Cooperative chiral salen Ti^IV catalysts with built-in phase-transfer capability accelerate asymmetric sulfoxidation in water

Abstract

A series of cooperative chiral salen Ti^IV catalysts with inherent phase-transfer capability were prepared by covalently bridging double chiral salen Ti^IV units with various polyethylene glycol (PEG)-based dicationic ionic liquid (IL) linkers. Characterization results suggested the presence of polyether-based IL spacers and intact active sites in the catalysts. The polyether-based dicationic IL spacer not only enforced an intramolecular, cooperative reaction pathway favored for the asymmetric suloxidation, but also endowed the bimetallic catalysts with built-in phase transfer capability. High yields of chiral sulfoxides (in the range of 74–90%) with excellent ee values (in the range of 85–91%) were achieved within 45 min when the asymmetric oxidation of methyl phenyl sulfide, methyl p-methoxyphenyl sulfide, and methyl o-methoxyphenyl sulfide were performed in water. The catalytic efficiency was significantly higher than that over neat complex (yields, 10–16% and ee values, 77–79%). More attractively, the phase transfer catalysts could be facilely recovered by solvent precipitation for efficient reuse.

---

Introduction

Asymmetric oxidation of sulfides is of great importance in fundamental chemistry and modern industrial processes, because the obtained chiral sulfoxides are versatile intermediates in organic synthesis and are useful for the preparation of biologically and medically important compounds.^1–5 The chiral salen Ti^IV complex is one of the most efficient catalysts for this transformation.^6–10 High yield of sulfoxides with excellent enantioselectivity was achieved when the reactions were carried out in dichloromethane using environmentally benign H₂O₂ as an oxidant. Environmental concerns required chemists to develop effective catalyst that could operate asymmetric catalysis in more ecofriendly reaction media compared to organic solvents, such as using water as a solvent.^11–13 Significant efforts have been made to fulfill the requirement. Nevertheless, the relatively low activity remained a key point to improve. The main issue lied in the poor accessibility of the hydrophobic substrates to the water-insoluble, catalytically active center in water. Variable attempts have been conducted to tackle these challenges, for example, adding surfactants or amphiphilic agents, using phase-transfer catalysts (PTCs), building emulsion systems, or modifying the hydrophobic/hydrophilic property of catalyst supports.^14–20 Among these solutions, PTCs were considered as the efficient strategy to circumvent the limited mass transfer, since they generally show excellent solubility in water and high affinity toward hydrophobic substrates. Although highly efficient, amphiphilic PTCs are often difficult to separate from aqueous reaction for reuse.^15,19 It thus restricted their industrial application in a large-scale.

Polyethylene glycol (PEG) with phase transfer capability provided an attractive solution to the problem of PTCs recovery, since the selective solubility of PEG facilitated the separation of catalyst through solvent precipitation.^21–23 Recently, we have covalently modified a chiral salen Mn^III complex by the PEG-based imidazolium ionic liquids (PEG-ILs).²² The obtained PTCs indeed achieved efficient catalysis coupled with facile recovery in a water/organic biphasic system. More importantly, the special ‘ionophilicity’ and polarity of imidazolium-based IL moiety played positive effects on activating and stabilizing the metallosalen active intermediate, which further enhancing catalytic efficiency and selectivity.²⁴ The featured advantages encouraged us to introduce PEG-based imidazolium IL into chiral salen Ti^IV complex to provide an efficient and reusable PTC for asymmetric sulfoxidation in water. It was well known that Ti(salen)-mediated asymmetric sulfoxidation proceed via cooperative bimetallic mechanisms, wherein a di-μ-oxo Ti(salen) species of [{(salen)Ti(μ-O)}₂] was the real catalytic precursor.^7,25 We thus decided to bridge double chiral salen Ti^IV units by a PEG-based dicationic IL linker, so as to enforce the intramolecular, cooperative reaction pathway favored for the catalysis.

Herein, PEG-bridged di-imidazole was synthesized and used to bridge double chiral salen Ti^IV units through N-alkylation of terminal imidazole groups with methyl chloride group (–CH₂Cl) at 5-position of asymmetric chiral salen ligand. Double chiral salen Ti^IV complex units were thus combined into a single molecule through a flexible PEG-based dicationic IL bridge, giving the PEG-based dicationic IL bridged bimetallic chiral salen Ti^IV catalysts (PIBC). The PEG-based dicationic IL spacer played double roles in the catalysis, i.e., endowing the PIBC with built-in phase transfer capability, and meanwhile assisting intramolecular cooperative reaction favored for the sulfoxidation. Significantly high activity and selectivity (in terms of chemo- and enantioselectivity) were observed over the PIBC in aqueous asymmetric sulfoxidation due to the phase transfer catalysis together with intramolecular cooperation. More importantly, the PTCs could be quantitatively recovered from the aqueous system by solvent precipitation for efficient reuse. The problems associated with mass transfer limitation, as well as catalyst recovery in aqueous organic reaction, could thus be well resolved. In addition, the diverse length of PEG spacer affected phase transfer capability and also cooperative interaction of the bimetallic catalysts, which further affected the catalytic performance of PIBC in the transformation.

