Remarkably Diastereoselective Synthesis of a Chiral Biphenyl Diphosphine Ligand and Its Application in Asymmetric Hydrogenation

Home , BINAP

Remarkably diastereoselective synthesis of a chiral SPECIAL FEATURE biphenyl diphosphine ligand and its application in asymmetric hydrogenation

Liqin Qiu*, Jing Wu*, Shusun Chan*, Terry T.-L. Au-Yeung*, Jian-Xin Ji*, Rongwei Guo*, Cheng-Chao Pai*, Zhongyuan Zhou*, Xingshu Li*, Qing-Hua Fan†, and Albert S. C. Chan*‡

*Open Laboratory of Chirotechnology of the Institute of Molecular Technology for Drug Discovery and Synthesis and Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong; and †Center for Molecular Science, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100080, China

Edited by Jack Halpern, University of Chicago, Chicago, IL, and approved February 17, 2004 (received for review November 22, 2003) Essentially complete atropdiastereoselectivity was realized in the (42–45) with central-to-axial chirality transfer, resulting in a preparation of biaryl diphosphine dioxide by asymmetric intramo- concomitant eight-membered ring closure. The introduction of lecular Ullmann coupling and oxidative coupling with central-to- the chiral bridge restricted the conformational rotation of the axial chirality transfer. A bridged C2-symmetric biphenyl phosphine diphosphine and made it more rigid than other biaryl diphos- ligand possessing additional chiral centers on the linking unit of phine ligands (5–15, 46). The diphosphine ligand was found to be the biphenyl groups was synthesized. No resolution step was highly effective in the asymmetric hydrogenation of ␣- and required for the preparation of the enantiomerically pure chiral ␤-ketoesters, 2-(6Ј-methoxy-2Ј-naphthyl)propenoic acid, ␤- ligand. These ﬁndings offer a general and practical tool for the alkyl-substituted ␤-(acylamino)acrylates, and enol acetates. development of previously uninvestigated atropdiastereomeric biaryl phosphine ligands. The diphosphine ligand was found to be Materials and Methods highly effective in the asymmetric hydrogenation of ␣- and ␤- General Information. Unless otherwise noted, all reactions were ketoesters, 2-(6؅-methoxy-2؅-naphthyl)propenoic acid, ␤-(acyl- carried out under an inert atmosphere of dry nitrogen and were CHEMISTRY amino)acrylates, and enol acetates. monitored by TLC. Glassware was flame-dried before use. Standard syringe techniques were applied to transfer dry sol- xially chiral biaryls are not only common structural motifs vents and reagents. The preparation of samples and the setup of Ain many natural products, but they are also the core of many high-pressure reactors were either carried out in a nitrogen- highly effective chiral ligands for asymmetric reactions. Atropi- filled, continuously purged Lab Master 230 glovebox someric, C2-symmetric diphosphine ligands have played a par- (MBRAUN, Shanghai, Republic of China) or by using standard ticularly crucial role in the development of asymmetric catalysis Schlenk-type techniques. 1H NMR, 31P NMR, and 13C NMR (1–4). Therefore, it is not surprising that considerable efforts spectra were recorded on a Varian 500 spectrometer (500, 202, have been taken for the design and synthesis of atropisomeric and 125 MHz, respectively). Chemical shifts (␦) are given in ppm ligands based on the biphenyl, binaphthyl, or other biaryl and are referenced to residual solvent peaks (1H NMR and 13C 31 backbones (5–15). Enantiomerically pure biaryls can be obtained NMR) or to an external standard (85% H3PO4, P NMR). by aryl-aryl coupling followed by a classical resolution of the High-resolution mass measurements were carried out with a VG resulting atropisomers. The disadvantage of the path by resolu- Micromass (Manchester, U.K.), Fison VG platform (VG Bio- tion is that the maximum yield of the desired atropisomer is only tech, Cheshire, U.K.), or a Finnigan model Mat 95 ST instrument 50% and the enantiomeric excesses (ee’s) are usually Ͻ100%, (Finnigan MAT, Bremen, Germany). Optical rotations were not to mention that resolution procedures are frequently tedious. recorded on a Perkin–Elmer 341 polarimeter in a 10-cm cell. From a practical standpoint, it is desirable to develop efficient Enantiomeric excesses of the asymmetric hydrogenation prod- atroposelective methodologies for the synthesis of biaryls ligands ucts were determined by chiral GC and HPLC. HPLC analyses to expand the scope of the useful catalysts. Various approaches, were performed by using a Waters model 600 with a Waters 486 including desymmetrization of prochiral biaryls (16), kinetic UV detector. Gas chromatographic analyses were performed on resolution of racemic substrates (17), asymmetric catalytic cou- an Hewlett–Packard 4890A GC with an flame ionization detec- pling (18–23), and chirality transfer from central, axial, and tor. Crystal structural data were collected by the single-crystal planar asymmetry have been reported (24–38). Oxazoline- x-ray diffraction method with a Bruker CCD Area Detector mediated asymmetric Ullmann coupling was studied by Meyers Diffractometer and PC computer with the BRUKER SMART and and coworkers (28–30) to produce diastereomerically pure BRUKER SHELXTL programs. Tetrahydrofuran (THF) and tolu- bis(oxazoline)s and corresponding enantiomerically pure biphe- ene were freshly distilled from sodium͞benzophenone ketyl, nols and binaphthols. Previously, most of the research was whereas DMSO, dimethylformamide (DMF), CH2Cl2, and Bu3N focused on the syntheses of biphenols and binaphthols; in were distilled from CaH2 under nitrogen atmosphere. MeOH contrast, less attention was paid to the atroposelective syntheses and EtOH were distilled from magnesium under nitrogen at- of biaryl diphosphine oxides, the precursors of widely used mosphere. All other chemicals were used as received from diphosphine ligands. Recently, we successfully developed two Aldrich, Acros (Geel, Belgium), or Strem (Newburyport, MA) diastereomeric diphosphine ligands for use in asymmetric hy- without further purification. All substrates used in moisture- drogenation reactions by stereoselective intermolecular Ull- mann coupling of two chiral phosphine oxide precursors with moderate atropdiastereoselectivity (39). However, very careful This paper was submitted directly (Track II) to the PNAS ofﬁce. separation of the diastereomers by column chromatography on Abbreviations: rt, room temperature; ee, enantiomeric excess; MeO-BIPHEP, 2,2Ј- silica gel was still needed. Herein, we report an example of bis(diphenylphosphino)-6,6Ј-dimethoxy-1,1Ј-biphenyl; BINAP, 2,2Ј-bis(diphenylphos- essentially complete atropdiastereoselectivity in the synthesis of phino)-1,1Ј-binaphthyl; DMF, dimethylformamide; THF, tetrahydrofuran. diphosphine dioxide by means of intramolecular Ullmann cou- ‡To whom correspondence should be addressed. E-mail: [email protected]. pling (27, 34, 40, 41) or Fe(III)-promoted oxidative coupling © 2004 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0307774101 PNAS ͉ April 20, 2004 ͉ vol. 101 ͉ no. 16 ͉ 5815–5820 Downloaded by guest on September 29, 2021 Scheme 1. Synthetic route. Path a: Cs2CO3, DMSO, 55°C, 53%. Path b: (i) n-BuLi, THF, Ϫ78°C; (ii)Ph2PCl, Ϫ78°C to rt. Path c: H2O2, acetone, 0°C, 85% from 3. Path d: (i) lithium diisopropylamide (LDA), THF, Ϫ78°C; (ii)I2, Ϫ78°C to rt, 85%. Path e: Cu, DMF, 140°C, 91%. Path f: LDA, FeCl3, Ϫ20°C to rt, 61% yield and 80% conversion. Path g: HSiCl3,Bu3N, toluene, reﬂux, 96%.

