodchem.2023.136138
Received 28 October 2022; Received in revised form 3 April 2023; Accepted 7 April 2023
Available online 11 April 2023
0308-8146/© 2023 Elsevier Ltd. All rights reserved.
Y. Liu et al. Food Chemistry 420 (2023) 136138

and the Al3+-based complexes have not further applied to the quanti­ carried out on a Shimadzu UV-2550 spectrophotometer (200–900 nm).
tative analysis of other objects, rendering them unsuitable for extensive The fluorescence spectra were measured on PerkinElmer LS55 and
practical testing. Therefore, developing a high efficiency fluorescent Hitachi F-4600 fluorescence spectrophotometers. An XtaLAB Synergy-R
chemosensor for Al3+ and utilization its Al3+-based complex as a sen­ diffractometer ketone target X-ray source Single crystal X-ray data was
sitive fluorogenic detector for other objects remains an area warranting performed to collect the single crystal X-ray data as well as the crystal
further investigation. structure was analyzed using the SHELXL 97 program (Sheldrick, 1997).
Water, as one of the most important natural resources, plays very Edinburgh FLS980 spectrometer was used to collect the time resolved
important role in human daily life. However, even trace amounts of fluorescence decay data.
water can be detrimental to various industrial processes, biological
processes and chemical reactions (Kumar et al., 2019; Ullman et al., 2.2. Synthesis and characterization of NPDI
2018; Wang et al., 2020; Wu et al., 2020). For example, trace amounts of
water are considered impurities in conventionally used chemical re­ As shown in Scheme S1, 4-nitrobenzaldehyde (302.3 mg, 2 mmol),
agents, as they can quench water-sensitive reactions, decrease the benzil (420.5 mg, 2 mmol) and ammonium acetate (1.23 g, 16 mmol) in
product yields and even bring about serious safety accidents, such as fire 15 mL glacial acetic acid solution were refluxed for 5 h with stirring at
and explosion (Gichinga & Striegler, 2008). Water presence in drugs 120 ◦ C. The reaction system was cooled and added into ice water. After
may reduce the physicochemical stability of active ingredients and then, the resulting solid precipitation was filtered. The final product was
preparations or induce impurities (Kay et al., 2021). Excessive water in dried in vacuum to give a yellow solid. Yield: 85.3%. FT-IR (cm− 1):
foods will promote hydrolysis of nutrients, producing some by-products 3369, 3057, 3024, 1678, 1599, 1520, 1497, 1440, 1332, 1103, 970, 853,
that reduce the taste or even provide a conducive environment for 767, 696. 1H NMR (d6-DMSO, 400 MHz) δ: 13.18 (s, 1H), 8.32–8.38 (m,
mycotoxins to grow (Wang et al., 2022). Consequently, accurately 4H), 7.40–7.57 (m, 7H), 7.24–7.35 (m, 3H); 13C NMR (d6-DMSO, 100
quantifying water content in chemical reagents, drugs, and food prod­ MHz) δ: 146.54, 143.39, 138.45, 136.10, 134.59, 130.49, 130.08,
ucts is critical for human safety and health. Although the traditional 128.75, 128.54, 128.25, 127.13, 126.92, 125.75, 124.29. ESI-MS: calcd
Karl-Fischer method has been widely employed for monitoring trace for C21H16N3O2: 340.1086 [M− H]-, found: 340.1091.
water (Liang, 1990), the shortcomings of specific instruments, toxic
reagents (SO2, I2, alcohol), complicated and time-consuming operating 2.3. UV–Vis and fluorescence selectivity to metal ions
severely limit its further applications. As an alternative, researchers
have developed fluorescent and colorimetric materials for water deter­ Stock solutions (1 × 10-2 mol/L) of NPDI and various metal ions
mination using fluorescence or absorption techniques (Chao et al., 2018; (Al3+, Ca2+, Tb3+, Zn2+, Co2+, Cd2+, Pb2+, Mg2+, Mn2+, Ni2+, Cr3+, K+,
Hou et al., 2021). Although these chemosensors are applicable to the Na+, Ag+) were prepared in dried acetonitrile. In the selective experi­
detection of water, some materials required complex synthesis, lengthy ment, 6 µL of NPDI stock solution and 30 µL (5.0 equivalents) of each
responsive time and intensive purification to acquire the final pure metal ion solution were added in a quartz optical cell with an optical
compounds (Chen et al., 2017). Therefore, there is an urgent need to path length of 1 cm. After being well mixed, the solutions were diluted to
exploit a low-cost, simple and convenient fluorescence method for 3 mL and equilibrated for a few minutes, whereafter, the UV–Vis ab­
detecting water content. Furthermore, identifying potential applications sorption and fluorescence spectra were recorded.
for the quantitative assessment of water content in food products is
equally crucial. 2.4. Fluorescence competition experiments
In the present work, a simple fluorescent chemosensor (NPDI) was
prepared by a one-step condensation reaction of 4-nitrobenzaldehyde For metal ion competition experiments, 30 µL (5.0 equivalents) of
and benzil. In acetonitrile solution, NPDI exhibited greater selectivity each metal ion solution was taken and added into 6 µL of NPDI aceto­
and sensitivity for Al3+ compared to other tested metal ions, accompa­ nitrile solution. Then, 30 µL (5.0 equivalents) of Al3+ was added to the
nied by a significant fluorescence enhancement. The possible detection above each mixed solution. Each system was diluted to 3 mL with dried
mechanism was thoroughly investigated using spectral experiments in acetonitrile. Fluorescence emission spectra of these solutions were
detail. Interestingly, upon the addition of trace amounts of water, the recorded at room temperature. The excitation wavelength was 360 nm.
fluorescence of the NPDI⋅Al complex in dried acetonitrile gradually
weakened and the emission color of the mixture became discernible to 2.5. UV–Vis and fluorescence titration experiments towards Al3+
naked eye. To the best of our knowledge, this is the first report of uti­
lizing an Al3+-based complex as a fluorescent chemosensor for trace In the titration experiment, 6 µL of NPDI stock solution and various
water detection. Based on the findings, the NPDI⋅Al complex demon­ concentration of Al3+ were mixed and the total volume was set to 3 mL
strated a lower limit of detection for water relative to previous studies. with dried acetonitrile. Then UV–Vis absorption and fluorescence
Finally, the application of NPDI⋅Al in detecting water content in food spectrum measurements were carried out at room temperature.
product samples was also performed.
2.6. Density functional theory (DFT) computation
2. Experimental section
The molecular structure optimization was carried out using Gaussian
2.1. Materials and instruments 16 (refer to the Supporting Information) at the B3LYP 6-311 G (d) level,
with the PCM solvation model in acetonitrile solvent, and Grimme’s
The materials, reagents and dried acetonitrile solvent (with molec­ DFT-D3 dispersion correction. TD-DFT calculations were performed
ular sieves, water ≤ 10 ppm) presented in this study were obtained using Gaussian 16 at the Cam- b3lyp /6-311G (d) level, with the PCM
commercially and used as received. Acetonitrile solution of each metal solvation model in acetonitrile solvent, and Grimme’s DFT-D3 disper­
ion were prepared with its nitrate or perchlorate salts. The NMR ex­ sion correction. Visualization was done using VMD and Multiwfn (Lu &
periments were carried out on a Bruker AVANCE III 400 spectrometer, Chen, 2012).
chemical shifts are reported in ppm using TMS as an internal standard.
Electro spray ionization mass spectroscopy (ESI-MS) were measured on 2.7. Detection of water by NPDI⋅Al
a Bruker esquire 6000 spectrometer. PerkinElmer FT-IR spectropho­
tometer (4000–400 cm− 1) was utilized to collect the FT-IR spectra by In a typical detection, 6 µL of NPDI and 3 µL of Al3+ (0.5 equivalents)
grinding technique. UV–Vis absorption spectra of all the solutions were stock solutions were mixed first and then 100 µL of water was added,

