The diterpenoids are fairly widely distributed in nature. The best known members are the resin acids obtained from the non-volatile portion of many oleoresins, especially those obtained fram conifers. Diteppenoid acids, alcohols, lactones, phenols, and alkaloids have been obtained from a variety of sources. The hydrocarbons are more restricted in occurrence; all are found in New Zealand and all but two (camphorene and cupressene) in endemic species. Their reported occurrences are almost solely limited to countries bordering on the Pacific.

The chemistry of the amidoximes (a) has long been of interest because of the possibility of tautomerism with hyroxyamidines (b)

Whole document restricted, see Access Instructions file below for details of how to access the print copy. / Attempts have been made to transform lanosterol, by both chemical and microbial means, into compounds of potential pharmacological importance. In part I some compounds with heterocyclic rings (pyrazoles, isoxazoles, and furazans) fused to ring-A have been synthesized by chemical methods. These have been tested for cytotoxic properties against lymphoid leukemia L-1210. The equilibria and n.m.r. spectra of some formyl ketones have also been studied. In part II attempts have been made to oxidize some lanosterol derivatives with microbial cultures in pursuit of synthetically useful intermediates. 3β-Hydroxy-11-keto-4, 4, 14α-trimethyl-5α-chol-8-enic acid methyl ester (35e) has been synthesized from lanosterol and transformed by Trichotecium roseum to a dihydroxy-5α-cholenic acid derivative. Extensive chemical transformations have been carried out on lanosterol to prepare substrates for microbial transformation. In particular a new efficient method for the mild degradation of the lanosterol side chain to a 17β-acetyl group has been developed.

This thesis investigated the formation of compositional gradients across 0.5 – 1 cm of porous silicon layers which had thicknesses of 2 – 10 μm. These compositional gradients were then characterised, and their potential use as vapour sensors was probed. Surface composition gradients have been reported on flat surfaces, but this is the first time that they have been reported on a three-dimensional material with controlled pore geometry. Chemical gradients have been generated across the surface of porous silicon by performing electrochemical attachment of organohalides with an asymmetric electrode arrangement, and by chemical hydrosilylation of alkenes in the presence of a diffusion gradient of diazonium salts across the porous silicon surface. Samples with electrochemical gradients of methyl, pentyl acetate, and decyl and using chemical hydrosilylation with gradients of undecanoic acid and decyl groups. The latter four gradient-modified porous silicon types have been ‘endcapped’ with methyl groups to give improved stability and greater hydrophobicity. The pentyl acetate and undecanoic groups have been converted into pentanol and undecanoate groups respectively to increase the hydrophilicity of these porous silicon surfaces. The gradients have been characterised using two-dimensional FTIR microspectrophotometry and water contact angle measurements. The interaction of these gradient porous silicon samples with ethanol, heptane, toluene and 2-hexanol vapours have been monitored either by UV-Vis reflectance spectroscopy at selected points across the surface or more globally using a digital camera. The undecanoate gradient porous silicon sample showed a large difference in optical response between the undecanoate end and the methyl end of the gradient when exposed to water vapour, showing that imposition of a chemical gradient can alter the sensing character of porous silicon in a controllable manner.

The diterpene hydrocarbon rimuene has exercised over the years, the interest of a number of laboratories engaged in natural product research. As is usual with this class of compound its occurrence is confined almost entirely to the native New Zealand flora and although only two types of tree have been used as sources, it has been identified, often only in small amounts, in at least ten different species. Its co-occurrence with other diterpenes of similar chemical and physical properties has often made it hard to isolate and this, together with its geographic distribution has probably tended to limit its availability for research purposes, and hence delayed the elucidation of its structure. The awareness of the existence of this compound of undetermined structure has, therefore, caused a noticable gap in the knowledge of the diterpenoids, and this hiatus has been accentuated by Wenkert who, believing it to have a pimarane skeleton, suggested that it might be the important biogenetic precursor of a number of tetracyclic diterpenoids. Recent work, confirmed by the present investigation has, however, shown that previous structural predictions were incorrect and in actual fact rimuene has a rosane skeleton. As Ruzicka has shown for this class of compound, it has probably been formed from a pimarane diterpenoid and therefore represents a later biogenetic step than the other diterpene hydrocarbons which are always found with it in nature.

