The synthesis, characterization and application of the binol-cages of R-/S-enantiomers

INTRODUCTION

The global attention on environmental pollution caused by heavy metal ions persists due to their potential to inflict severe harm on both animals and humans^[1]. Many highly sensitive and selective probes have been developed and successfully used in practical life in order to detect metal ions in the environment. For example, probes synthesized by Mohandoss et al. were widely used for detecting Fe³⁺, Al³⁺, Cu²⁺, and Ag⁺ ions^[2-4]. As is well known, mercury is extremely harmful to organisms and the environment, especially to mammals. It can not only enter the human body directly through the skin, digestive system, or respiratory system but also accumulate in the environment, ultimately entering the human body through the food chain and causing serious damage to the central nervous and endocrine systems of the human body. In addition, all stable forms of Hg (i.e., metallic, inorganic, and organic) cause neurological poisoning in fish, wildlife, and humans. It causes several disorders in the liver, kidney, heart, stomach, brain, intestine, and immune system^[5-7]. According to the United States Environmental Protection Agency (US EPA), the maximum possible limit of Hg²⁺ ions in drinking water is 2 ppb, but the amount of Hg²⁺ ions in drinking water excess to the limit may pose a risk to human health and the environment (Environmental Protection Agency, 2001). Despite these drawbacks, mercury and mercury salts still widely exist in nature in the form of elemental mercury, inorganic, and organic compounds. Human activities, industrial production, and natural phenomena can all lead to the production of various forms of mercury compounds^[8-10]. Therefore, it is necessary to seek efficient, simple, and fast new ways to detect mercury ions in the atmosphere and wastewater in order to protect the Earth's environment^[11-25].

In recent years, mercury ions have been detected by a large number of analytical methods, including cold vapor atomic fluorescence spectroscopy^[26], inductively coupled plasma mass spectrometry (ICP-MS)^[27], gas chromatography^[28], neutron activation analysis (NAA)^[29], and high-performance liquid chromatography^[30]. However, these technologies often require cumbersome operations and expensive instruments. As a result, the development of more efficient and facile approaches for mercury ion detection is highly appealing yet challenging.

Supramolecular chemistry, as an interesting field, has successively attracted much attention due to its unique structures and extensive applications in sensing^[31,32], catalysis^[33,34], separation^[35,36], gas adsorption^[37,38], biological imitation^[39], and so on^[40,41]. The intriguing three-dimensional molecular cages have proven to be superior alternative architectures, garnering considerable interest owing to their outstanding structures, properties, applications, and the diverse array of charming skeletons successfully generated via coordinative self-assembly, such as octahedrons^[42,43], tetrahedrons^[44-47], spirals^[48], capsules^[49], etc. Additionally, subcomponent self-assembly, as the most powerful bottom-up strategy, has attracted great attention by preparing highly organized structures with different functions in an "order-out-of-chaos" and well-controlled manner^[50-53]. Comparatively, the discrete supramolecular structures formed by covalent bonds remain underdeveloped, possibly due to the dynamic and unstable interactions^[54,55]. In this context, molecular cages formed by imine condensation have developed rapidly^[56-58] since the groundbreaking work of Quan et al. in 1991^[59]. Despite the impressive progress made by chemists in this field, the involvement of multiple functional components in a single supramolecular system causes further difficulties for synthesis and its evocative biological heuristic process. Meanwhile, there are relatively few examples of detecting mercury ions as fluorescent probes, although covalent supramolecular cages have been widely reported as examples of fluorescent probes^[60-64]. Therefore, it is crucial to create novel covalent supramolecular cages and use them for mercury ion detection.

Herein, we purposefully designed and constructed covalent organic cages with (R)- or (S)-binaphthyl skeletons as ligands and tri(2-aminoethyl)amine as vertices by amine aldehyde condensation-driven collaborative self-assembly. Notably, the ether chains were introduced on three symmetric binaphthyl ligands to form 20-crown-6. In addition, mercury ions were successfully recognized through fluorescence testing. To our delight, these covalent organic nanocages, which possess chemiluminescent properties, can serve as a fluorescence sensor to detect these ions. The successful synthesis of the cage had been demonstrated by NMR, mass spectrometry, Fourier transform infrared (FT-IR), and molecular simulations.