2
Y. Liu et al. Food Chemistry 420 (2023) 136138

which the mixed solution was diluted to 3 mL with dried acetonitrile. molecules played an important role in the lattice, and NPDI molecules
The resultant solution was placed in a quartz cell (1 cm width), and the were orderly arranged through non-covalent interaction between mol­
changes of the fluorescence was recorded at room temperature. For the ecules (The distance between nitrogen hydrogen on imidazole and hy­
water titration experiment based on NPDI⋅Al, 6 µL of NPDI and 3 µL of droxy oxygen is about 2.0 Å, and the distance between hydroxy
Al3+ stock solutions were mixed followed by different volumes of water hydrogen and nitro oxygen is 2.1 Å). The chromophore centers were
were added to the system. Subsequently, each solution was diluted to 3 almost parallel due to the CH…π (yellow dashed lines, CH…π length of
mL with dried acetonitrile. After the solutions were mixed thoroughly, 2.6 Å; green dashed lines, CH…O length of 2.6 Å) and π - π interactions
their fluorescence spectra under UV 360 nm excitation as well as UV–Vis (3.4 Å) of chromophores, as shown in Fig. 1a-b. Conversely, eight NPDI
absorption spectra were recorded. molecules constituted a single unit in orthorhombic system developed
from ethanol solution, as well as intermolecular hydrogen bonds existed
in imidazole units (NH…N length of 2.2 Å) and –NO2 units of neigh­
2.8. Detection of water content in food products
boring molecules with a bond distance of 2.5–2.6 Å (Fig. 1d). Adjacent
chromophore centers interacted via CH…π (the length of 2.6 and 3.4 Å),
The quantitative analysis of water in food products by NPDI⋅Al was
and the stable conformation was arranged in a 20.8◦ cross (Fig. 1e).
carried out using table salt, gourmet powder, peanut and aniseed as
Furthermore, the two crystals exhibited different fluorescence emission
examples. Initially, specific amounts of each food product (0.9338 g,
behaviors (Fig. S1), the triclinic and orthorhombic crystal displayed an
0.8629 g, 1.2324 g and 0.3155 g of table salt, gourmet powder, peanut
orange-red emission peaked at 606 nm and yellow emission peaked at
and aniseed, respectively) were immersed in 5 mL acetonitrile for a short
546 nm, respectively. This discrepancy might be attributed to the
period. The mixture was then centrifuged at 84 g for 20 min and sub­
different arrangement patterns of chromophore-conjugated molecules.
sequently filtered to obtain a clear solution. Each solution was added to
the NPDI⋅Al (molar ratio was 2:1) solution with the final volume
maintained at 3 mL. The fluorescence intensity at 546 nm of the tested 3.2. UV–Vis and fluorescence spectra characteristics for metal ions
samples was then measured. For the recovery study, the above food
products were thoroughly dried in a vacuum drying oven at 65 ◦ C for 24 The photophysical properties of the free chemosensor NPDI were
h and the treatment for these samples was the same as above. Subse­ initially investigated using UV–Vis spectroscopy. As seen in Fig. 2 and
quently, 0.30% and 0.80% volumes of water were spiked into the clear S2, NPDI exhibited a maximal absorption peak located at 387 nm, which
solutions, respectively. The total content of spiked water in the aceto­ was associated with the π-π* transition of the molecular skeleton.
nitrile solution of samples was determined according to the procedure Computational modelling showed that the HOMO and LUMO + 1 or­
described above. Each sample was measured in triplicate. bitals of free NPDI were distributed across the entire molecule, while the
LUMO orbital distributed on the terminal nitrobenzene ring (Fig. S3).
3. Results and discussion The simulated UV–Vis spectrum of NPDI demonstrated that the ab­
sorption of NPDI was primarily contributed by the HOMO → LUMO and
3.1. Single crystal X-ray structures of NPDI HOMO → LUMO + 1 transitions of molecule (Table S3). Next, the
spectroscopic properties of the chemosensor NPDI were inspected by
Crystal structure analysis is crucial for a deeper understand molec­ observing UV–Vis absorption and fluorescence changes in presence of
ular structure at the level of atomic and electronic distribution, and it various metal ions, including Al3+, Ca2+, Tb3+, Zn2+, Co2+, Cd2+, Pb2+,
also offers valuable insights into molecular properties. Fortunately, the Mg2+, Mn2+, Ni2+, Cr3+, K+, Na+ and Ag+ in acetonitrile solution at
single crystals of NPDI were successfully cultivated from the evapora­ room temperature (Fig. 2a). Upon addition of tested metal ions, treat­
tion process of methanol and ethanol solution, respectively. The single ment of NPDI only with Al3+ resulted in a decrease in absorption in­
crystal X-ray diffraction results showed that the diversified crystal sys­ tensity at 387 nm, while an increase at 340 nm with a blue shift of 47 nm
tems formed in distinct solvents, specifically triclinic and orthorhombic, was observed. In contrast, no significant changes in the UV–Vis spectra
as shown in Fig. 1c and 1f, respectively. Data collection and structure of NPDI were observed after adding other tested metal ions, signifying
refinement parameters were given in Table S1 and S2. the high selectivity of NPDI for Al3+.
In the crystal data derived from the methanol solution, methanol Fluorescence emission is also a powerful method for verifying the

Fig. 1. ORTEP drawings of single-crystal structures analysis of NPDI. (a, d) Non-covalent interaction between molecules of NPDI in crystals, (b, e) various π
conjugated group molecular stacking modes in different crystals, (c, f) the unit cell of NPDI obtained from methanol and ethanol respectively. (a-c, in methanol; d-f,
in ethanol).

3
Y. Liu et al. Food Chemistry 420 (2023) 136138

Fig. 2. (a) The UV–Vis absorption spectra of NPDI in CH3CN solution (2 × 10-5 mol/L) in the absence and presence of different metal ions, (b) fluorescence spectra of
probe NPDI (2 × 10-5 mol/L) in the presence of various metal ions in CH3CN medium (λex = 360 nm) and (b) Photographs of NPDI upon addition of 2 equiv. of
various metal ions in CH3CN under UV 365 nm irradiation.

