摘要:

Chemical Society Reviews Vol 11 No 2 1982 Page The Synthesis of Lignans and Neolignans 75By R. S. Ward The Thermolysis and Photolysis of Diazirines By M. T.H. Liu 127 CENTENARY LECTURE Cyclopentanoids:A Challenge for New Methodology By 33. M. Trost 141 NYHOLM LECTURE Solving Chemical Problems By M. J. Frazer 171 TILDEN LECTURE Carbon-Carbon Bond Formation Involving Boron Reagents By A. Pelter 191 The Royal Society of ChemistryLondon Chemical Society Reviews EDITORIAL BOARD Professor K. W. Bagnall, B.Sc., Ph.D., D.Sc., C.Chem., F.R.S.C. (Chairman) Professor K. R. Jennings, M.A., D.Phil., C.Chem., F.R.S.C. Professor G. W. Kirby, M.A., Ph.D., Sc.D., F.R.S.E., C.Chem., F.R.S.C. Professor G. Pattenden, Ph.D., C.Chem., F.R.S.C.Professor B. L. Shaw, B.Sc., Ph.D., F.R.S. Professor P. A. H. Wyatt, B.Sc., Ph.D., C.Chem., F.R.S.C. Editor: K. J. Wilkinson, B.Sc., M.Phi1. Chemical Society Reviews appears quarterly and comprises approximately 20 articles (ca. 500 pp) per annum. It is intended that each review article shall be of interest to chemists in general, and not merely to those with a specialist interest in the subject under review. The articles range over the whole of chemistry and its interfaces with other disciplines. Although the majority of articles are intended to be specially commissioned, the Society is always prepared to consider offers of articles for publication, In such cases a short synopsis, rather than the completed article, should be sub- mitted to The Managing Editor, Books and Reviews Section, The Royal Society of Chemistry, Burlington House, Piccadilly, London, W1 V OBN.Members of the Royal Society of Chemistry may subscribe to Chemical Society Reviews at E12.50 per annum; they should place their orders on their Annual Subscription renewal forms in the usual way. All other orders accompanied with payment should be sent directly to The Royal Society of Chemistry, The Distribution Centre, Blackhorse Road, Letchworth, Herts. SG6 1HN England. 1982 annual subscription rate U.K. E36.00, Rest of World E38.00, U.S.A. $85.00. Air freight and mailing in the U.S.A. by Publications Expediting Inc., 200 Meacham Avenue, Elmont, New York 11003. U.S.A. Postmaster: Send address changes to Chemical Society Reviews, Publications Expediting Inc., 200 Meacham Avenue, Elmont, New York 11003.Application to mail at second class postage is pending at Jamaica, New York 11431. All other despatches outside the U.K. by Bulk Airmail within Europe, Accelerated Surface Post outside Europe, Note to subscribers. Regrettably publication of the four issues has still not reverted to the usual quarterly dates. The cause of this is a persisting shortage of articles (the production problems of recent years have been largely overcome) but the setting-up of an Editorial Board should result in an increase in the commissioning of reviews in 1982. 0Copyright reserved by The Royal Society of Chemistry 1982 ISSN 0306--0012 Published by The Royal Society of Chemistry, Burlington House, London, W1V OBN Printed in England by Eyre & Spottiswoode Ltd, Thanet Press, Margate.

ISSN:0306-0012    

DOI:10.1039/CS98211FP003

出版商:RSC    

年代:1982

数据来源: RSC

ISSN:0306-0012    

DOI:10.1039/CS98211FX005

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Chemical Society Reviews Vol 11 No 2 1982 Page The Synthesis of Lignans and Neolignans By R. S. Ward 75 The Thermolysis and Photolysis of Diazirines By M. T. H. Liu 127 CENTENARY LECTURE Cyclopentanoids:A Challenge for New Methodology By B. M. Trost 141 NYHOLM LECTURE Solving Chemical Problems By M. J. Frazer 171 TILDEN LECTURE Carbon-Carbon Bond Formation Involviiig Boron Reagents By A. Pelter 191 The Royal Society of Chemistry London

