Preparation of Macrometallocycle and Selective Sensor for Copper Ion

Abstract

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.

Introduction

The detection of Cu²⁺ occupies an important position in host-guest chemistry because it plays a crucial part in chemistry, biology and environmental science^1,2,3. As a trace element in the body, copper are key components of hemocyanin and some enzymes. Ingesting excess or deficient Cu²⁺ will cause serious illness, such as Alzheimer’s and Wilson’s diseases, haematological manifestations and liver damage^{4,5,6,7,8,9,10,11,12}. Excess Cu²⁺ can also destroy the aquatic ecosystem, and disturb the nutrient absorption and transport of some plants¹³. Among the detection of Cu²⁺, the fluorescent chemosensor is one of significant tools due to its high sensitivity and the simplicity of equipment^14,15,16. So far, a variety of types of fluorescent chemosensors for Cu²⁺ have been reported, such as organic small molecules and MOFs^{17,18,19,20,21,22,23}. Besides, Liu and co-workers reported a sensor based on porous conjugated polymers for Cu²⁺, and it is high sensitivity and selectivity²⁴. Though some chemosensors for Cu²⁺ have appeared, the design and synthesis of new practical chemosensors are still desirable.

In the process of searching for suitable chemosensors for Cu²⁺, we focused on N-heterocyclic carbene (NHC) metal complexes because of their diverse structures, such as macrocycle^{25,26,27,28,29}, molecular rectangle^30,31,32 and groove^33,34. In a large number of complexes, cyclic NHC metal complexes have favorable recognition capability for metal ions^{35,36,37,38,39}, because this kind of host can capture effectively metal ions through several kinds of forces (electrostatic force, M···M interactions, M···X interactions and M···π interactions). Herein, we report the synthesis of bis-imidazolium salts 1,8-bis[2′-(N-R-imidazoliumyl)acetylamino]naphthalene chloride (L¹H₄·Cl₂: R = Et; L²H₄·Cl₂: R = ⁿBu), as well as the preparation and structure of four NHC complexes [L¹H₂Ag]Cl (1), [L¹Ni] (2), [L²Ni] (3) and [L¹H₂Hg(HgCl₄)] (4). Additionally, we studied the selective recognition of macrometallocycle 1 for Cu²⁺ with the methods of fluorescence and ultraviolet spectroscopy, ¹H NMR titrations, MS and IR spectra.

Results and Discussion

Synthesis and characterization of L¹H₄·Cl₂ and L²H₄·Cl₂

As shown in Fig. 1, 1,8-diaminonaphthalene reacted with chloroacetyl chloride to give 1,8-di(2′-chloroacetylamino)naphthalene, which further reacted with N-R-imidazole (R = Et or ⁿBu) to generate bis-imidazolium salts L¹H₄·Cl₂ and L²H₄·Cl₂. Precursors L¹H₄·Cl₂ and L²H₄·Cl₂ remain stable in the air, and can be dissolved in DMSO, dichloromethane and acetonitrile, but their solubility is poor in benzene, diethyl ether and petroleum ether. In the ¹H NMR spectra of L¹H₄·Cl₂ and L²H₄·Cl₂, the proton signals (NCHN) of imidazolium appear at δ = 9.47 and 9.50 ppm, and these values are analogous to those of known imidazolium compounds^{33,40,41,42,43,44,45,46}.

Figure 1

Preparation of Precursors L¹H₄·Cl₂ and L²H₄·Cl₂.

Synthesis and general characterization of complexes 1–4

The synthesis of NHC silver(I) complex [L¹H₂Ag]Cl (1) was accomplished via the reaction of L¹H₄·Cl₂ with Ag₂O in CH₃CN/DMSO (Fig. 2). The reactions of L¹H₄·Cl₂ or L²H₄·Cl₂ with NiCl₂ in the presence of K₂CO₃ in CH₃CN/DMSO afforded NHC nickel(II) complexes [L¹Ni] (2) and [L²Ni] (3). The reaction of L¹H₄·Cl₂ with HgCl₂ in the presence of KO^tBu in CH₃CN/DMSO gave NHC mercury(II) complex [L¹H₂Hg(HgCl₄)] (4).