selectivity of the chemosensors towards target ions. Therefore, the 3.3. Investigation of sensitivity and sensing mechanism
emission spectral responses of NPDI in the presence of different tested
metal ions were investigated. As depicted in Fig. 2b, the free NPDI Owing to its remarkable selectivity, the binding behavior and sen­
displayed a relatively weak emission peak at 606 nm upon excitation at sitive analysis of NPDI for Al3+ were meticulously scrutinized using
360 nm with a low quantum yield (ΦF(s) = 1.3). Upon addition of above fluorescence titration experiments with the results illustrated in Fig. 3a.
metal ions, the emission spectra revealed that only Al3+ induced a sig­ The fluorescence intensity of chemosensor NPDI at 546 nm exhibited a
nificant fluorescence enhancement at 546 nm with higher quantum gradual increase concomitant with the progressive addition of Al3+ (0 ~
yield (ΦF(s) = 6.5), accompanied by a substantial color change from 1.0 equiv.) and barely changed as the concentration of Al3+ was up to 10
orange-red to bright yellow (Fig. 2c). However, the addition of other µmol/L (0.5 equiv.), demonstrating that NPDI bound to Al3+ in a 2:1
tested ions did not cause any obvious changes in emission intensity, binding stoichiometry. From the results, there was a good linear rela­
except for Cr3+ and Tb3+ quenching the emission intensity and Pb2+ tionship between the concentration of Al3+ and the fluorescence in­
weakly enhancing the emission intensity. These fluorescence results tensity of NPDI at 546 nm in the range of 0 ~ 8 µmol/L (Fig. 3b). The
further suggested that NPDI could be employed as a fluorescent che­ limit of detection (LOD) value was calculated to be 7.25 × 10-8 mol/L
mosensor with high selectivity to Al3+. (based on 3 s/k, where s is the standard deviation of blank measure­
To better assess the selectivity of NPDI as a fluorescent chemosensor ments, and k is the slope of the intensity ratio versus sample concen­
to Al3+, the anti-interference experiment of NPDI for Al3+ was also tration plot) (Liu et al., 2018). Utilizing the Benesi-Hildebrand equation,
performed. From Fig. S4, it could be concluded that the emission in­ the binding constant for NPDI⋅Al complex was calculated to be 1.47 ×
tensity of NPDI⋅Al at 546 nm showed no obvious changes in the presence 105 L/mol (Fig. S5) (Benesi & Hildebrand, 1949). Additionally, a con­
of coexistent ions. The results demonstrated that these competitive spicuous transition in the emission color of NPDI from orange-red to
metal ions could hardly hinder the fluorescence response of NPDI to yellow occurred as the Al3+ concentration increased under UV 365 nm
Al3+. Thus, NPDI could serve as an efficient fluorescent chemosensor for lamp excitation (Fig. S6). In conjunction with the fluorescence colori­
Al3+ even in the presence of other tested metal ions. metric analysis of ion selectivity, these findings suggested that NPDI
could be employed for the detection and identification of Al3+ via naked

Fig. 3. (a) Fluorescence titration profiles (λex = 360 nm) of NPDI (2 × 10-5 mol/L) in the presence of increasing amounts of Al3+ ions in CH3CN and (b) Detection
limit of NPDI with Al3+ in CH3CN.

4
Y. Liu et al. Food Chemistry 420 (2023) 136138

eye observation. Moreover, an acceptable isosbestic point at 360 nm in rapidly quenched in less than 10 s accompanied by the fluorescence
UV–Vis spectra signified the formation of a stable complex in this system color vanished (Fig. 4a). This indicated that the prepared NPDI⋅Al
(Fig. S7). complex showed a swift response toward water, making it a promising
To elucidate the sensing mechanism of NPDI to Al3+, the Job’s fluorescent sensor for practical applications. To assess the possibility of
method was used to verify the binding stoichiometry between NPDI and quantitative detection of trace water, the fluorescence intensity changes
