2.1 AIEE performance of 2a
We studied the visual fluorescence response of 2a in DMF solution with different water fractions (ƒw, range 0%-90% (v/v)) (Fig. 1a). In DMF solution, 2a is yellow-green under 356 nm UV irradiation, and the maximum emission peak is at 520 nm. However, with the increase of water fraction (ƒw increased from 10–60%), the fluorescence emission intensity gradually decreased (Fig. 1c), the fluorescence color of the solution becomes yellow and the maximum emission peak is red-shifted (from 520nm to 542 nm ) (Fig. 1b). The decrease of fluorescence intensity can be attributed to the increase of the polarity of the mixed solvent with the increase of water fraction, which affects the conformation and charge distribution of the fluorescent molecules, resulting in the enhancement of intramolecular charge transfer (ICT) effect [28]. However, when the water fraction exceeds 70% (ƒw increases from 70–90%), the fluorescence intensity increases sharply. This phenomenon can be attributed to the increase of water fraction in the DMF-water mixed solvent, and the restriction of intramolecular motions (RIM) process limits the non-radiative decay of the aggregation state and promotes the transition of the excitation energy [29].
We also performed dynamic light scattering (DLS) studies on different DMF-water fractions (ƒw 60%, 70%, 80%, 90%). These studies have shown that the average particle size (Zav) decreases with the increase of water fraction, which provides strong evidence for the formation of aggregates (Zav at ƒw 60% = 693.7 nm ; at ƒw 70% = 588.5 nm ; at ƒw 80% = 406.7nm ; at ƒw 90% = 312.2nm (Fig. S13). Therefore, 2a can be considered as an AIEE material because it can emit brighter fluorescence in an aggregated state. Unfortunately, 2b does not have this phenomenon.
2.2 pH and Time Response
Figure S14a shows that probe 2a shows a stable fluorescence signal in the range of pH 5.0–12.0. However, after the addition of Cu2+, probe 2a has almost no fluorescence signal in this pH range. In addition, the pH range of probe 2b is 4.0–9.0 (Fig. S14b). Therefore, the subsequent experiments were carried out at pH = 7.3. Undoubtedly, 2a/2b is superior to most Cu2+/Cd2+ probes that can only obtain good sensing performance in a narrow pH range [30–35]. It is gratifying that when Cu2+/Cd2+ is rapidly added to the 2a/2b solution, the fluorescence is completely quenched/turned on within 5 s (Fig.S14c). Compared with the previously reported Cu2+/Cd2+ probes, [36–41] 2a and 2b showed shorter response time.
2.3 Research on sensing performance Sensing studies of 2a/2b to Cu2+/Cd2+
Due to the presence of imine (C = N) groups in 2a and 2b, it is often used as a binding site for various metal ions. [42–44] Therefore, we consider to develop it as a new type of square amide metal ion fluorescent probe, and test this idea through experiments. The selective analysis of nitrate solutions of different metal ions was carried out under 365 nm UV light, and the results were shown in Fig. 2a. Surprisingly, when Cu2+ was added, the original yellow-green fluorescence of 2a was quenched, while when other metal ions were added, the fluorescence did not change significantly. When Cd2+ was added to the DMF:H2O (5:3, v/v) 2b solution, the probe showed a fluorescence turn-on phenomenon, identically, when other metal ions are added, the fluorescence did not change significantly (Fig. 2d). Based on this phenomenon, we further studied the fluorescence emission spectra after adding different metal ions. The fluorescence emission spectrum shows that the signal peak of 2a at 520 nm disappeared after adding Cu2+ (Fig. 2b), which confirms that the fluorescence quenching occurs only in the presence of Cu2+. After the addition of Cd2+, the emission peak of 2b at 521 nm increased sharply, while the intensity of the emission peak did not change significantly when other metal ions were added (Fig. 2e). The above results show that the probes 2a/2b have high selectivity for Cu2+/Cd2+.