The diterpene hydrocarbon rimuene has exercised over the years, the interest of a number of laboratories engaged in natural product research. As is usual with this class of compound its occurrence is confined almost entirely to the native New Zealand flora and although only two types of tree have been used as sources, it has been identified, often only in small amounts, in at least ten different species. Its co-occurrence with other diterpenes of similar chemical and physical properties has often made it hard to isolate and this, together with its geographic distribution has probably tended to limit its availability for research purposes, and hence delayed the elucidation of its structure. The awareness of the existence of this compound of undetermined structure has, therefore, caused a noticable gap in the knowledge of the diterpenoids, and this hiatus has been accentuated by Wenkert who, believing it to have a pimarane skeleton, suggested that it might be the important biogenetic precursor of a number of tetracyclic diterpenoids. Recent work, confirmed by the present investigation has, however, shown that previous structural predictions were incorrect and in actual fact rimuene has a rosane skeleton. As Ruzicka has shown for this class of compound, it has probably been formed from a pimarane diterpenoid and therefore represents a later biogenetic step than the other diterpene hydrocarbons which are always found with it in nature.

This thesis investigated the formation of compositional gradients across 0.5 – 1 cm of porous silicon layers which had thicknesses of 2 – 10 μm. These compositional gradients were then characterised, and their potential use as vapour sensors was probed. Surface composition gradients have been reported on flat surfaces, but this is the first time that they have been reported on a three-dimensional material with controlled pore geometry. Chemical gradients have been generated across the surface of porous silicon by performing electrochemical attachment of organohalides with an asymmetric electrode arrangement, and by chemical hydrosilylation of alkenes in the presence of a diffusion gradient of diazonium salts across the porous silicon surface. Samples with electrochemical gradients of methyl, pentyl acetate, and decyl and using chemical hydrosilylation with gradients of undecanoic acid and decyl groups. The latter four gradient-modified porous silicon types have been ‘endcapped’ with methyl groups to give improved stability and greater hydrophobicity. The pentyl acetate and undecanoic groups have been converted into pentanol and undecanoate groups respectively to increase the hydrophilicity of these porous silicon surfaces. The gradients have been characterised using two-dimensional FTIR microspectrophotometry and water contact angle measurements. The interaction of these gradient porous silicon samples with ethanol, heptane, toluene and 2-hexanol vapours have been monitored either by UV-Vis reflectance spectroscopy at selected points across the surface or more globally using a digital camera. The undecanoate gradient porous silicon sample showed a large difference in optical response between the undecanoate end and the methyl end of the gradient when exposed to water vapour, showing that imposition of a chemical gradient can alter the sensing character of porous silicon in a controllable manner.

This thesis investigated the formation of compositional gradients across 0.5 – 1 cm of porous silicon layers which had thicknesses of 2 – 10 μm. These compositional gradients were then characterised, and their potential use as vapour sensors was probed. Surface composition gradients have been reported on flat surfaces, but this is the first time that they have been reported on a three-dimensional material with controlled pore geometry. Chemical gradients have been generated across the surface of porous silicon by performing electrochemical attachment of organohalides with an asymmetric electrode arrangement, and by chemical hydrosilylation of alkenes in the presence of a diffusion gradient of diazonium salts across the porous silicon surface. Samples with electrochemical gradients of methyl, pentyl acetate, and decyl and using chemical hydrosilylation with gradients of undecanoic acid and decyl groups. The latter four gradient-modified porous silicon types have been ‘endcapped’ with methyl groups to give improved stability and greater hydrophobicity. The pentyl acetate and undecanoic groups have been converted into pentanol and undecanoate groups respectively to increase the hydrophilicity of these porous silicon surfaces. The gradients have been characterised using two-dimensional FTIR microspectrophotometry and water contact angle measurements. The interaction of these gradient porous silicon samples with ethanol, heptane, toluene and 2-hexanol vapours have been monitored either by UV-Vis reflectance spectroscopy at selected points across the surface or more globally using a digital camera. The undecanoate gradient porous silicon sample showed a large difference in optical response between the undecanoate end and the methyl end of the gradient when exposed to water vapour, showing that imposition of a chemical gradient can alter the sensing character of porous silicon in a controllable manner.