sensitive reactions were predried twice with toluene azeotrope filtrate was washed successively with saturated aqueous ammo- before use. nium chloride and brine and was dried over anhydrous Na2SO4. After the solvent was removed, the residue was purified by silica Synthesis of Compound 6. The synthetic route for the ligand was gel column chromatography to give 7 as a white solid (406 mg, 1 ␦ ϭ shown in Scheme 1. Here we report the synthetic methods for 0.634 mmol, 91% yield). H NMR (500 MHz, CDCl3): 1.26 compounds 6, 7, and 8. The synthesis of compounds 3, 4, and 5, (d, J ϭ 6.0 Hz, 6H), 3.65–3.71 (m, 2H), 6.85 (d, J ϭ 7.5 Hz, 2H), the spectra and data for all new compounds, and the general 6.93–6.97 (m, 2H), 7.08–7.13 (m, 6H), 7.21–7.24 (m, 2H), procedures for the asymmetric hydrogenation are provided in 7.27–7.34 (m, 8H), 7.37–7.40 (m, 2H) and 7.63–7.67 (m, 4H). 31P ␦ ϭ 13 Supporting Text, which is published as supporting information on NMR (202 MHz, CDCl3): 28.46. C NMR (125 MHz, ␦ ϭ the PNAS web site. CDCl3): 18.97, 86.33, 123.80, 123.82, 127.16, 127.26, 127.66, A solution of lithium diisopropylamide (1.7 ml, 2.0 M) was 127.76, 127.79, 127.87, 129.34, 129.46, 130.20, 130.22, 130.74, added dropwise to compound 5 (1.00 g, 1.56 mmol) in THF (40 130.76, 131.97, 132.04, 132.33, 132.40, 132.92, 133.65, 133.76, ml) at Ϫ78°C for 0.5 h. After stirring for an additional 3 h, the 134.49, 134.72, 135.56, 158.97, and 159.08. HRMS (ESI): calcd ϩ ϩ ␣ 20 ϭ reaction mixture was cannulated into a flask containing I2 (1.584 for C40H35P2O4 [M H] 641.2011, found 641.1991. [ ]D Ϫ Ϫ ϭ g, 6.24 mmol) and 40 ml of THF at 78°C over 30 min. The 197.4 ° (c 1, CHCl3). HPLC conditions for the analysis of de: mixture was warmed to ambient temperature for 2 h, and the AD column, eluent ϭ 80:20 hexane͞i-PrOH, ␭ ϭ 254 nm, flow ϭ ͞ ϭ reaction was continued overnight. After the evaporation of rate 1.0 ml min); t1 13.2 min for 7. The other diastereomer, ϭ the solvent with a rotary evaporator, the residue was dissolved which has a retention time t2 18.8 min under otherwise in CH2Cl2 (50 ml). The resulting solution was washed succes- identical conditions, was not detected. sively with saturated aqueous ammonium chloride solution, Synthetic route II: Asymmetric oxidative coupling. A solution of lithium water, and saturated sodium thiosulfate solution, followed by diisopropylamide (0.68 ml, 2.0 M) was added dropwise to a drying over anhydrous Na2SO4, and concentration in vacuo gave solution of compound 5 (0.400 g, 0.622 mmol) in dried THF (6 a crude product that contained diiodide 6 (88.5%), monoiodide ml) in 4 ml of THF at Ϫ20°C for 30 min. After stirring for an Ϫ (9.3%), and compound 5 (2.2%) based on NMR spectral anal- additional 1.5 h at 20°C, 0.29 g (1.79 mmol) of FeCl3 (anhy- ysis. Purification by silica gel column chromatography afforded drous) was added in one portion to the dark-brown solution, pure compound 6 (1.185 g, 1.33 mmol, 85% yield). 1H NMR (500 while allowing the temperature to rise to room temperature. The ␦ ϭ ϭ MHz, CDCl3): 1.45 (d, J 6.0 Hz, 6H), 4.67–4.73 (m, 2H), solvent of the reaction mixture was evaporated. The dark, oily ϭ 6.68–6.73 (m, 2H), 7.05 (d, J 8.0 Hz, 2H), 7.19–7.23 (m, 2H), residue was taken up in 50 ml of CH2Cl2, and washed with 7.43–7.49 (m, 8H), 7.52–7.57 (m, 4H), and 7.65–7.71 (m, 8H). 31P ammonium hydroxide solution (25%, 1 ml) at 0°C. After stirring ␦ ϭ 13 NMR (202 MHz, CDCl3): 35.09. C NMR (125 MHz, for 30 min the iron salts were filtered off and rinsed with 50 ml ␦ ϭ CDCl3): 14.70, 77.49, 94.47, 94.53, 117.21, 117.23, 128.49, of CH2Cl2. The filtrate was dried over anhydrous Na2SO4, 128.59, 128.77, 128.88, 128.97, 131.13, 131.15, 131.83, 131.85, filtered, and evaporated. The residue was purified by silica gel 131.98, 132.00, 132.13, 132.16, 132.20, 132.24, 137.37, 138.21, column chromatography to give 7 as a white solid (241 mg, 0.376 ϩ 157.29, and 157.39. HRMS (ESI): calcd for C40H35I2P2O4 [M mmol, 61% yield) and starting material 5 (56 mg, 0.087 mmol, ϩ ϩ H] 895.0101, found 895.0002. calcd for C40H34I2P2O4Na [M 86.0% conversion). Analytical data corresponded with those ϩ ␣ 20 ϭϩ ϭ Na] 916.9920, found 916.9763. [ ]D 54.8 (c 1, CHCl3). obtained from the foregoing experiment.