Figure 2

Preparation of Complexes 1–4.

The crystals of complexes 1–4 were obtained via slow adding Et₂O to their solutions. Complexes 1–4 can be dissolved in DMSO and CH₃CN, but they are scarce soluble in benzene, diethyl ether and petroleum ether. The solution of complex 1 is slightly light-sensitive. The proton signals (NCHN) of imidazolium disappear in the ¹H NMR spectra of 1–4 due to the introduction of metals, and other proton signals are analogous to L¹H₄·Cl₂ or L²H₄·Cl₂. In the ¹³C NMR spectra of 1, no carbene carbon signal is found, and this phenomenon may be the fluxional behavior of the NHC silver(I) complexes^47,48,49. The carbene carbon signals of 2–4 are observed at 175.0–176.8 ppm, which are consistent with other NHC metal complexes in literatures^{50,51,52,53,54,55,56,57,58,59,60}.

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)°.

Figure 3

Perspective view of 1 and anisotropic displacement parameters depicting 50% probability. Selected bond lengths (Å) and angles (°): Ag(1)-C(3) 2.100(8), Ag(1)-C(20) 2.074(8); C(3)-Ag(1)-C(20) 175.3(3), N(1)-C(3)-N(2) 104.0(7), N(5)-C(20)-N(6) 104.5(6).

Figure 4

Perspective view of 2 and anisotropic displacement parameters depicting 50% probability. Selected bond lengths (Å) and angles (°):C(3)-Ni(1) 1.858(5), C(20)-Ni(1) 1.864(5), N(1)-Ni(1) 1.933(4), N(2)-Ni(1) 1.925(4); C(3)-Ni(1)-C(20) 91.2(2), N(1)-Ni(1)-N(2) 94.6(2), N(3)-C(3)-N(4) 104.9(4), C(3)-Ni(1)-N(1) 89.7(1), C(20)-Ni(1)-N(2) 88.1(2).

Figure 5

Perspective view of 3 and anisotropic displacement parameters depicting 50% probability. Selected bond lengths (Å) and angles (°): N(3)-Ni(1) 1.918(1), C(4)-Ni(1) 1.900(2), C(23)-Ni(1) 1.871(2), N(4)-Ni(1) 1.929(1); C(23)-Ni(1)-C(4) 97.9(1), N(3)-Ni(1)-N(4) 87.0(8), N(1)-C(4)-N(2) 103.5(1), N(5)-C(23)-N(6) 104.9(1), C(4)-Ni(1)-N(3) 91.8(9), C(23)-Ni(1)-N(4) 84.4(9).

Figure 6

Perspective view of 4 and anisotropic displacement parameters depicting 50% probability. Selected bond lengths (Å) and angles (°): Hg(1)-C(5) 2.073(6), Hg(1)-C(20) 2.081(7), Hg(1)-Cl(1) 2.880(1), Hg(2)-Cl(1) 2.514(1), Hg(2)-Cl(2) 2.557(1), Hg(2)-Cl(3) 2.437(1), Hg(2)-Cl(4) 2.418(1); C(5)-Hg(1)-C(20) 168.6(2), N(1)-C(5)-N(2) 106.3(5), N(5)-C(20)-N(6) 106.2(6).

In complex 1, the arrangement of C(3)-Ag(1)-C(20) is almost linear with the angle of 175.3(3)°, and the distances of Ag(1)-C(3) and Ag(1)-C(20) are 2.074(8) Å and 2.100(8) Å. Both are comparable with those of known NHC Ag(I) complexes^47,48,49.

