The intramolecular Heck reaction SUSAN E. GIBSON (nee Thomas) and RICHARD J. MIDDLETON Department of Chemist4 Imperial College of Science, Technology and Medicine, South Kensington, London SW7 2AI: UK Reviewing the literature published up to the end of 1995 1 2 2.1 2.1.1 2.1.1.1 2.1.1.2 2.1.1.3 2.1.2 2.1.2.1 2.1.2.2 2.1.3 2.2 3 3.1 3.1.1 3.1.2 3.2 4 4.1 4.2 4.3 4.4 4.5 5 6 7 Introduction Synthesis of heterocycles Monocyclisations Five-membered ring formation Synthesis of indoles and related heterocycles Other nitrogen-containing rings Rings containing other heteroatoms Six-membered ring formation Nitrogen-containing rings Rings containing other heteroatoms Synthesis of medium and large rings Multiple cyclisations Synthesis of carbocycles Monocyclisations With P-hydride elimination With anion capture Multiple cyclisations Regio- and stereo-control Mode of cyclisation - endo or axo? Control of isomerisation and the r d e of additives Unexpected products Diastereocontrol Enantiocontrol Total syntheses Conclusion References 1 Introduction For most of the 1980s the Heck reaction was a well known but little used organometallic reaction.Broadly designated as the palladium catalysed arylation or vinylation of alkenes it had great potential in organic synthesis, but, perhaps due to recognised problems such as lack of regioselectivity and the necessity for harsh reaction conditions, only a few research groups pursued its promise. The intramolecular version of the reaction was almost unheard of until the mid-eighties, when the synthesis of heterocycles by this method started to be properly explored; and the question of whether carbocycles could be synthesised via the intramolecular Heck reaction was left virtually unanswered until the late eighties.Since this time, however, there has been an explosion of interest in the intramolecular Heck reaction, culminating in the last three years in its widespread application in total synthesis. catalysed reaction of a vinyl or aryl halide 1 with an alkene 2 to form a new carbon-carbon bond as in 3 together with the formation of a hydrogen halide as a by-product, was developed as long ago as the early 1970s.' The Heck reaction, defined as the palladium 1 2 The currently accepted mechanism of the reaction was defined in the 1970s2 and has remained unmodified in everything but detail since then.The first step is oxidative addition of a 14 electron palladium(0) species 4 to a bond joining a carbon atom to an appropriate group, usually a bromide, iodide or triflate, to form a carbon-palladium o-bond as in 5 . The carbon atom is usually part of an aryl or vinyl group as any sp' hybridised carbon atoms bearing hydrogen atoms P to the palladium atom would lead to facile P-hydride elimination. After coordination of an alkene to the palladium, carbopalladation leads to another carbon-palladium o-bonded complex 6 . Provided there is now an sp' hybridised hydrogen P to the palladium atom, then /I-hydride elimination will occur and the product of \ I d 6 Gibson: The intramolecular Heck reaction 447the reaction 7 will be released from the cycle.The palladium(ii) complex 8 thus formed then undergoes reductive elimination to regenerate the palladium(0) species 4 which can then continue round the catalytic cycle. lacking an appropriate fi-hydrogen then the complex may be long-lived and thus susceptible to a further cross-coupling reaction. This step may take place with the use of anionic or neutral nucleophiles as well as organometallics but the term ‘anion capture’ is generally used to avoid confusion with the Heck coupling steps. The group bonded to the unsaturated carbon atom which undergoes the oxidative addition step is usually iodide, bromide or, more recently, triflate. Chlorides do not give very good results although some success has been achieved with aroyl’ and benzyl chlorides4 The standard Heck conditions of catalytic quantities of palladium (usually palladium acetate) and an amine base in a polar aprotic solvent have been found to be widely applicable. (Phosphine ligands are necessary to aid the oxidative addition step in the case of bromides but not iodides.) The reaction generally requires high temperatures under these conditions.More recently, modification of the conditions by Jeffery et al.s through use of an inorganic base together with a phase transfer agent has allowed the Heck reaction to occur at lower temperatures. These have come to be known as the Jeffery conditions and have widened the scope and effectiveness of the Heck reaction still further. Other modifications to conditions have sometimes been found to be advantageous and these will be described later.Since the real upturn in interest in the Heck reaction in the last few years there have been several reviews of the Heck reaction and related palladium catalysed coupling reactions.6 In spite of the number of examples and specific features of the intramolecular reaction, however, there has been no comprehensive review devoted to this area to date, and thus such a review is undertaken herein. Due to the variety of Heck-type couplings that now exist some definition of the scope of the review is necessary. Therefore, for the purposes of this review, an intramolecular Heck reaction is defined as one where: 1) the first step in the catalytic cycle involves oxidative addition of palladium(0) into a carbon-heteroatom a-bond to form a carbon- palladium(1r) a-bond, and 2) the next step involves coordination to and the carbopalladation of an alkene, alkyne or allene which is part of the same molecule.intramolecular Heck reactions falling within the above definition leading to the synthesis of, first, heterocycles (Section 2) and then carbocycles (Section 3). Section 4 deals with several areas of specific interest in the intramolecular Heck reaction and Section 5 provides a few examples of the application of the reaction in the area of total synthesis. If the palladium in a complex similar to 6 is In the next two sections, we review all 2 Synthesis of heterocycles 2.1 Monocyclisations 2.1.1 Five-membered ring formation 2.1.1.1 Synthesis of indoles and related heterocycles This section deals with the synthesis of indoles and heterocycles containing the indole skeleton.There are more examples of the synthesis of indoles by the intramolecular Heck reaction than of any other group of compounds. This is mainly due to the immense importance of indoles as a species in biologically active compounds. Other, more practical, reasons for their prevalence are the commercial availability of ortho-iodo and ortho- bromo anilines 10 and the ease of forming carbon- nitrogen bonds, thereby giving rise to a wide variety of possible substrates 9 for the intramolecular Heck reaction. It is in these types of substrates that many of the steric and electronic factors which affect the Heck cyclisation have been explored. R3 R3 In 1977 Mori et al. reported the first intramolecular Heck cyclisations.’ They found that aryl bromide 12 could be cyclised to the indole 13 in 43% yield when reacted neat with palladium acetate (2 mol%)/triphenylphosphine (4 mol%) as the catalyst and TMEDA (2 equiv.) as the base.Under these conditions other by-products were also formed (such as the deallylated product l l ) , but these could be largely avoided through the use of DMF as a solvent. The aryl chloride was found to give no cyclised product due to the difficulty of the initial oxidative addition step. Also worth noting here is the fact that the initially formed double bond would be exocyclic but this completely isomerises in the reaction mixture to the more stable endocyclic position, probably through the re-addition of the HPdBr species and a second b-hydride elimination.This isomerisation can be prevented by the addition of silver salts which give the exocyclic isomers in good yields;8 the use of silver and other salts will be discussed later. aBr 11 /WE NHAc + __I V 1Z Br Me02C 702Me \ Ir ’ Pd(OAc)P, PPh3 l!, ,AN TMEDA (2 equiv.) 125 O C , 5 h AC Ac 4 Q 43% A- 448 Conternporaiy Oqanic SynthesisSubsequently, in 1979, Mori et al.9 and Heck et al.'" published syntheses of indolones using the standard conditions of catalytic palladium, with a phosphine as ligand and an organic base, usually triethylamine. In Heck's paper N-cinnamoyl- and N-(P-methylcinnamoy1)-o-bromoaniline 14 and 15 were cyclised to the benzylideneindolones 16 and 17 with 1 mol% palladium acetate and 4 mol% tri- o-tolylphosphine at 100 "C.(Tri-o-tolylphosphine is sometimes used in place of triphenylphosphine in order to prevent the formation of phosphonium salts.) Once again, the preference for 5-exo-trig cyclisation over 6-endo-trig cyclisation is observed and the factors influencing this preference in the Heck cyclisation will be examined later. A reversal of alkene stereochemistry is also seen and this reflects the necessity for palladium to adopt a .syn relationship to the P-hydrogen as in 18 before ,!I-hydride elimination can occur. Apparent exceptions to this rule will also be discussed later. aZLR Pd(OAc)2, P(o-Tol)s Et3N, MeCN, 100 "C H H 14 R = H 16 R = H, 58% 15 R=Me 17 R =Me, 21% R Ph PdBr Ph\ .z - -R 'I-- PdBr d o - d o - 16or17 H H 18 All the examples described so far involve alkenes conjugated to a carbonyl group, which is known to activate alkenes towards Heck insertion.In order to synthesise indoles by the cyclisation of unactivated N-allylanilines, Hegedus et al. found that sequential addition of the catalyst at various times over the course of the reaction gave much improved yields." The catalyst appears to become deactivated as the reaction proceeds, possibly through the formation of colloidal palladium; where the cyclisation occurred, as in, for example, the cyclisation of 19 to 20, yields Pd(0AC)z (3 X lye), E1,N (1.5 + 0.5 + 0.5 equiv.) MeCN. 110 "C, 72 h 87% 19 20 n Pd(0Ac)p (3 x lYo), A4 Et3N (1.5 + 0.5 + 0.5 equiv.) 21 MeCN. 110 "C, 72 h of 50-87% were observed. Several other points were noted. First, use of phosphines was necessary in the case of aryl bromides but not aryl iodides as might be expected from the intermolecular examples;" second, the reactions exhibited a clear preference again for the 5-exo over the 6-endo cyclisation; and third, the complete failure of N-cyclohexenylaniline 21 to react suggests a stereoelectronic constraint on the cyclisation.Although, for some substrates, the Heck cyclisation can tolerate various substituents on the aromatic ring and the alkene and give good yields of indoles under the right conditions,'3 it can also be very sensitive. Small changes in either the reaction conditions or steric or electronic factors in the substrate can have a huge effect on the yield of indole pr0du~ed.l~ For example, Hegedus et al. reported a case where acetate substituents on the aromatic ring of 22 gave a good yield of the indole 23 but when they were replaced with the more electron donating methoxy substituents as in 24 no cyclised product was obtained.'' Another good example of this sensitivity is provided by the cyclisation of aryl iodide 25 which gave indole 26 in very good yield,lb whilst under very similar conditions the similar substrate 27 gave only a poor yield of the desired product 28.17 L$ EtsN, Pd( OAC) MeCN, 2, P 50-1 (o-Tol), 10 "C DLy$ Ac Br 64% Br $, OAc OAc 22 23 24 C02Et a;yo2Et- Pd(OAc)p, 120 "C, EtsN, 3 h.88% DMF Q$C02Et H C02Et H 25 26 fOTHP <OTHP Pd(OAC)z, EtsN, DMF 1 10 "c, 25% H H &: 27 28 The attractiveness of this method of indole synthesis was improved further by Larock et a1.who found that the use of Jeffery's phase transfer conditions instead of the standard Heck conditions gave improved yields at lower temperatures." The conditions employed used 2 mol% palladium Gibson: The intramolecular Heck reaction 449acetate, 1 equivalent of tetrabutylammonium chloride, DMF as solvent and 2.5 equivalents of base (either sodium carbonate, triethylamine or sodium acetate). It was also noted that substitution on either nitrogen or the double bond slowed the reaction. Heck conditions and Jeffery conditions have been used several times over the past few years in the synthesis of simply substituted target indoles,” one such example being the synthesis of an analogue of sumatriptan 29.20 M e N H S E I F Pd(OAC)p, E13N M e N H S d ” w / DMF, A