A Reversible Bis(Salamo)-Based Fluorescence Sensor for Selective Detection of Cd2+ in Water-Containing Systems and Food Samples

A novel, simple, highly selective, and sensitive fluorescence chemosensor for detecting Cd2+ that was constructed from a bis(salamo)-type compound (H4L) with two N2O2 chelating moieties as ionophore was successfully developed. Sensor H4L could show fluorescence turn-on response rapidly and significant selectivity to Cd2+ over many other metallic ions (Cu2+, Ba2+, Ca2+, K+, Cr3+, Mn2+, Sr2+, Co2+, Na+, Li+, Ni2+, Ag+, and Zn2+), and a clear change in color from colorless to yellow that can be very easily observed via the naked eyes in the existence of Cd2+, while other metallic ions do not induce such a change. Interestingly, its fluorescent intensity was increased sharply with the increased concentration of Cd2+. The detection limit of sensor H4L towards Cd2+ was down to 8.61 × 10−7 M.


Introduction
Metallic ions play a key role in daily life [1][2][3][4]. Cadmium, which is an essential resource in the earth, is widely used in all kinds of agricultural processes and industry containing chemical industry, electronics industry, nuclear industry semiconducting, quantum dots, phosphate fertilizers, and other fields [5]. However, it should be vigilant that Cd 2+ is a heavy metallic ion with highly toxic [6]. Some new studies reveal that a considerable amount of cadmium go into the environment from waste water, waste residue, and exhaust gas not only damages the environment, but also endangers human health [7]. Hence, with ever-increasing concern for environment and human health, there is now a much greater demand for the development of a rapid and convenient detection method for Cd 2+ ion [8,9].
As part of our ongoing interest in the development of new optical (both colorimetric and fluorescence) chemosensor, in the present paper, we report a bis(salamo)-type tetraoxime sensor H 4 L for detecting Cd 2+ by turn-on response in Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) solution. The sensor H 4 L has higher sensitivity for Cd 2+ than other metallic ions that are based upon change in color by naked eyes. The mechanism of fluorescence change has been well demonstrated. The sensor H 4 L as a reliable fluorescence probe displayed high sensitivity toward Cd 2+ in water-containing systems, and was able to detect Cd 2+ in food samples.

Materials and Methods
All of the reactions were performed under an air atmosphere. Boron tribromide (99.9%), methyl trioctyl ammonium chloride (90%), 2-hydroxy-3-methoxybenzaldehyde (99%), and pyridinium chlorochromate (98%) were gained from Alfa Aesar. Hydrobromic acid 33 wt % acetic acid solution was gained from J & K Scientific Ltd. (Beijing, China). Other solvents and reagents that were used in this work were analytical grade from Tianjin Chemical Reagent Factory (Tianjin, China). Melting points were measured using a microscopic melting point apparatus made in Beijing Taike Instrument Limited Company (Beijing, China) and were uncorrected. 1 H NMR spectra were made via German Bruker AVANCEDRX-400 spectrophotometer (Karlsruhe, Germany). All of the UV-vis and fluorescence spectra tests were measured on a Shimadzu UV-2550 (Kyoto, Japan) and Perkin-Elmer LS-55 spectrometer (Waltham, MA, USA). The solvent also has a great influence on the fluorescence of the complex. Through a series of experiments, we finally obtained the best selectivity for cadmium ions in Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) [59,60].

Preperation of Sensor H 4 L
The bis(salamo)-type tetraoxime sensor H 4 L was prepared on the basis of the reported methods [61][62][63][64][65][66][67][68]. The 1 H NMR, IR, and UV-vis spectra of H 4 L are nearly consistent with the previous result ( Figure 1). The major reaction steps of sensor H 4 L are demonstrated in Scheme 1.

pH Effect of Sensor H 4 L
For the sake of avoid the disturbance by protons in the recognition process of heavy metallic ions and to get optimum conditions, we concentrated on the pH influence on the fluorescence intensity. As depicted in Figure 2, the sensor H 4 L have barely changed in the fluorescent intensity from pH = 3.0 to 11.0, indicating that sensor H 4 L was consistent. The weak fluorescent sensor H 4 L may be owing to intra-molecular photo-induced electron transfer. However, H 4 L-Cd 2+ showed strong fluorescence, on account of the bonding of H 4 L with Cd 2+ lead to the inhibition of intra-molecular photo-induced electron transfer process. The results indicated that sensor H 4 L can be used to detect Cd 2+ and the process of detection was not affected greatly by pH values.