In complexes 2 or 3, two acetylamino groups (-CONH-) and two imdazolium moieties of precursors L¹H₄·Cl₂ or L²H₄·Cl₂ are deprotonated in the presence of K₂CO₃. 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 1.858(5)–1.900(2) Å and 1.918(1)–1.933(4) Å, respectively. The bond angles of C-Ni-C, N-Ni-N and C-Ni-N are 91.2(2)–97.9(1)°, 87.0(8)−94.6(2)° and 84.4(9)–169.9(9)°, respectively. Similar values were also reported in other literatures about NHC Ni(II) complexes⁶¹.

Both of Hg(1) and Hg(2) in complex 4 are tetra-coordinated. The distances of Hg(1)-C(5) and Hg(1)-C(20) are 2.073(6) Å and 2.081(7) Å, and the bond angle of C(5)-Hg(1)-C(20) is 168.6(2)°. The distances of Hg(2)-Cl (2.418(2)–2.557(1) Å) are shorter than that of Hg(1)-Cl(1) (2.880(1) Å). A distorted Hg₂Cl₂ quadrangular arrangement is formed by Hg(1), Cl(1), Hg(2) and Cl(2), in which the dihedral angle between the Cl(1)-Hg(1)-Cl(2) plane and the Cl(1)-Hg(2)-Cl(2) plane is 30.5(8)°. The Hg···Hg separation of 3.815(5) Å suggests the nonexistence of metal-metal interactions between both Hg(II) ions (van der Waals Radii of mercury = 1.70 Å)^62,63.

Recognition of Cu²⁺ using 1 as a chemosensor

The screening experiments of complexes 1–4 for some cations (Li⁺, Na⁺, K⁺, NH₄⁺, Ag⁺, Ca²⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Cd²⁺, Cr³⁺, Al³⁺, Pb²⁺ and Hg²⁺, and their anions are NO₃⁻) via fluorescence spectroscopy in CH₃CN 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 Cu²⁺, 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 Cu²⁺, the time-dependent plot was measured (Fig. 7). The results showed that the interactions between Cu²⁺ 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 L¹H₄·Cl₂ and complex 1 using 1-aminonaphthalene as fluorescence standard (Φ = 0.39) were measured⁶⁴. The fluorescence quantum yields of L¹H₄·Cl₂ 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.

Figure 7

Plot of fluorescence intensity of 1 (2.0 × 10⁻⁶ mol/L) and Cu²⁺ (20 × 10⁻⁶ mol/L) as a function of time in minutes.

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⁺, NH₄⁺, Ag⁺, Ca²⁺, Co²⁺, Ni²⁺, Zn²⁺, Cd²⁺, Cr³⁺, Al³⁺, Pb²⁺ and Hg²⁺ 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 Cu²⁺. In UV/vis experiment, upon addition of Cu²⁺ 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 Cu²⁺ from other cations effectively.

Figure 8

Fluorescence spectra of 1 (2.0 × 10⁻⁶ mol/L) and 10 equiv. of some cations (Li⁺, Na⁺, K⁺, NH₄⁺, Ag⁺, Ca²⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Cd²⁺, Cr³⁺, Al³⁺, Pb²⁺ and Hg²⁺) in CH₃CN at 25 °C.

In the fluorescence titration experiments (Fig. 9), upon the titration of Cu²⁺ into solutions of 1 in CH₃CN 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_Cu²⁺/C₁ 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 Cu²⁺ was added. The quenching behaviors of Cu²⁺ on the fluorescence of 1 were found to follow a conventional Stern-Volmer relationship^67,68 (equation (1)).

$${F}_{0}/F=1+{K}_{SV}{{C}_{Cu}}^{2+}$$

(1)

where F₀ and F are the fluorescence intensities of 1 in the absence and presence of Cu²⁺, and C_Cu²⁺ is the concentration of Cu²⁺. The equation reveals that F₀/F increases in direct proportion to the increasing concentration of Cu²⁺, and the Stern-Volmer constant K_SV defines the quenching efficiency of Cu²⁺.

