Phosphine-free Chiral Iridium Catalysts for Asymmetric Catalytic Hydrogenation of Simple Ketones

Please do not adjust margins a. Department of Chemistry, School of Life Sciences, University of Sussex, Brighton BN1 9QJ, UK. E-mail: pbhukumar@gmail.com. b. Inorganic Chemistry II, University of Bayreuth, 95440-Bayreuth, Germany. Email: kempe@uni-bayreuth.de. Electronic Supplementary Information (ESI) available: [details of any supplementary information available should be included here]. See DOI: 10.1039/x0xx00000x Received 00th January 20xx, Accepted 00th January 20xx


Introduction
The catalytic reduction of polar multiple bonds-mainly carbonyl (-C]O) functionalities into corresponding hydroxyl (-CH-OH) functionalities by molecular hydrogen has great signicance in modern synthetic chemistry.The stereoselective hydrogenation of ketones to yield enantiomerically pure alcohols is a key step, in the synthesis of ne chemicals, perfumes, and pharmaceuticals.This reaction is normally performed by complexes of precious metals (e.g.Ru, Rh, Ir) using either H 2 or iPrOH as the hydrogen source. 1 In particular, the reduction of ketones with gaseous hydrogen provides an atom-economical synthetic method.High enantioselectivity, low catalyst loadings, quantitative yields, atom economy and mild conditions are attractive features of this transformation as evident in the ever growing list of research articles that use these methods.The importance of this topic was recognized when Knowles 2,3 and Noyori 3,4 were awarded the 2001 Nobel prize for their instrumental contributions to this eld.The development of highly enantioselective chemocatalyst based on their low toxicity and lower price ligands could drastically increase the importance of asymmetric homogeneous catalysis. 5Therefore, it has been well established that organometallic complexes of ruthenium have been wellemployed in this scenario, recently in particular by the groups of Noyori 6 and Braunstein. 7Iridium and rhodium complexes have long been known to be highly effective homogeneous hydrogenation catalysts, with RhCl(PPh 3 ) 3 (Wilkinson's catalyst) 8 and the most widely used [Ir(cod)(py)(PCy 3 )]PF 6 [py ¼ pyridine, cod ¼ 1,5-cyclooctadiene, Crabtree's catalyst 9 ].In contrast, iridium phosphorus-free catalysts are extensively less efficient and/or enantioselective in this asymmetric hydrogenation. 10Relatively, few organometallic complexes derived from amine based ligands have been reported to be effective at the control of asymmetric hydrogenation reactions, 10,11 particularly in comparison to the large numbers of diphosphine 12, 13 and mixed amine/phosphine 14,15 ligands, which have been reported.In principle, amine-based ligands possess a potential advantage over phosphorus because they are relatively simple to prepare and less prone to decomposition and oxidation reactions.Furthermore, iridium(I) complexes of cycloocta-1,5-diene (cod) are of interest because the coordinated cod can be replaced by other ligands.Cod can also be hydrogenated to provide vacant coordination sites around iridium and thus increase the catalytic activities of the complexes. 16In this paper, we report the synthesis and structure of phosphorus-free iridium catalysts containing aminopyridinato and pyridylalkylamine (Scheme 1) and cod ligands for the hydrogenation of simple ketones.
General procedure.A 25 mL medium-pressure tube charged with 2-bromopyridine (1.00 mmol), or 6-methyl-2-bromopyridine (1.00 mmol) X, substituted amino alcohol (1.00 mmol) Y, N,N-diisopropyl ethylamine (1.16 mmol) and a stir bar was ushed with nitrogen several times, stoppered and heated at 160 AE 5 C for 2 days.The tube was allowed to cool to room temperature, the contents was diluted with a small amount of CH 2 Cl 2 and chromatographed (hexanes-EtOAc 2 : 1 / 1 : 1 / neat EtOAc) to afford the product (a, b) in good yield.