r 81% H O A C F 3 29 This methodology is not limited only to the formation of simple indoles as polycyclic fused ring and spirocyclic systems related to the indole system have also been synthesised.2’ The earliest examples were the syntheses of various carbazoles in 1980 by Iida et a1.22 but only poor yields were obtained.In 1987 Overman et al. published details of the synthesis of a number of spirocyclic indolones,” one of which was formed by cyclisation onto a tetrasubstituted alkene 30. Palladium catalysed arylations or vinylations of tetrasubstituted alkenes do not occur when performed intermolecularly and thus the versatility of the intramolecular reaction was again demonstrated. In recent years, Overman has shown that quaternary centres in oxindoles may be formed both diastereo~electively~~ and enantio~electively~~ as will be discussed later.5% Pd(OAC)p, 10% PPh3 0 2 equiv. Et3N * MeCN, 82 “C 56 h. 58% .Me 30 The versatility of this synthetic method has been extended still further by the use of the ‘anion capture’ methodology.26 Although there is one earlier example of hydride capture to give a 3,3-dimethylind0line,’~ the true scope of this method was first demonstrated by Grigg et al. who synthesised indole-related heterocycles by cyclisation onto geminally disubstituted alkenes with h ~ d r i d e , ~ ~ cyanidez8 and carbonylz9 capture, and onto alkynes with capture by organotin,” organozinc and organoboron reagents.” In one such example iodoaniline derivative 31 was cyclised to give the alkyl palladium species 33 which then underwent transmetallation with hexamethylditin, followed by a Stille type coupling to give 32.32 Luo et al.have recently reported conditions under which intramolecular cyclisations onto terminal alkynes such as 34 provide vinyl palladium intermediates 35 which may be trapped in situ with a wide range of organozinc reagents to provide products such as 36 in good yield.33 31 \ S02Ph 32 - J 33 / 34 36 72% 35 Finally, there have been several other palladium catalysed syntheses of indoles and related heterocycles which fall just outside the scope of this review but may be of interest to the reader.34 2.1.1.2 0 t her nit rogen-con taining rings Isoindolones were first synthesised by Grigg et al. as part of the work which demonstrated the feasibility of forming fused, spiro and bridged ring compounds by the intramolecular Heck rea~tion.~’ They were formed in good yields from vinyl amides of 2-iodobenzoic acids such as 37.As such amides are readily synthesised, a wide variety of substrates were thus available upon which a whole host of methodologies could be tested. Such vinyl amides were one of the main ways of testing the ‘anion capture’ methodology and were thus cyclised with hydride,” cyanide” and carb0ny1~~ capture. Furthermore, as bridgehead double bond formation is disfavoured, cyclisation onto amide substituted norbornenes provided a stable alkylpalladium intermediate which could then undergo organotid” or organoboron capture.31 Hence 37 was cyclised to 39 via 38 in good yield. Cyclisations onto dienes or allylic acetates gave n-allylpalladium complexes which then underwent capture with organozinc reagents”’ or carbon, nitrogen or oxygen nucleop hiles.36 For example, dienamide 40 was cyclised to 41.Other syntheses of isoindolones have been described, all occurring via 450 Contemporary Organic Synthesis37 \Ph 40 38 Pd(OAc)p, PPh3 2 equiv. NaCH(CN)2 MeCN, 80 "C 60% 39 A N Ph 41 _- 5-exo-trig cyclisations. '' Isoindolones have also been synthesised in studies on stereoselectivity3' and control of isomerisation7" and in the synthesis of indoloindoles via cyclisation onto an indole ring." The intramolecular Heck strategy has also been applied to the synthesis of a range of natural products featuring nitrogen-containing five membered rings such as camptothecin," magallane- sine," lentiginosine" and the mitoccnes."." A key step in the synthesis of an example of the latter was the cyclisation of an appropriately substituted N-vinyl- or N-allyl-bromoquinonc.For example, quinone 42 was cycliscd to the indolequinone 43 in excellent yield. The Heck cyclisation methodology 0 Meol?iBr Me0 *.I Pd(OAC)*, E13N * MeCN, 2 h, r.1. 97% 42 M e o w Me0 0 43 for the formation o f indoles has also been applied to thc synthesis of indole-type aromatic heterocycles where in place of a benzene ring there is a thiophene or selenophene to give thienopyrroles and s e 1 en o p y r r ol c s . ' ')" C y c 1 is at i o n s which construct n it rog e n - co n t a i n i ng heterocycles o n p re -for med indole rings as in thc synthesis of the indolo[3,2,1- i j ] [ 1,6)-naphthyridine ring system" and cyclisations onto the 2,3-double hond of i n d o l c ~ ~ ~ have been reported.Pyrrolidinones have also been formed from the Heck cyclisation of x-haloamides but only in poor to moderate Lastly, in a recent paper by Shibasaki c't ~ 1 . ' ~ the enantioselectivc formation of a five membered ring containing a nitrogen atom from B prochiral substrate was described; this work will be discussed morc fully I ate r . 2.1.1.3 Rings containing other heteroatoms This section is devoted almost exclusively to the formation of ox.ygcii-containir1~ rings. l'hc carlicst example involved the cyclisation of bromodialkenyl ethers and capture of the stable n-allylpalladium intermediate with a secondary amine, such as piper- idine.49 By this method, vinyl bromide 44 was cyclised to the 2,3-disubstituted dihydrofuran 46 via the n-allylpalladium intermediate 45 in good yield./-- Pd(OAC),, P(eT0l)3 O W 44 45 46- Larock et al. applied the conditions which had proved to be most successful for their synthesis of indoles'' to the synthesis of benzofurans.sO In this case, however, there is the complication that palladium(0) is known to react with aryl allyl ethers to give 7r-allylpalladium complexes. Together with one equivalent of sodium formate (whose rble may be to reduce any stable z-allylpalladium complexes that form thus releasing the palladium back into the catalytic cycle), the mild modified Jeffery conditions provided good yields of the desired benzofurans. In this manner iodoaryl allyl ether 47 gave 3-methyl- benzofuran 48.Negishi et al. also described the synthesis of bcnzofurans from aryl allyl ethers but used the standard Heck conditions. They obtained similar or slightly better yields but also obtained a mixture of isomers.5' P~(OAC)~, Na2C03 a:k Bu4NCI,HC02Na * DMF, 80 "C, 48 h 47 47% 48 Hoffmann el ai. described an intramolecular hydroarylation reaction on aryl ally1 ether 49 which gives the ABC ring system of the aflatoxins SO.'2 In this case there is n o syrz-[)-hydrogen for the palladium to eliminate arid so a hydride source was provided by preformed triethylammonium hydro- genformate. Pd(MeCN)2CI,, E13N H0 ~~ HC02H, DMF aLG0 50°C,3h 86% 49 50 Other benzofurans havc bcen synthesised by cycli- sations onto proximate alkynes with hydridc captures3 and spiro-fused benzofuran-2(3H)-ones have also been synthesised.2.i".s4 Recently an example of a cyclisation onto a carbohydrate template to form a fused five membered ring containing oxygen has also bccn The anion capture approach has been uscd by 1,arock et a!.in a synthesis of' prostaglandin-type corn pouri d s. "' I n t ra m ole cu I ar cycl i sa t ion of 5 1 leads to alkylpalladium intcrmcdiate 52 which lacks an appropriate syn-/&hydrogen for elimination and so is 3s IHO 51 PizNEt, DMF, 50 "C f l C O z E t o \ / b HO I . 53 42% available for anion capture. Both capture with alkenes (an intermolecular Heck) and organotin compounds (a Stille type coupling) have been used with alkenes giving the best results as exemplified by the synthesis of 53.There is only one publication, to date, on the synthesis of five-membered rings containing sulfur and this describes the synthesis of benzo- [blthiophenes by cyclisation of aryl allyl or aryl prop-2-ynyl thioethers." Although there are possible problems associated with the thiophilicity of palladium, under fairly standard Heck conditions at high temperature the desired benzothiophenes can be formed. For example, in a simple case, aryl allyl thioether 54 was cyclised to 3-rnethylbenzo[b]thio- phene 55 in good yield. tion is almost always preferred over 6-endo cyclisa- ion 15,35,58 the 6-endo cases must be dominated by other steric, electronic or stereoelectronic f d c t o r ~ . ~ ' ~ ~ ~ This topic will be dealt with more fully later. 6-Ex0 cyclisations of aryl bromides or iodides onto appropriately positioned alkenes can lead to substituted quinolines,lg as in the cyclisation of 56 to 57, or isoquinolines and related h e t e r o ~ y c l e s ~ ~ ~ ~ ~ " ~ ~ as in the cyclisation of 58 to 59.Where the alkene is itself part of a ring then spiro, fused23 or bridged35 rings may be formed as illustrated by the cyclisation of N-ally1 amide 60 to 61 in excellent yield. Fused ring systems have also been formed when the aryl iodide moiety was attached to the nitrogen atom of an indole ring bearing an alkene.45 When this methodology was directed towards the synthesis of dynemicin-A the fused quinolones 64 and 65 were successfully synthesised from 62.6' The BSA [N,O-bis(trimethylsilyl)acetamide] 63 was found to be crucial to the success of the cyclisation as it temporarily protects the secondary amide, a group which has been known to give problems in intra- molecular cyclisations.23 6-Ex0 cyclisation to form quinolones has also been employed in the total synthesis of (R, R)-crinan,62 racemic ly~oricidine~~ and ( + )-ly~oricidine.~~ Useful heterocycles can also be formed by 6-ex0 cyclisations onto in dole^.^^,^' Pd(OAc)p, Bu~NCI NaOAc. DMF I Et 57 55% 56 NH Pd(OAC)p, Bu~NCI Na2C03.DMF 54 a5 55 (J-j 80 "C, 24 h Pd(PPh&, Et3N MeCN, 140 "C. 15 h 70% 39% 58 59 2.1.2 Six-membered ring formation 2.1.2.1 Ni trogen-containing rings This section deals mainly with the synthesis of q N 9 ~~~K~~~ quinolines and isoquinolines and their partially either 6-endo or 6-ex0 cyclisations. As 5-exo cyclisa- 60 61 MeCN, 30-80 "C 88% 0 saturated analogues.These have been formed by 0 0 HN L Pd~(dba)~.CHCl~, P(*TOI)~ PII2NEt3, BSA-DMF (1:l) & 0 64 0 70 "C, 1 h 84% - n - o 62 5 : l BSA = N,Qbis(trimethyl)acetamide 63 452 Contemporary Organic SynthesisThe tandem cyclisation-‘anion capture’ methodo- logy can also be applied to the synthesis of six- membered rings containing nitrogen. This has been and capture with either h ~ d r i d e , ” . ~ ~ cyanide” or Et02C demonstrated by cyclisations onto proximate alkynes (PhaP)dPd, E13N organotin reagents3’ 6-Ex0 cyclisations of benzyl 73% halides onto alkenes with either hydride or sodium tetraphenylborate capture have also been used to prepare 3.3-disubstituted tetrahydroisoquinolines in good yield.4h Substituted piperidines may be synthesised by cyclisation of bromodialkenyl amines and subse- quent trapping of the n-allylpalladium intermediate with an in situ n~cleophile.“~ The cyclisation of an N-ally1 iodide onto a cyclic alkene to form a bridged ring product has also been successfully used in an approach to the Stlychnos alkaloid^.^^ Fused pyridi- nequinones have been synthesised as the minor product by a 6-endo cyclisation of bromoquinones where the major products were the desired indolo- quinones formed via the preferred 5-exo cyclisation.” 2.1.2.2 Rings containing other heteroatoms As for the five-membered rings, this section deals almost exclusively with the synthesis of oxygen- containing rings and again there are far fewer examples of this type than their nitrogen analogues.The earliest examples were reported by Heck et al.49 and included the cyclisation of a o-bromoaryl homoallyl ether 66 to give a mixture of isomers 67 and 68. 67 1 + Pd(OAc)l, P(eTo1)~ 47% EIsN, 100 “C, 48 h Br 28% 66 There was only one other publication in this area6’ until Negishi et ul. reported the synthesis of fused and spiro tetrahydrobenzopyrans from o-iodo- benzyl allyl ethers, and dihydropyrans from iododialkenyl ethers under standard Heck condi- tions.s’ On the whole, however, mixtures of isomers were formed. Overman et al. also used this method- ology (but with the addition of silver salts to prevent isomerisation) to form a spiro tetrahydrobenzopyran system in their total synthesis of (+)-tazettine and ( )-6a-epipreta~ettine.