Figure 9

Fluorescence titration spectra of 1 (2.0 × 10⁻⁶ mol/L) in the presence of different concentrations of Cu²⁺ in CH₃CN at 25 °C. C_Cu²⁺ for curves 1–21 (from top to bottom) are 0, 0.6, 1.2, 1.8, 2.6, 3.4, 4.2, 5.0, 5.8, 6.8, 7.8, 8.8, 10.0, 11.5, 13.0, 14.5, 16.0, 18.0, 20.0, 22.0, 24.0 × 10⁻⁶ mol/L (λ_ex = 330 nm). Inset: variation of fluorescence quenching F/F₀ of 1 with increasing Cu²⁺ concentration.

The K_SV value for 1·Cu²⁺ was calculated as 5.68 × 10⁵ M⁻¹ (R = 0.999) by using the equation (1) (Fig. S2). As shown in Fig. S3, the detection limit was estimated to be 1.5 × 10⁻⁷ mol/L³⁴. To furthur comfirm the complexation stoichiometry between 1 and Cu²⁺, a Job’s plot analysis at 214 nm was carried out (Fig. 10)^62,63. The χΔA values for 1·Cu²⁺ reached a maximum when molar fractions (χ) of 1 was 0.5, and it indicated stoichiometric ratio was 1:1. Where total concentration was a constant, and ΔA was the discrepancy of the absorption bands.

Figure 10

The Job’s plot of 1 toward Cu²⁺ at 214 nm. χ is the molar fraction of 1. It illustrates the host-guest fluorescence quenching occurs in 1:1 complexation.

To test the ability to resist interference of other cations, the competition experiments were conducted (Fig. S4), where 1 (2.0 × 10⁻⁶ mol/L) was mixed with 5 equiv. of Li⁺, Na⁺, K⁺, NH₄⁺, Ag⁺, Ca²⁺, Co²⁺, Ni²⁺, Zn²⁺, Cd²⁺, Cr³⁺, Al³⁺, Pb²⁺ or Hg²⁺, and then 5 equiv. of Cu²⁺ was added. The presence of other cations did not cause any significant changes in the emission of 1·Cu²⁺.

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⁻, SO₄²⁻, OAc⁻, Cl⁻, NO₃⁻ and CO₃²⁻) (Fig. S5). Thus, the different anions did not obviously influence on the binding between 1 and Cu²⁺. Reversible binding of 1 with Cu²⁺ 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 Cu²⁺ (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 Cu²⁺ again. This result showed that 1 was a good chemosensor for Cu²⁺ with admirable reversibility and regeneration capacity.

Interactions of 1 with Cu²⁺

The potential binding sites of 1 for Cu²⁺ may be oxygen atoms, nitrogen atoms and π systems (including O···Cu²⁺ interactions, N···Cu²⁺ interactions and π···Cu²⁺ interactions). To get detailed information on how 1 bound with Cu²⁺, we studied the data of ¹H NMR titrations (C_Cu²⁺/C₁ was from 0 to 2.0 equiv.) in DMSO-d₆ (Fig. 11). Upon the addition of 1 equiv. of Cu²⁺, 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 Cu²⁺ due to Cu²⁺···N interactions (Fig. 12). The proton signal of Hc on CH₂ attached to C=O shifted to downfield (ca. 0.25 ppm), which may be attributed to electron-withdrawing effect of Cu²⁺ due to Cu²⁺···O interactions. More equivalents of Cu²⁺ did not cause further change of chemical shifts of Hc-Hf (Fig. 11(v,vi)), which showed the combination ratio between 1 and Cu²⁺ was 1:1.

Figure 11

Partial ¹H NMR spectra in DMSO-d₆. (i) 1; (ii) 1 and 0.25 equiv. of Cu²⁺; (iii) 1 and 0.5 equiv. of Cu²⁺; (vi) 1 and 1 equiv. of Cu²⁺; (v) 1 and 1.5 equiv. of Cu²⁺; (vi) 1 and 2 equiv. of Cu²⁺.