Experimental section

Materials and reagents

(±)-1,2-Diaminocyclohexane, ethylenediamine and methyl acrylate were purchased from Alfa Aesar. Ti(Oⁱpr)₄ and aryl methyl sulfides were obtained by J&K. Other commercially available chemicals were laboratory grade reagents from local suppliers. All of solvents were purified by standard procedures. Ti(Oⁱpr)₄ was distilled and diluted to 0.1 M in CH₂Cl₂. Chloride-terminated PEG was prepared according to the reported procedures. (R,R)-[N-(3,5-Di-tert-butylsalicylidene)-N′-(3-tert-butyl-5-chloro-methylsalicylidene)-1,2-cyclohexanediaminato]titanium(IV) di-isopropyl (denoted as asymmetric chiral salen Ti^IV complex) was prepared according to the described procedures.²⁶ (R,R′)-[N,N′-Bis(3,5-di-tert-butyl salicylidene)-1,2-cyclohexanediaminato]titanium(IV) di-isopropyl (denoted as neat complex, as shown in Chart 1) were prepared according to described procedures.²⁷

Chart 1 Structures of PIBC-0, IL-complex and neat complex.

Methods

FT-IR spectra were obtained as potassium bromide pellets with a resolution of 4 cm⁻¹ and 32 scans in the range 400–4000 cm⁻¹ using an AVATAR 370 Thermo Nicolet spectrophotometer. The thermogravimetric and differential thermogravimetric (TG-DTG) curves were obtained on a NETZSCH STA 449C thermal analyzer. Samples were heated from room temperature up to 800 °C under flowing air using alumina sample holders. The sample weight was ca. 10 mg and the heating rate was 10 K min⁻¹. ¹H NMR spectra of samples were recorded on a Varian-500 spectrometer with TMS as an internal standard. Thin layer chromatography (TLC) was conducted on glass plates coated with silica gel GF₂₅₄. The content of titanium in the samples was determined by inductively coupled plasma mass spectrometry (ICP-MS) on a NexION 300X analyzer (Perkin-Elmer Corp.). The optical rotation of catalysts was measured in dichloromethane on a WZZ-2A Automatic Polarimeter. HPLC analyses were performed on a TECHCOMP L2000 or JASCO 2089 liquid chromatograph using the Daicel chiralpak AD column.

Preparation of PIBC-n (n = 0, 4, 8, 13, where n represented the repeated units number of ethylene oxide units in various PEG chains)

The preparation of PIBC-n (n = 0, 4, 8, 13) was outlined in Scheme 1.

Scheme 1 Synthesis of PIBC-n (n = 0, 4, 8, 13).

Synthesis of PIBC-n (n = 4, 8, 13). An anhydrous ethanol solution (10 mL) of sodium ethylate (0.68 g, 10 mmol) was dropwise added into imidazole (0.68 g, 10 mmol) in anhydrous ethanol (10 mL). The resulting mixture was stirred at 60 °C for 8 h under nitrogen atmosphere. Chloride-terminated PEG-n (5 mmol, n = 4, 8, 13, which is the corresponding average number of ethylene oxide units in PEG chain of PEG200, PEG400 and PEG600) was then slowly added into the stirring mixture. The mixture was stirred at 70 °C for another 12 h, resulting in the formation of some white precipitate. After filtration, the filtrate was collected and concentrated in vacuum. The gummy residue was dissolved in dichloromethane (100 mL) and washed with deionized water for several times to completely remove the insoluble salt of sodium chloride. The organic phase was dried over anhydrous Na₂SO₄ and concentrated under vacuum. The residue was further dried at 40 °C under vacuum, giving the sticky brownish red products of PEG-bridged di-imidazole (n = 4, 8, 13). PEG-bridged di-imidazole (n = 4): ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 7.47 (s, 2H, ring N=CHN), 7.05 (s, 2H, ring C=NCH=C), 6.96 (s, 2H, ring N–CH=C),4.13 (t, 4H, O–CH₂–CH₂–N_ring), 3.72–3.76 (t, 4H, ring O–CH₂–CH₂–N), 3.54–3.72 (m, 8H, CH₂–O–C₂H₄–O–CH₂); FT-IR (KBr): γ_max/cm⁻¹ 2879, 1657, 1511, 1443, 1361, 1242, 1100, 945, 880, 831, 710, 667, 623.

PEG-bridged di-imidazole with various PEG chains (5 mmol) was mixed with asymmetric chiral salen Ti^IV complex (7.0 g, 10.1 mmol) in dry toluene (20 mL). The mixture was refluxed for 48 h under nitrogen atmosphere. After removal of the solvent in vacuum, the mixture was washed with ether for several times. The gummy residue was dried in vacuum at 40 °C to provide PIBC-n as the yellow powder. The PIBC containing the average numbers of ethylene oxide unit in PEG moiety of 4, 8 and 13 were denoted as PIBC-4, PIBC-8 and PIBC-13, respectively.