General UV-vis Measurements
UV-vis spectra of sensor H 4 L in the existence of 14 metallic ions (Cd 2+ , Cu 2+ , Ba 2+ , Ca 2+ , K + , Cr 3+ , Mn 2+ , Sr 2+ , Co 2+ , Na + , Li + , Ni 2+ , Ag + , and Zn 2+ with nitrate anions) were investigated. As shown in Figure 3a, sensor H 4 L have two intense absorption bands at 313 and 358 nm, which could be attributed to π → π* transition and reveals that sensor H 4 L includes a larger conjugation system. However, when 3.0 equiv. Cd 2+ were added into the mixed solution, a new absorption peak emerged at 375 nm ( Figure 3a). Meanwhile, the change in absorbance is almost same for Cd 2+ , Ni 2+ , Zn 2+ , and Mn 2+ ions, indicating that the H 4 L were also involved in coordination to the Cd 2+ , Ni 2+ , Zn 2+ , and Mn 2+ ions, respectively. However, when Cd 2+ ion was added, sensor H 4 L could show a clear change in color from faint yellow to green that can be very easily observed via the naked eyes under UV light at 365 nm, while other metallic ions do not induce such a change. The colour change that is observed is mainly due to the charge transfer transition of the complex. The results revealed sensor H 4 L could be applied to selective identification of Cd 2+ according to the color changes. The responses of sensor H 4 L to Cd 2+ (1 × 10 −3 M) were studied further via UV-vis titration experiments, as depicted in Figure 3b. With the addition of Cd 2+ from 0.0-3.0 equiv. in Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) solutions, two peaks that were observed at 342 and 358 nm were clearly declined and a new band at 375 nm emerged. Meanwhile, one definitive isosbestic point was also noted at 326 nm, indicating that a new species is produced. When 3.0 equiv. Cd 2+ was added, the absorption no longer change, which showed 1:3 stoichiometry between Cd 2+ and sensor H 4 L (Figure 3b).

General Fluorescence Measurements
For the sake of evaluating the selectivity of sensor H 4 L, original screening of H 4 L to the bonding ability of metallic ions was performed in Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) solutions. The fluorescent spectra of sensor H 4 L in the existence of a series of metallic ions (Cd 2+ , Cu 2+ , Ba 2+ , Ca 2+ , K + , Cr 3+ , Mn 2+ , Sr 2+ , Co 2+ , Na + , Li + , Ni 2+ , Ag + , and Zn 2+ with nitrate anions) were gained, followed via excitation at 323 nm. As shown in Figure 4a, when Cd 2+ ion was added, the position of the emission peak was red-shifted from 412 to 486 nm, the fluorescence intensity increased from 125 to 712, and the fluorescence quantum yield increased from 2.6 to 13.8%. Significantly, H 4 L showed excellent selectivity for Cd 2+ in Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) solutions with a strong fluorescence response, with either a very weak or no fluorescence response to other metallic ions (Figure 4a). It is different from the previously reported sensors for Cd 2+ showing a wide range of response to Zn 2+ , sensor H 4 L nearly non-fluorescent respond to Zn 2+ could be owing to its rigid cavity with bigger size. Therefore, this rigid cavity could be definitely suitable for bonding with Cd 2+ , but not suitable for bonding with Zn 2+ that possess smaller ionic radius. Furthermore, according to the corrected Benesi-Hildebrand formula, the bonding constant for the bonding of Cd 2+ to sensor H 4 L was estimated as 4.98 × 10 4 M −1 and Zn 2+ to sensor H 4 L was estimated as 3.89 × 10 4 M −1 [69]. The result suggested that sensor H 4 L displays outstanding selectivity for Cd 2+ than other metallic ions.   Figure 5. With the addition of increasing Cd 2+ at an excitation wavelength of 323 nm, the fluorescent emission intensity at 485 nm gradually raised while the intensity at 407 nm reduced. Furthermore, after the addition of Cd 2+ , the absorbance at 485 nm exhibited a sharp increase when the ratio of [Cd 2+ ]/[H 4 L] is below 3:1, and no longer change when the ratio reaches 3:1. In addition, the fitting curve of fluorescence emission intensity with Cd 2+ concentrations was obtained by the data obtained of the fluorescence titration experiment ( Figure 5). We have performed 1 H NMR spectra of sensor H 4 L and in which the presence of 3.0 equiv. Cd 2+ ( Figure S1). Moreover, the detection limit is a considerably significant parameter in molecular recognition, the LOD and LOQ parameters were estimated to be 8.61 × 10 −7 M and 2.87 × 10 −6 M, respectively [61,69]. The LOD and LOQ were calculated based on the following equations: where S represents the standard deviation of the blank measurements, and δ is the slope of the intensity versus sample concentration curve. where F min , F max , and F are the emission intensities in the absence, presence of saturated Cd 2+ , and the addition of a given amount of Cd 2+ concentration, respectively. [Cd 2+ ] is the concentration of free Cd 2+ .
To know the stoichiometry between the sensor H 4 L and Cd 2+ in Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) solutions, job's plot has been performed ( Figure 6). When the molar fraction of Cd 2+ was 0.75, the intensity at 495 nm reached an extreme value, indicating the formation of a 3:1 complex between Cd 2+ and H 4 L. Furthermore, according to the reversible fluorescent switch of the sensor H 4 L to Cd 2+ in a coordination compound solution, we regard it as a two-input molecular logic gate, while the emission at 373 nm serves as the output. As depicted in Figure 7 and Table 1, when the output was zero (both the Cd 2+ and EDTA are absent), corresponding the gate being closed and this system shows weak fluorescence. When Cd 2+ alone was existent, the output is one and the relevant to the gate being open, so intense fluorescence was observed. Thus, the sensor H 4 L is able to serve as a logic gate. This result demonstrates that the sensor H 4 L as a reversible fluorescence probe.

