Recently, porous organic materials, including porous organic polymers (POPs), metal-organic frameworks (MOFs), and covalent organic frameworks (COFs) have emerged as versatile platforms for the deployment of catalysts because of their porous nature and high surface area [1-7]. However, MOFs and COFs are unstable under many circumstances such as in acidic, basic, and moist environments because of the connection of the coordination bonds in MOFs and the presence of imine or boronate ester groups in COFs, which severely limits their applications in catalysis [8, 9]. In contrast, POPs are advantageous compared with MOFs and COFs in terms of hydrothermal stability and chemical robustness because of the stronger linkage of their versatile covalent bonds apart from imines and boronate esters [10-14]. Thus, POPs can be designed as an ideal platform for incorporating molecular catalytic modules into highly stable and recyclable heterogeneous catalytic systems by taking advantage of their permanent porosity and the ability to tune their compositions and properties at the molecular level [15-17].

Compared with the large number of reported POPs, chiral POPs remain scarce. Lin et al. [18] developed the synthesis process for binaphthol-ligated, Ti-incorporating porous cross-linked polymers for catalytic asymmetric diethylzinc addition to aldehydes. Wang and co-workers [19] described the construction of a α, α, α′, α′-tetraaryl-1, 3-dioxolane-4, 5-dimeth-anol (TADDOL)-embedded chiral POP catalyst and its structure-activity relationship. In addition, to date, most research has been focused on microporous chiral POPs [18, 20-22]; studies aimed at the syntheses of advanced hierarchical (in combination with micropores and mesopores) chiral POPs are even more scarce [23, 24]. Abundant micropores generally favour a high density of catalytically active sites distributed in the catalyst particles, whereas mesopores improve the mass transfer efficiency to enable free contact of the substrate with catalytically active sites in heterogeneous catalysis [25-28]. Therefore, the synthesis of hierarchical chiral POPs is of great interest. The C₂-symmetric BINAP (2, 2'-bis (diphenylphosphino)-1, 1'-binaphthyl) ligand, which was reported by Noyori et al. [29] in 1980, represents one of the most efficient ligands. BINAP and its derivatives have been widely used in asymmetric catalysis and extensively investigated, especially in asymmetric hydrogenation. The heterogenization of BINAP has been a long-term research area in heterogeneous asymmetric catalysis [30-38]. Using the BINAP ligand, Xiao et al. [23] synthesized a vinyl-modified BINAP dioxide through the steps of oxidation, nitration, reduction, and acylation, which was denoted as 5, 5'-diacryloylamino BINAP dioxide. After copolymerization with divinyl benzene (DVB) and reduction with HSiCl₃, porous cross-linked p olymers (PCPs) with BINAP (PCP-BINAP) were obtained. The prepared heterogeneous catalyst Ru/PCP-BINAP exhibited high activity and recyclability in the asymmetric hydrogenation of β-keto esters. Inspired by this creative work, we recently reported excellent asymmetric hydroformylation catalysts through radical polymerization of (S)-5, 5'-divinyl-BINAP and a series of comonomers.

Herein, we report the construction of two advanced hierarchical chiral POPs, which were based on two synthesized vinyl-functionalized chiral BINAP ligands, (S)-4, 4'-divinyl-BINAP and (S)-5, 5'-divinyl-BINAP. The chiral BINAP-based POPs, which were denoted as 4-BINAP@POPs and 5-BINAP@POPs, were efficiently prepared via the copolymerization of vinyl-functionalized BINAP with DVB under solvothermal conditions. After exploring different synthetic routes and impregnating the POPs with a series of Ru species, the resultant materials were systematically studied for the heterogeneous asymmetric hydrogenation of β-keto esters.

All the solvents were analytical grade and were purified by distillation under Ar atmosphere before use. Unless otherwise noted, all the manipulations were performed under an Ar atmosphere either in a glove box or using standard Schlenk techniques. DVB (80%) was provided by J & K Chemical as a mixture of 1, 2-, 1, 3-, and 1, 4-DVB.

The synthesis routes of (S)-5, 5'-divinyl-BINAP have been previously reported. Here, we describe the synthesis routes of (S)-4, 4'-divinyl-BINAP in Scheme 1.

Compound 1, (S)-BINAP dioxide: (S)-BINAP (3 mmol, 1.87 g) was dissolved in dichloromethane (DCM, 60 mL) followed by the dropwise addition of H₂O₂ (18.32 mmol, 6.06 mL). The reaction was monitored by thin-layer chromatography (TLC). After being stirred for 30 min at room temperature, the solution was extracted with water (30 mL). The organic phase was washed with 10% NaHSO₃ aqueous solution (50 mL) and dried over Na₂SO₄. The solvent was removed under vacuum, and 1.95 g of a white solid (99% yield) was obtained, which was denoted as (S)-BINAPO.