This thesis examines the preparation and chemistry of osmium and ruthenium complexes with ligands featuring the group 14 donor atoms carbon and silicon, and the group l5 donor atom phosphorus. Aryl, alkenyl, and alkynyl complexes of osmium and ruthenium, prepared via mercury reagents, are discussed in Chapter One. 5-Coordinate 2-halophenyl complexes M(C6H4X-2)CI(CE)(PPh3)2 (M = Os; X = C1, Br; E = O, S; M = Os; X = I E = O; M = Ru; X = CI, Br, E = O) were synthesised by reaction of organomercury reagents Hg(C6H4X-2)2 (X = C1, I, Br) with MHC1(CO)(PPh3)3: (M = Os, Ru). Os(C6H4X-2)C1(CO)(PPh3)2 (M = Os; X = C1, I, Br) were characterised structurally and the interaction between X and M examined. Attempted benzyne syntheses using these complexes were not successful. 6-Coordinate complexes M(C6H4X-2)C1(CO)(CE)(PPh3)2 (M = Os, X = C1, E = O, S; Br, E = O, S; M = Os, E = O, X = I; M = Ru, E = O, X = C1, Br) were prepared by the addition of carbon monoxide to the corresponding 5-coordinate precursors. Approaches towards reduction of these complexes are discussed. The structure of Os(C6H4C1-2)C1(CS)(CO)(PPh3)2 revealed that the thiocarbonyl and the aryl halide ligands were cis and therefore in an ideal geometry to rearrange and form a substituted thioacyl ligand. Indeed, on heating Os(C6H4X-2)C1(CS)(CO)(PPh3)2 (X = C1, Br) the corresponding thioacyl complexes Os(η2-CS{C6H4X-2})C1(CO)(PPh3)2 (X = C1, Br) were formed. The decreased electron density in the halo aryl rings of these thioacyl complexes, combined with the fact that the halide substituents were no longer bonded to the metal, enabled facile lithiation of the aryl rings, even at low temperature. Quenching the appropriate lithiated intermediate with Bu3SnC1 gave Os(η2-CS{C6H4SnBu3-2})C1(CO)(PPh3)2. These results suggested that M(C6H4{CH2X}-2)C1(CO)(PPh3)2 (M = Os, Ru) were worthwhile target complexes for lithiation studies. To this end, Hg(C6H4{CH2OH}-2)C1 was prepared but attempts to convert the alcohol into a tosylate (for subsequent reaction with LiX) were unsuccessful and so this chemistry was not pursued further. Hg(C6H4{CH2OH}-2)C1 transferred the benzylic alcohol groups to osmium and ruthenium, albeit in low yields. Oxidation of Hg(C6H4{CH2OH}-2)C1 with PCC provided the benzaldehyde-containing mercury complex Hg(C6H4{CHO}-2)C1, symmetrization of which gave Hg(C6H4{CHO}-2)2. The latter compound was used to prepare aldehyde complexes of osmium and ruthenium, as well as mercury(II) benzaldoxime and benzaldimines. The aldehyde oxygen atoms in the osmium and ruthenium complexes were bound to the metals and were unaffected by amines, and by attempts to displace them from the metals. The addition of dimethyldithiocarbamate to Ru(C6H4{CHO}-2)C1(CO)(PPh3)2 displaced a triphenylphosphine ligand and Ru(C6H4{CHO}-2)(CO)({CH3}2NCS2)(PPh3) was formed. Transfer of the benzaldoxime ligand, and various benzaldimine ligands [C6H4{C[H]=NR] R = Me, CH2CH2NEt2, CH2CH2NMe2], to osmium or ruthenium gave M(C6H4{C[H]=NR}-2)C1(CO)(PPh3)2 (M = Ru, R = OH; M = Os, Ru; R = Me, CH2CH2NEt2;M= Ru, R = CH2CH2NMe2). The derived cationic complexes Ru(C6H4{C[H]=NCH2CH2NHR2}-2)C1(CO)(PPh3)2]BF4 (R - Me, Et) were prepared by protonation of the benzaldimine complexes with HBF4+ and Ru(C6H4{C[H]=NCH2CH2NR2}-2)(Co)(PPh3)2]BF4(R = Me, Et)were prepared by addition of AgBF4. Bromination of Ru(C6H4{C[H]=NMe}-2)C1(Co)(PPh3)2 