ISSN:0306-0012    

DOI:10.1039/CS98211BX007

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

The Synthesis of Lignans and Neolignans By R. S. Ward CHEMISTRY DEPARTMENT, UNIVERSITY COLLEGE OF SWANSEA, SINGLETON PARK, SWANSEA, SA2 8PP 1 Introduction Lignans and neolignans are formed in nature by the oxidative dimerization of various cSc3 phenols [e.g. (l)-(4)J.1 Indeed some lignans and neolignans can be prepared from cinnamyl alcohols and propenyl phenols, either by enzymic oxidation or by using conventional oxidizing agents such as FeC13 and AgZO. MeoT(a; R = H) (b; R = OMe)HO HO R R (I; X = CH,OH) (4) (2; X = C02H) (3; X = Me) The distinction between the terms lignan and neolignan has led to much confusion over the years, since at least two conflicting definitions have been proposed. Traditionally the term lignan was reserved for compounds such as (5) and (6) in which the two CsC3 units are linked by a bond connecting the central (/3) carbon atoms of each side chain.2 The term neolignan was introduced to designate compounds such as (7) and (8) in which the two cSc3 units are not linked by a /3-/3 According to the most recent definition, however, lignans [e.g.(5) and (7)] are formed by oxidative coupling of cinnamyl alcohols and/or cinnamic acids, whereas neolignans [e.g. (6) and (8)] are formed by oxidative coupling of propenylphenols and/or ally1 phenol^.^.^ Since the latter definition does not identify any fundamental chemical difference between the two series of compounds so defined, the former definition is to be preferred and is adopted in this review. ‘Chemistry of Lignans’, ed.C. B. S. Rao, Andhra University Press, 1978. a R. D. Haworth, J. Chem. Soc., 1942,448. 0. R. Gottlieb, Phytochemisfry,1972, 11, 1537. 0.R. Gottlieb, Rev. Latinoamer. Quim., 1974, 5, 1. 0. R. Gottlieb, in ‘New Natural Products and Plant Drugs with Pharmacological,Bioiogical or Therapeutical Activity’, Springer-Verlag, Berlin-Heidelberg. 1977, p. 227-248. 0. R. Gottlieb, Fortsch. Chem. Org. Naturst., 1978, 35, I. 75 The Synthesis of Ligti~tismid Nc.oligmru (5) HOCH, OMe (7) The many varied structures which lignans and neolignans possess can be accounted for by the different possible modes of coupling of the phenoxy- radicals generated by oxidation of (1)-(4). Thus, lignans are formed by further modification (e.g.reductioil, cyclization, hydration) of the intermediate obtained by coupling together two canonical forms of type E, whereas neolignans (as defined above) are formed from other combinations of the canonical forms A-E.It is also of interest to note that different modes of coupling do indeed occur in oxidative coupling reactions depending upon the method of oxidation or the oxidizing agent employed (see Sections 1 and 2). Lignans and neolignans have attracted much interest over the years on account of their widespread ocurrence in nature,? and on account of their broad range of biological activity.5 Thus, several lignans and neolignans are known to exhibit anti-tumour activity,8-12 while others function as growth inhibitors and anti- fungal agents.13114 Of possibly even greater importance is the recent isolation J.R. Cole and R. M. Wiedkopf, ref. 1, Chap. 2. C. Keller-Juslen, M. Kuhn, A. von Wartburg, and H. Stahelin, J. Med. Chem., 1971, 14, 936. * S. M. Kupchan, R. W. Britton, M. F. Ziegler, C. J. Gilmore, R. J. Restivo, and R. F. Bryan, J. Am. Chem. SOC.,1973,95, 1335. lo S. G. Weiss, M. Tin-Wa, R. E. Perdue, and N. R. Farnsworth, J. Pharm. Sci., 1975, 64, 95. l1 J. L. Hartwell, Cancer Treat. Rep., 1976, 60, 1031. S. K. Carter and R. B. Livingston, Cancer Treat. Rep., 1976, 60, 1141. l3 T. Kamikado, C. F. Chang, S. Murakoshi, A. Sakurai, and S. Tamura, Agric. Bid. Chem. (Jpn.), 1975, 39, 833. l4 G. B. Russell, P. Singh, and P. G. Fenemore, Am!. J. Bid. Sri., 1976, 29, 99.Ward X X X c (E) (D) of lignans from animals, including human beings,15-18 which has led to the suggestion that such compounds may be examples of a new type of hormone controlling cell growth. The many varied types of structure that lignans and neolignans can possess have presented a considerable challenge to organic chemists over the years and indeed many elegant syntheses have been reported. In the present review the methods that have been used to prepare lignans and neolignans are classified according to the types of reaction employed rather than according to the types of compound prepared. Thus, it can be seen that most of the syntheses that have been carried out depend in fact upon a limited number of key reactions, which have been used to construct the basic 18-carbon skeleton.Particular emphasis has been placed on new methods of synthesis involving, for example, non-phenolic oxidative coupling, cycloaddition to quinone monoketals, and conjugate addition by acyl anion equivalents. However, due prominence is also accorded to classical routes involving phenolic oxidative coupling, Diels Alder reactions, and Stobbe condensations, which are, in many ways, still unsurpassed as versatile routes to a whole range of lignans and neolignans. l5 S. R. Stitch, P. D. Smith, D. Illingworth, and K. Toumba, J. Endocrinol., 1980, 85, 23P. Is S. R. Stitch, J. K. Toumba, M. B. Groen, C. W. Funke, J. Leemhuis, J. Vink, and G. F. Woods, Nature, 1980, 287, 738. K.D. R. Setchell, R. Bull, and H. Adiercreutz, J. Steroid Biochem., 1980, 12, 375. K. D. R. Setchell, A. M. Lawson, F. L. Mitchell, H. Adlercreutz, D. N. Kirk, and M. Axelson. Nature, 1980, 287, 740. The Synthesis of Lignans and Neolignans 2 Phenolic Oxidative Coupling The oxidative coupling reactions of coniferyl alcohol and sinapyl alcohol constitute biomimetic syntheses of lignans and neolignans. Thus, enzyme catalysed oxidation of coniferyl alcohol (la) yields a mixture of products including pinoresinol (5) and dehydrod icon ifery 1 alcohol (7), while oxidative coupling of sinapyl alcohol (16) affords a high yield of syringaresinol (9).20 (1 b) (9) (Ar = 3,5-dimethoxy4-hydroxyphenyl) The oxidative coupling reactions of ferulic acid and sinapic acid are much more useful from a synthetic point of view, however, since the dilactones produced have unambiguous structures and are amenable to a wide range of structural modifications (Schemes 1 and 2).The dilactone (lo), for example, is produced in 30% yield by oxidation of ferulic acid (2a).21Treatment with methanolic HCI causes rearrangement to the aryldi hydronaphthalene (1 l), which can be further converted into a mixture of the arylnaphthalene lactones (12) and (13).22 Alternatively, hydrogenation followed by dehydration affords the anhydride (14), which is converted by reduction into matairesinol (1 5).22 Reduction of the acetylated dilactone (1 6) with lithium aluminium hydride (LAH) yields the tetrol (17), which can be cyclized to give pinoresinol (9,and the methylated dilactone (18) can be similarly converted into eudesmin (19).23 Partial reduction of the dilactones (10) and (18) can be achieved using di-isobutyl aluminium hydride (DIBAL), which affords the dilactols (20) and (21).(a) K. Freudenberg and H. H. Hubner, Chem. Ber., 1952, 85, 1181. (6) C. J. Sih, P. R. Ravikumar, F.-C. Huang, C. Buckner, and H. Whitlock, J. Am. Chem. Sor., 1976, 98, 5412. to (a) K. Freudenberg and G. Grion, Chem. Ber., 1959, 92,1355. (b) E.E. Dickey, J. Org.Chem., 1958,23,179. a1 (a) H. Erdtman, Svensk. Kem. Tidskr., 1935, 47, 223. (b) N. J. Cartwright and R. D. Haworth, J. Chem. SOC.,1944, 535. 22 Y. Takei, K. Mori, and M. Matsui, Agric. Biol. Chem. (Jpn.), 1973, 37, 637.p3 A. Pelter, R. S. Ward, D. J. Watson, P. Collins, and I. T. Kay, J. Chem. Soc., ferkin Trans. I, 1982, 175. Ward 79 The Synthesis of Lignans and Neolignans These can be further converted by tosylation followed by LAH reduction into the parent 2,6-diaryl-3,7-dioxabicyclo[3.3.0]octane~.~~Furthermore, since this last sequence does not involve ring-opened intermediates it constitutes an un- ambiguous synthesis of this class of lignans.23 As would be expected, oxidation of a mixture of coniferyl alcohol (la) and ferulic acid (2a) affords a mixture of pinoresinol (5) and the dilactone (lo), along with the monolactone (22) which is a potent germination inhibit0r.2~ Anodic oxidation of ferulic acid gives only a 6% yield of the dilactone (10) but gives a high yield (70%) of the asatone-type compound (23),25 formed by cyclodimerization of a dienone intermediate corresponding to canonical form C of the phenoxy-radical.(Ar = 4-hydroxy-3-methoxyphenyl) 0 The dilactone (24) is produced in 80% yield by phenolic oxidation of sinapic acid (2b).26 Treatment with aqueous acid yields thomasadioic acid (25), whereas treatment with methanolic HCl gives the dimethyl ester (26). Further modifi- cation of the benzyl ether (27) or the methyl ether (28) affords thomasic acid (29) and its dimethyl ether (30). Hydrogenation of (28) affords a mixture of the isomeric aryltetralins (31) and (32) of which the former has been converted into lyoniresinol dimethyl ether (33).27 Anodic oxidation of sinapic acid (2b) gives a low yield (9 %) of an isoasatone derivative (34)in addition to the dilactone (24).25 24 R.Cooper, H. E. Gottlieb, D. Lavie, and E. C. Levy, Tetrahedron, 1979, 35, 861. M. iguchi, A. Nishiyama, H. Eto, Y. Terada, and S. Yamamura, Chem. Lett., 1979, 1397. ae (a) K. Freudenberg and H. Schraube, Chem. Ber., 1955,88, 16. (6) R. Ahmed, M. Lehrer, and R. Stevenson, Tetrahedron Lett., 1973, 747; Chem. Ind. (London), 1973, 1001; Tetrahedron, 1973, 29, 3753. 27 A. F. A. Wallis, Aust. J. Chem., 1973, 26, 585 and 1571. 80 Ward MeoQ-co2HHO FeCI, 1.. OMe ROTco2HOMe Me0 *TH,OH H--RO Me0 Me0 OMe (24; R = H)(35; R = Ac) RO (29; R=H) (30; R = Me) Me0 CH,OH MeoT H,OH Me0 ' OMe RO Me0 (26; R=H) (27; R = CH,Ph) (33) (28; R= Me) C0,MeMe0 Me0 'C0,Me'C0,Me + Me0 OMe Me0 Me0 Me0 Scheme 2 The Synthesis of Lignans and Neolignans HO ' anodiL -(24) +oxidation in MeOHMeorco2HMe0 (2b) Several substituted dilactones have been prepared by oxidative coupling of appropriately substituted cinnamic acid^.^^^^^ Some of the diiactones so produced or indeed produced by direct halogenation of the parent dilactones [(16) and (35)] are useful for preparing other lignan types, as for example by acid catalysed rearrangement to tetrahydrofuran derivatives (Scheme 3).30 Tn this instance, the R Me0 OMe AcO Br R R R (36) 1.LAH (37) R R (38; R = H) galbelgin (39; R = OMe) grandisin Scheme 3 28 Y.Kumada, H. Naganawa, T. Takeuchi, H. Umezawa, K. Yamashita, and K. Watanabe, J. Antibiotics, 1978, 31, 105. 2s R. Ahmed, F. G. Schreiber, R. Stevenson, J. R. Williams, and H. M. Yeo, Tetrahedron, 1976,32, 1339. so R. Stevenson and J. R,Williams, Tetrahedron, 1977, 33, 285. Ward presence of the halogen atom serves to divert the normal course of the cyclization reaction. In other instances, and depending on the reaction conditions (Scheme 4), the presence of the halogen atom can be used to produce a different oxygen- ation pattern in the aryldihydronaphthalene (42), and hence in the aryltetralins (43) and (44),by blocking the normal position of ring cl~sure.~O OMe-HCI-MeOH-CHCl, Me0 OH (40) (42) HCI-MeOH.1 2.LAH 3. NaH-Me1 Me0 ,C C0,Me Meon+oH OMeHO (41) OMe + CH ,OMe ‘CH,OMe OMe OMe (44) Scheme 4 Phenolic oxidation of methyl sinapate (45) yields a mixture of compounds of which the major component (61 %) is the 4-hydroxyaryltetralin (4Q.27 Acid catalysed dehydration affords the dimethyl ester of thomasadioic acid (26).27 The Synthesis of Lignans and Neolignans C0,Me Me0MeoT-FeCI, HO Me0 (45) OH (46) IH’ Me0 OT C02MeH C0,Me Me0 OMe HO (26) The dihalogenated aryldihydronaphthalenes (48) can be directly prepared from the corresponding cinnarnate esters and the iodo-derivative (48; X = I) forms the starting point for a synthesis of isolariciresinol dimethyl ether (49).29 C0,Me Me0 FeC1, j **CO,Me X (47; X = I or Br) Me0 HO (48) (7) Me0 CHzOH ,OH 84 Me0 Me0 Wurd C0,Me C0,Me HO Br OH (50) Br Br Me Me 1.LAH 2. TsCL 3. LAH (Ar = 3,4-dimethoxyphenyl) rn contrast, phenolic oxidation of the dibromoferulate (50) affords the tetra- hydrofuran derivative (5 1) directly, leading eventually to veraguensin (52).29 Phenolic oxidation of the cinnamate ester (53) yields the bisquinonemethides (54) and (55),which both undergo tautomerization to afford the diarylbutadiene (561, but on hydrogenation yield the diarylbutanes (57) and (58) respectively (Scheme 5).31 The oxidative coupling of various ally1 and propenyl phenols is involved in the biosynthesis of several lignans and neolignans and affords a convenient route to a number of such compounds.Thus, (E)-isoeugenol (3a) is converted by oxidation into a mixture of compounds of which dehydrodi-isoeugenol (59) is the major component.32- 36 (2)-Isoeugenol gives a similar range of pr0ducts.3~ In contrast, chemical or enzymic oxidation of the (E)-2,6-dimethoxypropenyl-phenol (3b)yields a mixture of two tetrahydrofuran derivatives (60) and (61) as major products (Scheme 6).32Their methyl ethers rearrange in acid to give the aryldihydronaphthalene (62), illustrating once again the ready interconversion of the various lignan types under acid conditions. Oxidation of the (Z)-isomer of the same phenol yields four diastereoisomeric tetrahydrofuran derivative^.^^ 31 K.V.Sarkanen and A. F. A. Wallis, J. Chem. Soc., Perkin Trans. I, 1973, 1878. 32 K. V. Sarkanen and A. F. A. Wallis, J. Chem. Soc., Perkin Trans. I, 1973, 1869. 33 H. Cousin and H. Herissey, Compt. rend., 1908, 147, 247; Bull. Soc. Chim. Fr., 1908, 3, 1070; J. Pharm. Chim., 1908, 28, 93. K. Eskins, C. Glass, W. Rohwedder, R. Kleiman, and J. Slonekar, Tetrahedron Left., 1972,861. 35 1. J. Miller, Tetrahedron Lett., 1972, 4955. 3K M. Iguchi, A. Nishiyama, M. Hara, Y. Terada, and S. Yamamura, Chem. Lett., 1978, 1015. The Synthesis of Lignans and Neolignans C0,Me Me0,C CH,Ar H--H--*HOX-Q” ArCH, C0,Me ij3) (57) K ,I;e(CN), H/ MeOzC CHAr ArCHHC0,Me MeOK COzMe H->-(-Ht ////(54) ArCH, CH,Ar+ H2/ Me0,C C0,Me (58)$-’yo(Ar = 3,5-di-t-butyl-4-hydroxyphenyl) 0 (55) Scheme 5 Oxidation of (36) with ferric chloride yields (64) as the major product along with a number of other minor products as indi~ated.~7 Oxidation of the (2)-isomer with ferric chloride also gives a similar range of products.Anodic oxidation of eugenol (4a) and the allylphenol (46) gives various dimeric products, including the asatone derivatives (66) and (67).37938 Oxidation of (4a) with thallium nitrate in methanol also yields asatone derivative^.^^ Finally, a mixed phenolic oxidation is involved in the synthesis of eusiderin (68),40 and a remarkable reaction involving the generation of five contiguous 37 M. Iguchi, A. Nishiyama, Y. Terada, and S. Yamamura, Chem. Lett., 1978, 451.38 M. Iguchi, A. Nishiyama, Y. Terada, and S. Yamamura, Tetrahedron Lett., 1977,451 1. 39 M. Niwa, H. Noda, H. Kobayashi, and S. Yamamura, Chem. Lett., 1980, 85. *O L. Merlini and A. Zanarotti, Tetrahedron Lett., 1975, 3621 ;L. Merlini, A. Zanarotti, A. Pelter, M. P. Rochefort, and R. Hansel, J. Chem. Soc., Perkin Trans. I, 1980, 775. 86 Ward HO (34 Ar OMe (9) + (4 %, (17 %I (9 %) (Ar = 4-hydroxy-3-methoxyphenyl) Me Me HZ02 HO OR K,Fe(CN), + Me0 Ar' At-' 'Ar Me0 'Me H2Me0 Me0Pd-C / Me0 'Me0 OMe OMe OMe OMe (63) (62) (Ar = 3,5-dimethoxy-4-hydroxyphenyl) Scheme 6 The Synthesis of Ligiians orid N~~oligntrtrs Me0 I ArCHOH OMe (64; 55%) + OH Et Me0HOT:e 3-+ Me0 OH HO HO (Ar = 35dirnethoxy-4-h y droxy phenyl) asymmetric centres with the correct relative stereochemistry is involved in the synthesis of carpanone (70) by phenolic oxidation of the propenylphenol (69). Intramolecular cycloaddition of the intermediate bis-quinonemethide is held to be responsible for the high stereoselectivity of this rea~tion.~~.~~ 3 Non-phenolic Oxidative Coupling Several lignan syntheses are based on the recently developed technique of non-phenolic oxidative coupling using reagents such as thallium(1rr) trifluoroacetate and vanadium oxyfluoride.For example, treatment of the diarylbutane (71) with either of these reagents generates the dibenzocyclo-octadiene (72), which 41 M. Matsumoto and K. Kuroda, Tetrahedron Lett., 1981, 4437.42 0. L. Chapman, M. R. Engel, J. P. Springer. and J. C. Clardy, J. Am. Chetn. Soc., 1971, 93, 6696. Word Meoranodic7zzz HO inMeOH Me0 (4a) HO OH (65) + OMeMeoranodic MeO HO in MeOH MeO (4b) 0 OMe (Ar= 3,5-dimethoxy-4-hydroxyphenyl) OMeTo"' OH OMe OMe IJ 89 The Synthesis of Lignans and Neolignans ri' <qo<a0,GzT Me PdCl / Me 0' Lt,b0 has been used by Kende et al. for the synthesis of steganone and isosteganone (Scheme 7).43t44 The formation of a biaryl link by non-phenolic oxidative coupling is also a key step in Schlessinger's synthesis of isostegane (see section 6),45 and Stevenson's synthesis of deoxyschizandrin (see section Indeed Stevenson has demonstrated the general applicability of this reaction for the synthesis of dibenzocyclo-octadienes.46 In contrast, oxidation of the monophenol (73) affords the aryltetralin (74), which has been converted by means of a similar series of reactions into picro- podophyllone (Scheme 8).47 Clearly the presence of one phenolic group rep- resents a useful control element that may be of general use in dictating the size of ring produced.Another non-phenolic oxidative coupling reaction, which has strong similarities to many of the reactions discussed in the previous section, is the oxidative dimerization of cinnamic acids using thallium(1Ir) trifl~oroacetate.~~ 43 A. S. Kende and L. S. Liebeskind, J. Am. Chem. SOC.,1976,98, 267. '* A. S. Kende, L.S. Liebeskind, C. Kubiak, and R. Eisenberg, J. Am. Chem. Soc., 1976, 98, 6389 and R. C. Cambie, M. G. Dunlop, P. S. Rutledge, and P. D. Woodgate, Synth. Commun., 1980, 10, 827. 45 R. E. Damon, R. H. Schlessinger, and J. P. Blount, J. Org. Chem., 1976, 41, 3772. 46 T. Biftu, B. G. Hazra, and R. Stevenson, J. Chem. SOC.,Perkin Trans. I, 1979, 2276. 47 A. S. Kende, L. S. Liebeskind, J. E. Mills, P. S. Rutledge, and D. P. Curran, J. Am. Chem. Soc., 1977, 99, 7082 and A. S. Kende and P. S. Rutledge, Synth. Commun., 1978, 8, 245. 48 E. C. Taylor, J. G. Andrade, G. J. H. Rall, and A. McKillop, Tetrahedron Lett., 1978, 3623 and E. C. Taylor, J. G. Andrade, G. J. H. Rall, K. Steliou, G. E. Jagdamann, and A. McKillop, J. Org. Chem., 1981, 46, 3078.Ward - TTFA OR VOF, OMe OMe (72) 1. NBS 2.fI*O 3. Cr03-py 4. NaOH 5 CI1,O-OH + Me0 Me0 isosteganone steganone Scheme 7 91 The Synthesis of Lignans and Neolignans COz Et 1. TTFA C02 Et 2. NafiS03 3. Me, SO, ' Me0 QOMe OMeMe0 0 OH OMe (73) (74) I. NBS 2. HZO 3. CrOB--pyI 4.NaOH g*o CHI0- OH Q0 Me0 OMe Me0 OMe OMe OMe Scheme 8 This reaction has been used to prepare 4,s-dihydroxysesamin (76), which is at the present time the only naturally occurring example of a 4,s-dihydroxy-furofuran.23 Oxidation of the cinnamate ester (77) affords a high yield of the diary1 butadiene (78).48 7iTTFA oT40 Ar DlBAL HOR Ar<wco2"->H----H _I) H----H (15) Ar' 0 0 Ar' 0 OH (75) (76) (Ar = 3,4-methylenedioxyphenyl)rCozMe~ HArTTFA MeOzC \ Me0 Ar COzMe (77) (78) (Ar = 4-methoxyphenyl) Wwrd 4 The Use of Quinone Ketals Biichi has shown that various neolignans can be prepared by thc reaction of elect ron-rich al kenes with appropriately substituted qu inone nioiiokct aIs under acidic conditions.By varying the reaction conditions, and in particular the acid catalyst cmploycd. the reaction can be directed to give predominantly either thc burclwllin. guianin. or futoenone series (Scheme 9).4!'.50ii has been suggested Me aOMe Ar 'I '0 guianin burchellin 3 OMe Ar p + ou)Ar + Ar Me0 0 fut oenone (Ar = 3,4-met hylenedioxy phenyl) Scheme 9 49 G.Buchi and C.-P. Mak, J. Am. Clirnr. Sor., 1977, 99, 8073. G. Biichi and P.-S. Chu, J. Org. CIiem., 1978, 43, 3717. 93 The Synthesis of Lignans and Neolignans that the formation of the bicyclo[3.2. I ]octane series involves a concerted [4 +21 cycloaddition of a cyclohexadienyl cation to the alkene and that the dihydro- benzofuran and futoeone series result from subsequent isomerization of these products. The close relationship between the dihydrobenzofurans and the bicyclo[3.2.l]octanes is further emphasized by their ready interconversion under acidic conditions.51 Gottlieb has suggested that the preferred direction of these rearrangements is determined by steric interactions in the vicinity of the tetra- substituted sp3 hybridized C atom.51 The bicyclo[3.2.l]octane derivatives can also be further converted into tropolone deri~atives.5~ A similar scheme to that outlined above has been utilized by Buchi et a!.to synthesize megaphone,53 a compound known to exhibit anti-tumour activity. Eventual ring opening of the tetrahydrobenzofuran derivative affords the required cyclohexenone (Scheme 10). OMS -30 'C Ar OMe I. Hz/Kh-C2.ArSe-2. MsCl f---3. H,O megaphone (Ar= 3,4,5-trimethoxyphenyl) Scheme 10 5 Diels Alder and Related Reactions Diels Alder reactions have been widely used to synthesize lignans of the aryl- naphthalene and aryltetralin series. The first approach, involving cyclization of an acetylenic acid anhydride (Scheme ll), is illustrated by the synthesis of justicidin E (79) and taiwanin C (80).54.55 61 M.A. de Alvarenya, U. Brocksom, 0. R. Gottlieb, M. Yoshida, R. Filho, and R. Figliuolo, J. Chem. Soc., Chem. Commun., 1978, 831. 