~~ Also under standard Heck conditions bromoaryl 69 gave tricyclic 70 in good yield.7o High temperatures were necessary to isomerise the double bond to the thermodynamically preferred (2) product.Other examples of the formation of oxygen-containing six-membered rings via the 6-ex0 cyclisation of aryl halides onto proxi- mate alkenes have been rep~rted.~”’’ cyclised onto an allyl ether in very good yield as a An iodopyridine derivative was successfully 69 70 key step in the synthesis of 20-(S)-~amptothecin.~* Six-membered cyclic ethers have also been formed by cyclisations of iodoaryls onto vinyl s~lfones’~ (such as the cyclisation of 71 to 72) and nitroalk- e n e ~ . ~ ~ In both cases the bromoaryl gave no cyclised product, the addition of silver salts gave improved yields and phosphines were necessary.It is worthy of note that a nonpolar solvent was found to give by far the best results for the cyclisation onto the nitroalkene. O*O Pd(PPh&, Et3N AgN03 (5 equiv.) MeCN, reflux, 3.5 h c w 97% o-oSo2Ph 71 S02Ph 72 A six-membered ring containing a nitrogen- oxygen bond has been synthesised by the intra- molecular Heck reaction in the synthesis of FR 900482 by Danishefsky et al.” This proceeded in excellent yield with a large excess of triethylamine but otherwise under standard Heck conditions. Lastly, a cyclisation of a sulfonamide by Grigg et al. gave a 1 : 1 mixture of 6-ex0 and 7-endo products with the nitrogen-sulfur bond as part of the bridged ~ystern.’~ Thus, together with the other examples, this demonstrates that the intramolecular Heck reaction can be used not only in the presence of carbon-heteroatom bonds but also heteroatom- heteroatom bonds.2.1.3 Synthesis of medium and large rings Due to there being relatively few examples, the synthesis of medium and large ring heterocycles via the intramolecular Heck reaction will be dealt with together in this section. (Some of the factors governing these cyclisations will be dealt with in more detail in Section 4.) The first example of the synthesis of a large ring by the intramolecular Heck reaction was reported in 1981 when the 16-membered lactone 74 was prepared from 73 in 55% yield.77 This reaction used one equivalent of palladium and a slow addition of the substrate 73 to the reaction mixture in order to prevent the forma- tion of dimers and oligomers by the intermolecular reaction.A seven-membered cyclic ether was formed by a 7-endo cyclisation with piperidine capture of a n-allylpalladium intermediate in the same manner as the capture of 45 to give 46.49 The ?’-ex0 cyclisation Gibson: The intramolecular Heck reaction 453II 1 0 II PdC12(MeCN)2 (1 equiv.) I fi HC02H (3 equiv.) I \ Et3N (8 equiv.) MeCN, 25 “C 55% Me 73 74 has been used to provide b e n z a ~ e p i n e s ~ ~ and cyclic etherS,jl amideS23,S3.7R and sulfonamides.‘3 In the last year, two reports of the synthesis of seven-, eight- and nine-membered rings using the Heck reaction have appeared. Gibson (nee Thomas) et al. cyclised aryl iodides tethered to dehydroala- nine units by two to four methylene groups 75 under anhydrous Jeffery conditions to give seven-, eight- and nine-membered rings 76 by endo ring clos~re.’~ Negishi et al.have synthesised seven-, eight- and nine-membered cyclic ethers by cyclisation onto allenes in reasonable yields but have found that the corresponding cyclisations onto alkenes only proceed in very poor yield.” Thus allene 77 cyclised to the eight-membered ring 78 in good yield but alkene 79 gave no isolated yield of 80. The reported results seem to indicate that allenes are more reactive than either alkenes or alkynes towards intramolecular carbopalladation and thus the versa- tility of the Heck reaction is increased still further. P~(OAC)~, NaHC03 Bu4NCI, 3 A mol. sieves 5 MeCN, 95 ‘C, 16.5 h n=1-3 7s 5b60% 76 0.05 M DMF. 100 “C 2 h, 52% 77 78 C12Pd(PPh3)2, K2CO3, 10 equiv.EtOH 0.05 M DMF, 100 “C 21 h 80 c 5% by NMR 79 Sundberg et aZ. have formed an eight-membered ring via an 8-endo cyclisation of a 3-iodo-(N)-methy- lindole onto a pendant alkene under Jeffery condi- tions.*’ Following this, in the last year, Rigby et al. have reported that, in the cyclisations of aryl iodides onto enamides to form medium sized rings, Jeffery conditions gave the product of endo cyclisation and standard Heck conditions gave the product of exo cycli~ation.~~ During the course of their investiga- tion, they formed both seven- and eight-membered rings containing nitrogen in good yield. Lastly, the synthesis of 16- to 22-membered lactones (also containing amide functionality) in 24 to 42% yield has been recently reported by Stocks et aLX2 They cyclised aryl iodides onto alkenes under standard Heck conditions; the tendency for large rings to form by exclusively endo cyclisations was noted.2.2 Multiple cyclisations The development of multiple Heck cyclisations in the synthesis of heterocycles has been carried out largely by the groups of Grigg and Overman. Bis- cyclisations were first reported by Overman et al. in 198Sg3 in the synthesis of spiro and fused ring systems. The importance of silver salts in promoting the desired reaction was also shown. The next several papers on the subject were by Grigg et al. who demonstrated the powerful nature of the methodology by combining a bis-cyclisation with an ‘anion capture’ ~tep.~~”,’~.*‘ Cirigg has coined several terms to describe the various species that must be present in a multiple Heck cyclisation.The ‘starter species’ is that into which palladium undergoes initial oxidative addition (e.g. an aryl or vinyl halide or triflate); the ‘relay species’, of which there may be more than one, is that which undergoes carbopalladation in a ring forming step (e.g. an alkene or alkyne) to give a carbopalladated intermediate; the ‘terminating species’ is that part of the substrate from which the palladium can return to the catalytic cycle by either b-hydride elimination or anion capture. There are two obvious but important restrictions that must be noted. Firstly, the relay species must be such that after carbopalladation, 1-hydride elimination cannot occur (e.g. an alkene must be appropriately substi- tuted).Secondly, anion capture must take place at a slower rate than the various cyclisation steps and thus alkynes are more likely to be successful as relay species due to their higher reactivity toward carbo- palladation than alkenes. in the synthesis of many varied ring structures. For example 81 was cyclised via 82 to 83 in good yield with hydride capture.27h The importance of using either silver or thallium salts was again demon- strated and the use of tetraethylammonium chloride This methodology has been successfully employed PdI 81 82 I I.. 83 454 Contemporary Oiganic Synthesisled to faster reaction times but also to an increased straightforward intermolecular Heck reaction with amount of premature capture product. The reactivity of certain alkylpalladium inter- mediates has also been demonstrated.".8' For example, after vinyl bromide 84 cyclised onto the proximal alkene to give a 'neopenty1'-palladium species, this then underwent another cyclisation to give the cyclopropane 85." Alkylpalladium inter- mediates will even undergo cyclisations onto aromatic rings, a process which has been termed 'Friedel-Crafts alkylation'.Spiroindolines have been synthesised by this method,s6 as have bridge~i-ring'~ and angularly fused-ring systems, as in the synthesis of 87 from 86.87 (&--Me H Pd(OAc)2, PPh3, KOAc, anisole 80 "C, 12 h 78% Ph02SN 84 Pd(OAC)p, PPhs TI2CO3, PhMe llO"C, 16h- 7&85% 85 /=3 87 86 The powerful nature of this methodology has been demonstrated further in a cascade cyclisation incorporating an intermolecular step." A 6-ex0 cycli- sation onto an appropriately positioned alkyne 88 provides the vinyl palladium intermediate 89 which can then undergo an intermolecular Heck reaction with norbornene to give 90.The alkylpalladium intermediate so formed cannot undergo P-hydride elimination and so cyclises to form a cyclopropane and another alkylpalladium intermediate 91 which can fi-eliminate to form the final product 92. An alkyne is necessary as the first relay species as it will undergo carbopalladation much more rapidly than an alkene, thus diminishing the importance of the IPd, ,H NMe Et3N, MeCN reflux 0 89 0 88 ia c-- \ \ 0 N. Me Me Me 40% 0 0 0 92 91 90 norbornene as a competing reaction. palladium catalysed cyclisation cascade of certain diene-ynes an electrocyclic ring closure can follow the 8-hydride elimination Thus dienyne 93 underwent palladium catalysed cyclisation to give 94 which then underwent electrocyclic ring closure to 95.De Meijere et al. have shown that in the 6 r 4 93 04 8- 94 95 Overman et al. have applied the palladium catalysed cascade cyclisation approach to the synthesis of spirocyclic polyethers.'" They found that the Jeffery conditions were the best for the synthesis of tricyclic diether 97 but gave only a poor yield when applied to the synthesis of tetracyclic triether 98. An important competing side reaction appears to be the palladium catalysed isomerisation of the allylic ether to the more stable enol ether moiety. 60% 96 97 ~ o T o T o 3 I \ as above 98 3 Synthesis of carbocycles 3.1 Monocyclisations 3.1.1 With fi-hydride elimination The intramolecular Heck reaction has been exploited far less in the synthesis of carbocycles than the synthesis of heterocycles.This was especially so in the earlier years of the 1980s. The imbalance is fast being remedied, however, as the reaction is applied not only to the synthesis of biologically interesting molecules, natural products, substituted aromatics etc., but also to cascade cyclisations. There is no synthetic reason why carbocycles may Gibson: The intramolecular Heck reaction 455not be as easily prepared by this reaction as hetero- cycles but it was not until the late 1980s that the area really started to flourish. In 1984 and then later in 1988 Grigg et al. reported the cyclisations of several substituted 2-brorno-x,w-diene~.'~ In 1987, as part of their paper on the synthesis of polycyclic systems containing quaternary centres, Overman et al.reported the formation of the fused ring carbocycle 100 from iodoarene 99.23 In 1988, Negishi et al. published their research on the formation of carbocycles by the cyclisation of iododiene~.'~ They found that the best conditions used triphenylphosphine as the ligand and a mixture of acetonitrile and THF as the solvent. Using dibenzylideneacetone as the Iigand gave a much slower reaction, again underlining the sensitive nature of this reaction to the conditions used. Pd(OAc)z, PPh3 MeCN, 30 h, 23 "C 71% 99 100 Concurrently, Larock et al. reported their study on the effect of different conditions on the forma- tion of carbocycles from iodoarene substituted alkenes.'? They found that in the cyclisation of iodoarene 101 to tricyclic 102 and 103, modified Jeffery conditions gave the desired product in good yield but as a mixture of the two double bond regioisomers in a 1 : 5.1 ratio.Overman's condi- tion~,'~ which had proved successful in the synthesis of spiro and fused ring systems, and controlled isomerisation with silver salts were found to slow the reaction down but at the same time gave a > 95 : 5 ratio of the two regioisomers. Standard Heck conditions do, however, give good yields of carbocycles with certain s ~ b s t r a t e s ~ ~ ' ~ ~ ~ although sometimes care in the choice of phosphine is neces- ~ a r y . ~ ~ i. 3% Pd(OAc)2, Bu~NCI, KOAC, DMF, ~ 5.1 2 days, 80 "C, 63% ii.1 % P~(OAC)~, 3% PPh3 + additions every 24 h, Ag2C03, MeCN, 80 "C, > 95 : 5 96 h, 72% A further paper by Negishi et al. in 1988 reported that the position of the doubIe bond in the cyclisa- tion product could sometimes be controlled by cyclising onto alkenes bearing carbonyl example, vinyl bromide 104 was cyclised to 105 in good yield under standard Heck conditions. Negishi et aZ. also examined the use of benzyl halides for the initial oxidative insertion step.