384 Acta Chim. Slov. 2005, 52, 384–390 Minireview Molecular Rearrangements During Terpene Sj^ntheses1 Ajoy K. Banerjee,"* William J. Vera," Henry Mora," Manuel Laya," Liadis Bedoya," Carlos Melean," PO. S. Poon NG," and EMa \. Cabrera* " IVIC, Centro de Química, Apartado 21827, Caracas 1020-A, Venezuela. E-mail: abanerje@ivic.ve b Universidad del Zulia, Departamento de Química, Facultad Experimental de Ciencias, Maracaibo, Venezuela Received 12-08-2005 t To Mr. Victor Marcos Simon, a beloved gardner of 90 years old. Abstract Rearrangements of several organic compounds with acids, iodine, boron trifluoride etherate, thionyl chloride-pyridine, manganese(III) acetate and silicon dioxide have been discussed. These rearrangements were observed during our studies on the synthesis of natural products related to diterpenes and triterpenes. The mechanism of these molecular rearrangements has been suggested in order to explain the formation of the obtained structures. Key words: rearrangement, terpenes, natural products Introduction Molecular rearrangements are frequently associated with undesired skeletal transformations leading to totally unexpected products. These rearrangements have provided many of the keys which have opened the doors to our understanding of organic chemistry. As the title implies we have not tried to include aH recorded rearrangements of terpenoid chemist^.1 In relation with our studies on terpenoid compounds we have observed the molecular transformation of several organic compounds (ketones, dienones, alcohols) leading to the formation of unexpected products. This information which remained scattered in several journals is brought together now in the form of a short review article. This review also includes a brief description of similar transformations observed by other authors. The majority of the rearrangements reported are somewhat related to the well-known Grob-type fragmentation particularly those in the first part. Rearrangement of bicyclic diones with iodine and methanol Several unsaturated ketones and esters undergo aromatization with iodine. Kotnis2 reported the aromatization of a variety of Hagemann's esters 1-2 to the corresponding p-methoxybenzoate derivatives ?>-A (Scheme 1). Tamura and Yoshimoto3 have observed the aromatization of cyclohexenone with iodine and methanol at reflux. OMe Me COOEt 1 R=H 2R=Me Me COOEt 1 R=H (90%) 2 R=Me (87%) Scheme 1 These interesting observations gave us the impetus to study4 the rearrangement of the commercially available dione 5 expecting its transformation to tetralone 6 which may prove to be a potential intermediate for the synthesis of the naphthoic acid 75 (Scheme 2). Mel MeO' Me 0 Scheme 2 MeO' Me COOH 7 We observed that dione 5 on heating with iodine and methanol afforded a product (70%) which was assigned to structure 8 on the basis of spectral data (Scheme 3). The diones 9, 10 and 11 under similar conditions afforded anisole derivatives 12 (65%), 13 (70%) and 14 (68%) respectively (Scheme 3). o R R o 5 6 Banerjee et al. Molecular Rearrangements During Terpene Syntheses Acta Chim. Slov. 2005, 52, 384–390 385 Mell Me 5+ .1-1 O' Mell (a) i-r .-o- (b) Scheme 4 © MeO OMe MeTo-l (c) Me|| Me OBz Me OBz Me Me 15 HO' Me 16 Scheme 5 Me Me OH Me Me 17 Me Me OH Me Me 18 Rt Mell R2 5 R1=R2=H 9 R1=H,R2=Me 10 R1=R2=Me 11R1=Me,R2=H Rt MeO Me COOMe R2 8 R1=R2=H 12R1=H,R2=Me 13R1=R2=Me 14R1=Me,R2=H Scheme 3 and has been applied to the conversion of steroidal dienones to aromatic steroids.8 This finding encouraged us to study the transformation of dienone 15 to phenol 16 which can be of considerable utility in synthesis of rishitinol9 17 and occidol10 18 (Scheme 5). Dienone 15 was synthesizedn from dione 9 but on heating with hydrochloric