62 C.-P. Mak and G. Buchi, J. Org. Chem., 1981, 46, 1. 63 G. Buchi and P.-S. Chu, J. Am. Chem. SOC.,1981, 103, 2718. 64 T. L. Holmes and R. Stevenson, J. Org. Chem., 1971, 36, 3451). 55 D. Brown and R. Stevenson, Tetrahedron Lett., 1964, 3213, and J. Org. Chem., 1965, 30, 1759. 94 Ward 0 1. LA11 2. Ag,CO, -celite (0 ~ 0+ Scheme 11 By using the 2-bromophenylpropiolic acid (81) the direction of cyclization can be controlled in such a way as to yield the 7,8-dioxygenated system as found in helioxanthin (82), otobain (6), and dehydro-otobain (83) (Scheme 12).56*57 seT.L. Holrnes and R. Stevenson, Tetrahedron Lett., 1970, 199 and J. Chem. Sor. (C), 1971,2091. 57 I. Maclean and R. Stevenson, Chem. lnd. (London), 1965. 1379 and J. Chem. Sor. (C), 1966, 1717. The Synthesis of Lignans and Neolignans (6) 2. LAH 3. TsCl 4. LAH A DCCl ’ I. LAH AICI, 2. H,/Pd-C A second approach, involving the cyclization of a doubly unsaturated ester was used extensively by Klemm and co-workers for the synthesis of several aryltetralin lactones (Scheme 1 3).58-61The same approach has also been used by 58 L. H. Klemm and K. W. Gopinath, Tetrahedron Lett., 1963, 1243. 5s L. H. Klemm, K. E. Gopinath, D. Hsu-Lee, F. W. Kelly, E. Trod, and T. M. McGuire, Tetrahedron, 1966, 22, 1797.6o L. H. Klemm and P. S. Santhanam, J. Org. Chem., 1968, 33, 1268. 81 L. H. Klemm, D. R. Olson, and D. V. White, J. Org. Chem.. 1971, 36, 3740. Ward Stevenson et al. for the synthesis of collinusin and justicidin B,62t63 and by Joshi et a/. for the synthesis of attenuol (84) (Scheme 14).(j4 (Om0 I -clcctlo-0I reduction ’ / \ Me0 OMeMe0 OMe OMe OMe Me0 Me0Me0 OMe Scheme 13 Not surprisingly it is the multiple bond adjacent to the carbonyl group which acts as the dienophile in most cases, although in the case of the doubly acetylenic ester (85) both possible lactones are obtalned.58~5~ Thus, whereas the trans-6a E. Block and R. Stevenson, Chem. Ind. (London), 1970, 894 and J.Org. Chem., 1971, 36, 3453. 63 F. Kohen, I. Maclean, and R. Stevenson, J. Chem. Sor. (C), 1966, 1775. 64 B. S. Joshi, N. Viswanathan, V. Balakrishnan, D. H. Gawad, and K. R. Ravindranath, T(>rruhedron,1979, 35, 1665. 97 The Synthesis of Lignans and Neolignans RaNii OR 1. NaOAc 2. LAH<yMe1 TsCl CHzOH<T/ ‘Me 2. LAH / HzOH \ \ OH OR (84) Scheme 14 cinnamyl moiety functions only as a diene, the phenylpropargyl group may serve as either diene or dienophile. The oxidative cyclization of a doubly unsaturated succinic anhydride can also be used, as illustrated by the synthesis of justicidin E (79),taiwanin C (80),and helioxanthin (82).e5 This process formally involves an electrocyclic ring-closure followed by dehydrogenation.Indeed, cyclization of the di(arylmethy1ene)-succinic anhydrides and the corresponding lactones can also be achieved photo- 65 A. s. R. Anjaneyulu, V. Kameswara Rao, P. Satyanarayana, and L. R. Row, Indian J. Chem., 1973, 11, 203 and A. S. R. Anjaneyulu, V. Kameswara Rao, A. Madhusudhana Rao, and L. R. Row, Curr. Sci., 1974, 43, 542. Ward Me0 Me0 0 + -0 Me0Me09 Me0 MeoTOMe OMe (13) ~hemically,66~6~and in the case of the lactones at least, the initial products obtained are the expected 1,4-dihydroarylnaphthalenes[e.g. (86)].67 Photo-sensitized oxygenation of the doubly unsaturated lactones gives a low yield of the corresponding 4-hydroxyarylnaphthalene lactones.G8 0 Pb(OAc), -Q--(79) + (80) (18) HOA~-H+ 1.LAH 2. Ag,COs -silica OJO An elegant recent application of the Diels Alder reaction for the synthesis of lignans involves the generation in situ of an isobenzofuran such as (87) and its reaction with dimethyl acetylenedicarboxylate (DMAD), leading for example to dehydropodophyllotoxin (88) (Scheme 15).69 The same general approach has 66 D. C. Ayres, B. G. Carpenter, and R. C. Denney, J. Chem. SOC.,1965, 3578. 67 H. G. Heller and P. J. Strydom, J. Chem. SOC.,Chem. Commun., 1976, 50. 68 Z.-I. Horii, K. Ohkawa, and C. Iwata, Chem. Pharm. Bull., 1972, 20, 624. 6B H. P. Plaumann, J. G. Smith, and R. Rodrigo, J. Chem. Soc., Chem. Commun., 1980, 354, and S. 0. de Silva. C. St. Denis, and R. Rodrigo, J. Chem. Suc., Chem. Commun., 1980,995.99 The Synthesis of Lignans and Neolignans I. 1.AH 2. Ag2COa-silica (82) also been used to prepare taiwanin E, chinensinaphthol and diphyllin. In the absence of a strong acid the initial adduct can be isolated and selectively reduced leading eventually, in the case illustrated, to podophyllotoxin (89).70 Photoenolization provides an alternative way of generating a quinone-dimethane intermediate suitable for cycloaddition to a dienophile, and two distinctly different approaches have been developed using this reaction. The first starts from an appropriate benzophenone derivative (90)71whereas the second involves a formyldiphenylmethane [e.g. (92)].72 The latter route has the added attraction that it automatically generates a 4-hydroxy-substituent, leading R.Rodrigo, J. Org. Chem., 1980, 45, 4538 and D. Rajapaksa and R. Rodrigo, J. Am. Chem. SOC.,1981, 103,6208. 71 E. Block and R. Stevenson, J. Chem. SOC.,Chem. Commun., 1971, 71 1 and J. Chem. SOC., Perkin Trans. I, 1973, 308. B. J. Arnold, S. M. Mellows, and P. G. Sammes, J. Chem. SOC.,Perkin Trans. I, 1973, 1266. 100 Ward H+ Me0 Me0QOMC OMe C02Me coz Me C02Me Me0 QOMe Me0 QOMe OMe OMe BI I 3.Me S &OC*20H OH C02Me I Me0 G+*OMe OMe OH 2. HO- 3. H30+ OMe OMe Scheme 15 (89) 101 The Synthesis of Lignans and Neolignans for example to taiwanin E (93), whereas the former offers a potentially useful route to various dihydroarylnaphthalenes [e.g.(9I )I. OH COzMe<qCHO0 C0,Me H+ C'tC -(79) + (80) 0J0 6 Conjugate Addition Reactions The conjugate addition of a thioacetal carbanion to butenolide followed by trapping of the enolate anion so generated with a suitable electrophile provides a short, efficient approach to the construction of the basic lignan skeleton. Thus Ziegler et 01.73 have used benzyl halides and aromatic aldehydes to trap the enolate anions and have studied various cyclization reactions of the dibenzyl- butyrolactones produced (Schemes 16 and 17). The synthesis of isostegane by 73 F. E. Ziegler and J. A. Schwartz, J. Org. Chem., 1978, 43, 985. 102 Ward this route was first accomplished by Schlessinger et al.,45who suggested the possible involvement of a spirodienone intermediate in the cyclization step.Me0 isostegane 1. RaNi1.2. VOF, RS SR Me0 @OMe OMe OMe I. HgO--BF, 2. NaBHl 3. TFA 0 Me0 Scheme 16 Pelter et al.74have utilized a similar scheme to provide a general route to dibenzylbutyrolactones, including 'Factor X' (94), which is the first lignan to be isolated from animal ~ources.~~-~~ It has also been shown that treatment of the 74 A. Pelter, R. S. Ward, and P. Satyanarayana, Tetrahedron Lett., 1981, 1549. The Synthesis of Lignans and Neolignans RS SR RS SR OMe OMe deox yisopodoph yllotoxin isopodophyllotoxone Scheme 17 bis( thiophenyl) derivatives of the di benzylbutyrolactones with mercury(i1) trifluoroacetate affords the corresponding arylnaphthalenes (95) in good yield.75 No trace of the intermediate dihydroarylnaphthalenes was detected.Gonzalez ef al.76 have utilized a similar sequence to synthesize isodiphyllin, and Mpango and Snieckus have carried out similar conjugate addition reactions on an N,N-dimethyl a,P-unsaturated amide.77 Of possibly greater importance is the use of a chiral butenolide to carry out asymmetric syntheses of the lignans ( +)-burseran (96), ( -)-isostegane (97), and ( + )-steganacin (98).78 In what can be regarded as a logical extension of this approach Kende et ~1 have employed conjugate addition of an aryl-lithium reagent and intramolecular trapping by a benzyl bromide to prepare the aryltetralin system directly, as shown in Scheme 18.75 A. Pelter, R. S. Ward, P. Satyanarayana, and P. Collins, Tetraheclron Lett.. 1982, 571. 7e A. G. Gonzalez, J. P. Perez, and J. M. Trujillo, Tetrahedron. 1978, 34, 101 I. 77 G. B. Mpango and V. Snieckus, Trtruhvdron Lett., 1980, 4827. " K. Tomioka, T. Ishiguro, and K. Koga. Tcrrahrdron Lctt., 1980, 2973, and J. ChiJm.Soc. Chrm. Commim., 1979, 652. "A. S. Kende, M.L. King, and D. P. Curran, J. OrR. Chrm.. 1981, 46, 2826. 104 .~ Ward PhS SPh PhS SPh OMe 1-1 (94; R = H) PhS SPh PhS SPh R' R2 R4 R' R2 (95) The Synthesis of Lignans and Neolignans Me0 OMe OMe Me0 (96) (97) PO Me0 7 Alkylation and Acylation of Monotactones Several syntheses of lignans are closely related to the conjugate addition re-actions described in the last section in that they involve treating an enolate anion of a benzylbutyrolactone with an alkylating or acylating agent.Thus, the reaction of the anion derived from 4-(3’,4’-methy1enedioxybenzyl)butyro-lactone (99) with an aromatic aldehyde furnishes an a-hydroxybenzylbutyro-lactone derivative (loo),which can be cyclized under acidic conditions to give an aryltetralin [e.g. (84)].80 The same reaction can also be carried out in an intramolecular fashion starting from a biaryl derivative and leading, as indicated, to picrostegane (102)and isopicrostegane (103)F Two other examples of the condensation of an aromatic aldehyde with a benzylbutyrolactone are included in the next section (see Scheme 24).The chiral monolactones (104)and (105) have been used as a starting point for asymmetric syntheses of several lignans including (+)-podorhizon, (-)-podorhizol, (-)-isodeoxypodophyllotoxin,(+)-isosteganone, and (-)-steganone (Schemes 19 and 20).82-83 E. Brown, M. Loriot, and J.-P. Robin, Tetrahedron Lett., 1979, 1389: cf: E. Brown, J.-P. Robin and R. Dhal, J. Chem. SOC.,Chem. Commun., 1978, 556. 81 E. Brown and J.-P. Robin, Tetrahedron Lett., 1978, 3612:cf. E. Brown, R. Ihal, and J.-P. Robin, ibid., 1979, 733. 82 K. Tomioka, H. Mizuguchi, and K. Koga, Tetrahedron Lett., 1978, 4687 and K Tomioka and K. Koga, ibid., 1979, 3315. 83 J.-P. Robin, 0. Gringore, and E. Brown, Tetrahedron Lett., 1980, 2709. Ward OMeOMe 0 CO2Et OMeMe0 QOMe OMe 1. H30+ 2.CrOl 3. OH 4.CHzO 0 (Om00 Me0 OMe OMe LiAl H(OR), c---1. R,SiCI 2. LDA 3. pyHCl 4. Et,N.HF Me0 OMe OMe OMe podophyllotoxin picropodophyllin Scheme 18 The Synthesis of Lignans and Neolignans 7 0<“q0 (99) / \“y-J-$ OR (100) H+ CfC (Ar = 4-benzyloxyphenyl) OH ro -1. LHDS -1 Me0 2. H,/Pd-C Me0 0 + Me0 Me0 108 Ward ArCO. CO.Et -1 LDA LHDS ArCHO OMe (+)-podorhizon1 It)~~(Ar = 3,4,5-trimethoxyphenyl) OMe (-)-podorhizol and (-)-isodeoxy podophyllo toxin (-)-epipodorhizol Scheme 19 Condensation reactions involving y-lactones of a rather different type are involved in Munakata et aZ.’s syntheses of arylnaphthalene lactones, which are shown in Schemes 21 and 22.*4 The scope and mechanism of the latter reaction sequence, involving as it does a most unusual rearrangement. process, has been further studied by Ayres et al.85 84 K.Munakata, S. Marumo, K. Ohta, and Y.-L.Chen, Tetrahedron Lett., 1967, 3821 and Agric. Biol. Chem. (Jpn.), 1971, 35, 431. B5 D. C. Ayres and J. W. Mundy, J. Chem. SOC.(C), 1969, 637. The Svnthesis qf Lignuns and Neolignans (105; X= H) Me0 (106; X= I) 1. I,IIDS 2.Cr0313. Oll Me0 Me0 ."CH OH .CH,OH HMe0 Me0 Me0 Me0 (+)-isosteganone (-)-st eganone Scheme 20 I10 Ward OMe OMe I .NaOMe 12. "+ -( Oqc02H o\OMe OMe I OMe OMe retrochinensin (Ar = 3,4-methylenedioxyphenyl) Scheme 21 The Synthesis of Lignoris and Neolignans H -Me0 Me0 Ar Pb(OAc),1 Me0 Ar (Ar= 3,4-methylenedioxyphenyl) Scheme 22 8 Stobbe Condensation The Stobbe condensation has been frequently used to construct the basic lignan ~keleton,~~-~~and can be illustrated by Crombie's syntheses of (-) -cubebin and Ward related compounds (Scheme 23).86A similar approach was also used to prepare the various isomers of dihydroguiaretic acid dimethyl ether,87 and has been used more recently by Stevenson et a1.88to prepare the aryltetralins nirtetralin and hypophyllanthin (Scheme 24). The structures of the last compounds, which occur together in Phyllunthus niruri, have been the subject of much controversy over the years.Further elaboration of the benzylbutyrolactones is achieved by reaction with an aromatic aldehyde as outlined in the last section, and final reduction followed by methylation affords the isomeric lignans, having spectral characteristics identical to the naturally occurring materials. In contrast, the Japanese group of Horii et ~1.899~~have used the Stobbe condensation of a benzophenone derivative, an approach pioneered by Gensler et al.,gl to prepare arylnaphthalene lactones as shown in Scheme 25 and 26. In this case the isomeric arylnaphthalene lactones can be prepared from a common precursor by protecting the formyl group at C-3 as an isoxazole derivative, while reducing the ester group at C-2.9From Chalcones Chalcones contain a basic 15-carbon skeleton and require the addition of an extra three carbon atoms to afford the lignan framework. They are, however, conveniently prepared by condensation of aromatic aldehydes with acetophenones and hence represent useful precursors for the preparation of natural products containing two aromatic rings linked by an aliphatic chain. It has been shown that the basic aryltetralin framework can be constructed by the addition of a two-carbon unit in the form of a sulphonium ylide as shown in Scheme 27.92Rearrangement of the intermediate cyclopropane, which proceeds in a stepwise manner, affords an aryltetralin directly, requiring only the addition of a final one-carbon unit to complete the synthesis of the full carbon skeleton of the aryltetralin lactones, and providing yet another elegant approach to this medicinally important series of compounds. 86 J.E. Batterbee, R. S. Burden, L. Crombie, and D. A. Whiting, J. Chem. SOC.(C), 1969, 2470. A. W. Strecker, J. Am. Chem. Soc., 1951, 79, 3823. P. A. Ganeshpure and R. Stevenson, J. Chem. SOC.,Perkin Trans. I, 1981, 1681 and G. E. Schneiders and R. Stevenson, ibid., 1982, 999. Z.-I. Horii, M. Tsujiuchi, and T. Momose, Tetrahedron Lett., 1969, 1079 and Z.-I. Horii, M. Tsujiuchi, K.-I. Kanai, and T. Momose, Chem. Pharm. Bull., 1977, 25, 1803. Z.-1. Horii, K. Ohkawa, S.-W. Kim,andT. Momose, Chem. Pharm. Bu//.,1968,16,2404; 1969, 17, 1878; 1971,19, 535. s1 W. J. Gensler, C. M.Samour, S. Y. Wang, and F. Johnson, J. Am. Chem. Sac.,1960,82, 1714. ge W. S. Murphy and S. Wattanasin, J. Chem. Soc., Perkin Trans. I, 1981, 2920; 1982, 271 and 1029. The Synthesis of Lignans and Neolignans 1. H21Pd-C Ar-ArCHO TOzH 2. EtOM-H+ C02H HO 0 r MnOzhF Ypd-' AF -Ar Ar (+)-isohinokinin pH I;'ArTAr (-)-hinokinin Ar Ar Ar (->cubebin (-)-dihydrocubebin (Ar = 3,4-methylenedioxyphenyl) Scheme 23 COzEt ArCHOI Ar 1 .resolve 2. reduce ArTCH2OH 'CH20H Ar HzlPd-C ,*CHzOHArT.'CH20H Ar meso-dihydrocubebin Ward I1 JPd-1 c 0) Ar Br I. LDA 2. ArCHO 3. TFA1 nirtetralin CH2 OMe 'CHzOMe (Ar = 3,4-dimethoxyphenyl) h ypophyllanthin Scheme 24 115 The Synthesis of Lignans and Neolignans TCozH fCoPH Ar+C02EtArYoAr -Ar Ar \AcCl HCO Et-NaOH , CO2Et I Ar Ar 1.Brz NHIOH 2. --HBr \3. NaBH, ORI ! COzEt Ar Ar (R = H) taiwanin E I. LAH(R = Me) justicidin F I2. NaOEt Ar justicidin D (Ar = 3,4methylenedioxyphenyl) Scheme 25 Ward (Iqco2HCOzH I. DES KOBut 2. hydrolysis OMe Me0 OMeMe0 ' Me0 Me0 1. H, (one isomer) 2. AcCl -<q 0 I. EtOH-Ht 2. HC02Et-NaFl I 0 3. NaBH,Ar 4. -H*O a-apopicropodophyllin Me0 QOMe ti 30+ Me0 ?H I0Ar picropodophyllin Scheme 26 1 I7 0 9 hlc0 \/OMc 0(Om< ‘COZ Et+-----(3Me0 \ Me0 OMeQOMe OIe OMc podophyllotoxin Scheme 27 10 From Furans and 1,4-Diketones 1,4-Diketones are readily converted into a number of lignan types (Scheme 28).Thus, partial reduction followed by cyclization affords tetrahydrofuran or aryltetralin derivatives, whereas complete reduction affords di benzyl butanes, the stereochemistry of which in general depend upon the configuration of the original diket0ne.~~9~3$94 The dibenzyl butanes can also be subjected to non-phenolic oxidative coupling (section 3) to afford dibenzocyclo-octadienes of the deoxyschizandrin type. 1,6Diketones can also be prepared by the coupling together of P-keto-esters (Scheme 29). This method, which is based on the early work of Knorr, yields a mixture of the threo and erythro isomers of the diketo-diesters that can in most cases be separated by fractional crystallization.Reduction followed by cyclization then affords either the naturally occurring 2,6-or the unnatural 2,4-diaryl-3,7-dioxabicyclo[3.3.0]octane~.~~Cyclization to a furan followed by 93 C. W. Perry, M. V. Kalnins, and K. H. Deitcher, J. Org. Chem., 1972, 37, 4371. w T. Biftu, B. G. Hazra, R. Stevenson, and J. R. Williams, J. Chem. Sor., Perkin Trans. I, 1978, 1147. 95 A. Pelter, R. S. Ward, D. J. Watson, and 1. R. Jack, J. Chem. Soc., Perkin Trans. 1. 1982, 183. 118 Ward 0 0 MehMeN~oM c Ar +-EhYAr+ Me Arc0 COAr ArCO COAr Ar Ar Ar Ar veraguensin MeS02 c1 or Ph I'Br V OMe galbulin (Ar = 3,4-dirnethoxyphenyl) Scheme 28 reduction has also been used in one case to prepare a 2,4-diarylmonolactone (Scheme 30).24 EtO2C COAr ArCO COAr ArCOCH2C02Et +H->-<--H + ,----<--Id ArCO C02Et EtO2C COZEt I.LAH 1. LAH 2. H+ 2. Ht Ar H--H--Scheme 29 119 The Synthesis of Lignans and Neoligrians PhCO COPh Ph H+PhCOCHCOzEt -HHH -EtO2C CO2Et EtO2C CO2Et Scheme 30 Brownbridge and Change have used the disiloxyfuran (107) to prepare dilactones of type (108) and (109) having diequatorial and equatorial-axial configurations respectively. The proportions of the two isomers obtained depend upon the number of moles of TiCI, used and the nature of the aryl group. The unusual enedione, diethyl furoguaioxidin (1 13), has been prepared by sequential introduction of alkoxy-functions into the methyl groups of a 2,5-diaryl-3,4-dimethylfuran (110).97 Aerial oxidation of the furan affords an enolizable enedione (111) which reacts with ethanol to give the ethoxy- substituted compound (112).Repetition of the sequence then affords (1 13). g6 P. Brownbridge and T.-H. Chan, Tetrahedron Lett., 1980, 3427. P. Majumbar and M. Bhattacharyya, J. Chem. Soc., Chem. Commun., 1975, 702. Ward I I V Arc0 COAr 11 From Biphenyl and Phenanthrene Derivatives The main group of lignans containing a biaryl linkage is the dibenzocyclo- octadiene group of which steganone, schizandrin, and kadsurin are members. One approach to the synthesis of such compounds, which has already been dealt with in section 2, involves the formation of the biaryl linkage at a late stage in the synthesis by oxidative coupling.An alternative approach is to start from a simple biphenyl derivative and proceed to form the eight-membered ring and the remainder of the molecule at a later stage in the synthesis (cf. section 6). Further examples of this approach are illustrated in Schemes 31 and 32.g8-l02 Two main methods have been used to complete the synthesis of the cyclo- octadiene ring, using either malonic esters (Scheme 31) or Wurtz coupling (Scheme 32). The biphenyl'derivatives can be obtained either by the standard or byUllmann coupling rnethod~,80~81~83~98~~~the oxidative cleavage of phenant hrene derivatives (see Scheme 32). l00-l O2 Another elegant way of proceeding from a phenanthrene derivative to the dibenzocyclo-octadiene series involves the ring expansion of an enamine by cycloaddition to dimet hyl acetylenedicarboxylate (Scheme 33).lo3 Subsequent conversion of the enamine into a ketone and of the diester into a lactone produces steganone and isosteganone. 98 N. K. Kochetkov, A. Ya. Khorlin, and 0. S. Chizhov, Izv. Akad. Nauk SSSR, Otd. Khim., 1962, 856 and Izv. Akad. Nauk SSSR, Ser. Khim., 1964, 1036. 9Q F. E. Ziegler, K. W. Fowler, and N. D. Sinha, Tetrahedron Lett., 1978, 2767. loo M. Mervic and E. Ghera, J. Am. Chem. SOC.,1977, 99, 7673. Io1 E. Ghera, Y. Ben-David, and D. Becker, Tetrahedron Lett., 1977, 463 and M. Mervic, Y. Ben-David, and E. Ghera, ibid., 1981, 5091. Io2 E. Ghera and Y. Ben-David, J.Chem. Soc., Chem. Commun., 1978,480. loa D. Becker, L. R. Hughes, and R. A. Raphael, J. Chem. Snc., Perkin Trans. I, 1977, 1674; CJ E. R. Larson and R. A. Raphael, ibid., 1982, 521. The Synthesis of Lignans and Neolignans Me0 CO Me 1. 26aCH(C02Me)z 1. Br, 3. NaOEl _.____) Me0 MeoQIleO H2Br Me0 Me0 5 LAll Me0 Me Me0 Me0 Me Me0 Me0 CH2CH(C0~Me), Me0 Me0 Me0 steganone heat isosteganone Scheme 31 122 Ward Me0 Me0 Me0 MeoqI us04 Me0 COCHBrMe 2. so3 py =-3. I’tMgBrMe0 4.Ph(OAc), Meo@COCH Br Me 5. BIL Me0 Me0 Me0 Zll Cll Me0 Me0 I 1 LiAIH(OR), Me0 7 tIJl’d c Me0 ‘-Me0 Me0 Me / /Me0 Me0 OMe Me0 deoxyschzandrin Me0 Me0 Me0 -Me0Meo@Me \ e Me0 I.0~01 Me0Me 2. MsCI \ Me3. NaBH, /Me0 Me0 Me0 Me0 schizandrin Scheme 32 123 The Synthesis of Lignans and Neolignans ro0 CO,Me DMAD ___jMe0 Me0 C02Me I Me0 Me0 MeOHiHCI r? I. RaNi Me0 C0,MeMeowMe0 Me0 Me0 isosteganone 0 Me0 Me0 steganone Scheme 33 12 Miscellaneous An alternative approach to futoenone (1 16) starts from a benzofuran derivative (1 14).lo4 Cyclization of the dihydrobenzofuran (1 15) with expulsion of a p-toluenesulphonate group supplies the third ring of the tricyclic framework (Scheme 34). Io4 A. Ogiso, M. Kurabayashi, A. Taka.iashi, H. Mishima, and M. C. Woods, Chem. Pharm. Bull., 1970, 18, 105. Ward ArCH,COMe CHO PhCHsO COMe (114) .1H2 COMe Me0 oCHzPh\ 1.TsCl (115) 3. baseyzIe8;io Me0 0 Scheme 34

