Use this URL to cite or link to this record in EThOS:
Title: Synthesis and rearrangement of alicyclic systems
Author: McHale, Philip James
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 1973
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This thesis is concerned with intramolecular rearrangements of epoxides of bicyclic systems: in the presence of acids, epoxides will cleave to generate electron deficient sites which can promote rearrangements involving intramolecular nucleophiles such as methyl groups with their bonding electrons, "hydride" ions, parts of the carbon skeleton, and the π-electrons of olefinic double bonds. The latter interaction leads to ring closure, and this is an important process in the biogenesis of many polycyclic natural products. The first part of chapter one of this thesis describes the role of epoxides in the biosynthesis of polyisoprenoids, with particular reference to the cyclisation of squalene 2,3-epoxide and other long chain polyolefin epoxides. The second part of chapter one is concerned with Lewis acid catalysed rearrangements of epoxides of cyclic systems, particularly steroidal epoxides. Early workers [H.B. Henbest and T.I. Wrigley, J.Chem.Soc., 1957, 4596, 4765] emphasised the formation of ketones by 1,2-shifts, and when using boron trifluoride as the Lewis acid, they often obtained diaxial fluorohydrins formed by direct attack of a fluorine species on the epoxide; rigorous purification of the reagent to remove fluoride reduces the occurrence of such side-reactions to a minimum. Other more polar rearrangement products tended to remain unidentified, but more recent work [for recent reviews see D.N. Kirk in "Terpenoids and Steroids", ed. K.H. Overton, Specialist Periodical Report, The Chemical Society, London, Vol. 1, 1971, p.365; Vol. 2, 1972, p.306] has led to a reappraisal of earlier results, and these products have been largely identified; most are hydroxy-olefins formed from methyl and skeletal migrations. The rearrangements of steroidal epoxides give a wide variety of products, which depend on the environment of the oxiran ring and the positions and nature of other substituents in the molecule. The products obtained can be rationalised to a certain extent by considering nteric, electronic, and conformational factors; thus the direction of formation of ketones from tetrasubstitued epoxides can be predicted by the "rule of axial cleavage", which envisages the epoxide rearrangement as proceeding without the development of a full carboniurn ion, and retaining some internal SN2 character in the transition state. Migrations in which the oxiran oxygen becomes a hydroxy-group do appear to proceed by a full carbonium ion; thus in some cases [e.g. T.G. Halsall, Sir Ewart R.H. Jones, E.L. Tan, and G.H. Chaudhry, J.Chem.Soc. (C), 1966, 1374], groups have migrated from a position originally cis to the C-O bond which cleaves, which must require total C-O cleavage before migration can occur. Kinetic isotope studies [H.W. Whitlock and A.H. Olson, J.Amer.Chem.Soc., 1970, 92, 5383] suggest that these rearrangements proceed by a rate determining formation of a set of rapidly interconverting carbonium ions, which collapse to the observed products, even when the migrating group is trans and antiparallel with the C-O bond which breaks and concerted cleavage and rearrangement might be possible. The aim of the present work has been to synthesise a series of simple epoxides and to rearrange them with a Lewis acid with a view to correlating the course of the rearrangement with the nature of the functional groups present in the molecule. Epoxides of bicyclic systems were chosen as they would contain several potential migrating groups, without the possibility of back-bone rearrangements occurring, since these often complicate steroidal epoxide rearrangements; it was also thought that their rearrangements might yield novel and synthetically useful skeletons. A series of indan-epoxides (i)-(vi), and the trimethyldecalin-epoxides (vii) were synthesised and rearranged with boron trifluoride-ether complex in anhydrous benzene. The tetramethyldecalin epoxides (viii) and (ix) were also prepared from a readily available starting material to see whether their rearrangement would give the products (x) or (xi) which would be very useful ring (A + B) precursors in tri- and sesquiterpene synthesis. The syntheses of the model epoxides are outlined in Scheme 1. The indane systems were prepared by carrying out Robinson annelation reactions on 2-ethoxycarbonyl- and 2-methyl-cyclopentanone, and gem-dimethylation was effected by treatment with potassium t-butoxide in t-butanol, followed by methyl iodide. The use