The limit of detection (LOD) is an important index to evaluate the sensitivity of fluorescent probes. Therefore, we further carried out the fluorescence titration experiment of 2a/2b by Cu2+ / Cd2+. As shown in Fig. 2c, with the addition of Cu2+, the fluorescence emission peak of 2a at 520 nm gradually weakened. According to the experimental results, we use 1/[Cu2+] as the x-axis and the fluorescence intensity [1/( I-I0 )] at 520 nm as the y-axis, as shown in Fig. S15a. Then the binding constant between the fluorescent probe 2a and Cu2+ was calculated to be 2.26 × 105 M− 1 by Benesi-Hildebrand equation [45]. The slope of the fluorescence titration curve is -72.166, and the adjusted R2 is 0.994 (which indicates that there is a good linear relationship between the fluorescence intensity and the Cu2+ concentration) (Fig. S15c). According to the formula LOD = 3σ/S, where σ is the standard deviation of the maximum fluorescence intensity measured by 10 2a blank samples, and S is the slope of the ratio of the fluorescence intensity of probe 2a at 520 nm to the Cu2+ concentration [46], The LOD of the fluorescent probe for Cu2+ was calculated to be 1.26 × 10− 8 M, which is lower than the World Health Organization (WHO) standard [47]. On the contrary, with the addition of Cd2+, the fluorescence emission peak of 2b at 521 nm gradually increased (Fig. 2f and Fig. S15e). When the amount of Cd2+ reached 1 Eq. (10µM), the fluorescence intensity basically did not change. Similarly, the binding constant between 2b and Cd2+ is 2.31 × 105 M− 1 (Fig. S15b), and the LOD is 2.04 × 10− 8 M, which is significantly lower than the (WHO) standard [48]. The above results indicate that 2a/2b are highly sensitive Cu2+/Cd2+ fluorescent probes.
In addition, in order to further understand the selectivity of probe 2a/2b to Cu2+/Cd2+, we investigated the effect of adding different metal ions of 5 eq. , and then adding Cu2+/Cd2+ of 5 eq. , about the emission intensity, as shown in Fig. 3. When there are other metal ions in the solution, it has little effect on the fluorescence intensity of probe 2a/2b. However, after the addition of Cu2+/Cd2+, the fluorescence signal of probe 2a was rapidly quenched, and the fluorescence signal of 2b was rapidly enhanced. The experimental results show that the probe 2a/2b exhibits high anti-interference and can detect Cu2+/Cd2+ with high selectivity.
2.4 Sensing mechanism of 2a/2b to Cu2+/Cd2+
We studied the UV-vis absorption spectra of probe 2a for Cu2+ in DMF solution and probe 2b for Cd2+ in DMF/H2O solution (2:8, v/v), respectively. As shown in Fig. 4, with the increase of Cu2+ addition, the absorption peak at 410 nm gradually weakens, while the absorption peaks at 330 nm and 265 nm gradually increase. Two isoabsorptive point were observed at 334 nm and 446 nm. At the same time, with the increase of Cd2+ addition, the absorption peak at 369 nm gradually weakened, while the absorption peak at 239 nm gradually increased. Three isoabsorptive point were observed at 305 nm, 462 nm and 553 nm. These phenomena indicate that there is an interaction between the probes 2a/2b and Cu2+/Cd2+, which changes the original conjugated structure and forms a new product.
In order to further explore the complexation ratio and complexation mechanism between probes 2a/2b and Cu2+/Cd2+, we first used Job’s plot to reveal the chemical relationship between them (Fig. S16). Job’s plot showed that the binding ratios of 2a/2b to Cu2+/Cd2+ were both 1:1. In addition, the characteristic peaks in ESI-MS can also confirm the above proposed stoichiometric relationship (Fig. S9-Fig. S12). Based on the results of Job’s plot and ESI-MS, we proposed a possible coordination model (Scheme 2). In order to further confirm the complexation form of probes 2a/2b with Cu2+/Cd2+, we performed 1H NMR titration in DMSO-d6. As shown in Fig. 5, when 1 eq. Cu2+ is added, Only the proton signal of the O-H group of 2a disappeared, while the other proton signals were almost unchanged. In the presence of 1 eq. Cd2+, the imine proton (H1) and aromatic proton (H2) of 2b undergo significant low-field shifts (Fig. S17). These results indicate that Cu2+ can interact with the oxygen atom of the hydroxyl group of probe 2a, while Cd2+ binds to 2b through the nitrogen atom of the imine. It is basically in accordance with the binding mode proposed by Scheme. 2.