ISSN:0306-0012    

DOI:10.1039/CS9821100075

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

The Thermolysis and Photolysis of Diazirines By Michael T. H. Liu DEPARTMENT OF CHEMISTRY, UNIVERSITY OF PRINCE EDWARD ISLAND, CHARLOTTETOWN, PRINCE EDWARD ISLAND, CANADA, C1A 4P3 1 Introduction The chemistry of diazirines has a relatively short history dating from the dis- covery of these compounds in 1960 by PauIsen,l and Schmitz and Ohme.2 Since then the three-membered ring isomers of diazoalkanes have received increasing interest and the progress in this field has been reviewed.3-11 The structure of the three-membered ring compounds was first elucidated in 1962. Both the dipole moment and the quadrupole coupling constants were reported and shown to be consistent with a cyclic structure.12 It was also demonstrated that the two nitrogen atoms are equivalent.l3 The mode of preparation of diazirines depends on their functional groups.Alkyl and aryl diazirines are easily synthesized by using the appropriate aldehyde or ketone [equation (l)].14 Graham reported that the oxidation of amidines by hypochlorites (or hypobromites) gives chloro- (or bromo) -diazirines.l5 Spectral characteristics are most commonly used in the identification of these compounds. For example, in the infrared absorption spectrum the N=N stretching frequency is 1560-1 585 cm-l whereas the ultra-violet absorption is consistently in the 350-400 nm range. These cyclic compounds are remarkably stable towards organic and inorganic reagents,l6 in sharp contrast to the behaviour of their linear isomers. They are S. R.Paulsen, Agnew. Chem., 1960,72, 781. E. Schmitz and R. Ohme, Angew. Chem., 1961, 73, 115. E. Schmitz, Adv. Heterocycl. Chem., 1979, 24, 63. * E. Schmitz, ‘Dreiringe mit Zwei Heteroatomen’, Springer-Verlag, New York, 1967. P. S. Engel, Chem. Rev., 1980, 80, 99. (I H. M. Frey, Adv. Photochem., 1966, 4, 225. S. Braslavsky and J. Heicklen, Chem. Rev., 1977, 77, 473. H. Meier and K.-P. Zeller, Angew. Chem., Znt. Ed. Engl., 1977, 16, 835. J. B. Moffat, in ‘The Chemistry of Diazonium and Diazo Groups’ Pt. 1, ed. S. Patai, Wiley-Interscience, New York, 1978. lo W. Kirmse, ’Carbene Chemistry’, Academic Press, New York, 1971. l1 M. Jones Jr. and R. A. Moss, ‘Carbene’ Vol. 1, Wiley-Interscience, New York, 1973. l2 L.Pierre and V. Dobyns, J.Am. Chem. SOC., 1962, 84, 2651. l3 E. Schmitz, R. Ohme, and R. D. Schmidt, Chem. Ber., 1962,95,2714. l4 E. Schmitz and R. Ohme, Tetrahedron Lett., 1961, 612. l5 W. H. Graham, J. Am. Chem. SOC.,1965, 87,4396. l6 It is advisable to take extreme precautions when working with neat diazirine in all phases. It is recommended that all diazirines should be handled in diluted state only. cf. M. T. H. Liu, Chem. Eng. News, Sept. 9, 1974, p. 3; M. J. Leleu, Cahiers De Notes Documentaires No. 93, 1978, p. 569. The Thermolysis and Photolysis of Diazirines C,H,CH2NH2 R’\ R’ ‘c=o -,C=N,P-TsOHR2’ R2 Me NHz/NH20SO,H1 resistant to alkalis or strong acids and are only decomposed by 80% sulphuric acid. Because of their stability, diazirines have been investigated as a source of carbene following thermolysis or photolysis.Most photochemical studies have been carried out in the gas pha~e,~?~ but many thermal decompositions have been studied both in the gas phase17-21 and in solution.22-26 Reports have appeared on the photolysis of diazirines trapped in solid mat rice^,^^*^* in which the nature of the intermediate formed during the de- composition was studied. The first examples of substituted diazirinyl radicals29 have also been obtained by photolysis of the parent bromides in the presence of hexa-n- bu t yldi t in. In this review the kinetic aspects of thermolysis and photolysis of diazirines are discussed, as these two topics compiement each other in the understand- ing of the underlying mechanisms. 2 Thermolysis of Diazirines Frey and Stevens21 have investigated the gas-phase thermal decomposition of 3,3-dimethyldiazirines over the range 124-174 “C.They found that the reaction is homogeneous and first-order and gives only propene and nitrogen. Propene does not inhibit the reaction; thereby providing evidence for the absence of any radical-chain contribution and strongly suggesting that the decomposition is simple and unimolecular. The decomposition is pressure-dependent below l7 H. M. Frey and A. W. Scaplehorn, J. Chem. SOC.,A, 1966, 968. la E. W. Neuvar and R. A. Mitsch, J. Phys. Chem., 1967, 71, 1229. Is M. R. Bridge, H. M. Frey, and M. T. H. Liu, J. Chem. SOC.,A, 1969, 91. ao H. M. Frey and M. T. H.Liu, J. Chem. SOC.,A, 1970, 1916. a1 H. M. Frey and I. D. R. Stevens,J. Chem. Sor., 1962, 3865. M. T. H. Liu and K. Toriyama, Int. J. Chem. Kinet., 1972, 4, 229. *3 M. T. H. Liu and K. Toriyama, J. Phys. Chem., 1972,76, 797. B. M. Jennings and M. T. H. Liu, J. Am. Chem. SOC.,1976, 98, 6416. M. T. H. Liu and B. M. Jennings, Can. J. Chem., 1977, 55, 3596. N. P. Smith and I. D. R. Stevens, J. Chem. Soc., Perkin Trans. 2, 1979, 213. C. B. Moore and G. C. Pimentel, J. Chem. Phys., 1964,41, 3504. lSO.L. Chapman, C. C. Chang, J. Kolc, M. E. Jung, J. A. Lowe, T. J. Barton, and M. L. Tumey, J. Am. Chem. SOC.,1976,98, 7844. Y.Maeda and K. U. Ingold, J. Am. Chem. SOC.,1979, 101, 837. Liu Table 1 Arrhenius parameters for the thermnl decomposition of 'diazirines R1 R2 Condit iotts log A E/kcal mol-1 Ref: Me Me 4 mm 13.89 33.17 21 F F 200 mm 13.1 32.20 18 Et Et 10 mm 13.73 31.89 17 -(cH2)4- 7 mm 13.40 30.50 17 -(cH2)5- 5 mm 13.34 30.87 17 Me CI 4 mm 11.71 27.18 19 Me CI SO mm diazirine and 14.14 3 1.07 19 300 mm Et c1 perfluoropropane 20 mm diazirine and 13.99 30.45 20 60 nun n-butane Prn CI 10 mm diazirine and 13.98 30.98 20 20 mm butane Pri c1 10 mm diazirine and 14.01 30.59 20 20 mm butane But c1 4 mm 13.36 29.50 20 CCI3 CI cc14 13.8 29.20 22 CCI 3 CI iso-octane 13.8 29.00 22 Ph Ph Ph Br c1 c1 cyclohexene dimethyl sulphoxide cyclohexene 13.75 14.1 1 13.87 27.36 28.1 1 28.00 23 23 23 p-MeOCeH4 CI cyclohexene 13.41 26.49 34 p-MeC6H4 CI cyclohexene 13.94 27.87 34 p-CIC 6H 4 P-NOZC 6H4 CI c1 cyclohexene c ycl ohexene 13.81 13.80 27.72 27.85 34 34 m-CIC 6H 4 CI cyclohexene 13.56 27.58 35 m-MeCsH4 p-MeC6t-h Cyclopropy1 CI c1 c1 cyclohexene acetic acid 2 mm 13.80 13.81 14.76 27.85 27.61 29.83 35 35 36a Cyclopropyl Cyclopropy1 Cyclopropyl Br c1 c1 2 mm cyclohexene cyclohexene 13.70 13.64 13.53 27.58 27.49 29.1 1 36a 35 35 C2H 3 Me cc14 13.70 25.60 37 CsH 3 Me dimethyl formamide 13.60 25.40 37 C2H 3 Me ethanol 13.70 25.60 37 Bun Ph dimethyl sulphoxide 13.26 28.08 25 Me Me0 Ph c1 dimethyl sulphoxide MeOH 13.45 13.52 28.50 24.16 25 45 Me0 PhCHz CI CI gascc14 13.06 13.82 25.26 29.60 45 62a 129 The Thermolysis and Photolysis of Diazirines 100mm, but at 100 mm it has virtually reached the high pressure limit.At 4 mm pressure an activation energy of 33.17 kcal mol-l and a log A value of 13.89 were reported. The Arrhenius parameters for these compounds will be discussed (Table 1). Frey and Stevens favoured a mechanism involving a concerted elimina- tion of N2 in which dimethylcarbene is formed as an intermediate. The results of Bottomley and N~berg,~O and those of Frey and Stevens,21 after adjustment to infinite pressure, fit a single Arrhenius line. Bottomley and Nyberg concluded that the transition state is considerably looser than the initial configuration, is non-cyclic, and is probably free radical in nature.The gas-phase decomposition of difluorodiazirine was studied by Neuvar and Mitsch.18J1 They found that the decomposition rate constant at a total pressure of 200 mm is k = 1013.1exp(-32 ZOOIRT) s-l. Although the data do not reveal the mechanism, Neuvar and Mitsch nevertheless suggested a mechanism in-volving the transient existence of a diazomethane intermediate prior to loss of nitrogen. Frey and Scaplehorn17 from their study on 3,3-tetra-3,3-pentamethylene-diazirine and 3,3-diethyldiazirine conclude that all the reactions are homogeneous and first-order, and probably unimolecular. Bridge, Frey, and Lidg from their further studies on the decomposition of 3-chloro-3-methyldiazirinefound that the products, vinyl chloride and nitrogen, are formed quantitatively in a first-order reaction.Substitution by a chlorine atom at C-3 resulted in a stabilization of 2 kcal mol-l. Although the reaction mechanism was not clear, the authors favour a carbene intermediate. The exothermicity on the decomposition of 3-chloro-3-methyldiazirinewas discussed and further examined by Archer and Tyler.32 Frey and Liu20 also studied the thermal decomposition of several chlorodiazirines. The rate constants were determined in the high pressure region and were shown to be kinetically first- order. The experimental evidence on the pyrolysis of chlorodiazirines indicates that a carbene is formed, which gives products by the concerted elimination of nitrogen. Although kinetic data were available for the gas-phase decomposition of diazirines prior to 1970, none were recorded for decomposition in solution.In keeping with the growing interest in diazirines as a source of free carbenes, Liu and Toriyama22r23 carried out investigations of in solution decomposition of diazirines to seek information about the nature of such strained ring systems. The decomposition of 3-chlor0-3-trichloromethyldiazirine~~in carbon tetrachloride and iso-octane was investigated over the temperature range 75-1 15 "C. The only products were nitrogen and tetrachloroethylene. The rate was shown to be invariant in a variety of solvents. Similarly, decomposition of 3-chloro- and 3-brorn0-3-phenyldiazirines~~in several solvents gave nitrogen and the corresponding carbene.The carbene can either react with diazirine giving a dimeric product, or with the solvent cyclohexene to yield 7-chloro-7- 8o G. A. Bottomley and G. L. Nyberg, Aust. i. Chem., 1964, 17, 406. 31 R. A. Mitsch, J. Heterocyclic Chem., 1964, 1, 59; 1966, 3, 245. 3p W. H. Archer and B. J. Tyler, J. Chem. Soc.., Furaduy Trunv. I, 1976, 72, 1448. 130 Liu phenylnorcarane and the products of insertion iit the C-3 and C-4 positions (-1 :l). This study indicates that the activation cncrgy of 3-chloro-3-phenyl-diazirine is lowered by -3 kcal mol 1 comparcd to that of 3-chloro-3-methyl-diazirine. This difference may be attributed to the resonance stabilization afforded by the phenyl group on the transition state for diaiirinc decomposition.The value of the A factor of -1014 s-' (ASt = +3 cal K 'mol 1) is typical for a unimolecular reaction having a relatively tight transition state. As the rate of decomposition does not change with solvent polarity, the rate-determining step is likely to be radical rather than ionic. The fact that the Arrhenius para- meters in the gas phase and in solution have similar values suggests that the transition state for all the diazirine decompositions is identical.z2,z:3 This con- clusion is further corroborated by studies of the decomposition of 3-chloro-3- hasethyldiazirine, in which the ratio kgas/kcyciohexene a value 10°,17exp (-0.85/RT).33 Liu and T~riyama~~ examined the effect of puru substituents on the thermal decomposition of a series of 3-chloro-3-aryldiazirinesin various sol-vents.They found that the activation energies of 3-chloro-3-aryldiazirinesand 3- chloro-3-alkyldiazirinesare nicely explained by the different resonance stabiliza- tions of the transition states I, 11, and 111. The observed effect of substituents I I1 I11 on rates is in the following order: p-Me0 > p-Me > p-CI > H 31 p-NO2. Transition state I was proposed by Frey et 01.19 to be the simplest case for diazirine decomposition. Schmitz3 argued, however, that as the electron distribution on the carbon atom in transition state 1 must tend towards a sextet, it cannot be stabilized by electron withdrawal. If this were so, the order of the rates would be expected to be p-Me0 > p-Me > H > p-CI =-p-NOz, which is not in agree-ment with the observed results. Schmitz3 has postulated instead an ionic transi- tion state, 11, as being more likely in view of the greater instability of a-keto-pentamethylenediazirine with pentamet h ylenediazir ine.If the react ion proceeds by transition state XI, the rate should be: p-NO2 > p-CI > H > p-Me > p-MeO; unfortunately this order is the exact reverse of that observed. Finally, Liu and T0riyarna3~ concluded that the best transition state is 111. Experimentally, a Hammett correlation was determined for the series but it did not fit with either o or a+ constants.34 However, a smooth curve was obtained for all points except for the parent compound. On plotting log k vs o+ --a a straight line was obtained.34 This result is in keeping with transition state 111.Liu and Chien35 examined the thermal decomposition of mera-substituted 3-chloro-3-aryldia- 33 M. T. H. Liu and D. H. T. Chien, Can. J. Chem., 1974, 52, 246. 34 M. T. H. Liu and K. Toriyama, Can. J. Ch~rn.,1972, 50, 3009. R5 M.T. H. Liu and D. H. T. Chien, J. Clrrvn. Soc.. Perkin Trms. 2. 1974. 937. 131 The Thermolysis and Photolysis of Diuzirines zirine and found that, in contrast to the para-substituted derivatives, the rates were linearly related to their respective o values. Lower rate constants were also observed for the meta-substituted diazirines since the substituents are unable conjugatively to interact with the transition state.In fact, the data fall into two categories: (i) the metu-substituted derivatives (including the unsubstituted compound) and (ii) the pam-substituents. The results are again consistent with a polarized, radical-like transition state 111. The extent of the polarization depends on the diazirine ring substituent. The thermolysis of 3-halo-3-cyclopropyldiazirinesin the gas phase and in s0lution3~ give, respectively, nitrogen and 2-halobuta-l,3-diene and nitrogen and halocyclobutene, together with traces of the adduct from the cyclopropyl- halocarbene and halocyclobutene. These results are conveniently interpreted in terms of vibrationally excited halocyclobutene molecules. In the gas phase, the excited halocyclobutene isomerizes completely to 2-halobuta-l,3-diene on account of its inherent instability.In solution, on the other hand, complete deactivation and consequently formation of the halocyclobutene occurs. In general, the activation energy for the bromo-compound is lower than that for the chloro-compound.23~36 The thermal decompositions of 3-chloro-3-cyclopropyl- and 3-chloro-3-cycloheptyldiazirine35 in cyclohexene depend on the substituents. The rate constant for the decomposition of the cyclopropyl compound is an order of magnitude greater than for the diazirines containing cycloheptyl or alkyl substituents. The most convincing argument for a step-wise mechanism (transition state 111) is provided by the thermal isomerization of 3-methyl-3-vinyldia~irine.~~Mainly 3-methylpyrazole is formed in a first-order isomerization in various solvents [equation (2)].The fact that solvent effects and entropies of activation for the c,1I H isomerization of 3-methyl-3-vinyldiazirine and 3-chloro-3-phenyldiazirine are similar points to a common transition state for both reactions. Since isomeriza- tion can only proceed by cleavage of a single C-N bond,38 it can be concluded that the two C-N bonds in diazirine break consecutively. Although evidence is sufficient for a stepwise mechanism in the present case, it is not certain whether as W. J. Engelbrecht and S. W. J. Van Der Merwe, J. S. African Chem. Inst. (a) 1975,28, 144; (b) 1975, 28, 148; (c) 1975, 28, 158. 37 M. T. H. Liu and K. Toriyama, Can. J. Chem., 1973, 51, 2393.38 An alternative pathway for this isomerization can be described since a concerted rearrange- ment such as a -1,3-sigmatropic shift (suprafacial for C-I, C-2, and C-3, antarafacial for N-I), not requiring the intermediacy of diazopropene is also possible. The possibility of such process is under investigation by M. T. H. Liu. Liu the intermediate is a radical or an ion. The small solvent effect observed has been discussed in terms of the compensation in net polarity of the dia~o-intermediate.~~ The formation of diazomethane on thermal decomposition has always been inferred, since no diazomethane intermediate has ever been isolated from diazirine thermolysi~.3~+35,39~~~ Although an attempt to synthesize 3,3-diphcnyl- diazirine41 resulted in the formation of diphenyldiazomethane, the precursory existence of a diazirine was never verified.The first isolation of a diazomethane was reported by Jennings and Liu24 for the case of phenyl-n-butyldiazirine (1) (Scheme 1). The thermal decomposition of (1) in dimethyl sulphoxide at 100“Cgave nitrogen and cis-and trans-1-phenyl-pent-1-enes (cis :trans = 1 :5). The kinetic parameters25 for the isomerization and decomposition in dimethyl sulphoxide of (l), kl and k2, together with other experimental evidence showed that almost all diazirine (1) in dimethyl sulphoxide decomposes via the diazo- compound (2). The isomerization of phenylmethyldiazirine to 1-phenyldiazo-ethane is first order25 and probably unimolecular, but the kinetics for the subse- quent reactions of ‘l-phenyldiazoethane are complicated by several competing rate proces~es.~~~~~ The distribution of azine and cyclopropane products depends on solvent polarities.42 Azine arises from the dimerization of 1 -phenyldiazo- ethane.43 Bradley, Evans, and Stevens44 found that cycloalkanespirodiazirines decomposed giving the corresponding cycloalkylidenes, which rearrange intra- molecularly to cycloalkanes and bicycl oal kanes.3,3-Pentamethylenediazi rine, on heating in acetic acid, yielded cyclohexyl acetate (23%) and cyclohexene (77 %).44 These findings permit the conclusion that cycloalkanespirodiazirines thermolyse by concurrent one- and two-bond cleavage. Smith and Stevens45 observed that 3-chloro-3-methoxydiazirinesin thc gas phase and in solution undergoes thermolysis regardless of the phase or the polarity of the solvent, in agreement with the’result of Liu and Chien.The stabilization of the carbene centre by both chloro- and methoxy-substituents persuaded 3B E. Schmitz, C. Horig, and C. Grundemann, Chem. Ber., 1967, 100, 2093. 40 E. Schmitz, ‘23rd International Congress of Pure and Applied Chemistry’ Vol. 11, Butter-worths, London, 1971, p. 283. 41 C. G. Overberger and J.-P AnseIme, Tetrahedron Lett., 1963, 1405. 4z M. T. H. Liu and K. Ramakrishnan, J. Org. Chem., 1977, 42, 3450. 43 M. T. H. Liu and K. Ramakrishnan, Tetrahedron Lett., 1977, 3139. 44 G. F. Bradley, W. B. L. Evans, and I. D. R. Stevens, J. Chem Soc., Perkin Trans.2, 1977, 1214. 46 N. P. Smith and I. D. R. Stevens, J. Chem. Soc., Perkin Trans. 2, 1979, 213. 133 The Thermolysis and Photolysis of Diazirines Smith and Stevens to favour a simultaneous two-bond rupture. The thermal decomposition of phenyl-n-butyldiazirine in the presence of mera-chloroper-benzoic acid (MCPBA) gave cis-and trans-1-phenylpent-1 -ene oxides and valer~phenone.~~*~~These products together with the kinetic data indicate that the oxidation does not take place on the diazirine ring, but rather that attack occurs on 1-phenyl-1-diazopentane. To date, all the data suggest that either the carbenic or the diazo or both pathways for the thermal decomposition of diazirine depend on (i) the nature of the diazirine ring substituents, (ii) the stability of the intermediate diazo- compound, and (iii) the ability of the solvent molecules to stabilize the carbene or diazo-intermediate.3 Photolysis of Diazirines The mechanism of photolysis of diazirine is similar to that for thermolysis. However, because the energy of the light quanta are greater than the thermal energy of activation, the products contain excess internal energy which results in further decomposition or isomerization occurring. A very detailed review on the photolysis of diazirines was prepared by Frey6 in 1966. The early studies mainly concerned the nature of products and the chemical processes involved in their formation. The primary products are carbenes and molecular nitrogen. The resulting carbenes can be captured when saturated or unsaturated hydrocarbons are present.For example, the photolysis of 3H-diazirine in the presence of excess cyclobutane48 yielded methylcyclo- butane, ethylene, and propylene, whereas photolysis alone yielded ethylene and nitrogen. These observations were rationalized by Frey and Stevens who suggest that the mechanism involves the insertion of methylene to cyclobutane to give excited methylcyclobutane which is then deactivated or decomposes to give ethylene and propylene. The formation of ethylene when diazirine is photolysed alone is explained by the attack of methylene on diazirine. Although the formation of diazomethanes as intermediates during the photo- lysis of 3-substituted diazirines is unquestioned,6 their existence on photolysis of diazirine itself has been controversial.Amrich and Bell49 photolysed 3H-diazirine in the gas phase under nitrogen using monochromatic light of -320 nm. By monitoring the u.\. spectrum they detected the formation of diazomethane and they claimed that most of the excited diazirine decomposed to methylene and nitrogen while some isomerized to the excited diazomethane. The latter could either decompose to methylene and nitrogen or be stabilized by collision. From the quantum yields of products they concluded that about 20% of the primary decomposition of diazirine proceeds by isomerization to diazomethane. These results are at variance with those of Moore and Pimentel5O who carried 46 M.T. H. Liu and I. Yamamoto, Can.J. Chem., 1979, 57, 1299. 47 M. T. H. Liu, G. E. Palmer, N. H. Chishti,J. Chem. Soc., Perkin Trans. 2, 1981, 53. 48 H. M. Frey and 1. D. R. Stevens, Proc. Chem. SOC.,1962, 79. 49 M. J. Amrich and J. A. Bell, J. Am. Chem. SOC.,1964, 86, 292. j"C. B. Moore and G. C. Pimentel, J. Chrm. Phj~..1964, 41, 3504. 134 Liu out the photolysis of diazirine in a solid nitrogen matrix. They detected diazo- methane, but concluded from results using a 15N matrix that it derived from the reaction of metkylene with nitrogen. In thc case of 3-methyldia~irine,~l Frey and Stevens comidered that the diazirine molecule initially fragmented to ethylene and methyl carbene. The carbene could first rearrange to vibrationally excited ethylene, and then to acetylene and hydrogen.This mechanism predicts that at sufficiently low pres- sures the yield of acetylene will tend to 100%. However, a plot of acetylene yield as a function of pressure' extrapolates only to 60.2 %. Consequently there must be an additional reaction leading to ethylene which is incapable of de- composing to acetylene and hydrogen. They found no evidence for the isomeri- zation to diazoethane during the photolysis. Frey and Stevens later studied the photolysis and pyrolysis of 3-methyl-3- ethy1diazirine,s2 3,3-diethyldiazirine,s3 3-methyl-3-isopropyldiazirine,and 3-t-b~tyldiazirine.5~The product ratios are different for the photolysis and pyrolysis reactions, which can be attributed to the formation of a 'hot' carbene in the photolysis reactions.Difluorodiazirine on phot~lysis~~using a medium pressure mercury arc lamp (GE-AH4) in inert matrices at 4-20 K generated CF2 free radicals and N2 as the principal products. On warming to 20 K they observed C2F4, un- doubtedly derived from the recombination of free radicals. Irradiation of the atsame reactant in a cryogenic mass spe~trometer5~ -139 "C gave small amounts of cyclic compounds, presumably from the addition of CF2 to F2C-N-N-=CF2. The gas-phase photolysis of 3-chloro-3-methyldiazirine57gave vinyl chloride, acetylene, hydrogen chloride, 1,1 -dichloroethane, and nitrogen. Photolysis with added hydrogen bromide led to the formation of 1 -bromo-1-chloroethane. Clearly the formation of 1,l-dihalo-derivatives arise by the reaction of the intermediate carbene with hydrogen chl0ride.5~ From the results of photolysis of 3-chlor0-3-methyIdiazirine,~~Frey and Penny further argued that 1,l-dichloroethane arose from the reaction of methyl- chlorodiazomethane, formed by photoisomerization from diazirine with HCI.Cadman et aZ.59 from a study on the photolysis of 3-chloro-3-methyldiazirine favour a statistical distribution for the energy dispersion. Subsequently, Figuera and co-workers60 questioned the validity of this statistical distribution and proposed a mechanism involving two pathways leading to vinyl chloride. 61 H. M. Frey and I. D. R. Stevens, J. Chem. Soc., 1965, 1700. 52 H. M. Frey and I. D. R. Stevens, J. Am. Chem. SOC.,1962, 84,2647.53 A. M. Mansoor and I. D. R. Stevens, Tetrahedron Lett., 1966, 1733. 54 H. M. Frey and I. D. R. Stevens, J. Chem. SOC.,1965, 3101. 55 D. E. Milligan, D. E. Mann, M. E. Jacox, and R. A. Mitsch, J. Chem. Phys., 1964,41, 1199. 56 S. S. Cristy and G. Mamantov, Int. J. Mass Spectram. Ion Phys., 1970, 5, 319. 57 W. E. Jones, J. S. Wasson, and M. T. H. Liu, J. Photochern., 1976, 5, 311. 58 H. M. Frey and D. E. Penny, J. Chem. SOC.,Faradny Trans. I, 1977, 2010. 5D P. Cadman, W. J. Engelbrecht, S. Lotz, and S. W. J. Van der Merwe, J. S. African Chem. Inst., 1974, 27, 149. 6o J. M. Figuera, J. M. Perez, and A. Tobar, J. Chem. SOC.,Faraday Trans. I, 1978, 809. The Thermolysis and Photolysis of Diuzirincv In the gas-phase photolysis of pentamethylenediazirine,61 the products are cyclohexene, bicyclo[3.1 .O]hexane, methylenecyclopentane, butadiene, and ethylene.The mechanism proposed involves a carbene rearrangement generating a ‘hot’ species which decomposes unimolecularly or becomes deactivated by collision. Normally butadiene and ethylene arise by cycloreversion of excited cyclohexene; however, in the presence of a large excess of nitrogen, this process is virtually eliminated. In the liquid-phase photolysis of cycloalkanespirodia-zirines, Stevens and co-~orkers~~provided evidence for the formation of diazocycloalkane in the initial phases of the reaction. The photolysis of penta-methylenediazirine in acetic acid and in [2H]acetic acid showed that 59% of the diazirine decomposed via the diazonium ion (Scheme 2) and the remaining 41 % ’ ‘H MeC0,H Scheme 2 via the diazo-intermediate (Scheme 3).Accordingly, acetate resulted from the diazonium ion only whereas the alkene could arise from the diazonium ion or the carbene intermediate. * Scheme 3 Liu and co-~orkers~~ have further tested the validity of these schemes by conducting the photolysis of phenyl-n-butyldiazirine in acetic acid and C2H4]- acetic acid. They found that the photolysis of the diazirine in acetic acid pro- ceeded only via Scheme 2, without any participation of the carbene intermediate (Scheme 3). In the thermal decomposition of phenyl-n-butyldiazirine at various 6’ H. M. Frey and I. D. R. Stevens. J. Chem. Sor... 1964. 4700. Liu concentrations of acetic acid,47 the alkene-acetate product ratio is insensitive to changes in acetic acid concentration, thereby suggesting that the products are derived mainly from the reactions of 1-phenyl-1 -diazopentane.However, further work by Liu and Chishti62a on the photochemical decomposition of 3-chloro-3-benzyldiazirine(4)in acetic acid and [2H4]acetic acid supports a carbenic mechanism. The products of the chlorodiazirine were exclusively cis-and trans-/?-chlorostyrene and 1-chloro-2-phenylethylacetate(5) according to equations (3) and (4). PhCH, PhCH, 0,CMe \C/i 3-MeC0,H -PhHC=CHCl + 'c' (3) cis and frans 'HCl '" c1' PhHC =CHCl PhCH, c1'" cis and trans PhCH, \C/i f CD,CO,D + + PhHC=CDCl cis and trans (4) (5) The photolysis of (4) was also carried out at different concentrations of acetic acid in carbon tetrachloride.The a1kene:acetate product ratio varied with the acetic acid concentrations. This result points to chlorobenzylcarbene as the key intermediate (Scheme 4). Laser flash photolysis of 3-chloro-3-aryldiazirinesin the presence of acetic acid at room temperature 62b showed that the chlorophenylcdrbene inserts into the 0--H bond of the acetic acid62c with a quenching rate-constant k4 -2 x lo9 M-l s-~.If the insertion of chlorobenzylcarbene into 0-H bond of the acetic acid is assumed to be diffusion controlled, then the product concentra- tions indicate a rate constant of the order -los M-l s-l for the 1,Zhydrogen shift. Revealing observations have been obtained from studies on diazirines as possible photo-affinity labelling compounds for biological systems.63 Spectral evidence for the diazo-intermediate was forthcoming.Irradiation in acetic acid indicated that it underwent fragmentation to arylcarbene and photoisomeriza- tion to the linear diazo-compound.63a (a) M. T. H. Liu and N. H. Chishti, to be published. (b) D. Griller, M. T. H. Liu, J. C. Scaiano, and P. C. Wong, to be published. (c) An alternative mechanism is the addition of the carbene to the oxygen lone-pair to form an ylide, which then gives product by O+C proton transfer. However, this mechanism is not in keeping with our laser flash photolysis experiments. 63 R. A. G. Smith and J. R. Knowles, J. Chem. SOC.,Perkin Trans.2, 1975, 686; H. Bayley and J. R. Knowles, Biochemistry, 1978,17,2420; J. Brunner, H. Senn, and F. M. Richards, J. Biof. Chem., 1980, 255, 3313; B. Erni and H. G. Khorana, J. Am. Chem. SOC.,1980, 102, 3888. 137 The Thermolysis and Photolysis of Diazirines PhCH, c1 PhCH, ‘C: -I-Nz/ Ph H ’ PhCH, Protonation ‘H c1 cis and trans (5) PhCH, MeC0,-f c1 Scheme 4 Although the photochemical transformation of diazirines into diazo-com- pounds is feasible, the first example of the reverse process was only published in 1971. Lowe and Parker64 showed that N-diazoacetylpiperidine on irradiation with visible light gave N-diazirinylcarbonylpiperidine in about 20 % yield. .~~Franich et ~ 1 have also suggested that this photochemical isomerization is restricted to diazoamides, in which the excited state is stabilized by the amide group.Indeed, such isomerizations have been observed for a few linear65 and cyclic66967 a-diazoamides. Similar isomerization of a-diazoketones was virtually unknown until 1978 when Mukai and co-workers68 reported the reversible photochemical valence isomerization between a-diazoketones and a-keto-diazirines. The stabilization by the amide or n-bond participation has been invoked.68 Ab initio SCF calculations on the photochemical behaviour of dia-zirine69 are in general agreement with experiment. 4 Diazirine as a Source of Carbene Few methods are available for generating ‘free’ carbenes as most methods O4 G. Lowe and J. Parker, J.Chem. Soc., Chem. Commun., 1971, 1135. O5 R. A. Franich, G. Lowe, and J. Parker, J. Chem. SOC.,Perkin Trans. I, 1972,2034. 88 E. Voigt and H. Meier, Angew. Chem., Int. Ed. Engl., 1975, 14,.103. E. Voigt and H. Meier, Chem. Ber., 1975, 108, 3326. T. Miyashi, T. Nakajo, and T. Mukai, J. Chem. Soc., Chem. Commun., 1978,442. B. Bigot, R. Ponec, A. Sevin, and A. Devaquet, J. Am. Chem. Sor., 1978, 100, 6575. produce only carbenoids. Our understanding of the reactivity and spin states of carbenes has benefited significantly from studies of photolysis of diazirines. Photolysis of phenylbrom~diazirine~~ in various alkenes gave essentially con- figurationally pure cyclopropane adducts. Additions to cis-butene were highly stereospecific.Methylchlorocarbene generated by the photolysis of 3-chloro-3-methyldia~irine~ladded to cis- and trans-butene giving the corresponding cyclopropanes in stereospecific manner. Photolysis of 3,3-difl~orodiazirine~~ in alkenes also afforded stereospecific adducts, showing that the difluorocarbene is formed and is reacting as a singlet. It is generally held that if a singlet carbene can undergo intramolecularly an internal insertion or cycloaddition it will not be efficiently trapped by an external reagent.ll In contrast, Moss and M~njal~~ have shown that the carbenes formed by photolysis of 3-chloro-3-ethy1, 3-chloro-3-isopropy1, and 3-chIoro-3- t-butyldiazirine added readily to external alkenes. Intermolecular addition was found to be far more efficient than intramolecular reaction for chloroethyl- carbene and chloro-t-butylcarbene.The reverse trend was observed for chloro- isopropylcarbene. It has been suggested that the stabilization of positive charge at the migration origin74 is necessary for efficient 1 ,Zhydrogen shift. However, the thermolysis of 3-chloro-3-isopropyldiazirine20gave only 1-chloro-2-methyl-propene. Moreover, the thermolysis of is~propyldiazirine~~ afforded nearly equal amounts of methylcyclopropane and isobutene. It follows that the chlorine atom disfavours C-H insertion. Little attention has been lent to solvent effects on 1,Zhydrogen shift to divalent carbon generated from diazirines or diazo-compounds. The rate of thermal decomposition of 3-chloro- and 3-brom0-3-ethyIdiazirine~~is insensitive to the nature of the solvent, but the subsequent reactions of the carbene are atl’ected. Continuing their studies on capture of carbene by alkenes Moss and Fantina75 photolytically generated cyclopropylchlorocarbenefrom cyclopropylchlorodiaz- irine and added it to a variety of alkenes.Additions to both cis- and trans- butene were stereospecific. Stabilizing factors such as the dual interactions of the chlorine lone-pair and the ‘bent’ a-bonds of the cyclopropyl group have been invoked. The singlet carbene adopted the ‘bisected’ conformation which disfavoured the hydride migration. Cyclobutylchlorocarbene photolytically generated from the corresponding dia~irine~~ was intercepted by a variety of alkenes accompanied by ring expansion and hydride shift products. Although carbenes can be classified into two categories, electrophilic and 7u R.A. Moss, Tetrahedron Left., 1967, 4905. ” R. A. Moss and A. Mamantov, J. Am. Chem. Soc., 1970, 92, 6951. 72 R. A. Mitsch, J. Am. Chrm. SOC.,1965, 87, 758. 7x R. A. Moss and R. C. Munjal, J. Chem. SUC.,Chem. Commun., 1978, 775. 74 T. T. Su and E. R. Thornton, J. Am. Chem. Soc., 1978, 100, 1872. 75 R. A. Moss and M. E. Fantina, J. Am. Chcm. Soc., 1978, 100, 6788. iR R. A. Moss. M. E. Fantina, and R. C. Miinjal, Tefrahrrlron Lrvr., 1979, 1277. 139 The Thermolysis and Photolysis of Diazirines nucleophilic, diazirines offer the mechanistically interesting possibility of giving ‘am biphil ic’ carbenes.7797r( 5 Concluding Remarks It is established that diazoalkanes are formed in the thermolysis and photolysis of some diazirines. It is to be expected that further investigation of the thermo- chemistry, for example, will resolve the many conflicting reportsY concerning the value of dHt O for diazirine. Advances are also anticipated from the technique of laser flash photolysis. Use of alternative modes in inducing decomp~sition~~ are also to be envisaged. The low energy (5-20 eV) electron impact studies of 3-chloro-3-methyldiazirineby Kuwata and co-workers,80 where NdC3vu), CN(B2C+),and CH(A2d) are formed, look promising. The reaction of 3-chloro-3-methyldiazirinewith hydrogen atoms, produced by microwave dis- charge, gave HCI and acetonitrile.81 Here the activation of the methyl group is responsible for the reaction course.In summary, recent years have witnessed extensive investigations into the thermolysis and photolysis of diazirines. Experiments have been performed in which intermediates have been trapped and by which mechanistic pathways have been elucidated. New compounds have been synthesized. As a result, diazirines, as a class, have become better understood. Acknowledgements The author wishes to thank Dr. H. Tomioka, Professors C. W. Jefford, W. Ando, K. J. Laidler, and E. Schmitz for their valuabYe suggestions and Ms. J. Tojo and N. Comeau for preparation of the manuscript. Grateful acknowledgement is made for a grant-in-aid of research from the Natural Sciences and Engineer- ing Research Council of Canada and from the University of Prince Edward Island. 77 R. A. Moss and W.-C Shieh, Tetrahedron Lett., 1978, 1935. N. P. Smith and I. D. R. Stevens, Tetrahedron Lett., 1978, 1931. 79 K. T. Chang and H. Shechter, J. Am. Chem. Soc., 1979, 101, 5082. 8o M. T. H. Liu, T. Tanaka, T. Hirotsu, K. Fukui, 1. Fujita, and K. Kuwata, J. Phys. Chem., 1980,84, 3 184. C. D. Burkholder, W. E. Jones, J. S. Wasson, and M. T. H. Liu, J. Am. Chem. SOC.,1980, 102,2841.

