Combining the catalytic enantioselective reaction of visible-light-generated radicals with a by-product utilization system

We report an unusual reaction design in which a chiral bis-cyclometalated rhodium(iii) complex enables the stereocontrolled chemistry of photo-generated carbon-centered radicals and at the same time catalyzes an enantioselective sulfonyl radical addition to an alkene.


General Information
All catalytic reactions were carried out under an atmosphere of nitrogen with magnetic stirring in a Schlenk tube (10 mL). The catalysts Δ-IrS 1 and Δ-RhS 2 were synthesized according to our published procedures. Δ/Λ-RhO were synthesized with some modifications (see Section 2). 3 Solvents were distilled under nitrogen from calcium hydride (CH3CN, CH2Cl2), sodium/benzophenone (THF,Et2O). HPLC grade of acetone, methanol, and ethanol was used without further purification. Dry 1,4-dioxane was bought from Alfa-Aesar. Reagents that were purchased from commercial suppliers were used without further purification. Flash column chromatography was performed with silica gel 60 M from Macherey-Nagel (irregular shaped, 230-400 mesh, pH 6.8, pore volume: 0.81 mL  g -1 , mean pore size: 66 Å, specific surface: 492 m 2  g -1 , particle size distribution: 0.5% < 25 m and 1.7% > 71 m, water content: 1.6%). 1 H NMR, 19 F NMR and proton decoupled 13 C NMR spectra were recorded on Bruker Avance 300 (300 MHz), or Bruker AM (500 MHz) spectrometers at ambient temperature. NMR standards were used as follows: 1 H NMR spectroscopy:  = 7.26 ppm (CDCl3). 19 F NMR spectroscopy:  = 0 ppm (CFCl3). 13 C NMR spectroscopy:  = 77.0 ppm (CDCl3). IR spectra were recorded on a Bruker Alpha FT-IR spectrophotometer. High-resolution mass spectra were recorded on a Bruker En Apex Ultra 7.0 TFT-MS instrument using ESI technique. HPLC chromatography on chiral stationary phase was performed with an Agilent 1200 or Agilent 1260 HPLC system. Optical rotations were measured on a Krüss P8000-T polarimeter with []D 22 values reported in degrees with concentrations reported in g/100 mL. The EPR spectrometer is from Bruker (model esp300), with a modified Varian rectangular X-band cavity and the modulation frequency was set to 100 kHz, the modulation amplitude was 0.1 mT. The Stern-Volmer quenching experiments were recorded on a Spectra Max M5 microplate reader in a 10.0 mm quartz cuvette.

Light sources and emission spectra of the lamps
A 21 W compact fluorescent lamp (CFL, OSRAM DULUX ® SUPERSTAR Micro Twist) or 24 W Blue LEDs (Hongchangzhaoming from Chinese Taobao, https://hongchang-led.taobao.com) served as light sources. Figures S1 and S2 display their emission spectra. S3 Figure S1. Emission spectrum of the 21 W CFL lamp. Figure S2. Emission spectrum of the 24 W blue LEDs. S4

Modifications for the Synthesis of Λ/Δ-RhO
Racemic RhO complex was synthesized according to our previous procedures, 3 in which the enantiopure RhO was obtained through a proline-mediated route resulting in a loss of at least 50% of rhodium complex. Herein, we modified the resolution process using a chiral auxiliary (R)-Aux, namely (R)-3-fluoro-2-(4-phenyl-4,5-dihydrooxazol-2-yl)phenol, instead of proline. The corresponding complexes Λ/Δ-(R)-RhO are stable and could be separated by flash chromatography, thus improving the atom economy of the catalyst synthesis.