Figure 12

The interactions of 1 with Cu²⁺.

Additional evidence for the combination ratio between 1 and Cu²⁺ was obtained through high-resolution mass spectra of 1·Cu²⁺ (Fig. S7). The observation of m/z (318.3) for (1·Cu²⁺)/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·Cu²⁺ were measured for more information about how 1 bound with Cu²⁺. In Fig. S8, we found that several absoption bands have changed after adding Cu²⁺. The υ_(C=O) varied from 1660 cm⁻¹ to 1683 cm⁻¹, υ_(N-H) varied from 3378 cm⁻¹ to 3382 cm⁻¹, and δ_(N-H) varied from 1617 cm⁻¹ to 1629 cm⁻¹, respectively.

By analyzing the structure of 1 and above experiment results, we can conclude that 1 bound with Cu²⁺ mainly through Cu²⁺···O and Cu²⁺···N interactions. Once complex 1·Cu²⁺ 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·Cu²⁺, but unsuccessful.

Conclusion

In conclusion, we prepared and characterized two bis-imidazolium salts L¹H₂·Cl₂ and L²H₂·Cl₂, 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 Cu²⁺ was studied with the methods of fluorescence and ultraviolet spectroscopy, ¹H NMR titrations, MS and IR spectra. The experimental results displayed macrometallocycle 1 can distinguish Cu²⁺ from other cations effectively. K_SV value of 5.68 × 10⁵ M⁻¹ for 1·Cu²⁺ 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 Cu²⁺. In literatures, some peptide sensors for Cu²⁺ were reported^{71,72,73,74,75,76}, and their association constants and detection limits were in the ranges of 10⁴–10⁶ M⁻¹ and 10⁻⁵–10⁻⁷ mol/L. Compared with these sensors, sensor 1 showed similar binding ability and good sensitivity to Cu²⁺. Further investigation for new NHC metal complexes from L¹H₂·Cl₂, L²H₂·Cl₂ and similar to precursors are still under way.

Experimental Section

General procedures

N-ethyl-imidazole and N-ⁿbutyl-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. ¹H and ¹³C 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 ¹H and ¹³C 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.

Synthesis of 1,8-bis(2′-chloroacetyl)diaminonaphthalene

A suspension of 1,8-diaminonaphthalene (10.000 g, 63.2 mmol) and triethylamine (21.0 mL, 151.6 mmol) in CH₂Cl₂ (120 mL) was stirred for 30 min at 0 °C. Then chloroacetyl chloride (11.4 mL, 151.7 mmol) was dropwise added to the suspension above and stirred continually for 3 h at ambient temperature. The mixture was filtered and washed by water to afford 1,8-bis(2′-chloroacetyl)diaminonaphthalene as a yellow powder. Yield: 15.731 g (80%). M.p.: 265–267 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 4.36 (s, 4H, CH₂), 7.52 (t, J = 3.4 Hz, 6H, PhH), 7.90 (t, J = 4.6Hz, 2H, PhH), 10.10 (s, 2H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 43.8 (CH₂), 126.0 (PhC), 127.8 (PhC), 132.18 (PhC), 135.9 (PhC), 165.6 (C=O).

Preparation of 1,8-bis[2′-(N-ethylimidazoliumyl)acetylamino]naphthalene chloride (L¹H₄·Cl₂)

