ACCOUNT 2547 Regioselectivity, Stereoselectivity and Catalysis in Intermolecular Pauson–Khand Reactions: Teaching an Old Dog New Tricks IntermolecularSabine Pauson–Khand Reactions Laschat,* Armand Becheanu, Thomas Bell, Angelika Baro Institut für Organische Chemie, Universität Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany Fax +49(711)6854285; E-mail: [email protected] Received 18 July 2005 Dedicated to Prof. Henning Hopf on the occasion of his 65th birthday. method in organic synthesis. Cyclopentenones are very Abstract: Since its discovery in the early seventies the inter- versatile building blocks for natural products, pharma- molecular Pauson–Khand reaction has made considerable progress 2 towards a powerful synthetic method. This account describes the ceuticals and fine chemicals such as prostaglandins 4, major accomplishments with respect to reactivity, stereoselectivity isocarbacyclins 6, sesquiterpenes 5, and jasmonates 7 and catalytic versions, which have been achieved over the last (Scheme 2). decade and summarizes mechanistic information being obtained by theoretical and experimental studies. H 1 Introduction CO2H 2 Reactivity: Improvement by Additives O HO CO2H 3 Regioselectivity: Some Mechanistic Struggles H 4 Stereoselectivity 4.1 Chiral Precursors 4.2 Chiral Auxiliaries HO OH C H 4.3 Chiral Cobalt Complexes 5 11 4 5 4.4 Chiral Amine N-Oxides 5 Catalysis H O 5.1 Cobalt-Catalyzed Reactions 5.2 Ruthenium-Catalyzed Reactions 5.3 Rhodium-Catalyzed Reactions H 5.4 Iridium-Catalyzed Reactions H CO2Me 6 Theoretical Studies and some Mechanistic Curiosities 6 7 7 Conclusions Scheme 2 Key words: Pauson–Khand reaction, metal-catalysis, stereoselec- tivity, theoretical studies The mechanism of the Pauson–Khand reaction, initially proposed by Magnus,3 has now been widely accepted (Scheme 3). In the presence of Co2(CO)8 the alkyne forms 1 Introduction the tetrahedral dicobalt complex 8. After loss of CO, the alkene is coordinated to give the alkene complex 9, which During their studies of cobalt–alkyne complexes in 1971 undergoes insertion of the alkene moiety into the sterical- Pauson and coworkers reported the formation of cyclo- ly least hindered Co–C bond. Subsequent CO insertion pentenone 3 by treating acetylene dicobalthexacarbonyl gives rise to the cobalt acyl complex 11. Extrusion of one 2a with ethylene (Scheme 1).1 Co(CO)3 fragment yields the cobaltacyclopropene com- plex 12, which is finally converted to the cyclopentenone O 13 by reductive cleavage of Co2(CO)6. Except the forma- −2 CO + 8 H H + Co (CO) tion of the stable and isolable cobalt–alkyne complex , 2 8 ∆ Co2(CO)6 however, there was no experimental evidence for this 1a 2a 3 mechanism until recently. Scheme 1 Since the early reports by Pauson it was found that several other complexes containing transition metals such Fe,4 5 6 7 8 9 10 11 Originally discovered by serendipity, the cobalt-mediated Ru, Rh, Ni, Cr, Mo, W, Ti, and Zr can be used for [2+2+1] cocyclization of an alkyne, an alkene and carbon this cocyclization. monoxide to cyclopentenone, commonly known as the In 1981 Schore reported the first example of an intra- Pauson–Khand reaction, has grown into a powerful molecular Pauson–Khand reaction (Scheme 4).12 Despite this initial delay the intramolecular variant made much more progress over the last two decades particularly with SYNLETT 2005, No. 17, pp 2547–2570 17.10.2005 regard to reactivity, stereoselectivity and catalysis as Advanced online publication: 05.10.2005 DOI: 10.1055/s-2005-918922; Art ID: A38505ST compared to the intermolecular counterpart. © Georg Thieme Verlag Stuttgart · New York 2548 S. Laschat et al. ACCOUNT Based on our own research efforts in this area, the follow- gio- and stereoselectivity are often difficult to control. ing article will focus on the various above-mentioned And last but not least a catalytic version is out of sight. issues of the intermolecular Pauson–Khand reaction with some selected examples of the intramolecular version. Although the intermolecular Pauson–Khand reaction pro- 2 Reactivity: Improvement by Additives ceeds in a highly convergent fashion and tolerates a vari- ety of functional groups such as ethers, alcohols, tertiary One of the earliest attempts to improve the reactivity was amines, thioethers, ketones, acetals, esters, amides, aryl described by Smit13a (Scheme 5). By adsorption of the co- and alkyl halides, heterocycles, vinylethers and -esters, it balt–alkyne complex 16 to silica gel and performing the is limited to reactive alkenes such ethylene, allene and reaction without any solvent (dry state adsorption condi- strained cyclic alkenes. Sterically hindered alkenes are tions) reaction