Synthesis of Compound 7. Synthetic route I: Asymmetric Ullmann cou- Synthesis of Compound 8. A 100-ml, two-necked flask equipped pling. DMF (5 ml) was added into a flask containing Cu powder with a magnetic stirring bar and a reflux condenser was charged (100 mesh, 0.352 g, 5.54 mmol) and 6 (0.620 g, 0.693 mmol). The with 7 (480 mg, 0.749 mmol). Dry and degassed toluene (10 ml), resulting mixture was stirred at 140°C for 12 h under a nitrogen tributylamine (3.7 ml, 15.5 mmol), and trichlorosilane (1.6 ml, atmosphere. After the removal of the DMF solvent under 15.5 mmol) were added to the flask by use of a syringe. The reduced pressure, the residue was boiled for 5 min with hot mixture was stirred at reflux temperature overnight. After the ϫ CHCl3 (10 ml 3). The insoluble solid was removed by filtration solution was cooled to 0°C, a 30% aqueous sodium hydroxide ϫ and was washed with hot CHCl3 (5 ml 3). The combined solution (23 ml) was carefully added. The mixture was then

5816 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0307774101 Qiu et al. Downloaded by guest on September 29, 2021 chiral tether bridging over two aryl groups that are to be coupled

at a later stage. Thus, chiral bis(bromoether) 3 was prepared by SPECIAL FEATURE treatment of (2R,3R)-butanediol dimesylate 2 with bromophenol 1. Lithiation of 3 with n-butyllithium followed by the introduction of the phosphinyl moiety with chlorodiphenylphosphine gave compound 4, which was oxidized with H2O2 to give phosphine oxide 5. Subsequent ortho-lithiation͞iodination with lithium diisopropylamide and iodine gave diiodide 6 (7). An intramolecular Ullmann coupling of 6 led to the final ring closure, affording the diphosphine dioxide 7 in high yield. Compound 7 could also be obtained by a one-step oxidative coupling by using lithium diisopropylamide and anhydrous FeCl3 in 61% yield (11, 12). The key feature in these two coupling reactions was a substrate-directed asymmetric reaction. The center chirality was translated into axial chirality with high efficiency from which only one atropdiastereomer 7 was obtained, and the synthesis successfully avoided the resolution step Fig. 1. ORTEP drawing of compound (Rax)-7. for enantiomers or a separation process for diastereomers. The other diastereomer was not found in the crude products and the stirred at 60°C until the organic and aqueous layers became clear. chromatographed product by NMR and HPLC analyses. (An The organic product was extracted with toluene (10 ml ϫ 3) and authentic sample of the other diastereomer has been prepared the combined extracts were washed successively with water and by using a different method, which is beyond the scope of this brine and dried over anhydrous Na2SO4. The filtered organic article.) The high yield in biphenyl coupling and the complete solution was concentrated under reduced pressure to give a atropdiastereoselectivity may have resulted from the significant light-yellow solid. This residue was washed with hexane to give directing effect of the bridging moiety in 6 in comparison with 8 as a pure white powdery product (439 mg, 0.72 mmol, 96% the intermolecular coupling reaction (39). This methodology is 1 ␦ ϭ ϭ yield). H NMR (500 MHz, CDCl3): 1.28 (d, J 6.0 Hz, 6H), ϭ ϭ obviously useful for the development of enantiomerically pure CHEMISTRY 3.71–3.78 (m, 2H), 6.80 (d, J 7.5 Hz, 2H), 6.89 (d, J 8.0 Hz, phosphine ligands. To the best of our knowledge, this example 2H) 7.01–7.04 (m, 4H), 7.13–7.19 (m, 8H), 7.35–7.38 (m, 6H), 31 ␦ ϭϪ of such extremely high atropdiastereoselectivity in the prepara- and 7.58–7.61 (m, 4H). P NMR (202 MHz, CDCl3): 8.42. tion of atropisomeric biaryl diphosphine dioxide by means of 13C NMR (125 MHz, CDCl ): ␦ ϭ 19.09, 86.33,122.20, 127.74, 3 Ullmann coupling or oxidative coupling reactions has been 127.76, 127.78, 128.24, 128.26, 128.96, 129.50, 133.21, 133.29, 133.38, 133.87, 133.96, 134.05, 135.08, 135.20, 135.32, 137.39, previously uninvestigated. The molecular structure of 7 in 137.42, 137.46, 138.33, 138.40, 138.46, 138.57, 138.60, 138.63, (Rax)-form was confirmed by single-crystal x-ray diffraction (Fig. 1). Reduction of compound (R )-7 with trichlorosilane and 159.69, 159.72, and 159.75. HRMS (ESI): calcd for C40H35P2O2 ax ϩ ϩ ␣ 20 ϭϪ ϭ tributylamine gave diphosphine 8. Treatment of the latter in [M H] 609.2113, found 609.2089. [ ]D 341.6° (c 1, toluene). refluxing toluene for 24 h, followed by reoxidation with H2O2, showed that 8 had a high stereochemical stability, because no Results and Discussion epimerization was detected by NMR and HPLC analyses. Synthesis of Ligand 8. The synthetic pathway is outlined in Scheme One key factor in the ortho-lithiation͞iodination process needs 1. The salient feature of this route is the early introduction of a to be mentioned further. The sequence of addition has a great

Table 1. Asymmetric hydrogenation of ␤-ketoesters

ee,* % (conﬁguration)

Entry R1 R2 R3 8 (S)-MeO-BIPHEP (S)-BINAP

1* Me H Me 99.3 (R) 97.7 (S) 97.7 (S) 2* Me H Et 99.4 (R) 98.0 (S) 98.0 (S) 3* Me H Bn 99.3 (R) 98.1 (S) 96.1 (S)

4* ClCH2 H Et 97.2 (S) 95.5 (R) 94.2 (R) 5† Ph H Et 95.6 (S) 93.0 (R) 89.3 (R) 6† Ph Cl Et 92.7 (22.0% anti) 70.8 (16.0% anti) 16.3 (10.4% anti) (2R,3R)(2S,3S)(2S,3S) 26.2 (78.0% syn) 5.3 (84.0% syn) 9.8 (89.6% syn) (2S,3R)(2R,3S)(2R,3S)