A solution of N-ethyl-imidazole (1.538 g, 16.0 mmol) and 1,8-bis(2′-chloroacetylamino)naphthalene (2.000 g, 6.4 mmol) in DMF (150 mL) was heated to reflux for 7 days with stirring, and precipitated a black powder. The precipitate was collected by filtration and washed with a small portion of DMF to give 1,8-bis[2′-(N-ethyl-imidazoliumyl)acetylamino]naphthalene chloride. Yield: 1.480 g (48%). M.p.: 260–261 °C. Anal. Calcd for C₂₄H₂₈N₆O₂Cl₂: C, 57.25; H, 5.60; N, 16.69%. Found: C, 57.20; H, 5.56; N, 16.68%. ¹H NMR (400 MHz, DMSO-d₆): δ 1.48 (t, J = 7.2 Hz, 6H, CH₃), 4.32 (m, 4H, CH₂), 5.50 (s, 4H, CH₂), 7.59 (s, 2H, PhH), 7.92 (t, J = 15.6 Hz, 4H, PhH), 9.47 (s, 2H, 2-imiH), 11.07 (s, 2H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 15.6 (CH₃), 44.7 (CH₂), 52.1 (CH₂), 121.6 (PhC), 124.7 (PhC), 125.8 (PhC), 126.8 (PhC), 127.9 (PhC), 131.4 (PhC), 136.0 (PhC), 137.7 (PhC), 164.9 (C=O) (imi = imidazolium).

Preparation of 1,8-bis[2′-(N-ⁿbutyl-imidazoliumyl)acetylamino]naphthalene chloride (L²H₄·Cl₂). L²H₄·Cl₂

Was prepared according to the methods of L¹H₂·Cl₂, only N-ethyl-imidazole was replaced by N-ⁿbutyl-imidazole (1.984 g, 16.0 mmol). Yield: 1.790 g (50%). M.p.: 240–242 °C. Anal. Calcd for C₂₈H₃₆N₆O₂Cl₂: C, 60.10; H, 6.48; N, 15.01%. Found: C, 60.22; H, 6.32; N, 15.23%. ¹H NMR (400 MHz, DMSO-d₆): δ 0.93 (s, 6H, CH₃), 1.30 (m, 4H, CH₂), 1.82 (s, 4H, CH₂), 4.28 (s, 4H, CH₂), 5.55 (s, 4H, CH₂), 7.59 (t, J = 7.4 Hz, 4H, PhH), 7.97 (m, 6H, PhH), 9.50 (s, 2H, 2-imiH), 11.16 (s, 2 H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 13.7 (CH₃), 19.2 (CH₂), 31.8 (CH₂), 49.0 (CH₂), 52.2 (CH₂), 121.9 (PhC), 124.8 (PhC), 125.8 (PhC), 126.8 (PhC), 127.8 (PhC), 131.4 (PhC), 136.0 (PhC), 138.0 (PhC), 164.9 (C=O).

Preparation of [L¹H₂Ag]Cl (1)

The mixture of L¹H₄·Cl₂ (0.100 g, 0.2 mmol) and Ag₂O (0.046 g, 0.2 mmol) in DMSO (2.5 mL) and CH₃CN (12.5 mL) was heated to reflux for 24 h with stirring. After filtration, the solvent was evaporated to 5 mL, and the yellow powder of 1 was obtained after adding 5 mL of diethyl ether. Yield: 0.040 g (36%). M.p.: 192–194 °C. Anal. Calcd for C₂₄H₂₆AgN₆O₂Cl: C, 50.23; H, 4.56; N, 14.64%. Found: C, 50.44; H, 4.42; N, 14.52%. ¹H NMR (400 MHz, DMSO-d₆): δ 1.43 (t, J = 17.5 Hz, 6H, CH₃), 4.20 (q, 4H, CH₂), 5.06 (s, 4H, CH₂), 7.29 (m, 4H, PhH), 7.60 (d, J = 88 Hz, 4H, PhH), 8.34 (s, 2 H, PhH), 9.29 (s, 2H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 17.3 (CH₃), 46.2 (CH₂), 121.0 (CH₂), 124.1 (PhC), 125.5 (PhC), 135.9 (PhC), 166.3 (C=O).