Preparation of pyridylalkylamine ligands
General procedure.To ice-cooled solution of amino alcohol (1.00 mmol) (Y), pyridine-2-carbaldehyde or 6-methylpyridine-2carbaldehyde (1.00 mmol) (X 0 ) was added slowly in 20 mL of methanol.The resulting solution was stirred at room temperature for 3 h and subsequently treated with sodiumtetrahydridoborate (2.5 mmol) in small portions.Aer additional 2 h, water (50 mL) was added and the reaction mixture was concentrated to about 10 mL using rotovap.The remaining solution was extracted with dichloromethane (2 Â 20 mL), and the organic layer was separated and subsequently dried with sodium sulphate and concentrated in vacuum to afford the product (c-h) in good to high yield.
2.2.4.General procedure for the asymmetric hydrogenation.All catalytic hydrogenation experiments were carried out in stainless steel autoclaves (Parr instrument N-MT5 300 mL) and in a mmol scale.All experiments were carried out in a glove box under exclusion of oxygen and moisture.Stock solutions of the pre-catalysts were prepared in THF via alcohol elimination reaction of the ligand (stock solution in THF) and 0.5 equiv. of [IrOCH 3 (cod)] 2 .The solutions were prepared and stored in a glove box.A high pressure steel autoclave (Parr Instruments; 300 mL, 200 bar, 350 C) with an aluminium insert for multiple reaction tubes (5 or 20) was taken into a glove box.Then the reaction tube (placed in a 20-or 5-well insert for the autoclave, equipped with a magnetic stir bar) was loaded with additive (base if required), the pre-catalyst solution (e.g. 100 mL ¼ 0.05 mol%) and 350 mL (2.29 mmol) of the substrate solution.Then the autoclave was sealed and taken out of the glove box.The autoclave was attached to a high-pressure hydrogen line and purged with H 2 .The autoclave was sealed under the appropriate H 2 pressure and the mixture was stirred for e.g.48 h at the appropriate pressure at room temperature or at the appropriate temperature (external heating mantle).In order to stop the hydrogenation reaction, the pressure was released and water and dodecane (standard for GC) were added to the reaction.
2.2.5.Single crystal X-ray structure determinations.Crystals suitable for single crystal X-ray diffraction analyses for 1a, 1b, 2c, 2d and 2e contained toluene as a solvate.Preliminary data on the space group and unit cell dimensions as well as intensity data were collected by using a STOE-IPDS II equipped with an Oxford Cryostream low temperature unit using graphite monochromatized Mo-Ka radiation.Structure solution and renement were accomplished using SIR97, 18 SHELXL-97 (ref.19) and WinGX. 20The non-hydrogen atoms were rened with anisotropic thermal parameters.Hydrogen atoms were geometrically xed and allowed to rene using riding model.
CCDC deposition numbers 1437480-1437485 contain supplementary crystallographic data in CIF format for this paper.†

Ligand synthesis
The targeted ligands were obtained from commercially available reactants in one or two steps based on previous studies. 21he ligand precursors 4-methyl-2-(6-methyl-pyridin-2-yl-amino)pentan-1-ol (a) and 4-methyl-2-(pyridin-2-yl-amino)-pentan-1-ol (b) were synthesized in a one-pot reaction starting from 6methyl 2-bromopyridines and 2-bromopyridines with enantiopure amino alcohols using N,N-diisopropylethyl amine as a base at 165 C for 3 days (Scheme 2) in quantitative yields. 22Ligands (c-h) were prepared by reductive amination of heteroaromatic aldehydes (pyridine-2-carbaldehyde or 6-methylpyridine-2carbaldehyde) with enantiopure aminoalcohols over two steps including the formation of the corresponding ketimine and further reduction with sodium borohydride (Scheme 3).Both methodologies allowed us to prepare a family of substituted ligands based on a pyridylalkylamine and aminopyridinato core.All the synthesized ligands are air stable and soluble in polar as well as non-polar solvents and characterized by 1 H, 13 C NMR and elemental analysis.

