Vitamin B12 and hydrogen atom transfer cooperative catalysis as a hydride nucleophile mimic in epoxide ring opening

SUMMARY Epoxide ring-opening reactions have long been utilized to furnish alcohol products that are valuable in many subfields of chemistry. While many epoxide-opening reactions are known, the hydrogenative opening of epoxides via ionic means remains challenging because of harsh conditions and reactive hydride nucleophiles. Recent progress has shown that radical chemistry can achieve hydrogenative epoxide ring opening under relatively mild conditions; however, these methods invariably require oxophilic metal catalysts and sensitive reagents. In response to these challenges, we report a new approach to epoxide ring-opening hydrogenation using bio-inspired Earth-abundant vitamin B12 and thiol-centric hydrogen atom transfer (HAT) co-catalysis to produce Markovnikov alcohols under visible light irradiation. This powerful reaction system exhibits a broad substrate scope, including a number of electrophilic and reductively labile functionalities that would otherwise be susceptible to reduction or cleavage by hydride nucleophiles, and preliminary mechanistic experiments are consistent with a radical process.


Reaction Set-Up
A 427-nm, blue LED light (Kessil®, PR160L) was used at 100% intensity to irradiate the reactions. The reaction vial was positioned approximately 3 cm from the surface of the lamp such that the solventcontaining subsection was centered in the light beam. Though not explicitly shown, a three-sided "igloo" (tin foil taped to the inside of a cardboard 'C') was placed surrounding the stir plate-LED apparatus to contain the light. For all reactions (0.2 mmol and 0.4 mmol scale), 8-mL vials with screw-top septa-caps (Chemglass® CG-4909, 17x60 mm clear borosilicate glass w/ TFE septa) were used. No cooling fans nor light filters were employed for this reaction apparatus ( Figure S1).
The spectral irradiance of the Kessil PRL160L-427 is available at the following address: https://kessil.com/products/science_PR160L.php To an 8 mL septum screw-capped vial equipped with magnetic stir bar, was added (2,3-epoxypropyl)benzene (26.8 mg, 0.2 mmol, 1.0 equiv), Cobalt Catalyst, Zn (39.2 mg, 0.6 mmol, 3.0 equiv), NH4Cl (32.1 mg, 0.6 mmol, 3.0 equiv), hydrogen atom transfer (HAT) donor, and methanol solvent (2 mL). The resultant mixture was sparged with a nitrogen balloon for 10 min and the vial was sealed with parafilm. The mixture was then stirred and irradiated with a 427 nm LED (Kessil®) for 24 hr. The reaction mixture was then concentrated in vacuo, re-suspended in dichloromethane, and sonicated before being passed through a cotton pipette filter. The resulting filtrate was concentrated in vacuo. Yield was determined by 1 H NMR analysis using 1,3,5trimethoxybenzene as an internal standard.

Mechanistic Considerations TEMPO Radical Inhibition Experiment
To an 8 mL septum screw-capped vial equipped with magnetic stir bar was added (2,3-epoxypropyl)benzene (53.7 mg, 0.400 mmol, 1.0 equiv), Vitamin B12 (5.4 mg, 1 mol %), Zn (78.5 mg, 1.201 mmol, 3.0 equiv), NH4Cl (64.2 mg, 1.200 mmol, 3.0 equiv), and TEMPO (78.2 mg, 0.500 mmol, 1.25 equiv). 2,4,6-triisopropylbenzene thiol ("TRIP thiol") was then added from a 0.1 mmol/mL stock solution in deuterated methanol (0.4 mL, 10 mol %) followed by addition of the deuterated methanol solvent (3.6 mL). This solution was sparged with a nitrogen balloon for 10 min and the vial was sealed with parafilm. The mixture was then stirred and irradiated with a 427 nm LED (Kessil®) for 48 hr. The reaction mixture was then concentrated in vacuo, resuspended in dichloromethane, and sonicated before being passed through a cotton pipette filter. The filtrate was concentrated in vacuo. Yield was determined by 1 H NMR analysis using 1,3,5-trimethoxybenzene as an internal standard. The significant decrease in yield when TEMPO is present relative to the control is consistent with radical inhibition, and suggests the reaction proceeds through a radical mechanism. However, we are cognizant of the findings of Sen and coworkers demonstrating that non-radical reactions can be inhibited by TEMPO, showing that these results must be interpreted with care. 1

Radical Clock Experiment
To further investigate the proposed radical mechanism of our reaction, a radical clock experiment was conducted with 1,2-epoxy-5-hexene as the substrate. If indeed a radical is formed at the alpha position to the alcohol upon light-mediated homolytic cleavage of a Co-C bond, the intermediate species would be primed to undergo a 5-exo-trig and/or 6-endo-trig radical cyclization because of its position relative to the olefin in the starting material.

