The catalytic hydrogenation of nitrobenzene is an industrially important reaction used in the commercial production of aniline for use in the polyurethane industry. A mechanism for the reaction was first proposed by Haber in 1898 and has been widely accepted despite never being fully delineated. During this study the nitrobenzene hydrogenation reaction was investigated, over a range of metal catalysts, to probe the mechanism of hydrogenation and catalyst deactivation. Initial investigations over Pd/C catalysts revealed the reaction to be sensitive to the solvent and the nature of the carbon support. However more importantly it was shown that the first intermediate in Haber’s scheme, nitrosobenzene, could not act as an intermediate to nitrobenzene hydrogenation. As a result, a new reaction mechanism was proposed where the hydrogenation of nitrobenzene and nitrosobenzene proceed via separate mechanistic routes, linked by a common adsorbed intermediate; the surface concentration of this adsorbed species controls the hydrogenation pathway followed. Further investigation over Raney nickel suggests this mechanism to be valid over other metals and not specific to palladium. A series of novel bimetallic catalysts were also prepared for use in this study. Characterisation of these catalysts was carried out to determine the nature of the metal-metal interaction on the surface. the evidence suggests mixed metal particles may have been formed on some catalysts. The activity of these catalysts was found to be greatly enhanced following pre-treatment with water vapour in a hydrogen atmosphere. It was postulated that partial oxidation of the metal active sites was occurring and that these systems were more active due to the enhanced adsorption of nitrobenzene. The copper nickel/systems were found to show enhanced catalytic activity, whereas all systems containing cobalt displayed irreversible deactivation following water treatment, which was attributed to the formation of irreducible cobalt aluminium spinel from the CoO formed on the surface.

A one-pot, two step tandem process involving an Overman rearrangement and a ring closing metathesis reaction has been utilised for the efficient synthesis of various cyclic allylic trichloroacetamides from simple allylic alcohols. Various methods were then investigated for the allylic oxidation of a carbocyclic amide using TBHP along with different transition metals such as Pd, Se, Mn and Cr. This was required for the synthesis of the important building blocks for the construction of structurally diverse antiviral and anticancer carbocyclic nucleosides and natural products. The oxidation of (1S)-N-(cyclohexenyl)trichloroacetamide was then studied leading to the preparation of two diol analogues in excellent stereoselectivity. The cyclohexene derivative was also stereoselectively functionalised using Upjohn dihydroxylation conditions or by a directed epoxidation/hydrolysis sequence of reactions to generate two aminocyclitols, the enantiomer of dihydroconduramine C-1 and dihydroconduramine E-1 in excellent stereoselectivity. In addition to this, a one-pot tandem process involving a substrate-directed Overman rearrangement and ring closing metathesis reaction was developed for the stereoselective synthesis of a functionalised carbocyclic allylic trichloroacetamide. The functionalised carbocyclic amide was employed in the successful synthesis of a syn-(4aS,10bS)- phenanthridone framework using a Pd-catalysed cross-coupling reaction. Stereoselective epoxidation and dihydroxylation of the syn-(4aS,10bS)-phenanthridone was then investigated leading to the preparation of new analogues of 7-deoxypancratistatin. Attempts were also made to use the functionalised carbocyclic amide in the total synthesis of the Amaryllidaceae alkaloid (+)-γ-lycorane. Further studies were then investigated to expand the scope of the one-pot tandem process to include heterocyclic derived substrates. This led to a seven-membered carbocyclic amide, which has been modified to create a diastereomeric core of balanol.

Herein, we report the results of research continuing previous successI in the field of enantioselective organocatalytic reduction of imines with trichlorosilane. Syntheses of various precursors (ketones) and substrates (imines) for the reduction reaction and their reduction following the protocol (Scheme A1) are described in this thesis. 1. Aromatic heterocycles containing nitrogen – good yields of the reduced product (68-85 %), the enantioselectivity depended on steric bulk in proximity to the nitrogen, steric bulk improved the enantioselectivity (up to 78 % ee), probably due to thwarting the coordination of the nitrogen to HSiCl3. 2. Aromatic heterocycles containing sulfur – sulfur in the ring was tolerated well (89 % ee). 3. Aromatic heterocycles containing oxygen – generally good yields (62-90 %), dependence on position isomer was observed: furan-2-yl-derived substrates were reduced in moderate enantioselectivity (45-85 % ee), possibly due to the problem of coordination; in contrast, furan-3-yl derivatives were reduced in good enantioselectivity (77-91 % ee). 4. Non-heterocyclic aromatic or aliphatic – good yields (62-98 %) but varied enantiomeric excess (10-97 %). The high enantioselectivity values (76-97 % ee) were for substrates with significant contrast of the steric hindrance of the groups next to the reaction centre. Furthermore, an example of practical utilisation of the method is presented. Naturally occurring alkaloid N-acetylcolchinol was synthesised in 9 steps and overall 8 % yield (Scheme A2). The stereogenic centre was introduced using our method and afforded the desired enantioenriched amine in 96 % ee.