& They found that benzyl chlorides gave the best results as iodides and bromides gave a greater amount of For Pd(OAc)Z, PPh3 NaHC03,DMF, * 8OoC, 36 h 0 68% 0 104 105 double bond regioisorners. The formation of five- and seven-membered ring carbocyclic products as well as a bis-cyclisation under standard Heck condi- tions were reported.For example, benzyl chloride 106 underwent 7-exo cyclisation to give 107 in good yield. Negishi et al. have also found that cyclisation of vinyl halides onto alkenes can sometimes lead to a cyclopropylcarbinyi-homoallyl rearrangement via a second cyclisation of an alkyl-palladium speces and so an initial 6-ex0 cyclisation can lead to a seven- membered ring and a 5-ao cyclisation to a six- membered ring. This rearrangement will be dealt with in more detail in Section 4.3. 106 107 The intramolecular Heck reaction was applied to the synthesis of bioactive molecules in an approach to the ergot alkaloids by Hegedus et al.98 Brorno- indoline 108 containing a monosubstituted alkene underwent 6-endo cyclisation to 109 in 50% yield under standard Heck conditions.This yield was increased to 64% in the case of the geminally disub- stituted alkene 110 which cyclised to 111. It appears to be the conjugation of the alkene in 110 to an electron withdrawing group rather than its disub- stitution which leads to the increase in yield, as cyclisation of r,P-unsaturated ketone 112 to 113 aIso proceeded smoothly. Br ylR A Pd(OAC)Z, Et3N P(eTol)s, MeCN Ts Ts 108 R = H 110 R=C02Et Go Ts 109 R = H, 50% 111 R = COP Et, 64"/0 Pd(OAC)*, Et3N P( ~ T O I ) ~ , MeCN Ts 691 112 113 Cyclisation onto allylic alcohols has also been reported.99 When a o cyclisation onto an allylic alcohol occurs the alkylpalladium intermediate can undergo fl-hydride elimination to form an enol. This enol can then tautomerise to the aldehyde as in the cyclisation of vinyl bromide 114 to cyclopentane 115.456 Contemporary Organic SynthesisE*2c?sco2Et Pd(OAC)*, PPh3 E'02cfico2E EtSN, MeCN 80°C,4h 60% HO CHO 114 115 Overman et ul. have more recently used vinyl and aryl triflates for the synthesis of cis angularly fused ring products. "') Although, for example, aryl triflate 116 cyclised in a 6-exo manner to give a mixture of regioisomers 117 and 118 in good yield, the condi- tions of the reaction required careful development. Bidentate phosphines were found to be necessary as monodentate phosphines gave only partial convcr- sion and amine bases were found to have a deleterious effect in giving less selectivity in the formation of the two isomcrs and also leading to the reduced product 119.(Amine bases arc thought to be hydride donors in palladium catalysed reactions."") Other bases also gave problems: silver salts led to decomposition of the starting material and bases such as sodium hydrogen carbonate gave side reactions such as the formation of 120. Lastly none of the desired product was observed at temperatures below 100 "C. With respect to tuning conditions, it has recently been shown that by varying the phosphine used, both the regio- and stereo-selectivity of an intramolecular Heck reaction can be altered.'"2 It is also worthy of mention that although almost none of the reactions indicated so far for the synthesis of carbocycles have used the Jeffery conditions, these can, in certain circum- stances, give excellent yields of carbocyclic products.'I)' W O B n TBDMSO 9, 116 R =OTf 119 R = H 120 R=OH Pd(dppb), KOAc Me2NCOMe, 120 "C 30 h, 68% 1 OTBDMS OTBDMS 117 20 : 1 118 Both medium and large ring carbocyclic products have been formed by the intramolccular Heck reaction. Although Roberts et al. had difficulty in controlling the regioselectivity of the reaction, they cyclised a bromoindolc onto a monosubstituted alkene to form, in a 2: 1 ratio, the 8-endo and 7-a0 products respectively in an 88% yield under standard Heck Negishi et al. have synthesised both medium and large ring carbocyclic products viu cyclisation onto both alkenes and allenes under high dilution Jeffery conditions.*' They found that they could only synthesise medium sized rings in very low yield with alkenes but that large rings, such as the 21-membered 121, could be prepared in good yield by this method.The large rings were formed almost exclusively in an endo fashion. The cyclisations with allenes proved to be very successful and 7- through 20-membered rings were formed, on the whole, in good to very good yields under high dilution Jeffcry conditions. For example, the 20-membered ring 123 was formed in 86% yield from allene 122. It was shown that carbon-carbon bond formation takes place at the central carbon atom of an allene to form a n-allylpalladium intermediate which can be trapped with a variety of nucleophilcs. Indeed intra- ClzPd(PPh& KzC03, Bu~NCI t DMF, 120 "C, 12 h 6670 ClzPd(PPh& K2CO3, Bu4NCI DMF, 120 "C, 5 h 66% E E 121 E 123 molecular nucleophilic capture can take place to afford another cyclisation.This is exemplified in the cyclisation of allene 124 to n-allylpalladium inter- mediate 125 which is then trapped with piperidine to form eight-membered ring 126. The rate of allene cyclisation was shown to be much faster than alkene cyclisation and it was proposed that it was due to this factor, rather than the lack of competing side reactions, that such good yields of large rings could be obtained. Lastly the geometry of the endocyclic double bond in the newly formed ring was shown to be (2) in eight- and less-membered rings, ( E ) in 1 1 - and more membered rings and to depend on other factors in nine- and ten-membered rings. Thus, the combination of allenes and the intramolecular Heck reaction appears to be a potentially excellent strategy for macrocyclic synthesis. the intramolecular asymmetric Heck reaction has involved the synthesis of carbocycle~.~' The majority of the work on the development of These will Gibson: Tjze intramolecular Heck reaction 4573.1.2 With anion capture Nuss et al. have described an approach to the skeleton of vitamin D3 which combines the intra- an anion capture that introduces the rest of the the vinylpalladium intermediate 134 which can then ClzPd(PPh3)~ II Kzco3, p i d Bu~NCI p i n DMF E molecular Heck cyclisation to form the A-ring with 66% E 80°C,Qh pi PtJ E 124 125 m01ecule."~ Cyclisation of the alkyne 133 leads to piperidine I E 126 undergocapture by the vinyltin reagent to give the product 135.TBSO- be discussed in detail later.Several total syntheses have employed the intramolecular Heck mediated formation of carbocycles as a key step: the synthesis of ( f )-y-apopicropodophyllin,'06 formed by a 6-end0 cyclisation; the synthesis of optically pure opioids,'"' via 6-a0 cyclisations; ( & )-duocarmycin,'"' via a 6-ex0 cyclisation; ( +)-aphidicolin,"l' via a 5-exo cyclisation; ( sation; taxol,"' via an 8-ex0 cyclisation and ( f)-dehydrot~bifoline,"~ via a 6-ex0 cyclisation. Several groups have synthesised variously substi- tuted and protected forms of the vitamin D1 A-ring via an intramolecular Heck 6-ex0 cyclisation."' The A-ring synthon 128 can be obtained by disconnec- tion of lr,25-dihydroxyvitamin D, 127 and the synthesis of 127 from 128 has been described.'14 The synthon 128 can itself be disconnected at the bond joining the two double bonds to reveal a substrate for the intramolecular Heck reaction where either C(5) or C(6) can bear the halide or triflate.Both approaches have been taken: for example, Shimizu under standard conditions with potassium acetate as base, whilst Hatakeyama et al. cyclised vinyl iodide 131 to 132 under similar conditions with triethyla- mine as base.'IY )-cis-trikentrin A,"" via a 5-a0 cycli- cyclised 129 to 130 in very good yield 113a.h 133 134 135 The earliest example of an intramolecular Heck carbocycle synthesis followed by anion capture was reported in 1983 by Heck et al. They cyclised bromodienes and captured the moderately stable n-ally1 palladium intermediate with piperidine thus forming five-membered rings.'I6 Several years later Grigg et al.exploited the tandem cyclisation-anion capture approach not only in the synthesis of heterocycles but also carbocycles. This was initially via 5-ex0 cyclisation onto 1,3-dienes followed by capture of the intermediate n-allylpalladium inter- mediate with either organotin reagents'" or carbon nucleophiles.36 Hydride capture has also been shown to be useful in the synthesis of carbocycles containing quaternary carbon centres.'I7 halides onto a range of terminally substituted alkynes followed by capture with phenylzinc chloride.Its They found that aryl iodides invariably gave better results than aryl bromides and that substitution on the alkyne can have a significant effect on the yield with the trimethylsilylated substrate giving only moderate yields.Thus 136- 138 were cyclised to 139-141 in the yields indicated. Wang et al. described the Heck cyclisation of aryl Y = halide or Off; 2 = H Y = H; 2 = halide or OTf n 127 128 roH C02Et Pd(PPh314 Et-N. MeCN 1\1, - Pd(0Ac)Z. PPh3 KzCO3, MeCN reflux TBSO'. &TBS 86% OT I c r r I -. -. 129 130 131 132 458 Contemporary Organic SynthesisI z 60% 68% 46% Negishi et al. studied the efficiency of various metals for the introduction of a range of organic functional groups in the anion capture process."9 As has been noted before, the main competing reaction in this process is anion capture of the initial inter- mediate formed by oxidative addition of palladium(0). The cyclisation reaction needs to occur significantly faster than this competing reaction for high yields of the desired product to be obtained.Alkynes (or even allenes) are therefore a better substrate for carbopalladation than alkenes and the vinylpalladium species thus formed is always stable enough to undergo anion capture. Negishi et al. therefore captured the product of an intra- molecular 5-exo cyclisation of an aryl iodide onto an alkyne with alkenyl-, alkynyl- and aryl-metals. They found that tin was good for introducing either alkenyl or alkynyl groups but that zirconocene chlor- ides were the best for alkenyls. They also found that the best metal for the introduction of aryl groups was aluminium and that organozincs reacted too fast and therefore gave too high a yield of the competing premature capture product. This methodology has been used in the construc- tion of the ene-yne system found in the neocarzinos- tatin chromophore. Torii et al.reported that cyclisation of the geminally dibrominated alkene 142 onto the proximate alkyne gave the vinylpalladium intermediate 143 which then underwent anion capture with the alkynylstannane to give 144 in reasonable yield.12" Palladium catalysed cross- coupling of 144 with another alkyne gave conjugated ene-yne 145. Thus two different acetylenic append- ages may be introduced by this technique. Nuss et al. found that with 1,l-diiodoalkenes the two possible organostannane coupling steps may occur in a one-pot reaction."' For example when diiodoalkene 146 was reacted in THF with 5 equiva- B n O e O T H P 144, 51 % \PdCI*(PPh&, CUI 145 HO I 1 46 Pd(PPh&, THF HO' 1 47 lents of the alkynylstannane then 147 could be formed in 32% yield.The earliest examples of the anion capture process involved, as mentioned previously, the trapping o f a n-ally1 intermediate by a secondary amine. In the last few years it has been shown that the formation of the intermediate n-ally1 complex is regiospecific and thus the product may be formed regioselectively.'*' Thus vinyl bromide 148 cyclises to 149 which then rearranges to 150. Intramolecular nucleophilic attack by the sulfonamide then generates 151 in good yield. Spirocyclic products may also be formed by this method. It is not only sulfonamides that can be used as nucleophiles. Carbon-centred nucleophiles derived from alkyl sulfonesl" and sc-substituted malonates have also been successfully intramolecular capture of n-allylpalladium inter- mediates formed by the intramolecular Heck reaction in their synthesis of morphine.'*5 In this case the internal nucleophile was a hydroxy group.Overman et a f . have used BrPd, 148 149 I 151 150 Lastly, a palladium catalysed rearrangement followed by an intramolecular Heck reaction has been reported by Watson et al. The palladium catalysed reaction of alkenyl ally1 ether 152 was expected to provide the product of a 7-ex0 cyclisa- tion, but in fact a palladium(0) catalysed 1,3-allyl shift occurred followed by a 5-exo cyclisation to give the spirocyciic product 153. I 152 153 Gibson: The intramolecular Heck reaction 4593.2 Multiple cyclisations The area of palladium catalysed cascade cyclisations is a relatively new and exciting one whereby, with the correctly tailored substrate, several rings may be formed in one reaction under diastereo- and, poten- tially, enantio-c~ntrol.'