acid in methanol it did not yield the expected phenol 16. Aromatization followed by ring cleavage provided ketal 19 whose formation has been explained in Scheme 6. The above mentioned observations may lead one to think that hydroiodic acid produced by the reaction of iodine caused aromatization and fragmentation. This assumption proved incorrect on finding that diones on heating with hydroiodic acid and methanol were recovered unchanged. On heating with iodine and methanol diones 5, 9, 10, 11 did not generate practically any acid (litmus paper test). A reasonable mechanism of the transformation of dione 5 to anisole derivative 8 is depicted in Scheme 4. Probably the iodine activates the carbonyl groups of the dienone (a) formed by halogenation and dehydrohalogenation of the dione 5 yielding intermediate (b) and this by the nucleophilic attack of methanol suffered cleavage to the anisole derivative 8. To the best of our knowledge this is the first report of aromatization and fragmentation of the cyclic diones with iodine and methanol. Rearrangement of dienone with acids The acid catalyzed rearrangement of cyclohexadienone to phenol is a well known reaction6'7 Me OBz H+/MeOH <^ 15 -H20 ' Me°' Med ^ O-Bz Me <^ Me\OBz Me H+ p A. MeO'© Me MeO OBz OMe Me MeO Me Me Me 19 OMe OMe Scheme 6 Heating the dienone 15 with acetic anhydride and p-toluene-sulphonic acid yielded compound 20 by a mechanism illustrated in Scheme 7. 8 o o 5 o o 9 Banerjee et al. Molecular Rearrangements During Terpene Syntheses 386 Acta Chim. Slov. 2005, 52, 384–390 Ac2o 15 ------r^- ^ AcO OBz OBz Me |J AcO Me Me -QAc, ^^Me OBz AcO H OAc Me 20 Scheme 7 Alkaline hydrolysis of ester 20 did not give the expected aldehyde 21 but afforded a dimeric product 22 instead. We assume that aldehyde 21 formed during alkaline hydrolysis yielded via aldol condensation the dimeric adduct 22 (Scheme 8). The above mentioned observations support the previous finding that the rearrangement of dienone depends upon the structural features,6 nature of the subtituent and the reaction condition.12 Me 20 AcO CHO Me 21 „Me Mex AcO OH Me CHO Me 22 Scheme 8 Rearrangement of alcohols with thionyl chloride Thionyl chloride, a valuable synthetic reagent in preparative organic chemistry has been frequently utilized for the dehydration13 of tertiary alcohols for the preparation of alkenes but does not always produce14 the desired result. Thionyl chloride-pyridine mediated rearrangement of tertiary alcohols 23 and 25 to olefins 24 and 26 respectively has been reported (Scheme 9).1516 Alcohol 27 upon treatment with thionyl chloride and pyridine yielded a mixture of tetrahydronaphthalene 29 and diene 30 (Scheme 10). The formation of hexahydronaphthalene 30 which occurred through the intermediate 27i (Ha) was expected on the basis of published result.16 The origin of tetrahydronaphthalene 29 involves dehydration of alcohol 27 to diene 28 via the intermediate 27i (Hb), followed by aromatization. OH Mej,Me Me Me Me'MeH 23 HO Me Me' Me 24 Me Me M6MeH 25 Me' Me 26 Scheme 9 Me OH HaN MeMe-OSOCI Hb Me" Me 27 Me' Me 27i Me. Me Ha; Hb~> Me' Me 30 Me Mej Me Me Me Me 28 Scheme 10 Me Me 29 It is only too obvious that 29 would be far more stable than 28 by virtue of aromaticity. This transformation can also be explained on the basis of computational calculation.17 Octahydroindenol 32,16 prepared from ketone 31 by Grignard reaction afforded 33 (Scheme 11) with thionyl chloride and pyridine. This appeared strange from past experience1516 but the phenomenon may be explained by analyzing the orientations of the various bonds involved in the rearrangement. The trans ring juncture of alcohols 23 and 25 permits an antiperiplanar rearrangement of the migrating methyl group and the