Figure S2. GC-MS Chromatogram of Radical Clock Experiment
Notably, the signal at t = 3.617 min displayed a mass spectrum ( Figure S3) that very closely resembles (73% match with the NIST mass spectrum library of organic molecules) that of 3-methylcyclopentanol ( Figure  S4). 2 This strongly suggests that 1) a radical is generated at the terminal carbon of the epoxide substrate during the reaction mechanism, and 2) the 5-exo-trig radical cyclization product, 3-methycyclopropanol is present in the crude reaction mixture when 1,2-epoxy-5-hexene is subjected to our epoxide ring-opening reaction. Additionally, a standard of cyclohexanol was analyzed on GC-MS resulting in the chromatogram shown below ( Figure S5). The standard had an acquisition time of 4.582 minutes. This is significant because a minor signal is present at ~4.57 minutes in the chromatogram of the radical clock reaction ( Figure S6) which gives reasonable evidence that the 6-endo-trig cyclization byproduct is also present.

Deuterium Labeling Experiment
We wondered if perhaps the methanol solvent was active in replenishing the thiol with hydrogen to continue the catalytic HAT process, as opposed to solely being supplied by NH4Cl, the strongest acid in the reaction system. To gain some clarity in this matter, we sought to run the control reaction in deuterated methanol.
Presumably if the methanol solvent was serving as a proton source in the HAT catalytic cycle, we would see deuterium incorporation of the product when reacted in a deuterated methanol solvent.
The NMR spectra of the isolated product ( Figures S7 and S8) strongly suggest deuterium incorporation. Relative to the 1 H NMR spectrum of the unlabeled alcohol ( Figure S9), the methyl proton peak displays a notable decrease in integration 2.42 which indicates approximately 58% deuterium incorporation. Additionally, in the 13 C NMR spectrum of the deuterated alcohol, two signals are visible corresponding to the 1-cyclohexanol?
terminal carbon of the alkyl chain. One signal occurs at 22.8 ppm, identical to that seen in the unlabeled alcohol's 13 C NMR ( Figure S10). However, a 1:1:1 triplet centered at 22.5 ppm is also observed; this is further evidence of partial deuterium incorporation in the alcohol product when the epoxide is reacted in deuterated methanol.
For each time point, a 100 µL aliquot was taken from the reaction mixture and dissolved in 800 µL of CDCl3. This solution was extracted with 3 mL of H2O; the organic layer was removed and used directly to acquire a 1 H NMR spectrum.

Syntheses of Epoxide and Oxetane Substrates
General Procedure 2 3-chloroperbenzoic acid (2.5 equiv) was added to a 150-mL round bottom flask equipped with stir bar and dissolved in dichloromethane (15 mL). The flask was sealed with a rubber septum and equipped with a vent needle. In a separate round bottom flask, alkene (1.0 equiv) was dissolved in dichloromethane (10 mL). The alkene/DCM solution was slowly added (over ~1 min) to the m-CPBA mixture via syringe and the reaction was allowed to stir overnight. The reaction was quenched with sat. NaHCO3 and allowed to stir for 20 min and then extracted with dichloromethane. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated in vacuo. The epoxides were isolated using silica gel column chromatography.

General Procedure 3
To a 100-mL, two-necked round bottom flask equipped with a magnetic stir bar was added p-substituted phenol (1.0 equiv) and potassium carbonate (2.0 equiv). One neck was plugged with a rubber septum and the flask was placed under nitrogen using Schlenk technique. Anhydrous acetone (15 mL) and allyl bromide (1.1 equiv) were added via syringe and the reaction solution was refluxed (oil bath) overnight. After cooling to room temperature, the solvent was concentrated in vacuo. The resulting residue was diluted in 30 mL (each) of ethyl acetate and H2O, then extracted with ethyl acetate. Combined organic layers were washed with brine, dried over Na2SO4, filtered, and concentrated in vacuo. 8 The resulting crude product was used in General Procedure 4 without further purification.

General Procedure 4
To a 100-mL round bottom flask equipped with a magnetic stir bar was added 3-chloroperbenzoic acid (1.8 equiv) and dichloromethane (15 mL), and the mixture was stirred until m-CPBA fully dissolved. The reaction was cooled to 0 ⁰C (ice bath) and allyl ether (crude from General Procedure 3) was added. The reaction flask was sealed with a septum/vent needle and stirred overnight. Reaction was quenched with 10 mL H2O and 10 mL aqueous NaOH (2M), then extracted with dichloromethane. The combined organic layers were washed with brine, dried over Na2SO4, filtered, and concentrated in vacuo. 8 The epoxide products were purified by silica gel column chromatography.