Synthesis and structure of the complexes
The dinuclear iridium complexes 1a, b and 2c, d can be obtained via alcohol elimination route with [IrOCH 3 (cod)] 2 (ref. 23) (Scheme 2 and 3).Addition of 0.5 equiv. of the metal precursor to a solution of a-d in THF gives rise to 1a, b and 2c, d which is accompanied by a color change to yellow/reddishbrown.While under the same reaction condition mononuclear iridium complexes 2e-h can be synthesized by reaction with [IrOCH 3 (cod)] 2 (Scheme 3).This synthetic approach shows the presence of the methyl group, at sixth position of pyridine ring, and the methylene (-CH 2 NH) group, favour the formation of mononuclear iridium complex.
X-ray crystal structure analyses of the dinuclear complexes (1a, b and 2c, d) and the mononuclear (2e) were performed to determine the molecular structures (Fig. 1-5).Selected crystallographic data and geometrical parameters are summarized in the Fig. 1-5, respectively and details about the data collection, solution, and renement are summarized in the ESI (Table S1 †).In complexes 1a and 1b (Fig. 1 and 2) the Ir1-atoms are stabilized by the formation of a ve membered chelate (Ir1, N1, C9, C10, O1) and Ir2-atom is stabilized by the formation of a four membered chelate (Ir2, N1, N2, C15).It was observed that Ir-N amido bond  3 and 4).