ISSN:0306-0012    

DOI:10.1039/CS9821100127

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

CENTENARYLECTURE* Cyclopentanoids:A Challenge for New Methodology By Barry M. Trost MCELVAIN LABORATORIES OF ORGANIC CHEMISTRY, DEPARTMENT OF CHEMISTRY, UNIVERSITY OF WISCONSIN, 1101 UNIVERSITY AVENUE, MADISON, WI 53706, U.S.A. 1 Introduction The chemistry of six-membered rings plays a dominant role in synthetic organic chemistry, in part, because of the widespread occurrence of such a structural feature in many biologically important natural products. Prior to the early sixties, except for a few classical examples such as the iridoids and cedranes and as a singular appendage to six-membered rings as for the steroids, cyclo- pentanoids and especially polycondensed cyclopentanoid natural products were rare.1 Development of synthetic strategy to such systems was overshadowed by the pre-occupation with their six-membered ring counterparts.The identity of prostaglandins as cyclopentane derivatives [e.g. PGE2, (1)2J and the elucidation of the structure of hirsutic acid3 (2) as a terpene possessing three fused five- membered rings initiated a continuing revelation of many varied cyclopentane natural products, some of which are listed in Figure 1. Synthetic solutions to such a marvellous array of structural types will un-doubtedly benefit from a plethora of five-membered ring forming reactions. Classical reactions such as Dieckmann cyclizations, Friedel-Crafts acylations, and aldol condensations clearly apply to these systems as well as other ring sizes. However, in these and other reactions, e.g. intramolecular alkylations, special problems accrue to the formation of the somewhat strained five-membered ring.For example, whereas application of the aldol condensation for the synthesis of (3; equation l), a well-known building block in natural products synthesis known as the Wieland-Miescher ketone, succeeds admirably, the corresponding reaction for formation of (4; equation 2) requires quite special conditions and then only proceeds in 30-4 % yield.4 Intramolecular alkylations such as that represented in equation 3 suffer from a stereoelectronic bias for 0-*The present text is based upon the lecture delivered on the 11 March 1982 at a RSC Perkin Division Meeting at the Scientific Societies’ Lecture Theatre, Savile Row, London W1.T. K. Devon and A. I. Scott, ‘Handbook of Naturally Occurring Compounds,’ Vol. I and 11, Academic Press, New York, 1972. * A. Mitra, ‘The Synthesis of Prostaglandins,’ John Wiley & Sons, New York, 1977. F. W. Comer, F. McCapra, 1. H. Qureshi, and A. I. Scott, Tetrahedron, 1967, 23, 4761; F. W. Comer and J. Trotter, J. Chem. SOC.B, 1969, 11. W. G. Dauben and D. J. Hart, J. Org. Chem., 1977, 42, 3787. Cyclopentanoids: A Challengefor New Methodology L7C c5 OH1l 2 H Hb bH Cedrene Pent alenolactone &OH OH& &/ # 0 HO H02C‘ A I # I H OH C;H (2) Coriolin Capnellane Isocomene Modhephene Iaurene *q--C02H Retigeranic acid Figure 1 Representative cyclopen tanoid natural products 0 0 Trost 0 0 lii KF 18-crown-6 reflux rather than C-alkylation, again in contrast to the six-membered ring case, which leads smoothly via normal C-alkylation.5 2 Cyclization via Acyl Anions Many methods that apply to six-membered rings do translate over to five- membered rings.One of the most general cyclization methods known is the acyloin condensation, which appears to apply to virtually any ring size.6 A related reaction is the redox coupling of an m,w-dialdehyde with a thiazolium salt (equation 4).7 Such a reaction would appear to proceed through the OHC'(CH,)n 80OC ~ 'CHO OH+ "+ -(, X-0 OH equivalent of an acyl anion formed in situ between the thiazolium salt and the aldehyde.8 The ability to add such acyl anion equivalents in conjugate fashion to unsaturated carbonyl systems suggests the possibility of an intramolecular version of the type illustrated in equation 5.Such an approach would lend itself 0 J. E. Baldwin and L. I. Kruse, J. Chem. SOC.,Chem. Commun., 1977, 233; H. 0. House, W. V. Phillips, T. S. B. Sayer, and C. C. Yau, J. Org. Chem., 1978, 43, 700. J. J. Bloomfield, D. C. Owsley, and J. M. Nelke, Org. React., 1976, 23, 259. R. C. Cookson and R. M. Lane, J. Chem. SOC., Chem. Commun., 1976, 804. H. Stetter and H. Kuhlmann, Chem. Ber., 1976, 109, 2890: H. Stetter, W. Basse, and K. Wiemann, ibid., 1978, 111,431. Cyclopentanoids: A Challengefor New Methodology to an attractive solution of polycondensed cyclopentanoid natural products such as hirsutic acid (9.9 As Scheme 1 illustrates, a tetracycle such as (6) Rfi2He\i@co2H0 CHO (7)0 (8) Scheme 1 Retrosynthetic analysis of hirsutic acid possesses the basic carbon skeleton of this terpene.The tricycle (7) represents a logical precursor that, by application of the principle represented in equation 5, simplifies to a bicyclo[3.2.l]octane (8). Simplification to a bridged bicyclic skeleton allows the stereochemical bias of such a system to resolve the problem of controlling the stereochemistry of hirsutic acid. Scheme 2 summarizes this synthetic approach. Gratifyingly, the crucial cyclization of (10) to (1 1) proceeded routinely in 65-70% yields using triethyl- amine in hot isopropanol. The conjugate addition to the ynoate in (9) to produce the first five-membered ring offers an opportunity to explore an asymmetric synthesis.Indeed, use of quinine in this step gave a 65:35 mixture of enantiomers. Thus, intramolecular conjugate additions create five-membered rings with an efficiency comparable to other ring sizes. * B. M. Trost, C. D.Shuey, and F. DiNinno, Jr., J. Am. Chem. Soc., 1979, 101, 1284. Trost CN e d f--@*c--@C0,Me CHO C0,Me co,H +(Y 0 Me0,C (11) ii i / -yJyo&OH I HO2C I 0 Me0,CH H Reagents: (a):i, LDA, TMS--r-CH,I, THF; ii, KOH, MeOH; iii, LDA, THF, TMEDA, CO,; iv, CH,N,, ether; v, H,SO,, H,O, 78%. (6); Et,N, PhCH,, 75%. (c): i, H,, Pd/BaCO,, EtOH-EtOAc; ii, BrZnCH2C02C2H5,PhH, ether; iii, NaOMe, MeOH, 77%.(d): i, BH,.THF; ii, Ac,O, C,H,N, 67%. (e): i, NBS, CCI,; ii, LiBr, Li,CO,, DMF; iii, PCC, CH,CI,, NaOAc, 64 %. (f);3.4-dimethyl-5-p-hydroxy-ethylthiazolium iodide, PPOH, 67 %. (g): i, K,CO,, MeOH ; ii, NaBH,, MeOH, THF; iii, Ph,P, MeO,CN=NCO,Me,92%. (h):i, HCI,MeOH, ether; ii, O,, MeOH then Me,S; iii, MeSH, BF,.ether; iv, W-5 Raney Ni, EtOH, 75%. (i): i, KOH, THF, H,O; ii, MeMgBr, ether; iii, PCC, CH,CI,, NaOAc; iv, KOBut, THF, 46”/(j); performed by the procedure of Matsumoto er ol. ref. 63 Scheme 2 A synthesis of hirsutic acid 145 Cyclopentanoids: A Challenge for New Methodology 3 Cyclization via Olefination Reactions A cyclopentenone annulation viu a 1,4-diketone such as (12) requires a 2-oxopropylene equivalent.The general applicability of the Wacker oxidationlo fl CO Me COzMe H Me hirsutic acid -pCOzMe 0 H allows a simple ally1 group to serve this purpose.11 However, this method relies on the applicability of the direct aldol condensation for the final cyclization. The synthesis of a potentially useful building block, the bis-nor-Wieland- Miescher ketone (1 3), cannot be approached by this method. Circumvention of this problem invokes the use of an olefination procedure, which required development of a three-carbon synthon that permitted chemo- selective conversion into the desired ylide. An enol ether such as (14) would be ideal since it would permit the chemoselective conversion of the side chain carbonyl group to the desired functionality.Although (14; X = Br) has been lo J. Tsuji, I. Shimizu, and K. Yamamoto, Tetrahedron Lett., 1976, 2975. l1 M. Yamazaki, M. Shibasaki, and S. Ikegami, Chem. Lett., 1981, 1245. Trost OEt OEt used as a.2-oxopropylene equivalent,12 the difficulties associated with its synthesis, its lability, and the fear that 0-alkylation of 1,3-dicarbonyl substrates would dominate led to the synthesis of (14; X = OAc) from ethyl vinyl ether and formaldehyde.13 Palladium(0)-catalysed alkylation of 2-carboethoxycyclo-pentanone led to the smooth generation of the alkylated product (1 5). Sequen-C0,Et COzEt + (14; X= OAc) -pdo QoEt86% COzEt COZEt 6 62% 0 -a;ph3 tial treatment of the enol ether with NBS in moist DMSO, triphenylphosphine in hot benzene, and aqueous potassium carbonate produces the stabilized phosphorus ylide, which smoothly cyclizes in refluxing methylene chloride.I3J4 Application of this identical series of reactions to 2-methylcyclopentan-I ,3- dione led via (1 6) to bis-nor-Wieland-Miescher ketone (1 3 ; m.p.41-43 "C). l2 R. M. Jacobson, R. A. Raths, and J. H. McDonald 111, J. Org. Chem., 1977, 42, 2545. l3 B. M. Trost and D. P. Curran, J. Am. Chem. Sur., 1980, 102, 5699. R. 0. Clark, L. G. Kozar, and C. H. Heathcock, Synth. Commim., 1975, 5, 1; E. Piers, B. Abeysekera, and J. R. Scheffer, Tetrahedron Lett.. 1979, 3279; H. H. Aldenback, Angrw. Chem.,lnr. Ed. Eng., 1979, 18, 940. 147 Cyclopentanoids: A Challengefor New Methodology The use of chiral phosphines produced (13) with enantiomeric excesses of up I0 77 %.15 By employing the 0-methylmandelate ester (17), n.m.r.spectroscopy permitted assignment of the absolute configuration. Using the Mosher model'" depicted in an 'extended Newman' projection in which the circle represents 0 I1 -0-C-in an anti-array, the absolute stereochemistry depicted in (17a) can be assigned to the compound having more upfield shifts for the methyl group and the methylene group to the carbonyl group, compared to (17b). This little H Me0 H"-+ Ph .a'OMe H used method has proved generally valid in assigning and analysing the degree of absolute stereochemistry, which when combined with the ease of resolution of 0-methylmandelate esters on h.p.l.c., make such derivatives the ones of choice to resolve alcohols.4 Cyclization via Alkylation Methods Attempts to cyclize /3-ketoesters such as (I 8) not unexpectedly generated the products of 0-rather than C-alkylati0n.~7 This same observation plagued the cyclization of (19), a type of substrate that can readily be derived by straight- l5 B. M. Trost and D. P. Curran, Tetrahedron Left., 1981, 22,4929. I6 J. A. Dale and H. S. Mosher, J. Am. Chem. SOC.,1973, 95, 512. J. Martel, A. Blade-Font, C. Marie, M. Vivat. E. Toromanoff, and J. Buendia, Bull. Soc. Chim. Fr., 1978, 11, 131. Trost dMe 0 forward manipulation of the aldol condensation product of a ketone (e.g. 6-bromotetralone) and 1-aryl thiocyclopropane- 1-car boxaldehyde.18 A1though an imine derivative of (18) resolved this problem in this case, a more direct approach envisages a [1.3] rearrangement of the 0-alkylated product to the thermodynamically more stable C-alkylated one.Thermal reorganization of 2-alkylidene-5-vinyltetrahydrofuranssuch as (22) leads via a [3.3] pathway to produce cyclohepten~nes.~~ On the other hand, palladium(0) catalysts permute the reaction profile of these allylvinyl ethers so that they now follow a [1.3] rearrangement pathway to 3-vinylcyclopentanones. Indeed, exposure of (20) to B. M. Trost and L. N. Jungheim, J. Am. Chem. SOC.,1980,102, 7910. l9 B. M. Trost and T. A. Runge, J. Am. Chem. SOC.,1981, 103, 7550. Cyclopetitanoids: A Challenge for New Methodology such a catalyst isomerizes it to the cyclopentanone (21), a steroid precursor.Since palladium(0) catalysts can initiate the 0-alkylation and then subsequently rearrange that product, a one-pot cyclization of an acyclic precursor to a cyclopentanone can be accomplished as shown in equation 6. (dppe)2PdA& DMSO, 130 OC ’ 8 52% A general cyclopentanone synthesis hinges on the availability of 2-alkylidene- S-vinyltetrahydrofurans.20 One such route highlights the utility of the 1-arylthiocyclopropane-1-carboxaldehyde as a useful conjunctive reagent for synthesis. A totally different approach via the aldehyde (23) as an alternative conjunctive reagent takes cognizance of the potential of an intramolecular Michael addition of an alcohol onto an ynoate, i.e.(24) -(25). Simple base catalysed addition was unsatisfactory. On the other hand, sodium benzenesulphi- nate proved to be an exceptionally efficient nucleophilic trigger that permitted obtention of the requisite substrates in excellent yields. Palladium-initiated OYOl ao B.M. Trost and T. A. Runge, J. Am. Chem. SOC.,1981, 103, 7559. Trost isomerization then completes the cyclopentanone synthesis, which in this case generates a prostaglandin intermediate, (26). A chiral cyclopentanone synthesis emerges from the utilization of lactones that derive from carbohydrates as precursors to 2-alkylidene-5-vinyltetra-hydrofurans. Olefination of (27), derivable from D-mannose, proceeded via two J COZBut /OLi ‘OBu‘ 2.MeS0,CI DBU 91% (dppe),Pd LDMSO 0 pathways; (a)condensation of the lactone with an ynamine21 in the presence of a Lewis acid to give directly a vinylogous urethane (28), and (6)addition of an ester enolate followed by dehydration to give a vinylogous carbonate (29). Subjection of (28) to a soluble palladium(0) catalyst in dioxane leads to the cyclopentanone (30) with faithful translation of the chirality of (28). A similar rearrangement of (29) to cyclopentanone (3 1) occurred with a polymerically bound palladium(0) catalyst; whereas, the soluble palladium(0) catalyst in DMSO gave the product of [3.3] rearrangement, (32). The ability to orientate the reaction pathway between the [I .3] and [3.3] rearrangements by variation of ligand and solvent should prove generally useful.2‘ J. Ficini, J. P. Genet, and J. C. Depezay, Bit//. Soc. Chirn. Fr., 1973, 3367. Cyclopentanoids: A Chullengefor New Methodology 5 Cyclopentanone Annulation via a Vinylcyclopropane Rearrangement In addition to general cyclization methods, there exist potential routes that are unique for five-membered rings. One of the least appreciated until recently derived from the vinylcyclopropane to cyclopentene rearrangement.22 The utility of such a method hinges, to a large extent, on the accessibility of the requisite vinylcyclopropane. Diphenylsulphoniumcyclopropylide (33) offers H H a simple approa~h.~39~4 Its adduct with carbonyl partners, an oxaspiropentane, is a highly reactive epoxide, which easily suffers base initiated elimination to form a vinylcyclopropanol.Thermal reorganization of the corresponding tri-methylsilyl ether produces a cyclopentene bearing a trimethylsiloxy-group on the double bond, i.e. an enol silyl ether of the corresponding cy~lopentanone.~~~~~ Equation 7 illustrates the overall transformation that permits not only a regio- controlled annulation of the cyclopentanone ring, but also allows for further elaboration of the original carbonyl carbon as a result of the cyclopentanone being initially generated in the form of an enolate equivalent. A similar annulation emanates from use of the related reagent, l-lithio-cyclopropylphenyl sulphide (equation 8).27 Dehydration of its carbonyl adduct a2 M. R. Willcott, R.L. Cargill, and A. B. Sears, Prog. Phys. Org. Chem., 1972,9, 25; J. J. Gajewski, Mech. Mol. Migration, 1971, 4, 7. 23 B. M. Trost and M. J. Bogdanowicz, J. Am. Chem. SOC., 1973, 95, 289, 531 1. 24 P. H. Scudder, Ph.D. Thesis, University of Wisconsin, 1977. 25 B. M. Trost and P. H. Scudder, J. Org. Chem., 1981,46, 506. 26 Also see J. P. Barnier, B. Garnier, C. Girard, J. M. Denis, J. Salaun, and J. M. Conia, Tetrahedron Left., 1973, 1747; J. M. Conia and C. Girard, ibid., 1973, 1767; C. Girard, P. Arnice, J. P. Barnier, and J. M. Conia, ibid., 1974, 3329. 27 €3. M. Trost and D. E. Kee1ey.J. Am. Chrm. Soc., 1976, 98, 248. 152 Trost H followed by pyrolysis creates a specific enol thioether of the cyclopentanone as a single stereoisomer.The chemo- and regio-selectivity of this strategy permitted transformation of the ketoester (34) to the prostaglandin intermediate (35).28 Scheme 3 outlines a retrosynthetic analysis of the novel antitumour compound aphidicolin (36)29 TMSOWCozH based upon this methodology.30 From the recognition that standard methodology exists for converting (37) into aphidicolin emerges the definition of this synthesis as a problem in regiocontrolled cyclopentanone elaboration. Thus, by applying the structural interconversion represented in equation 7 to structure (37), this molecule simplifies to (38), which in turn should derive from the common octahydronaphthalenedione(39). Realization of this plan is summarized in Scheme 4.In the key transformation of ketone (40) to trimethylsiloxycyclopentene(43), the opening of the oxaspiro- 28 B.M. Trost and S. Kurozumi, Tetrahedron Lett., 1974, 1929. 2@ W. Dalziel, B. Hesp, K. M. Stevenson, and J. A. Jarvis, J. Chem. SOC.,Perkin Trans. I, 1973,2841. 30 B. M. Trost, Y. Nishimura. and K. Yamamoto, J. Am. Chem. Soc., 1979, 191, 1328. 153 Cyclopentanoids: A Challenge for New Methodology HO HO iA Scheme 3 Retrosynthetic analysis of aphidicolin 154 Trost (44) Reagents: (a) i, Li, NH,, ButOH, THF, TMS-Cl; ii, MeLi, ether, CH,O; iii, LiBui,ButAIH, hexane-ether; iv, HCI, H,O, THF; v, MeCOMe, TsOH, 62%. (b); (33), DMSO. (c): i, PhSeNa, DME; ii, MeC(OTMS)=NTMS, 56% overall for b and c. (d):i, FVP; ii, Pd(OAc),, MeCN; iii, Li, NH,, THF, ButOH then TMS-CI, 58%.(e); BuLi, THF, HMPA, CH8=CHCH,I, 35 %. (f): i, Me,CHCMe,BH,, diglyme, then NaOH, H202; ii, PCC, NaOAc, CH,Cl,; iii, KOH, MeOH, 31 %. (g): i, DHP. TsOH, CHCl,; ii, KOH, HO(CH,CH,O),H; iii, TsOH, MeCOMe; iv, PCC, NaOAc, CH,Cl,, 62 %. (h); See Dalziel et al. ref. 29 Scheme 4 A synthesis of aphidicoiin pentane required a deviation from the normal base-catalysed process. For simple epoxides, such a reaction involves a cis syn elimination preferring removal of an axial proton (see equation 9).31In (41) [depicted in (41a)], the only hydrogen cis to the epoxide oxygen is an equatorial one; indeed, the base catalysed ring opening fails! A simple resolution of this problem derived from 31 C. L. Kissel and B.Rickborn, J. Org. Chem., 1972, 37, 2060; R. P. Thummel and B. Rickborn, ibid., 1972, 37, 3919, 4250: 1971, 36, 1365. 155 Cyclopentanoik: A Challenge for New Methodology Ph (41a) use of a merged substitution-elimination pathway32 employing phenylselenide anion under aprotic conditions as shown in equation 10. The stereochemistry of the vinylcyclopropane rearrangement of (42) proved equally fascinating. Whereas conformational considerations strongly suggested preferential formation of the desired product (43; p-H at C-8), experimentally the major product of rearrangement proved to be (44;a-Hat C-8), in which ring B was OTMS forced into a boat conformation. Resolution of this stereochemical problem involved obliteration of the stereochemical centre at C-8 by converting the mixture of enol silyl ethers that resulted into the corresponding enone, followed by its reconstitution with the correct p-H stereochemistry as shown in Scheme 4.With the key ring-structure completed, the remaining stages of synthesis followed more traditional lines as Scheme 4 illustrates. The cyclopropyl reagents outlined herein offer an alternative cyclopentenone annulation based upon conversion of the initial adducts into spirocyclo-butan0nes.~3 Such strained ketones undergo particularly facile Baeyer-Villiger oxidation to y-butyrolactones, which upon subjection to acid, transform to 3p P. Beltrame, G. Biale, D. J. Lloyd, A. J. Parker, M. Ruane, and S. Winstein, J. Am. Chern. SOC.,1972,94,2240.ss B. M. Trost and M. J. Bogdanowicz, J. Am. Chem. Soc., 1973,95, 5321. Trost cyclopentenones (equation 1 l).34 This sequence proved to be a critical elaboration in the synthesis of d~decahedranes.~~ I” 6 A Cycloaddition Strategy Although cyclization by single bond formation represents a major entry into any ring system, the merit of a cycloaddition strategy provides strong impetus for development of such approaches. The power of the Diels-Alder reaction, whose pre-eminence as a six-member ring forming method becomes increasingly clearer, induces its exploitation for both six-membered and non-six-membered natural products by adjustment of ring sizes or other clever manipulations of Diels-Alder adducts.36 An analogue of the Diels-Alder reaction, shown in its barest essentials in equation 12, for five-membered ring formation is stated in equation 13.Although 1,3-dipolar cycloadditions for heterocycles are already tEWGOEWG__j ingrained in our dictionary of strategies, such an approach for five-membered carbon rings remains elusive. Trimethylenemethane (TMM), a species well-studied from a physical per- spective,37 offers many synthetic attractions. In addition to providing the requisite ring-system, it provides a functional group of sufficient flexibility that a diverse range of structural types would be approachable. Its absence as a synthetic method stems from the lack of suitable methods for its generation and the low yields of its ‘cycloadditions.’38 34 P.E. Eaton, G. F. Cooper. L. C. Johnson, and R. H. Mueller, J. Org. Chem., 1972, 37, 1947. 35 L. A. Paquette and D. W. Balogh, J. Am. Chem. SOC.,1982, 104, 774. 36 For a recent elegant example, see M. Demuth, P. R. Raghavan, C. Carter, K. Nakano, and K. Schafner, Helv. Chim. Acta, 1980, 63, 2434; W. C. Still and M.-Y. Tsai, J. Am. Chem. Soc., 1980, 102, 3654. 37 P. Dowd, AcL.. Chcm. Res., 1972, 5, 242; J. A. Berson, ibid., 1978, 11, 446. rs For some elegant recent examples of substituted systems see R. D. Little, G. W. Muller, M. G. Venegas, G. L. Carroll, A. Bukhari, L. Patton, and K. Stone, Tetrahedron, 1981,37, 4371; R. D. Little and G. W. Muller, J. Am. Chvm. Soc.., 1981. 103, 2744; R. D. Little and G. L. Carrol, Tetrcrhdwn Lpft,. 1981. 22.4389. I57 Cyclopentanoidy: A ChulliwAv jijr Nc w Mcthocloiogy I x---,ar' In circumventing the perceived problems of TMM, a mild method for its generation from readily available precursors as well as a more selective reactivity towards cycloaddition are required. The propensity of silicon to transfer to a silylophile when bound to an electronegative carbon raised the possibility of the desilylation of an intermediate such as (45). The propensity of any silylophile to effect a desilylation [path a in (45)] compared to simple charge neutralization [path b in (45)]suggested decreasing the probability of the latter pathway by delocalizing the positive charge or, even better, making such a pathway totally reversible. Palladium(0) initiated ionization of (46) offers both!39 However, TMM itself would not be formed, but rather its palladium complex (47).4*-42 With the hope that such a species would participate in a cycloaddition, (46) was exposed to a catalytic quantity of a palladium(0) complex in the presence of an ambident trap (48);one possessing a strained electron-rich double bond 39 B.M. Trost, Tetrahedron, 1977, 33, 2615; Pure Appl. Chem., 1979,51, 787. For reactions of olefins and alkylidenecyclopropanes catalysed by nickel see R. Noyori,Y. Kumagai, I. Umeda, and H. Takaya, J. Am. Chem. SOC.,1979, 94, 4018; P. Binger, Synthesis, 1973, 427. 41 For reaction of olefins and alkylidenecyclopropanes catalysed by palladium see P. Binger and U. Schuchardt, Angew. Chem., Int.Ed. Eng., 1977, 16, 249; Chem. Ber., 1980, 113, 3334; 1981, 114, 3313; P. Binger and A. Germer, ibid., 1981, 114, 3325. 42 For reaction of trimethylenemethaneiron tricarbonyl see A. C. Day and J. T. Powell, Chem. Commrm., 1968, 1241 ;K. Ehrlich and G. F. Emerson, J. Am. Chem. Sor.. 1972,94, 2464. 158 Trost and an electron-deficient one.43 A single product involving only addition to the electron-deficient double bond, i.r.adduct (49), emerged. Equation 15summarizes the general features of the reaction.44145 The dipolarophile requires at least one electron-withdrawing group (EWG). The reaction involves a two-step process; conjugate addition of the negative end of the dipole onto the dipolarophile followed by charge neutralization.The degree of stereoselectivity depends upon the rate of the ring closure relative to the rate of bond rotation; in the case of trans traps a high selectivity is observed. Among the successful EWG's stand a ketone, an ester, a nitrile, and a sulphone. The question of the substitution on the dipole determines the full scope of this process. As simple a molecular modification as incorporation of a methyl group is not trivial. As equation 16 emphasizes, several potentially unfavourable wEWC(16) TMsL YEWG& competitions can doom the desired path. For example, (50) can deprotonate to form a stable molecule (path a) or desilylate to form a high energy intermediate (51; path b). The latter also can experience a unimolecular hydrogen shift to form isoprene (path c) compared to a bimolecular trapping to form the alkylidenecyclopentanes (path d).Finally, any cycloaddition can produce a total of six isomers; two stereoisomers each of three regioisomers. In order to probe such systems, general methods of synthesis need to be developed.Scheme 5 summarizes the three general approaches and introduces 43 D. M. T. Chan, Ph.D. Thesis, University of Wisconsin, 1982. 44 B. M. Trost and D. M. T. Chan, J. Am. Chcm. Soc., 1979, 101, 6429, 6432. 45 B. M. Trost and D. M. T. Chan, J. Am. Chrm. Soc.. 1980. 102, 6359. 159 Cyclapentanaiak A Challenge far New Methodalogy OH 6 c-fTMS -c'OAc RCHO Scheme 5 Synthetic approaches to silyl alcohols two new bifunctional conjunctive reagents, aldehyde (52)43 and bromide (53).46 Path a involves the direct metallation with n-butyl-lithium followed by silylation with trimethylchlorosilane. Path b envisions (52), which has its origins from methallyl alcohol, as .an acceptor reagent requiring a donor reactant.Path c reverses the electronic characteristics of the two reacting partners, i.e. the silyl component, which has its origins from 2,3-dibromopropene, is the donor re- act ant . TM S OAc TM S (55)TOAc (54) 46 B. M. Trost and D. M. T. Chan, J. Am. Chrm. SOC.,1982, 104, in press. 160 Trost The methyl substituted precursor is prepared by path a of Scheme 5.47 Tn the event, it participated with a facility equal to the parent system to give a single regioisomer (>20: l), albeit as a mixture of stereoisomers (55) with cyclopentenone.Most strikingly is the location of the methyl substituent in the cycloadduct. To verify that this regiochemistry is independent of the structure of the silyl acetate, an alternative regioisomeric starting material was employed with the same result. In consonance with earlier results, these observations suggest that the intermediate is a rapidly equilibrating pair of TMM-Pd complexes, (51) and (54),and that product formation derives from (54).Fenske-Hall calculations indicate that (54)is more stable than (51).4* Thus, in contrast to normal intuition, the electron releasing methyl substituent prefers to be on the most electron-rich carbon atom of the TMM-Pd intermediate! Such a reordering of normal organic preferences clearly derives from the effect of the transition metal on chemical reactivity.n-Conjugating substituents such as phenyl (equation 17)43949 and carbonyl (equation 18)50 show a similar preference. The case of a vinyl substituent is most curious because of its possible direct participation. Nevertheless, it proceeds without complications (equation I9).49 Ph TMSbOk+ a.5 Ph *' B. M. Trost and D. M. T. Chan, J. Am. Chem. Soc., 1981, 103, 5972. 48 D. J. Gordon, R. F. Fenske, T. PI.Nanninga, and B. M. Trost, J. Am. Chem. Soc., 1981, 103, 5974. 49 T. N. Nanninga, unpublished results in these laboratories. 6o T. Satoh, unpublished results in these laboratories. Cyclopentanoids: A Challenge for New Methodology It is appropriate to contrast these observations with the palladium catalysed condensations of alkylidenecyclopropanes with ole fin^.^^ Although it is tempting to interpret the two reactions as proceeding through the same intermediates, many differences are noted; one of the most striking being a different regioisomeric result (equation 20).The potential for simplification of synthetic strategy appears exciting. For example, the addiict with dimethyl (E,E)-muconate has a striking similarity to C0,Me 'C02Me the antiturnour macrolide, brefeldin A. Albene, originally assigned structure (56), for which a Diels-Alder strategy would be had its structure revised to (57),52p53for which such a strategy was clearly less satisfactory. On the other 51 K.Vokac, Z. Samek, V. Herout, and F. Sorm, Tetrahedron Lett., 1972,1665; P. T. Lansbury and R. M. Boden, ibid., 1973, 5017. 54 W. Kreiser and L. Janitschke, Chem. Ber., 1979, 112, 408; W. Kreiser, L. Janitsche, W. Voss, L. Ernst, and W. S. Sheldrick, ibid., 1979, 112, 397. 53 J. E. Baldwin and T. C. Barden, J. Org. Chem., 1981, 46, 244. 162 Trost hand, the normal bias for norbornenes to undergo ex0 addition would make this cycloaddition approach much more likely. Scheme 6 summarizes the C0,Me 01. t (57) P =P (NMe2), Reagents: (a);(46), (PriO),P, Pd(OAc),, THF. (6): i, LAH, ether; ii, 0,, CH,Cl,, MeOH. (c); KN(TMS),, DME, HMPA,’[Me,N],P(O)CI. (d);Li, EtNH,, But OH Scheme.6 A synfhesis of( +) albene realization of this approach, which permits a five-step synthesis in an overall yield of 21 x.54 The substituted analogues also have potential applications. For example, the methyl substituted adduct (55) undergoes chemoselective addition of methyl-lithium, ozonolysis of the exocyclic methylene group, and base equilibration of the secondary methyl group to produce (58),4? a known precursor55 of chryso-melidial (59), a constituent of the defensive secretion of the chrysomelide beetle.(55) -54 P. Renaut, unpublished work in these laboratories. 65 K. Kon and S. Isoe. Tctmhedrron Lett.. 1980, 3399. 163 Cyclopentanoids: A Challenge for New Methodology A loganin synthesis from (55) can also be envisaged.56 Ricyclic systems (60; n = 1 or 2) result from intramolecular versions of this process.4fi TMS, 7 An Electrophilic TMM Equivalent The palladium-catalysed reactions of the silyl alcohols represent a type of behaviour initiated by a nucleophilic TMM partner 1i.e. (61 )I.Equally versatile would be a reverse reactivity profile as represented in (62). In fact, such a reactivity is even simpler to envisage since it only requires a selection of X in (63) such that it is sufficiently reactive to be displaced by a nucleophile, but sufficiently unreactive such that (63) does not undergo self-annihilation. This balance is achieved with (63; X = I) in that it reacts quite smoothly with anions such as that derived from 2-phenylsulphonylcyclopentanoneto PhSO 2 81% TMs%PhSO 2 (64) SO, Ph6ZIO](66)(67) SO2Ph (65) 56 K.Kon and S.Isoe. Trnnon Yrrki Kagohutsu Toronkai Kom Yoshishrr 23rd, 1980, 49. Trost which in turn can be induced to cyclize to the methylenecyclopentane (65) by unmasking the anion portion of the system with fluoride ion? Strategic place- ment of the anion stabilizing phenylsulphonyl group and the electron donating hydroxy-group weakens bond 'a' such that treatment with potassium hydride initiates cleavage of this bond to give, presumably via (66), cyclo-octadienone (67). The overall sequence constitutes a three-carbon intercalation between a carbonyl carbonand an cx-carbon atom as represented in equation 21. Applied to 0 EWG cyclododecanone, this sequence provides an extraordinarily efficient approach (62% overall yield) to muscone (Scheme 7).In the fluoride activation of the Muscone d SO2 Pho& -& (69) Reagents: (a):i, Br,, CHCI,; ii, PhSO,Na, (CBHI3),NBr, DMF, 85%. (b): NaH, DME, (63; X = I), 83%. (c); Bu,NF, THF, 92%. (d): i, H,, 5% Pd/BaSO,. EtOH; ii, 6% Na/Hg. Na,HPO,, MeOH, 95% Scheme 7 A synthesis of muscone ''B. M. Trost and J. E. Vincent, J. Am. Chem. Soc., 1980,102, 5680. Cyclopentanoids: A Challengefar New Methodolagy allylsilane (68), the intermediate methylenecyclopentane is not observed, the direct product being the ring enlarged (69). Such an approach provides a potential solution to the taxane family as rep- resented by taxinine (70) and taxol (71).5* The bicyclo[5.3.1 Iundecane skeleton OAcI# o*o AcO -4O -Ph Ph (70) (71) bcaring a gem dimethyl group embedded in the centre of the ring system can be atialysed vicr this type of intercalation procedure as shown in equation 22.0 (-J&-(yTMS+(22) X Y X X 2-Methyl-3-hydroxycyclohexene silylates selectively at the methyl group which then, by a series of [3.3] and [2.3] sigmatropic rearrangements, leads directly to the cyclization precursor (72; Scheme 8).59 Whereas, fluoride induced cyclization failed, Lewis acid initiated cyclization produced the methylene- cyclopentane analogue (73). Its fragmentation proved most instructive (equation 23). Treatment of (73) with potassium hydride in DME followed by protonation led via only endo-exo interconversion into (74).On the other hand, a mixture of (74) and (75), which ranged from 1 :1 to 1 :9, resulted by addition of 18-crown-6 and [2.2.2]cryptand respectively. As implied by equation 23, these observations are nicely accommodated by the dependency of the equilibria among the salts on the structure of the ion pairs, which permits an extraordinary level of control. On the other hand, equilibrating the neutral hydrocarbons by simply using a catalytic amount of potassium t-butoxide in DMSO converted (73) exclusively into the ring enlarged (75). As shown in Scheme 8, application of this latter procedure to (76), which derives from (73), smoothly generates the bridged bicyclic portion of the taxanes. 58 R. W. Miller, J. Nut. Prod., 1980, 43, 425.59 B. M. Trost and H. Hiemstra, J. Am. Chem. Soc., 1982, 104, 886. 1 66 Trost TMS WTMS-b OH SH 3. e d c-gH($ SO; Me -S0,Me S02Me (77) (76) (73) Reagents: (a):i, BuLi, ether, TMS-Cl; ii, NaH, CS, then Mel; iii, A; ivyLAH, ether, 38 %. (b): i, 2-chlorocyclopentanone, NaH, DMF; ii, KH, DME, A then MeI; iii, MCPBA, CH,Cl,, NaHCO,, 41 %. (c); EtAICl,, PhMe, 73 %. (d):i, Et,Zn, CH,12, PhH, air; ii, H,, Pt02, NaOAc, HOAc, 81 %. (e);cat ButOK, DMSO, 95% Scheme 8 A synthesis of a taxane model (73) ,---SO, Me 2gSOzMe z%-MeSO2K+ ? 1 (23) SO, Me MeSO, (75) (74) Although the application of this electrophilic synthon of TMM for three- carbon intercalation is clearly promising, it also permits evolution of efficient strategy towards polycylic cyclopentanoid natural products.For example, (78), 167 Cyclopentanoids: A Challenge for New Methodology hirsutic acid (24) t which may derive from a synthon for an oxatrimethylenemethane as well as the electrophilic version of trimethylenemethane alluded to herein condensing with 2-methylcyclopentan-1,3-dione,can be envisaged to be a common precursor to both hirsutic acid and coriolin. As detailed earlier, Zethoxyallyl acetate serves nicely as a synthon for oxatrimethylenemethane in providing ready access to bis-nor-Wieland-Miescher ketone (13).13J5 The next annulation required the introduction of a substituent that would serve as a stereochemical anchor, i.e. a substituent that would fix the stereochemistry of a third five- membered ring and be easily dismissed once its mission was accomplished.An alkylthio-group nicely served this purpose as detailed in Scheme 9 since (79; X = H) showed a propensity to isomerize.60 Introduction ,of the methylene- cyclopentane ring was best achieved by alkylating the b-ketosulphide with (63; X = I) but by cyclizing the p-ketosulphone. Tricycle (80) provides a con-venient divergence point to either hirsutic acid or coriolin, the latter being the most complicated member of the hirsutane family.61 Choosing coriolin as the target, the exocyclic methylene group simply becomes a gem dimethyl group via cyclopropanation and hydrogenolysis. With the remaining structural modi- fications being mainly adjustment of oxidation level as outlined in Scheme 9, dienone (81), the penultimate intermediate in all the previous syntheses of coriolin,62 and thus coriolin becomes available using more standard method- ology.8 Conclusions While much attention focuses on medium and large rings, the plethora of biologically significant cyclopentanoids reawakens concerns regarding their synthesis. The fact that viewing synthetic approaches to them as simple extra- polations from methods available to make six-membered rings can be painfully deceptive and that the most powerful method for making six-membered rings does not apply directly to five-membered rings opens the challenge for new methodology. The methods detailed contribute to meeting this challenge. 'O B.M. Trost and D. P. Curran, J. Am. Chem. SOC.,1981, 103, 7380. 'l H. Nakamura, T. Takita, H. Umezawa, K. Mamuru, N. Yuga, and Y. Itaka, J. Anribiol., 1974, 27, 301, 6a K. Tatsuta, K. Akimoto, and M. Kinoshita, J. Antibiot., 1980, 23, 100; S. Danishefsky,R. Zamboni, M. Kahn, and S. J. Etheridge, J. Am. Chem. Sac., 1981, 103, 3460; M. Shibasaki, K. Iseki, and S. Ikegami, Tetrahedron Lett., 1980, 3587. 63 H. Hashimoto, K. Tsuzuki, F. Sakan, H. Shirahama, and T. Matsumoto, Tetrahedron Letr., 1974, 3745. 168 Trost JI HO H? I, ..r;ctrj-.-vog .# -h --,’yypo I’ I I OH OH (81)li,,* OH Reagents: (a): i, Et,N, MeSH, MeOH; ii, HOCH,CH,OH, camphorsulphonic acid, PhH, 92%. (b): i, KH, MeSSMe, DME; ii, KH, DME, (63; X = I), 57%.(c): i, MCPBA, CH2CI,, NaHCO,; ii, Bu,NF, THF, 55 %. (d): i, Et,Zn, CHJ,, PhH, air; ii, H,, PtO,, HOAc, NaOAc; iii, SOCI,, C,H,N; iv, MCPBA, CH,Cl,, 62”/6,(el: i, 10% HCIO,, acetone; ii, DBU, CH,CI,, 91 %. (f): i, Na naphthalenide, DME then DBU, CH,Cl,; ii, Li, NH,. (g): i, MCPBA, CH,CI,; ii, DBU, CH,CI,, 43% overall for f and g. (h): i, CF,C(OTMS)=NTMS, DMF; ii, LDA, THF- HMPA, TMS-Cl; iii, Me,NCH,l, CHCI,; iv, Mel, ether; v, DBU, CH,CI,, 46”/,. (i); As per Danishefsky et al. ref. 62 Scheme 9 A synthesis of coviolin I69 Cyclopentanaids: A Challenge .for New Methodology Acknowledgment. I am most grateful for a highly talented group of collaborators who have made this programme possible. They are individually acknowledged in the references.The National Institutes of Health and the National Science Foundation have provided the financial resources for the programme for which we are deeply indebted.

