A Double-Site Chemodosimeter for Selective Fluorescence Detection of a Nerve Agent Mimic

A novel two-site chemodosimeter (SWJT-4) based on fluorescein skeleton to detect diethyl chlorophosphate (DCP) was designed and synthesized. It is a turn-on fluorescent probe for DCP with good selectivity and obvious color change in aqueous solution. Interestingly, the two oxime groups of SWJT-4 as dual response sites initiated different reactions with DCP to form a cyano group and an isoxazole ring, respectively. The corresponding mechanism was confirmed by 1H NMR, MS and DFT calculation. Moreover, SWJT-4 could be used as a fluorescent test paper to detect DCP vapor.


Introduction
Organophosphate nerve agents refer to a class of chemical substances composed of organophosphorus compounds (OPs). They have been widely used in pesticides and chemical warfare agents, such as parathion, systemic phosphorus, malathion, dimethoate, DDVP, tabun, sarin and soman [1,2]. These chemicals can result in a range of neurological symptoms such as headache, dizziness and agitation [3][4][5][6]. Organophosphorus compounds were used as a chemical weapon in war to threaten the safety of human life seriously. In a word, organophosphorus nerve agents are not only a potential threat of biochemical warfare, but also a usual weapon of terrorist organizations. Therefore, the detection of these substances by a convenient method is very necessary.
In our previous work, we synthesized a series of DCP probes based on ON-OFF fluorescent responses [40,41]. In the present work, a novel fluorescent probe SWJT-4 with two response sites was designed and synthesized for selective detection of DCP. It has good photochemical stability in an aqueous solution. The turn-on fluorescent responses could be achieved by tethering two oxime groups to a fluorescein skeleton. Moreover, SWJT-4 could be used to detect DCP vapor by a fluorescent test paper. good photochemical stability in an aqueous solution. The turn-on fluorescent responses could be achieved by tethering two oxime groups to a fluorescein skeleton. Moreover, SWJT-4 could be used to detect DCP vapor by a fluorescent test paper.

Design and Photoproperties of SWJT-4
According to the Duff reaction, two aldehyde groups were formed at the ortho positions of two hydroxyl groups of fluorescein [42]. Two oxime groups as the recognition sites were then added to the fluorophore to construct SWJT-4 (Scheme 1). The C = N rotation in SWJT-4 would weaken the fluorescence. With the addition of DCP, an isoxazole or cyano group would be formed, which would inhibit the rotation of C = N and therefore enhance the fluorescence. The corresponding 1 H and 13 C NMR spectra and ESI-MS of SWJT-4 are demonstrated in Figures S1-S3 (Supplementary Materials). In order to study its solvent effect on fluorescence properties of SWJT-4, five common organic solvents, namely acetone, methanol, tetrahydrofuran (THF), N,N-dimethyl formamide (DMF) and dimethyl sulfoxide (DMSO), were selected to test their performance under excitation at 520 nm. As shown in Figure S4 (Supplementary Materials), SWJT-4 in DMSO has a longer emission wavelength. After the addition of DCP, the fluorescence intensity was enhanced clearly. Subsequently, the impact of pH on SWJT-4 was also studied ( Figure S5, Supplementary Materials). SWJT-4 did not react with DCP under acidic conditions, and the best reaction condition was pH 7.0 to 8.0. Therefore, DMSO-HEPES (1/1, v/v, pH 7.4) buffer solution was determined as the optimal condition. Fluorescence Response of SWJT-4 to DCP As shown in Figure 1a, the UV-visible absorption spectrum of SWJT-4 exhibited prominent absorption at 522 nm, which was attributed to the formation of intramolecular hydrogen bonds between the hydroxyl groups on fluorescein and the nitrogen atom on oxime, and the hydrogen bonds would enlarge the conjugation of the probe [43]. When DCP was added to the solution of the probe, the absorbance was blue-shifted to 508 nm, which is the absorption of the ring-opening structure of fluorescein [44]. These results showed that the reaction occurred between SWJT-4 and DCP to break the intramolecular hydrogen bonds. The color of the solution changed from pink to pale yellow (Figure 1a, inset). For the fluorescence spectrum, under the excitation of 520 nm, SWJT-4 showed weak emission at 557 nm (Φ = 5.7%) (Figure 1b). After the addition of DCP, the fluorescence was enhanced at 545 nm (Φ = 26.7%) with a slight blue-shift [45]. The fluorescence color of the solution was observed to turn chartreuse to green (Figure 1b, inset). These results showed that SWJT-4 was a turn-on fluorescent probe and could be used for the detection of DCP.
The fluorescence titration experiment of SWJT-4 was then studied. As shown in Figure 1c, the fluorescence intensity gradually increased with the increase in DCP. There was a good linear relationship between DCP concentration and the fluorescence intensity in the range of 0-140.0 μM (Figure 1d). The detection limit was calculated as 53.0 nM ( Figure  S6, Supplementary Materials), which was much lower than the reported lethal dose (0.01 Fluorescence Response of SWJT-4 to DCP As shown in Figure 1a, the UV-visible absorption spectrum of SWJT-4 exhibited prominent absorption at 522 nm, which was attributed to the formation of intramolecular hydrogen bonds between the hydroxyl groups on fluorescein and the nitrogen atom on oxime, and the hydrogen bonds would enlarge the conjugation of the probe [43]. When DCP was added to the solution of the probe, the absorbance was blue-shifted to 508 nm, which is the absorption of the ring-opening structure of fluorescein [44]. These results showed that the reaction occurred between SWJT-4 and DCP to break the intramolecular hydrogen bonds. The color of the solution changed from pink to pale yellow (Figure 1a, inset). For the fluorescence spectrum, under the excitation of 520 nm, SWJT-4 showed weak emission at 557 nm (Φ = 5.7%) (Figure 1b). After the addition of DCP, the fluorescence was enhanced at 545 nm (Φ = 26.7%) with a slight blue-shift [45]. The fluorescence color of the solution was observed to turn chartreuse to green (Figure 1b, inset). These results showed that SWJT-4 was a turn-on fluorescent probe and could be used for the detection of DCP.
The fluorescence titration experiment of SWJT-4 was then studied. As shown in Figure 1c, the fluorescence intensity gradually increased with the increase in DCP. There was a good linear relationship between DCP concentration and the fluorescence intensity in the range of 0-140.0 µM (Figure 1d). The detection limit was calculated as 53.0 nM ( Figure S6, Supplementary Materials), which was much lower than the reported lethal dose (0.01 mg/L) [46]. This result indicated that SWJT-4 had high sensitivity to DCP and could detect DCP at lower concentrations in an aqueous solution. Moreover, as shown in Figure  S7