~~ Much of the initial work on multiple Heck cyclisations was performed in the area of heterocycle synthesis.In 1988, however, Overman et al. reported a series of bis-cyclisations from aryl iodides which formed spiro, fused and bridged ring compounds in good yields using silver carbonate as the base.'28 This was followed, in the next year, by a report of the use of vinyl triflates under standard Heck conditions in bis-cyclisations to form spiro carbocycles.'2y For example, vinyl triflate 154 was cyclised via alkylpalladium inter- mediate 155 to the tricyclic product 156 in good yield.Investigations into performing the cyclisation enantioselectively were also made. 72% 154 155 156 Negishi et al. used alkynes as the relay species with P-hydride elimination to form an alkene as the terminating step in a cascade cyclisation."" For example, the tricyclisation of vinyl iodide 157 to 158 was performed in excellent yield under standard Heck conditions.'31 As had been shown earlier in the synthesis of heterocycles, the 'neopenty1'-palladate intermediate is highly reactive and so, if the palladium is y to a double bond then a cyclopropane ring can be formed. Grigg et al. have demonstrated that this is the case in carbocyclic synthesis as well.85 MGN, reflux, 4 h 95% 1 57 158 De Meijere et al. have shown that palladium catalysed multiple cyclisations followed by a pericyclic reaction can be used in the construction of fused heterocyclic systems.89 Only moderate yields were obtained, however, and they propose that this was due to the presence of the oxygen atom as, in purely carbocyclic systems, very clean reactions may be obtained.132 The pericyclic reaction can take the form of either a 6n electrocyclic rearrangement to form a cyclohexadiene, or an intramolecular Diels-Alder to form a strained ring system.In a typical example, 159 underwent Pd(OAC)p, PPh3 KzCO3, MCN 130°C, 1 day Et02C E tO2C 159 160 Et02C C02Et 47% 161 palladium catalysed cyclisation to cislirans 160 and at the higher temperature of 130 "C trans 160 under- went a Diels-Alder cyclisation to give tetracyclic product 161.'33 cyclisations to the formation of pentacycles with carbonylative esterification as the termination step.134 As carbon monoxide insertion into a carbon-palladium rs bond and subsequent acyl palladation is a possibility in the presence of carbon monoxide then the relay species must be alkynes. This is because carbopalladation of alkynes occurs more quickly than insertion of carbon monoxide which occurs more quickly than carbopalladation of alkenes. Nucleophilic capture of the terminating acylpalladium species can be either intermolecular with, for instance, methanol or intramolecular with the use of a pendant hydroxy function to form a lactone.For example, vinyl iodide 162 was cyclised in good yield to the pentacyclic lactone 163. Negishi et al. have extended the area of multiple J>Bu ClpPd(PPh3)pp EtaN CO (1.1 atm), MeOH 7OOC.1 day * coyo It is also possible to form benzene rings through the use of multiple Heck cross-couplings.This can be done by either a mix of inter- and intra- molecular s t e p ~ ' ~ ' . ' ~ ~ or in a completely intra-molec- ular fashion.'36 For example, vinyl bromide 164 undergoes a Heck cyclisation to form a vinyl- palladium intermediate which then undergoes an intermolecular cross coupling with an alkyne 165 to form another vinyl palladium intermediate 166. This intermediate then undergoes either electrocyclic ring closure and palladium hydride elimination or 6-end0 carbopalladation to give the benzene deriva- 460 Contemporary Organic Synthesis\' DMF, 63% 164 Et2(HO)C = -.$ Et02C T? C02Et 1 68 C(OH)Et2 I Pd(OAC)z, PPh3 me K2C03, MeCN 120 "C, 67% Et02C C02Et 169 tive 167.This occurs with high regioselectivity.'3" In an intramolecular version enediyne 168 is cyclised to bisannelated benzene derivative 169 in good yield.'36 Finally, Overman's group have shown that the palladium catalysed cyclisation approach to the formation of carbocycles can be used successfully in the synthesis of natural products such as the synthesis of the scopadulcic acids.'37 4 Regio- and stereocontrol 4.1 Mode of cyclisation - endo or exo? There are several problems commonly associated with the Heck reaction, many of which have been at least partially overcome in the past few years. For example, reaction substrates were originally limited to aryl or vinyl bromides or iodides but more recently it was discovered that the often more accessible aryl or vinyl triflates may also bc used.138 Also, the previous general necessity for high reaction temperatures is now sometimes avoidable through the use of the Jeffery conditions.The problem of regioselectivity in the carbopalladation step is another area to which much attention has been paid as this is one of the most fundamentally important aspects of the Heck reaction. Before discussing the intramolecular reaction it is important to establish the factors influencing regio- selectivity in the intermolecular version. The Cabri- Hay ashi modelh". 1 ,139 for the coordination and carbopalladation steps will be very helpful to these discussions. (It is of note that the carbopalladation step is presented as irreversible in the Cabri- Hayashi model and that kinetic studies seem to provide some support for this hypothesis.'40) After oxidative addition to give complex 170 the reaction may take either Path A or B and the factor determining which occurs is the nature of 'X' in complex 170. If palladium undergoes oxidative addition into an aryl or vinyl halide bond then 'X' is a halide (usually either bromide or iodide) and the reaction proceeds down Path A. In order for the reacting alkene to coordinate to the palladium, then, due to the strong nature of the palladium- 166 1 67 > 98% regioselectivity /'- \ L, .L P4 I R X 170 H A 171 R Y 172 R halide bond, another ligand must dissociate as in 171. This may be a phosphine or a solvent molecule or if a chelating bisphosphine is present then one of its phosphorus atoms must dissociate before alkene coordination can take place.If 'X' is a triflate then Path B is followed. As the triflate group is only weakly coordinated to the palladium, it can dissociate to form the cationic complex 172 to which the alkene can then coordinate. In a simple reaction like that of an aryl iodide with an x-substituted alkene such as 174 then the reaction follows Path A and the alkene coordinates to give the neutral complex 171. In this case steric considerations dominate and the new carbon- carbon bond tends to be formed at the least substi- tuted end of the alkene to give disubstituted alkene 175. In the reaction of an aryl triflate with 174 the reaction follows path B to give cationic complex 172.The cationic nature of the palladium increases the polarisation of the double bond and so electronic factors dominate the carbopalladation step and the aryl group is transferred to the end of the alkene with the lowest electron density to give product 173. Thus it can be seen that the regio- selectivity of the intermolecular Heck reaction may be influenced by the choice of substrate for the reaction. &o",rOTf.Pdo doH ArI, Pdo L O H 'Ar Path B Path A / electronic steric Ar 175 173 control 174 control In the intramolecular version conformational constraints also exist, as well as the steric and electronic constraints already considered. The alkene should ideally be coplanar with the palladium and the carbon atom of the aryl or vinyl species, i.e.in an eclipsed conformation as Gibson: The intramolecular Heck reaction 46 1illustrated by 176 as opposed to a twisted conforma- tion 177. This is borne out by observation of inter- molecular reaction^'^' and Overman et al. have also provided evidence to support this theory in their synthesis of the Amaryllidaceae alkaloids where the eclipsed and twisted conformations would lead to different diastereomers of the p r o d ~ c t . ~ ~ ~ ~ " ~ Pd eclipsed 176 twisted 177 Accessing an eclipsed conformation in an inter- molecular reaction is not normally a problem and it is thus only the orientation of the alkene, which is governed by steric and/or electronic considerations, that determines the regioselectivity of the carbo- palladation step. In an intramolecular reaction, however, the alkene may be limited to only one eclipsed conformation due to the size and/or shape of the substrate.In this case only one of the possible regioisomers will be formed. In the formation of large rings via the intra- molecular Heck reaction, only endo selectivity has been observed to date.77*x0.82 Due to the large and flexible nature of the substrate very little conforma- tional constraint is present and so both possible eclipsed conformations of the alkene are accessible. The reaction therefore behaves in a manner similar to an intermolecular reaction. As all cases reported to date have employed Path A conditions, steric factors have dominated and endo cyclised products have been formed. As we move into the area of medium sized rings, there may still be a degree of flexibility in the substrate chain but now the conformational constraints become more important than they were for the larger rings and therefore it may prove energetically more difficult to adopt the alkene conformation favoured on steric or electronic grounds alone.A fine balance of steric, electronic and conformational influences may in fact lead to mixture of products. For example, in their synthesis of the conformationally constrained tryptophan derivatives,'04 Roberts et al. isolated a 2 : 1 mixture of the products 178 and 179/180 derived from 8-endo and 7-exo cyclisations respectively. For the synthesis of five-, six- and seven- membered rings, the reaction almost always occurs via the exo mode of cyclisation as conformational considerations now outweigh any steric or electronic factors in the system.It is possible, however, to construct substrates for the Heck reaction which give only the 6-endo cyclisation. For example, Hegedus et a/. observed exclusive 6-endo cyclisation of substrates such as 112 in their approach to the synthesis of ergot alkaloid^.'^ Other similar examples were reported by Black et al." in their cyclisations of 7-bromo-N-allylindoles. For example, 2 x 5 mol% Pd(OAc), 13 mol% (@TOI)~P H D H Et3N (2 equiv.), MeCN, 85 "C 2 x 3 h. 88% flNHCO2BZ C02Me +*.. H &j H 179 180 178: (179/180) = 2 : 1 cyclisation of the N-crotyl derivative 181 gave the tricyclic compound 182 in 94% yield. This particular preference for the endo cyclisation has been investi- gated more fully by Dankwardt et a1.2' who have shown that this selectivity occurs when the alkene component is an ally1 substituent on a ring fused to the aromatic ring bearing the halide and where the halide and the alkene are peri to each other as in 183.They have also shown that this high selectivity exists only when the ring fused to the aromatic ring is five-membered; with six-membered systems a mixture of ex0 and endo cyclised products is obtained. ?Me Ph J$& 43rn0l%Pd(OAc)~ Ph 73 rnol% (O-TOI)~P Me0 - EtsN, MeCN, 100 "C Ph ' N I Br \I 15h,94% Me0 181 \ 1\/1 182 p Br I Po 183 Although the structure of the substrate frequently dictates the regioselectivity of the intramolecular Heck reaction, it appears that sometimes the reaction conditions can be used to favour one mode of cyclisation or the other.For example, a study of the conversion of 184 to 185 and 186 suggests that small changes in catalyst concentration and base can lead to a reversal of regioseIectivity.3'"'" In one reaction (i), a 2: 1 :20 ratio of triphenylphosphine to palladium acetate to substrate with 2 equivalents of potassium carbonate gave a 2.5 : 1 ratio of exo:endo cyclised products. whilst, in a second reaction (ii), a 462 Contemporury Organic Synthesis5-eXO 6-endo -' ' M x h Me COPh i or ii 185 186 184 i. 5 mol% Pd(OAc)*, 10 mol% PPh3 2.5 I K2CO3 (2 equiv.), MeCN, 80 "C, 35 h, 73% ii. 10 mol% P~(OAC)~, 20 mol% PPh3 1 2.5 Et3N (2 equiv.), MeCN, 80 "C, 48 h, 81 % 2: 1 : 10 ratio of phosphine to palladium to substrate with 2 equivalents of triethylamine gave a 1 : 2.5 ratio of exo:endo products.Similarly, it has recently been shown that, for certain substrates, the use of the Jeffery conditions can lead to the formation of endo cyclised products whereas standard conditions yield ex0 cyclised products.s8h Thus aryl iodide 187 forms the seven-membered ring 188 under Jeffery conditions and the six-membered ring 189 under standard Heck conditions. XPd 1 90 C02Et m C 0 2 E t "0 191 t: i + Pd tI rn.m I1 r. C02Et C02Et t: Pd + I X Jeffery conditions NHCy Me0 Me0 standard conditions 6-ex0 32% 189 The regioselectivity of the intramolecular Heck reaction is thus still an area with many uncertainties. Although the structure of the substrate frequently dictates the regioselectivity, it is evident that when two competing pathways are energetically similar then the conditions may be altered to favour one mode of cyclisation over the other.Further rigorous studies are required before a clear understanding of the effects of any given variable will emerge. 