Preparation of Macrometallocycle and Selective Sensor for Copper Ion

Two bis-imidazolium salts 1,8-bis[2’-(N-R-imidazoliumyl)acetylamino]naphthalene chloride (L1H4·Cl2: R = Et; L2H4·Cl2: R = nBu), as well as their four NHC metal complexes [L1H2Ag]Cl (1), [L1Ni] (2), [L2Ni] (3) and [L1H2Hg(HgCl4)] (4) have been synthesized. In each of the cationic moieties of complexes 1 or 4, there is a groove-like 14-membered macrometallocycle, and each macrometallocycle is consisted of one biscarbene ligand L1H2 and one metal ion (silver(I) ion for 1 and mercury(II) ion for 4). Three 6-membered cycles are contained in each molecule of complexes 2 or 3. Additionally, the selective recognition of macrometallocycle 1 for Cu2+ was studied with the methods of fluorescence and ultraviolet spectroscopy, 1H NMR titrations, MS and IR spectra. The experimental results display macrometallocycle 1 can discriminate Cu2+ from other cations effectively.


Synthesis and general characterization of complexes 1-4. The synthesis of NHC silver(I) complex
[L 1 H 2 Ag]Cl (1) was accomplished via the reaction of L 1 H 4 ·Cl 2 with Ag 2 O in CH 3 CN/DMSO (Fig. 2). The reactions of L 1 H 4 ·Cl 2 or L 2 H 4 ·Cl 2 with NiCl 2 in the presence of K 2 CO 3 in CH 3 CN/DMSO afforded NHC nickel(II)      solution of complex 1 is slightly light-sensitive. The proton signals (NCHN) of imidazolium disappear in the 1 H NMR spectra of 1-4 due to the introduction of metals, and other proton signals are analogous to L 1 H 4 ·Cl 2 or L 2 H 4 ·Cl 2 . In the 13 C NMR spectra of 1, no carbene carbon signal is found, and this phenomenon may be the   Structure of complexes 1-4. In complexes 1-4 (Figs 3-6), the N-C-N angles are between 103.5(1)° and 106.3(5)°, and these values are consistent with those of literatures [47][48][49]61 . One 14-membered macrometallocycle is contained in each of the molecules of complexes 1 or 4. By contrast, three 6-membered cycles in each molecule of 2 or 3 are observed. In the same ligand for 1-4, the naphthalene plane and two imidazole planes form the dihedral angles of 51.5(5)-75.9(8)° (Table S1 in Supporting Information). Two imidazole planes in the same NHC-metal-NHC unit form the dihedral angles of 9.6(5)-14.2(4)° for 1 and 4. In complexes 2 and 3, the dihedral angles formed by two imidazole planes are in the range of 74.9(1)-83.4(3)°.
In complex  In complexes 2 or 3, two acetylamino groups (-CONH-) and two imdazolium moieties of precursors L 1 H 4 ·Cl 2 or L 2 H 4 ·Cl 2 are deprotonated in the presence of K 2 CO 3 . As a result, Ni(II) ion is coordinated to two carbene atoms and two nitrogen atoms to adopt a quadrilateral geometry with slight distortion. The bond distances of C-Ni and N-Ni are 1. Recognition of Cu 2+ using 1 as a chemosensor. The screening experiments of complexes 1-4 for some cations (Li + , Na + , K + , NH 4 + , Ag + , Ca 2+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Cd 2+ , Cr 3+ , Al 3+ , Pb 2+ and Hg 2+ , and their anions are NO 3 − ) via fluorescence spectroscopy in CH 3 CN at 25 °C were carried out. The fluorescence intensities of complexes 2-4 didn't change after adding cations. However, the fluorescence emission of complex 1 decreased remarkably after adding Cu 2+ , and other cations did not have similar phenomenon. Therefore, complex 1 was selected as a chemosensor to process recognition investigation of cations.