Over the past thirty years, numerous fused polycyclic ether natural products have been isolated from small marine organisms. These compounds revealed a variety of interesting biological properties, which has attracted the interest of many synthetic groups. The common structural characteristic of these compounds is an array of trans-fused ether rings. Several iterative strategies have been reported to build six- and seven-membered cyclic ethers, which are the two most common units in these natural products. The objective of work described in the first part of this thesis was to develop new synthetic methodology to access these motifs. Each unit must be built in the minimum number of steps and the new methodology must be flexible enough to obtain both six- and seven-membered rings from a common precursor. They key reaction of the strategy will be ring-closing metathesis, as previous work in our group showed that this is a powerful reaction to create cyclic ethers. The second part of this thesis focus on the total synthesis of one of the marine polyether natural products, gambieric acid A. This compound was isolated in 1992 and has so far eluded total synthesis. First, the synthesis of the tricyclic B-D core, which has been developed in our group, was optimised and performed on a large scale. The synthetic route relies on the ring-closing metathesis reaction to construct two of the cyclic ethers from a commercially available glucose derivative. Several strategies for the coupling with the tetrahydrofuran A ring were then investigated. The initial method envisioned for the fragment coupling was a nucleophilic substitution of an alkyl iodide by a lithiated dithiane. However, this strategy revealed unsuccessful. Instead, a more converging approach using a diastereoselective Tsuji reaction was developed, which allowed the formation of the complete carbon skeleton of the A-D fragment.

This dissertation is divided into five chapters. Chapter One consists of an introduction to radical cyclisation and rearrangement reactions. Chapter Two investigates the reactions of substituted arylsulfonamides 278a-l with copper bromide and an amine ligand-TPA. This reaction involves an alkyl radical generated from the copper (I) bromide/TPA complex, which can then undergo a 1,5- ipso attack onto the sulfonamide leading to a cyclohexadienyl radical intermediate. Re-aromatisation and extrusion of sulfur dioxide leads to an amidyl radical intermediate. This can undergo either cyclisation back onto the aromatic ring to give the 6-substituted oxindole 336, or reduction from H-atom abstraction by the solvent leading to rearranged amides. A minor product identified as a 5-substituted oxindole 333 may be formed from direct radical cyclisation onto the sulfonamide followed by extrusion of sulfur dioxide. An unambiguous synthesis of 333 was obtained through the Stollé method in order to rigorously identify this product. For completion, the rearranged amide 280e was also unambiguously synthesised from known literature sources. It has been shown that the selectivity towards either rearrangement or cyclisation is dependent upon the solvent used and temperature. For example, toluene induces excellent selectivity towards cyclisation (to furnish oxindoles), while using dichloromethane (DCM) induces a greater selectivity towards rearranged amides. Chapter Three explores the effects of varying the alkyl chain length on the nitrogen atom on selectivity, while keeping both the aryl group and initiator the same. It has been shown that selectivity towards the rearrangement (or decrease in cyclisation) occurred when the alkyl chain was increased from N-butyl to N-dodecyl. In addition a similar solvent effect on selectivity was observed as discussed in Chapter 3, notably relatively more rearranged amide was produced with DCM and oxindoles with toluene. Chapter four involves investigating the copper-mediated radical cyclisation of haloamides to give oxindoles directly. The final chapter consists of the experimental.