PIBC-4: ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 8.29 (s, 4H, Ar-CH=N), 7.47 (s, 2H, ring N=CHN), 7.31–7.29 (s, 8H, Ar-H), 7.05 (s, 2H, ring C=NCH=C), 6.96 (s, 2H, ring N–CH=C), 4.13 (t, 4H, O–CH₂–CH₂–N_ring), 3.72–3.76 (t, 4H, O–CH₂–CH₂–N_ring), 3.68 (m, 4H, (CH₃)₂–CH–O), 3.54–3.72 (m, 24H, CH₂–O–C₂H₄–O–CH₂); 3.15 (s, 4H, C=N–CH–CH₂), 2.08 (s, 4H, Ar-CH₂–N), 1.48–1.95 (m, 16H, C=N–CH–(CH₂)₄–CH), 1.40 (m, 54H, (CH₃)₃C–), 1.22 (m, 24H, (CH₃)₂–CH–O); FT-IR (KBr): γ_max/cm⁻¹ 3130, 2950, 2864, 1630, 1541, 1440, 1360, 1319, 1270, 1207, 1168, 1096, 1028, 805, 773, 754, 710, 660, 627, 558, 447; titanium content: 1.03 mmol g⁻¹ (theoretical value: 1.36 mmol g⁻¹); α²⁸_D = −277 (C = 0.03, CH₂Cl₂); PIBC-8: ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 8.29 (s, 4H, Ar-CH=N), 7.47 (s, 2H, ring N=CHN), 7.31–7.29 (s, 8H, Ar-H), 7.05 (s, 2H, ring C=NCH=C), 6.96 (s, 2H, ring N–CH=C), 4.13 (t, 4H, O–CH₂–CH₂–N_ring), 3.72–3.76 (t, 4H, O–CH₂–CH₂–N_ring), 3.68 (m, 4H, (CH₃)₂–CH–O), 3.54–3.72 (m, 8H, C₂H₄–O–CH₂); 3.15 (s, 4H, C=N–CH–CH₂), 2.08 (s, 4H, Ar-CH₂–N), 1.48–1.95 (m, 16H, C=N–CH–(CH₂)₄–CH), 1.40 (m, 54H, (CH₃)₃C), 1.22 (m, 24H, (CH₃)₂–CH–O); FT-IR (KBr): γ_max/cm⁻¹ 3132, 2944, 2866, 1630, 1541, 1472, 1443, 1397, 1360, 1273, 1252, 1208, 1096, 1035, 847, 805, 750, 627, 549, 467; titanium content: 0.99 mmol g⁻¹ (theoretical value: 1.20 mmol g⁻¹); α²⁸_D = −261 (C = 0.03, CH₂Cl₂); PIBC-13: ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 8.29 (s, 4H, Ar-CH=N), 7.47 (s, 2H, ring N=CHN), 7.31–7.29 (s, 8H, Ar-H), 7.05 (s, 2H, ring C=NCH=C), 6.96 (s, 2H, ring N–CH=C), 4.13 (t, 4H, O–CH₂–CH₂–N_ring), 3.72–3.76 (t, 4H, O–CH₂–CH₂–N_ring), 3.68 (m, 4H, (CH₃)₂–CH–O), 3.54–3.72 (m, 44H, C₂H₄–O–CH₂); 3.15 (s, 4H, C=N–CH–CH₂), 2.08 (s, 4H, Ar-CH₂–N), 1.48–1.95 (m, 16H, C=N–CH–(CH₂)₄–CH), 1.40 (m, 54H, (CH₃)₃C), 1.22 (m, 24H, (CH₃)₂–CH–O); FT-IR (KBr): γ_max/cm⁻¹ 3135, 2946, 2871, 1630, 1541, 1499, 1475, 1443, 1397, 1360, 1248, 1209, 1170, 1096, 1035, 849, 805, 750, 707, 627, 560, 444; titanium content: 0.88 mmol g⁻¹ (theoretical value: 1.08 mmol g⁻¹); α²⁸_D = −220 (C = 0.03, CH₂Cl₂).

Synthesis of PIBC-0. A bimetallic analog of PEG-free counterpart, in which double ((R,R)-salen)Ti^IV(Oⁱpr)₂ units were bridged with an ethyl-based dicationic IL, was prepared as a control catalyst according to a similar preparation procedure to that of PIBC-n, as shown in Scheme 1. Instead of the chloride-terminated PEG, 1,2-dichloroethane was used to bridge imidazole moiety through nucleophilic substitution. The obtained ethyl-bridged di-imidazole was readily N-alkylated with chloromethyl groups (–CH₂Cl) at 5-position in asymmetric chiral salen Ti^IV complex, giving the ethyl-based dicationic IL bridged bimetallic chiral salen Ti^IV complex (denoted as PIBC-0). The structure of PIBC-0 is shown in Chart 1. FT-IR (KBr): γ_max/cm⁻¹ 3138, 2941, 2867, 1630, 1541, 1471, 1440, 1395, 1361, 1319, 1266, 1206, 1169, 1103, 1034, 805, 769, 748, 707, 632, 557, 468; titanium content: 1.26 mmol g⁻¹ (theoretical value: 1.54 mmol g⁻¹); α²⁸_D = −332 (C = 0.03, CH₂Cl₂).

Synthesis of monomeric IL-functionalized chiral salen Ti^IV complex. For comparison, a monomeric counterpart of 1-methyl-3-methyleneimidazolium chlorine-functionalized chiral salen Ti^IV complex (denoted as IL-complex, as shown in Chart 1) was also prepared as the control catalyst. The preparation procedure was similar to that of PIBC, except for the use of N-methylimidazole instead of PEG-bridged di-imidazole during N-alkylation.

Catalyst testing

The selected catalyst (1.0 mol% of substrate, based on titanium content in the catalyst) and sulfides (1.0 mmol) were added into water (1 mL) under stirring. H₂O₂ (30 wt%, 1.2 mmol) was then dropwise added within 15 min at room temperature. The oxidation progress was monitored constantly by TLC. After the reaction, the reaction solution was extracted with CH₂Cl₂ for three times (3 × 3 mL). The combined organic layer was dried over anhydrous Na₂SO₄, and was concentrated in vacuo. Further purification of the residue by chromatography on silica gel (petroleum ether/ethyl acetate, 1.5/1) afforded pure sulfoxides. ee values of corresponding chiral sulfoxides were determined by HPLC analysis using the Daicel chiralpak AD column. Detailed NMR and HPLC analysis for obtained sulfoxides are available in the ESI.† The separated aqueous layer containing catalyst was treated with n-hexane (5 mL). Catalyst could be readily precipitated out from the aqueous phase. The recovered catalyst was washed with ether (3 × 5 mL), and dried in vacuo at 30 °C overnight for consecutive reuse.