Al3+ (Fig. S8). It revealed that the maximum emission intensity of the were investigated as a function of water content. The fluorescence
NPDI⋅Al solution was observed when the proportion of Al3+ reached spectra of NPDI⋅Al in dried acetonitrile with water fractions ranging
0.33, indicative of a 2:1 binding mode between NPDI and Al3+. Addi­ from 0 to 5.0% (v/v) were examined (Fig. 4b), revealing a 95.8% sup­
tionally, a satisfactory 2:1 stoichiometry of chemosensor NPDI with Al3+ pression of fluorescence intensity and a 30 nm red-shift. The bath­
was corroborated using the free bindfit software (Fig. S9). The stoi­ ochromic shift and significant quenching should ascribe to the
chiometric complexation was further substantiated by a m/z peak at hydrogen-bonding interaction of nitro substituent with water (Shugrue
783.1973 in the ESI-MS spectrum of NPDI with Al3+ (Fig. S10), assigned et al., 2016). In addition, a nice linear relationship was established be­
to [2NPDI + Al3++H2O + NH+ + + 1
4 +CH3CN-3H ] . H NMR titration tween the emission intensity and water content within the range of
experiment was implemented for a comprehensive elucidation of the 0–1.17% with a correlation coefficient (R2) of 0.995 (Fig. 4c). The LOD
binding mechanism by gradual addition of Al3+ to a CD3CN solution of for water using the NPDI⋅Al complex was calculated as 0.012% by 3 s/k
NPDI (Fig. S11). In the presence of 0.5 equiv. of Al3+, the proton signal formula, a value lower than most previously reported fluorescence
(Ha) of the imidazole NH at 11.09 ppm markedly down-field shifted to sensors (Basak & Das, 2022; Huo et al., 2021; Mandal et al., 2016; Men
12.84 ppm. Simultaneously, the proton signals of Hb and Hc on the ar­ et al., 2013). The rapid response of NPDI⋅Al in detecting water content
omatic group with an attached nitro group at 8.35 ppm and 8.23 ppm also surpassed some earlier reported chemosensors, as summarized in
shifted down-field by 0.19 ppm and 0.01 ppm, respectively. The signals Table S4. Concurrently, the quenching type was analyzed by the Stern-
assigned to the two adjacent benzene protons shifted downfield toward Volmer equation (F0/F = KsvC + 1) (Fig. 4d), where Ksv and C are the
7.56 ppm, suggesting that the configuration of benzene rings was altered quenching constant and the content of water, respectively. The
due to the coordination between NPDI and Al3+ (Zong et al., 2021). quenching constant Ksv was estimated to be 7.5 × 105 L/mol. Moreover,
Furthermore, no significant shift in proton signals was observed upon the emission color of NPDI⋅Al visibly transitioned from yellow to dark as
the addition of 1.0 equiv. of Al3+, reaffirming the 2:1 stoichiometry water content increased from 0 to 3% (v/v) under UV 365 nm irradiation
between NPDI and Al3+. The complexation behavior of NPDI with Al3+ (Fig. S13). Considering these findings, the NPDI⋅Al complex demon­
was also analyzed by FT-IR technique (Fig. S12), the -C– – N- stretch of strated potential for quantitative and accurate analysis of trace water in
imidazolyl in free NPDI, initially at 1680 cm− 1, shifted to 1715 cm− 1 in environmental samples.
the NPDI⋅Al complex, signifying the coordination of imidazole N with To elucidate the mechanism underlying water-induced fluorescence
Al3+. quenching in NPDI⋅Al complex, UV–Vis spectral measurements were
conducted on NPDI⋅Al in dried acetonitrile containing different water
contents (Fig. S14). As the water content rose from 0% to 5%, the ab­
3.4. Detection of trace water content using NPDI⋅Al
sorption peak redshifted from 340 nm to 346 nm, rather than returning
to the original position of free NPDI. This observation suggested that the
Intriguingly, upon introducing a trace amount of water to the
addition of water did not induce hydrolysis of the NPDI⋅Al complex, a
NPDI⋅Al acetonitrile solution, the fluorescence intensity of NPDI⋅Al was