Chapter One provides a review of the literature relating to the isolation, structure elucidation, biological activity, and mode of action of the azinomycins. It also provides an overview of the synthesis and biological activity of simplified synthetic analogues. Chemical strategies towards the natural products are presented, as are proposals for the biosynthetic construction of these complex metabolites. Chapter Two describes our synthetic efforts towards simplified analogue 90 of azinomycin B, containing an amide substituent at N5, using the N16–C17 amide bond disconnection. These studies involved the synthesis of α-azido-β-hydroxy ester 125 in eight steps and 5% overall yield from L-methyl serinate. Further progression of 125 to fully elaborate the right-hand domain could not be achieved. Chapter Three explores a new approach to the azinomycins, based around the formation of the C7−N16 enamide bond using copper mediated cross-coupling methodology. Amides related to the left-hand domain of the azinomycins, 140 and 142, were demonstrated to undergo copper-mediated cross-couplings with phenyl iodide in yields of 19% and 64% respectively, demonstrating the potential of this approach. Methods for the synthesis of suitable coupling partners were explored. The most efficient route utilised the Eschenmoser sulphide contraction to elaborate Rpyroglutamic acid to bromoalkylidenepyrrolidine 184 in 7 steps and 16% overall yield. Initial studies into the amidation of 184 suggest that protection of the pyrrolidine nitrogen will be required to facilitate cross-linking. Chapter Four details the synthesis and biological evaluation of a range of new bisepoxides based on the azinomycin “left-hand” domain, which possess rigid linkers. The most potent of these analogues was identified as bisepoxide 203 (GI50 = 0.33 μM). All the new analogues displayed reduced potency compared to those based around aliphatic linkers, suggesting that they cannot readily adopt viable conformations for ISC formation. Chapter Five contains detailed experimental procedures for the new compounds described within this thesis.

The spontaneous formation of supramolecular assemblies has been viewed as a potential route to the creation of functional nano-scale architectures for a number applications in electronics. In this thesis a number of assemblies formed from molecular constituents deposited from the solution phase have been studied. The structures formed by two carboxylic acid derivatives on the highly oriented pyrolytic graphite (HOPG) surface from nonanoic acid solutions are presented. Quaterphenyl-tetracarboxylic acid (QPTC) molecules are observed to form a supramolecular network where all the constituents lay parallel to one another on the surface. The network is stabilised by four carboxylic acid dimer bonds per molecule in addition to admolecule-substrate interactions. Terphenyl-tetracarboxylic acid (TPTC) molecules form a much more complex structure with individuals orientating themselves in one of three directions to form a network with hexagonal symmetry but no translational order. To characterise such an unusual supramolecular morphology we introduce a rhombus tiling representation of the network where each molecule is schematically replaced with a lozenge rhombus producing a tiling. Such tilings have been studied previously in the literature and utilising this we are able to determine that the morphology is stabilised by entropic contributions to the free energy. In addition to this we present the tip-induced manipulation of the TPTC supramolecular network. The manipulation is performed by imaging the structure within a specific voltage bias range resulting the TPTC molecules reordering into a close packed structure. Returning the voltage to that conventional used for imaging causes the network to relax back into the open structure although with a different morphology. We then discuss the changes induced in these supramolecular networks when additional molecular species are introduced to the system. First, coronene and perylene are separately codeposited with QPTC resulting in the formation of a hexagonally ordered network with coronene or perylene located at the vertices of six QPTC molecules. This new structure is observed to form even when QPTC is deposited first. Second, the adsorption of coronene into the porous TPTC network is presented. When the TPTC network forms before the introduction of coronene we note little effect on the network morphology. However, when the molecules are mixed in the solution phase and deposited simultaneously we observe the non-uniform adsorption of coronene into the TPTC structure. At higher coronene concentrations we note the network forms with a different morphology shifted towards a more ordered state suggesting that when the molecules are deposited sequentially the system is kinetically trapped in the originally formed structure. We then present a series of studies of molecular adsorption on the Au (111) surface. First, hexaazatrinaphthylene (HATNA) molecules are observed to form stable supramolecular structures when deposited from ethanol solutions. A core hydrogen bonding junction is identified. The network switches between two domain orientations and we identify a linear defect where the two domains meet. Second, we report the adsorption of Tri( 4-bromophenyl) benzene (TBPB) on the Au (111) surface. TBPB forms three different structures at room temperature. When samples are heated during the deposition stage we observe the covalent coupling of pairs of molecules to form dimers. This reaction is confirmed by ToFSIMS experiments. The substrate is confirmed to play a significant role in the coupling process as subsequent experiments on HOPG failed to yield dimer formation. Finally we demonstrate the potential of a UHV-prepared sample by templating the adsorption of adamantanethiols. Finally, we demonstrate the adsorption of a solubilised derivative of perylene tetracarboxylic dianhydride (PTCDA). PTCDA molecules have poor solubility in most solvents commonly used for liquid deposition. The addition of alkane chains attached to the sides of the perylene core promotes the solubility of the molecule in these solvents whilst leaving the anhydride functionality intact. Deposition is performed from 1-phenyloctance solutions on HOPG. The molecules form an ordered structure characterised by a single molecule unit cell. The results presented in this thesis show that the understanding of supramolecular networks has progressed to the point where changes in the morphology can be induced via a variety of processes.

No description available.