To evaluate the response time of complex 1 to Cu 2+ , the time-dependent plot was measured (Fig. 7). The results showed that the interactions between Cu 2+ and 1 can cause fluorescence quenching, in which fluorescence intensity quickly reduced within 6 minutes, and then the tendency slowed down. The fluorescence quantum yields (Φ) of L 1 H 4 ·Cl 2 and complex 1 using 1-aminonaphthalene as fluorescence standard (Φ = 0.39) were measured 64 . The fluorescence quantum yields of L 1 H 4 ·Cl 2 and complex 1 were determined to be 0.16 and 0.21, and the latter was higher than the former. It may be originated to the incorporation of metal-ligand coordination interactions 65,66 .
As shown in Fig. 8, complex 1 showed a fluorescence emission band at ca. 415 nm, which originated from conjugated bis(acetylamino)-naphthalene (λ ex = 330 nm). When 10 equiv. of Li + , Na + , K + , NH 4 + , Ag + , Ca 2+ , Co 2+ , Ni 2+ , Zn 2+ , Cd 2+ , Cr 3+ , Al 3+ , Pb 2+ and Hg 2+ were added, the fluorescence intensity of 1 had no observable change. However, the significant fluorescence quenching of 1 was observed after adding 10 equiv. of Cu 2+ . In UV/vis experiment, upon addition of Cu 2+ to the solution of 1, the absorption of 1 at ca. 250-350 nm increased remarkably, but other cations had no similar influence on the absorption of 1 (Fig. S1 in the Supporting Information). The experiment results showed that 1 can discriminate Cu 2+ from other cations effectively. In the fluorescence titration experiments (Fig. 9), upon the titration of Cu 2+ into solutions of 1 in CH 3 CN at 25 °C, the fluorescence intensities of 1 at ca. 415 nm decreased gradually. In the inset of Fig. 9, the fluorescence intensities of 1 went down quickly in the ratios of C Cu 2+ /C 1 being 0 to 10:1. When the ratio ascended to 20:1, the quenching rate slowed down. Finally, fluorescence intensities remained unchanged even though more Cu 2+ was added. The quenching behaviors of Cu 2+ on the fluorescence of 1 were found to follow a conventional Stern-Volmer relationship 67,68 (equation (1)).
where F 0 and F are the fluorescence intensities of 1 in the absence and presence of Cu 2+ , and C Cu 2+ is the concentration of Cu 2+ . The equation reveals that F 0 /F increases in direct proportion to the increasing concentration of Cu 2+ , and the Stern-Volmer constant K SV defines the quenching efficiency of Cu 2+ .
Analogous to Fig. 8, the decrease of fluorescence intensities of 1 were also observed after the addition of other copper(II) salts (1.0 × 10 −5 mol/L) with different counter anions (Br − , SO 4 2− , OAc − , Cl − , NO 3 − and CO 3 2− ) (Fig. S5). Thus, the different anions did not obviously influence on the binding between 1 and Cu 2+ . Reversible binding of 1 with Cu 2+ was also carried out (Fig. S6). The addition of 10 equiv. of EDTA to a mixture of 1 (2.0 × 10 −6 mol/L) and Cu 2+ (20 × 10 −6 mol/L) resulted in the increase of fluorescence intensity at 415 nm, and the fluorescence intensity was approximately equal to that of 1, which signified the regeneration of the free 1. The fluorescence intensity decreased upon the addition of Cu 2+ again. This result showed that 1 was a good chemosensor for Cu 2+ with admirable reversibility and regeneration capacity.
Interactions of 1 with Cu 2+ . The potential binding sites of 1 for Cu 2+ may be oxygen atoms, nitrogen atoms and π systems (including O···Cu 2+ interactions, N···Cu 2+ interactions and π···Cu 2+ interactions). To get detailed information on how 1 bound with Cu 2+ , we studied the data of 1 H NMR titrations (C Cu 2+ /C 1 was from 0 to 2.0 equiv.) in DMSO-d 6 (Fig. 11). Upon the addition of 1 equiv. of Cu 2+ , the proton signal on NH (Hd) had a large downfield shift by 0.92 ppm (Fig. 11(iv)), and the proton signals of He and Hf on naphthalene ring also shifted to downfield (ca. 0.27 ppm), which may be attributed to electron-withdrawing effect of Cu 2+ due to Cu 2+ ···N interactions (Fig. 12). The proton signal of Hc on CH 2 attached to C=O shifted to downfield (ca. 0.25 ppm), which may be attributed to electron-withdrawing effect of Cu 2+ due to Cu 2+ ···O interactions. More equivalents of Cu 2+ did not cause further change of chemical shifts of Hc-Hf (Fig. 11(v,vi)), which showed the combination ratio between 1 and Cu 2+ was 1:1.