gave Ru(C6H3{C[H]=NMe}-2, Br-4)C1(Co)(PPh3)2 which was lithiated at low temperature. The aryllithium was quenched with Bu3SnC1 to give Ru(C6H3 { C[H]=NMe}-2,SnBu3-4)C1(CO)(PPh3)2. The reaction of alkynylmercury reagents with osmium and ruthenium complexes are discussed in the following sections. Treatment of RuHCI(CO)(PPh3)2 with Hg(C≡CPh)2 has been reported previously, the result being formation of an α-phenylethynyl-trans-β-styryl ligand. However, the corresponding reaction with OsHCI(CO)(PPh3)3 resulted in catalysed coupling of the alkyne. This reaction was re-examined and the 6-coordinate α-phenylethynyl-trans-β-styryl osmium complex was prepared by direct reaction of the 5-coordinate complex with acetate ion. A dicarbonyl complex containing the α-phenylethynyl-trans-β-styryl ligand, Os(C{C≡Ph}=CHPh)CI(CO)2(PPh3)2, was prepared by the addition of carbon monoxide in the presence of LiC1 to the acetate complex. Thiocarbonyl complexes containing an α-phenylethynyl-trans-β-styryl ligand were prepared. Addition of carbon monoxide to solutions containing these complexes gave the thioacyl analogues M(η2-CS{C[C≡CPh]=CHPh})C1(CO)(PPh3)2 (M = Os, Ru). The remaining sections in Chapter One examine the reactions of mercury(II) reagents with the osmium(O) complexes Os(CO)2(PPh3)3 and OsCI(NO)(PPh3)3. Oxidative addition of the mercury-carbon bond of HgR2 (R = C6H4CH34, C≡CPh, trans-CH=CHPh) to Os(CO)2(PPh3)3 gave OsR(HgR)(CO)2(PPh3)2. Reaction of the acetylide or styryl complexes with iodine resulted in cleavage of the osmium-mercury bond and yielded either Os(C≡CPh)I(CO)2(PPh3)2 or Os(trans-CH=CHPh)I(CO)2(PPh3)2. Similar complexes were not accessible from reaction of the osmium(II) complex OsHC1(CO)(PPh3)3 with the appropriate mercury reagent. Whereas the mercury reagents reacted with Os(CO)2(PPh3)3 to give the simple oxidative addition products, the corresponding reactions of OsC1(NO)(PPh3)3 with HgR2 did not always give the analogous products OsR(HgR)C1(NO)(PPh3)2. Addition of Hg(C6H4CH3-4)2 to OsC1(NO)(PPh3)3 gave a mixture of the bis(p-tolyl) complex Os(C6H4CH3-4)2C1(NO)(PPh3)2 and the mono(p-tolyl) complex Os(C6H4CH3-4)CI2(NO)(PPh3)2. The structure of the bis(p-tolyl) complex revealed that the p-tolyl ligands were trans and the metal-carbon(aryl) bond lengths were extremely long. Addition of pyridine to the bis(p-tolyl) complex gave Os(C6H4CH3-4)2(C5H5N)CI(NO)(PPh3), which contained two cis p-tolyl ligands. The reactions of Hg(C6H4C1-2)2 and Hg(C≡CPh)Ph with OsCI(NO)(PPh3)3 gave complexes containing a single organic ligand. In contrast, treatment of OsCI(NO)(PPh3)3 with either Hg({C4H4S-2})2 or Hg([C4H4SMe-5]-2})2 gave the dithienyl complexes OsR2CI(NO)(PPh3)2 [R = (C4H4S)-2, ({C4H4SMe-5})-2. Furthermore, the reaction of Hg(CF3)2 with OsC1(NO)(PPh3)3 gave Os(CF3)(Hg{CF3})CI(NO)(PPh3)2. Treatment of OsCI(NO)(PPh3)3 with Hg(trans-CH=CHPh)2 gave Os(trans-CH=CHPh)CI2(NO)(PPh3)2 and an osmaindene complex, Os(C6H4CH=CH)H(NO)(PPh3)2, which in turn gave Os(C6H4CH=CH)Cl(No)(PPh3)2 on treatment with HCl. Chapter Two examines osmabenzene chemistry. Spectroscopic data were collected for the known complexes Os(η2-C[S]CH=CHCH=CH)(CO)(PPh3)2, and Os(C[SH]CH=CHCH=CH)(cis-CI)(CO)(PPh3)2. Oxidation of the osmabenzene thiol to a sulfinic acid was attempted. Methylation of the parent complex, Os(η2-C[S]CH=CHCH=CH)(CO)(PPh3)2, gave the product of kinetic control