Fig. 4. (a) Fluorescence emission changes observed in NPDI⋅Al solution upon the addition of water, (b) Fluorescence intensity changes of NPDI⋅Al in dried CH3CN
with the incremental increase of water content, (c) the linear relationship of the fluorescence intensity at 546 nm with the water content and (d) plot of the
fluorescence intensity changes of NPDI versus water content in CH3CN.

5
Y. Liu et al. Food Chemistry 420 (2023) 136138

conclusion further supported by 1H NMR experiments (Fig. S11). Upon Table 1

increasing water content to 8% (v/v), an upfield shift of the proton (Hb) Determination of water content in food products.
in the ortho position of the nitro group from 8.537 ppm to 8.486 ppm Samples Weight Intensity Water content RSD (%, n =
was observed, while other proton signals remained relatively stable. (g) (au) (%) 3)
These results provided additional evidence for the existence of hydrogen table salt 0.9338 270.93 3.20 % 0.93
bonding interactions between water and the nitro substituent. gourmet 0.8629 325.71 1.26 % 1.12
powder
peanut 1.2324 123.82 6.56 % 1.89
3.5. Analytical performance of NPDI⋅Al toward water in food products aniseed 0.3155 208.91 6.28 % 2.25

To validate the practical applicability of the NPDI⋅Al complex, its

feasibility for quantitative determination of water content in common deviations from this range cause immense problems in industrial and
food products, such as table salt, gourmet powder, peanut and aniseed, biological processes as well as even lead to health issues. Motivated by
was examined. After the preparation of the clear solutions (as detailed in the response of NPDI⋅Al to water, its potential application for environ­
the experimental section), the emission intensity of each solution was mental humidity detection (in % RH) was further explored. Upon
measured. As shown in Fig. 5a-b and Table 1, the samples exhibited exposure of the dried acetonitrile solution of NPDI⋅Al in an open beaker
significant changes in emission intensity compared to the blank NPDI⋅Al to an environment with 56% RH, the fluorescence intensity was sup­
complex solution with calculated water contents ranging from 1.26% to pressed after 225 min (Fig. 5c). In addition, the required time for fluo­
6.56%. To substantiate the accuracy of NPDI⋅Al in sample analysis, the rescence quenching could adjust from 90 to 390 min over a RH range of
recovery rate of moisture detection using standard addition method was 70%~43% (Fig. 5d). And more interestingly, an acceptable linear cor­
assessed. These food products were completely dried, 0.30% and 0.80% relation (R2 = 0.980) of the fluorescence quenching time as a function of
volumes of water spiked to the clear acetonitrile solutions, respectively. RH was observed, implying that NPDI⋅Al might hold potential advan­
The total water content of each sample was derived from calibration tages for environmental moisture detection.
equation and the recoveries were summarized in Table S5. The re­
coveries (98.2 ~ 103.5%) of water content was satisfactory and relative 4. Conclusion
standard deviation (RSD) was below 3.22%. These findings demon­
strated the reliability and practicability of NPDI⋅Al as a sensing material In summary, we have successfully synthesized and characterized a
for water detection in food products. facile diphenyl imidazole-based fluorescent chemosensor NPDI, exhib­
iting high selectivity and sensitivity for Al3+ in acetonitrile. NPDI
showed selective response toward Al3+ over other tested metal ions via
3.6. Detection of moisture in the atmosphere
an emission enhancement effect. NPDI coordinated with Al3+ through a
2:1 binding stoichiometry with a LOD of 7.25 × 10-8 mol/L. Intriguingly,
Moisture in the air and surrounding environment plays a crucial role
the prepared NPDI⋅Al complex effectively detected water content in
in human daily activities and well-being. The optimal relative humidity
dried acetonitrile solution, exhibiting pronounced fluorescence
for our surroundings is 40 to 60% RH (Kumar et al., 2019) and