hydrogen at a bridgehead position, thus fulfilling the stereoelectronic requirement for a more or less synchronous elimination rearrangement reaction. As a result of the cis juncture a similar rearrangement cannot be expected in alcohol 32. Banerjee et al. Molecular Rearrangements During Terpene Syntheses Acta Chim. Slov. 2005, 52, 384–390 387 Me// Me 31 MeUOH H 32 Scheme 11 Me Me Mer 33 Similarly alcohols 44 and 45 prepared from ketones 42 and 43 respectively experienced aromatization with thionyl chloride and pyridine to afford compounds 46 and 47 respectively.1819 The conversion of compounds 41, 46 and 47 to the potential intermediates for (+) occidol, (±) occidol 48 and (±) platphyllide 49 respectivefr/ has already been reported (Scheme 14).1819 Tertiary alcohol 35, prepared from ketone 34 by treatment with potassium cyanide and alcohol, afforded via intermediate (35i) the product 36 in major yield.17 (Scheme 12). The unsaturated nitrile 37 was obtained in inferior yield and the formation of the expected16 rearrangement product 38 could not be detected. Me I mVh Me'Me' 34 H H-C-H Me' Me 35 ,CN OSOCI J Me Me [35i] CH2 CN Mel Me'Me H 36 NC Me MeV 37 Scheme 12 Me' Me 38 It can be seen that the mentioned rearrangement opened an efficient route for the construction of the fundamental ring skeleton of cis-fused perhydroazulenes. During our studies on the synthesis of sesquiterpenes, the rearrangement of several cyclohexadienyl alcohols with thionyl chloride and pyridine has been observed. The alcohol 40, obtained from a-santonine 39, undenvent aromatization18 with thionyl chloride and pyridine and yielded hyposantonin 41 via the intermediates 40i and 40ii (Scheme 13). Mel O' ^T ^Y V-Me Me 6- O ,— 39:R=0 L- 40: R=H, OH .Me Me ciosa Me O r Me 40i Me 40ii -Me Me O 41 Scheme 13 OBz Me| Me OBz I I Me I rj 0- 00: -*- a; -L,Me Me T Me T Me LOH r— 42: R=0 L*- 44: R=H, OH 46 48 OBz Mel Me OBz 1 1 ClT -i r— 43: R=0 L-*- 45: R=H, OH ca 47 ^~o 49 CH2 Scheme 14 In addition of our above mentioned work, many other interesting rearrangements of tertiary alcohols with thionyl chloride have been reported in a spate of papers. Some selected examples are discussed below. The rearrangement of tertiary alcohols presents in a heterocyclic system has been reported by Jackson.20 On reaction with hot thionyl chloride the alcohols 50 and 51 undenvent cyclization yielding the benzothiophenes 52 and 53 but the tertiary alcohol 54 produced a dimeric product 55 (Scheme 15) vvith thionyl chloride at room temperature. The above mentioned rearrangements are very interesting and permit a new route for the synthesis of various benzothiophenes. V s OH -(-Me R 50: R= i-propyl 51: R= i-butyl 54: R= Me S R 50j: R= i-propyl 51j:R=i-butyl 54j: R= Me ,CI ^%/^S Me 55 Scheme 15 O Banerjee et al. Molecular Rearrangements During Terpene Syntheses 388 Acta Chim. Slov. 2005, 52, 384–390 The transformation of alcohols 50, 51, 54 to compounds 52, 53 and 55 respectively can probably be explained by assuming the formation of intermediates 50i, Sli, 54i. Dimerization of 54i (R = Me) occurred at room temperature to yield the compound 55. Such dimerization was not possible with intermediates 50i and Sli (R=i-propyl, i-butyl) probably due to steric repulsion of the bulky groups. Therefore the compounds 50 and 51 did not undergo dimerization. Sesquiterpene cartol 56 with thionyl chloride and pyridine suffered ring contraction21, and produced acoradienes 57 and 58 along with other products. The acetylenic carbinols 59 and 60 on similar treatment afforded rearragements products 61 and 62, respectively (Scheme 16).22 Me Me^ ^Me CH2 Me Me Me Me 56 HO Ph-C-CsC-Ph Ph' 59 HO Ph-C-C=C-H Ph' 60 57 Me 58 Ph Cl Ph PhH Ph 61 Ph Ph"" ^C-CHCI PfK ^C-CHCI Ph 62 Scheme 16 Rearrangement of carbonyl compounds with boron trifluoride etherate The chemical literature records several interesting rearrangements of organic compounds with boron trifluoride etherate. A few examples are cited below. The conversion of isophorone oxide 63 to 64 was observed23 (Scheme 17) when treated with boron trifluoride etherate on inert solvent like benzene or dichloromethane. 