General procedure A
To a mixture of diethyl (cyanomethyl)phosphonate (20 mmol) and a 37% aqueous solution of formaldehyde (80 mmol), a saturated aqueous solution of potassium carbonate (37.5 mmol) was added at room temperature dropwise over 30 min. After stirring for an additional 2 h, the reaction was quenched with saturated aqueous ammonium chloride (20 mL). Afterwards, the reaction mixture was extracted with diethyl ether (3 × 12.5 mL). The organic layers were combined and dried over sodium sulfate. The solvent was evaporated using a rotary evaporator, and the remaining colorless oil was purified by flash chromatography using pentane/CH2Cl2 (2/1) giving the pure product S1 as a colorless oil (70% yeild).
To a solution of S1 (14 mmol) in dry ether (20 mL) was added phosphorus(III) bromide (5 mmol) at 10 C. The temperature was allowed to rise to 20 C and stirring was continued for 3 h.
Water (10 mL) was then added and the mixture was extracted with diethyl ether (3 × 30 mL). The organic phase was washed with brine (20 mL), dried with sodium sulfate and concentrated under reduced pressure. The crude product was purified by column chromatography on silica gel (pentane/ CH2Cl2, 1/1) to give S2 as a colorless oil (89% yield).
To a solution of S2 (2.0 mmol) in methanol (5 mL) was added corresponding sodium aryl sulfinate (3.0 mmol). After 2.5 h of reflux, the mixture was concentrated under reduced pressure, the thereby obtained residue was dissolved in EtOAc and the mixture was washed with water, brine, dried with Na2SO4, filtered and the filtrate was evaporated and purified by chromatography (EtOAc/n-hexane, 1/1) to give corresponding products 2a-h. The characteristic data of 2a are in accord with literature. 4

General procedure B
To a solution of corresponding alcohol (ROH, 10 mmol) and triethylamine (15 mmol) in acetone (15 mL) was added acryloyl chloride (13 mmol) dropwise at 0 C. After stirring at 0 C for 30 min, the reaction mixture was warmed to room temperature and stirred for additional 5 h. The resulting mixture was concentrated, then taken up in EtOAc (50 mL) and washed with brine (3  10 mL). The organic extracts were dried over anhydrous Na2SO4, concentrated by rotary evaporation.
To a solution of a 37% aqueous solution of formaldehyde (7.0 mmol) and ester S3 (5 mmol) in 5 mL 1,4-dioxane-water (1:1, v/v) was added DABCO (7.0 mmol) and the reaction progress was monitored by TLC. Upon completion, the reaction mixture was partitioned with EtOAc (50 mL) and water (20 mL). The organic layer was separated and washed with brine (5 mL), dried over anhydrous Na2SO4 and concentrated under reduced pressure. The crude product was purified by column chromatography on silica gel (EtOAc/n-hexane, 1/1) to afford corresponding alcohol ester

S4.
To a solution of S4 (5 mmol) in dry ether (10 mL) was added phosphorus(III) bromide (1.7 mmol) dropwise at 10 C. The temperature was allowed to rise to 20 C and stirring was continued for 3 h. Water (20 mL) was then added and the mixture was extracted with diethyl ether (3 × 10 mL).
The organic phase was washed with saturated sodium chloride solution (5 mL), dried with sodium sulfate and concentrated under reduced pressure. The crude product was purified by column chromatography on silica gel (pentane/CH2Cl2, 1/1) to give corresponding brominated compound S8

S5.
To a solution of S5 (2.0 mmol) in methanol (5 mL) was added corresponding sodium aryl sulfinate (3.0 mmol). After 2.5 h of reflux, the mixture was concentrated under reduced pressure, the thereby obtained residue was dissolved in EtOAc and the mixture was washed with water, brine, dried with Na2SO4, filtered and the filtrate was evaporated and purified by chromatography (EtOAc/n-hexane, 1/1) to give corresponding products 2j-m.