Fig. 5. (a) The food products used in the test, (b) fluorescence intensity changes of NPDI⋅Al towards food products, (c) The fluorescence intensity changes of NPDI⋅Al
in dried CH3CN solution under 56 %RH humidity. Inset: photo of the hygrometer used in the humidity test and (d) linear relationship of NPDI⋅Al quenching time as a
function of relative humidity.

6
Y. Liu et al. Food Chemistry 420 (2023) 136138

quenching and color change. NPDI⋅Al displayed a rapid response time Goshisht, M. K., Patra, G. K., & Tripathi, N. (2022). Fluorescent Schiff base sensors as a
versatile tool for metal ion detection: Strategies, mechanistic insights, and
toward water with a calculated LOD value of 0.012%. The quenching
applications. Materials Advances, 3(6), 2612–2669. https://doi.org/10.1039/
mechanism was attributable to the formation of hydrogen bonds be­ D1MA01175H
tween water and the nitro substituent. Additionally, NPDI⋅Al responded Hou, F., Liu, X., Hao, X., Li, G., Lu, F., & Deng, Q. (2021). New benzotriazole-based D-
to ambient humidity and proved applicable for the quantitative detec­ A–D type solvatochromic dyes for water content detection in organic solvents. Dyes
and Pigments, 195, Article 109667. https://doi.org/10.1016/j.dyepig. 2021.109667
tion of water content in food products. In essence, this is the first Huo, Z., Xu, W., Wang, Z., & Xu, S. (2021). A carbonized polymer dot (CPD) nanosensor
endeavor to develop an Al3+-based complex as fluorescence chemo­ for trace water detection with a wide detection range. Dyes and Pigments, 196, Article
sensor for water detection in practical samples. This work offers prom­ 109805. https://doi.org/10.1016/j.dyepig.2021.109805
Kay, J., Thomas, R., Gruenhagen, J., & Venkatramani, C. J. (2021). Simultaneous
ising insights into the construction of straightforward and versatile quantitation of water and residual solvents in pharmaceuticals by rapid headspace
practical fluorescent sensing materials for environmental analysis. gas chromatography with thermal conductivity detection (GC-TCD). Journal of
Pharmaceutical and Biomedical Analysis, 194, Article 113796. https://doi.org/
10.1016/j.jpba.2020.113796
CRediT authorship contribution statement Kumar, P., Ghosh, A., & Jose, D. A. (2019). A simple colorimetric sensor for the detection
of moisture in organic solvents and building materials: Applications in rewritable
Yucun Liu: Conceptualization, Funding acquisition, Writing - orig­ paper and fingerprint imaging. Analyst, 144(2), 594–601. https://doi.org/10.1039/
C8AN01042K
inal draft, Investigation. Ning Li: Methodology, Data curation. Yon­ Kumar, V., Kumar, P., Kumar, S., Singhal, D., & Gupta, R. (2019). Turn-on fluorescent
gling Zhang: Validation, Visualization. Yuan Wang: Funding sensors for the selective detection of Al3+ (and Ga3+) and PPi ions. Inorganic
acquisition, Resources, Software, Writing - review & editing. Chemistry, 58(15), 10364–10376. https://doi.org/10.1021/acs.inorgchem. 9b01550
Lan, L., Chen, D., Yao, Y., Peng, X., Wu, J., Li, Y., … Ying, Y. (2018). Phase-dependent
fluorescence quenching efficiency of MoS2 nanosheets and their applications in
Declaration of Competing Interest multiplex target biosensing. ACS Applied Materials & Interfaces, 10(49),
42009–42017. https://doi.org/10.1021/acsami.8b15677
Lan, L., Yao, Y., Ping, J., & Ying, Y. (2017). Recent progress in nanomaterial-based
The authors declare that they have no known competing financial optical aptamer assay for the detection of food chemical contaminants. ACS Applied
interests or personal relationships that could have appeared to influence Materials & Interfaces, 9(28), 23287–23301. https://doi.org/10.1021/
acsami.7b03937
the work reported in this paper. Liang, Y. Y. (1990). Automation of Karl Fischer water titration by flow injection
sampling. Analytical Chemistry, 62(22), 2504–2506. https://doi.org/10.1021/
Data availability ac00221a018
Liu, H., Liu, T., Li, J., Zhang, Y., Li, J., Song, J., … Wong, W.-Y. (2018). A simple Schiff
base as dual-responsive fluorescent sensor for bioimaging recognition of Zn2+ and
Data will be made available on request. Al3+ in living cells. Journal of Materials Chemistry B, 6(34), 5435–5442. https://doi.
org/10.1039/C8TB01743C
Lu, T., & Chen, F. (2012). Multiwfn: A multifunctional wavefunction analyzer. Journal of
Acknowledgements Computational Chemistry, 33(5), 580–592. https://doi.org/10.1002/jcc.22885
Mal, K., Naskar, B., Mondal, A., Goswami, S., Prodhan, C., Chaudhuri, K., &
This work was supported by the National Natural Science Foundation Mukhopadhyay, C. (2018). Dihydroindeno[1,2-b]pyrroles: New Al3+ selective
off–on chemosensors for bio-imaging in living HepG2 cells. Organic & Biomolecular
of China (No. 21703078), the Natural Science Foundation Project of Chemistry, 16(32), 5920–5931. https://doi.org/10.1039/C8OB01411F
Jilin Province (No. YDZJ202301ZYTS311, YDZJ202201ZYTS425), the Mandal, D., Dinda, S., Choudhury, P., & Das, P. K. (2016). Solvent induced
financial support from BMS Junior Fellow of BNLMS (No. morphological evolution of cholesterol based glucose tailored amphiphiles:
Transformation from vesicles to nanoribbons. Langmuir, 32(38), 9780–9789. https://