The first chapter focuses on the introduction of iodine mediated carbocyclisations and their applications continue to present a stimulating challenge in target- and diversity-oriented synthesis. The second chapter discusses applications and brief literature overview about classical approaches towards the syntheses of indene derivatives. Herein the syntheses of 3-iodo-1 H-indene derivatives via iodonium-promoted 5-endo-dig carbocyclisation of 2-substituted ethynylmalonates as a key starting material are described. Within this study, we were able to show that the 3-iodo-1 H-indene can be used as a synthetic platform not only for the palladium chemistry but also as a catalyst for the in situ generation of [text unavailable] hypervalent iodine reagent. Additionally, 3-iodo-1 H-indene derivatives have the potential to perform asymmetric synthesis. Third chapter demonstrates tandem iodine mediated carboannulation of the stilbene malonate derivatives via either 5-exo- or 6-endo-trig mode under basic reagents with subsequent lactonisation to structurally complex indanes and tetrahydronaphthalenes with three new stereogenic centres. In the present study, a unique stereochemistry was observed in the case of tetrahydroindenofuranones and confirmed by single crystal X-ray analysis. These cyclisations proceed exclusively with the retention of configuration to form tetrahydroindenofuranones. Both the compounds formed as a single diastereomers as judged from their 1H and 13C NMR spectrum. In fourth chapter the synthesis of novel simplified analogues of IBA by oxidation of [text unavailable] diiodoacrylic acids are described. Additionally, the ligand exchange resulted in tosylate derivative. The new reagents have been utilized in various well established oxidative transformations with superior or similar reactivity as conventional hypervalent iodine(III) reagents.

The effect of scale on the conversion to products in flash vacuum pyrolysis experiments was studied using four model reactions. Overall, conversion was dependent on rection scale up to 0.5 g, but above 0.5 g, the effect was minimal and within experimental error. This effect was shown to be due to variations in contact time of the molecules in the furnace tube. Altering the furnace tube diameter had the effect of increasing the conversion to product. The pyrolysis of ortho-anilinomethylene Meldrum’s acid derivatives was investigated. Typically the 8-substituted quinolin-4-one was obtained as the major product, with a few exceptions. Certain substituents (such as nitro) were found to react with the ketene produced in the reaction to give alternative products, while others (such as chloro- and N-unsubstituted amides) gave the 3-substituted quinolin-4- ones via ipso-cyclisation and migration of the substituent. The regioselectivity of the pyrolysis of meta-anilinomethylene Meldrum’s acid derivatives, to give 5- and 7-substituted quinolin-4-ones, was studied. In general a 3:1 – 4:1 ratio of regioisomers was obtained, in favour of the 7-substituted quinolin-4-one. Substituents capable of hydrogen bonding, such as hydroxy-, were shown to give the 5-substituted quinolin-4-one exclusively and DFT calculations were employed to show that, in these examples, the 5-substituted product was favoured energetically. The pyrolysis of the methylene Meldrum’s acid derivative of 3-aminophenol gave 8-hydroxyquinolizinone as the sole product at low temperatures, with 5- hydroxyquinolinone as the major product formed at higher temperatures. The mechanism involves a regioselective electrocyclisation, followed by a hydrogen transfer and generation of a new ketene. Cyclisation of this ketene gives the quinolizinone. The scope of this reaction was explored, with a number of derivatives synthesised, and substitution on the aminophenol and the ketene generator was tolerated. The reactivity of the quinolizinones was also explored. The hydroxygroup was found to be phenol-like and underwent similar reactions, such as alkylations and acetylations. The compound was found to be highly reactive towards electrophiles, reacting in the 1- and 3- positions of the ring system, often in both positions. The pyrolysis of the aminomethylene Meldrum’s acid derivatives of certain pyridazinones was shown to give pyridopyrazidinediones. In some examples, a 4 second product based on a pyrrolopyridazine ring system was observed and DFT calculations show that the mechanism involves probably an electrocyclisation, followed by a decarboxylation reaction. Pyrolysis of amino acid ester derivatives of methylene Meldrum’s acid were shown to give N-unsubstituted 3-hydroxypyrroles and 1H-pyrrol-3(2H)-ones. Different amino acids were tolerated in the reaction, as were different electronwithdrawing groups in place of the amino acid ester. DFT calculations were employed to explore the mechanism of the reaction. 3-Hydroxypyrrole was also synthesised from the pyrolysis of Meldrum’s acid derivative of glycine tert-butyl ester, and the reactivity of the compound explored for the first time. The compound was found to be reactive towards electrophiles, such as diazonium salts, and could be O-acetylated under appropriate conditions.