Methyl phenyl sulfoxide. Yield: 90%; ee value: 85%, determined by HPLC (i-PrOH/n-hexane = 1 : 9 (v/v)); flow rate = 1.0 mL min⁻¹; 25 °C; λ = 254 nm; major enantiomer t_R = 18.1 min and minor enantiomer t_S = 21.1 min; ¹H NMR (CDCl₃, 500 MHz): δ (ppm): 2.56 (s, 3H, Me), 7.37–7.52 (m, 5H, ArH). ¹³C NMR (CDCl₃, 125 MHz): δ (ppm): 43.8 (SCH₃), 123.4, 129.3, 131.0, 145.5 (ArC).

Methyl p-methoxyphenyl sulfoxide. Yield: 84%; ee value: 90%, determined by HPLC (i-PrOH/n-hexane = 2 : 8 (v/v)); flow rate = 1.0 mL min⁻¹; 25 °C; λ = 254 nm; major enantiomer t_R = 14.2 min and minor enantiomer t_S = 17.2 min; ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 3.01 (s, 3H, SCH₃), 3.91 (s, 3H, OCH₃), 7.04 (d, 2H, ArH), 7.89 (d, 2H, ArH). ¹³C NMR (CDCl₃, 125 MHz): δ (ppm): 44.9 (SCH₃), 55.7 (OCH₃), 114.5, 129.6, 132.3, 163.7 (ArC).

Methyl o-methoxyphenyl sulfoxide. Yield: 74%; ee value: 91%, determined by HPLC (i-PrOH/n-hexane = 2 : 8 (v/v)); flow rate = 1.0 mL min⁻¹; 25 °C; λ = 254 nm; major enantiomer t_R = 13.8 min and minor enantiomer t_S = 16.8 min; ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 2.67 (s, 3H, SCH₃), 3.78 (s, 3H, OCH₃), 6.84–7.37 (m, 4H, ArH). ¹³C NMR (CDCl₃, 125 MHz): δ (ppm): ¹³C NMR (CDCl₃, 125 MHz): δ (ppm): 41.1 (SCH₃), 55.7 (OCH₃), 118.6, 121.5, 124.3, 132.0, 154.7 (ArC).

Methyl p-nitrophenyl sulfoxide. Yield: 22%; ee value: 63%, determined by HPLC (i-PrOH/n-hexane = 3 : 7 (v/v)); flow rate = 1.0 mL min⁻¹; 25 °C; λ = 254 nm; major enantiomer t_R = 10.8 min and minor enantiomer t_S = 20.1 min; ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 2.57 (s, 3H, SCH₃), 7.30 (d, 2H, ArH), 8.16 (d, 2H, ArH). ¹³C NMR (CDCl₃, 125 MHz): δ (ppm): 43.9 (SCH₃), 113.9, 125.0, 144.7, 148.9 (ArC).

Methyl p-bromophenyl sulfoxide. Yield: 38%; ee value: 98%, determined by HPLC (i-PrOH/n-hexane = 5 : 5 (v/v)); flow rate = 1.0 mL min⁻¹; 25 °C; λ = 254 nm; major enantiomer t_R = 8.4 min and minor enantiomer t_S = 9.9 min; ¹H NMR (CDCl₃, 500 MHz) δ (ppm): 3.07 (s, 3H, SCH₃), 7.84 (d, 2H, ArH), 7.74 (d, 2H, ArH). ¹³C NMR (CDCl₃, 125 MHz): δ (ppm): 44.5 (SCH₃), 129.0, 132.7, 139.5 (ArC).

Results and discussion

Preparation of catalysts

Chiral phase transfer catalysis is a feasible approach that allows asymmetric catalytic reaction to be efficiently carried out in water. Low molecular weight liquid PEG is a fascinating phase transfer reagent with tunable solubility.²⁸ Chiral catalysts containing the PEG moiety often undergo chiral phase transfer catalysis in aqueous organic reaction and also can be facile recovered by solvent precipitation.^22,29,30 The inherent advantages encouraged us to introduce a PEG moiety into a framework of chiral salen Ti^IV complex to fabricate a reusable PTC for asymmetric sulfoxidation in water using H₂O₂ as an oxidant. ILs derived from N,N-dialkylimidazolium were found to stabilize chiral metallosalen complex during catalytic process and/or enhance their catalytic efficiency.^24,31 We thus attempted to modify chiral salen Ti^IV complex with PEG-modified imidazolium IL to make chiral salen Ti^IV complex efficient and reusable. Given that asymmetric sulfoxidations catalyzed by chiral salen Ti^IV complex occurred via a bimetallic, cooperative mechanism,^7,25 a strategy that we have designed here is to bridge double chiral salen Ti^IV units with a PEG-based imidazolium IL spacer, so as maximize cooperative reaction favored in the catalysis.