as Os(C[SMe]CH=CHCH=CH)(cis-I)(CO)(PPh3)2, reported previously, which rearranged on heating to give the trans isomer, Os(C[SMe]CH=CHCH=CH)(trans-I)(CO)(PPh3)2. Approaches to auration of the sulphur in Os(η2-C[S]CH=CHCH=CH)(CO)(PPh3)2, are described. Although the metallabenzenes reported previously have physical properties comparable with those of benzene itself, little evidence has been reported to suggest that the chemical reactivity of metallabenzenes is similar to that of benzene. The research described in this chapter provides the first example of a metallabenzene complex that undergoes aromatic electrophilic substitution. Thus, the metallabenzene complex Os(C[SMe]CH=CHCH=CH)(cis-I)(Co)(PPh3)2was brominated, chlorinated, and even iodinated. Crystal structure determinations and NMR studies showed that C5, which was activated by the thioether functionality, was the preferred site of electrophile attack. Even more significantly, Os(C[SMe]CH=CHCH=CH)(cis-I)(Co)(PPh3)2 was nitrated with either Cu(NO3)2/acetic anhydride, or with the more potent reagent NO2CF3SO3.CF3SO3H. The site of nitration was identical with that of halogenation, namely, C5. Previous syntheses of metallabenzenes had reported the use of the simplest alkyne, ethyne. This chapter describes the first metallabenzene complex prepared from propyne, giving the metallabenzene Os(η2-C[S]C{CH3}=CHCH=C{CH3})(CO)(PPh3)2 and the oxidative addition product Os(C≡CCH3)H(CO)(CS)(PPh3)2. Ten of the metallabenzene complexes were characterised structurally and the significance of the carbon-carbon bond lengths in the metallacyclic rings are discussed. The complete characterisation of these complexes by NMR spectroscopy revealed that the ring protons in the metallabenzene complexes, excepting H6, were at chemical shifts similar to those expected for normal aromatic carbons. Chapter Three examines the coordination of the strongly ח-accepting tris(N-pyrrolyl)phosphine ligand to osmium. Two tris(N-pyrrolyl)phosphine complexes of Os(II), OsHCI(CO)(PPh3)2(P{NC4H4}3) and OsH(C6H4CH3-4)(CO)(PPh3)2(P{NC4H4}3), were prepared. Both of these showed significantly higher infrared carbonyl stretching absorptions than the analogous triphenylphosphine complexes, reflecting the ח-acceptor nature of the tris(N-pyrrolyl)phosphine ligand. The osmium(O) complexes Os(CE)(CO)(PPh3)2P(NC4H4)3 (E = O, S) were prepared and the carbonyl complex was characterised structurally. The osmium-phosphorus(pyrrolyl) bond length of this complex was relatively short. The tris(N-pyrrolyl)phosphine ligand was in the equatorial plane as were the two carbonyl ligands. Chapter Four examines silyl and siloxane complexes of osmium and ruthenium. Although triethoxysilyl complexes of ruthenium have been prepared previously through ethanolysis of the coordinated SiCl3 group, the osmium analogues could not be prepared this way. It was found that Os(Si{OEt}3)CI(CO)(PPh3)2 could be prepared successfully by direct treatment of Os(Ph)Cl(CO)(PPh3)2 with triethoxysilane. Addition of carbon monoxide to the 5-coordinate triethoxysilyl complex afforded the dicarbonyl complex. The triethoxysilyl nitrosyl complex, OsH(Si{OEt}3)CI(NO)(PPh3)2, was prepared by oxidative addition of triethoxysilane to OsCI(NO)(PPh3)3, and the siloxane nitrosyl complex, Os(O[Si{OEt}3])CI2(NO)(PPh3 )2 was also characterised fully. Prior to this work only a single silatranyl