Compound 2, (S)-4, 4'-diBr-BINAPO: (S)-BINAPO (10 mmol, 6.55 g) was dissolved in DCM (150 mL) followed by the addition of liquid bromine (30 mmol, 4.8 g) and pyridine (10 mmol, 0.79 g). After being stirred for 24 h at room temperature, the solution was washed with 10% NaHSO₃ aqueous solution, saturated brines, and saturated sodium bicarbonate aqueous solution, dried over Na₂SO₄, and evaporated under vacuum to remove the solvent. After repeating the above steps twice, 6.10 g of a white solid (75% yield) was obtained, which was denoted as (S)-4, 4'-diBr-BINAPO.

Compound 3, (S)-4, 4'-divinyl-BINAPO: (S)-4, 4'-diBr-BINAPO (1.0 mmol, 0.82 g), potassium vinyltrifluoroborate (2.4 mmol, 0.32 g) and PdCl₂(dppf) CH₂Cl₂ (0.08 mmol, 0.058 g) were placed in a three-necked flask. N-propanol (n-PrOH, 10 mL) and triethylamine (2.0 mmol, 0.20 g) were added to the flask. The reaction was monitored by TLC. After being refluxed for 3 h, the solvent was removed under vacuum. The precipitate was passed through a silica gel column and dried under vacuum to produce 0.565 g of a white solid (80% yield), which was denoted as (S)-4, 4'-divinyl-BINAPO.

Compound 4, (S)-4, 4'-divinyl-BINAP: (S)-4, 4'-divinyl-BINAPO (1.0 mmol, 0.70 g), trichlorosilane (3.0 mmol, 0.41 g) and phenylsilane (3.0 mmol, 0.32 g) were placed in a three-necked flask with the addition of toluene (10 mL). The reaction was monitored by TLC. After being refluxed for 3 h, the solvent was cooled to 0 ℃, followed by the slow addition of NaOH aqueous solution. Then, the solution was extracted with water and dried over Na₂SO₄. The solvent was removed under vacuum. The precipitate was passed through a silica gel column and dried under vacuum to produce 0.15 g of a white solid (21% yield), which was denoted as (S)-4, 4'-divinyl-BINAP. ¹H NMR (400 MHz, CDCl₃) δ 5.38 (d, 1H, J = 10.9 Hz), 5.52 (d, 1H, J = 17.1 Hz), 6.85-6.96 (m, 2H), 7.00-7.20 (m, 10H), 7.34-7.41 (m, 1H), 7.42-7.49 (m, 1H), 7.55-7.63 (s, 1H), 8.09 (d, 1H, J = 8.5 Hz); ³¹P NMR (161 MHz, CDCl₃)-14.9 Hz.

The synthesis of the BINAP@POPs and catalysts is summarized in Scheme 2. In an autoclave, 0.1 g (S)-4, 4'-divinyl-BINAP and 1.0 g DVB were dissolved in 10 mL of tetrahydrofuran (THF), followed by the addition of 25 mg of 2, 2'-azoisobutyronitrile (AIBN). The mixture was first stirred for 10 min at room temperature and then heated at 100 ℃ for 24 h. The solvent was removed under vacuum at 65 ℃, and a white solid was obtained, which was denoted as 4-BINAP@POPs.

Using 0.1 g of (S)-5, 5'-divinyl-BINAP instead of (S)-4, 4'-divinyl-BINAP in the solution above, 5-BINAP@POPs were obtained via the same synthesis method.

First, 0.41 g 4-BINAP@POPs and 0.0069 g [Ru (benzene) Cl₂] ₂ (0.028 mmol Ru) were added to 6 mL of N, N-dimethylformamide (DMF). After being refluxed for 3 h at 100 ℃, the solvent was removed under vacuum at 65 ℃, and a dark-green solid was obtained, which was denoted as Ru/4-BINAP@POPs.

Using 0.41 g of 5-BINAP@POPs instead of 4-BINAP@POPs in the solution above, Ru/5-BINAP@POPs-1 were obtained via the same strategies.

