Counterion Enhanced Organocatalysis: A Novel Approach for the Asymmetric Transfer Hydrogenation of Enones

Abstract We present a novel strategy for organocatalytic transfer hydrogenations relying on an ion‐paired catalyst of natural l‐amino acids as main source of chirality in combination with racemic, atropisomeric phosphoric acids as counteranion. The combination of a chiral cation with a structurally flexible anion resulted in a novel chiral framework for asymmetric transfer hydrogenations with enhanced selectivity through synergistic effects. The optimized catalytic system, in combination with a Hantzsch ester as hydrogen source for biomimetic transfer hydrogenation, enabled high enantioselectivity and excellent yields for a series of α,β‐unsaturated cyclohexenones under mild conditions. Moreover, owing to the use of readily available and chiral pool‐derived building blocks, it could be prepared in a straightforward and significantly cheaper way compared to the current state of the art.


General remarks
All purchased chemicals from commercial suppliers were used without further purification. Dry solvents were pre-distilled and desiccated on aluminium oxide columns (PURESOLV, Innovative Technology).
Column chromatography was performed on standard manual glass columns using Merck (40-60 µm) silica gel with pre-distilled solvents (PE : petrolether, EtOAc : ethyl acetate, Et2O : diethyl ether). For TLC analysis, precoated aluminium-backed plates were purchased from Merck (silica gel 60 F254). UV active compounds were detected at 254 nm. Non-UV active compounds have been detected using vanillin staining solution (5% vanillin in EtOH + H2SO4). 1 H, 13  GC analysis have been performed on a Thermo Scientific Focus on BGB5 column by using FID detector. Chiral GC measurements were performed on chiral BGB columns (BGB173 or BGB175) by using FID detector.
Optical rotation was measured on an Anton Paar MCP500 polarimeter at the specific conditions and the results have been compared to literature values. Concentrations are given in g / 100 ml.
HR-MS analysis was performed using HTC PAL system auto sampler, an Agilent 1100/1200 HPLC and Agilent 6230 AJS ESI-TOF mass spectrometer.
Microwave reactions were performed on a Biotage Initiator Classic in 20 ml pressure tight glass vials.
Melting points above room temperature were measured on an automated melting point system OPTI MELT of Stanford Research Systems and are uncorrected.
Infrared spectra were recorded on a Perkin-Elmer Spectrum 65 FT IR spectrometer equipped with a specac MK II Golden Gate Single Reflection ATR unit.

Synthesis of L-amino acid derivatives 2.1 General procedure for the synthesis of L-amino acid tert-butyl esters
Compound have been prepared by the following literature procedure: 1 To a solution of the corresponding L-amino acid (1.0 equiv.) in tert-butyl acetate (18.0 equiv.), HClO4 (60 % aqueous solution, 1.5 equiv.) was added dropwise at 0°C. The reaction mixture was stirred 18 h at room temperature. Distilled H2O was added, and the reaction mixture was extracted with distilled H2O (2×) and 1 N HCl (3×). The combined aqueous phases were then adjusted to pH 10 with a saturated Na2CO3 solution. The resulting solution was then extracted with CH2Cl2. The combined organic phases were dried over anhydrous Na2SO4 and concentrated under reduced pressure to give the L-amino acid tert-butyl esters, which were find to be pure without further purifications.