Reaction conditions: solvents ϭ 12.5 ␮lofCH2Cl2 plus 987.5 ␮l of MeOH or EtOH; reaction time ϭ 24 h, except for entry 6 (100 h); substrate͞[Ru] ϭ 667:1 (mol͞mol), except for entry 6 (substrate͞[Ru] ϭ 100:1); substrate concentration ϭ 0.5 M; T ϭ 70°C, except for entry 6 [room temperature (rt)] and entry 4 (optimum temperature, 70–90°C); P ϭ 50 psi H2, except for entry 6 (1,000 psi H2); complete conversions were obtained in all cases. *The ee values were determined by chiral GC with a Varian CP Chirasil-DEX CB column (25 m ϫ 0.25 mm) after converting the products to the corresponding acetyl derivatives. †The ee values were determined by chiral HPLC with a Daicel Chiralcel OD column; ratio of anti-tosyn-products was determined by 1H NMR.

Qiu et al. PNAS ͉ April 20, 2004 ͉ vol. 101 ͉ no. 16 ͉ 5817 Downloaded by guest on September 29, 2021 6 Table 2. Asymmetric hydrogenation of ␣-ketoester with [RuL(␩ -C6H6)Cl]Cl

Ligand Substrate͞catalyst T, °C PH2, atm Time, h ee,* % (conﬁguration)

8 600 60 35 15 93.5 (R) 8 600 rt 75 15 95.7 (R) 8 600 rt 40 15 97.0 (R) (S)-BINAP 560 30 100 94 79.0 (S)†

*The ee values were determined by chiral GC with a Lipodex-A (50 m ϫ 0.25 mm); all conversions were Ͼ99%. †Data from ref. 48.

influence on the yield of compound 6. The addition of an iodine 15 h at 40 atm and 30°C, giving the product methyl mandelate solution in THF to lithiated 5 at Ϫ78°C gave 6 in only 30–40% with 97.0% ee. In contrast, 94 h was required for the reaction to yield along with a complex mixture. However, the reverse complete at 100-atm hydrogenation pressure, and 79.0% ee was addition of lithiated 5 in THF to the iodine solution at Ϫ78°C obtained by using BINAP as ligand (Table 2). dramatically improved the chemical yield up to 85%. Ligand 8 was efficient not only for carbon–oxygen double- bond hydrogenation but also for carbon–carbon double bond Applications of the Ligand. Chiral ␤- and ␣-hydroxy esters are (prochiral olefins) hydrogenation. The results of the asymmetric useful building blocks for the synthesis of biologically active hydrogenation of 2-(6Ј-methoxy-2Ј-naphthyl)propenoic acid compounds and natural products. For this reason we first chose (48) are summarized in Table 3. Lower reaction temperature or ␤- and ␣-ketoesters as model substrates to examine the capability higher hydrogenation pressure was beneficial to obtain high of enantiomeric differentiation for ligand 8 and the correspond- enantioselectivity. The best result was obtained in up to 96.0% ␩6 ing catalysts Ru8Cl2(DMF)n (47) and [Ru8( -C6H6)Cl]Cl (48). ee at 0°C and 1,500 psi H2 with [Ru8(p-cymene)Cl]Cl as catalyst. The results of the asymmetric hydrogenation for ␤-ketoesters This result compared favorably with those obtained with MeO- (summarized in Table 1) compared favorably with those ob- BIPHEP or BINAP ligand under otherwise identical conditions. tained by using benchmark 2,2Ј-bis(diphenylphosphino)-6,6Ј- Enantiomerically pure ␤-amino acid derivatives are important dimethoxy-1,1Ј-biphenyl (MeO-BIPHEP) or 2,2Ј-bis(diphe- building blocks in the synthesis of ␤-peptides and ␤-lactams and nylphosphino)-1,1Ј-binaphthyl (BINAP) ligands under some natural products (49). One of the most facile routes to otherwise identical conditions. As compared with analogous ␤-amino acids involves the asymmetric hydrogenation of the C2TunaPhos (46) (Table 1, entries 1 and 5, 99.3% and 95.6% ee corresponding dehydroamino acid precursors. Although a wide vs. 90.9% ee and 76.8% ee, respectively), the added stereogenic range of optically active natural and unnatural ␣-amino acids centers in ligand 8 obviously had beneficial effects in transmit- have been synthesized in excellent ees by catalytic asymmetric ting chiral information. The dynamic kinetic resolution of ␣-sub- hydrogenation, the asymmetric hydrogenation of ␤-(acylamino)- stituted ␤-ketoester ethyl-2-chlorobenzoyl acetate by hydroge- acrylates is relatively less developed. Good-to-excellent ee’sof nation was also investigated. The levels of enantioselection were ␤-alkyl-substituted ␤-amino acids have been obtained by using much higher when using 8 as ligand for both the anti- and some chiral phosphane–rhodium complexes (such as Rh com- syn-products in comparison with those using MeO-BIPHEP or plexes of BICP, DuPhos, MiniPhos, BDPMI, TangPhos, and BINAP (entry 6). MonoPhos; ref. 50). In contrast, the application of ruthenium The preliminary study also indicated that catalyst [Ru8(␩6- complexes as catalysts in the asymmetric hydrogenation of ␤ C6H6)Cl]Cl was much more active and enantioselective than -(acylamino)acrylates is rather limited (51, 52). ␩6 ␣ [Ru(BINAP)( -C6H6)Cl]Cl in the hydrogenation of -ke- Ligand 8 in combination with a cationic Ru(II) complex was toesters. For instance, the hydrogenation of methyl benzoylfor- found to be efficient for the asymmetric hydrogenation of ␩6 ␤ ␤ mate by using [Ru8( -C6H6)Cl]Cl catalyst was completed in -alkyl-substituted -(acylamino)acrylates. Optimization studies