rate and yield of the intramolecular Pau- disfavored. In many cases, yields are only moderate. Re- son–Khand reaction could be dramatically increased. Biographical Sketches Sabine Laschat studied of Professor Larry Over- Technische Universität in chemistry at the Universität man, she finished her habil- Braunschweig from 1997– Würzburg from 1982–1987 itation at the Universität 2002. Since 2002 she holds and received her PhD at the Münster (mentor: Professor a Chair in Organic Chemis- Universität Mainz in 1991 Gerhard Erker) in 1996. try at the Universität under the supervision of Then she was appointed as Stuttgart. Professor Horst Kunz. After an Associate Professor of a postdoc year in the group Organic Chemistry at the Armand G. Becheanu was his engineer diploma. He sität in Braunschweig, Ger- born in 1977 in Lipova, Ro- started his PhD on ‘Pauson– many, under the supervision mania. He studied at the Khand Reactions as Key of Professor Sabine Laschat ‘Politehnica’ University Steps for Pharmaceutically and currently continues Timisoara between 1995 Relevant Target Molecules’ work in Stuttgart after the and 2000, where he received at the Technische Univer- group moved there in 2002. Thomas Bell was born in versität in Braunschweig, ‘Cellulose Modification London, UK, in February Germany, under the super- Using Ionic Liquids’ with 1981. He studied for a vision of Professor Sabine Professor A. P. Abbott and MChem (Europe) degree at Laschat. He graduated in Dr. S. Handa in cooperation the University of Leicester, 2003. He is currently study- with Procter and Gamble, UK, which included one ing at the University of Lei- UK. year at the Technische Uni- cester for his PhD on Angelika Baro was born in she received her PhD in tific staff member at the Stadtoldendorf, Germany. 1987 in Clinical Biochemis- Institut für Organische Che- She studied chemistry at the try under the supervision of mie, Universität Stuttgart, Georg-August-Universität Professor H. D. Söling. Germany. Göttingen, Germany, where Since 1991 she is a scien- Synlett 2005, No. 17, 2547–2570 © Thieme Stuttgart · New York ACCOUNT Intermolecular Pauson–Khand Reactions 2549 RL However, a disadvantage of the amine N-oxides is their + Co2(CO)8 need in large excess (3–6 equiv). A solution to this prob- S L R R (CO) Co Co(CO)3 – 2 CO 3 lem has been recently reported by Kerr, who developed a RS polymer-supported morpholine-N-oxide 21 which could 18be recycled up to five times by treatment with Davies re- 16 + R – CO agent (Scheme 7). This result is particularly relevant for the use and recycling of chiral amine N-oxides for stereo- L RL R selective Pauson–Khand reactions (see chapter 4.4). + CO Co(CO) Co(CO)3 (CO)3Co 3 (CO)2Co RS + 21 R RS OH R + 10 9 OH THF, r.t., 0.5 h Co2(CO)6 O + CO 2b 20 22 1. cycle: 91% RL 5. cycle: 86% RL O O Co(CO)3–Co(CO)3 Co(CO)3 O O (OC)3Co N N S S R H O R O O R R = Agrogel 11 12 21 O – Co2(CO)6 Scheme 7 RL A plethora of other Lewis bases such as dimethyl sulfox- O ide,17 cyclohexylamine,18,19 aqueous ammonium hydrox- RS ide (in dioxane),18 and sulfides20 has been developed in R 13 order to enhance Pauson–Khand reaction rates. Particular- ly n-butyl methyl sulfide turned out to be successful, Scheme 3 where other additives failed (Scheme 8).20 O Co2(CO)8 O Ph Ph ∆ + Co (CO) 14 15 (31%) 2 6 2c 23 24 Scheme 4 Scheme 8 Reagents and conditions: (i) Toluene, reflux, 3 d, 24: 23%; (ii) NMO (6 equiv), CH2Cl2, r.t., 10 min, 24: decomplexation; Co2(CO)6 (iii) n-BuSMe (4 equiv), Cl(CH2)2Cl, 83 °C, 1.5 h, 24: 85% method O O O A or B In order to circumvent the major disadvantage of n-butyl 16 17 methyl sulfide, its unpleasant odor and high volatility, Kerr prepared a sulfide 25 tethered to a Merrifield resin Scheme 5 Reagents and conditions: Method A: isooctane, 24 h, (Scheme 9).21 Even with sterically hindered cobalt– 17 17 60 °C, : 29%; Method B: SiO2, O2, 0.5 h, 45 °C, : 75% alkyne complexes such as 2d high yields could be main- tained over at least five cycles. Later Schreiber,13b Krafft,14 and Chung15 discovered the accelerating effect of tertiary amine N-oxides, which is probably due to the oxidative removal of one CO ligand + 25 from the cobalt–alkyne complex (Scheme 6). + Cl(CH2)2Cl Co2(CO)6 83 °C, 0.5 h O 2d 20 26 Ph method SMe 1. cycle: 89% Ph + O 5. cycle: 88% A or B 25 1b 18 19 O = Merrifield resin Scheme 6 Reagents and conditions: Method A: DME, 4 h, 60–70 °C, 19: 45%; Method B: Me3NO (3 equiv), CH2Cl2, 2 h, Scheme 9 0 °C, 19: 80% Synlett 2005, No. 17, 2547–2570 © Thieme Stuttgart · New York 2550 S. Laschat et al. ACCOUNT A completely different approach to rate enhancement uti- silanes for isoprostan synthesis (Scheme 12).27 Surpris- lizes solvent effects.