General procedure for the synthesis of L-amino acid esters via DCC/DMAP coupling
L-Valine mentyl esters (all three diastereomers), L-valine cyclohexyl ester and L-valine 4-(tertbutyl)cyclohexyl ester have been prepared by the following two step general procedure: 3 A solution of L-valine (5.0 g, 43 mmol, 1.0 equiv.) in 2 N NaOH (43 mL) was cooled to 0°C and Boc2O (11.3 g, 51.6 mmol, 1.2 equiv.) was added slowly via syringe. The ice bath was removed and the reaction mixture stirred for 18 h at room temperature. The mixture was acidified (pH = 2) by using a 4 N HCl and extracted with EtOAc (3 × 50 mL). The combined organic phases were dried over anhydrous Na2SO4 and concentrated to give (tert-butoxycarbonyl)-L-valine as a colorless oil (8.3 g, 90%).
To a solution of (tert-butoxycarbonyl)-L-valine (1.0 equiv.) in dry CH2Cl2, DCC (1.1 equiv.) was added at room temperature. The suspension was cooled to 0°C followed by the addition of the corresponding alcohol (1.0 equiv.) and DMAP (0.1 equiv.). The reaction mixture was stirred for 18 h at room temperature followed by the addition of EtOAc. After 10 min of stirring, the dicyclohexyl urea was filtered off, rinsed with EtOAc and the filtrate was successively washed with 1 N HCl solution, saturated NaHCO3 solution, dried over anhydrous Na2SO4 and concentrated under reduced pressure. 4 N HCl in dioxane was added and the resulting mixture was stirred for 3 h at room temperature. The solvent was removed, Et2O was added and stirred for 10 min. The amino acid HCl salt was filtered off, washed with Et2O (3×) and dried in vacuo. The salt was suspended in CH2Cl2, saturated NaHCO3 was added and stirred for 2 h at room temperature. The organic phase was separated, washed with saturated NaHCO3 (3×) and concentrated under reduced pressure to yield the corresponding esters.

Synthesis of other L-amino acid esters
Methyl L-valinate 5 To a slurry of L-valine methyl-ester hydrochloride (1.27 g, 7.6 mmol, 1.0 equiv.) in dry MeOH (1 mL), Et3N (1.6 mL, 12 mmol, 1.6 equiv.) was added. After 10 min of stirring, dry Et2O (30 mL) was added, the solution was cooled to 0°C and it was stirred for another 30 min. The triethylamine hydrochloride salt was filtered off and the filtrate was concentrated to yield the product as a colorless oil (800 mg, 81%). Isopropyl L-valinate 6 To a solution of L-valine (1.0 g, 8.5 mmol, 1.0 equiv.) in isopropanol (26 mL, 341 mmol, 40 equiv.) was added thionyl chloride (3.1 mL, 43 mmol, 5.0 equiv.) at 0°C. The reaction mixture was then refluxed for 18 h. After cooling to room temperature, the solvent was evaporated, Et2O (50 mL) was added and the amino acid hydrochloride salt was filtered off. CH2Cl2 (50 mL) was added, followed by the addition of sat. NaHCO3 (50 mL) and the two phases were stirred for 30 min. The organic phase was washed several times with sat. NaHCO3 (3 × 25 mL), dried over anhydrous Na2SO4 and concentrated to give the product as a colorless liquid (380 mg, 23%).

Benzyl L-valinate 7
To a solution of L-valine (1.0 g, 8.5 mmol, 1.0 equiv.) in benzyl alcohol (26 mL, 341 mmol, 40 equiv.) was added thionyl chloride (3.1 mL, 43 mmol, 5.0 equiv.) at 0°C. The reaction mixture was then stirred for 18 hours at room temperature. Et2O (50 mL) was added and the amino acid hydrochloride salt was S9 filtered off. CH2Cl2 (50 mL) was added, followed by the addition of sat. NaHCO3 (50 mL) and the two phases were stirred for 30 min. The organic phase was washed several times with sat. NaHCO3 (3 × 25 mL), dried over anhydrous Na2SO4 and concentrated under reduced pressure to give the product as a colorless liquid (1.12 g, 62%). 1