4.2 Control of isomerisation and the rble of additives The question of regioselectivity arises not only in the carbopalladation step, as discussed above, but also in the P-hydride elimination step. Here the direction of elimination may be a problem as illus- trated by the potential conversion of intermediate 190 to either 191 or 192. Moreover, the palladium hydride formed from B-elimination may subse- quently coordinate to the newly-formed double bond to give 193, add again to give another alkyl- palladium intermediate 194, and re-eliminate with the formation of a different double bond.Repeti- tion of this process may lead to a variety of different products. The first problem is not as serious as the second as there is generally a preferred direction of elimination. It has been found, however, that both problems can be dealt with to a greater or lesser extent through the use of either silverI4' or thallium'33 salts. They are used to sequester halide ions from the palladium complex 170 formed in the initial oxida- tive addition step, thus forming a series of cationic palladium complexes. Due to its cationic nature the palladium not only coordinates the alkene more rapidly, hence partly explaining the rate enhancing effect of these salts, but it also effectively undergoes more selective P-hydride elimination.This may be explained by faster p-elimination and/or a shorter lifetime of the hydridopalladium species thus formed. The first reported use of silver salts to control isomerisation in the intramolecular Heck reaction was by Overman et aLz3 Since then their use has become more widespread in the synthesis of both hetero- and carbo-cycles in m ~ n o - ~ ~ " " ~ ~ and poly- c y c ~ ~ s a ~ ~ o n s ~ X ~ . l ~ ~ . 1 " 6 . ~ 3 7 A good example of their use is in the control of isomerisation in the synthesis of 3-meth~leneindolines.~ Without silver salts the cycli- sation of 195 would form substituted indole 197 via initial elimination followed by re-addition of the hydridopalladium species and re-elimination to bring the double bond into conjugation with the aromatic ring.With the addition of silver carbonate, however, exclusive formation of the desired exo- methyleneindoline 196 was achieved. Gibson: The intramolecular Heck reaction 4633% Pd(OAc)2,6% PPh3 d( S02Ph a:T Ag2C03 (2 equiv.), DMF * S02Ph ~ t . . 5 h, 80% 1 95 196 ($ 0% S02Ph 197 Denmark et al. found that not only were silver salts necessary for good yields of the desired products from intramolecular Heck arylations of nitroalkenes, but also that the nature of the anion was very important.74 For instance, silver carbonate worked well but silver nitrate gave none of the desired product. The importance of the anion has also been noted in asymmetric cyclisations but, as yet, there is little understanding of its r6le. The use of silver salts with vinyl triflates has led to decom- position of the starting material."" (However, silver salts should not be necessary in such cases, as when triflate is used as the leaving group then the reaction should follow Path B of the Cabri-Hayashi model to form a cationic palladium complex anyway.) Thallium salts have been found to have a similar effect to silver salts and have been exploited in the intramolecular Heck reaction, predominantly by Grigg, in nion~cyclisations,~~~~~~~'~~~~~~ and polycyclisa- Tietze et al.have found that ally1 silanes may be used to control isomerisation.j3 Under Jeffery condi- tions the silyl group eliminates, but with the addition of silver oxide P-hydride elimination occurs to yield vinyl silanes as in the cyclisation of 198 to 199.t ions.29.S7.8S L S i M e , I R Jeffery conditions, R = H 198 with Ag20, R = SiMe3 199 Finally, the addition of water to the reaction mixture when a phase transfer agent is has been found to accelerate the reactionSs and in one case its presence was essential for the reaction to proceed."" 4.3 Unexpected products Occasionally, the products formed from a Heck reaction do not fit the accepted mechanism. Often there is a reasonably obvious explanation but sometimes the proposed explanation raises as many questions as it answers. One of the very earliest examples of an unexpected product involved an attempted cyclisa- tion of anilide 200 to quinolone 201 by Heck et al.'" The isomeric quinolone actually formed, 204, was clearly not produced via the usual mechanism and so an alternative mechanism was proposed. This relied on the preferential formation of the 5-exo intermediate 202. Lacking P-hydrogen atoms, the palladium undergoes a P-carbonyl elimination to form an acylpalladium intermediate 203 which then cyclises via a 6-endo process to form 204.H H 200 201 202 \ / the 203 An interesting rearrangement was discovered by Rawal et al. in their efforts directed towards the synthesis of Strychnos alka10ids.l~~ At first sight, the cyclisation of 205 appeared to have proceeded via a 7-endo route to give 206 in pxeference to the 6-ex0 product, but on further inspection, the geometry of the exocyclic double bond is seen to be incorrect for such a cyclisation. It was therefore poposed that the cyclisation had occurred via the 6-ex0 mode but that the alkylpalladium intermediate thus formed had not undergone the expected /I-hydride elimination but instead formed a six-membered palladocycle with the neighbouring carbamate group'46 to give 207.This intermediate could then rearrange with the aid of the pendant exocyclic double bond via the cyclopropylcarbinylpalladate intermediate 208 to form the alkylpalladium intermediate 209 and the observed product 206. Examples of homoallyl- palladate-cyclopropylcarbinylpalladate rearrange- m e n t ~ ' ' ~ have also been observed by Negishi et al. during the course of studies on the cyclisations of vinyl halides onto geminally disubstituted terminal a1 kenes. ')7 sation should result in an alkylpalladium inter- mediate lacking a syn P-hydrogen atom for elimination and yet elimination still occurs.The reasons for this are varied. In cyclisations onto alkenes coordinated to carbonyl groups then oxo- n-allylpalladium intermediates have been through which the palladium can orient itself syrz to a P-hydrogen. Where the alkyl- palladium intermediate has palladium bound to a In several other cases the apparent mode of cycli- i nvoked 1 Ojg.7 lh.94U 464 Contemporay Organic SynthesisMe02C I Me026 H 205 1 t 206 In a recent example, 5-exo cyclisation of 213 gave 214 when stoichiometric amounts of palladium acetate and triphenylphosphine were used with a large excess of triethylamine in refluxing THF.’” It was proposed that the alkylpalladium intermediate underwent intramolecular oxidative addition into the benzylic carbon-hydrogen bond to form 215 which then underwent sequential reductive elimina- tion and P-hydride elimination of the palladium.Pd(OAC)P, PPh3 EtsN,THF * reflux, 10 h PdI Me CJ2Me 207 \ Me02C / 209 Me02C 208 benzylic carbon then ‘stereomutation’ has been proposed to allow syn P-hydride eliminati~n.””.~~ In syntheses of lycoricidine63.3,64 6-ex0 cyclisation is apparently followed by anti P-hydride elimination to give the desired product and so an intermediate such as 210 is proposed which would lead to the product via a reductive elimination step. E O M O M NMPM 210 MPM = CH2C6H4(pOMe) Danishefsky et al. have reported the formation of an aldehyde via cyclisation onto an enol ether.”” They propose that the elimination of palladium from 211 occurs via an ‘Arbuzov-like unravelling’ to give aldehyde 212.OMe CHO 21 1 21 2 6i Br 21 5 Lastly an example of what appears 21 4 // to be syn dealkoxypalladation has been reported.55 Although for palladium( 11) this is unprecedented, an example of syn elimination of palladium(0) from a p-hydroxy organopalladium intermediate is known.’4x 3.4 Diastereocontrol The intramolecular Heck reaction is often highly diastereoselective as has been shown in Overman’s elegant synthesis of scopadulcic acid The use of chiral auxiliaries in providing stereocontrol, however, has not received much attention. Grigg et al. have shown that amino ethers may be effective chiral auxiliaries for the Heck reaction.’8 Thus, in the cyclisation of iodoarene 216 onto the pendant cycloalkene, no diastereoselectivity was observed for the two chiral auxiliaries indicated when achiral phosphines were used, but when (S)-BINAP was used diastereoselectivities of between 48 and 55% were achieved.When the acyclic alkene 217 bearing SAMP (or RAMP) as the chiral auxiliary was 21 6 0 with (S )-BINAP when R = SAMP, de = 48% when R = CH(Ph)CH20Me, de = 55% 0 OMe de > 95% Gibson: The iritramolecular Heck reaction 465subjected to standard cyclisation conditions, diaster- eoselectivities of 95% were achieved with achiral p hosp hines. In the cyclisation of dienyne 218 carried out by de Meijere et al. 133 the stereoselectivity of the final electrocyclisation step is controlled by the pendant chiral ether to give a diastereomeric excess of > 95%. yh lO%Pd(OAC)p ph 20% PPh3 Et02C EQC OMe 21 8 Overman et al.have reported an example where the diastereoselectivity of an intramolecular Heck reaction was controlled by the conditions used.24h Thus, when triethylamine was used as the base in the cyclisation of aryl bromide 219, an 89: 11 ratio of 220 to 221 was formed in very good yield. With silver phosphate as the base, however, a 3:97 ratio of 220 to 221 was formed in good yield. Their proposal to account for this reversal in stereose- lectivity was that the cationic palladium inter- mediate formed by the action of the silver salts can coordinate to both alkenes present in the substrate to form intermediate 222 and therefore carbopalla- dation is directed in this case onto the upper face of the alkene. SEM \\ O v N , .