General procedure C
To a solution of methyl phenyl sulfone (1.25 g, 8.0 mmol) in THF (40 mL) cooled at -78 ºC, n-BuLi (1.6 M in n-hexane, 5.5 mL, 8.8 mmol) was added dropwise under argon atmosphere. The resulting solution was stirred at 0 ºC for 30 min, and then cooled back to 78 ºC. A solution of 2,3,4,5,6-pentafluorobenzaldehyde (1.72g, 8.8 mmol) in THF (2.0 mL) was added dropwise and the temperature was allowed to slowly raise to room temperature, and the solution was stirred until methylphenylsulfone disappeared by TLC. A saturated aqueous solution of NH4Cl (20 mL) was added, the organic layer was separated and the aqueous layer was extracted with CH2Cl2 (3  10 mL). The combined organic layers were dried with Na2SO4 and evaporated under reduced pressure.
Without further purification, the resulting alcohol was dissolved in dry CH2Cl2 (25 mL) under argon atmosphere, cooled to 0 ºC, then Et3N (11.2 mL, 80 mmol) and methanesulfonyl chloride (0.93 mL, 12 mmol) were added continuously. After stirring at room temperature for 90 min, a saturated aqueous solution of NH4Cl (30 mL) was added, the organic layer was separated and the aqueous layer was extracted with CH2Cl2 (3  15 mL). The combined organic layers were dried (Na2SO4) and the solvent was evaporated. The residue was purified by flash chromatography (n-hexane/EtOAc, 5/1) to afford compound 5 (1.73g, 65%) as a white solid. S9

General Procedure
A dried 10 mL Schlenk tube was charged with 2a (20.7 mg, 0.10 mmol), Δ-RhO (6.6 mg, 8.0 mol%) and HE-1 (42.2 mg, 0.15 mmol, synthesized following a reported procedure 6 ). The tube was purged with nitrogen for three times. Then, 1,4-dioxane (1.0 mL, 0.10 M, bubbling with nitrogen gas for five minutes before addition) was added via syringe followed by addition of 1a (32.8 mg, 0.2 mmol) under nitrogen atmosphere. The tube was sealed and positioned approximately 5 cm away from a 21 W compact fluorescent lamp. The reaction was stirred at room temperature for the indicated time (monitored by TLC) under nitrogen atmosphere. Afterwards, the mixture was diluted with CH2Cl2. The combined organic layers were concentrated under reduced pressure. The residue was purified by flash chromatography on silica gel (n-hexane/EtOAc) to afford the products 3a and 4a. Racemic samples were obtained by carrying out the reactions with rac-RhO. The enantiomeric excess was determined by HPLC analysis on a chiral stationary phase. S16 Table S1. Effect of Lewis acid catalysts a a Reaction conditions: 1a (0.20 mmol), 2a (0.10 mmol), Lewis acid (8.0 mol%) and HE-1 (0.15 mmol) in 1,4-dioxane (0.1 M) were stirred at room temperature for 24 h with a 21 W CFL. b Δ-RhPP was synthesized according to our previous report. 7 c (20 mol%) of Sc(OTf)3 was employed. d (200 mol%) of LiBF4 was employed.  The reduced product, 1-(3,5-dimethyl-1H-pyrazol-1-yl)butan-1-one 8 , was detected in less than 5% yield.   Figure S1 for emission spectrum. c See Figure S2 for emission spectrum. S18

Identification of RhO-1a Intermediate
To a solution of rac-RhO (83.1 mg, 0.1 mmol) in CH2Cl2 (2 mL) was added α,β-unsaturated N-acylpyrazole 1a (16.4 mg, 0.1 mmol). The mixture was stirred at room temperature for 1 minute and then the solvent was removed in vacuum. The procedure was repeated for another 3 times until the ligand exchange finished completely (detected by 1 H NMR). The resulting solid was recrystallized from CH2Cl2/Et2O giving pure RhO-1a, which was characterized by single crystal X-ray diffraction (see Section 11).

Using BHT as a radical trap
As shown above, when BHT (3.0 equiv) was added to the reaction 1a+2a3a+4a under standard conditions, the reaction was significantly inhibited delivering 3a and 4a in decreased yields. S19

Using TEMPO as a radical trap
When TEMPO (3.0 equiv) was added to the reaction 1a+2a3a+4a under standard conditions, the reaction was completely inhibited.

Using 1,1-diphenyl ethylene a radical trap
When the 1,1-diphenylethylene (3.0 equiv) was added to the reaction 1a+2a3a+4a under standard conditions, the reaction was partly inhibited and the yields of the products were decreased to 20% for 3a and 22% for 4a.
All these control experiments indicate that radical processes might be involved in the present transformation.