Synthesis of benzyl-10,11-epoxy-undecanoate (1K)
Undec-10-en-1-oic acid (S1, 1.843 g, 10.0 mmol, 1.0 equiv), K2CO3 (1.658 g, 12.0 mmol, 1.2 equiv), and benzyl bromide (1.1 mL, 9.3 mmol, 0.93 equiv) were added to a 50-mL two-necked round bottom flask equipped with a magnetic stir bar. One neck was plugged with a rubber septum, and the flask was placed under vacuum and flushed with nitrogen (Schlenk technique). The reagents were then dissolved in anhydrous DMF (22 mL) and stirred at room temperature overnight. The reaction was quenched with 1M HCl (15 mL), then extracted with hexane. The combined organic phase was dried over MgSO4, filtered, and concentrated in vacuo. The crude product was used without further purification.
In a 100-mL round bottom flask equipped with magnetic stir bar, benzyl undec-10-en-1-oate (S2, crude from previous step) was dissolved in DCM (50 mL) and stirred at 0 ⁰C, then m-CPBA (2.243 g, 13.0 mmol, 1.3 equiv) was added slowly and the reaction was allowed to warm to room temperature and stirred overnight. The reaction was quenched with 1M NaOH (50 mL) and extracted DCM. The combined organic phase was dried over MgSO4, filtered, and concentrated in vacuo. The product (1K) was obtained without further purification as a clear oil (2.421 g, 83% over 2 steps

General Procedure 5
In a 100-mL round bottom flask equipped with a magnetic stir bar, X-H substrate (1.0 equiv) was dissolved in dichloromethane (30 mL). To this mixture was added K2CO3 (1.5 equiv) and epichlorohydrin (1.5 equiv). The reaction was allowed to stir overnight at room temperature. After such time, the reaction was quenched with H2O (25 mL) and extracted with DCM. The combined organic layers were dried over MgSO4, filtered, and concentrated in vacuo. Purification by silica gel column chromatography (20% ethyl acetate in hexane) afforded the epoxide products.

Synthesis of 2-(oxiran-2-yl)-1-phenylethan-1-one (1M)
A 500-mL round bottom flask equipped with a magnetic stir bar was charged with Zn (3.062 g, 46.83 mmol, 3.0 equiv) and the flask was placed under vacuum and seal with a rubber septum. Anhydrous tetrahydrofuran (50 mL) was added and the mixture was brought to 0 ⁰C. After stirring for 10 min at 0 ⁰C, allyl bromide (6.352 g, 52. 50 mmol, 3.5 equiv) was slowly added and the solution was allowed to stir at room temperature for 1 hr. Benzaldehyde (S3, 1.591 g, 14.99 mmol, 1.0 equiv) was then added to solution via syringe and the reaction was stirred overnight. The reaction was quenched with sat. NH4Cl and extracted with ethyl acetate. The combined organic layers were wash with brine, dried over MgSO4, filtered, and concentrated in vacuo. The resulting residue was used without further purification.
To a new 500-mL round bottom flask equipped with a magnetic stir bar, the crude allyl alcohol product (S4) was dissolved in dichloromethane (50 mL) and the solution was brought to 0 ⁰C. Dess-Martin periodinane (DMP, 6.360 g, 14.99 mmol, 1.0 equiv) was added and the solution was stirred for 10 min at room temperature. The reaction was quenched with sat. NaHCO3, and extracted with dichloromethane. The combined organic layers were washed with brine, dried of MgSO4, filtered, and concentrated to afford the crude product.

Synthesis of N-methyl-N-glycidyl-aniline (1N)
To a 500-mL round bottom flask equipped with a magnetic stir bar was added N-methylaniline (S6, 2.21 mL, 2.0 mmol, 1.0 equiv) and DCM (50 mL). A solution of KOH (2.24 g, 40.0 mmol, 20.0 equiv) in H2O (20 mL) was added to the mixture. Epichlorohydrin (2.00 mL, 25.0 mmol, 12.5 equiv) was also added. The reaction mixture was stirred at room temperature overnight. After such time, the reaction was quenched with H2O (100 mL) and extracted with DCM. The combined organic layers were dried with MgSO4, filtered, and concentrated in vacuo. The crude product (S7) was used in the next step without further purification.

Synthesis of 2-benzyloxymethyloxetane (S11)
To a 25-mL two-necked round-bottom flask fitted with a magnetic stir bar was added NaH (60% dispersion in oil) (60.0 mg, 1.50 mmol, 1.5 equiv). One neck was plugged with a rubber septum and the flask was placed under nitrogen using Schlenk technique. 2-hydroxymethyloxetane (100.0 mg, 1.13 mmol, 1.0 equiv) was added to the flask, the mixture was dissolved in anhydrous THF (6 mL), and it was allowed to react at 0 °C for 1 hr. To an 8 mL septum screw-capped vial equipped with magnetic stir bar was added epoxide (1A-Q, 1.0 equiv), Vitamin B12 (cyanocobalamin, 1 mol %), Zn (3.0 equiv), and NH4Cl (3.0 equiv). 2,4,6-triisopropylbenzene thiol ("TRIP thiol") was then added from a 0.1 mmol/mL stock solution in methanol (10 mol %) followed by the methanol solvent (0.1 mmol epoxide/mL methanol). This solution was sparged with a nitrogen balloon for 10 min and the vial was sealed with parafilm. The mixture was then stirred and irradiated with a 427 nm LED (Kessil®) for 2-48 hr. After reaction completion, the reaction mixture was concentrated in vacuo, suspended in dichloromethane, and sonicated before being passed through a cotton pipette filter. The filtrate was concentrated in vacuo and purified by silica gel column chromatography or preparative TLC to isolate the alcohol products featured below.