Preparation of [L¹Ni] (2)

NiCl₂ (0.052 g, 0.4 mmol) was mixed with L¹H₄·Cl₂ (0.100 g, 0.2 mmol) and K₂CO₃ (0.138 g, 1.0 mmol) in DMSO (2.5 mL) and CH₃CN (12.5 mL), and the reaction kept going for 24 h at 60 °C with stirring. After filtration, the solvent was evaporated to 5 mL, and the pale yellow powder of 2 was obtained after adding 5 mL of diethyl ether. Yield: 0.040 g (40%). M.p.:>320 °C. Anal. Calcd for C₂₄H₂₄NiN₆O₂: C, 59.16; H, 4.96; N, 17.25%. Found: C, 59.32; H, 4.87; N, 17.43%. ¹H NMR (400 MHz, DMSO-d₆): δ 1.06 (t, J = 7.2 Hz, 6H, CH₃), 3.41 (q, J = 6.9 Hz, 4H, CH₂), 4.50 (t, J = 3.2 Hz, 4H, CH₂), 6.70 (s, 2H, PhH), 7.11 (t, J = 7.6 Hz, 2H, PhH), 7.28 (d, J = 2.0 Hz, 2H, PhH), 7.40 (d, J = 3.0 Hz, 2H, PhH), 7.55 (d, J = 0.5 Hz, 2H, PhH). ¹³C NMR (100 MHz, DMSO-d₆): δ 15.6 (CH₃), 44.5 (CH₂), 65.3 (CH₂), 121.5 (PhC), 122.2 (PhC), 124.5 (PhC), 135.4 (PhC), 166.6 (C=O), 175.0 (2-imiC).

Preparation of [L²Ni] (3)

[L²Ni] (3) was prepared according to the methods of 2, only L¹H₄·Cl₂ was replaced by L²H₄·Cl₂ (0.100 g, 0.2 mmol). Yield: 0.020 g (20%). M.p.: >320 °C. Anal. Calcd for C₂₈H₃₂NiN₆O₂: C, 61.90; H, 5.93; N, 15.46%. Found: C, 61.78; H, 5.84; N, 15.58%. ¹H NMR (400 MHz, DMSO-d₆): δ 0.70 (t, J = 23 Hz, 6H, CH₃), 1.07 (m, 4H, CH₂), 1.44 (m, 4H, CH₂), 3.80 (t, J = 48.4 Hz, 4H, CH₂), 5.03 (s, 4H, CH₂), 6.78 (s, 2H, PhH), 7.09 (t, J = 7.8 Hz, 2H, PhH), 7.29 (d, J = 8.0 Hz, 2H, PhH), 7.36 (s, 2H, PhH), 7.55 (d, J = 0.8 Hz, 2H, PhH). ¹³C NMR (100 MHz, DMSO-d₆): δ 13.3 (CH₃), 19.1 (CH₂), 30.3 (CH₂), 49.5 (CH₂), 53.7 (CH₂), 112.4 (PhC), 112.5 (PhC), 113.2 (PhC), 116.0 (PhC), 121.7 (PhC), 122.2 (PhC), 123.9 (PhC), 132.0 (PhC), 165.1 (C=O), 175.0 (2-imiC).

Preparation of [L¹H₂Hg(HgCl₄)] (4)

HgCl₂ (0.110 g, 0.4 mmol) was mixed with L¹H₄·Cl₂ (0.100 g, 0.2 mmol) and KOBu^t (0.056 g, 0.5 mmol) in DMSO (2.5 mL) and CH₃CN (12.5 mL). The solution was heated to 80 °C for 24 h with stirring. After filtration, the solvent was evaporated 10 mL, and the pale brown powder of 4 was obtained after adding 5 mL of diethyl ether. Yield: 0.080 g (40%). M.p.: > 320 °C. Anal. Calcd for C₂₄H₂₆Hg₂N₆O₂Cl₄: C, 29.61; H, 2.69; N, 8.63%. Found: C, 29.76; H, 2.58; N, 8.77%. ¹H NMR (400 MHz, DMSO-d₆): δ 1.46 (t, J = 7.2 Hz, 6H, CH₃), 4.56 (m, 4H, CH₂), 5.57 (s, 4H, CH₂), 7.51 (d, J = 4.8 Hz, 4H, PhH), 7.77 (d, J = 18.8 Hz, 4H, PhH), 7.88 (t, J = 6.4 Hz, 4H, PhH), 10.22 (s, 2H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 16.0 (CH₃), 45.6 (CH₂), 52.7 (CH₂), 122.2 (PhC), 125.0 (PhC), 125.4 (PhC), 125.5 (PhC), 127.1 (PhC), 131.8 (PhC), 135.4 (PhC), 165.0 (C=O), 176.8 (2-imiC).