\ O A k J i - 4 21 9 Dr Br’ 222 4.5 Enantiocontrol I TNSE5 0 Me02C’ w:‘“ Br’ 221 I The area of the asymmetric intramolecular Heck reaction has been explored mainly by the Shibasaki group. They have shown that prochiral vinyl iodides and vinyl triflates can be cyclised to products with high optical purity in the presence of enantiopure phosphines. For vinyl iodides the addition of silver salts is important to obtain a good eelo5” as would be expected from the Cabri-Hayashi model.When a chiral bisphosphine is used, obviously the highest degree of influence over the enantioselectivity of the reaction is exerted when both phosphorus atoms are bound to the palladium atom. For this to be the case when the palladium coordinates to the alkene, the reaction must proceed along Path B. It was also found that the nature of the silver anion was very important and that silver phosphate gave the best results.The best phosphine for this reaction has been found to be the BINAP ligand and better ees were obtained when the preformed palladium dichloride- BINAP catalyst was The best solvents were polar and aprotic and N-methylpyrrolidinone was found to give consistently good results. The use of calcium carbonate as an additional base was also beneficial.1o5e Under these conditions vinyl iodide 223 was cyclised to 224 in good yield and with a good ee.Iosd Tris(dibenzy1ideneacetone)bispalladium has also been found to be a good catalyst for asymmetric Heck reactions of vinyl iodides. Experi- mentation with various ligands and salts often proves beneficial as in the synthesis of indolizi- dines48 where BPPFOF { 1-[ 1 ‘,2-bis(diphenyl- phosphino)ferrocenyl]ethanol} was found to be the best chiral ligand and silver-zeolite the best source of silver ions.,OTBDMS ,OTBDMS 10% PdCld(R )-binap] CaC03 (2.2 equiv.) NMP, 60 “C, 100 h 87% 0e OAc 67% OAc 223 224 It has also been found that the use of vinyl triflates, which obviates the need for silver salts, gives good enantioselectivities with BINAP as the ligand.lo5‘ When used with a phase transfer agent such as tetrabutylammonium acetate then the cycli- sation can be followed by anion capture with the acetate anion as in the conversion of 225 to 226.’05c6 It has been found that nonpolar solvents give the best results when triflates are used’05g and under such conditions, with THF as the solvent and potas- sium carbonate as the base, asymmetric quaternary centres can be formed as in the cyclisation of aryl triflate 227 to tetrahydronaphthalene 228.105”.i*k 1.7% Pd(OAc)p, 2.1% (3-BINAP BulNOAc (1.7 equiv.), DMSO Me 20 “C, 2.5 h, 89% Me 80% ee 225 226 3% Pd~(dba)~ K2C03, THF 8Yo (R)-BINAP m 9 227 50 “C, 336 h.97% OTBDPS 466 Contemporary Organic SynthesisThe solvent that gave the highest ees with triflates was 1,2-dichloroethane but under the standard range of conditions only a poor conversion of starting material was possible. This conversion was improved, however, upon the addition of alcohols, the most effective being pinacol or the acetate anion."" It appears that Pdo is oxidised readily in 1,2-dichloroethane to Pd"C12L, and that the addition of pinacol or potassium acetate prevents this process.Thus a new set of conditions have been developed for the intramolecular asymmetric Heck reaction of triflates: namely, palladium acetate as the catalyst, BINAP as the chiral ligand, two equiva- lents of potassium carbonate as the base together with either 15 equivalents of pinacol or one equiva- lent of potassium acetate in 1,2-dichloroethane. Overman et al. have obtained extremely interesting results in their syntheses of indolones with asymmetric induction.2' With the same enantiomer of a chiral ligand (BINAP) they have obtained either enantiomer of the cyclic product derived from 229 by varying the conditions used. With silver phosphate as the base and dimethylace- tamide as solvent both significant ee's and yields of the (S)-indolone can be obtained, whilst with 1,2,2,6,6-pentamethylpiperidine and no silver salt present the (R)-indolone can also be formed in significant ee.Overman et al. have also carried out a bis-cyclisation with moderate enantio~electivity,"~ but although enantioselective polycyclisations are in principle a very powerful technique, this area is still largely unexplored. 5% Pd2(dba)3 (9-(+) 71%ee I 81% 229 10% Pd2(dba)3 c 20% (R)-BINAP 0 MeCONMe2, 80 "C, 140 h Me Me Me Tietze et al. have combined asymmetric conditions with their use of ally1 silanes for controlling the regioselectivity of the elimination step to create an asymmetric intramolecular Heck reaction of use in their total synthesis of a norsesquiterpene."'.1"5' In the key step, aryl iodide 230 was cyclised to 231 in excellent yield and enan tioselectivi ty.The asymmetric intramolecular Heck reaction is thus fast becoming a very useful and reliable 2.5% Pd2(dba)3.CHC13 91% Me 92% ee 230 231 catalytic asymmetric carbon-carbon bond forming reaction. 5 Total syntheses In the last three years the intramolecular Heck reaction has been widely used in the total synthesis of natural products with the ring generated being both hetero~yclic~~,~'.~~-~~.~~' and carbocy- described which demonstrate the versatility of the intramolecular Heck reaction. The total synthesis of taxol 233 has been described by Danishefsky et al. in a series of papers."' The key step in this synthesis was the closure of the eight-membered ring via an intra- molecular Heck reaction of the vinyl triflate 232.The closure was achieved in 49% yield under fairly standard conditions and clearly demonstrated the utility of the reaction for the cyclisations of highly functionalised substrates. In this section, four examples are clic.43.1U5.1 12.125.127h.137 90"C,49% WH 'r c A/ 1- 1 '' 232 0 OH 232 BzHN u 0 233 The synthesis of morphine 237 by Overman et al. has as its key step an intramolecular Heck reaction followed by an intramolecular anion capture process.'25 Thus, aryl iodide 234 undergoes a stereo- selective Heck cyclisation onto a 1,3-diene to form the n-allylpalladium intermediate 235 which is then captured by an appropriately positioned hydroxy group to give the pentacyclic product 236. has been demonstrated by the synthesis of scopa- dulcic acid A 240 by Overman et al.synthesis vinyl iodide 238 undergoes regioselective bis-cyclisation to give tricyclic 239. The synthesis of (+)-vernolepin 243 by Shibaski et al. 1('5' corresponds to the first asymmetric total synthesis of this molecule and demonstrates the usefulness of the asymmetric intramolecular Heck reaction. Using the conditions described earlier The usefulness of the polycyclisation methodology In this Gibson: The intramolecular Heck reaction 467235 -PdI 237 236 TBDMSO 238 \ A / 239 OH HO 240 5% Pd(OAc)p, 10% (R)-BINAP K2CO3 (2 equiv.), KOAC (1 equiv.) CICH2CH2Ci. 60 "C. 41 h H 70% TfO 241 86%ee 242 / 0 -+y 0 243 (potassium acetate as the additive and 1,2-dichloro- ethane as the solvent) vinyltriflate 241 was cyclised to bicyclic product 242 in 70% yield and 86% ee.6 Conclusion In the past few years, the intramolecular Heck reaction has started to reveal its full potential as a powerful tool for the synthetic chemist interested in constructing heterocyclic and carbocyclic compounds. Together with its modifications which promote anion capture processes, multiple cyclisa- tions and asymmetric syntheses, and its tolerance of a wide range of functional groups, this reaction is surely now one of the most powerful available to the synthetic organic chemist. 7 References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 ( a ) T. Mizoroki, K. Mori and A. Ozaki, Bull. Chem. SOC. Jpn., 1971, 44, 581; (6) R. F. Heck and J. P. Nolley, Jr., J. OR. Chem., 1972, 37, 2320. ( a ) R. F. Heck,Acc. Chem.Rex, 1979, 12, 146; (b) R.F. Heck, Org. React., 1982, 27, 345; (c) J. Tsuji, Organic Synthesis with Palladium Compounds, Springer, Berlin, 1980; ( d ) B. M. Trost and T. R. Verhoeven, in Comprehensive Organometallic Chemistry, Pergamon, Oxford, 1982, vol. 8, p. 799. ( a ) G. P. Chiusoli, Pure Appl. Chem., 1980, 52, 635; (b) C.-M. Andersson and A. Hallberg, Tetrahedron Lett., 1987, 28, 4215. ( a ) G.-Z. Wu, F. Lamaty and E. Negishi, J. (2%. Chem., 1989, 54, 2507; (b) R. Grigg, S. Sukirtha- lingam and V. Sridharan, Tetrahedron Lett., 1991, 32, 2545. (u) T. Jeffery, J. Chem. Soc., Chem. Commun., 1984, 1287; (b) T. Jcffery, Tetrahedron Lett., 1985, 26, 2667; (c) T. Jeffery, Synthesis, 1987, 70. ( a ) G. D. Daves, Jr. and A. Hallberg, Chem. Rev., 1989, 89, 1433: (6) R.F. Heck, in Compretzensive Organic Synthesis, Pergamon, Oxford, 1991, vol. 4, p. 833; ( c ) A. de Meijere and F. E. Meyer,Angew. Chem., Int. Ed. Engl., 1994, 33, 2379; ( d ) W. Cabri and I. Candiani, Acc. Chem. Res., 1995, 28, 2; ( e ) B. C. Sodcrberg in Comprehensive Organometallic Chemistty II, Pergamon, Oxford, 1995, vol. 12, p. 259. M. Mori, K. Chiba and Y. Ban, Tetrahedron Lett., 1977, 1037. T. Sakamoto, Y. Kondo, M. Uchiyama and H. Yama- naka, J. Cjiem. SOC., Perkin Trans. 1, 1993, 1941. M. Mori and Y. Ban, Etrahedron Lett., 1979, 1133. M. 0. Terpko and R. F. Heck, J. Am. Chem. SUC., 1979, 101, 5281. R. Odle, B. Blevins, M. Ratcliff and L. S. Hegedus, J. 0%. Chem., 1980, 45, 2709. H. A. Dieck and R. F. Heck, J. Am. Chem. Soc., 1974, 96, 1133.A. Kasahara, T. Izumi, S. Murakami, H. Yanai and M. Takatori, Bull. Chern. Suc. Jpn., 1986, 59, 927. J .P. Michael, S.-F. Chang and C. Wilson, Tetrahedron Lett., 1993, 34, 8365. L. S. Hegedus, T. A. Mulhcrn and A. Mori,J. 0%. Chem., 1985, 50, 4282. T. Sakamoto, T. Nagano, Y. Kondo and H. Yamanaka, Synthesis, 1990, 215. K. Koerber-PI6 and G. Massiot, Synlett, 1994, 759. R.C. Larock and S. Babu, Tetrahedron Lett., 1987, 28, 5291. ( a ) P. Martin, Helv. Chim. Actu, 1989, 72, 1554; (h) R. J. Sundberg and W. J. Pitts, J. OR. Chem., 1991, 56, 3048; (c) A. Arcadi. S. Cacchi, F. Marinclli and P. Pace, Synlert, 1993, 743; ( d ) L. F. Tietze and T. Grote, J. Org. Chem., 1994, 59, 192; (e) D. Wensbro, U. Annby and S. Gronowitz, Tetmhedron, 1995, 51, 10323. 468 Contemporary Organic Synthesis20 J.E. Macor, D. H. Blank, R. J. Pos and K. Ryan, 21 J. W. Dankwardt and L. A. Flippin, J. 0%. Chem., 22 H. Iida, Y. Yuasa and C. Kibayashi, J. 0%. Chem., 23 M. M. Ableman, T. Oh and L. E. Overman, J. 0%. 24 (a) W. G. Earley, T. Oh and L. E. Overman, Tetra- Tetrahedron Lett., 1992, 33, 8011. 1995, 60, 2312. 1980,45, 2938. Chem., 1987, 52,4130. hedron Lett., 1988, 29, 3785; (b) A. Madin and L. E. Overman, Tetrahedron Lett., 1992, 33, 4859. 25 ( a ) A. Ashimori and L. E. Overman, J. 0%. Chem., 1992, 57, 4571; (b) A. Ashimori, T. Matsuura, L. E. Overman and D. J. Poon, J. Org. Chem., 1993,58, 6949. 26 R. Grigg, J. Heterocycl. Chem., 1994, 24, 2139. 27 (a) B. Burns, R. Grigg, V. Sridharan, P. Stevenson and T. Worakun, Tetrahedron Lett., 1988, 29, 4329; (b) B.Burns, R. Grigg, V. Santhakumar, V. Sridharan, P. Stevenson and T. Worakun, Tetrahedron, 1992, 48, 7297. hedron Lett., 1993, 34, 3163. hedron Lett., 1992,33, 7789. P. Stevenson, S. Sukirthalingam and T. Worakun, Tetrahedron Lett., 1988, 29, 5565. P. Stevenson, S. Sukirthalingam and T. Worakun, Tetrahedron Lett., 1989, 30, 11 35. 32 R. Grigg, A. Teasdale and V. Sridharan, Etrahedron Lett., 1991, 32, 3859. 33 F.-T. Luo and R.-T. Wang, Heterocycles, 1991, 32, 2365. 34 (a) J. M. O’Connor, B.J. Stallman, W. G. Clark, A. Y. L. Shu, R. E. Spada, T. M. Stevenson and H. A. Dieck, J. Org. Chem., 1983, 48, 807; (6) E. C. Taylor, A. H. Katz, H. Saldora-Zamora and A. McKillop, Tetrahedron Lett., 1985, 26, 5963; (c) R. C. Larock, N. Berrios-Pena and K. Narayanan, J.0%. Chem., 1990,55, 3447; ( d ) R. C. Larock, Pure Appl. Chem., 1992, 62, 653; (e) L. S. Hegedus, Angew. Chem., Int. Ed. Engl., 1988, 27, 1113. T. Worakun, J. Chem. SOC., Chem. Commun., 1986, 1697. 36 R. Grigg, V. Sridharan, S. Sukirthalingam and T. Worakun, Tetrahedron Lett., 1989, 30, 1139. 37 ( a ) R. Grigg, V. Sridharan, P. Stevenson and S. Sukir- thalingam, Tetrahedron, 1989, 45, 3557; (b) R. Grigg, V. Sridharan, P. Stevenson and S. Sukirthalingam and T. Worakun, Tetrahedron, 1990, 46, 4003. laussavaratana, W. D. J. A. Norbert and V. Sridharan, Tetrahedron Lett., 1990, 31, 3075. haran and A. Teasdale, Tetrahedron Lett., 1991,32, 687. 40 (a) A. P. Kozikowski and D. Ma, Tetrahedron Lett., 1991,32, 3317; (b) G. A. Kraus and H. Kim, Synth. Commun., 1993, 23, 55.41 D. L. Comins, M. F. Baevsky and H. Hong, J. Am. Chem. SOC., 1992, 114, 10971. 42 R. Yoneda, Y. Sakamoto, Y. Oketo, K. Minami, S. Harusawa and T. Kurihara, Tetrahedron Lett., 1994, 28 R. Grigg, V. Santhakumar and V. Sridharan, Tetra- 29 R. Grigg, P. Kennewell and A. J. Teasdale, Tetra- 30 B. Burns, R. Grigg, P. Ratananukul, V. Sridharan, 31 B. Burns, R. Grigg, P. Ratananukul, V. Sridharan, 35 R. Grigg, V. Sridharan, P. Stevenson and 38 R. Grigg, M J. R. Dorrity, J. F. Malone, T. Mongko- 39 R. Grigg, V. Loganathan, V. Santhakumar, V. Srid- 35, 3749. 43 M. Shibasaki and M. Sodeoka, J. Synth. 0%. Chem. Jpn., 1994, 52, 596. 44 J. R. Luly and H. Rapoport, J. OF. Chem., 1984, 49, 1671. 45 A. L. Germain, T. L. Gilchrist and P. D. Kemmitt, Heterocycles, 1994, 37, 697.46 E. Desarbre and J.-Y. MCrour, Heterocycles, 1995, 41, 1987. 47 (a) M. Mori, N. Kanda, I. Oda and Y. Ban, Tetra- hedron, 1985,41,5465; (b) M. Mori, I. Oda and Y. Ban, Tetrahedron Lett., 1982, 23, 5315. Tetrahedron, 1994, 50, 371. R. F. Heck, J. 0%. Chem., 1983, 48, 3894. 29, 4687. 1989, 28, 55. 48 Y. Sato, S. Nukui, M. Sodeoka and M. Shibasaki, 49 L. Shi, C. K. Narula, K.J. Mak, L. Kao, Y. Xu and 50 R. C. Larock and D. E. Stinn, Tetrahedron Lett., 1988, 51 E. Negishi, T. Nguyen and B. O’Connor, Heterocycles, 52 S. Wolff and H. M. R. Hoffmann, Synthesis, 1988, 760. 53 L. F. Tietze and R. Schimpf, Chem. BeK, 1994, 127, 2235. 54 R. Anacardio, A. Arcadi, G. Danniballe and F. Mari- nelli, Synthesis, 1995, 831. 55 J.-F. Nguefack, V. Bolitt and D. Sinou, J.Chem. Soc., Chern. Commun., 1995, 1893. 