of potassium t-amyloxide under more vigorous conditions gave a greater yield of the trimethylated product (xxiv). Reduction of keto-ester (xvi) with lithium aluminium hydride gave the diol (xxv), which was also obtained when an excess of sodium borohydride was used, along with the hydroxy-ester (xxvi) and the lactone (xxvii). The hydroxy-ester (xxvi) could be obtained as the only product by using a stoicheiometric amount of sodium borohydride at 0°. Reduction of the methyl-indanone (xxii) with lithium aluminium hydride gave a mixture of the epimeric 5-alcohols (xxviii), which could not be separated. The octalone (xxx) was prepared by saponification and Jones oxidation of the benzoate of octalol (xxix), which was available in this laboratory. The tetramethyloctalol (xxxiii) was prepared by the method of Ireland at al. [G. Saucy, R.E. Ireland, J. Bordner, and R.E. Dickerson, J.Org.Chem., 1971, 36, 1195] by treating luciferin aldehyde (xxxii) with phosphoric acid. The octalone (xxxiv) was obtained by Jones oxidation of octalol (xxxiii). Epoxidation was effected by treatment of the appropriate olefin with either m-chloro- or p-nitro-perbenzoic acid, and usually gave a mixture of the α- and β-isomers, which were separated by p.l.c.; in one case (iii) separation was not possible, and in some cases, only one epoxide was formed. The stereochemistry of most of the epoxides could be tentatively assigned from the expected mode of attack of the peracid. Formation of one of the epoxides of keto-ester olefin (xvi) was accompanied by loss of equivalence of the ester methylene signals in the n.m.r. spectrum; the methylene signal in the other epoxide appeared as a simple quartet. The stereochemistry of the lactone epoxide (vi) could not be assigned with certainty. The octalol (xxxiii) gave a single epoxide, which was probably the β-isomer, while the octalone (xxxiv) gave two, and the former epoxide was correlated with the more abundant of the latter epoxides by oxidation with chromium trioxide in pyridine. The hydroxy-ester epoxides (ii) were unstable and converted to the lactone epoxides (vi) on standing. The decalone epoxides (viia) and (viib) were of known stereochemistry, having been previously prepared and characterised. The results of the boron trifluoride catalysed rearrangements are shown in Scheme II. A wide variety of products v/as obtained, and some generalisations can be made: compared v/ith the rearrangements of steroidal epoxides, rearrangements of bicyclic epoxides give more ring fragmentation reactions, and carbonyl forming reactions are not so common. Most of the rearrangements which were terminated by dissipation of the positive charge at a site away from the oxiran ring and resulted in the oxiran oxygen becoming a hydroxy-group could be rationalised in terms of a mechanism proceeding through a fully developed carbonium ion; an analysis of conformational, electronic, and stereochemical factors indicates which groups are likely to migrate to the electron deficient site. Thus suitable migrating groups must lie parallel or nearly parallel to the axis of the empty p-orbital at the carbonium ion site; their migration should not develop positive charge at an unfavourable position e.g. α to an electron-withdrawing group; and their migration path should not be sterically impeded or lead to a sterically hindered structure. In the absence of other competing reactions, the relative epoxide geometry need not determine absolutely the course of the rearrangement; thus the decalone α- and β-epoxides (viia) and (viib) both gave products from the migration of the some suitably aliened groups, although each epoxide did give one characteristic rearrangement not given by the other. The directing effects of functional groups on these rearrangements are amenable to analysis: electron withdrawing groups e.g. -CO2Et tend to reduce reactivity and to retard development of a positive charge at an adjacent site; carbonyl groups have the same effect, but may become involved in the rearrangement by giving acyl migrations; hydroxy-groups do not have such a marked effect on the initial migrations which occur, but they can terminate the rearrangement by giving a fragmentation leading to an aldehyde (iii), or by "trapping" a suitable carbonium ion to give a cyclic ether (both of these reactions can also be given by the oxiran oxygen). In one case, this ring closure reaction gave a novel heterocyclic "propellane" (xl). Rearrangements in which the oxiran oxygen became involved in a carbonyl group, with no migration of positive charge away from the oxiran site were not so amenable to analysis.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available