A Highly Selective Perylenediimide-Based Chemosensor: “Naked-Eye” Colorimetric and Fluorescent Turn-On Recognition for Al3+

A novel “turn-on” fluorescent probe (PCN) was designed, synthesized, and characterized with perylene tetracarboxylic disimide as the fluorophore and Schiff base subunit as the metal ion receptor. The probe demonstrated a considerable fluorescence enhancement in the presence of Al3+ in DMF with high selectivity and sensitivity. Furthermore, the considerably “off–on” fluorescence response simultaneously led to the apparent color change from colorless to brilliant yellow, which could also be identified by naked eye easily. The sensing capability of PCN to Al3+ was evaluated by the changes in ultraviolet–visible, fluorescence, Fourier transform–infrared, proton nuclear magnetic resonance, and high-resolution mass spectrometry spectroscopies. The linear concentration range for Al3+ was 0–63 μM with a detection limit of 0.16 μM, which allowed for the quantitative determination of Al3+.


Materials and Instruments
All chemicals and solvents were purchased from commercial providers and used without purifying. FT-IR spectra were measured using a Bruker ALPHA-T spectrometer (KBr, Bruker, Germany). The 1 H NMR and 13 C NMR spectra were recorded on a Bruker AVANVE 400 MHz system (Bruker, Germany) using CF 3 COOD as the solvent. The HRMS was carried out on an FTMS Ultral Apex MS spectrometer (Bruker Daltonics Inc., USA). The ultraviolet-visible (UV-vis) spectra were gained on a UV-2550 ultraviolet spectrophotometer (Shimadzu, Japan). The fluorescence spectra were obtained through the PerkinElmer SCHEME 2 | Synthesis of perylene-3,4,9,10-tetracarboxylic diimide probe PCN.

Spectrophotometric Studies
The stock solution of PCN (10 −3 M) was dissolved in DMF and then diluted to 10 −5 M for the spectroscopic measurements of Al 3+ . The stock solutions of 10 −2 mol L −1 concentration of metal ions were provided from AlCl 3 , FeCl 3 , CrCl 3 , MgCl 2 , PbCl 2 , ZnCl 2 , Hg(OAc) 2 , CaCl 2 , CuCl 2 , CoCl 2 , MnCl 2 , SnCl 2 , NiCl 2 , BaCl 2 , NaCl, KCl, and AgNO 3 using ultrapure water. In the selectivity measurement, 10 mL of the PCN solution (1 × 10 −5 M) and 50 µL of each metal ion stock solution (10 −2 M) were added to volumetric flasks. The probe PCN stock solution (10 mL) was mixed with gradual incremental Al 3+ solution separately for the titration experiments. These resulting solutions were well mixed, and then the spectral properties were recorded after 6 h. The excitation was set at 500 nm for the measurement of fluorescence, and slit width of the excitation was 10 nm. In spectral experiment, the concentration titration was measured at least twice to ensure consistent results. All the measurements were obtained at 25 • C.

UV-Vis and Fluorescence Spectral Characteristics Studies
The solvent effect of PCN has been studied through fluorescence measurement in different solvents (Supplementary Figure 8). The probe exhibits weak fluorescence properties in almost all the investigated solvents except EtOH. The compound exhibits weak yellow fluorescence emission with peaks from 541 nm (MeCN) to 558 nm (DMSO) in most solvents, but much no fluorescence in EtOH was observed. Based on the response mode of the fluorescent molecular probe's "turn on" and "turn off, " the solvent was selected for further study. The probe had failed to show the good selectivity toward various ions in DMSO and other solvents. The photophysical property of the fabricated fluorescent chemosensor in DMF was investigated. Free probe (PCN) demonstrated weak fluorescence at about 550 and 590 nm. To estimate the selectivity and sensitivity of PCN (10 µM), the UV-vis and fluorescence spectra of PCN toward different metal cations (such as Al 3+ , Fe 3+ , Cr 3+ , Mg 2+ , Pb 2+ , Zn 2+ , Hg 2+ , Ca 2+ , Cu 2+ , Co 2+ , Mn 2+ , Sn 2+ , Ni 2+ , Ba 2+ , Na + , K + , and Ag + ) have been investigated. UV-vis spectra of PCN were obtained in the existence of 5 equiv. of the tested cations. The absorption of PCN increased significantly in the presence of Al 3+ at 490 and 525 nm, and the colorless solution of PCN changed to yellow under natural light, indicating that the "naked eye" is visible (Figure 1). The fluorescence responses of PCN were measured in the presence of fivefold excess of various metal ions (Figure 2). Upon the addition of different cations (5 equiv.), only when added the Al 3+ into the solution can it induce a significant fluorescence enhancement at 550 and 590 nm. Moreover, a yellow-colored visual fluorescence change was observed after adding Al 3+ ion to the PCN solution. In contrast, most of the other metal ions, including some mono-, Frontiers in Chemistry | www.frontiersin.org di-, and trivalent metal ions (Ag + , Na + , K + , Mg 2+ , Pb 2+ , Zn 2+ , Hg 2+ , Ca 2+ , Cu 2+ , Co 2+ , Mn 2+ , Sn 2+ , Ni 2+ , Ba 2+ , Cr 3+ , and Fe 3+ ), were unresponsive to this system. The imine bond in PCN was hydrolyzed by the addition of Al 3+ due to the Lewis acid character of Al 3+ , and compound 2 with strong fluorescent was released. It resulted to a prominent "light-on" yellow solution and fluorescence emission of PCN, which allowed for naked-eye detection of Al 3+ under natural light and UV light of 365 nm. This mechanism was proven by FT-IR, 1 H NMR, and HRMS experiments. Therefore, PCN showed "off-on" response to Al 3+ ions in the DMF solution. All these showed the good selectivity of PCN toward Al 3+ over other cations.