The synthesis route for PIBC-n (n = 4, 8, 13) was outlined in Scheme 1. PEG-bridged di-imidazole with various length of PEG chain was obtained by nucleophilic substitution of chloride-terminated PEG-n (n = 4, 8, 13) with imidazolinium sodium salt under alkaline condition. Successive N-alkylation of PEG-bridged di-imidazole with the chloromethyl groups (–CH₂Cl) at 5-position of asymmetric chiral salen Ti^IV complex afforded the intramolecular bimetallic catalyst of PIBC-n (n = 4, 8, 13), where double chiral metallosalen units were bridged by a PEG-based dicationic IL spacer, as shown in Scheme 1. For comparison, a PEG-free counterpart of PIBC-0, in which an ethyl-based dicationic IL was acted as the bridge, was also prepared according to a similar procedure, except for the use of 1,2-dichloroethane instead of chloride-terminated PEG, as shown in Scheme 1.

As expected, PIBC-n (n = 4, 8, 13) were water-soluble (Fig. 1a–c), although the chiral salen Ti^IV backbone was entirely insoluble in water. We speculated that the water-solubility was related with the PEG-based dicationic IL spacer. Interestingly, despite of dicationic IL bridged dimeric analogue, the PEG-free counterpart of PIBC-0 was practically insoluble in water (Fig. 1d).

Fig. 1 Photographs of PIBC-4 (a), PIBC-8 (b), PIBC-13 (c) and PIBC-0 (d) in water (0.5 g sample in 5 mL water) and aqueous solution of PIBC-8 treated with n-hexane (e) (0.5 g PIBC-8 in 1 mL water treated with 4 mL n-hexane).

Therefore, amphiphilic PEG chain was responsible for the water-solubility of PIBC-n (n = 4, 8, 13). Actually, the water-solubility of PIBC-n (n = 4, 8, 13) decreased as the length of polyether chain increased (Fig. 1a vs. b vs. c). Furthermore, we noticed that PIBC-n (n = 4, 8, 13) could be precipitated from water with the addition of n-hexane, as shown by the typical PIBC-8 (Fig. 1e). The special solubility thus offered a reliable method for subsequent separation and recycling of the chiral salen Ti^IV-based PTCS.

Characterization of samples

FT-IR. The synthesized PIBC-n (n = 0, 4, 8, 13), as well as neat complex for comparison, were characterized by FT-IR spectra (Fig. 2). Obviously, PIBC-n (n = 0, 4, 8, 13) exhibited characteristic vibration bands associated with C=N (1630 cm⁻¹), C–O (1541 cm⁻¹), and Ti–O (805 cm⁻¹) in FT-IR spectra,^10,27 suggesting the presence of intact chiral salen Ti^IV center in PIBC-n (Fig. 2a–c). Notably, the characteristic bands appeared as red shift as compared with those of neat complex (Fig. 2a–c vs. e). It was mainly due to the electron-deficient substitute of imidazolium cation at the 5-position of double salen ligand units.^23,32 The observations suggested successfully grafting imidazolium IL moiety on 5-position of chiral salen Ti^IV complex in PIBC-n (n = 0, 4, 8, 13). The presence of imidazolium IL moiety was also evident from a newly formed band at 627 cm⁻¹ associated with stretching vibration mode of C–N–C in imidazolium cation (Fig. 2a–c).³¹ Apart from characteristic C–N–C vibration band, an additional distinct band at around 1096 cm⁻¹ was observed in the FT-IR spectra of PIBC-n (n = 4, 8, 13) (Fig. 2a–c). It was attributable to C–O–C stretching vibration in PEG chains in PIBC-n (n = 4, 8, 13).³³ While, the C–O–C stretching vibration was absent in FT-IR spectrum of PIBC-0 (Fig. 2d vs. a–c), due to the absence of the PEG moiety in the framework, as shown in Chart 1. It was the amphipathic polyether chain that responsible for the phase transfer capacity of PIBC-n (n = 4, 8, 13) in aqueous asymmetric sulfoxidation.

Fig. 2 FT-IR spectra of PIBC-4 (a), PIBC-8 (b), PIBC-8 after five runs (b′), PIBC-13 (c), PIBC-0 (d) and neat complex (e).

TG-DTG. Thermal analysis has been used to monitor the decomposition profiles of typical PIBC-8, as well as neat complex and PEG-bridged di-imidazole for comparison. The results obtained were depicted in Fig. 3. Neat complex showed three distinct steps of weight loss in the combined TG-DTG curves, when heated from room temperature to 800 °C under airflow (Fig. 3A). The first loss in weight was centered at 188 °C, which was assigned to the cleavage of tert-butyl groups.³⁴ The second loss in weight occurs at 360 °C and was followed by an additional large weight loss at 488 °C. The two steps were associated with oxidative decomposition of residual salen ligand moiety.³⁴ The decomposition profile of neat complex was completed at ca. 598 °C with the residue (ca. 32.3 wt%) amounting to titanium oxides (Fig. 3A). PIBC-8 showed six distinct steps of weight losses upon being heated from room temperature to 800 °C under airflow (Fig. 3B). As mentioned, the first weight loss centered at 202 °C accounted for the cleavage of the tert-butyl groups. Notably, the decomposition temperature got increased due to mutual stabilization of salen ligand and ILs moiety. The second loss in weight occurred at 264 °C and was followed by an additional weight loss at 335 °C. The two steps, which overlapped in TG curve but were well distinguished in the corresponding DTG curve, were associated with oxidative decomposition of the polyether chain in PIBC-8, since the PEG-bridged di-imidazole showed similar weight loss in the TG-DTG curves (Fig. 3B vs. C). Successive cleavage of residual salen ligand was occurred in the range of 399–578 °C. Notably, the onset decomposition temperature of salen ligand in neat complex was only 292 °C. Obviously, the cleavage of salen ligand in PIBC-8 was retarded (from 292 to 399 °C). This pointed to a stabilization of salen Ti^IV complex moiety originated from the functionalization of imidazolium IL.