Table 3. Asymmetric hydrogenation of 2-(6؅-methoxy-2؅-naphthyl)propenoic acid with [RuL(p-cymene)Cl]Cl

ee,* % (conﬁguration)

Entry PH2, psi T, °C Time, h 8 (S)-MeO-BIPHEP (S)-BINAP

1 1,000 rt 4 89.5 (R) 89.3 (S) 89.0 (S) 2 1,500 rt 4 91.2 (R) 89.5 (S) 89.7† (S) 3 1,000 0 24 95.2 (R) 94.2 (S) 94.1 (S) 4 1,500 0 24 96.0 (R) 94.4 (S) —

Reaction conditions: solvent ϭ MeOH (2.5 ml); substrate͞catalyst ϭ 100:1 (mol͞mol); substrate concentration ϭ 2.0 mg͞ml; complete conversions were obtained in all cases. *The ee values were determined by chiral HPLC with a Sumichiral OA-2500 column. † P ϭ 1,600 psi H2.

5818 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0307774101 Qiu et al. Downloaded by guest on September 29, 2021 Table 4. Asymmetric hydrogenation of ␤-dehydroamino acids Table 5. Asymmetric hydrogenation of enol acetates catalyzed 6 with [Ru8(␩ -C6H6)Cl]Cl by (NMe2H2)[(Ru8Cl)2(␮-Cl)3] SPECIAL FEATURE

Entry Ar T, °C Time, h Conversion, % ee,* % Entry R1 R2 T, °C Time, h ee,* % 1 Naphthyl rt 12 Ͼ99 96.7 1 Me Me rt 6 96.4 2 p-FC6H4 rt 12 Ͼ99 97.1 2 Me Me 0 48 97.7 3 p-ClC6H4 rt 14 Ͼ99 96.5 3 Me Et rt 6 95.7 4 p-MeOC6H4 50 48 Ͼ99 92.6 4 Me Et 0 48 98.1 5C6H5 50 96 75 94.9 5 Et Me rt 6 94.8 6 Et Me 0 48 96.9 Reaction conditions: EtOH͞CH2Cl2 (4:1) was used as solvent; substrate͞Ru ϭ ͞ ϭ 7 i-Pr Me rt 6 93.5 100:1 (mol mol); PH2 50 psi. ϫ 8 i-Pr Me 0 48 98.8 *ee’s were determined by chiral GC with a Varian 25 m 0.25 mm CP- CYCLODEX B 236M column; all products were in (R)-conﬁguration. 9 n- Et rt 6 94.0 Pr 10 n- Et 0 48 96.0 (Table 5), we found the enantioselectivities using (R,S,S)-8 to be Pr similar to those obtained using the Ru-TunaPhos system (entries 11 t- Me rt 6 99.8 1–3). However, in the hydrogenation of relatively electron-rich Bu substrates, such as 1-(4-methoxyphenyl)-1-(acetyloxy)ethylene Reaction conditions: 1.7 mg of substrate; substrate͞catalyst ϭ 100 (mol͞ and 1-phenyl-1-(acetyloxy) ethylene, although no reaction was mol); substrate concentration ϭ 0.05–0.09 M in MeOH, PH2 ϭ 250 psi; the observed in Ru-TunaPhos (54), Ru-8 still gave the desired conversions were determined by NMR and GC analysis and were found to be products in high ee’s (92.6% and 94.9%, respectively, entries 4 Ͼ99.9% in all cases. and 5). *The ee values were determined by chiral GC with a 25 m ϫ 0.25 mm CHEMISTRY Chirasil-DEX CB column or 30 m ϫ 0.25 mm ␥-DEX-225 column; all products Conclusions were in (S)-conﬁguration. In conclusion, we have realized a completely atropdiastereo- selective synthesis of a biaryl diphosphine dioxide by asymmetric intramolecular Ullmann coupling or Fe(III)-promoted oxidative indicated that methanol was the best solvent for this catalyst coupling. This synthesis utilizes central-to-axial chirality trans- system. The hydrogen pressure had little influence on the fer, a process enabling the synthesis of a diastereopure C - enantioselectivity. Lower reaction temperature afforded higher 2 symmetric biphenyl phosphine ligand without incurring other- enantioselectivity at the expense of the reaction rate. Excellent wise tedious optical resolution steps. The strategy introduced in enantioselectivities in the hydrogenation of (E)-␤-alkyl- ␤ this study has thus provided a convenient pathway for the design substituted -(acylamino)acrylates were achieved, and the re- and synthesis of diphosphine ligands based on biaryl atropiso- sults were summarized in Table 4. Substrate with a bulky alkyl merism, with the opportunity of tuning the torsion angle by using substituent gave the best ee (up to 99.8%, Table 4, entry 11). other chiral diols with various chain lengths. Although the Asymmetric hydrogenation of enol acetate is an attractive effectiveness of the new ligand has been demonstrated in a alternative to the direct hydrogenation of unfunctionalized ␲ limited range of asymmetric hydrogenation reactions, with the ketones. In addition to a -donor olefin group, this type of advent of this strategy, it is anticipated that this class of easily substrate supplies a secondary donor group for chelation, which accessible modular ligands may cover a wide spectrum of is helpful for obtaining high enantioselectivities in hydrogena- catalytic asymmetric reactions. tion. Most of the studies on this reaction are focused on using Rh-phosphines as catalysts. The use of Ru-phosphine system in We thank the University Grants Committee Areas of Excellence Scheme this reaction is rarely mentioned in the literature (53, 54). in Hong Kong (AoE P͞10-01) and the Hong Kong Polytechnic Univer- In our study of the asymmetric hydrogenation of enol acetates sity Area of Strategic Development Fund for financial support.