General synthesis of the phosphoric acids
The phosphoric acids have been prepared following either a one or a two-step reaction pathway: 1.) One step procedure: 8 To a solution of the corresponding diol (1.0 equiv.) in dry Et2O or THF pyridine (2.0 equiv.) was added.
The reaction mixture was cooled to 0°C and POCl3 (1.4 equiv.) was added slowly via syringe. The resulting reaction mixture was stirred for 18 h at room temperature. The pyridinium hydrochloride was filtered off and it was rinsed several times with Et2O. The filtrate was concentrated and recrystallized from n-heptane. The white solid was filtered off, washed with n-heptane (3×) and hydrolyzed with 6 N HCl for 3 h at reflux temperature. Filtration gave the products as a white solid. 2.) Two step procedure: 9,10 In a 20 mL microwave vial equipped with a stir bar, the corresponding phenol (1.0 equiv.) was dissolved in chlorobenzene (1.66 M). Di-tert-butyl peroxide (1.05 equiv.) was then added via syringe. The vial was capped and the reaction mixture was stirred for 15 min at room temperature. The vial was then placed in the microwave reactor and heated to 160°C (high absorption setting) and stirred for 15 min.
To a solution of the corresponding 2,2'-diol (1.0 equiv.) in pyridine, POCl3 (2.0 equiv.) was added slowly via syringe at 0°C. After stirring the reaction mixture for 24 h at 95°C, it was cooled to room temperature and distilled H2O was added slowly. The resulting clear solution was stirred for another 18 h at 95 °C. After cooling down to room temperature, 4 N HCl was added slowly. The precipitate of the product was filtered off and washed with 4 N HCl. The solid product was further hydrolyzed by redissolving it in CH2Cl2 and washing several times with 4 N HCl (3-4×) After being dried over anhydrous Na2SO4, removal of the solvent gave the phosphoric acids (Step B). 11 Column chromatography (1.5% EtOAc in light petrol) afforded the product as pale yellow solid (53% yield).  12 Column chromatography (2% EtOAc in light petrol) afforded the product as light yellow oil (58% yield). 1 13 Column chromatography (1.5% EtOAc in light petrol) afforded the product as white solid (45% yield).  14 Grey solid (63% yield).  16 Compound was prepared by the following modified literature procedure.
The remaining 3-alkylcyclohexenones and 3-arylcyclohexanones have been prepared by the following procedure, starting from 3-ethoxycyclohexenone (Acros Organics): 17 A 3 M solution of the Grignard reagent was prepared freshly from freshly ground Mg (1.0 equiv.), alkyl/aryl bromide (1.0 equiv.) in dry THF, and it was refluxed for 1 h. After being cooled to 0°C, 3ethoxycyclohexenone (1.0 equiv.) was added slowly and the reaction mixture was stirred for 18 h at room temperature. The reaction was quenched with 1 N HCl solution (50 mL) at 0°C. Et2O was added and the organic phase was washed with 1 N HCl solution (3 × 25 mL), sat. NaHCO3 (3 × 25 mL), dried over anhydrous Na2SO4 and concentrated under reduced pressure. The crude products were purified by column chromatography (Et2O: PE, UV TLC visualization) to provide the pure products. 18 Column chromatography (25% Et2O in PE) afforded the product as a yellow oil (67%

3-Phenethyl-2-cyclohexenone 24
The formation of the grignard-reagent and the reaction have been carried out at and Hantzsch ester (552 mg, 2.2 mmol, 1.2 equiv.). The reaction mixture was stirred at 50°C for 48 h.
After cooling to room temperature, Et2O (5 mL) and 4 M HCl (10 mL) was added and the mixture was stirred until the phases were transparent (30 min). The phases were separated and the organic phase S19 was washed with 4 M HCl (3 × 20 mL). The organic phase was dried over Na2SO4 and concentrated under reduced pressure. Purification by column chromatography (Et2O: PE, vanillin staining agent or UV visualization) gave the 3-substituted cyclohexanones.

Analytical data of ATH products (S)-3-Methylcyclohexanone
Column chromatography (25% Et2O in PE) afforded the product as a colorless oil.

Choice of method and basis set
Although several studies on the calculation of rotational barriers in substituted biphenyls have been published 27,28,29 , the only systematic protocol so far has been established by Masson. 30 A key finding of Masson's study is that for accurate barrier energies, inclusion of vibrational corrections is crucial. Since frequency computations as well as transition state optimizations for molecules as large as the substituted phosphoric acids can become quite expensive, a compromise between accuracy and efficiency is necessary in terms of the method.
The popular as well as efficient B3LYP functional 31 has been shown to perform well for the estimation of rotational barriers for unsubstituted and substituted biphenyls 29,30,32 .However, as also noted by However, Masson emphasizes that solvent effects cannot be confidently neglected for negatively charged biphenyls. Since solvent effects and likely specific interactions with both solvent and cation S33 need to be considered when computing rotational barriers of the anionic forms of phosphoric acids 4-7, the barrier of the corresponding anions will be the topic of more detailed future work.