Fig. 3 TG-DTG curves of neat complex (A), PIBC-8 (B), and PEG-bridged di-imidazole (C) ((a) thermogravimetric curves; (b) differential thermogravimetric curves).

It provided an indirect proof for the successful grafting of PEG-ILs moiety on the chiral salen ligand. The sixth weight loss appeared at 637 °C was logically assigned to the complete decomposition of the imidazolium IL moiety. Complete decomposition of PIBC-8 occurred at 733 °C, leaving non-removable white residue belonged to the formed titanium oxides.

Catalytic performances

Despite the advantages of being safe, environmentally benign, and inexpensive, water as a solvent in asymmetric sulfoxidations was limited by the incompatibility of sulfides and chiral salen Ti^IV complex in water. Catalysts of PIBC-n (n = 4, 8, 13) were designed as the PTCs to circumvent the incompatibility and enhance catalytic efficiency in the aqueous sulfoxidation. Catalytic performances of PIBC-n (n = 4, 8, 13) were investigated in asymmetric sulfoxidation of methyl phenyl sulfide in water using H₂O₂ as an oxidant. The results were summarized in Table 1. PEG-free analogue of PIBC-0 were prepared as a control catalyst to elucidate the function of PEG moiety. Furthermore, traditional chiral salen Ti^IV complex (denoted as neat complex), as well as the IL-functionalized monometallic Ti-(salen) complex (IL-complex, as shown in Chart 1), was also employed for comparison.

Table 1 Results of asymmetric sulfoxidation over different chiral salen Ti^IV complexes in water^a

Entry	Catalyst	Substrate	Product	Yield^b (%)	ee^c (%)
a Catalyst (1.0 mol% of substrate, based on titanium content), substrate (1.0 mmol), H2O2 (1.2 mmol, added in 15 min), water (1 mL), 45 min, 25 °C.b Isolated yield after column chromatography.c Determined by HPLC (Daicel chiralpak AD column).d The mixture of PIBC-8 (1.0 mol% of substrate, based on titanium content) and methyl p-bromophenyl sulfide (1.0 mmol) in water (1 mL) was pretreated by ultrasonic for 10 min.
1	Neat complex			16	77 (R)
2	IL-Complex			64	79 (R)
3	PIBC-0			71	81 (R)
4	PIBC-4			81	83 (R)
5	PIBC-8			90	85 (R)
6	PIBC-13			82	82 (R)
7	Neat complex			17	77 (R)
8	IL-Complex			52	82 (R)
9	PIBC-0			80	88 (R)
10	PIBC-8			84	90 (R)
11	Neat complex			10	79 (R)
12	IL-Complex			50	71 (R)
13	PIBC-0			67	89 (R)
14	PIBC-8			74	91 (R)
15	Neat complex			—	—
16	IL-Complex			9	31 (R)
17	PIBC-0			18	59 (R)
18	PIBC-8			22	63 (R)
19	Neat complex			—	—
20	IL-Complex			17	88 (R)
21	PIBC-0			35	91 (R)
22	PIBC-8			38	98 (R)
23^d	PIBC-8			42	98 (R)

---

As expected, neat complex with poor water-solubility was inactive in the aqueous asymmetric sulfoxidation, giving only 16% yield of sulfoxide with 77% ee value (Table 1, entry 1). Although still limited water-solubility, IL-complex offered higher yield (64%) and ee value (79%) than neat complex (Table 1, entry 2 vs. 1) due to positive effect of imidazolium IL moieties on activating chiral salen metallosalen complex. However, it was less efficient than the dicationic IL bridged bimetallic catalyst of PIBC-0 (Table 1, entry 2 vs. 3), which gave 71% yield of methyl phenyl sulfoxide with 81% ee value under identical condition. The improved performance of bimetallic catalyst pointed to a cooperative bimetallic mechanism of the asymmetric sulfoxidation system. Combination of double chiral salen Ti^IV units into a single molecule enforced an intramolecular, cooperative reaction pathway resulting in enhanced reaction rate and higher selectivity.^25,35–37 Notably, also as bimetallic catalyst, the PEG-based dicationic IL bridged catalysts of PIBC-n (n = 4, 8, 13) were far more active and selective than PIBC-0 (Table 1, entry 2 vs. entries 4–6). A catalyst loading of 1.0 mol% was sufficient to afford excellent yields of methyl phenyl sulfoxide (in the range of 81–90%) with high enantioselectivities (in the range of 82–85%) within 45 min (Table 1, entries 4–6). Enhanced catalytic efficiency over PIBC-n (n = 4, 8, 13) should arise from bridging spacers of PEG chains. Different from ethyl bridging spacer in PIBC-0, amphipathic PEG chain acted as a phase transfer promoter to carry the organic sulfides into aqueous phase, enhancing the concentration of substrate in water for efficient catalysis. Therefore, PIBC-n (n = 4, 8, 13) with desirable built-in phase transfer capacity circumvented the mass transfer restriction associated with aqueous phase catalysis. More importantly, the flexible PEG spacers allowed the active sites to adopt their preferred conformation during catalytic reaction, which was crucial in effective bimetallic cooperative catalysis; otherwise the selectivity would be lost.^25,38–40