Competition Experiments
In order to investigate the plausible interference of other organophosphorus reagents or nerve agent mimics on the detection of DCP, phosphoric acid (PA), cyanomethyl diethyl phosphate (DCMP), cyanoyl diethyl phosphate (DCNP) and ethyl dichlorophosphate (DCEP) were selected to study the selectivity of SWJT-4 ( Figure 2). With the addition of other phosphate-containing substances, the fluorescence intensity of SWJT-4 changed little. However, after the addition of DCP to the solution of SWJT-4, the fluorescence at 545 nm was significantly enhanced. These results clearly showed that the fluorescent probe SWJT-4 could recognize DCP effectively even in the presence of other war agent mimics.

Competition Experiments
In order to investigate the plausible interference of other organophosphorus reagents or nerve agent mimics on the detection of DCP, phosphoric acid (PA), cyanomethyl diethyl phosphate (DCMP), cyanoyl diethyl phosphate (DCNP) and ethyl dichlorophosphate (DCEP) were selected to study the selectivity of SWJT-4 ( Figure 2). With the addition of other phosphate-containing substances, the fluorescence intensity of SWJT-4 changed little. However, after the addition of DCP to the solution of SWJT-4, the fluorescence at 545 nm was significantly enhanced. These results clearly showed that the fluorescent probe SWJT-4 could recognize DCP effectively even in the presence of other war agent mimics.

Response Mechanism
In order to identify the reaction mechanism between SWJT-4 and DCP, 1 H NMR titration was performed. As shown in Figure 3a, the peaks at 11.90 ppm and 11.10 ppm belonged to the proton signals of two hydroxyl groups (H a ) on oxime moieties and two hydroxyl groups (H b ) on fluorescein in SWJT-4, respectively. The chemical shift of the proton (H c ) of the aldoxime group was at 8.80 ppm. After the addition of DCP, the original signals H a and H b in the probe disappeared, and one aldoxime proton H c moved from 8.80 to 9.44 ppm downfield. The above results indicated that different reactions occurred at the two reaction sites of SWJT-4 to form product 2. One oxime group reacted with one DCP to form the nitrile [47], and the other oxime group reacted with another DCP to form an isoxazole ring. Firstly, the hydroxyl group in the oxime attacked the phosphorus center of DCP to form phosphate oxime. Then, another hydroxyl group in the adjacent position of the oxime group could intramolecularly attack this generated intermediate, and the isoxazole ring was then formed through a release of phosphate moiety [48]. Meanwhile, the mixture of SWJT-4 and DCP was also characterized by ESI-MS, and the peak at m/z 381.3 corresponding to product 2 was observed ( Figure S10, Supplementary Materials).