complex was known. This chapter reports fifteen new silatranyl complexes and examines the unique properties conferred upon the silatrane by coordination to the metal. The silatranyl-containing complexes OsH(Si{OCH2CH2}3N)CI(No)(pph3)2 and OsH(Si{OCH2CH2}3N)(CO)2(pph3)2 were formed by oxidative addition of silatrane to the appropriate osmium(O) complex. Neither was suitable for further research because the chloro nitrosyl complex ejected silatrane in the presence of oxygen, and the dicarbonyl complex was isolated as a mixture of three isomers. In contrast, the unsaturated complexes M(Si{OCH2CH2}3N)C1(CO)(PPh3)2 (M = Os, Ru) were excellent materials for further study. The crystal structures of both of these complexes reveal typical metal-silyl distances with atypical silatranyl N→Si bond lengths. In both cases the N→Si bond length is elongated, the nitrogen is planar, and the cage is best described as quasi-silatranyl. The unsaturated nature of the metal in Os(Si{OCH2CH2}3N)CI(CO)(PPh3)2 offered access to derived complexes. The ח-acid carbon monoxide added to the vacant site forming Os(Si{OCH2CH2}3N)C1(co)2(pph3)2. Methylation of the 5-coordinate silatranyl complexes, gave [M(Si{OCH2CH2}3NMe)CI(CO)(PPh3)2]CF3SO3(M = Os, Ru). This reaction has not been achieved previously for silatrane derivatives. These methylated complexes have the longest recorded N→Si distances for any silatrane derivatives, and display a tetrahedral bridgehead nitrogen which points out of the cage at the methyl substituent. Protonation of the 5-coordinate silatranyl complexes gave [M(Si{OCH2CH2}3NH)CI(CO)(PPh3)2]CF3SO3 (M = Os, Ru). The thiocarbonyl-containing derivatives Os(Si{OCH2CH2}3N)CI(CS)(PPh3)2 and [Os(Si{OCH2CH2}3NMe)CI(CS)(PPh3)2]CF3S03 were prepared and these complexes showed spectroscopic properties similar to those observed for the carbonyl-containing analogues. The 5-coordinate silatranyl-thiocarbonyl complex rearranged in the presence of carbon monoxide to form Os(η2-C{S}Si{OCH2CH2}3N))CI(CO)(PPh3)2, which did not contain a metal-silicon bond. The silatranyl cage in the structure of this complex showed a very short N→Si bond with the nitrogen centre tetrahedral and pointing into the cage and towards the silicon atom. The osmium(IV) complex OsH3(Si{OCH2CH2}3N)(pph3)2 was prepared from OsH4(PPh3)2 and silatrane. The structure of this complex showed that the hydride ligands were oriented trans to a single triphenylphosphine in each case. Treatment of this complex with methyl iodide gave the quaternary salt [OsH3(Si{OCH2CH2}3NMe)(PPh3)3]I, and protonation gave [OsH3(Si{OCH2CH2}3NH)(PPh3)3]CF3SO3.

A theoretical analysis of the Be9(pd)Be8 reaction at a proton energy of 333kev assuming compound nucleus formation of a 1- state in B10 predicts that if protons are unpolarised the deuterons have zero vector but maximum possible tensor polarisation i.e. the magnetic substate population ration is 1:0:1 and Pzz =1. furthermore it predicts that both the polarization and the angular distribution of these deutrerons are isotropic. The polarization of these deuterons emitted at 30 in the lab. has been measured by allowing them to initiate the reaction He3 (dp)He4 at 430 kev and determining the angular distribution of the resulting protons. / Note: parts of this thesis have since been published in Nuclear Physics A 95 663-673 http://dx.doi.org/10.1016/0375-9474(67)90859-7