Actually, the length of PEG spacer was found to obviously influence activity and selectivity of the cooperative catalysts. PIBC-8 showed the highest catalytic efficiency, giving almost quantitative yield (90%) of methyl phenyl sulfoxide with 85% ee value within 45 min (Table 1, entry 5). Further lengthening or shorting the bridged polyether chain led to a decreased activity in the reaction accordingly (Table 1, entries 4–6). A possible explanation might be that the length of polyether chain directly affected not only the inherent phase transfer capability of catalyst, but also the intramolecular, cooperative interaction of active sites. Though increasing the total length of PEG spacer may allowed for conformational freedom of active sites favored for the cooperative catalysis, it reduced the water-solubility of corresponding catalysts, as shown in Fig. 1a vs. b vs. c, which accordingly was detrimental to the phase transfer catalysis. Therefore, it was reasonable that PIBC-8 with a middling length of PEG spacer gave highest overall activity in the asymmetric sulfoxidation of methyl phenyl sulfide in water.

Benefiting from the inherent phase transfer capability and enforced intramolecular cooperation, PIBC-8 also exhibited highest catalytic efficiency in the case of methyl o-methoxylphenyl sulfide (Table 1, entry 10 vs. entries 7–9), methyl p-methoxylphenyl sulfide (Table 1, entry 14 vs. entries 11–13), methyl p-nitrophenyl sulfide (Table 1, entry 18 vs. entries 15–17), and methyl p-bromophenyl sulfide (Table 1, entry 22 vs. entries 19–21), as shown in Table 1. High yield of sulfoxides (74–84%) with excellent chiral induction (90–91%) was observed for the methoxyl group substituted methyl aryl sulfides (Table 1, entries 10 and 14). Whereas the yields were not encouraging in the case of the methyl p-nitrophenyl sulfide and methyl p-bromophenyl sulfide, probably due to that the substrates were solid and insoluble in water. Poor water-solubility of the sulfides lowered the concentration of substrate in aqueous phase, which thus decreased the reaction rate of the aqueous phase catalysis. Simple pretreatment of the solid substrate in water by ultrasonic resulted in the increase of activity due to enhanced compatibility (Table 1, entry 23 vs. entry 22). The water-solubility of substrate had a minimal effect on the enantioselectivity.

Reusability

Apart from endowing the PIBC with built-in phase transfer capability and assisting intramolecular cooperation, the PEG bridging spacer with selective solubility also facilitated the separation of catalysts through solvent precipitation. PIBC-n (n = 4, 8, 13) could be precipitated out from the reaction system by adding sufficient n-hexane to reaction mixture, as shown by the typical PIBC-8 in Fig. 1e. Fig. 4 showed the results of the recovery and reusability of PIBC-n (n = 4, 8, 13) in asymmetric sulfoxidation of methyl phenyl sulfide in water using H₂O₂ as an oxidant (Fig. 4A–C).

Fig. 4 Reuse of PIBC-4 (A), PIBC-8 (B) and PIBC-13 (C) in asymmetric oxidation of methyl phenyl sulfide in water using aq. H₂O₂.

We noticed that the reusability of PIBC-n (n = 4, 8, 13) was associated with the total length of polyether chain of corresponding complex. PIBC-4 gave slight decrease in sulfoxide yield upon successive use (Fig. 4A), whereas there was negligible loss of activity if PIBC-8 or PIBC-13 was used as catalyst (Fig. 4B and C). The difference should be related with solubility of corresponding catalysts. PIBC-4, which was well soluble in water, suffered minor leaching loss in aqueous system, resulting in slight activity decrease. About 0.45 ppm of titanium content was indeed determined in PIBC-4-based reaction medium via chemical analysis (ICP). Whereas, no titanium (less than 0.1 ppm by ICP) was detected in the supernatant of PIBC-8 or PIBC-13, which revealed negligible leaching loss of titanium species during the oxidation. Furthermore, all PIBC-n (n = 4, 8, 13) showed retentive ee values during the asymmetric sulfoxidation. The results suggested that the catalysts were perfectly stable during sulfoxidation in this work, and ligand degradation, a main reason for deactivation of metallosalens in oxidation reactions,⁴¹ did not occur. FT-IR spectra of the typical fresh and recovered PIBC-8 (Fig. 2b vs. b′) demonstrated the stability of catalyst. No significant changes of the catalyst took place even after reuse for five times. Imidazolium IL moiety may play a positive effect on stabilizing the complex. Furthermore, short reaction time (within 45 min) and low temperature (25 °C) used in our studies would prevent or at least slow catalyst degradation.

Conclusions

We have first prepared the binuclear chiral salen Ti^IV catalysts of PIBC-n (n = 4, 8, 13) by using PEG-based dicationic IL as a bridge. The amphiphilic bridging spacer not only enforced the intramolecular cooperation favored in the catalysis, but also endowed the chiral salen Ti^IV catalyst with built-in phase transfer capability. The obtained PIBC-n (n = 4, 8, 13) thus functioned as the cooperative, phase transfer catalysts in asymmetric sulfoxidation in water. Remarkable enhancement of catalytic efficiency with higher ee values was observed over the PIBC-n (n = 4, 8, 13), especially, PIBC-8, for the aqueous organic reaction. Furthermore, the PTCs could be facilely recovered from the aqueous reaction mixture for efficient reuse.