Optimization of transition states
The systematic search for rotational transition states is not straightforward. Usually, fixing the torsional dihedral angle around the aryl-aryl bond to 0 degrees and optimizing this constrained structure yields good guesses for rotational transition states of substituted biphenyl. 30 The situation is trickier when the planar transition state is hindered by bulky ortho-substituents such as the isopropyl and tert-butyl groups in phosphoric acids 6 and 7. For all phosphoric acids, we optimized the structure while fixing one of the dihedral angles to 0° and used the result as guesses for the transition state optimizations. In addition, for phosphoric acid 6 we performed a systematic transition state search with the nudged elastic band (NEB) method implemented in ORCA. 40,41 We also used the result of this NEB calculation to construct another transition state guess for phosphoric acid 7. The imaginary frequencies of all transition states corresponded to an out-of-plane torsion of the phenyl rings.
For phosphoric acids 4 and 5, fully planar transition states as depicted in Figure S1 are obtained when using the constrained structures as transition state guesses.
As already mentioned, compounds 6 and 7 have methyl groups in the 6,6'-position, which hinder the planar transition state. Instead, they adopt a conformation in which the aryl-aryl-bond is out-of-plane and the two rings roughly coplanar. This peculiar transition state geometry for ortho-substituted biphenyls has been anticipated by Baddeley as early as 1946. 42 Ling and Harris proposed the associated mechanism in 1964. 43 They suggested an asynchronous nature of the racemization, with 6,6'-groups passing over each other subsequently instead of synchronously, minimizing the barrier energy. The outof-plane transition states as well as the asynchronous nature of the racemization has been confirmed by Masson's calculation. 30 The asynchronous mechanism is also visible in the case of both phosphoric acids 6 and 7. By constraining different dihedral angles during pre-optimization, two different transition states can be obtained. The first transition state corresponds to passing of the methyl groups prior to  Since no force field for MTBE was available, we chose n-methylbutylether as a solvent. The solvent parameters were taken from the CGenFF force field 46 . Simulation boxes contained a single ion pair and 216 molecules of n-methylbutylether. To consider all possible isomers and stereisomers, both the (E)and (Z)-form of the cation as well as both anion enantiomers were considered, yielding four different systems (see Table S8.). 10 replica simulations of every system were conducted under the following conditions: All simulations were performed with the program package CHARMM. 47 Initial configurations were created with PACKMOL. 48 All simulations were conducted in the NpT ensemble

S35
and employed the CHARMM CPT algorithm, the leapfrog integrator and the Hoover thermostat. A temperature of 330 K, a pressure of 1 atm and a non-bonded cutoff of 11 Å were used. Electrostatics were treated using the Particle Mesh Ewald method, with a grid size of approximately 1 Å, cubic splines of order 6 and a Κ of 0.41 Å -1 . The SHAKE algorithm was employed to fix the length of bonds to hydrogens.
Boxes were first equilibrated for 2 ns using a timestep of 1 fs, followed by production simulations of at least 38 ns (see Table S8 for system details) under the same conditions. Coordinates were written to disk every 250 steps. For trajectory analysis, the MDAnalysis library 49 as well as self-written python code was used. Plots were created with R (ggplot). 50 Hydrogen bonds were analyzed using CHARMM and a cutoff distance of 2.4 Å and a cutoff angle of 135°. Every 10 th step of the trajectory was considered in the analysis.

Statistical considerations
To ensure statistical convergence of the simulations, several measures were taken. For both cation isomers (E and Z) 10 different conformers opimized for RMSD diversity were generated using a genetic algorithm implemented in OpenBabel. 51 Sampling of anion conformations was not done due to the structural rigidity of the anion. For each of the 10 replica/systems, a different cation conformer was used to create an initial configuration. Since we were especially interested in ion-ion-interactions, we monitored the distribution of several atom-atom interionic distances to assess sampling. Every system was simulated until the distributions reached convergence, the respective times for every system are listed in Table S8. These converged distributions are depicted in Figure S4 and show that especially the N23-O48 and N23-O49 distance distributions show full overlap.
To illustrate the distribution prior to convergence, the distributions of the E/S system after 38 ns and the full simulation time of 78 ns are depicted in Figure S5. It can be clearly seen that the N23-O48 and S36 N23-O49 distance distributions are different at 38 ns, but after a simulation time of 78 ns the distributions have equalized. Figure S4. Interionic distances illustrated for the E-cation/S-anion pair (top) and the distributions after the full simulation time (bottom). Simulation times for the different systems are listed in Table S8.