2021BMS20053), China Postdoctoral Science Foundation (No.
doi.org/10.1021/acs.langmuir.6b02165
2022M713201). Men, G., Zhang, G., Liang, C., Liu, H., Yang, B., Pan, Y., … Jiang, S. (2013). A dual
channel optical detector for trace water chemodosimetry and imaging of live cells.
Analyst, 138(10), 2847–2857. https://doi.org/10.1039/C3AN36887D
Appendix A. Supplementary data Perl, D. P., & Brody, A. R. (1980). Alzheimer’s disease: X-ray spectrometric evidence of
aluminum accumulation in neurofibrillary tangle-bearing neurons. Science, 208
Supplementary data to this article can be found online at https://doi. (4441), 297–299. https://doi.org/10.1126/science.7367858
Sanz-Medel, A., Soldado Cabezuelo, A. B., Milačič, R., & Bantan Polak, T. (2002). The
org/10.1016/j.foodchem.2023.136138.
chemical speciation of aluminium in human serum. Coordination Chemistry Reviews,
228(2), 373–383. https://doi.org/10.1016/S0010-8545(02)00085-1
References Shugrue, C. R., DeFrancisco, J. R., Metrano, A. J., Brink, B. D., Nomoto, R. S., &
Linton, B. R. (2016). Detection of weak hydrogen bonding to fluoro and nitro groups
in solution using H/D exchange. Organic & Biomolecular Chemistry, 14(7),
Basak, M., & Das, G. (2022). Amine-incorporated quinoxaline based fluorescent sensor
2223–2227. https://doi.org/10.1039/C5OB02360B
for detection of trace water: Solvent influenced self-assembly. Spectrochimica Acta
Sivaraman, G., Iniya, M., Anand, T., Kotla, N. G., Sunnapu, O., Singaravadivel, S., …
Part A: Molecular and Biomolecular Spectroscopy, 280, Article 121521. https://doi.
Chellappa, D. (2018). Chemically diverse small molecule fluorescent chemosensors
org/10.1016/j.saa.2022.121521
for copper ion. Coordination Chemistry Reviews, 357, 50–104. https://doi.org/
Benesi, H. A., & Hildebrand, J. H. (1949). A spectrophotometric investigation of the
10.1016/j.ccr.2017.11.020
interaction of iodine with aromatic hydrocarbons. Journal of the American Chemical
Song, G., Li, J.-C., Majid, Z., Xu, W., He, X., Yao, Z., … Cheng, N. (2022). Phosphatase-
Society, 71(8), 2703–2707. https://doi.org/10.1021/ja01176a030
like activity of single-atom CeNC nanozyme for rapid detection of Al3+. Food
Cassella, R. J., Magalhães, O. I. B., Couto, M. T., Lima, E. L. S., Neves, M. A. F. S., &
Chemistry, 390, Article 133127. https://doi.org/10.1016/j.foodchem.2022.133127
Coutinho, F. M. B. (2005). Synthesis and application of a functionalized resin for
Tang, X., Wang, Y., Han, J., Ni, L., Zhang, H., Li, C., … Qiu, Y. (2018). A novel
flow injection/F AAS copper determination in waters. Talanta, 67(1), 121–128.
fluorescent probe based on biphenyl and rhodamine for multi-metal ion recognition
https://doi.org/10.1016/j.talanta.2005.02.019
and its application. Dalton Transactions, 47(10), 3378–3387. https://doi.org/
Chao, D., Lyu, W., Liu, Y., Zhou, L., Zhang, Q., Deng, R., & Zhang, H. (2018). Solvent-
10.1039/C7DT04629D
dependent carbon dots and their applications in the detection of water in organic
Ullman, A. M., Jones, C. G., Doty, F. P., Stavila, V., Talin, A. A., & Allendorf, M. D.
solvents. Journal of Materials Chemistry C, 6(28), 7527-7532. doi: 10.1039/
(2018). Hybrid polymer/metal–organic framework films for colorimetric water
C8TC02184H.
sensing over a wide concentration range. ACS Applied Materials & Interfaces, 10(28),
Chen, L., Ye, J.-W., Wang, H.-P., Pan, M., Yin, S.-Y., Wei, Z.-W., … Su, C.-Y. (2017).
24201–24208. https://doi.org/10.1021/acsami.8b07377
Ultrafast water sensing and thermal imaging by a metal-organic framework with
Wang, H., Yu, Z., Wang, Z., Hao, H., Chen, Y., & Wan, P. (2011). Preparation of a
switchable luminescence. Nature Communications, 8(1), 15985. https://doi.org/
preplated bismuth film on Pt electrode and its application for determination of trace
10.1038/ncomms15985
aluminum(III) by adsorptive stripping voltammetry. Electroanalysis, 23(5),
Dey, S., Purkait, R., Pal, K., Jana, K., & Sinha, C. (2019). Aggregation-induced emission-
1095–1099. https://doi.org/10.1002/elan.201000748
active hydrazide-based probe: Selective sensing of Al3+, HF–2, and nitro explosives.
Wang, J., Li, Y., Li, K., Meng, X., & Hou, H. (2017). Highly selective turn-on fluorescent
ACS Omega, 4(5), 8451–8464. https://doi.org/10.1021/acsomega.9b00369
chemodosimeter for AlIII detection through AlIII-promoted hydrolysis of C=N
Erdemir, S., & Kocyigit, O. (2018). Dual recognition of Zn2+ and Al3+ ions by a novel
double bonds in the 8-hydroxyquinoline aldehyde Schiff base. Chemistry – A
probe containing two fluorophore through different signaling mechanisms. Sensors
European Journal, 23(21), 5081-5089. doi: 10.1002/chem.201606024.
and Actuators B: Chemical, 273, 56–61. https://doi.org/10.1016/j.snb.2018.06.019
Wang, J., Zhang, X., & Liu, H.-B. (2020). Highly sensitive pyrene–dansyl conjugate-based
Gichinga, M. G., & Striegler, S. (2008). Effect of water on the catalytic oxidation of
fluorescent sensor for discriminative detection of water in organic solvents. Dyes and
catechols. Journal of the American Chemical Society, 130(15), 5150–5156. https://doi.
Pigments, 182, Article 108685. https://doi.org/10.1016/j.dyepig. 2020.108685
org/10.1021/ja078057+