1. Noyori, R. (1994) in Asymmetric Catalysis in Organic Synthesis (Wiley, New 13. de Paule, S. D., Champion, N., Ratovelomanana-Vidal, V., Geneˆt, J.-P. & York), pp. 1–93. Dellis, P. (2001) Fr. Patent 2,830,254. 2. Ojima, I., ed. (2000) Catalytic Asymmetric Synthesis (Wiley, New York), 2nd Ed. 14. Pai, C.-C., Li, Y.-M., Zhou, Z.-Y. & Chan, A. S. C. (2002) Tetrahedron Lett. 43, 3. Noyori, R. & Ohkuma, T. (2001) Angew. Chem. Int. Ed. Engl. 40, 40–73. 2789–2792. 4. Bringmann, G., Breuning, M. & Tasler, S. (1999) Synthesis 525–558. 15. Gelpke, A. E. S., Kooijman, H., Spek, A. L. & Hiemstra, H. (1999) Chem. Eur. 5. Noyori, R. & Takaya, H. (1990) Acc. Chem. Res. 23, 345–350. J. 5, 2472–2482. 6. Tang, W. & Zhang, X. (2003) Chem. Rev. 103, 3029–3070. 16. Tuyet, T. M. T., Harada, T., Hashimoto, K., Hatsuda, M. & Oku, A. (2000) J. 7. Schmid, R., Foricher, J., Cereghetti, M. & Schonholzer, P. (1991) Helv. Chim. Org. Chem. 65, 1335–1343. Acta 74, 370–389. 17. Bringmann, G. & Menche D. (2001) Acc. Chem. Res. 34, 615–624. 18. Yin, J. & Buchwald, S. L. (2000) J. Am. Chem. Soc. 122, 12051–12052. 8. Schmid, R., Broger, E. A., Cereghetti, M., Crameri, Y., Foricher, J., Lalonde, 19. Hayashi, T., Hayashizaki, K., Kiyoi, T. & Ito, Y. (1988) J. Am. Chem. Soc. 110, M., Muller, R. K., Scalone, M., Schoettel, G. & Zutter, U. (1996) Pure Appl. 8153–8156. Chem. 68, 131–138. 20. Barhate, N. B. & Chen, C.-T. (2002) Org. Lett. 4, 2529–2532. 9. Pai, C.-C., Lin, C.-W., Lin, C.-C., Chen, C.-C. & Chan, A. S. C. (2000) J. Am. 21. Nicolaou, K. C., Li, H., Boddy, C. N. C., Ramanjulu, J. M., Yue, T.-J., Bra¨se, 122, Chem. Soc. 11513–11514. S.&Ru¨bsam, F. (1999) Chem. Eur. J. 5, 2584–2601. 10. Benincori, T., Brenna, E., Sannicolo`, F., Trimarco, L., Antognazza, P., 22. Cammidge, A. N. & Cre´py, K. V. L. (2000) Chem. Commun. 18, 1723–1724. Cesarotti, E., Demartin, F. & Pilati, T. (1996) J. Org. Chem. 61, 6244–6251. 23. Kano, T., Ohyabu, Y., Saito, S. & Yamamoto, H. (2002) J. Am. Chem. Soc. 124, 11. Saito, T., Yokozawa, T., Ishizaki, T., Moroi, T., Sayo, N., Miura, T. & 5365–5373. Kumobayashi, H. (2001) Adv. Synth. Catal. 343, 264–267. 24. Itoh, T. & Chika, J.-i. (1995) J. Org. Chem. 60, 4968–4969. 12. de Paule, S. D., Jeulin, S., Ratovelomanana-Vidal, V., Geneˆt, J.-P., Champion, 25. Lipshutz, B. H., Kayser, F. & Liu, Z.-P. (1994) Angew. Chem. Int. Ed. Engl. 33, N. & Dellis, P. (2003) Eur. J. Org. Chem. 1931–1941. 1842–1844.