ISSN:0306-0012    

DOI:10.1039/CS9821100141

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

NYHOLM LECTURE* Solving Chemical Problems By M. J. Frazer SCHOOL OF. CHEMICAL SCIENCES, UNIVERSITY OF EAST ANGLIA, NORWICH, NR4 7TJ 1 Introduction Chemists spend much of their time solving, or attempting to solve, problems! This review is about how they do it, why they succeed and why they fail; and what lessons we can learn so that we can help all those studying chemistry or working as chemists to improve their ability to solve chemical problems. In the next section, the nature and extent of chemical problem solving is explained. Then in sections 3 and 4 the research methods for investigating problem solving and the main findings as they apply to chemical problems are described. Experience with teaching chemical problem solving is the subject of section 5.There is a vast literature on problem solving in general and it would not be possible to include references to it all in this review. Therefore the paper is comprehensive about chemical problem solving, but only key references are given to research into general problem solving. Many authors,l-4 most recently Guy5 writing about university chemistry courses, have provided opinion and sometimes research evidence that students, teachers, and employers are dissatisfied with the ability of chemistry students at all levels to identify? and to solve problems. What do they mean by solving problems in chemistry ? 2 What is Chemical Problem Solving? No problems exist in isolation-a problem is perceived by an individual. This is illustrated by Figure 1 and by the illustrative examples (Problems 1-7 which appear in this text).$ These statements or questions have been problems to at least one individual at some time.*First delivered at a RSC Education Division Meeting on 23 February 1982, at the Scientific Societies' Lecture Theatre, Savile Row, London W 1. tln fact this review is as much about identifying and recognizing problems as it is about solving problems. $The Nyholm lecture was illustrated with 20 examples provided in a separate booklet. A. H. Johnstone, F. Percival, and N. Reid, Stud. Higher Educ., 1981, 6, 77. H. L. Youmans,J. Chem. Educ., 1971,48, 387. G. L. Gilbert, J. Chem. Educ., 1980, 57, 79. M. J. Frazer, C. R. Palmer, and R. J. Sleet. Educ.. Chcm., 1976, 13, 44.J. G. Guy, ChtJm.Br.. 1982, 18, 44. 4" Answer Decision Task or goal ‘Chemical completed com fort’ so1 Problem solving is overcoming obstacles or barriers or bridging this gap by using IN FORMATION and REASONING PROBLEM Question Indecision Task or desired ‘Chemical goal discomfort’ Figure 1 The transition from problem to solution requires (a) information and (b) reasoning There are many times when a pupil at school, a student in higher education, or a professional chemist is faced: (a) with a question to which he does not immediately know the answer, and/or (b)with indecision and cannot immediately make a decision, and/or (c) with a task or a goal that he cannot accomplish or reach immediately and/or (d) with a feeling of ‘chemical discomfort’ (e.g.Problem 1) and cannot immediately feel ‘chemically comfortable’. Problem I (a) A red solid of composition PBr7 exists (b) There is an oxide of carbon ClzOs All these situations can be summarized by the statement that the individual has a problem for which he cannot immediately find a solution. There is an obstacle or barrier in the path from problem to solution. In general, problem solving is bridging this gap, or is overcoming the obstacle or barrier. To a greater or lesser extent, every problem requires the individual (a)to possess information and (6) to reason with this information in order to progress from the state of having a problem to the state of having a solution. Chemical problem solving is the process of using chemical knowledge and chemical skills to bridge the gap between problem and solution.It is important for the problem solver to recognize that the required knowledge and skills do not depend on him alone. Memory is one source of information for bridging the gap, but other sources are: (a) the problem statement itself; (b)experts in the problem area (e.g.consider the best way of solving Problem 2); (c) the literature; and (d) observation and experiment (Problem 3 was solved by laboratory simulation). Frazer Problem 2 How can I find the composition ofa 1981 50 petice piece? Problem 3 What was the cause of explosions on two siyxrrcire occasions when crude oil from Qatar was in the early stages of‘ king discharged at a port in Thailand? In order to emphasize the nature and extent of chemical problems it is worth attempting a classification (Table 1).This classification was developed at a recent seminar on chemical problem solving.6 Many problems, indeed most problems outside the classroom, do not have unique, unambiguously correct solutions- these can be called ‘open problems’. On the other hand, problems that have Table 1 A classification of chemical problems-canre.\-r tirid ‘c-loscri’ or ‘open’ Artificial Problems (i) The solution is known at least to the person (teacher, textbook author) who has presented the problem. Such problems are used in teaching for two purposes (a) helping students to learn by applying their knowledge, and (b) preparing students to solve real problems.(ii) Artificial problems may be further classified according to the nature of the solution (closed or open). Closed problems: There is a single unique solution (e.g. numerical problems, identification of a compound either by experiment or from given data). Open problems: There are a number of possible solutions (e.g. alternative synthetic routes, a1 ternative experimental designs for a practical exercise, alternative courses of action in a simulation exercise concerning chemistry and its impact on society). Real Problems (i) The solution is not known to anyone. There may not even be a solution; or on the other hand, there may be several reasonable courses of action and the problem then becomes one of selecting the best solution.(ii) Real problems may be further classified according to the context (mission directed or not mission directed). Mission direc ted: The solution is of consequence to industry, some other enterprise, or to society. Not mission directed: The solution is of no immediate consequence to anyone except to the problem solver. IJ M. J. Frazer, Report of a Seminar on Chemical Problem Solving, University of East Anglia, 1981. (Available from the author). I73 Solving Chemical Problems a unique answer may be described as ‘closed problems’. We offer our students plenty of experience with closed problems but hardly prepare them at all for dealing with open problems.It should now be clear that chemical problem solving involves one or more of the skills of analysis, selection, pattern recognition, experimental design, synthesis, process design, etc. Faced with such a wide range of types of chemical problem, some may ask is there any hope of being able to teach chemical problem solving? Would it not be better to concentrate on imparting chemical knowledge and letting the individual learn to cope with problems of particular types when first he meets them? It is the purpose of this review to show that it is both desirable and possible to help students become better at problem solving. But in order to achieve this our teaching needs to be based on the results of research on problem solving.3 Research Methods for Investigating Problem Solving A. Overview.-In order to appreciate the results of research on problem solving, it is worth reviewing briefly the methods that have been used. The motives for this kind of research are varied. For some, the studies are fundamental and are firmly placed in cognitive or developmental psychology and are moves in the attempt to answer the questions: ‘How do we think?’ ‘How does the brain work?‘ Others are much more pragmatic and are simply trying to discover the strategies that lead to success and the reasons for failure in problem solving with the ultimate aim of developing better teaching methods. Although it is not possible to classify each study unambiguously, roughly we can label research with these two motives as ‘descriptive’ and ‘prescriptive’ respectively. Basically, in the descriptive approach the researcher records the behaviour of individuals in defined problem situations with the intention of describing as far as possible the mental processes that are occurring during problem solving.Often the intention is to fit the description into one of the theories or models of cognition. For example Simon and Newel17 in a major, and now classical, study fit their observation and recordings of individuals solving problems into an ‘information processing theory’ in which the human problem solver is likened to a computer. Both are examples of information processing systems and are characterized by having an input and output for symbol structures, a processor including a short term memory, and a long term memory capable of storing and retaining symbol structures.The prescriptive approach is aimed at generating advice to pass on to others. Such advice if followed is likely to lead to success in solving problems. Such advice may be general, may be related to a particular subject, or may refer to a particular type of problem (e.g. balancing a redox reaction, interpreting an n.m.1. spectrum) within a subject. In order to concentrate on the ‘reasoning’ component and to reduce the ‘information’ component of bridging the gap from problem to solution, most ’I A. Newel1 and H. A. Simon, ‘Human Problem Solving’, Prentice Hall, New Jersey, 1972. Frazer studies have been in a context that is subject free.Researchers have confronted individuals with the Towers of Hanoi, with computer variations of the popular ‘Dungeons and Dragons’ game, with numerous combinations of missionaries, cannibals, domestic animals, and wild beasts crossing and recrossing rivers, and with many other ingenious situations in which previously learnt knowledge is of little use in reaching a solution. Problem solving research in a subject context has been mainly in mathematicss-10 and engineering.11 In fact very little has been published on problem solving in chemistry. Part of the discussion at a recent conference on problem solving researchI2 centred on the question of the relative merits of subject-free and subject-based studies. With our present state of knowledge in this field all types of research (pre- scriptive and descriptive, as well as subject-free and subject-based) are needed and will contribute to understanding.The choice is very much a question of the personal interests and priorities of the researcher. Because of the urgent need to improve the teaching of chemical problem solving and because, ultimately, real people have to solve real problems, the reviewer gives priority to research that can be described broadly as ‘chemical and prescriptive’ and so this review will be biased in this way. Research concerned with finding ways to overcome students’ learning diffi- culties and misconceptions in chemistry is often relevant in attempts to improve chemical problem solving abilities. Examples of this type of research were described in the last Nyholm lecture by J0hnst0ne.l~ Four main methods for research into problem solving can be identified.B. Empirical Methods.-Included here are the numerous publications usually written in the form of advice about strategies based on the analysis by an expert of his accumulated experiences of problem solving. There are several books7~14-18 describing general strategies for problem solving. The background of the authors is so varied, and yet the recommended strategies are so similar in their essentials that we can be confident that the approach is correct. A summary is given in section 4B.Purists might not accept these publications as research reports because there are no pre-planned experiments, no testing of hypotheses, no controls, and no statistical data.On the other hand, critical G. Polya, ‘How to Solve It’, Doubleday Anchor Books, New York, 1957. M. P. Cohen and J. E. Bernard. Int. J. Math. Educ. Sci. Technol., 1981, 12, 169. D. J. Goldberg, Int. J. Math. Educ. Sci.Terhnol., I981 ,12, 21 1. D. R. Woods, J. D. Wright, T. W. Hoffman, R. K. Swartman, and I. D. Doig, Ann. Eng. Educ., 1975, 1, 238. l2 Problem Solving and Education: issues in teaching and research, (Proceedings of a conference), ed. D. T. Tuma and F. Reif, Lawrence Erlbaum Associates, New Jersey, 1980. l3 A. H. Johnstone, Chem. Soc. Rev., 1980, 9, 365. l4 K. Raaheim, ‘Problem Solving and Intelligence’, Universitetsforlaget, Bergen, 1978.K. F. Jackson, ‘The art of solving problems’, Heinemann, London, 1975. Is F. H. George, ‘Problem Solving’, Duckworth, London, 1980. R. W. Samson, ‘Problem Solving Improvement’, McGraw Hill, New York, 1970. l8 W. A. Wickelgren, ‘How to solve problems’, Freeman, 1974. Solving Chemical Problems reflection about experiences is an approach to research and, if the different authors are in broad agreement, their views cannot be ignored. C. Studying Individuals Attempting to Solve Problems.-In this type of research, analysis is made of data collected about the behaviour of individual subjects attempting to solve problems. The subjects may be either novices or experts. However, there are obvious dangers in expecting to be able to transfer the skills and strategies used by an expert directly to the novice.Various methods of data collection have been used : (a) Recording on audio-tape, the subject’s comments after he has been invited to ‘think aloud’ during problem solving (subsequent analysis of the transcript of the subject’s comments is called protocol analysis); (6) Observing the subject during problem solving either directly or by television; (c) Obtaining the subject’s written solution and notes; (d) Interviewing the subject immediately after he has attempted the problem; (e) Testing the subject either before or after problem solving to investigate whether he has the requisite chemical knowledge and skills (‘information’ in Figure 1) to solve the problem; (.f) A variation of (e) is to provide the subject with a highly structured problem in order to investigate which part of the problem (e.g.item of chemical knowledge, chemical skill, reasoning step) causes the difficulty ; (8) Providing the subject with an opportunity to ask for more information as he attempts the problem-this information can be provided directly by the observerlresearcher, by card selection, or by selection from a computer memory. Most often a combination of these data collection methods is used. There are various ways of recording and analysing the data. Sometimes a simple record showing the sequence of behaviour or a transcript is felt to be sufficient, but where possible the researcher will try to probe deeper and produce a map or network purporting to show the connections between the words, concepts, and actions used by the subject.At the University of East Anglia we have been attempting to develop the use of problem solving networkslg~20 as a method of both recording and analysing individual subjects’ attempts at chemical problem solving. These methods of data collection and analysis do not lend themselves to dealing with large groups and most research in this area is of the type sometimes described as ‘clinical’. For example, in recent research into solving problems in lB A. D. Ashmore, M. J. Frazer, and R. J. Casey, J. Chem. Educ., 1979, 56, 377 zo M. J. Frazer, J. Morris, M. P. B. A. Pereira, M. E. M. Pestana, A. V. Powell, and T. F. Wallace, Higher degree theses at the University of East Anglia.Details available from the author. Frazer physics21922 the 'think aloud' protocols of only one expert and one novice were compared. Nevertheless some useful insights were gained. D. Evaluation of Courses and Activities that claim to Teach Problem Solving.- There are so few courses concerned principally with problem solving that it is not surprising that there are very few published evaluations. If we are to improve the problem solving abilities of our students we shall need to develop more courses and activities 'with this specific aim. It is to be hoped that such courses will be evaluated. A course in general problem solving techniques for post-graduate students23 and problem solving courses in mathematicsg and engineering] * have been evaluated. E.Systematic Callection of Teachers' Views.-Teachers spend a considerable amount of time marking, watching, and correcting students, attempts at problem solving. Few, if any, attempts have been made to tap this potentially rich source of information about students' strategies, difficulties, and reactions to various approaches. In the next section the results obtained by these various research methods, particularly as they apply to problems in chemistry, are brought together. 4 Results of Research into Chemical Problem Solving A. The Stages of Problem Solving.-Several a~thor~~J~J~~~~-~~have taken an overview of the problem solving process and have described the stages that an individual must pass through in order to progress from problem to solution.In Table 2 a summary of the views of different authors is displayed and, although the words are different, a clear pattern emerges. However Table 2 is not altogether satisfactory because, with the exception of the scheme due to Jackson, emphasis is on the class-room type of problem in which a well defined problem with a unique answer is presented to students. A scheme, adapted from one first presented by Ca~ey,~~ is more suitable for describing the stages for real chemical problems. This scheme, which is still related to Table 2 by the three main phases, is shown in Figure 2. Important features are: (a) recognizing that a problem exists is an important stage that is often overlooked, (b)all the stages are interrelated and the problem solver may return to each stage several times clarifying and refining each time, (c) arriving at the best solution often leads to 81 J.H. Larkin, J. McDermott, D. P. Simon, and H. A. Simon, Science, 1980, 208, 1335. 22 J. H. Larkin and F. Reif, Eur. J. Sri. Edur., 1979, 1, 191. 23 M. F. Rubinstein in 'Problem Solving and Education: issues in teaching and research, ed. D. T. Tuma and F.Reif, Lawrence Erlbaum Associates, New Jersey, 1980, p. 25. R. J. Casey, personal communication, 1980. 25 D. P. Ausubel, 'Educational Psychology-a cognitive view', Holt, Rinehart, and Winston, New York, 1970. Be R. M. Gagne,'The Conditions of Learning', Holt, Rinehart, and Winston, New York, 1970. 27 J.P. Guilford and R. Hoepfner, 'The Analysis of Intelligence,' McGraw Hill, New York, 1971, 104. 2x C. T. C. W. Mettes, A. Pilot, H. J. Roossink, and H. Kramers-Pals, J. Chem, Edur., 1980, 57, 882. 177 Table 2 Words used by various authors ro describe the stages of problem solving Ashmore, Ausubelb Gagntc Guilford and Jacksone Mettes, Pilot, PolyagCasey, and Hoepfnerd Roossnik, and Frazer" Kramers-Palsf PHASE I 1. Defining I. Setting 1. Presenting 1. Preparing 1. Formulating 1. Analysing 1. UnderstandingRECOGNIZING 2. Defining 2. Defining 2. Analysing, 2. Interpreting PHASE 11 2. Collecting 3. Gap filling 3. Formulating 3. Producing 3. Constructing 2. Planning 2. DevisingSOLVING information hypotheses courses of the process a plan3. Reasoning action PHASE Ill 4.Checking 4. Verifying 4. Verifying 4. Verifying 4. Making 3. Executing 3. Executing a CHECKING hypotheses decisions and planAND 5. Reapplying 5. Implementing checking 4. ReviewingIMPLEMENTING and the reviewing soh tion aRef. 19; bref. 25; Cref. 26; dref. 27; eref. 15; fref. 28; gref. 8 Frazer PHASE I Receive and Recognize RECOGNIZING a Problem I I 1 --.--PHASE 11 Factors: ----+ Assemble Tnformation Relevant SOLVING Chemistry to the Problem PHASE III CHECKING AND IMPLEMENTING I I Select best Solution I experience this problem for future I I Figure 2 A madel af the phases in real chemical problem solving action and to the recognition of further problems, and (d) people should learn from the experience of solving a problem.We can never know how may times the failure to recognize a problem has delayed the advance of knowledge or has led to inefficiency in industry. The importance of problem recognition cannot therefore be overemphasized. Furthermore, team work and providing opportunities for contact between individuals with different backgrounds and outlooks is important. It is unlikely that all stages of a complex chemical problem could be completed by one in-dividual. Someone with the experience to recognize a problem may not have Solving Chemical Problems the experience and knowledge to solve it and vice versa. This is one of the strong arguments in favour of co-operation between higher education and in- dustry. In some cases the problem may not be recognized until all the chemistry of the solution has been worked out (e.g.Problem 4-here was a ‘solution’ waiting for a problem.) Problem 4 Can the compound AIP04.HCI.(EtOH)4 which was discovered by chance at the Mond Division of ICI in the early 1970’s be exploited? Its X-ray crystal structure was determined, it is soluble in water, and decomposes at about 70 “C to give the inert Alp04 B. General Strategies of Problem Solving.-Research methods described in 3B and 3C have led to the formulation of a number of general strategies to be adopted when faced with a problem. They are listed in Table 3. They are Table 3 General strategies or advice to problem solvers (1) Work backwards from the goal not forwards from the given information.(2) Break down the problem into sub-goals and work at each separately. Do not try to cope with too much information at any one time. (3) Convert an unfamiliar problem into a familiar problem and then apply an already learnt procedure. (4) Make a guess at the solution and work backwards to see if the guessed solution is consistent with all the information available. (5) Check that all the information stated in the problem has been used and that all other sources of information (memory, literature, experts, experiment) have been exhausted. . (6) Check that all the stages of problem solving (Table 2 and Figure 2) have been used. (7) Check whether there are any guidelines (4C) or algorithms (4D) applicable to this problem.(8) Try to see the problem as a whole. (9) Draw diagrams, verbalize the problem, convert a statement into a question, convert statements into mathematical expressions. (10) ‘Brainstorm’ i.e. write down all the ideas that come to you however foolish or irrelevant they seem. (11) Rest to allow time for ‘incubation’ of the problem. applicable to all problems whatever the subject content but are not all necessarily appropriate for every particular problem. Indeed some are contra- dictory. They are best seen as advice or ideas to try if the problem solver is not 180 Frazer making progress and does not know what to do next in order to proceed from problem to solution, C.Guidelines for Chemical Problem Solving.-These refer to procedures that are more specific than general strategies but more general than algorithms (see 4D). Many teachers and students find general strategies of little help when they are immersed in a chemical problem or at a ‘dead-end’. The dangers of relying on specific algorithms will be outlined in the next section. There is therefore a need to generate guidelines, based on research, for chemical problem solving. Not much has been published yet, but a start has been made with problems in the areas of (a)synthesis of organic compounds,29JO (b) identifying organic compounds from given analytical and spectral data,29 (c) elementary thermodynamic~,3~9~~(d) numerical problems in general chemistry,33 and (e) simple stoicheiometric problems.34135 The guidelines for organic problems developed by the groups at the Universities of East Anglia and Le~ven~~.~~ are shown in Table 4.The approaches developed by the group at Twente University of Te~hnology~~J19~5J~ and by Selvaratnam and Frazer33 for solving numerical problems in general chemistry are similar and are shown in Figure 3 and Table 5 respectively.Five simple steps for solving stoicheiometric problems34 are shown in Table 6. It is surprising, however, how many students are either unaware of this approach or are unable to apply it. D. Algorithms in Chemical Problem Solving.-An algorithm is a set of rules which are to be learnt and which if applied correctly to an appropriate standard problem will lead automatically to a solution of the problem.Most authors would consider that once a problem has been reduced to the stage of only needing the application of an algorithm then there is no longer a problem. The obvious danger of teaching problem solving by using algorithms is that a student is lulled into a false sense of security and is completely unable to cope when meeting a novel situation. Students trained to use V1N1 = V2N2 to solve titration problems have difficulties when faced with titrations using all of a solution made by weighing out a solid into an unknown volume of water. Some algorithms may be useful (e.g. converting % composition figures into as L. Brandt, H. Fierens, R.A. Y.Jones, and P. J. Slootmaekers, Paper given at International Conference on Chemical Education, Dublin, 1979. 30 P. J. Slootmaekers, L. Brandt. H. Fierens, R. A. Y. Jones, and M. J. Frazer, Paper given at 6th International Conference on Chemical Education, Maryland, U.S.A., 198I. 31 C. T. C. W. Mettes, A. Pilot, H. J. Roosi,ik, and H. Kramers-Pals, J. Chem. Educ., 1981, 58, 51. 32 C. T. C. W. Mettes, A. Pilot, and H. J. Roosink, Instruct. Sci., 1981, 10, 333. 33 M. Selvaratnarn and M. J. Frazer, ‘Problem Solving in Chemistry’, Heinemann Educational Books, 1982. 34 M. J. Frazer and D. Servant, unpublished. 35 H. Kramers-Pals, J. Lambrechts, and P. J. Wolff, ‘Conversion of Quantitative Problems in General Chemistry to Standard Problems’, personal communication, 198 I.36 H. Kramers-Pals, J. Lambrechts, and P. J. Wolff, ‘Recurrent Difficulties of Students in Solving Quantitative Problems in General Chemistry’, J. Chern. Edur., 1982, 59, (June issue). Solving Chemical Problems Table 4 Guidelinesfor solving some problems in organic structure analysis and synthesis Organic structure analysis (1) Set out the problem in the form of a flow-scheme. (2) Calculate the ‘unsaturation index’ of each compound and interpret this in terms of possible combinations of rings and multiple bonds. (3) Write down possible explanations for each step in the flow-scheme, and exclude any contradictions in these explanations. (4) Work through the scheme, starting from the part where there is most information and using the explanations derived in (3).(5) Check for alternative solutions, and check that the proposed solution is chemically correct and fits all the information in the problem statement. Organic synthesis (1) Write out in full the formula for the target molecule. (2) Examine target molecule for its main features. (3) Select ‘equivalent molecules’ (i.e. the same carbon skeleton) to target molecule. (4) Split up equivalent molecule into possible precursors. (5) Combine possible starting molecules into these precursors. (6) Select synthetic route. an empirical formula), but in general it is not recommended to teach students algorithms for solving chemical problems. E.Reasons for Failure to Solve Chemical Problems.-In sections A-D emphasis was on what leads to success. We now turn to the results of research that indicate reasons for failing to solve problems. It is assumed in this section that a problem has been recognized because it is hardly meaningful to refer to failure when the individual is unaware that he has a problem. Let us then assume that someone has recognized a chemical problem but fails to bridge the gap to obtain a satisfactory solution. There can be three reasons: (a) failing to start, (b) starting but not finishing, and (c) finishing but with a ‘solution’ that is in- correct, or that is not a solution to the original problem. We take each of these in turn. (i) Not starting.This may be due to: (a)lack of confidence, (b)lack of motivation, (c) having too much information, (d) not obtaining an overview of the problem and thus not identifying goals and sub-goals. Problem 5 is a good example for illustrating failure to identify the goal. Frazer Figure 3 Principal phases of the programme of actions and methads far systematic problem solving in science (PAM) ---4--------------r--------1 I I 2. Planning the 'II problem solving 2b. writing down possibly 1 useful relations; i problem situation I I I i interpretation of Table 5 Guidelines far salving numerical prablems in general ckmistry Step I Clarifv and define the prablem Step 2 Select the key equation This relates the required physical quantity to some or all of the physical quantities available from the data given in the problem.Step 3 Derive the equation for the calculation This is derived from the key equation and is in the form of the required physical quantity on the left hand side and only known physical quantities on the right hand side. Step 4 Collect the data, check the units, and calculate Step 5 Review, check, and learn from the solution Problem 5 3.00 g of phosphorus pentachloride (vapour) are heated in a closed 1 .OO dm3 vessel at 300 "C. The degree ojdissociation according to the equation: PCIs(g) +PCI3(g) +Clz(g) is then 0.300. Calculate the density of the equilibrium mixture Solving Chemical Problems Table 6 Guidelines for solving stoicheiometric problems Write balanced equations for all the processes.Hence find the stoicheiometric ratio of the unknown to the known species. Convert all the given quantities (masses, volumes, concentrations eic.) into moles of specified chemical species. Find the moles of the specified unknown species. Convert moles of the unknown species into the required quantity (mass, volume, concentration etc.) The problem was first suggested by Selvaratnam37 and has since been used by him and the author to bewilder chemists at all levels. Experienced chemists have been found to cover a page or so of algebra based on before they realized that the goal-was to find the density of a stated mass of gas in a closed container of fixed and stated volume.It is a common mistake to assume, on the basis of a superficial reading of the problem statement, that here is a problem of a particular type and then to embark on some known procedure (e.g. in numerical problems this may take the form of writing down a known equation). The successful problem solver, on the other hand, obtains an over- view of the problem and identifies the goal. One of the major differences between the novice and the expert is the greater amount of information the expert can handle.21922 Through greater knowledge and experience the expert sees patterns (‘chunks’) in the given information. He is able to work with these chunks as if they are single items of information. On the other hand, the novice does not see the pattern and tries to cope with considerably more items of information.The question of processing chemical information by ‘chunking’ has been discussed by Johnstone.13J8 (ii) Starting but not Finishing. This may be due to: (a)any of the reasons for not starting, (b)absence of, or failure to recall, required knowledge, (c) knowledge incorrectly recalled or applied, (d) failure to use items of knowledge that are available to the individual (e.g. information given in the problem statement), (e) failure to make approximations, (f) becoming ‘set’ (i.e. fixed in a particular mode of thought) as a result of either failing to make a guess or of imposing unnecessary constraints. Of these, (b) is the most important. The author and ~o-workers~~~~~~~~ have now tested many secondary school and university level chemistry students with a range of problems.The subjects’ written attempts were analysed using the 37 M. Selvaratnam, Educ. Chem., 1974, 11, 201. 38 A. H. Johnstone and N. C. Kellett, Eur. J. Sci. Educ., 1980, 2, 175. 38 M. J. Frazer and R. McCabe, ‘Students’ difficulties with chemical problem solving’, Paper presented at the international seminar: Chemical Education in the Coming Decades- Problems and Challenges, Ljubljana, 1977. Frazer network methodl9 (see section 3C). Subjects were interviewed and also took a test, which included items to see whether or not they possessed the knowledge and skills required to solve the problem. It is then possible to divide the subjects into four groups as shown in Figure.4. Figure 4 Division of students into four groups according to problem and test results All attempts I Problem Problem correct incorrect I I I I Test items Some test Test items Some test for required items for for required items for knowledge required knowledge required are all knowledge are all knowledge correct are correct are incorrect incorrect GROUP A GROUP B GROUP C GROUP D Students in groups A and D need not concern us at present. It is students in group C who are of most interest. They fail to solve the problem, yet they possess all the required knowledge. In a typical result20 taken for Problem 6, 54 % of the students were in group C. There is always a small number of students in group B.This is not surprising since the test is unlikely to be 100% reliable; the context of the problem may prompt the student to recall the information whereas the test item did not prompt him; and most likely, the student was able to obtain the solution either by guessing, or by assuming, the missing required knowledge. Problem 6 The haemoglobin from the red corpuscles of most mammals contains approximately 0.33 % iron by mass. Physical measurements indicate that haemoglobin has a relative moleciilar mass (molecular weight) of 68 x 103. How many iron atoms are there in one haemoglobin molecule? Examples of all the other causes of starting but failing to finish have been identified by the work at East Anglia.20~39 (iii) Finishing but with an Incorrect Solution.This may be due to: (a)any of the reasons for not starting or failing to finish, (6)errors in arithmetic, (c) failure to check final answer (e.g. for orders of magnitude, for correct number of decimal places, for correct units, for ‘pentavalent’ carbon for a chosen reagent that will attack, or be attacked by, some other part of the system, etc.). Solving Chemical Problems 5 Teaching Chemical Problem Solving A. General Considerations.-If, as seems generally agreed, we want to improve the chemical problem solving abilities of our students, then the conditions listed below need to be followed as far as possible. (i) Practice. Students must be given opportunities to practise solving problems. Not much will be learnt about problem solving by reading about it or by listening to someone talk about it or demonstrate it.More time in courses needs to be allocated for students to experience at first hand the deployment of their chemical knowledge and skills in order to move from the problem state to the solution state. Testing recall of knowledge and testing the ability of students to substitute numbers into an equation or to follow an algorithm is problem solving at the lowest level. (ii) Develop Confidence. It is important to develop the student's confidence that he can solve problems. A number of points are worth noting. (a) The student needs to be confronted as far as possible with problems care- fully selected to provide tasks which are not beyond his knowledge and level of skill.All too often the problems presented to students require the use of some obscure piece of knowledge or the application of an only recently acquired concept that is still insecure in the student's framework of knowledge. Problem solving is not something to be met by the student for the first time in an exam- ination. The teacher must select from the literature chemical situations that will be real problems to the student because the solution(s) will not be obvious and yet the necessary information and reasoning is likely to be well within his grasp. In the course at UEA described in section B problems of this type have been used. For example Problem 7 does not require any great depth of knowledge or skill and yet the compounds X, Y,and Z are unlikely ever to have been met by the second year students taking the course.Problem 7 Sulphur tetrafluoride and ammonia react at -95 "C to give nitrogen, ammonium fluoride, and a yellow solid X (N, 30.4%; S, 69.5%; relative molecular mass 1 84). Reduction of X with sodium dithionite gives a white solid Y (N, 29.8%; S, 68.1 %; H, 2.1%; relative molecular mass 188). X reacts with chlorine to give 2 (N, 17.2%; S, 39.3%; CI, 43.6%; relative molecular mass 245). What are the formulae of X, Y, and Z? The infrared spectrum of Y shows bands at 3320 and 3285 cm-1. Complete reduction of X using hydrogen iodide gives quantitatively hydrogen sulphide (infrared bands at 2684 and 1290 cm-I) and ammonia (infrured band at 3336,1628, and 950 cm-l).Using this information what can you deduce about the structures of X, Y, and Z? (b) The intention should be for students to succeed and not to fail. Of course problems should be challenging, and teachers must be constantly pressing their Frazer students, but unless students quite frequently experience the pleasure of bridging the gap and reaching a solution, they will lose confidence and interest. (c) There is a tendency for teachers to ignore the students who have obtained a correct solution, just putting a tick at the end. This does not help. Teaching problem solving is about teaching the processes of problem solving. The student with a correct answer needs guidance, comment, and encouragement about his approach and the strategies used, just as much as the student with no solution or with an incorrect one.The suggestion in step 5 of the guidelines shown in Table 5 is an important one. It is check and learn from the solution. Teachers should try to inculcate this habit in their students, with the hope that it might then stay with them for life. Certainly, successful researchers and industrial chemists are those who are constantly examining, and learning, from their own experiences. Students need to be shown what they have achieved by their solutions or attempted solutions. (d) One of the most common ways of developing confidence in problem solving is to use small groups and peer tea~hing.l~J~~~~-43 Students are more willing to make guesses, and to try trial and error methods in the absence of their teacher.They are likely to be patient as they explain to each other their ideas and learn from one another as they try to bridge the gap from problem to solution. The expert who has crossed the gap and is ‘looking back’ from the solution is likely to be less able to explain the problem and to see the student’s difficulties. (e) Presenting a complete solution in the form of a network19 allows the student to see the many possible routes from problem to solution. Some students lose confidence if the route they chose does not correspond to the one presented in a linear fashion in a text book or by the teacher. (iii) Use Guidelines. In the early stages of teaching problem solving it is necessary to provide the student with some general guidelines (see 4C).The group at Twente University of Technology proposes the use by students of a key-relations ~hart,~5 which is a summary of the major equations relating the various physical quantities in the topic area. (iv) Limit the Amount of Information. In the early stages, the teacher should as far as possible present problems in which the amount of information the student has to handle at any one time is not too large44 (three or four separate items at the most). Later, with experience, the student will see patterns in data and will be able to handle more and more information as a consequence. (v) Provide Realistic Problems. Too often the problems presented to students are academic (closed problems with most of the required information given in the problem statement).This arises because of the false relationship between problem solving and assessment. The student is not given any experience of problem recognition or definition, gains no experience of deciding what 40 M. Brewer, SIMIG, Stud. Higher Educ., 1977, 2, 33. O1 K.G. Collier, Stud. Higher Educ., 1980, 5, 55. 4* G. D. Moss and D. McMillen, Stud. Higher Edur., 1980,5, 161. 45 A. D. Ashmore and M. J. Frazer, ‘The Evaluation of a Problem Solving Course’, in Research for the Classroom and Beyond, The Chemical Society, 1977. 44 A. E. Mihkelson, Educ. Chem., 1982, 19, 24. 187 Salving Chemical Problems information he should obtain from the literature or by experiment, and does not have to choose the best from a range of possible solutions.There is no shortage of books giving worked examples and exercises in closed academic type prob- lems33$45but we need to give students more opportunities to experience problem solving as it really is outside the classroom. There is, however, a shortage of suitable examples and of experience on how best to use the material that is available. A set of fourteen case studies of problem solving in the chemical industry,*46 examples of design,47 the Scottish Chemistry Teaching Materials,4s and the extension study to the S304 Open University Course49 are some of the few examples of material available at the present time. B. A Chemical Problem Solving Course.-The author has tried to include as many as possible of the principles listed in 5A in a course for second year B.Sc. honours chemists at the University of East Anglia.The chemical theme of the course is non-transition elements but the main aim is to help to develop the students’ problem solving skills. The course is now in its eighth year and an early version has been described and evaluated.43 The course is short, lasting for ten one hour sessions. There are five components of the course. (a) One lecture on strategies in chemical problem solving. (b) Five sessions, described in more detail below, in which students attempt to solve problems (Problem 7 *is a typical example). (c) One session which takes the form of two games in which students experience problem recognition and working in a syndicate under pressures of time and finance (information to solve the problem has to be ‘bought’ with Monopoly money).(d) The equivalent of two sessions spent at a computer terminal, solving problems in which no initial information is given and the student has to decide what information to request from the computer. (e) One session, which is a course test consisting of three problems, for which there are course marks. No course marks are given for the other components of the course. *Problems 3 and 4 are examples. 45 M. C. V. Cane and M. J. Tomlinson, ‘Organic Chemistry: A problem solving approach’, Mills and Boon, London, 1977; G. C. Long and F. C. Hentz, ‘Problem exercises for general chemistry’, J.Wiley and Sons, New York, 1978; E. 1. Peters, ’Problem Solving for Chemistry’, W. B. Saunders Co., Philadelphia, 1976: C. H. Sorum and R. S. Boikess, ‘How to solve general chemistry problems‘, Prentice Hall, New Jersey, 1976; C. J. Willis, ‘Problem solving in general chemistry,’ Houghton Miffin Co., Boston, 1977. 46 R. J. Casey and M. F. Frazer, ‘Case studies of problem solving in industrial chemistry’, to be published. Details from the author. 47 C. J. Suckling, K. E. Suckling, and C. W. Suckling, ‘Chemistry through models’, Cambridge University Press, 1978. 48 N. Reid, ‘New Chemistry Teaching Materials’, Scottish Council for Educational Technology, 1980. 411 Extension Study I, to the Course S304, The Open University, 1976.188 Fruzer The five sessions in which a single problem is presented take the followinr: form. Each problem has a closed and an open part. The closed part should be possible to solve in about ten minutes. At the beginning of the session each student is presented with the problem and with carbonized paper. For the fir\t twenty minutes the students work individually writing their solutions on the carbonized paper. They are encouraged to write down all their ideas and gue\\e\ as they work at the problem. At the end of twenty minutes they hand in one COPY of their solution. These are marked by the tutor, who makes written comments about the chemistry and the problem solving strategies, and in due course returns them to the student.The students take the remaining copy of their solution into a peer group with three other students. For the next twenty minutes they then share and discuss their solutions in these groups. For the twenty minutes, there is a plenary session during which a spokesman for each group presents the solutions for the closed and open parts of the problem. Finally, every student is given a handout showing the solution in network form.]" Although the course is short, students do seem to gain in confidence and in their ability to solve this type of problem. They also learn some non-transition element chemistry. Each year the formal and informal 'feedback' from the students is highly favourable. However, perhaps the best testimonial for the course comes from the number and quality of attempts at the inorganic problem5 included each year in the final examination.6 Conclusion It is widely accepted that professional chemists and chemistry students at all levels should be able to identify and solve problems. The nature of chemical problems and the skills needed to solve them are extremely varied. Research into general, and chemical, problem solving is revealing not only successful strategies and guidelines but also the causes of difficulties. Results from this type of research are needed so that the teaching of chemical problem solving can be improved. There is no doubt that, with the right experiences, students can become better at solving problems. Furthermore, although there is no strong evidence, a problem solving approach to teaching may help with all aspects of learning chemistry.Finally, it must be emphasized that a problem is something that an individual perceives. A given chemical situation may not be a problem at all for one individual, but may require high orders of creativity, or even serendipity (discovery by chance), for another. It was Louis Pasteur who wrote: '. . . chance only favours the prepared mind.' This quotation highlights one of the main points of this paper-chemical problem solving requires chemical knowledge. We must not allow the problem solving approach to teaching to cause our students to think that knowing and understanding the facts, concepts, and principles of chemistry is unimportant.Acknowledgment. It is a great honour to have been awarded the Nyholm medal and 1 wish to express my gratitude to the Royal Society of Chemistry. Solving Chemical Problems Sir Ronald Nyholm was a great inspiration to me, not only in my teaching and research in inorganic chemistry, but also because of his enthusiasm for efforts to improve the teaching and learning of chemistry. It was in no small part his encouragement that led me to commit myself professionally to chemical education. I should also like to thank all my colleagues and students who have taught me so much during the last twenty-five years and I would particularly like to thank all those who have worked, or are working with me, on chemical problem solving.