7
Y. Liu et al. Food Chemistry 420 (2023) 136138

Wang, M., Lu, L., Song, W., Wang, X., Sun, T., Zhu, J., & Wang, J. (2021). AIE-active Yuan, L., Lin, W., Zheng, K., He, L., & Huang, W. (2013). Far-red to near infrared analyte-
Schiff base compounds as fluorescent probe for the highly sensitive and selective responsive fluorescent probes based on organic fluorophore platforms for
detection of Al3+ ions. Journal of Luminescence, 233, Article 117911. https://doi.org/ fluorescence imaging. Chemical Society Reviews, 42(2), 622–661. https://doi.org/
10.1016/j.jlumin.2021.117911 10.1039/C2CS35313J
Wang, S., Yu, L., Wei, Z., Xu, Q., Zhou, W., & Xiao, Y. (2022). Dual-response ratiometric Zhou, Y., Zhang, J. F., & Yoon, J. (2014). Fluorescence and colorimetric chemosensors
fluorescence based ligand-functionalized CsPbBr 3 perovskite quantum dots for for fluoride-ion detection. Chemical Reviews, 114(10), 5511-5571. doi: 10.1021/
sensitive detection of trace water in edible oils. Sensors and Actuators B: Chemical, cr400352m.
366, Article 132010. https://doi.org/10.1016/j.snb. 2022.132010 Zong, Z., Wang, J., Bin, Y., Wu, Y., & Huang, G. (2021). A bifunctional fluorescent probe
Wu, Y., Ji, J., Zhou, Y., Chen, Z., Liu, S., & Zhao, J. (2020). Ratiometric and colorimetric for sensing of Al3+ and H2S. Analytical Methods, 13(18), 2157–2164. https://doi.org/
sensors for highly sensitive detection of water in organic solvents based on hydroxyl- 10.1039/D1AY00096A
containing polyimide-fluoride complexes. Analytica Chimica Acta, 1108, 37–45.
https://doi.org/10.1016/j.aca.2020.02.043