UV-Vis absorption spectra and luminescence emission spectra
As shown in Figure S3, both HE-1 and RhO-1a absorb visible light with wavelength < 425 nm.
In order to simulate the reaction conditions, the luminescence quenching experiments were performed with the photoredox mediator Hantzsch ester alone (see section 6.3.2) and with the mixture of Hantzsch ester and RhO in a molar ratio of 2 : 1 (see section 6.3.3), respectively. Figure S3. UV-Vis absorption spectra and luminescence emission spectra. Concentration for absorption spectra in 1,4-dioxane: HE-1 = 0.05 mM, RhO = 0.05 mM, RhO-1a = 0.05 mM. Concentration for emission spectra of HE-1 in 1,4-dioxane = 0.5 mM.

Quenching experiments with the Hantzsch ester alone
The solutions of HE-1 (0.5 mM in 1,4-dioxane) were excited at  = 360 nm and the emission was measured at 455 nm (emission maximum). For each quenching experiment, after degassed with a nitrogen stream for 5 minutes, the emission intensity of the solution (1 mL  are not capable of quenching ( Figure S4). RhO might quench the luminescence of HE-1 via competitive absorption (inner filter effect). 10 Considering the similar absorption of RhO-1a and RhO ( Figure S3), the in situ generated RhO-1a can quench the luminescence of the mixture of HE-1 and RhO slightly ( Figure S5), indicating RhO-1a might undergoes a photoinduced electron transfer with HE-1. Furthermore, RhO-1a as the major existing species of rhodium complexes, is most likely responsible for oxidative quenching of photoexcited HE-1, which is further supported by cyclic voltammetry studies (see Section 6.4).

Cyclic Voltammetry
All cyclic voltammetry experiments were carried out using analytical grade CH2Cl2 as the solvent containing 0.1 M Bu4NPF6 as the electrolyte and 1 mM of the analyte. Cyclic voltammetry experiments were conducted with a computer controlled Eco Chemie Autolab PGSTAT302N potentiostat in a Metrohm electrochemical cell containing a 1 mm diameter planar glassy carbon (GC) disk electrode (eDAQ), a platinum wire auxiliary electrode (Metrohm) and a silver wire miniature reference electrode (eDAQ) that was connected to the test solution via a salt bridge containing 0.5 M nBu4NPF6 in CH3CN. Accurate potentials were referenced to the ferrocene/ferrocenium (Fc/Fc + ) redox couple, which was used as an internal standard. All solutions used for the voltammetric experiments were deoxygenated by purging with high purity argon gas and measurements were performed in a Faraday cage at room temperature (22 ± 2 o C).
Substrate 1a showed one chemically irreversible reduction process with a cathodic peak potential (Ep red ) at -2.59 V vs. Fc/Fc + ( Figure S6, red curve). RhO-1a could be reduced with an Ep red at approximately -1.62 V vs. Fc/Fc + and oxidised with an Ep ox at approximately +1.32 V vs.
Fc/Fc + , both in chemically irreversible processes ( Figure S6, blue curve). It is noteworthy that coordination of the cyclometalated rhodium catalyst could significantly decrease reductive potential of 1a.
Besides, HE-1 could be oxidised in a chemically irreversible process with an anodic peak potential (Ep ox ) at approximately 0.50 V vs. Fc/Fc + ( Figure S7). According to luminescence emission spectra ( Figure S3, maximum wavelength = 455 nm, corresponding to 2.73 eV), the redox potential of photoexcited HE-1 is estimated as -2.23 V vs. Fc/Fc + , which is feasible to selectively reduce RhO-1a instead of free 1a. Figure S6. CV of compound 1a and RhO-1a. Figure S7. CV of compound HE-1.