2,0 O^^TjVle 3 4 Me' Me 63 OHC. Me ^ Me Me 64 Scheme 17 The coordination between boron trifluoride etherate and the oxygen of the epoxide leads isophorone oxide 63 to attain a transition state geometry24 which facilitates the 1,2 carbonyl migration to give compound 64. Upon treatment with boron trifluoride etherate in acetic anhydride ether 65 exhibited skeletal rearagement25 giving alcohol 66 which has turned out to be a potential intermediate in the synthesis of the sesquiterpene sesquifenchene 67 (Scheme 18). Me. Me -Me Me OAc OH 65 66 H2C; MeY-\^-wMe -/'/~Ky Me 67 Scheme 18 On treatment with boron trifluoride etherate hydroquinone diester 68 manifested Fries rearrangement26 and yielded 69 (Scheme 19). OAc OH O S-^Me OAc OAc 68 69 Scheme 19 Upon treatment with boron trifluoride etherate in dichloromethane solution at room temperature a variety of ketoxime ethyl carbonates undergo Beckmann rearrangement27 in excellent yields (generally 75-99%). On treatment with boron trifluoride etherate the ketoxime ethyl carbonates 70-72 yielded amides 73-75 (Scheme 20). N-o-co2Et R^R2 70: Rl=R2=Ph 71:R1=Ph,R2=Me 72: R1=R2=B R1-NH-CO-R2 73: Rl=R2=Ph 74: R1=Ph, R2=Me 75: R1=R2=B Scheme 20 In contrast to a literature report28 treatment of tetralones 76 and 77 with boron trifluoride etherate and acetic anhydride at 0 °C unexpectedly produced29 compounds 78 and 79 respectively instead of acetates 80 and 81 (Scheme 21). o Banerjee et al. Molecular Rearrangements During Terpene Syntheses Acta Chim. Slov. 2005, 52, 384–390 389 OMe O OMe O OMe OCOMe OCOMe OMe COMe 80 76 78 O MeMe Mejj H 90 OH Me I ,OH SiQ2 MeMe' 91 OCOMe OMe MeO^ ^\ /L MeO^ ^\ J\ MeO^ O O 81 77 Scheme 21 OCOMe 79 CH2 Me'1 H Mer 91 i Me Me Me Me 92 Scheme 23 Rearrangement by silica gel Many organic compounds undergo diverse molecular rearrangements induced by silica gel. A great variety of interesting rearrangements effected by silica gel have been reviewed by McKillop and Young.30 Therefore for space limitations their discussions have been omitted in this review. We shall confine ourselves to a few cases only. The addition product 84 of buta-l,3-dien-l-yl acetate 82 and chlorobenzoquinone 83 through a column of silica gel led to rearrangement and concomitant aromatization31 affording to naphthoquinone 85 (Scheme 22). Similarly the addition product 88, of diene 86 and chlorobenzoquinone 87 in a column of silica was transformed into naphthoquinone 89 (Scheme 22). AcO Cl- R2 R3 82: R1=H 83: R2=CI, R3=H 86: Rl=Me 87: R2=H, R3=CI R1' AcO O R2 R3 R1- R2 R3 O 84: R1=R3=H, R2=CI 88: R1=Me, R2=H, R3=CI 85: R1=R3=H, R2=CI 89: R1=Me, R2=H, R3=CI Scheme 22 An attempt to purify the diol 91, obtained32 from ketone 90 over silica gel afforded tetralin 92 in high yield (Scheme 23). It can be assumed that the dehydration of diol formed the intermediate 91i which undenvent aromatization to afford tetralin 92. Rearrangement with Manganese(III) acetate Oxidation of cyclic olefins with manganese(III) acetate in acetic acid in the presence of a metal halide catalyst (potassium bromide) yields allylic acetates.33'34 This method proved unsatisfactory with the compound35 93 which yielded tetralin 94 (60%) and compound 95 (40%) instead of expected allylic acetate 96, which can be of considerable utility in the synthesis10 of occidol 18 (Scheme 24). Me Me 93 Me Me Me OAc i ^T ^1 A^T ^l ^ \kJ Me Me 94 95 ,Br Me OAc Me 96 Me Me 18 Me OH Scheme 24 Concluding remarks The above mentioned molecular rearrangements have provided some interesting results. Though the desired products were not obtained for their further transformations to terpenoid compounds, the valuable information obtained would definitely constitute an important contribution to organic chemistry. O O O O Banerjee et al. Molecular Rearrangements During Terpene Syntheses 390 Acta Chim. Slov. 2005, 52, 384–390 Acknowledgements The senior author expresses heart felt thanks to the referees for valuable suggestions which helped to improve the standard of the review. References 1. J. F. King, P. de Mayo, “Terpenoid Rearrangements in Molecular Rearrangements”, (ed. de Mayo, P.), Interscience Publishers, New York, 1964, 771-840. 2. A. S. Kotnis, Tetrahedron Lett. 1990, 31, 481-484. 3. Y. Tamura, Y. Yoshimoto, Chem. Ind. 1980, 888-889. 4. A. K. Banerjee, E. V. Cabrera, J. A. Azocar, Synth. Commun. 2000, 30, 3815-3821. 5. A. G. Myers, V Subramanian, M. Hammond, Tetrahedron Lett. 1996, 37, 587-590. 6. R. B. Woodward, T. Singh, /. Am. Chem. Soc. 1950, 72, 494—500. 7. J. Berson, “The Dienone-Phenol Mysteries”, in “Chemical Creativity”, Wiley-VCH, Weienheim, Germany, 1999, 77-107. 8. R. Shopiro, in “Steroid Reactions”, (Edit. Djerassi,), Holden-Day, San Francisco, 1963, 371-402. 9. N. Katsui, A. Matsunaga, K. Imaizumi, T. Masamune, K. Tomiyama, Buli. Chem. Soc. Jpn. 1972, 45, 2871-2877. 10. T. -L. Ho, /. Chem. Soc, Perkin L 1973, 2579-2582 and related references. 11. A. K. Banerjee, H. Hurtado, Tetrahedron 1985, 41, 3029-3032. 12. B. P. J. Kropp,/. Am. Chem. Soc. 1965, 87, 3914-3921. 13. J. D. Roberts, W. G. Young, S. Winstein, /. Am. Chem. Soc. 1942, 64, 2157-2164. 14. J. W. ApSimon, K. Yamasaki, Chem. Lett. 1977, 1453-1456. 15. S. C. Welch, A. S. C. Prakasa Rao, /. Org. Chem. 1978, 43, 1957-1961. 16. A. K Banerjee, W. Vera, Rec. Trav. Chim., Pays-Bas 1995, 114, 87-90. 17. A. K. Banerjee, W. Vera,itec. Trav. Chim., Pays-Bas 1993, 112, 259-261. 18. A. K. Banerjee, W. Vera, Synth. Commun. 1996, 26, 4641-464. 19. A. K. Banerjee, J. A. Azocar, W. Vera, Synth. Commun. 1999, 29, 2995-3001. 20. J. L. Jackson,/. Heterocyclic Chem. 1991, 28, 109-112. 21. A. Lardan, T. Reichstern, Helv. Chim. Acta 1962, 45, 943-962. 22. P. D. Landor, S. R. Landor, /. Chem. Soc. 1963, 2707-2711. 23. F. Kunisch, K. Hobert, P. Welzel, Tetrahedron Lett. 1985, 26, 6039-6042. 24. R. D. Bach, R. C. Klix, Tetrahedron Lett. 1985, 26, 985-988. 25. Y. Bessiere-chreiten, C. Grisan, CR. Acad. Sci. Ser C. 1972, 275, 503-504. 26. J. Boyer, J. Krum, M. Myers, A. Fazal, C. Wigal, /. Org. Chem. 2000, 65, 4712-4714. 27. R. Anilkumar, S. Chandrasekhar, Tetrahedron Lett. 2000, 41, 5427-5429. 28. C. R. Narayanan, K. N. Iyer, /. Org. Chem. 1965, 30, 1734—1736. 29. A. K. Banerjee, L. Bedoya, W. J. Vera, C. Melean, H. Mora, M. S. Laya, M. Alonso, Synth. Commun. 2004, 34, 3399-3408 30. A. McKillop, D. W. Young, Synthesis, 1979, 401-422. 31. I. Boisvert, P. Brassard, Tetrahedron Lett. 1983, 24, 2453-2456. 32. A. K. Banerjee, V E. Khalil, W. Vera, Synth. Commun. 2000, 30, 4375-4385. 33. J. R. Gilmore, J. R. Mellor, /. Chem Soc. 1971, 2355-2357. 34. A. K. Banerjee, W. Vera, M. S. Laya, Russ. J. Org. Chem. 2002, 38, 855-859. 35. A. K Banerjee, W. Vera, M. S. Laya, Synth. Comm. 2004, 34, 2301-2308. Povzetek Opisane so molekulske premestitve nekaterih organskih spojin z reagenti, kot so kisline, jod, BF3.Et2O, tionil klorid-piridin, magnezijev(III) acetet in silicijev dioksid. Te premestitve so bile opažene pri sintezi naravnih produktov sorodnih diterpenom in triterpenom. Predlagani so mehanizmi reakcij. Banerjee et al. Molecular Rearrangements During Terpene Syntheses