As shown in path B of Scheme 2, the ligand (S)-5, 5'-divinyl-BINAP was coordinated with [Ru (benzene) Cl₂] and then copolymerized with DVB to obtain the catalyst Ru/5-BINAP@POPs-2. First, 0.1 g (S)-5, 5'-divinyl-BINAP and 0.0186 g [Ru (benzene) Cl₂] ₂ (0.074 mmol Ru) were dissolved in 4 mL of DMF, and the mixture was stirred for 3 h at room temperature. Then, 1.0 g DVB, 0.025 g AIBN, and 6 mL of DMF were added to the mixture. The mixture was first stirred for 10 min at room temperature and then heated at 100 ℃ for 24 h. The solvent was removed, and a dark-green solid was obtained, which was denoted as Ru/5-BINAP@POPs-2.

First, 0.35 g 5-BINAP@POPs and 0.0072 g [Ru (p-cyme) Cl₂] ₂ (0.024 mmol Ru) were added to 5 mL of DMF. After being refluxed for 3 h at 100 ℃, the solvent was removed and a dark-green solid was obtained, which was denoted as Ru/5-BINAP@POPs-3.

First, 0.27 g 5-BINAP@POPs and 0.0050 g RuCl₃∙ x H₂O (0.018 mmol Ru) were added to 5 mL of DMF. After being refluxed for 3 h at 100 ℃, the solvent was removed and a dark-green solid was obtained, which was denoted as Ru/5-BINAP@POPs-3.

The liquid-state NMR spectra were recorded on a 400-MHz spectrometer. The chemical shifts are reported in ppm. The 1H NMR spectra were referenced to CDCl₃ (0 ppm), and the ¹³C NMR spectra were referenced to CDCl₃ (77.0 ppm). All the ¹³C NMR spectra were measured with complete proton decoupling. The signal patterns were indicated as s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br, broad singlet; and J, coupling constant in Hz. The solid-state NMR spectra were obtained using a Varian Infinity-Plus 400 spectrometer. The ³¹P MAS NMR spectra were recorded with a 5-mm probe at a frequency of 161.8 MHz under a magic angle spinning rate of 10 kHz and a delay of 4 s. The chemical shifts were referenced to 85% H₃PO₄. The ¹³C CP/MAS NMR spectra were recorded under a magic angle spinning rate of 6 kHz. Nitrogen isotherms at 77.3 K were measured using a Quantachrome Autosorb-1. Thermogravimetric analysis (TGA) was performed using a NETZSCH STA 449F3, and the samples were heated from 40 to 1000 ℃ at a rate of 10 ℃/min under air. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) analysis of the liquid samples was performed on a Perkin-Elmer 7300DV. High-angle annular dark-field scanning transmission electron microscopy (HAADF STEM) images were obtained using a JEM-ARM200F. The K-edge extended X-ray absorption fine structure spectra (EXAFS) of Ru were obtained at the BL14W1 beamline of SSRF, SINAP (Shanghai, China) using a Si (311) crystal monochromator. The storage ring was operated at 3.5 GeV with injection currents of 200 mA. The data were recorded in the fluorescence mode. The EXAFS data were analysed using the Demeter software package. Fourier transformation of the EXAFS data was applied to the k³-weighted functions. The theoretical scattering amplitude and phase-shift functions of all the paths for fitting the EXAFS data were calculated using the F EFF6 code.

As a typical run for asymmetric hydrogenation of β-keto esters, 0.026 g Ru/5-BINAP@POPs-1 catalyst, 0.20 g methyl acetoacetate, and 2 mL of isopropanol (ipro) were added to a 30-mL autoclave in a glove box. After the reactor was purged with H₂ four times, its pressure was finally adjusted to the desired value, heated from room temperature to the reaction temperature of 50 ℃, and stirred for 10 h. The catalyst was separated by centrifugation, and the product was analysed using gas chromatography (GC; Agilent 7890B gas chromatograph equipped with a flame ionization detector and a Cyclosil-B capillary column).

For recycling the catalyst, 0.052 g Ru/5-BINAP@POPs-1 catalyst, 0.40 g methyl acetoacetate, and 4 mL of ipro were used. The catalyst was separated by centrifugation (performed in a glove box), washed with ipro (3 × 2 mL); the catalyst was then used directly for the next catalytic reaction.

The pore properties of the three catalysts were analysed using N₂ physisorption isotherms (Fig. 1(a)). The catalysts were insoluble and porous with large surface areas of 1058-1070 m²/g (A _BET) and pore volumes of 1.19-1.64 cm³/g. The Ru/5-BINAP@POPs-2 contained larger mesopores than the Ru/4-BINAP@POPs and Ru/5-BINAP@POPs-1. Meanwhile, the polymers of 4-BINAP@POPs and 5-BINAP@POPs had surface areas greater than 1600 m²/g. The high P/Ru molar ratio (P/Ru = 2) can explain the different pore properties of the polymers and catalysts. Notably, polyDVB, which is obtained by the self-polymerization of DVB, only gives a A _BETof 816 m²/g. These results demonstrate that the addition of the rigid and sterically chiral divinyl-BINAP ligand is beneficial for increasing the surface areas and pore volumes of polymers. The three catalysts all contained micropores and mesopores, as further confirmed by the pore size distribution curves calculated using density functional theory (DFT; Fig. 1(b)).