ISSN:0306-0012    

DOI:10.1039/CS9821100171

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

TILDEN LECTURE* Carbon-Carbon Bond Formation Involving Boron Reagents By A. Pelter DEPARTMENT OF CHEMISTRY, UNIVERSITY COLLEGE OF SWANSEA. SINGLETON PARK, SWANSEA, SA2 SPP, U.K. 1 Introduction The uses of boron reagents in organic synthesis are so various and the subject so vast that it is impossible to cover all aspects in a single lecture. In Figure 1, are listed some of the topics that are being omitted, each of which is interesting and useful. Some, such as boron reagents as reductants are worthy of many lectures and their omission simply reflects the organic chemist’s abiding and necessary interest in the central theme of the efficient welding together of carbon frameworks. TOPICS OMITTED 1. Boron reagents as reductants. 2. Boron reagents as protecting groups.3. The preparation of organoboranes. 4. Formation of 0-C, N-C, Hal-C, S-Cand Metal-C bonds. 5. The photochemistry of organoborates. Figure 3 The themes that will be discussed are summarized in Figure 2. Although, for convenience, the topics are labelled separately they interact with each other and their use in tandem can lead to unique synthetic routes. 2 Electrocyclic Reactions The basis of these reactions is summarized in Figure 3, in which X can be CR2, NR, or 0, and the double or triple bond may be C=C, C=O, C-N, CcC-Z, or C=N. [In the Figures that follow a ‘thick’ bond denotes a new *First presented at a meeting of the Perkin Division of the Royal Society of Chemistry at the Scientific Societies’ Lecture Theatre, London, on 15 December 1981.Curbon-Curbon Bond Formation Involving Boron Reagents CARBON-CARBON BOND FORMING REACTIONS. Classes to be discussed. 1. Electrocyclic reactions of organoboron compounds. 2. 1,2-Migrations involving organoboranes (B(II1)). 3. Radical reactions of organoboron compounds. 4. 1,2-Migrations involving organoborates (B(1V) ). 5. Boron stabilised carbanions. Figure 2 carbon-carbon bond made in the reaction. Wedged and dashed bonds retain their usual significance.] Figure 3 A. Ally1boranes.-The usefulness of allyl boranes can be diminished by permanent allyl migrations, which can not only lead to mixtures of the type -B--CR2CH--=CH2 and --B-CH~CH-:CRZ but also undermine the con-figurational identity of a double bond.This can be overcome in three ways: (a) by substituting at boron with substituents (NRz, OR) capable of 7~ back-donation,] (Figure 4), (b) by complexing with amines,2 and (c) by large steric interactions, as with 9-BBN-alIylb0ranes.~ The insertion reactions of allylboranes have been extensively st~died.~~~ Examples from a wealth of chemistry are shown on Figures 5 and 6. In Figure 5 K. G. Hancock and J. D. Kramer, J. Organomet. Chem., 1974,64, C29, cf. J. Am. Chem. SOC.,1973, 95, 6463. a G. W. Kramer and H. C. Brown, J. Organomet. Chem., 1977, 132, 9. B. M. Mikhailov, Pure Appl. Chem., 1974,39, 505-523. B. M. Mikhailov, Sov.Sci. Rev.,Sect, B, Chem. Rev., 1980, 283-355. Pelter Et -Me2N.BMe2N€!I+ E, = 24-6 Kcal.mol-' -15OoC/3 h A S= -20.6 e.u. M e2N.B -MqN.8 17OoC/24h, c,a 2% conversion 200°C, no rearrangement. Figure 4 the first intermolecular insertion reaction is followed by a second intramolecular insertion at higher temperature. Finally, at yet higher temperatures, a reorganization ensues, which gives borabicyclic compounds with great ease. The reaction is general, and when X = R, the temperatures required for the second and third rearrangements are lowered. A similar reaction occurs with allenes, and by suitable manipulation (Figure 6), boradamantane is readily produced from allene and triallylborane. Substituted allenes yield the corresponding boradamantanes. By gentle processes, which will be illustrated later, the boron atom can be replaced by carbon, thus giving rise to unusual but rational syntheses of difficultly available carbocycles.Benzylboranes behave similarly to allylboranes5 (Figure 7) with the happy limitation that, due to a ready prototropic shift, the reaction proceeds only to the first stage, giving rise to the equivalent of pure ortho-substitution. This situation can be induced more generally by use of a dialkyl allylborane6 (Figure 8). Of course there are very many variations on this general theme one such, leading to a 1,3-diketone synthesis is shown in Figure 9.7 B. M. Mikhailov, Y. N. Bubnov, S. A. Korobeinikova, and S. I.Frolov, J. Orgunornet. Chem., 1971, 27, 165. B. M. Mikhailov, Y. N. Bubnov, and S.A. Korobeinikova, Izv.Akud.Nuuk. SSSR, Ser Khim., 1970, 2631; J.Prakt.Chem., 1970, 312, 998. V. A. Dorokhov and B. M. Mikhailov, Dokl.Akad.Nurrk SSSR, Sw Khim., 1969, 187, 1300, Chem. Ahstr., 1969, 71, 124551r. I93 Bond Formation Involving Boron Reagents ,OEtBa2O0C Intermolec u lar Allyl insertion react ion Intramolecular insertion reaction 543 120OC 130-18OOC ___) a71 Figure 5 Allyl 1. MeOH 2. 'Et 2 BH'--[@I -Allyl car bonylation-or cyanoborat e Figure 6 Pelter (Bzl = 6enzy) Figure 7 Figure 8 The known cyclic insertion reactions of aldehydes with allylboranes have been combined with knowledge of the configurational stability of dialkoxy- and diamino-allylboranes to produce enantio- and stereo-selective syntheses that proceed in high yield in mild conditions.Figure 10 shows the general process, of which three examples (a)*,(!I)~, and (cV0 are given in Figure 11. Of course ozonolysis of the alkenes produces yields of carbonyl compounds that would otherwise require synthesis by highly selective aldol condensations. 8 T. Herold, U.Schrott, and R. W. Hoffmann, Chrm. Blv., 1981, 114. 359. 9 R. W. Hoffmann and W. Ladner. Tctrruhdron L~it.,1979,4653. '(1 R.W. Hoffmann and B. Kemper. Tr.tralwr/ron Lett., 1980, 4883. J 95 Cnrbon-Carbon Bond Formation involving Boron Rrugents __I) HO.CHR?CHR!COX Erythro and Threo Figure 10 1. CH3CH0, -1IOOC L-(a1 2. N (CHZCH~OH),&?JPh Opt. purity,86OIo configurationally stable I.CH3CHO 2. N(CHzCH2OHI3 * 9-+OR 93% yield,98% eryt hro 65% Opt.purity MeS SMe - Figure 11 SEt Pelter These, and other reactions to be discussed, proceed via well defined transition state@ that are particularly tight due to the short B-0 bond (Figure 12). RCHO RCHO -H I 4I Figure 12 B. Viny1oxyboranes.-These are well defined and readily available compounds12 that do not suffer from the danger of allylic rearrangement and also undergo electrocyclic reactions. The contrast between the reactions of vinyloxyboranes and vinyloxysilanes is shown in Figure 13.13 Clearly the borane is not acting as a simple enolate derivative, but instead is undergoing an electrocyclic reaction analogous to those of allylboranes.Once more an electrocyclic reaction involving a rigid transition-state allows control of stereochemistry in the synthesis of aliphatic systems. The precise organoborane used can be important (Figure 14) as, when the vinyloxyborane is produced by an equilibrium reaction, the bulk of the groups can control the stereochemistry of the double bond and hence the relative stereochemistry of the product.14 Ketene acetals and thioacetals15 (Figure 15) also undergo the same type of reaction leading directly to important intermediates on the route to the l1 R. W. Hoffmann and H. J. Zeiss, J. Org. Chem., 1981, 46, 1309. l2 R. Koster and W. Fenzl, Angew. Chem., Int. Ed. Engl., 1968,7, 735. l3 S. Kuwajima, M. Kato, and A. Mori, Tetrahedron Lett., 1980, 4291.l4 D. E. van Horn and S. Masamune, Tetrahedron Lett., 1979, 2229. l5 M. Hirame, D. S. Garvey, L.D.-L. Lu, and S. Masamune, Tetrahedron Lett., 1979, 3937. 197 6 Curbon-Carbon Bond Formation Involving Boron Reagents OSiMe3 OBBUZ Bu2BOTf* R'+ 4-Me3SiOTf f4--R erythro :threo 49 : 51 Rl+R OH R H R2 R2 R&O eryt h ro, 95 threo, 5 Y;,kcBu R Bu Figure 13 97% eryt hro PrjNEt RCHO -85% threo Figure 14 production of macrolide antibiotics, perhaps the most active field of current synthetic effort. The enantio- and stereo-selectivity of the reactions has recently resulted in a Pelter d~~~ent2RCHO~ SBut + Pentzf30Tf -- R z SBu' -- 5s- 0 1. 'SPh Me 02C uCHO 2. CF3C02H * +.*h gE-SPh Two isomers only Figure 15 OSiMezBu' OB(O), RCHO .%-3lI,oButMezSi0 II absolute configuration H absolute and relative configuration Figure 16 199 Curbon-Carbon Bond Formution In volvitw boron Neuhvwts synthesis of 6-deo~yerythronoIide;~~ja portion of this synthesis is shown on Figure 16.The stereochemical control in the build up of highly functionalized aliphatic systems effected by the use of electrocyclic insertion reactions of allyl- and vinyl-oxyboranes is unique and its importance in organic synthesis, large already, is rapidly growing. Vinyloxyboranes can also be advantageously used as simple enolates for halogenation and alkylation (ref. 17, p. 920). 3 1,2-Migrations of Organoboranes If a transiently four-covalent boron atom bearing an organyl group is adjacent to an electron deficient carbon atom or one bearing a leaving group, then 1,Zmigration of the organyl group from boron to carbon can occur.The migration most frequently proceeds with retention of configuration of the migrating group and therefore any stereochemistry built in by, say hydro- boration, is retained in the product borane. There are several ways of attaining the situation leading to migration, perhaps the most general being shown in equations 1 and 2 of Figure 17. Equation 1 illustrates the reaction of a tri- organylborane with a carbenoid i.e. an anion bearing an appropriate leaving group, whilst equation 2 shows a-substitution of an organoborane followed by migration induced by nucleophilic attack on b0ron.1~ R IR,B + XCHR'---+ [R3i--CH(XjR'] -R2B-CHR' (1) X R I I-RB.CR'R2 (2)R2BCHR'RZ-R2B.C(X)R'R2 '-b[P2;1-CR1R2] I Y Y Figure 17 There are many examples of equation 1, including the use of anions derived from a-halocarbonyl compounds 18~and a-halonitrilesl*b (Figure 18).It should be noted that the intermediate organylboranes may not have the exact structures shown owing to the possibility of oxyallylborane rearrangement or its equivalent. An example showing a double alkylation of a dihal~nitrilel~is shown and also the direct production of a carboxylic acid in which phenoxide acts as a leaving group.20 The use of alkyl-9-BBN derivatives is successful in both types l6 S. Masamune, W. Choy, F.A. J. Kerdesky, and B. Imperiali, J. Am. Chern. SOC.,1981, 103, 1566. l7 A. Pelter and K. Smith, 'Comprehensive Organic Chemistry' Vol. 3, ed. D. N. Jones, Pergamon Press, Oxford, 1979, p. 791-904. I* (a) H. C. Brown, H. Nambu, and M. M. RogiC, J. Am. Chem. SOC.,1969, 91, 6852. (6) H. C. Brown, H. Nambu, and M. M. RogiC, ihid., 1969, 91, 6854. I9 H. Nambu and H. C. Brown, J. Am. Chem. SOC.,1970,92, 5790. 2o S. Hara, K. Kishirnura, and A. Suzuki, Tetrahedron Lett., 1978, 2891. Pelter of reaction, so allowing full utilization of the alkyl groups.21 As compared with alkylations of enolate anions with alkyl halides or their equivalents, reactions using organoboranes have the advantage of allowing arylation, vinylation, and alkylations with groups such as cyclohexyl and 2-norbornyl that are resistant to sN2 displacement but migrate readily with retention of configuration.PX @? + XCHY -[ CB-CH-Y R-CHzY I R X = I, Br,CL; Y = CN,COR,C02R R CL R R' ---+ R-CHCKN -R-~H.CN CH3 CH:CH.(C H 7 13-CH2 C02H (90% overall) Figure 18 Use of the anion from dichloromethoxymethane (Figure 19) allows three migrations to proceed with even very hindered alkyl groups,21 whilst use of alkoxydialkylboranes yields ketones.22 a1 H. C. Brown and B. A. Carlson, J. Org. Chem., 1973, 38, 2422. 22 B. A. Carlson and H. C. Brown, Synthesis, 1973, 776. 201 Carbon-Carbon Bond Formation Involving Boron Reagents CI Cl R R -1 II I R,B + CCI,OMe w R3B-C-OMe -B-C-R -R-C.OH I 1ICI OMe'R R OR I 0 Figure 19 The readily available tris(pheny1thio)methane anion similarly can be manipulated to give three migrations from boron to carbon.23 Two .migrations are spontaneous however, and the reaction is readily stopped to give ketones on oxidation (Figure 20).The reaction bears a generic similarity to the use of benzo- dithiolium carbanions, which can also yield ketones and trialkylcarbinols, but this time with only one and two migrations respectively from boron to carbon (Figure 2O).24 In each case the initial migration produces a nucleophilic species that can co-ordinate to boron and induce a further migration. The attacking species can be an ylide (equation 1;X positively charged) as in Figure 21, in which the leaving groups attached in the a-position are +SMe225 and +N2.26 The reactions proceed efficiently in mild conditions, utilization of the alkyl or aryl groups being enhanced by use of the readily available chloro- or dichloro-organoboranes.27 Into this category comes the carbonylation reactions of organoboranes 23 A.Pelter and J. Madhusudhana Rao, J. Chem. SOC.,Chem. Commun., 1981, 1149. a4 S. Ncube, A. Pelter, and K. Smith, Tetrahedron Lett., 1979, 1893; 1895. 25 J. J. Tuffariello, P. Wojtkowski, and L. T. C. Lee, Chem. Commun., 1967, 505. " J. HOOZet al., J. Am. Chem. SOC.,1968, 90, 5936; 6891. Can. J. Chem., 1970, 48, 868; 357i, 49, 2371.'' J. Hooz, J. N. Bridson, J. G. Calzada, H. C. Brown, M. M. Midland, and A.B. Levy, J. Org. Chem., 1973, 38,2574. 202 Pelter R I R,B + c(SPh13 --t[R3i-C(SPh),] -----+b2B-I C(SPh),] SPh SPh R8 -R B-C. 9 S Ph Hg" * f3-S-RII SPh R XR R'>=O R R R R' R8.C-R 'C'II XR R' 'OH Figure 20 (Figure 22).28 These, as well as providing facile routes to aldehydes29 and primary-carbinols30 of known regio- and stereo-chemistry, yield, by two- and three-bond connection reactions, either ketones31 or tertiary carbinols32 respectively. In this way the facility of production of cyclic organoboranes can be transferred to the synthesis of unique functionalized carbocyclic compounds. Examples of these transformations are given in Figure 23. Many substituted boradamantanes (see Section 1) have been transformed to the corresponding adaman tanes .4 aB M.E. Hillman, J. Am. Chem. SOC.,1962, 84, 4715; 1963, 85, 982, 1626. H. C. Brown, 'Boranes in Organic Chemistry', Cornell University Press, Ithaca, 1972. as H. C. Brown, M. M. RogiC, M. W. Rathke, and G. W. Kabalka, J. Am. Chem. SOC., 1969, 91, 2150; H. C. Brown and R. A. Coleman, ibid., 1969, 91, 4604. 30 H. C. Brown, T. M. Ford, and J. L. Hubbard, J. Org. Chem., 1980, 45, 4067. 31 H. C. Brown and E. Negishi, Chem. Commun., 1968, 594. sz H. C. Brown and E. Negishi, J. Am. Chem. SOC.,1967, 89, 5478; 1969,91, 1224. 203 Carbon-Carbon Bond Formation Involving Baron Reagents Ph + -1-+ Ph3B 4-CHzSMez-Ph,B -CH2-SMe, -Ph2BCH2-Ph0 R R,B t i2&X [R~t3CHXX] R-CH2 X + N,’3 X = CH0,C02R’,COR’ PhBCIz + NzCH.CO2Et ____t_.) Ph-CHzCO2Et Figure 21 -+ -[R38-C?;)]R,B + C=O --+ [R,B-CO-R] (1) A reducing RZB-CH-R--RZBCHZR 5 -agents I Figure 22 204 Pelter H H H H HQ+eH Figure 23 Equation 2 (Figure 17) is exemplified by the facile photochemical a-bromi- nation of organylboranes in the presence of water.The bromination proceeds without cleavage of the B-C bond and if water is present migration is immediate. The process may be repeated to build up highly hindered carbinols with great ease.33 If it is desired to stop the reaction at one migration only, then full utilization of the alkyl groups can be achieved by use of dialkylhydroxy-boranesNa or thexylboranes.346 Ring structures may also be produced by this reaction35 (Figure 24).A further method for the production of a-haloboranes is the hydroboration of vinyl halides. The results shown in Figure 25 indicate that the migration may be controlled so that clean inversion occurs at the migration terminus. This 33 C. F. Lane and H. C. Brown, J. Am. Chem. SOC.,1971,93, 1025. 34 (a) H. C. Brown, Y. Yamamoto, and C. F. Lane, Synthesis, 1972, 303. (b) H. C. Brown, Y. Yamamoto, and C. F. Lane, ibid., 1972, 304. 35 Y.Yamamoto and H. C. Brown, J. Org. Chem., 1974,39,861. 205 Carbon-Carban Bond Formation Involving Boron Reagents CH3 CH3 Br OH CH3 Br OH CH3 I/ Br2/ hS Ill CH3CH2CH. B.C.CHzCH3 CH3CH2y-B -t.CH2CH3 II CH3 CHCHzCH3 CH3 CH.CH2CH3 c/H3 c/H2 OHCH3 blH20 I I -(HO),B.C.CH2CH-j -CH,.C.CH*CH,I I CH-yC CH2CH3 CH3.C .CH,CH, i I CH3C .CH 2 CH 3 CH3-C .C H 2CH 3 I I H H Figure 24 result is sensitive to the complexing agents present, however, and in the presence of THF or Me2S almost complete stereochemical scrambling occurs.36 It is important to note that nucleophilic attack on an a-haloalkenylborane also readily induces migration in which the migrating group retains its con- figuration and there is clean inversion at the migration terminus (Figure 26).36 M. M. Midland, A. R. Zolopa, and R. L. Halterman, J. Am. Chem. Soc., 1979, 101, 248. Pelter it it Br Me Me, Figure 25 Such processes are of great interest in forming alkenes37 and diene~~~ of known stereochemistry in mild conditions.4 C-C Bond Formation by Radical Reactions of Organoboranes A.-Alkyl Coupling Reactions.-Both trialkylboranes and a1 kyldihydroxy- boranes react with alkaline silver nitrate to give coupled alkane~3~~~~ (Figure 27). The reaction appears to proceed through silver alkyls and all alkyl groups are utilized. Mixtures of boranes or mixed trialkylboranes give statistical mixtures of coupled products showing the intermolecular nature of the coupling. Of course, use of excess of any trialkylborane can cause a corresponding increase in the percentage of mixed coupled product when this is required. The process has been used with readily available borocyclanes to produce cycloalkanes (Figure 27), this mild coupling method having potential in the synthesis of complex carbocyclic corn pound^.^^ 37 E. Negishi, J.-J.Katz, and H. C. Brown, Synthesis, 1972, 555; E. J. Corey and J. Ravindranathan. J. Am. Chem. Soc., 1972, 94, 4013. 38 E. Negishi and T. Yoshida, J. Chem. SOC.,Chem. Commun., 1973, 606 39 H. R. Synder, J. A. Kuck, and J. R. Johnson,J. Am. Chem. Soc., 1938,60, 105; 111. 40 H. C. Brown, C. Verbrugge, and C. H. Snyder, J. Am. Chem. Soc., 1961,83, 1001; H. C. Brown and C. H. Snyder, ibid., 1961, 83, 1002. 41 R. Murphy and R. H. Prager, Tetrahedron Lett., 1976, 463. 207 Carbon-Carbon Bond Formation Involving Boron Reagents OMe I+ + d A c=c -___) Br-C ,C.R2 /\Br RZ _____) MR’+ I+ .8 HCI -C =C R’ ‘H OMe bJaOMe H R2 H R2 Figure 26 208 Pelter R\ B~2 L F--or b b 79'lo 2.Ag NO3/ KOH 8'' Figure 27 B. Conjugate Addition Reactions of 0rganoboranes.