RESULTS AND DISCUSSION

As shown in Scheme 1, our investigation was commenced by using the compound 1 [3, 6, 9, 12-tetraoxatetradecane-1, 14-diol] and compound 3 (R)- or (S)-BINOL (binaphthol) as the starting substrates. Initially, (S)-4 was treated with potassium t-butoxide at 75 °C, and the resultant system was reacted with pentaethylene glycol di-p-toluenesulfonate 2 to afford the product (S)-5 in 40% yield. A Suzuki coupling reaction of bromide (S)-5 with 4-formylphenylboronic acid gave an important precursor (S)-7 in 30% yield, which has two aldehyde functionalities attached to the binaphthyl backbone. On the basis of the successful synthesis of (S)-7, covalent organic nanocage cage (S)-8 was prepared by co-self-assembly with tris(2-aminoethyl)amine in a yield of 55% as a yellow solid. The same method was used to synthesize (R)-8 for a 53% yield. The chemical structures of these chiral cages were adequately characterized by ¹H NMR, mass spectrometry, and correlation spectroscopy (COSY) spectral information.

Scheme 1. Synthetic route to chiral organic cages.

The desired covalent organic cages were successfully synthesized according to ¹H NMR spectroscopy [Supplementary Figures 1-20]. As shown in Figure 1, the protons H^a-H^g located on the binaphthyl backbone and phenyl moiety were represented by eight independent sets of peaks in the spectra of both ligand 7 and cage 8. It is worth noting that the characteristic peak of aldehyde in the 7 disappeared due to the condensation with amine, indicating that compound 7 was completely reacted. Moreover, the signal of Schiff base hydrogen appeared at 8.01 ppm, also proving the successful synthesis of the target cage structure. Chemical shifts were observed for protons H^a (from 7.96 ppm to 7.43 ppm) and H^b (from 7.84 ppm to 6.99 ppm), as compared with the free aromatic ligand 7, because of the electronic effect on the imine group. 2D NMR spectroscopies, such as COSY and diffusion-ordered spectroscopy (DOSY), provided more precise structural data in support of the formation of covalent organic cages [Supplementary Figures 21 and 22]. Moreover, additional evidence for the formation of [2+3] imine cages could also be provided by electrospray ionization mass spectrometry (ESI-MS). Experimental isotope patterns for the peak of [8+2H]²⁺ are strongly consistent with theoretical isotope patterns, suggesting that covalent organic supramolecular cages were constructed successfully via self-assembly. All main peaks of mass spectra also indicated the formation of corresponding cage architectures [Figure 1E and Supplementary Figure 23].

Figure 1. Part of the ¹H NMR (600 MHz, CDCl₃, 22 ^oC) spectra of (S)-7 (A); (S)-8 (B); (R)-8 (C); (R)-7 (D) and experimental ESI-MS of the (S)-8 of [8+2H]²⁺ (blue) and simulated mass spectra (red) (E).

FT-IR characterization has also proven to be an important tool for detection of changes in functional groups [Figure 2]. The infrared spectrum of compound (S)-7 displayed a strong aldehyde stretching vibrational peak at 1,695 nm alongside a stretching vibration peak at 1,642 nm due to the imine formation.

Figure 2. Fourier transform infrared spectroscopy of cage (black line) and compound 7 (red line).

The UV-Vis and fluorescence spectra also show the structural characteristics of the cage, as shown in Figure 3; the UV-Vis absorption spectra for the cage display two main peaks at 322 nm and 280 nm, which correspond to the n-π*transitions (C=N) and π-π* transitions (C=C). The UV-Vis absorption spectra for the ligand exhibited a unique peak at 337 nm, which was attributed to the existence of n-π* transitions (C=O). Furthermore, the emission wavelengths of (S)-8 and (S)-7 are 414 nm and 458 nm in the fluorescence spectra, respectively.

Figure 3. UV-Vis and fluorescence spectra of compound (S)-7 and (S)-8.

A simulated molecular model of cages was constructed due to many unsuccessful efforts to obtain the crystal structure of cages [Figure 4 and Supplementary Figure 24]. Analysis of the simulated cages showed the expected [2+3] structure, with two amine moieties connected by three side arms based on binol in the top and bottom views. The structure of cages had a C₃ symmetric topology and three C₂ axes perpendicular to the C₃ axis.

Figure 4. The simulated crystal structures of (S)-8 (A) and (R)-8 (B). (N, blue, O, red, C, gray, and hydrogens are omitted for clarity).

The inherent chirality of imine cages was characterized by circular dichroism (CD) spectroscopy. As shown in Figure 5, (R)-8 and (S)-8 exhibited mirror CD spectra with highly pronounced Cotton effects in the wavelength range of 240 to 400 nm. The CD spectra of cages displayed one strong band at 286 nm, which indicated the absorption of naphthalene groups on the binaphthyl backbone. Therefore, we have successfully synthesized the covalent supramolecular cages through NMR, ESI-MS, molecular simulations, CD, and FT-IR characterization methods.

Figure 5. (A) The UV-Vis spectra of (R)-8 and (S)-8 in CDCl₃ (c = 0.01 mM); (B) CD spectra of (R)-8 and (S)-8 in CDCl3 (c = 0.1 mM).