Trapping experiments with but-3-en-2-one
To trap the sulfonyl radical, but-3-en-2-one which is not able to bind the Rh catalyst was added to act as a sulphonyl trap. As shown, the expected radical trapping product 4a' could be obtained in S24 28% yield, along with the formation of 3a and 4a, indicating the involvement of sulfonyl radical.  133.9, 129.4, 127.9, 50.5, 35.8, 29.8. All characteristic data are consistent with the literature report. 11

EPR experiments
EPR spectra were recorded at room temperature using DMPO (5,5-dimethyl-1-pyrroline N-oxide) as free radical spin trapping agent. According to general procedure, the reaction of 1a and 2a under standard conditions with the addition of 10 µL DMPO solution (1M in H2O) was stirred with 21 W CFL for 30 min. Then, a portion of the reaction mixture was taken out to an EPR tube and measured by EPR (9.18142 GHz; Mod. Frequency = 100 kHz; Mod. Ampl. = 0.08 mT).
As shown in Figure S8, two sets of signals were observed, one of which is simulated as signals 1 with 6 lines (g = 2.006; αN = 9.5 G, αH β = 12.9 G) and further signed as EPR signals of sulfonyl radical adducts. 12 These results suggest more than one radical species including sulfonyl radicals are involved in this transformation.

Determination of the Quantum Yield
The quantum yield of the title reaction 1a+2a3a+4a was determined by a method and setups developed by Prof. Dr. Eberhard Riedle's Group. 13 As light source 420 nm LEDs were employed.
A powermeter was used as detector. The measurement was accomplished in a dark room with a 1.1 W red LEDs.
Step 1: The radiant power of light transmitted by the cuvette with a blank solution was measured as Pblank = 46.25 mW.
Step 4: The overall quantum yield can be calculated as following: where Nproduct is the number of product 3a formed; Nphoton is the number of photons absorbed; NA is Avogadro's constant; nproduct is the molar amount of product 3a formed; Pabsorbed is the radiant power absorbed; t is the irradiation time; h is the Planck's constant; c is the speed of light; λ is the wavelength of light source, Pblank is the radiant power transmitted by the cuvette with a blank solution; Psample is the radiant power transmitted by the cuvette with reaction mixture.

Single-Crystal X-Ray Diffraction Studies
X-ray data were collected with a Bruker 3 circuit D8 Quest diffractometer with MoKα radiation (microfocus tube with multilayer optics) and Photon 100 CMOS detector at 100 K. Scaling and absorption correction was performed by using the SADABS software package of Bruker. Structures were solved using direct methods in SHELXT and refined using the full matrix least squares procedure in SHELXL-2014. The hydrogen atoms were placed in calculated positions and refined as riding on their respective C atom, and Uiso(H) was set at 1.2 Ueq(Csp 2 ) and 1.5 Ueq(Csp 3 ).
Disorder was refined using restraints for both the geometry and the anisotropic displacement factors.
The absolute configuration of 4d and 9 have been determined.

Crystal structure of RhO-1a
Single crystals of RhO-1a suitable for X-ray diffraction were obtained by slow diffusion from a solution of raacemic RhO-1a (20 mg) in CH2Cl2 (2.0 mL) layered with ethyl ether (1.0 mL) at room temperature for several days in a NMR tube.
Crystal structure, data and details of the structure determination for RhO-1a are presented in the Figure S102 and Table S6.

Crystal structure of 4d
Single crystals of 4d suitable for X-ray diffraction were obtained by slow diffusion from a solution of 4d (20 mg) in ethyl ether (0.5 mL) layered with n-hexane (0.5 mL) at 4 o C for several days in a NMR tube.
Crystal structure, data and details of the structure determination for 4d are presented in the Figure S103 and Table S7. Figure S103. Crystal structure of 4d.

Crystal structure of 9
Single crystals of compound 9, which was obtained via transamidation of (R)-3a (obtained from the reactions catalyzed by Λ-RhO, suitable for X-ray diffraction were obtained by slow diffusion from a solution of 9 (30 mg) see Section 8.3), in CH2Cl2 (0.5 mL) layered with n-hexane (0.5 mL) at room temperature for several days in a NMR tube.
Crystal structure, data and details of the structure determination for 9 are presented in the Figure S104 and Table S8. Figure S104. Crystal structure of 9. S100 Table S8. Crystal data and structure refinement for 9.