-Many ap-unsat urated carbonyl compounds undergo ready 1,6addition of organoboranes, the reaction proceeding, in general, through a free-radical chain mechanism (Figure 28)42 though B-alkenyl-9-BBN derivatives seem to react by a concerted mechanism involving a six-membered cyclic tran~ition-state.~~ The radical reactions proceed well even with p-unsubstituted enones, which give poor results using copper reagents, and with methyl vinyl ketone and various a-substituted acroleins the reactions are spontaneous. Radical initiation, however, is required for reaction with p-substituted enones. Related reactions proceed well with ynones and up-unsaturated imines as well as up-alkenyl- and alkynyl-epoxides.44 However, electrolytic conditions are required to induce reaction of up-unsaturated e~ters.4~ para-Quinones react readily in the presence of 0xygen.~~7~7 5 1,2-Migrations Involving Organoborates (BIV) Organoboranes may be attacked by simple carbanions to yield co-ordinatively saturated organoborates, which show no tendency to undergo migration.However, electrophilic attack 18 to boron creates a dipole so that migration can 4a H. C. Brown et al., J. Am. Chem. SOC.,1970,92, 710; 712; 714. 43 P. Jacob and H. C. Brown, J. Am. Chem. SOC.,1976,98, 7832. 44 A. Suzuki, N. Miyaura, M. Itoh, H. C. Brown, G. W. Holland, and E. Negishi, J.Am. Chem. SOC.,1971, 93, 2792, CJ Synthesis, 1973, 305. 46 T. Takahashi, K. Yuasa, M. Takuda, M. Itoh, and A. Suzuki, Bull. Chem. SOC.Jpn., 1978, 51, 339. 46 M. F. Hawthorne and M. Reintjes, J. Am. Chem. SOC.,1964,86,951; 1965, 87,4585. 47 G. W. Kabalka, J. Organomet. Chem., 1971, 33, C 25; Tetrahedron, 1973, 29, 1159. Carbon-Carbon Bond Formation Involving Boron Reagents R,B -R' R3B + CH,=CH.CH-CH, -R-CH,CH=CH.CH2OH '0' R3B + CHEC.CH-CH2 -RICH=C=CH.CH~OH '0' 0 OH R3B -4-CHz=C(R1)COzEt RZNX / MeCN b R-CHZCHR 2 COzEt electrolysis(R =H or Me) Figure 28 occur (Figure 29). The situation differs from that involving organoboranes inasmuch as fhe organoborates are stable intermediates and migration is induced by electrophilic rather than nucleophilic attack.Moreover, the nature of the electrophile may be controlled so as to potentiate further reaction after the Pelter initial migration. General equations (3)-(5) (Figure 29) were set up48 and from them a large number of synthetically useful reactions have re~ulted.~~-~' R' 2-+R1R:B-RR2B 1 2-a X=Y+RIE+I R~B-X=Y-EER:B-X=Y I -E (3) R' IE+I 12 IR'R;B-R1R:6-X=Y+R R2B-X.L.Y-E -R~B-X--Y -E (4) For carbon -carbon bond formation, X = C in all cases. Figure 29 The migrating group R1 can be aryl, alkenyl, or alkynyl as well as alkyl, and R2 need not be organyl groups but may be OR, NR2, halogeno, or SR provided these groups do not migrate or interfere with the required reactions. In this way valuable alkyl groups may be conserved. Further variations arise when R1R2R3B is used rather than R22R1B or R3B.A. The Cyanoborate (Cyanidation) Process.-Cyanoborates are readily available, stable salts that are soluble in most organic solvents. Protonation leads to two migrations and oxidation of the product yields ketones.50 Yelds, however, are only ca. 50% as a result of abstraction of HCN from Rd%--C=NH by the basic dimer resulting from two migrations (Figure 30). Replacement of the proton by acylating agents such as trifluoroacetic anhydride (TFAA), trichloroacetyl chloride, or benzoyl chloride as electrophiles leads to the high yielding and three-4gmigration cyanoborate processes (Figure 31). Many types of structure can readily be produced in mild conditions in good yields.These include fused,4*$52,53 bridged,48954 and medium ring ketonesS5 often with striking stereospecificity (Figure 32). 48 A. Pelter, K. Smith, M. G. Hutchings, and K. Rowe, J. Chem. SOC.,Perkin Trans. I, 1975, 129. 49 A. Pelter, K. Smith, M. G. Hutchings, and K. Rowe, J. Chem. SOC.,Perkin Trans. I, 1975, 138. 50 A. Pelter, M. G. Hutchings, and K. Smith, J. Chem. Soc., Perkin Trans. I, 1975, 142. 51 A. Pelter, M. G. Hutchings, K. Smith, and D. J. Williams, J. Chem. Soc., Perkin Trans., I, 1975, 145. 51 G. W. Kabalka, Synth. Commun., 1979, 607. 53 T. A. Bryson and C. J. Reichel, Tetrahedron Lett., 1980, 2381. s4 H. C. Brown, J. A. Sikorski, S. U. Kulkarni, and H. D. Lee,J. Org. Chem., 1980,45,4542. 55 M.E. Garst and J. N. Bonfiglio, Tetrahedron Lett., 1981,22,2075. 211 Carbon-Carbon Bond Formation Involving Boron Reugents R R H+ -1-+ I 2 R3B -2 R$ -CN -2 [R,B-C=N-H] -+2[R2B -C= N H]u RR ‘C’ RB’ ‘NH 2BR3-I I HN, ,BR/“\RR Figure 30 R‘ R’ 0-c t!CORZ 0 RA 1. MCN Figure 31 An interesting conjunction of the two migration cyanoborate process with transition-metal stabilized cation chemistry results in the ready construction of a tetracyclic system (Figure 33).56 It is noteworthy that tris(4-bromobutyl) 5* E. Mincione, A. J. Pearson, P. Bovicelli, M. Chandler, and G. C. Heywood, Tetrahedron Lett., 1981, 22, 2929. PeIrer I, II,III, IV0--E + ,\-.fXoA & III,Cp1,H-BHz l,ll,lll,lv* IllMCNTFAA IV, NaOH/H202 H I, 11, III,IV -6 (-yJI CBz CBZ (CBz = carbobenzyloxy) Figure 32 213 Carbon-Carbon Bond Formation Involving Boron Reagents methanol, though easily produced in the mild conditions of the cyanoborate process, could not be made at all by the carbon monoxide protocol57 (Figure 33).-eoMe4 Steps Me0 0' Cyanoboratg -AcO ACOV U 1. KCN [BH 3 /(CH l4 Br3-Br AB(C4H&3r)3 2.TFAA / O°C H0.C-(CH2),Br3-roi '(CH21,Br Figure 33 B. Alkyny1borates.-Electrophilic attack on alkynylborates may lead to migration and to products in which the entire alkyne unit and the electrophile are incorporated. Prot0nation5~ or alkylation59 give rise to valuable syntheses of ketones (Figure 34) in which the organyl groups have separate origins and in which the regiospecificity of the alkyl groups in R1COCHR2R3 is completely assured.Neither protonation nor alkylation with simple alkylating reagents is stereo- specific, however, and protonation of the borane intermediates yields mixtures of t r isu bsti tuted a1 kenes. However, when propargyl bromide, iodoacetonitrile, and a-bromocarbonyl compounds are used as alkyating agents60 the reactions are completely stereo- 57 J. E. Hallgren and G. M. Lucas, Tetrahedron Lett., 1980, 3951. 58 A. Pelter, C. R. Harrison, C. Subrahmanyam, and D. Kirkpatrick, J. Chem. SOC.,Perkin Trans., I, 1976, 2435. 59 A. Pelter, T. W. Bentley, C. R. Harrison, C. Subrahmanyam, and R. J. Laub, J. Chem.SOC., Perkin Trans. I, 1976, 2419. 6o A. Pelter, K. J. Gould, and C. R. Harrison, J. Chem. SOC.,Perkin Trans. I, 1976, 2428. 214 Pelter R 1%R,B + tiCrCR' ----+R~B--C~CR'-----, R~B-CC-; JJR1 \El 1E Y I" R 0El R IR' 1'c=c + 'C =c R-CO.CH-E I H \Rl H R' E=HX or R2X:E'= H,R2 Figure 34 specific (Figure 35) giving direct syntheses of functionalized trisubstituted alkenes as well as y-keto-esters, nitriles, and alkynes and I ,6diketones, all useful synthetic intermediates. R$C=CR~ .+ XCHzY X=Br, Y = COCH3,COPh,C02Et,C=CH X=I, Y=CN, + BrCH2 COCH, Bun 70 "lo 74"lo one isomer only Figure 35 215 Carbon-Carbon Bond Formation Involving Boron Reugents Alkynyiborates are readily manipulated. Some examples of their uses are shown on Figures 36 and 37.The first example is of interest in that, by the use of a dihalide, two migrations are induced in an alkynylborate with the formation of three C-C bonds in all. A first allylic rearrangement is followed by another on hydrolysis.61 Alkynylborates will attack electron-deficient rings in a regiospecific fashion as shown by interactions with pyridinium62 and metal-stabilized cyclohexadienyl cation ring-sy~tems.~~ The latter is of particular interest as the boron may be removed at different stages of manipulation by either hydrolysis or mild oxidation giving rise to a wide variety of products. ~\6c =CR* R: BcR'=c R R~CHCO-R' t _I_) CH,COCI lPyridine COCH3 I i.Pr'COZH ii. H2021NaOd R~C:CH=R'6 R$C=CR~ + -FeKO), Figure 36 61 A Pelter and C.R. Harrison, J. Chem. SOC.,Chem. Commun., 1975, 828. 6a A. Pelter and K. S. Gould, J. Chem. SOC.,Chem. Commun., 1974, 347. 6s A Pelter, K. J. Gould, and L. A. P. Kane-Maguire, J. Chem. SOC.,Chem. Commun., 1974, 1029. Pelter An alkynylborate is readily attacked by organometal halides, often with complete regio-and stereo-specificity to yield 1,Zdimetallovinyl boranes (Figure 37) from which, for example, the boron may be readily removed to give stereochemically defined vi ny 1-metal der ivatives.l 9 R' R2 R:BC-CR~ + CIMR: ---+ 'c=c / /\ RiB MR: MR: = BR2,PPh2,S;Me3,SnBu3 ,SePh Figure 37 Ethynylborates themselves may be protonated (Figure 38) to yield the vinylboranes that would be obtained if it were possible to carry out Markovnikov hydroboration of a terminal alkyne by a dialkylborane.These intermediates may be manipulated in the usual ways.66~6~ Reaction with butyl-lithium yields a dianion which may be alkylated, this being a general route to alkynylborates.68~69 t-HCI R Lo1LiR3BCGCH )C=CH2 -R-COCH3 Figure 38 64 P. Binger and R. Koster, Tetruhedron Lett., 1965, 1901; J. Organomet. Chem., 1974, 73, 205; Synthesis, 1973, 309. 85 J. Hooz and R.Mortimer, Tetrahedron Lett., 1976, 805. 68 H. C. Brown, A. B. Levy, and M. Mark Midland, J. Am. Chem. SOC.,1975, 97,5017. 67 H. C. Brown and M. Mark Midland,J. Org. Chem., 1975, 40, 2845. 68 K. Utimoto, M. Kitai, M. Naruse, and H.Nozaki, Tetrahedron Lett., 1975, 4233. 69 K. Utimoto, Y. Yabuki, K. Okada, and H. Nozaki, Tetrahedron Left., 1976, 3969. Carbon-Carbon Bond Formation Involving Boron Reagents Alkynylborates react with iodine with migration followed by elimination of R2B1,to give the alkylated alkyne in excellent yield70 (Figure 39). This has been extended to the coupling of alkynes by use of a dialkynylb~rate.~~ It is note- worthy that unsymmetrical diynes with long alkyl side-chains can be readily produced by this route, these being available only in low yields by the more R’ R’ -1C-C CR2R’zBX i-2 LiC CR2 -LiX + Li +RiB( CE CR’), ---+ 1 R2C R2 I H R2 c=c H1 ‘R2 I NHCOCF, Figure 39 70 A. Suzuki, N. Miyaura, S. Abiko, M.Itoh, H. C. Brown, J. A. Sinclair, and M. Mark Midland, J. Am. Chem. SOC.,1973, 95, 3080. 71 A. Pelter, K. Smith, and M. Tabata, J. Chem. SOC.,Chem. Commun., 1975, 857. Pelrer usual meth0ds.7~9~~ Other reactions of alkynylborates bearing functional groups are shown in Figure 3974975it being of interest that an alkenyl group migrates in preference to a cyano-group under the influence of TFAA.75 In order to test whether the ionic chemistry of organoborates was wholly contained in the general equations (3), (4), and (9,unique Michael reactions involving migration from boron to carbon were successfully carried out, showing that neutralization of charge in a dipolar intermediate is not necessary but that stabilization of the charge is sufficient for 1,Zmigration (Figure 40).76977 These R’ R2 t /x I/LiR\BCrCR2 + R3CH=C -RSB-C=C ‘tHR! cI -Y lPriC02H X JR2 .YR ~ CHRZO cHR ~CHX R’-CH= c ‘C H R 3CHX H R’ R2 R3L? R\BkR2 Ill -R;B-CH-CH-CH-~X Y+x R3CH=C / ‘Y lR2 R3 R2 R3 1 I I t R1-CH2-CH -CH CHX Y R’-CH -CH -CH.CHX.Y 1 OH Figure 40 78 J.A. Sinclair and H. C. Brown, J. Org. Chem., 1976, 41, 1078. 73 A. Pelter, R. Hughes, K. Smith, and M. Tabata, Tetrahedron Lett., 1976, 4385. 74 E. Negishi, G. Lew, and T. Yoshida, J. Chem. SOC.,Chem. Commun., 1974, 1411. 76 A. Pelter, A. Arase, and M. G. Hutchings, J. Chem. SOC.,Chem. Commun., 1974, 346. 7e A. Pelter and L. Hughes, J. Chem. SOC.,Chem Commun., 1977, 913. 77 A. Pelter and J. Madhusudhana Rao, Tetrahedron Lett., 1981, 22, 797.219 Carbon-Carbon Bond Formation Involving Boron Reagents reactions have opened up a wholly new field for exploration and, for example, have recently been shown to be applicable to alkenylborates.78 The successful Michael acceptors, RCH=CXY in general are those in which the pK, of CH2XY is not higher than ca. 9-11. C. Alkeny1borates.-The reaction of iodine on triorganylethenylborates leads to alkene~.~~ Both secondary alkyl and aryl groups migrate, and by the use of 9-BBN derivatives organyl groups may be fully utilized. If the boron atom is substituted by alkoxy-groups these do not interfere with the reaction and allow full utilization of the alkoxy groups (Figure 41).80 Migration proceeds with retention of configuration of the migrating group and inversion at the mi- gration terminus.Pent ORA4 0siMe2But OSi Me2But t’ CIi, LiCH =CHCHPent; ii, LiCH =CHCHPent. Figure 41 Protonation of dialkylalkenylborates with dry hydrogen chloride in either gives the expected one-migration produ~t.~~?~~?~~ The reactions of alkenylborates with epoxides and aldehydes yield products that on oxidation yield 1,4- and 1,3-diols respectivelys2 (Figure 42). Aromatic borates behave to some extent like alkenylborates, this making possible some very specific substitution and coupling reactions, examples of which are shown in Figure 43.83984 The variety of substitution reactions on the 78 A. Pelter and F. Gibbs. Unpublished results. 79 G. Zweifel et al., J.Am. Chem. SOC.,1967, 89, 3652; 1968, 90, 6243; 1971, 93, 6309; 1972, 94,6560. D. A. Evans, F. C. Crawford, R.C. Thomas, and J. A. Walker, J. Org. Chem., 1976, 41, 3974. cf. Tetrahedron Lett., 1976, 1427 and also A. G. Abatjoglou and P. S. Portoghese, ibid., 1976, 1427. 81 G. Zweifel and R. P. Fisher, Synthesis, 1974, 339. K. Utimoto, K. Uchida, and H. Nozaki, Tetrahedron Lett., 1973,4527. 83 E. R. Martinelli and A. B. Levy, Tetrahedron Lett., 1979, 2313. 84 1. Akimoto and A. Suzuki, Synthesis, 1979, 146. 220 Pelter --2 R~CHOIR' I R~B-CH \ p 6-CH I R2 indole moiety is particularly n0teworthy.~5 Aromatic groups can be coupled with conservation of the organo-groups by use of readily available crystalline ethanolamine complexes derived from diarylalko~yboranes.g~+~~ mixedThe diary1 compounds are available by stepwise displacement of alkoxy-groups from trial koxy boranes.2-Bromo-6-lithiopyridine reacts with trialkylboranes to yield borates that undergo extremely facile ring-cleavage, ( Figure 44),x8a reaction similar to those of alkenylborates bearing a leaving group in the u-position.xY Somewhat sur- prisingly, a-lithiofuran can behave similarly.g0 D. a-Thio-0rganoborates.-It is possible to produce a stable organoborate that contains a functional group capable of being transformed into a good leaving group, so inducing migration. This very general principle is exemplified in a new homologation procedureY1 (Figure 45), which uses t hiomet hoxymet hyl-l i thi um to form a stable organoborate.Methylation of the sulphur in mild conditions produces an ylide that undergoes migration to give a homologous organoborane, which can be manipulated as required. Use of disiamyl- and 9-BBN groups was successful in conserving groups and novel homologations of aryl and alkenyl (retention of configuration) groups may be carried out in good yields. H6 A. B. Levy, Tvtruhrrlron Loft., 1979, 402 I. R6 G. M. Davies, P. S. Davies, W. E. Paget, and H. M. Wardleworth, Tetruhedron L~ft., 1976, 795. Hi A. Pelter. G. M. Davies, and H. Williamson, Unpublished experiments. RX K. Utimoto. N. Sakai, M. Obayashi, and H. Nozaki, TcJtruhrdron,1976, 32, 769. 8y G. Zweifel and H. Arzoumanian, J. Am. Chrm. Sot.., 1967, 89, 5086.8o A. Suzuki, N. Miyaura, and M. Itoh. Tetruhedron, 1971. 27, 2775. B1 E. Negishi. T. Yoshida, A. Silveira, Jr., and B. L. Chiou, J. Org. Chcm.. 1975, 40, 814. 22 I Carbon- Carbon Bond Format ion Involving Bar on Reagents X = 0,NMe QN[cH2coNHzQ)-fxMeI Et I Et R’ R’ Figure 43 Pelter H CN -'R VCN['WH]-[.4--...]RBOHR2 -B\ OH Figure 44 Bun H Y" /H -Ll -BU"CECH c=c H/\ H/BSia2 !Hz BSia 2 Figure 45 Carbon-Carbon Band Formation Involving Boron Reagents 6 Reactions of Boron Stabilized Carbanions Overlap between the empty orbital on boron and an adjacent non-bonding filled orbital should stabilize a carbanionic centre a to b0ron.9~ In practice it is difficult to produce such carbanions as the base used for deprotonation will normally complex directly with the boron to yield an organoborate (Figure 46).To overcome this problem a very general approach is to use a heavily hindered non-nucleophilic base and in this way B-methyl-9BBN has been introduced as a methylenation reagent in the boron equivalent to the Wittig reaction, (Figure 46) Y MY IXzBCHR, -X2g-CHR2 M+ kv+ HVX2B-cR2 M+ LTM Pa)B-CH, -1. LTM P 2. R2XI Mes 2BCH3 LDCA w MeszBCH, Mes2BCH2-Oct" m HOCH2-0ct" Figure 46 s2 A. Pross, D. J. DeFrees, B. A. Levi, S. K. Pollack, L. Radom, and W. J. Hehre, J. Org.Chem., 1981,46, 1693. Pelter using N-lithio-2,2,6,6-tetramethylpiperidine (LTMP) as base.93 Similarly, a methylene group bearing two dioxaboryl groups, which do not readily complex with bases, yields a stabilized carbanion.94 An alternative, and equally general approach is to use a very hindered borane so that a simple base may be used. Alkyldimesitylboranes may be handled with great ease as, owing to the extreme hindrance of the environment around boron, it is difficult for complexes to be formed.In this case carbanions a to boron are readily produced at room temperature and behave as expgcted. Thus B-methyldimesitylborane is an excellent homologating reagent whose other reactions are actively under investigation in our laborat0ries.~5 A quite different method for the production of carbanions a to boron is the cleavage of l,l-di96397 or I ,I ,l-tri-boryl97 compounds with base (Figure 47).The metal exchange reaction occurs with great ease and the carbanions may be used in the usual ways. One notable reactiong7 sequence uses trisethylenedioxy- ' borylmethane to accomplish the homologation of ketones R1COR2 to the aldehydes R1R2CH.CHO. Li BuYi I R'C ECH + 2CxzBH -R'CH~CH(BCX~)~-R'CHZCH.BCx2 R 'C H 0 R'CH2CH=C H R (20-50'10 ).L Figure 47 33 M. Rathke and R. Kow, J. Am. Chem. SOC.,1972,94, 6854. 94 D. S. Matteson and R. J. Moody, J. Am. Chem. SOC.,1977, 99, 3196. 95 J. Wilson, J. Organomet. Chem., 1980, 186, 297; A. Pelter, L. Williams, and J. Wilson. Unpublished experiments. g6 G. Cainelli, G. Dal Bello, and G. Zubiani, Tetrahedron Lett., 1965, 3429; 1966, 4315. G. Zweifel et al.. J. Am.Chem. SOC.,1967, 89, 291. Synthesis, 1973, 37. 91 D. S. Matteson et al., J. Am. Chem. SOC.,1975, 97, 5608; J. Organomet. Chem., 1975, 93, 21 ;Synthesis, 1975, 147.

ISSN:0306-0012    

DOI:10.1039/CS9821100191

出版商:RSC    

年代:1982

数据来源: RSC