The ¹³C MAS NMR spectra of 4-BINAP@POPs and 5-BINAP@POPs both contained clear resonance peaks at 145, 137, and 128 ppm, which were assigned to the carbon of naphthalene and benzene rings of the BINAP ligand and DVB (Fig. 1(c)). The resonance peaks near 44.0 ppm with relatively large peak areas are mainly attributed to the carbon in the alkyl linker formed by the copolymerization of vinyl functional groups. Notably, the spectra show small peaks at 114 ppm corresponding to vinyl groups, indicating the high degree of the copolymerization of the ligands and DVB. The ³¹P MAS NMR spectrum of 4-BINAP@POPs produced a signal at -14.0 ppm and that of 5-BINAP@POPs produced a signal at -14.6 ppm (Fig. 1(d)). No other obvious signals are observed for either polymer, which indicates that the phosphine in the polymer backbone is not oxidized. The ³¹P MAS NMR spectrum of Ru/5-BINAP@POPs-2, a representative catalyst, exhibited a broad peak at 52.8 ppm, which could be assigned to the unprotected phosphine coordinated with Ru species (Fig. 1(e)). In addition, the relatively less intensive peak at 24.9 ppm was attributed to phosphorus of phosphine oxide.

Excellent thermal stability is a crucial requirement for polymer-supported catalysts. The TGA curves under air demonstrate that the catalysts exhibited excellent thermal stability without any decomposition at 300 ℃ (Fig. 1(f)).

To confirm the state of the Ru species in our heterogeneous catalyst, EXAFS andHAADF-STEM characterization were employed for the fresh and used Ru/5-BINAP@POPs-1 catalyst. As observed in Fig. 2 and Table 1, no obvious Ru-Ru bonds were detected regardless of whether the sample was fresh or used, indicating that Ru species are dispersed as a single atom on the 5-BINAP@POPs, which was further confirmed by the HAADF-STEM images (Fig. 3). In addition, P, Cl, and O were coordinated with Ru, whose coordination numbers are all two [40-42]. Moreover, considering the uncertainty of the fitting results, we believe that the bond lengths of Ru-P and Ru-Cl in the fresh and used samples were the same because they are all from the same ligands. The Ru-O bond lengths of the fresh and used Ru/5-BINAP@POPs-1 sample were 1.90 and 2.13 Å, respectively. The distinct bond length of Ru-O is due to the difference between DMF and ipro. The Ru species in the fresh sample was coordinated with the O atom of DMF. Ipro was selected as the reaction solvent for the asymmetric hydrogenation of the-keto ester. Therefore, the Ru species in the used sample was coordinated with the O atom of ipro.

The effect of the catalysts for the hydrogenation of methyl acetoacetate on the reaction rate and enantioselectivity was studied using Ru/4-BINAP@POPs and Ru/5-BINAP@POPs-1 (Fig. 4). Ru/5-BINAP@POPs-1 produced a 74.3% yield and 89.1% enantiomeric excess (ee) within 1 h, whereas Ru/4-BINAP@POPs only produced a 29.1% yield and 84.6% ee. After 6 h, the yield was more than 98% for both catalytic systems. In particular, Ru/5-BINAP@POPs-1 produced a 99.5% yield and 94.3% ee after 10 h. Ru/5-BINAP@POPs-1 exhibited a higher reaction rate and similar enantioselectivity compared with Ru/4-BINAP@POPs. Considering the similarity in the pore structures of the two catalysts, the divinyl derivatives of BINAP, which were substituted in different positions, affected the chiral characters of the BINAP ligand in the polymer backbone and may ultimately affect their catalytic performance. The bulky building blocks formed in the 4, 4'-positions were closer to the phosphorus atoms of BINAP than the 5, 5'-positions, which made the catalyst of Ru/4-BINAP@POPs not easily accessible to the substrate.

