Asymmetric ruthenium tricarbonyl cyclopentadienone complexes; synthesis and application to asymmetric hydrogenation of ketones

A series of enantiomerically-pure ruthenium tricarbonyl cyclopentadienone complexes were prepared via the cyclisation of C2-symmetric dialkynes with Ru 3 (CO) 12 . Four complexes were characterised by X-ray crystallography and a hydride derivative of one of these was characterised. The complexes were tested in the asymmetric hydrogenation and transfer hydrogenation of acetophenone. Whilst the catalysts were active, acetophenone was reduced in low ee, however a reduction product of up to 46% ee could be obtained when a mild base ( i Pr 2 NEt or pyridine) was added to the reaction.


Introduction
Metal complexes of cyclopentadienones, notably ruthenium [1] and iron [2] derivatives of general structure 1 and 2, have been applied as catalysts in hydrogen transfer reactions [3], operating via their metal hydride derivatives [4]. Asymmetric versions of the ligands have also been reported to be effective however derivatives which deliver products in high ee remain elusive. In recent work, we [5] and others [6], have reported asymmetric iron cyclopentadienone complexes 3-6 and have found that these will reduce ketones in up to 77% ee [6c]. Although iron complexes are desirable due to their improved environmental properties compared to precious metal group catalysts, ruthenium derivatives are generally more active and stable under catalytic conditions, and can be used at lower loadings [5a]. In addition, very few examples of asymmetric ruthenium cyclopentadienone complexes have been reported, with field seemingly restricted to examples 7 [5a], 8 [7] and 9 [8], all of which rely on planar chirality to induce asymmetry in the reactions which they catalyse. In our studies [5a], complexes of type 7 achieved asymmetric transfer hydrogenation of acetophenone in no higher than 17% ee, whereas complex 8 reduced acetophenone in up to 21% ee under a 35 atm pressure of hydrogen gas [7]. Complex 9 catalysed the transfer hydrogenation of 1,1,1-trifluoroacetophenone in up to 56% ee and an imine in up to 64% ee [8]. During their catalytic cycles, the iron and ruthenium complex form active hydride species of general structures 10 (monomer) and 11 (dimer) respectively. In order to provide a comparison between iron and ruthenium systems, we have now prepared a series of ruthenium analogues of our earlier iron complexes 6 and have tested them as catalysts for ketone reduction (Fig. 1).

Results and discussion
The synthesis of four asymmetric ruthenium tricarbonyl cyclopentadienone complexes 13a-13d was achieved through intramolecular cyclisation of dialkoxy dialkynes precursors using Ru 3 (CO) 12 following an established procedure. Whilst dialkynes 12a-12c are known, compound 12d, containing methoxy substituents, is novel and was prepared via methylation of the known diol 12a. The isolation of the side product 14 from the synthesis of 13a, where a shorter reaction time was used (2 days instead of 5 days), indicates that this is likely intermediate in its formation (Scheme 1).
The reactions worked efficiently and delivered products in good yields and in crystalline form, allowing full characterisation by spectroscopic methods. X-ray crystallographic structures were obtained of all the ruthenium cyclopentadienone complexes and of the diruthenium complex 14 and these are shown in Fig. 2 (Full details in the Supporting Information). In all cases the anticipated ruthenium tricarbonyl structure was present, with the phenyl groups flanking the central C]O in a twisted orientation, presumably influenced by the adjacent alkoxy groups. The structures were similar to those reported for the iron analogues [5b,c]. An X-ray crystallographic structure of the diruthenium complex 14 was also obtained, which confirmed its interesting symmetric structure in which each ruthenium was bonded to three carbon monoxide ligands (Fig. 2 b).
A hydride complex 15 was formed from 13b upon treatment with the mild base NaHCO 3 , and this is likely to be an intermediate in the asymmetric reductions (Fig. 3). This was sufficiently stable to be isolated and was characterised by NMR spectroscopy.
Each of the new complexes 13a-13d was tested as a catalyst for asymmetric hydrogenation of acetophenone (Table 1) using the conditions we had previously reported, and a range of representative results are reported in Table 1. A detailed Table featuring the results of all our tests is reported in the Supporting Information. Using 13a as the catalyst, the addition of TMAO as an activator was required for full conversion within 18 h at 80°C, although good conversion was obtained without the activator. The product was essentially racemic however. The use of 5% K 2 CO 3 in place of TMAO gave lower conversion, and the ee remained low. A similar pattern of results was obtained using catalyst 13b, which was slightly less reactive than 13a. Pyridine N-oxide was also shown to be an effective initiator for this catalyst, however TMAO was retained for future tests as it is easier to remove from the reaction mixture after use. Catalysts 13c and 13d also gave full conversions under our standard conditions, however catalyst 13d gave full conversion to product both in the absence of TMAO and if an excess of this was used, or if 5% K 2 CO 3 was used. However the different reaction conditions had no effect on the ee of the product.
The effect of the addition of a range of additives to the reductions was studied, using the benzyloxy catalyst 13b throughout for consistency (Table 2). Triphenylphosphine reduced the overall conversion and had little effect on the product ee. 2-Picoline, lutidine, and i Pr 2 NEt additives increased the product ees, with a product of 45% ee being obtained with 10 mol% of i Pr 2 NEt, which is one of the highest recorded asymmetric inductions we have achieved with this class of catalyst. However full conversions were not achieved, and more than 10 mol% of the amine reduced the conversions notably. The addition of triethylamine at up to 10% did not reduce the conversion but the ee dropped to 0%, whilst dimethyl aniline had little effect on the conversion or ee. The addition of 2,2′-dipyridine as an additive was also investigated but gave product of just up to 4.8% ee, although with 97% conversion (see supporting information).
The addition of pyridine as an additive was investigated in more detail with catalysts 13b (Table 3), 13c (Table 4) and 13d (Table 5). Using catalyst 13b ( Table 3) the effect of the relative amounts of TMAO and pyridine was studied, with a steady increase in ee observed as both were increased, with > 99% conversion maintained, and an ee of 38.2% delivered using 5 mol%. The ee increased further (up to 46%) when 10 mol% of each additive was added however the conversion dropped. The reaction was carried out at 40°C however both the conversion and ee decreased. A similar pattern of result was observed for catalyst 13c, with the use of 50% of both TMAO and pyridine reducing the reaction efficiency (Table 4). The effect of the addition of pyridine to catalyst 13a was also investigated; although the ee was increased to 10.8% using 10% TMAO and 10% pyridine, the yield was slightly reduced (see Supporting information).
In the case of catalyst 13d, a similar pattern of results was observed, with slightly improved product ees observed using 5 mol% TMAO and