Qiu et al. PNAS ͉ April 20, 2004 ͉ vol. 101 ͉ no. 16 ͉ 5819 Downloaded by guest on September 29, 2021 26. Lipshutz, B. H., James, B., Vance, S. & Carrico, I. (1997) Tetrahedron Lett. 38, 42. Evans, D. A. & Dinsmore, C. J. (1993) Tetrahedron Lett. 34, 6029–6032. 753–756. 43. Feldman, K. S., Ensel, S. M. & Minard, R. D. (1994) J. Am. Chem. Soc. 116, 27. Lin, G.-Q. & Zhong, M. (1997) Tetrahedron Lett. 38, 1087–1090. 1742–1745. 28. Nelson, T. D. & Meyers, A. I. (1994) Tetrahedron Lett. 35, 3259–3262. 44. Feldman, K. S. & Ensel, S. M. (1994) J. Am. Chem. Soc. 116, 3357–3366. 29. Nelson, T. D. & Meyers, A. I. (1994) J. Org. Chem. 59, 2655–2658. 45. Maitra, U. & Bandyopadhyay, A. K. (1995) Tetrahedron Lett. 36, 3749– 30. Nelson, T. D. & Meyers, A. I. (1993) Tetrahedron Lett. 34, 3061–3062. 3750. 31. Rawai, V. H., Florjancic, A. S. & Singh, S. P. (1994) Tetrahedron Lett. 35, 46. Zhang, Z., Qian, H., Longmire, J. & Zhang, X. (2000) J. Org. Chem. 65, 8985–8988. 6223–6226. 32. Ku, Y.-Y., Grieme, T., Raje, P., Sharma, P., King, S. A. & Morton, H. E. (2002) 47. Kitamura, M., Tokunaga, M., Ohkuma, T. & Noyori, R. (1993) Org. Synth. 71, J. Am. Chem. Soc. 124, 4282–4286. 1–13. 33. Michaud, G., Bulliard, M., Ricard, L., Geneˆt, J.-P. & Marinetti, A. (2002) 48. Mashima, K., Kusano, K. H., Sato, N., Matsumura, Y., Nozaki, K., Kumoba- Chem. Eur. J. 8, 3327–3330. yashi, H., Sayo, N., Hori, Y., Ishizaki, T., Akutagawa, S. & Takaya, H. (1994) 34. Miyano, S., Fukushima, H., Handa, S., Ito, H. & Hashimoto, H. (1988) Bull. J. Org. Chem. 59, 3064–3076. Chem. Soc. Jpn. 61, 3249–3254. 49. Juaristi, E., ed. (1997) Enantioselective Synthesis of ␤-Amino Acids (Wiley-VCH, 35. Nelson, S. G. & Hilfiker, M. A. (1999) Org. Lett. 1, 1379–1382. New York). 36. Spring, D. R., Krishnan, S., Blackwell, H. E. & Schreiber, S. (2002) J. Am. Chem. Soc. 124, 1354–1363. 50. Pen˜a, D., Minnaard, A. J., de Vries, J. G. & Feringa, B. L. (2002) J. Am. Chem. 37. Kamikawa, K., Watanabe, T. & Uemura, M. (1996) J. Org. Chem. 61, Soc. 124, 14552–14553. 1375–1384. 51. Wu, J., Chen, X., Guo, R., Yeung, C.-H & Chan, A. S. C. (2003) J. Org. Chem. 38. Vorogushin, A. V., Wulff, W. D. & Hansen, H.-J. (2002) J. Am. Chem. Soc. 124, 68, 2490–2493. 6512–6513. 52. Zhou, Y.-G., Tang, W., Wang, W.-B., Li, W. & Zhang, X. (2002) J. Am. Chem. 39. Qiu, L., Qi, J., Pai, C.-C., Chan, S., Zhou, Z., Choi, M. C. K. & Chan, A. S. C. Soc. 124, 4952–4953. (2002) Org. Lett. 4, 4599–4602. 53. Ohta, T., Miyake, T., Seido, N., Kumobayashi, H. & Takaya, H. (1995) J. Org. 40. Dai, D. & Martin, O. R. (1998) J. Org. Chem. 63, 7628–7633. Chem. 60, 357–363. 41. Miyano, S., Tobita, M., Nawa, M., Sato, S. & Hashimoto, H. (1980) J. Chem. 54. Wu, S., Wang, W., Tang, W., Lin, M. & Zhang, X. (2002) Org. Lett. 4, Soc. Chem. Commun. 1233–1234. 4495–4497.

5820 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0307774101 Qiu et al. Downloaded by guest on September 29, 2021