The heavy metal ion Hg²⁺ poses a significant threat to the environment due to its high affinity for thiol groups in proteins and enzymes. Additionally, it is important to note that mercury ions are not biodegradable. As a result, even at extremely low concentrations, they can have detrimental effects on the kidneys and brain, potentially leading to various diseases. The US EPA stipulates that the maximum concentration of mercury ions in drinking water is 10 μM. Consequently, in this work, the cage can be used as a fluorescence probe because adding mercury ions to the solution does not decompose the cage [Supplementary Figure 25]. As shown in the results, the fluorescence intensity of the supramolecule cage gradually increases with the increasing concentration of mercury ions [Figure 6A and B]. Meanwhile, the covalent supramolecule cage shows a redshift of 48 nm in the emission spectra. The addition of mercury ions created an umbrella-shaped structure that made the molecule more rigid. As a result, electron delocalization between the crown ether and imine parts became more viable. This enhanced dipole moment and charge separation, leading to a significant redshift of 48 nm in the fluorescence emission spectrum [Figure 6C]^[65,66]. Moreover, the UV-Vis titration experiment provided important quantitative information [Supplementary Figures 26-30]. As shown in Supplementary Figure 26, the absorbance at 280 nm gradually decreased and underwent a shift to red with the addition of mercury ions. The change remained basically unchanged after adding 5 eq. of mercury ions. Therefore, the binding ratio of 1:5 could be obtained by using the molar ratio plot [Supplementary Figure 28]. Two linearly fitted lines intersected at a point with a horizontal coordinate of 5 eq., proving that the binding ratio of the cage to mercury ions was 1:5. Meanwhile, the UV-Vis titration data could be well fitted by the Hill function. Based on the fitting results, the binding constant K = 2.57 × 10⁵ M^-1 could be obtained [Supplementary Figure 29]. Finally, the detection limit was calculated by showing a good linear relationship between the absorbance of the cage and 0-5 eq. of mercury ions at 280 nm [Supplementary Figure 30]. It was worth noting that the CD spectrum of the cage could also detect these ions. The CD spectrum underwent significant changes with the continuous addition of mercury ions. On the contrary, no significant changes were observed after the addition of copper ions, suggesting that copper ions could not bind to the cage [Supplementary Figure 27]. In order to prove that the cage can effectively recognize mercury ions, common metal ions are also used for detection [Figure 7]. It is obvious from Figure 7 that Hg²⁺ ions have the most fluorescence enhancement effect on the supramolecule cage.

Figure 6. Fluorescence titration of (R)-8 (10 µM) upon addition of Hg²⁺ ion up to 10 µM in CHCl₃ (A) and the plot of fluorescence intensity versus Hg²⁺ concentration (B) (λ_ex = 322 nm, Silt = 2, 2); (C) Mercury ion detection mechanism.

Figure 7. (A) Emission spectra and (B) bar graph of (R)-8 (10 µM) with various metal ions (10 µM) in CHCl₃ (λ_ex = 322 nm, slit width = 2, 2 nm).

In order to further explore the sensing ability of the cage for the Hg²⁺ ion, selective recognition experiments of the cage (0.2 mM) were also conducted by using different concentrations of Hg²⁺ (0.7, 1.4, 2.1 mM) in CHCl₃ [Figure 8]. The results, as depicted in Figure 8, exhibit distinct changes in color that can be obviously observed by the naked eye. The color of the CHCl₃ with spiked Hg²⁺ transformed from colorless to light yellow. The ability of cages to detect mercury ions in different pH solutions was also tested separately, and similar results were observed [Supplementary Figure 31]. In addition, the cage had been compared with supramolecular materials that were used for detecting mercury ions in recent years [Supplementary Table 1]. According to [Supplementary Table 1], the diverse detection methods of the cage possessed unique advantages. Therefore, it is possible to detect the presence of mercury ions without the use of any additional instruments and make the process more efficient and expedient.

Figure 8. Hg²⁺ion detection diagram based on (R)-8.

CONCLUSIONS

In conclusion, we have demonstrated the formation of two chiral organic cages using chiral binaphthyl as the main building block. The structures of the chiral cages were characterized using ¹H NMR, 2D NMR, mass spectroscopy, molecular simulation, CD, and FT-IR characterization analyses. Furthermore, the absorbance of cages exhibited a good linear correlation with Hg²⁺ concentrations with detection limits of 1.9 × 10^-7 M. On the other hand, the cage could be used for selective recognition of Hg²⁺ ions as a supramolecular probe, especially allowing for the direct observation of changes in solution systems, thereby contributing to the protection of environmental pollution, which may contribute to the development of new approaches for utilizing supramolecular cages as probes in the future.