56 R. C. Larock and N. H. Lee, J. Org. Chem., 1991,56, 6253. s57 N. Arnau, M. Moreno-Manas and R. Pleixats, Tetra- hedron, 1993,49, 11 019. 58 (a) B. Burns, R. Grigg, P. Ratananukul, V. Sridharan, P. Stevenson and T. Worakun, Etrahedron Lett., 1988, 29,4329; (b) J. H. Rigby, R. C. Hughes and M. J. Heeg, J. Am. Chem. SOC., 1995, 117, 7834. 59 D. St. C. Black, P. A. Keller and N. Kumar, Tetra- hedron, 1992, 48, 7601. 60 ( a ) L. F. Tietze and R. Schimpf, Angew. Chem., Int. Ed. Engl., 1994, 33, 1089; (6) L. F. Tietze and 0. Burkhardt, Synthesis, 1994, 1331; (c) L. F. Tietze and 0. Burkhardt, Liebeigs Ann. Chem., 1995, 1153. 61 T. Okita and M. Isobe, Tetrahedron, 1994,50, 11 143. 62 R. Grigg, V. Santhakumar, V. Sridharan, M.Thornton-Pett and A. W. Bridge, Tetrahedron, 1993,49,5177. 63 S. F. Martin and H.-H. Tso, Heterocycles, 1993, 35, 85. 64 ( a ) N. Chida, N. Ohtsuka and S. Ogawa, Etrahedron Lett., 1991,32, 4525; (b) N. Chida, M. Ohtsuka and S. Ogawa, J. 0%. Chem., 1993, 58,4441; (c) T. Hudlicky and H. F. Olivo, J. Am. Chem. SOC., 1992, 114, 9694; ( d ) M. C. McIntosh and S. M. Weinreb, J. 0%. Chem., 1993, 58, 4823. 65 ( a ) P. Melnyk, J. Gasche and C. Thal, Tetrahedron Lett., 1993,34, 5449; (b) P. Melnyk, B. Legrand, J. Gasche, P. Ducrot and C. Thal, Tetrahedron, 1995, 51, 1941. 66 ( a ) B. Burns, R. Grigg, V. Sridharan and T. Worakun, Tetrahedron Lett., 1988, 29, 4325; (b) R. Grigg, V. Loganathan, S. Sukirthalingam and V. Sridharan, Tetrahedron Lett., 1990, 31, 6573. 32, 1695. Chem.Commun., 1987,510; (b) S. A. Ahmad-Junan, P. C. Amos and D. A. Whiting, J. Chem. SOC., Perkin Trans. 1, 1992, 539. J. Am. Chem. SOC., 1990, 112, 6959. hedron: Asymmetry, 1995, 6, 1527. G. Solari, Tetrahedron Lett., 1994,35, 5919; (b) G. K. 67 V. H. Rawal and C. Michoud, Tetrahedron Lett., 1991, 68 ( a ) P. C. Amos and D. A. Whiting, J. Chem. SOC., 69 M. M. Ableman, L. E. Overman and V. D. Tran, 70 B. Wunsch, H. Diekmann and G. Hofner, Tetra- 71 ( a ) M. Catellani, G. P. Chiusoli, M. C. Fagnola and Gibson: The intramolecular Heck reaction 469Friestad and B. P. Branchaud, Tetrahedron Lett., 1995, 36, 7047. 72 F. G. Fang, S. Xie and M. W. Lowery, J. Org. Chem., 1994, 59, 6142. 73 Z . Jin and P. L. Fuchs, Tetrahedron Lett., 1993,34, 5205. 74 S. E. Denmark and M.E. Schnute, J. 0%. Chem., 1995, 60, 1013. 75 ( a ) K. F. McClure and S. Danishefsky, J. Am. Chem. SOC., 1993, 115, 6094; ( h ) K. McClure, S. Danishefsky and G. K. Schulte, J. 0tg Chem., 1994, 59, 355. 76 R. Grigg, V. Sridharan, P. Stevenson, P. Teasdale, M. Thornton-Pett and T. Worakun, Tetrahedron, 1991, 47, 9703. feld, Tetrahedron, 1981, 37, 4035. 77 F. E. Ziegler, U. R. Chakraborty and R. B. Weisen- 78 L. F. Tietze and K. Schimpf, Synthesis, 1993, 876. 79 S. E. Gibson (nke Thomas) and R. J. Middleton, 80 ( a ) S. Ma and E. Negishi, J. Org. Chem., 1994, 59, J. Chern. SOC., Chem. Commun., 1995,1743. 4730; (b) S. Ma and E. Negishi, J. Am. Chem. SOC., 1995, 117, 6345. 1990,55,6028. hedron Lett., 1995, 36, 6555. SOC., 1988, 110, 2328. and S. Sukirthalingam, Tetrahedron Lett., 1990.31, 1343. 85 R. Grigg, V. Sridharan and S. Sukirthalingam, Tetra- hedron Lett., 1991, 32, 3855. 86 R. Grigg, P. Fretwell, C. Meerholtz and V. Sridharan, Tetrahedron, 1994, 50, 359. 87 D. Brown, R. Grigg, V. Sridharan and V. Tambyr- ajah, Tetrahedron Lett., 1995, 36, 8137. 88 R. Grigg and V. Sridharan, Tetrahedron Lett., 1992, 32, 7965. 89 F. E. Meyer, P. J. Parsons and A. de Meijere, J. Org. Chem., 199 1,56, 6487. 90 S. D. Knight and L. E. Overman, Heterocycles, 1994, 39, 497. 91 (a) R. Grigg, P. Stevenson and 1’. Worakun, J. Chem. SOC., Chem. Commun., 1984, 1073; (h) R. Grigg, P. Stevenson and T. Worakun, Tetrahedron, 1988, 44, 2033. hedron Lett., 1988, 29, 2915; (b) Y. Zhang, B. O’Connor and E. Negishi, J. Org. Chem., 1988, 53, 5588. 93 R.C. Larock, H. Song, B. E. Baker and W. H. Gong, 7ktrahedron Lett., 1988, 29, 2919. 94 (a) H. Ishibashi, K. Ito, T. Hirano, M. Tabuchi and M. Ikeda, Tetrahedron, 1993,49, 5471; (b) A. J . Davies, R. J. K. Taylor, D. J. Scopes and A. H. Wadsworth, Bioorg. Med. Chem. Lett., 1992, 2, 481. 95 0. Cornec, B. Joseph and J.-Y. Mkrour, Tetrahedron Lett., 1995, 36, 8587. 96 B. O’Connor, Y. Zhang, E. Negishi, F. T. Luo and J. W. Chang, Etrahedron Lett., 1988, 29, 3903. 97 Z. Owczarczk, F. Lamaty, E. J . Vawter and E. Negishi, J. Am. Chem. SOC., 1992, 114, 10091. 98 L. S. Hegedus, M. R. Sestrick, E. J. Michaelson and P. J. Harrington, J. 0%. Chem., 1989, 54, 4141. 99 J. M. Gaudin, Tetrahedron Lett., 1991, 32, 6113. 81 R. J. Sundberg and R. J. Cherney, J. OR. Chem., 82 M. J.Stocks, R. P. Harrison and S. J. Teague, Tetra- 83 M. M. Abelman and L. E. Overman, J. Am. Chem. 84 R. Grigg, M. J. Dorrity, J. F. Malonc, V. Sridharan 92 ( a ) E. Negishi, Y. Zhang and B. O’Connor, Tetra- 100 S. Ldschat, F. Narjes and L. E. Overman, Tetrahedron, 101 ( a ) G. E. Stokker, Tetruhedron Lett., 1987, 28, 3179; 1994, SO, 347. (b) J. P. Konopolski, K. S. Chu and G. R. Negrete, J. 0%. Chem., 1991, 56, 1355; (c). W. Cabri, 1. Candiani, S. De Bernardianis, F. Francalanci, S. Penco and R. Santi, J. Org. Chem., 1991,56,5796. 1995,51,9139. 1995, 78, 765. Tetrahedron Lett., 1994, 35, 939; (b) D. C. Horwell, P. D. Nichols, G. S. Ratcliffe and E. Roberts, J. Org. Chem., 1994, 59,4418. Chem., 1989,54,4738; (b) Y. Sato, M. Sodeoka and M. Shibasaki, Chem. Lett., 1990, 1953; ( c ) K.Kage- chika and M. Shibasaki, J. Org. Chem., 1991, 56,4093; ( d ) Y. Sato, S. Watanabe and M. Shibasaki, Tefra- hedron Lett., 1992,33, 2589; (e) Y . Sato, T. Honda and M. Shibasaki, Tetrahedron Lett., 1992, 33, 2593; ( f ) K. Kagechika, T. Oshima and M. Shibasaki, Tetra- hedron, 1993,49, 1773; (9) K. Kondo, M. Sodeoka, M. Mori and M. Shibasaki, Synthesis, 1993, 920; (h) T. Takemoto, M. Sodeoka, H. Sasai and M. Shibasaki, J. Am. Chem. SOC., 1993, 115, 8477; (i) K. Ohrai, K. Kondo, M. Sodeoka and M. Shibasaki, J. Am. Chem. Soc., 1994, 116, 11 737; ( j ) K. Kondo, M. Sodeoka and M. Shibasaki, J. Org. Chem., 1995, 60, 4322; ( k ) K. Kondo, M. Sodeoka and M. Shiba- saki, Tetrahedron: Asymmetry, 1995, 6, 2453; (/) L. F. Tietze and T. Raschke, Synlett, 1995, 597.106 14. Ishibashi, K. Ito, M. Tahuchi and M. Ikeda, Heterocycles, 1991, 32, 1279. 107 C. Y. Hong, N. Kado and L. E. Overman, J. Am. Chem. SOC., 1993, 115, 11 028. 108 H. Muratake, I. Abe and M. Natsumc, Tetrahedron Lett., 1994, 35, 2573. 109 (a) M. Toyota, Y. Nishikawa and K. Fukumoto, Tetra- hedron Lett., 1994, 35, 6495; ( b ) M. Toyota, Y. Nishi- kawa and K. Fukumoto, Tetrahedron, 1994, 38, 11 153. 110 P. Wiedenau, B. Monse and S. Blechcrt, Tetrahedron, 1995, 51, 1167. 11 1 (a) J. J. Masters, D. K. Jung, W. G. Bornmann, S. J. Danishefsky and S. Degala, Etrahcdron Lett., 1993, 34, 7253; (6) J.J. Masters, D.K. Jung, S. J. Danish- efsky, L. B. Snyder, J . K. Park, R. C. Isaacs, C. A. Alaimo and W. B. Young, Angew. Chem., Int. Ed. Engl., 1995, 34, 452; (c) J.J. Masters, J. T. Link, L. B. Snyder, W. B.Young and S. J. Danishefsky, Angew. Chem., Int. Ed. Engl., 1995,34, 1723; ( d ) W. B. Young, J.J. Masters and S.J. Danishefsky, J. Am. Chem. Snc., 1995, 11 7,5228. Chem. SOC., 1993, 115, 3030. I. Shimizu, Tetrahedron Lett., 1991, 32, 4937; (b) K. Nagasawa, €1. Ishihara, Y. Zako and 1. Shimizu, J. 0%. Chem., 1993,58, 2523; ( c ) C. Chen and D. Crich, Tetrahedron Lett., 1992, 33, 1945; (d) J. L. Mascarenas, A.M. Garcia, L. Castedo and A. Mourino, Tetrahedron Lett., 1992, 33, 4365; (e) T. Takahashi and M. Nakazawa, Synlett, 1993, 37; ( f ) S. Hatakeyama, H. Irie, T. Shintani, Y. Noguchi, H. Yamada and M. Nishizawa, Tetrahedron, 1994, 50, 13369. M. R. Uskokovic, J. Am. Chem. SOC., 1982, 104, 2945. Heravi and B. J.Mohr, Tetrahedron Lett., 1993, 34, 3079. 102 C. Liljebris, B. Resul and U. Hacksell, Tetrahedron, 103 D. Sperandio and €4. J. Hansen, Helv. Chim. Acta, 104 (a) D. C. Horwell, P .D. Nichols and E. Roberts, 105 (a) Y. Sato, M. Sodeoka and M. Shibasaki, J. OR. 112 V. H. Rawal, C. Michoud and K. Monestel, J. Am. 113 (a) K. Nagasawa, Y. Zako, H. Ishihara and 114 E. G. Raggiolini, 3. A. Iacobelli, B. M. Hennesy and 115 J. M. Nuss, M. M. Murphy, R. A. Rennels, M. H. 470 Contemporary Organic Synthesis116 C. K. Narula, K. J. Mak and R. F. Heck, J. Org. 117 J. K. Mukopadhyaya, S. Pal and U. R. Ghatak, Synzh. 118 R.-T. Wang, F.-L. Chou and F.-L. Luo, J. 0%. Chem., 119 E. Negishi, Y. Noda, F. Lamaty and E . J . Vawter, 120 S. Torii, H. Okumoto, T. Tadokoro, A. Nishimura 121 ( a ) J. M. Nuss, B. H. Levine, R. A. Rennels and Chem., 1983,48,2792. Commun., 1995, 25, 1641. 1990,55,4846. Tetrahedron Lett., 1990, 31, 4393. and M. A. Rashid, Tetrahedron Lett., 1993, 34, 2139. M. M. Heravi, Tetrahedron Lett., 1991, 32, 5243; (b) J. M. Nuss, R. A. Rennels and B. H. Levine, J. Am. Chem. SOC., 1993, 115, 6991. 122 G. D. Harris, R. J. Herr and S. M. Weinreb, J. Org. Chem., 1993,58,5452. 123 C. S. Nylund, J. M. Klopp and S. M. Weinreb, Tetra- hedron Lett., 1994, 35, 4287. 124 C. S. Nylund, D. T. Smith, J. M. Klopp and S. M. Weinreb, Tetrahedron, 1995, 51, 9301. 125 C.Y. Hong and L. E. Overman, Tetrahedron Lett., 1994,35, 3453. 126 S . P. Watson, G. R. Knox and N. M. Heron, Tetra- hedron Lett., 1994, 35, 9763. 127 (a) E. Negishi, Pure Apyl. Chem., 1992, 64, 323; ( 6 ) L. E. Overman, M. M. Ableman, D. J. Kucera, V. D. Tran and D. J. Ricca, Pure Appl. Chem., 1992, 64, 1813. 128 M. M. Abelman and L. E. Overman, J. Am. Chem. Soc., 1988, 110, 2328. 129 N. E. Carpenter, D. J. Kucera and L. E. Overman, J. 0%. Chem., 1989, 54, 5846. 130 Y. Zhang and E. Negishi, J. Am. Chem. SOC., 1989, 111,3454. 131 Y. Zhang, G. Wu, G. Agnel and E. Negishi, J. Am. Chem. SOC., 1990, 112, 8590. 132 F. E. Meyer, J. Brandenburg, P. J. Parsons and A. de Meijere, J. Chem. Soc., Chem. Commun,, 1992, 390. 133 F E. Meyer, H. Henniges and A. de Meijere, Tetra- hedron Lett., 1992, 33, 8039. 134 T. Sigihara, C. Coperet, Z. Owczarczyk, L. S. Harring and E. Negishi, 1. Am. Chem. SOC., 1994, 116, 7923. 135 ( a ) S . Torii, H. Okumoto and A. Nishimura, Tetra- hedron Lett., 1991,32, 4167; (b) E. Negishi, L. S. Harring, Z. Owczarczyk, M. M. Mohamud and M. Ay, Tetrahedron Lett., 1992, 33, 3253; ( c ) E. Negishi, M. Ay and T. Sugihara, Tetrahedron, 1993, 49, 5471. 136 F. E. Meyer and A. de Meijere, Synlett, 1991, 777. 137 ( a ) D. J. Kucera, S. J. O’Connor and L. E. Overman, J. 0%. Chern., 1993, 58, 5304; (b) L. E. Overman, D. J. Ricca and V. D. Tran, J. Am. Chem. SOC., 1993, 115,2042; ( c ) L. E. Overman, Pure Appl. Chem., 1994, 66, 1423. Lett., 1984, 25, 2271; ( 6 ) W. J. Scott, M. R. Pena, K. Sward, S. J. Stoessel and J. K. Stille, J. Org. Chem., 1985, 50, 2302; ( c ) S. Cacchi and A. Lupi, Tetrahedron Lett., 1992, 33, 3939. 139 (a) W. Cabri, I. Candiani, A. Bedeschi, S . Penco and R. Santi,J. Org. Chem., 1992, 57, 1481; (b) W. Cabri, I. Candiani, A. Bedeschi and R. Santi, J. Org. Chem., 1992,57, 3558; ( c ) F. Ozawa, A. Kubo and T. Hayashi, J. Am. Chem. SOC., 1991, 113, 1417. 140 E. G. Samsel and J. R. Norton, J. Am. Chem. Soc., 1983, 106, 5505. 141 ( a ) R. F. Heck, J. Am. Chem. SOC., 1969,91, 6707: (b) D. L. Thorn and R. Hoffmann, J. Am. Chem. Soc., 1978, 100, 2079. 142 (a) K. Karaberlas and A. Hallberg, Tetrahedron Lett., 1985, 26, 3131; (b) K. Karabelas, C. Westerlund and A. Hallberg, J. Org. Chenz., 1985, 50, 3896; ( c ) K. Karabelas and A. Hallberg, J. Org. Chem., 1986, 51, 5286; ( d ) K. Karabelas and A. Hallberg, J. 0%. Chem., 1988, 53, 4909; (e) K. Karabelas and A. Hall- berg, J. Oig. Chem., 1989, 54, 1773; (f) R. C. Larock and W. H. Gong, J. Org. Chem., 1989,54, 2047; (g) K. Nilsson and A. Hallberg, J. 0%. Chem., 1990,55, 2464: ( h ) K. Nilsson and A. Hallberg, J. Org. Chem., 1992, 57,4015. 143 (a) C. Carfagna, A. Musco, G. Sallese, R. Santi and G. Fiorani, J. Org. Chem., 1991, 56, 261; (b) W. Cabri, I. Candiani, A. Bedeschi and R. Senti, Tetrahedron Lett., 1991, 32, 1753. 138 (a) S. Cacchi, E. Morera and G. Ortar, Tetrahedron 144 T. Jeffery, Tetrahedron Lett., 1994, 35, 3051. 145 V. H. Rawal and C. Michoud, J. 0%. Chenz., 1993, 58, 5583. 146 J. J. Masters and L. S. Hegedus, J. 0%. Chem., 1993, 58, 4547. 147 (a) M. Green and R. P. Hughes, J. Chem. SOC., Dalton Trans., 1976, 1880; (b) W. A. Donaldson and C. A. Brodt, J. Organomet. Chem., 1987,330, C33. 148 U. Hacksell and G. D. Daves Jr, Organometallics, 1983, 2, 772. Gibson: The intramolecular Heck reaction 471