Catalytic studies
Based on the pioneering work of Kempe et al. group 24 where amido iridium complexes were employed as a catalyst for the asymmetric hydrogenation of the simple ketones, we have chosen to determine the catalytic potential of iridium-N^N^O complexes for such reactions.In general, the screening reactions were performed using 2.29 mmol of the substrate at 20 C for 48 h and the catalyst was prepared in situ from stock solutions of [IrOCH 3 (cod)] 2 and N^N^O ligand.
First of all, mononuclear iridium complexes 2e-h (Scheme 3) were used to determine the effect of substitution at the amino skeleton of the ligands on the asymmetric hydrogenation of simple ketones (Table 3).Evidently, the catalysts 2e and 2f containing benzyl and isopropyl substituents on the nitrogen (Table 1, entries 2 and 4) gave better results than those containing the methyl (2g) or phenyl (2h) substituent (Table 1, entries 6 and 8).The best catalyst for this reaction seems to be compound 2e (Table 1, entry 2), which achieved a >99% conversion and 89% ee at a very low catalyst loading (0.05 mol% iridium) and excess KO t Bu.
In this catalytic work, all optimization studies were carried out with a-methyl propiophenone as model substrates.As can be seen from Table 2, KO t Bu appears to be the most suitable base, because complete conversion and a good ee (89%) could only be achieved by using this base (Table 2, entry 6).This result shows that KO t Bu is needed as an additive for better enantioselectivity during hydrogenation.Presence of base KO t Bu is essential for catalysis, because base itself act as a catalyst for the hydrogenation of simple ketones under drastic conditions, 25 its potassium cations accelerate the reaction rate of phosphaneruthenium-diamine complexes. 26er these optimisations, we examined the addition of excess amounts of base was needed to allow complete conversion or whether catalytic quantities of base are sufficient.Therefore, the inuence of the base/catalyst ratio was investigated (Table 3, entries 1-5).The results shown in Table 3, entries 1-5, suggest that it is necessary to use a base/catalyst ratio of 500 : 1, because only in this case (Table 3, entry 5) was it possible to obtain complete conversion and a better ee (96%) within 48 h in the presence of catalytic amount of acetone (acetone : a-methylpropiophenone 2 : 1).
However, at a base loading of only 5 mol%, it was possible to achieve a conversion (>99%) and ee (90%) (Table 3, entry 4), which contradicts the aforementioned excess requirement for base.For this reason, we investigated whether it is possible to bring the reaction to complete conversion or better ee by the use of 25 mol% of KO t Bu.To this end, the reaction time was 48 h an ee of 96% were obtained.
The nal screening was performed on the catalyst loading to nd the minimum catalyst loading necessary to achieve full conversion and good ee (Fig. 6).As shown in Fig. 6, it was sufficient to use a catalyst loading of 0.05 mol% to obtain a very good ee (96%) for this reaction (Fig. 6).It was observed the addition of KO t Bu and acetone leads to quantitative conversion with 2e as a pre-catalyst with catalyst loadings of 1, 0.5, 0.2, 0.1, and 0.05 mol% (Fig. 6).It was examined that without KO t Bu and acetone with 1.0 mol% of 2e a conversion of 55% and 10% ee is obtained (Fig. 6, Table 1, and entry 1).The catalytic performance for hydrogenation of a-methylpropiophenone of 2e-h with KO t Bu and acetone was examined (Table 4).Upon addition of acetone in the presence of KO t Bu, 96% ee could be achieved under same reaction condition with 0.05 mol% catalyst loading (Table 4, entry 2).The increase in enantioselectivity could result from the fact that the more enantioselective catalyst species is formed during catalysis, the addition of (non-prochiral) ketones, which is hydrogenated in parallel and preferentially faster, should support the formation of a more enantioselective catalyst.As it can be seen in Table 4, catalyst 2e and 2f are active catalyst for asymmetric hydrogenation of a-methylpropiophenone.Further, we were interested to examine the effect of acetone on the reaction.It was observed that acetone/ pre-catalyst 2e ratio of 2 : 1 was use to obtain complete conversion (>99%) and good ee (96%).It was also observed that if we changed the acetone/catalyst ratio (1 : 1 and 3 : 1), considerably affect on the conversion (86% and 90%) and ee (80% and 86%).
To conrm the results we achieved for the hydrogenation amethylpropiophenone with catalyst 2e, different simple aromatic ketones were further examined (Fig. 7).These results show that the reduction of the hindered aromatic ketones proceeded with excellent enantioselectivity, although the expected trend of a decrease in conversion and enantioselectivity for the more bulky t-butyl substituent was also observed (Fig. 7).Also with the increase of the a-carbon chain (-CH 2 CH 2 CH 3 ) (Fig. 7, VIII), the selectivity and activity of the reaction drops signicantly.This result illustrates the importance of bulky groups next to the carbonyl for obtaining high enantioselectivity.The aromatic ketones (Fig. 7, VII) with a strong selectron withdrawing and weak p-electron donating chloro group and with the p-electron donating methoxy and methyl group in the para position showed a higher selectivity and lower activity compared to that with propiophenone.These results indicate that the para substituted ketones have major effect on the catalytic hydrogenation of ketones.
In contrast, the dinuclear iridium complexes 1a, b and 2c, d show lower catalytic activity and selectivity for the hydrogenation of simple ketones.For example asymmetric hydrogenation of a-methylpropiophenone catalyzed by 1a and 2c (0.05 mol%) in the presence of excess KO t Bu and acetone under same reaction condition (see Table S2 †) gave corresponding alcohol in 40% and 35% yield and 24% and 8% ee.While 1b and 2d gave 23% and 16% yield and 11% and 2.0% ee.Needless to say, no reaction occurred in the absence of KO t Bu.This may suggest that dinuclear iridium complexes act as catalyst precursor to promote enantioselective asymmetric hydrogenation under the inuence of base and acetone.

Conclusions
In conclusion the reported pyridylalkylamine mononuclear iridium complexes 2e-h represents a novel class of efficient and easily accessible catalysts for the asymmetric hydrogenation of simple ketones.Due to the modular ligand design, broad substitution patterns can be realized.The moderate to high efficiency and good selectivity combined with the novel structural motif opens up new prospects for the enantioselective hydrogenation of ketones.

Fig. 6
Fig. 6 Hydrogenation of a-methylpropiophenone with 2e in the presence of KO t Bu and acetone.