The catalytic performance of different 5-BINAP@POP-supported catalysts was also studied (Table 2). As previously mentioned, Ru/5-BINAP@POPs-1 and Ru/5-BINAP@POPs-2 were synthesized via different strategies (post-synthesis and one-pot, respectively), leading to distinct effects on the catalytic activity. The catalyst prepared through post-synthesis (Table 2, entry 1) exhibited better activity and enantioselectivity than the one-pot sample (Table 2, entry 2). This finding may be attributed to part of the Ru species being embedded in the framework of the Ru/5-BINAP@POPs-2 catalyst, which therefore restricted the accessibility of the reagent to the active sites of the catalyst [39]. Ru/5-BINAP@POPs-3 and Ru/5-BINAP@POPs-4 were obtained using the post-synthesis strategy using the Ru precursors of [Ru (p-cyme) Cl₂] ₂ and RuCl₃∙ x H₂O, respectively. The activity and enantioselectivity of the catalysts decreased in the order Ru/5-BINAP@POPs-1 > Ru/5-BINAP@POPs-3 > Ru/5-BINAP@ OPs-4. Therefore, the catalyst that was post-synthesized by the 5-BINAP@POPs polymer and [Ru (benzene) Cl₂] ₂ precursor were selected for further study.

The effects of various parameters (solvent, reaction temperature, and H₂ pressure) on the product yield and enantioselectivity were investigated (Table 3). The insoluble catalyst was well swollen by a wide range of organic solvents. Alcoholic solvents were better selections than the others, and the use of ipro resulted in slightly higher enantioselectivity than the use of methanol (Table 3, entries 1-5). The effect of temperature on the reaction was investigated in the range of 40-80 ℃ (Table 3, entries 5-9). The reaction rate slowed down with decreasing temperature; however, the enantioselectivity was not significantly affected by this factor. In addition, the enantioselectivity improved from 91.3% to 95.0% when the H₂ pressure increased from 1 to 6 MPa (Table 3, entries 10-14).

Next, we examined the scope of asymmetric hydrogenation of β-keto esters by Ru/5-BINAP@POPs-1 (Table 4). Excellent production yields were obtained with high enantioselectivities. In the hydrogenation of tert-butyl acetoacetate and isobutyl acetoacetate, the catalyst exhibited lower performances than the other substrates (Table 4, entries 5 and 6). Notably, the catalyst demonstrated satisfactory activity at very high molar ratios of the substrate and catalyst (S/C = 4000) for 10 h (Table 4, entry 8). When the amount of the reaction substrate was increased to 43 mmol, a long reaction time of 20 h was needed to obtain a satisfactory result (Table 4, entry 9), which was most likely due to the effect of mass transfer. The uniform distribution of the BINAP ligand in the polymer backbone and abundant pore structures of the catalyst may afford a quasi-homogeneous flexible character to the ligands in the framework while the reaction proceeds [43].

The asymmetric hydrogenation of methyl acetoacetate was selected to evaluate the reusability of Ru/5-BINAP@POPs-1 (Table 5). The catalyst can be reused for at least 6 cycles with complete conversion and high enantioselectivity. We further confirmed the heterogeneous nature of the system using ICP analysis. During the recycling, the amount of Ru species in the filtrate was less than 0.06 ppm for each round of hydrogenation. The morphologies of fresh and used Ru/5-BINAP@POPs-1 were examined using HAADF STEM (Fig. 3). In addition, no obvious Ru nanoparticles were formed after the sixth run, as evidenced by the EXAFS results (Fig. 2 and Table 1). Note that even with a high P/Ru ratio (P/Ru = 2), the catalyst exhibited high stability during the recycling. The hot filtration test results for the Ru/5-BINAP@POPs-1 in the asymmetric hydrogenation of methyl acetoacetate are presented in Fig. 5. Because the substrate conversion was no more than 15%, the ee value is not listed. The results indicate that when the insoluble catalyst was removed from the hot reaction mixture by filtration, the reaction stopped and no additional products were generated. This finding indicated that the reaction was indeed heterogeneous.

Two novel vinyl-modified BINAP ligands, 4, 4'-divinyl-BINAP and 5, 5'-divinyl-BINAP, were successfully synthesized and used as monomers for the preparation of hierarchical porous chiral polymers. The chiral POPs supporting Ru catalysts prepared using two different synthetic routes were systematically investigated to demonstrate their excellent catalytic performance for heterogeneous asymmetric hydrogenation of β-keto esters. Characterization by NMR and EXAFS confirmed that through copolymerization with DVB, the chiral ligands were incorporated into the polymer backbone by an alkyl linker. The alkyl linker formed by the vinyl groups in the prepared polymers could maintain the quasi-homogeneous flexible character of the BINAP ligand in the polymer backbone [43]. Ru species were firmly coordinated by chiral BINAP ligands fixed on the polymer backbone to prevent leaching during recycling usage. Meanwhile, abundant pore structures and uniformly implanted chiral BINAP ligands in the polymer were favourable for highly efficient chiral inducibility of β-keto esters on coordinated Ru sites.