The Detecting Mechanism of H 4 L for Cd 2+
According to the fluorescent spectra, the detecting mechanism of the sensor H 4 L for Cd 2+ was suggested, as follows ( Figure 8). The fluorescent intensity of the sensor H 4 L response to Cd 2+ may be assigned to CHEF and PET. Before being coordinated with Cd 2+ , sensor H 4 L displayed a weaker fluorescence due to the lone pair of electrons of nitrogen atoms, which gives rise to an intra-molecular PET. Furthermore, the lone electron pairs of the nitrogen atoms give rise to a nonradiative process by the n-π* state, which also led to a wide degree of fluorescent quenching. Conversely, after H 4 L was coordinated to Cd 2+ , the radiation process was primarily via the π-π* state and the coordination compound was more rigid [48,59]. In addition, the PET process was restrained by the addition of Cd 2+ analyte at the receptor site. It is obvious that the appearance of the ICT process influenced PET in this system, but fluorophore and receptor distances and the orientation between them can also contribute in the overall PET process.  Figure 9 illustrates that fluorescence intensity of sensor H 4 L did not vary with further prolong the reaction time. Besides, the fluorescent intensity of sensor H 4 L almost constant in the temperature range of 0-90 • C. Therefore, sensor H 4 L could be used for rapid response to Cd 2+ in room temperature, which is of significant practicability for the detection of Cd 2+ .

Test Strips Measurements
Currently, test strips analytical equipment have gained momentous attention on account of their high sensitivity, low-cost, and quick response. It is perhaps the most convenient of modern detection tools, as a change in color could be seen via the naked eyes. Hence, in this study, filter papers were performed via soaking filter papers into Tris-phosphate buffer (c = 0.2 M, DMF/H 2 O = 9:1, v/v, pH = 7.0) mixed solutions of probe H 4 L (1 × 10 −5 M), and then drying by exposure to air. The filter papers, including sensor H 4 L, were applied to detect Cd 2+ and other metallic ions. As depicted in Figure 10, after Cd 2+ and the other metallic ions were added on the test tools, respectively, the distinct color changes were seen merely with Cd 2+ solution under UV light at 365 nm, and potentially competitive metallic ions have no work on the detection of Cd 2+ via the filter papers ( Figure S2). It is a really actual convenient method that quickly measures the Cd 2+ , and it could be able to be satisfactorily used in the fields of food security and environmental surveillance.

Application in Food Samples
Furthermore, we investigated the applicability of sensor H 4 L in food samples. 500 mg of crushed naturally containing cadmium rice were put into the PTFE microwave digestion tank by the addition of 3 mL of HNO 3 and 4 mL of H 2 O 2 , and the mixture solution containing cadmium(II) needs be digested after keeping 30-60 min. Then, the mixture solution was dried in vacuo and was added into the mixed solution of DMF and H 2 O (c = 1 × 10 −3 M, DMF/H 2 O = 9:1, v/v, pH = 7). As shown in Figure 11, after added the sample solution to sensor H 4 L, the fluorescence intensity is on the increase.

Conclusions
In this paper, we presented a new bis(salamo)-type fluorescence sensor H 4 L, which could serve as a great promising analytical kit for measuring Cd 2+ via different fluorescence changes and changes in color from light yellow to green that can be measured via naked eyes. The detection limit about fluorescent response of probe H 4 L to Cd 2+ was down to 8.61 × 10 −7 M. As designed, the filter papers could conveniently, efficiently, and simply detect Cd 2+ in solution. Moreover, Cd 2+ in the food samples was detected by the sensor H 4 L simply and effectively. We believe this study will inspire the development of bis(salamo)-based chemosensor by optimizing unsaturated metal coordinating sites for practicability to many other metallic ions detecting in analytical chemistry, medical treatment, biological, and environmental fields.