Since the O-H proton signal of 2a disappeared after binding to Cu2+, based on this result, we propose that the fluorescence emission of probe 2a is attributed to the excited-state intramolecular proton transfer (ESIPT) process. The introduction of Cu2+ leads to the inactivation of the ESIPT process, and therefore, the fluorescence is quenched [49]. In addition, in DMF:H2O (5:3, v/v) solution, 2b has almost no fluorescence, which is due to the effect of photoinduced electron transfer (PET) effect. After combining Cd2+, the PET process of 2b is blocked, making its fluorescence turn on [50].
2.5 Real water sample testing for Cu2+/Cd2+
It is necessary to detect the level of Cu2+/Cd2+ in the natural environment. We selected tap water, Jialing River water and Qingxi River water as real water samples to test the detection ability of probe 2a/2b. Before the experiment, the water samples were filtered to remove suspended solids and precipitated impurities. Then, three different concentrations of Cu2+/Cd2+ (1, 1.5, 2 µM) were added to these water samples for recovery test, and each experiment was repeated three times under the same conditions. In order to recover Cu2+/Cd2+, we established a standard curve based on the fluorescence intensity between Cu2+/Cd2+ and probe 2a/2b. The standard curve has a good linear relationship (adj. R2 > 0.99) in the target ion concentration range of 1–10 µM, and it was applied to the recovery test of Cu2+ in actual water samples. The results are shown in Table 1, for all samples, the recoveries of probe 2a for Cu2+ and probe 2b for Cd2+ were 97.0%-103.5% and 95.0%-109.5%, respectively, and the relative standard deviation (RSD) was less than 5%. The above results show that the probe 2a/2b has good feasibility for detecting the concentration of Cu2+/Cd2+ in actual samples.
Table 1
The detection results of Cu2+/Cd2+ in actual water samples by probe 2a/2b
Samples
|
Cu2+ added (µM)
|
Cu2+ found (µM)a
|
Recovery (%)
|
RSD (%) (n = 3)
|
Tap water
|
1.00
|
0.97
|
97.0
|
1.52
|
Jialin Raver Water
|
1.50
|
1.48
|
98.7
|
0.73
|
Qinxi Raver Water
|
2.00
|
2.07
|
103.5
|
2.38
|
Samples
|
Cd2+ added (µM)
|
Cd2+ found (µM)b
|
Recovery (%)
|
RSD (%) (n = 3)
|
Tap water
|
1.00
|
0.95
|
95.0
|
3.01
|
Jialin Raver Water
|
1.50
|
1.46
|
97.3
|
1.06
|
Qingxi Raver Water
|
2.00
|
2.19
|
109.5
|
4.26
|
a Probe 2a detects Cu2+
|
|
|
|
b Probe 2b detects Cd2+
|
|
|
|
2.6 The visual application of 2a
In order to develop a portable field sensing platform of probe 2a, we immersed the filter paper in dichloromethane solution containing 2a for 10 min [51]. After soaking, the filter paper was dried in a vacuum oven, then the prepared filter paper was used for Cu2+ detection. As shown in Fig. 6, under 365 nm UV light irradiation, 2a filter paper showed yellow-green fluorescence. After adding 10 µL 1 mM metal ion solution, obvious black spots appeared on the filter paper added with Cu2+, that is, fluorescence quenching, while the fluorescence intensity on the filter paper added with other metal ions remained basically unchanged. This result shows that the 2a filter paper has high selectivity and specificity for the detection of Cu2+ in solution. Subsequently, we used 2a filter paper to test the response of different Cu2+ concentrations. Fig. S18 showed that the fluorescence quenching phenomenon could still be observed when the concentration of Cu2+ was 10 µM. This indicates that the 2a filter paper can be used for the detection of trace Cu2+, which further expands the application range of probe 2a.