Computational Studies
In order to further study the fluorescence responses of SWJT-4 with DCP and the influence of different solvent media on the absorption and emission, the B3LYP/6-31g method was conducted by Gaussian simulation (DFT) [49][50][51]. As shown in Figure 4, the main contribution for SWJT-4 was from HOMO -2 to LUMO + 1. The electron clouds of their HOMO -2 orbital in SWJT-4 were distributed on one benzene ring of the xanthrene group and the adjacent oxime group. The electron clouds of the LUMO orbital of SWJT-4 were mainly distributed on the whole molecule instead. These results suggested the weak fluorescence of SWJT-4. As for product 2, the main contribution was HOMO -2 to LUMO. The electron clouds of HOMO -2 and LUMO orbitals were all mainly distributed in the xanthrene moiety, indicating the strong fluorescence character of 2. These results were very consistent with the fluorescence turn-on change on the detection of DCP using SWJT-4. Then, the absorption and emission of SWJT-4 were calculated in DMSO or in water ( Figure  S11, Supplementary Materials). As shown in Figure S11a,c (Supplementary Materials), the maximum emission wavelength of SWJT-4 in DMSO or in water was about 287 nm and the maximum absorption wavelength of SWJT-4 in DMSO or in water was 280 nm ( Figure S11b,d, Supplementary Materials). Although these wavelengths greatly differ from the measured results, the results indicated that the solvent effect had almost no effect on absorption and emission spectra in different solvent media.

Response Mechanism
In order to identify the reaction mechanism between SWJT-4 and DCP, 1 H NMR titration was performed. As shown in Figure 3a, the peaks at 11.90 ppm and 11.10 ppm belonged to the proton signals of two hydroxyl groups (Ha) on oxime moieties and two hydroxyl groups (Hb) on fluorescein in SWJT-4, respectively. The chemical shift of the proton (Hc) of the aldoxime group was at 8.80 ppm. After the addition of DCP, the original signals Ha and Hb in the probe disappeared, and one aldoxime proton Hc moved from 8.80 to 9.44 ppm downfield. The above results indicated that different reactions occurred at the two reaction sites of SWJT-4 to form product 2. One oxime group reacted with one DCP to form the nitrile [47], and the other oxime group reacted with another DCP to form an isoxazole ring. Firstly, the hydroxyl group in the oxime attacked the phosphorus center of DCP to form phosphate oxime. Then, another hydroxyl group in the adjacent position of the oxime group could intramolecularly attack this generated intermediate, and the isoxazole ring was then formed through a release of phosphate moiety [48]. Meanwhile, the mixture of SWJT-4 and DCP was also characterized by ESI-MS, and the peak at m/z 381.3 corresponding to product 2 was observed ( Figure S10, Supplementary Materials).

Computational Studies
In order to further study the fluorescence responses of SWJT-4 with DCP and the influence of different solvent media on the absorption and emission, the B3LYP/6-31g method was conducted by Gaussian simulation (DFT) [49][50][51]. As shown in Figure 4, the main contribution for SWJT-4 was from HOMO -2 to LUMO + 1. The electron clouds of their HOMO -2 orbital in SWJT-4 were distributed on one benzene ring of the xanthrene group and the adjacent oxime group. The electron clouds of the LUMO orbital of SWJT-4 were mainly distributed on the whole molecule instead. These results suggested the weak fluorescence of SWJT-4. As for product 2, the main contribution was HOMO -2 to LUMO. The electron clouds of HOMO -2 and LUMO orbitals were all mainly distributed in the xanthrene moiety, indicating the strong fluorescence character of 2. These results were very consistent with the fluorescence turn-on change on the detection of DCP using SWJT-4. Then, the absorption and emission of SWJT-4 were calculated in DMSO or in water ( Figure S11, Supplementary Materials). As shown in Figure S11a  Although these wavelengths greatly differ from the measured results, the results indicated that the solvent effect had almost no effect on absorption and emission spectra in different solvent media.