Scheme 1.
Route to Ru complexes 13a-13d prepared in this project, and diruthenium complex 14 also formed during preparation of 13a. A. Del Grosso, et al. Inorganica Chimica Acta 496 (2019) 119043 pyridine ( Table 5). The opportunity was taken to examine the effect of alternative solvents, with several found to be compatible with the conditions. Using DMSO, the configuration changed sharply to 33.2% (S), although the conversion was low.

Conclusions
In conclusion, we have prepared a series of asymmetric ruthenium tricarbonyl cyclopentadienone complexes using an efficient intramolecular cyclisation reaction with Ru 3 (CO) 12 . These are stable complexes which are readily isolated and purified, and all were characterised by NMR, HRMS and X-ray crystallography. All of the complexes proved to be highly effective in the asymmetric hydrogenation of acetophenone, as a prototype substrate which allows comparisons to be made with other catalysts. Full conversion to the alcohol product was possible in most cases, and provided that TMAO activator was added, although the product ees were low. A study of additives revealed that the addition of pyridine could raise the product ee to 38.2% whilst maintaining full conversion. The use of 10 mol% i Pr 2 NEt as an additive raised the ee to 46% without a reduction in conversion, which is one of the highest ees we have obtained for acetophenone reduction using this class of ruthenium-based catalysts. The basic additives may be forming modified complexes or influencing the reaction in other ways, and this   A. Del Grosso, et al. Inorganica Chimica Acta 496 (2019) 119043 is currently under investigation.

General experimental
General; Solvents and reagents for the synthesis of complexes and catalytic reactions were degassed prior to use and all reactions were carried out under either a nitrogen or argon atmosphere. All heated experiments were conducted using thermostatically controlled oil baths. Reactions were monitored by TLC using aluminum backed silica gel 60 (F254) plates, visualized using UV 254 nm and phosphomolybdic acid (PMA), potassium permanganate or vanillin dips as appropriate. Flash column chromatography was carried out routinely using 60 µm silica gel. Reagents were used as received from commercial sources unless otherwise stated. 1 H NMR spectra were recorded on a Bruker DPX (300, 400 or 500 MHz) spectrometer. Chemical shifts are reported in δ units, parts per million relative to the singlet at 7.26 ppm for chloroform and 0.00 ppm for TMS. Coupling constants (J) are measured in Hertz. IR spectra were recorded on a Perkin-Elmer Spectrum One FT-

Asymmetric hydrogenation of acetophenone
The alcohol formed by reduction of acetophenone has been reported and our procedures and characterisation followed the protocols reported [5b]. A typical asymmetric hydrogenation procedure is as follows; acetophenone (200 mg, 1.67 mmol), catalyst (1 mol%), iPrOH (0.5 mL) and H 2 O (0.2 mL) were added to a small test tube containing a stirrer bar. TMAO (1.25 mg, 0.017 mmol, 1 mol%) was added, then the test tube was sealed in a Parr hydrogenator and charged with hydrogen to 30 bar, venting once. The sealed vessel was heated to 80°C and stirred for 18 h. At the end of this time, the reaction was allowed to cool to rt, the pressure was carefully released and the sample was worked up and analyzed by chiral GC [5b]. At the end of this time the reaction was allowed to cool to rt and EtOAc:hexane (1:4, ca. 10 mL) was added to dilute the sample. This solution was passed through celite and then silica gel to remove residues of catalyst. Removal of solvent gave the product which was analyzed by GC [5b]. The absolute configuration was assigned by comparison of GC spectroscopic data previously reported for this compound [5b].

Data sharing statement
The research data (and/or materials) supporting this publication can be accessed at http://wrap.warwick.ac.uk/.

Declaration of Competing Interest
None.
Project with support from the AWM and part funded by the ERDF. 1 H and 13 C NMR spectra. X-ray data (CCDC 1923113-1923117), Examples of Chiral GC of reduction products. Supplementary data to this article can be found online at https://doi.org/10.1016/j.ica.2019. 119043.