Acknowledgements

The work was supported by the National Natural Science Foundation of China (NSFC 21476069, 21003044), the Scientific Research Fund of Hunan Provincial Education Department (13B072), the Program for Excellent Talents in Hunan Normal University (ET14103), the Program for Science and Technology Innovative Research Team in Higher Educational Institutions of Hunan Province, the Foundation for Innovative Research Groups of the Hunan Natural Science Foundation of China, the Construct Program of the Key Discipline in Hunan Province.

References

1. B. Ferber and H. B. Kagan, Adv. Synth. Catal., 2007, 349, 493.
2. M. A. Fascione, S. J. Adshead, P. K. Mandal, C. A. Kilner, A. G. Leach and W. B. Turnbull, Chem.–Eur. J., 2012, 18, 2987.
3. Y. Gong and A. Ebrahim, Environ. Sci. Technol., 2012, 47, 568.
4. H. Srour, P. L. Maux, S. Chevance and G. Simonneaux, Coord. Chem. Rev., 2013, 257, 3030.
5. B. M. Trost and M. Rao, Angew. Chem., Int. Ed., 2015, 54, 5026.
6. B. Saito and T. Katsuki, Tetrahedron Lett., 2001, 42, 8333.
7. B. Saito and T. Katsuki, Tetrahedron Lett., 2001, 42, 3873.
8. T. Tanaka, B. Saito and T. Katsuki, Tetrahedron Lett., 2002, 43, 3259.
9. B. Saito and T. Katsuki, Chirality, 2003, 15, 24.
10. Y. Chen, R. Tan, W. Zheng, Y. Zhang, G. Zhao and D. Yin, ChemCatChem, 2015, 7, 4066.
11. K. M. Gligorich and M. S. Sigman, Chem. Commun., 2009, 3854.
12. A. Chanda and V. V. Fokin, Chem. Rev., 2009, 109, 725.
13. N. Yan, C. X. Xiao and Y. Kou, Coord. Chem. Rev., 2010, 254, 1179.
14. Y. Uozumi and R. Nakao, Angew. Chem., Int. Ed., 2003, 42, 194.
15. S. Luo, X. Mi, S. Liu, H. Xu and J. Cheng, Chem. Commun., 2006, 3687.
16. S. Crossley, J. Faria, M. Shen and D. E. Resasco, Science, 2009, 327, 68.
17. Z. P. Wang, M. C. M. van Oers, F. P. J. T. Rutjes and J. C. M. van Hest, Angew. Chem., Int. Ed., 2012, 51, 10746.
18. H. Yang, T. Zhou and W. Zhang, Angew. Chem., Int. Ed., 2013, 52, 7455.
19. R. Tan, Y. Dong, M. Peng, W. Zheng and D. Yin, Appl. Catal., A, 2013, 458, 1.
20. Y. Zhang, R. Tan, G. Zhao, X. Luo and D. Yin, Catal. Sci. Technol., 2016, 6, 488.
21. W. Shan, F. Meng, Y. Wu, F. Mao and X. Li, J. Organomet. Chem., 2011, 696, 1687.
22. Y. Chen, R. Tan, G. Zhao, W. Zheng, R. Luo and D. Yin, Appl. Catal., A, 2015, 491, 106.
23. R. Luo, R. Tan, Z. Peng, W. Zheng, Y. Kong and D. Yin, J. Catal., 2012, 287, 170.
24. C. E. Song and E. J. Roh, Chem. Commun., 2000, 10, 837.
25. R. M. Haak, S. J. Wezenberg and A. W. Kleij, Chem. Commun., 2010, 46, 2713.
26. J. H. Kim and G. J. Kim, Catal. Lett., 2004, 92, 123.
27. C. Lv, D. Xu, S. Wang, C. Miao, C. Xia and W. Sun, Catal. Commun., 2011, 12, 1242.
28. J. Chen, S. K. Spear, J. G. Huddleston and R. D. Rogers, Green Chem., 2005, 7, 64.
29. F. Alonso and M. Yus, ACS Catal., 2012, 2, 1441.
30. M. R. Nabid and Y. Bide, Appl. Catal., A, 2014, 469, 183.
31. R. Tan, D. Yin, N. Yu, H. Zhao and D. Yin, J. Catal., 2009, 263, 284.
32. F. Song, C. Wang, J. M. Falkowski, L. Ma and W. Lin, J. Am. Chem. Soc., 2010, 132, 15390.
33. H. Zhi, C. Lv, Q. Zhang and J. Luo, Chem. Commun., 2009, 2878.
34. L. Frunza, H. Kosslick, H. Landmesser, E. Hijft and R. Fricke, J. Mol. Catal. A: Chem., 1997, 123, 179.
35. Z. Zhang, Z. Wang, R. Zhang and K. Ding, Angew. Chem., Int. Ed., 2010, 49, 6746.
36. N. C. Maity, S. H. R. Abdi, R. I. Kureshy, N. H. Khan, E. Suresh, G. P. Dangi and H. C. Bajaj, J. Catal., 2011, 277, 123.
37. R. I. Kureshy, T. Roy, N. H. Khan, S. H. R. Abdi, A. Sadhukhan and H. C. Bajaj, J. Catal., 2012, 286, 41.
38. J. Park, K. Lang, K. A. Abboud and S. Hong, Chem.–Eur. J., 2011, 17, 2236.
39. Z. Sun, G. Li, H. Liu and L. Liu, Appl. Catal., A, 2013, 466, 98.
40. D. E. White, P. M. Tadross, Z. Lu and E. N. Jacobsen, Tetrahedron, 2014, 70, 4165.
41. C. Baleizão and H. Garcia, Chem. Rev., 2006, 106, 3987.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra01130f

---