Gas-Phase Detection of DCP
Considering its practical application, DCP vapors were used to determine the recognition ability of SWJT-4 (Figure 5a-d). The solution of SWJT-4 was placed in a glass bottle with a lid (Figure 5a,c). The color of the solution was orange-pink under visible light (Fig-Figure 4. The optimized structures and the molecular orbital plots of SWJT-4 and 2.

Gas-Phase Detection of DCP
Considering its practical application, DCP vapors were used to determine the recognition ability of SWJT-4 (Figure 5a-d). The solution of SWJT-4 was placed in a glass bottle with a lid (Figure 5a,c). The color of the solution was orange-pink under visible light (Figure 5a) and chartreuse in ultraviolet light (Figure 5c). As a contrast, the solution of SWJT-4 was placed in another glass bottle with a lid, in which there was a smaller bottle containing DCP (Figure 5b,d). When the DCP vapors came into contact with the solution of SWJT-4, the color of the solution changed from orange-pink to yellow in visible light (Figure 5b), and the chartreuse color turned to green under ultraviolet light (Figure 5d). These results indicated that SWJT-4 could detect DCP vapor.

Materials and Reagents
Related materials, reagents and the detail of detection are described in the Supplementary Materials.

Synthesis of Probe SWJT-4
The compound 1 was synthesized according to a known procedure [42]. Hydroxylamine hydrochlorde (24.4 mg, 0.62 mmol) was dissolved in ethanol (3 mL) and stirred at room temperature for 10 min. Then compound 1 (60.4 mg, 0.16 mmol) was dispersed in 5 mL of ethanol and dripped into the above solution. The reaction mixture was stirred at room temperature for 2 h. The organic solvent was removed by rotary evaporation, and the crude product was isolated by column chromatography (dichloromethane:methanol = 80:1) on silica gel to obtain the probe SWJT-4 (54.2 mg, yield 82.9 %) as a pink solid. 1

Conclusions
In summary, the fluorescence changes of a novel fluorescent probe SWJT-4 for the detection of DCP based on the dual reaction site were explored. SWJT-4 showed good selectivity for DCP and the obvious color change in an aqueous solution. Notably, the two reaction sites in the probe also triggered different reaction types. Moreover, SWJT-4 could be used for DCP Research Funds for the Central Universities (Nos. 2682020CX55,  (Figure 5f), which was observed by the naked eye. The green color in paper sensors gradually increased under ultraviolet light (Figure 5e), and under visible light, the color of the paper sensors changed from pink to pale-yellow (Figure 5f). These results showed that the probe can be used as a fluorescent test paper to detect DCP vapor and has potential application in the development of detection kits for DCP.

Materials and Reagents
Related materials, reagents and the detail of detection are described in the Supplementary Materials.

Synthesis of Probe SWJT-4
The compound 1 was synthesized according to a known procedure [42]. Hydroxylamine hydrochloride (24.4 mg, 0.62 mmol) was dissolved in ethanol (3 mL) and stirred at room temperature for 10 min. Then compound 1 (60.4 mg, 0.16 mmol) was dispersed in 5 mL of ethanol and dripped into the above solution. The reaction mixture was stirred at room temperature for 2 h. The organic solvent was removed by rotary evaporation, and the crude product was isolated by column chromatography (dichloromethane:methanol = 80:1) on silica gel to obtain the probe SWJT-4 (54.2 mg, yield 82.9 %) as a pink solid. 1

Conclusions
In summary, the fluorescence changes of a novel fluorescent probe SWJT-4 for the detection of DCP based on the dual reaction site were explored. SWJT-4 showed good selectivity for DCP and the obvious color change in an aqueous solution. Notably, the two reaction sites in the probe also triggered different reaction types. Moreover, SWJT-4 could be used for DCP vapor detection and as fluorescent test paper to detect DCP.
Supplementary Materials: The following supporting information can be downloaded online: Table S1: Some reported work for the detection of DCP. Figures S1- Figure S3: Copies of 1 H, 13 C NMR, ESI-MS spectra of SWJT-4. Figure S4: The changes of fluorescence in different solvents. Figure S5: The effect of pH on fluorescence intensity. Figure S6: The linear relationship between SWJT-4 and different concentrations of DCP. Figure S7