UV-Vis and Fluorescence Titration Experiments
The UV-vis and fluorescence titrations were measured by increasing the amount of Al 3+ (0-70 µM) to PCN in DMF. The intensity of absorbance at 490 and 525 nm increased gradually after the addition of an increasing amount of Al 3+ (Figure 3). The absorbance of PCN became steady at 525 nm with the Al 3+ concentration being 63 µM. The change of UVvis spectral could attribute to the binding affinity of PCN. As depicted in Figure 4, the independent PCN exhibited a weak emission in DMF. However, a distinct increase in fluorescence emission was observed after Al 3+ was added and a plateau with the addition of 63 µM of Al 3+ was achieved at 550 nm. The weak fluorescence may correspond to the photoinduced electron transfer (PET) process resulted from the N atom of imine (C=N) to the luminescent perylene unit (Upadhyay et al., 2018;Fu et al., 2019b). With the increase of Al 3+ , the PET effect of the sensor PCN is inhibited, and thereby the intense fluorescence of PDI units is restored. Also, it clearly indicates the structural change of PCN by the interaction of Al 3+ with PCN. As shown in Figure 5A, the fluorescence intensity showed a linear relationship (R 2 = 0.9956) with concentration of Al 3+ in the range of 0-63 µM, indicating that PCN could be used to determine the Al 3+ quantitatively. The detection limit of PCN for sensing Al 3+ was 0.16 µM based on applying equation LOD = 3σ/k, which was calculated from the linear regression curve of the fluorescence intensity and the Al 3+ concentration. This indicated that PCN had high fluorescence selectivity to Al 3+ . In addition, the binding constant, K, could also be obtained from the Stern-Volmer equation I 0 /(I 0 − I) = 1/A + 1/KA · 1/[Q]. The K for Al 3+ was calculated to be 7.75 × 10 4 M −1 in the DMF solution, as increased from the fluorescence titration curves of probe PCN with Al 3+ (Figure 5B). The comparison of probe PCN with other Al 3+ chemical sensors based on Schiff 's base was summarized in Table 1 with different sensing mechanisms (Gan et al., 2017;Roy et al., 2017;Tajbakhsh et al., 2017;Shen et al., 2018;Wang et al., 2018;Zhang et al., 2018;Fu et al., 2019). Compared with other sensors, the advantage of probe PCN was its lower detection limit, but its insolubility in water was its shortcoming, which might limit its application in biological and environmental chemistry to some extent.

The Competition Experiments Studies
The fluorescence spectra of PCN were investigated to examine its selectivity in the existence of other metal ions. First, 10 equiv. of background metal ions (100 µM) was added to the solution of PCN (10 µM) to form a PCN/M n+ system, and then, 5 equiv. of Al 3+ (50 µM) was added into the solution. As shown in Figure 6, Al 3+ detection by compound PCN was not influenced by the selected background metal cations. Therefore, the combined results clearly indicated that PCN exhibited remarkable Al 3+ signaling behavior and can function as a high selectivity and disturbance-free Al 3+ fluorescent probe even in the existence of most competing metal cations.