Additional evidence for the combination ratio between 1 and Cu 2+ was obtained through high-resolution mass spectra of 1·Cu 2+ (Fig. S7). The observation of m/z (318.3) for (1·Cu 2+ )/2 furthur comfirmed the formation of a 1:1 complex. This finding agreed with the result of Job's plot (Fig. 10). The IR spectra of 1 and 1·Cu 2+ were measured for more information about how 1 bound with Cu 2+ . In Fig. S8, we found that several absoption bands  By analyzing the structure of 1 and above experiment results, we can conclude that 1 bound with Cu 2+ mainly through Cu 2+ ···O and Cu 2+ ···N interactions. Once complex 1·Cu 2+ was formed, the photo-induced electron transfer (PET) process from the imidazole rings to naphthalene ring was switched on and it led to the quench of fluorescence emission of 1 69,70 . We tried to cultivate the single crystal of 1·Cu 2+ , but unsuccessful.

Conclusion
In conclusion, we prepared and characterized two bis-imidazolium salts L 1 H 2 ·Cl 2 and L 2 H 2 ·Cl 2 , as well as their four NHC metal complexes 1-4. In each molecule of 1 or 4, one 14-membered groove-like macrometallocycle was contained. Additionally, the selective recognition of macrometallocycle 1 for Cu 2+ was studied with the methods of fluorescence and ultraviolet spectroscopy, 1 H NMR titrations, MS and IR spectra. The experimental results displayed macrometallocycle 1 can distinguish Cu 2+ from other cations effectively. K SV value of 5.68 × 10 5 M −1 for 1·Cu 2+ based on a 1:1 association equation analysis was obtained through fluorescence titrations. The detection limit was calculated as 1.5 × 10 −7 mol/L, which indicated that 1 is sensitive for Cu 2+ . In literatures, some peptide sensors for Cu 2+ were reported [71][72][73][74][75][76] , and their association constants and detection limits were in the ranges of 10 4 -10 6 M −1 and 10 −5 -10 −7 mol/L. Compared with these sensors, sensor 1 showed similar binding ability and good sensitivity to Cu 2+ . Further investigation for new NHC metal complexes from L 1 H 2 ·Cl 2 , L 2 H 2 ·Cl 2 and similar to precursors are still under way.

Experimental Section
General procedures. N-ethyl-imidazole and N-n butyl-imidazole were prepared according to the methods of literature reported 67,77 . Schlenk techniques were used in all manipulations. All the reagents for synthesis and analyses were of analytical grade and used without further purification. Melting points were determined with a Boetius Block apparatus. 1 H and 13 C NMR spectra were recorded on a Varian Mercury Vx 400 spectrometer at 400 MHz and 100 MHz, respectively. Chemical shifts, δ, are reported in ppm relative to the internal standard TMS for both 1 H and 13 C NMR. J values are given in Hz. Elemental analyses were measured using a Perkin-Elmer 2400 C Elemental Analyzer. The fluorescence spectra were performed using a Cary Eclipse fluorescence   spectrophotometer. UV-vis spectra were recorded on a JASCO-V570 spectrometer. EI mass spectra were recorded on a VG ZAB-HS mass spectrometer (VG, U.K.). IR spectra (KBr) were taken on a Bruker Equinox 55 spectrometer.   X-Ray data collection and structure determinations. A Bruker Apex II CCD diffractometer were used for the collection of diffraction data of 1-4 78 . The structure was solved with the SHELXS program 79 . Figures 1-4 were formed via employing Crystal-Maker 80 . Other details for structural analysis and crystallographic data was listed in Tables 1 and 2.