Fluorescence titrations

The stock solution (1.0 × 10⁻⁴ M) of the host was prepared and diluted to the suitable concentration with CH₃CN. The stock solutions (1.0 × 10⁻³ M or 1.0 × 10⁻⁴ M) of guest were prepared and diluted in the same solvent. Test solutions were prepared through placing 0.2 mL of host stock solution into a 10 mL volumetric flask, and the appropriate amount of the stock solutions (1.0 × 10⁻³ M or 1.0 × 10⁻⁴ M) of guest were added with a microsyringe. The mixture solutions were diluted to 10 mL with CH₃CN to prepare test solutions. The concentrations of guest in the test solutions were from 0 to 24.0 × 10⁻⁶ M, and the concentration of host stayed the same (2.0 × 10⁻⁶ M). The test solutions were kept at 25 °C for 8–10 minutes, and then fluorescence spectra were recorded with the excitation wavelength at 330 nm, and the excitation and emission slits are 5 nm and 5 nm. Statistical analysis of the data was carried out using Origin 8.0. CH₃CN used in the titrations was freshly distilled.

Quantum yields

Fluorescence quantum yields (Φ) of L¹H₄·Cl₂ and complex 1 were determined by using 1-aminonaphthalene (Φ = 0.39) in CH₃CN as the standard compound. Fluorescence quantum yields could be calculated according to the equation (2) below⁶⁴.

$${{\rm{\Phi }}}_{{\rm{U}}}={{\rm{\Phi }}}_{{\rm{S}}}({A}_{S}/{A}_{U})({F}_{U}/{F}_{S}){({{\rm{n}}}_{{\rm{U}}}/{{\rm{n}}}_{{\rm{S}}})}^{2}$$

(2)

where Φ_U, A_U and F_U are the quantum yield, the absorbance and the emission intensity for L¹H₄·Cl₂ or complex 1. Φ_S, A_S and F_S are the quantum yield, the absorbance and the emission intensity for 1-aminonaphthalene. n_U and n_S are the average refractive index of the sample solution (n_U = n_S = n_acetonitrile).

Method for Job’s plot

The stock solution (1.0 × 10⁻⁴ M) of the host was prepared and diluted to the suitable concentration with CH₃CN. The stock solutions (1.0 × 10⁻⁴ M or 1.0 × 10⁻³ M) of guest were prepared and diluted in the same solvent. The molar fractions of host and guest in the test solutions were from 1 to 0 and 0 to 1, respectively. The total concentration is 4.0 × 10⁻⁵ M and different amounts of host and guest solutions were placed into a 10 mL volumetric flask using a microsyringe, and then diluted to 10 mL. The test solutions were kept at 25 °C for 8–10 minutes, and then absoption spectra were measured. Statistical analysis of the data was carried out using Origin 8.0.

X-Ray data collection and structure determinations

A Bruker Apex II CCD diffractometer were used for the collection of diffraction data of 1–4⁷⁸. The structure was solved with the SHELXS program⁷⁹. Figures 1–4 were formed via employing Crystal-Maker⁸⁰. Other details for structural analysis and crystallographic data was listed in Tables 1 and 2.

Table 1 Summary of crystallographic data for 1–4.

Table 2 Summary of crystallographic data for 3 and 4.