Sensing Mechanism of PCN for Al 3+
To elucidate the sensing mechanism, the IR, NMR, and HRMS spectra of PCN-Al 3+ were performed. The FT-IR spectroscopic analysis of PCN and PCN-Al 3+ complexes is shown in Figure 7.
Compared with the FT-IR spectra of PCN, the peaks appeared at 3,386 and 3,323 cm −1 in the FT-IR spectra after the addition of 5 equiv. of Al 3+ , which is consistent with the amino peak of intermediate 2.
The result suggested that intermediate 2 might be regenerated, which is consistent with the previous spectral analysis. The 1 H NMR spectroscopic analysis of PCN and the reaction product of PCN with Al 3+ are shown in Figure 8. The addition of Al 3+ resulted in different peak profiles, the signal of aldimine protons (H 4 ) at 8.99 ppm completely disappeared, and these peaks at 7.67-7.98 ppm corresponding to aromatic protons (H 5 ) also disappeared comparing with 1 H NMR spectra, which also confirmed that probe PCN was hydrolyzed. Moreover, by comparing the HRMS spectra in Figure 9, it was found that the original peak at [M+H] + 721.1407 for free PCN disappeared, and a new peak at [M+H] + 477.1552 emerged after the addition of Al 3+ . All results clearly delineated that Al 3+ induced the cleavage of imine. In order to get full insight into the mechanism, the fluorescence spectra of compound 2 and PCN in the absence and presence of Al 3+ were recorded separately in Figure 10.
Upon the spectral changes of PCN induced by Al 3+ , it was found that the spectral data were nearly identical with those of compound 2, which clearly confirmed the cleavage of the C=N of PCN in the presence of Al 3+ (Scheme 3).
In order to understand the sensing mechanism in the probe PCN in the absence or presence of Al 3+ ion, density functional theory (DFT) quantum mechanical approach was performed. A Gaussian program (Frisch et al., 2009) was employed for DFT calculations at the B3LYP-D3BJ/def2-SV(P) level (Stephens SCHEME 3 | Proposed sensing mechanism toward Al 3+ .   al., 1994;Weigend and Ahlrichs, 2005;Frisch et al., 2009;Grimme et al., 2010). The S1 state geometry was optimized, and the corresponding molecular orbitals were recorded by isosurfaces with the isovalue at 0.02. In the excite state of PCN, HOMO-1 (−6.62 eV) of the receptor unit is close to fluorophore HOMO (−6.16 eV) and located above the fluorophore HOMO-9 (−7.67eV). Hence, the electron of the HOMO-1 will be transferred to fluorophore regime through the reductive PET mechanism (Maity et al., 2019;Dos Santos Carlos et al., 2020). Significantly, the fluorescent "off " state of PCN is observed. After Al 3+ ion hydrolyzed the probe PCN, the HOMO-2 (−6.44 eV) energy levels of the primary amine are decreased than that of the fluorophore HOMO-1 (−6.18 eV), as also observed. Hence, PET could not efficiently operate from the HOMO-2 of the primary amine to the fluorophore's HOMO-1 upon the removal of imine moiety, resulting in the fluorescent "on" state ( Figure 11). Thus, the present calculation demonstrated that the electron donor imine leads to a highly efficient PET process.

CONCLUSION
In summary, a novel PDI-based Schiff base derivative PCN was synthesized and utilized as a fluorescent probe. PCN exhibited a selective turn-on response to Al 3+ over other coexisting competitive metal ions in DMF. DFT calculations showed that coordination of PCN to Al 3+ inhibits the PET process. The C=N of PCN was hydrolyzed by Al 3+ , leading to the return to the intermediate compound, which resulted naked-eye visible color changes from colorless to yellow and nonfluorescent to yellow fluorescent. The detection limit was sufficiently low to determine the micromolar levels of Al 3+ . This sensor is valuable for Al 3+ analysis in environmental samples.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
YL and SG constructed the workflow, performed the data analysis, and wrote the manuscript. X-ML synthesized and purified the compound. LY and Y-LL contributed to data analysis. FY and YF contributed to the conception of the study, revised the manuscript, and approved the final version. All authors contributed to the article and approved the submitted version.

ACKNOWLEDGMENTS
This work was supported by the Natural Science Foundation of China (51903032), the Natural Science Foundation of Heilongjiang Province (LH2019B002), Postdoctoral Science Foundation of Heilongjiang Province (No. LBH-Z19045), and Young Talents Project of Northeast Agricultural University (No. 18QC64). The authors are grateful to Prof. Min Zhang (Northeast Normal University) for the assistance with the DFT calculation and analysis.