X-ray crystallographic analyses of a group of salicylaldoxime ligands, and their copper(II) complexes were carried out. The three of the copper(II) complexes studied, bis-(2-hydroxy-5-methylbenzaldehydeoximato)copper(II), [(L1)2Cu]n, bis-(2-hydroxy-5-methylbenzophenoneoximato)copper(II), [(L5 )2Cu] and bis-(2-hydroxy-3-t-butyl-methylbenzaldehydeoximato)copper(II), [(L7 )2Cu], existed as centrosymmetric molecules, with the copper atom at crystallographic inversion centre. Two bidentate ligands produce a planar arrangement of donor atoms around the copper. The molecules reassemble macrocyclic systems, due to effective intramolecular hydrogen bonding linking the two ligands. To correlate chelating properties with intramolecular H-bond strengths an X-ray structural analysis was carried out on one ligand supplied by ICI, and thought to be 3,5-dimethyl-2-hydroxy acetophenone methoxyimine. Unexpectedly the molecule was found to be 3,5-dimethyl-4-hydroxy acetophenone methoxyimine. This was successful in explaining the anomolously low efficiency of extraction for copper(ll), shown by the related n-nonyl oxime ligand. It is thought that the presence of syn and anti isomers in 2-hydroxybenzophenone oximes accounts for the low efficiency of copper extraction shown by reagents which contain this chelating unit. The syn and anti forms, of 2-hydroxy-5-methylbenzophenone oxime (L5H) were prepared separately. Significantly both forms appear to yield [L5]2CH] when treated with Cu(CH3CO2)2.H20 in a homogeneous solution, although there are indications that formation of the complex from the syn-form is slower. Selectivity of complexation of metal ions by macrocyclic ligands were investigated. The stability constant data of zinc(II) and cadmium(II) complexes of a series of cyclic quadridentate ligands were compared. For the 15-membered N4 macrocyle L15, the propyl (L15p) and butyl (L15b) C-alkylated derivatives were synthesised for the first time, and the cadmium(II) complex of (L15b) was prepared. X-ray analysis were performed on the free ligand L15b, and the three metal complexes [Zn(L15)(NO3)(H2O)]NO3, [Cd(15)(NO3)NO3, and [Cd(L15b)(NO3)NO3. These results were then correlated with the results of stability constant measurements in discussing the probable behaviour of the ligands in the solution environment and hence to provide some insight into the effect of alkylation on the discrimination for zinc(II) over cadmium(II). The discrimation for zinc over cadmium is decreased on alkylation and the reason for this is discussed.

Experimental results are presented here for a 7017-T651 AlZnmg alloy where stress corrosion crack growth data on compact tension and double cantilever beam specimens has been obtained in distilled water, water saturated air vapour and aqueous chloride environments over a range of temperatures. The results, including fracture surface examination, are interpreted as indicating that a hydrogen embrittlement mechanism is operative whereby hydrogen, generated by a corrosion reaction, embrittles the alloy. Activation energy determinations have been made for stress corrosion crack growth at the free corrosion potential, and the activation energy results have been interpreted as indicating that the same process is rate controlling during both region I and region II crack growth. Hydrogen diffusion in AlZnMg has been investigated by electrochemical permeation and gas chromatography techniques. It was found that the sensitivity of the electrochemical permeation technique is not sufficient to detect a permeation transient due to hydrogen. However, the diffusion coefficient for hydrogen in AlZnMg was determined by gas chromatography over a range of temperatures. A critique is made of previous reported measurements of hydrogen diffusion in an AlZnMg alloy by electrochemical permeation. The average diffusion rate of hydrogen in AlZnMg alloy material using gas chromatography on hollow cylindrical specimens was determined. Finally, the surface reactions between polished AlZnMg alloy surfaces and distilled water, water saturated air vapour and aqueous chloride solutions have been Investigated to determine the rate of hydrogen production and the morphology of the reaction products. It was shown that the reaction of water saturated air vapour with the commercial AlZnMg alloy surface did not exhibit specificity with respect to grain boundary attach, unlike the specificity shown by high purity versions of these alloys. The contribution of these studies to the understanding stress corrosion crack growth by a hydrogen embrittlement mechanism is discussed.

The preparation and characterisation of several new types of tetraazamacrocycles and their copper(ll) complexes are described. Condensations of 4,8-diaza-2,3:9,10-dibenzoundecane-1,11-dione with a range of aliphatic and aromatic diamines has allowed the isolation of metal-free diimine macrocycles without recourse to metal template' or 'high dilution' techniques. In several cases these diimines were successfully reduced to yield the related tetraamine macrocycles. For the 15-membered macrocycle, 18,19,20,21-tetrahydro-17-H-tribenzo(f,j,n)(1,5,9,12)tetraazacyclopentadecine, derivatives bearing different substituents in one of the benzene rings were prepared. For the diimine ligands of this type both cationic and neutral copper(II) complexes were isolated and characterised; (the latter type result from loss of anilino protons on complex formation.) Compounds which are structurally related to the above tetraazamacrocycles are described. C-alkylation of the diimines using organolithium reagents appears to proceed regioselectively. An X-ray structure determination of the copper complex of 5,12-dimethyl-5,6,11,12, 18,19,20,21-octahydro-17-H-tribenzo (f,j,n)(1,5,9,12) tetraazacyclopentadecine shows this to be meso-dimethyl isomer. Condensation of 2,2 '-(oxalyldiimino)bisbenzophenone with diamines in the presence of copper(ll) acetate yields macrocyclic neutral copper(ll) complexes which contain a dioxo linkage between the anilino nitrogen atoms. An X-ray structure determination of [5,12-diphenyl-18,19-dioxo-9,10,17,18,19,20-hexahydrodibenzo (e,m) (1,4,8,11) tetraazacyclotetradecinato(2-)]-dimethylformamidocopper II dihydrate indicates steric crowding for the phenyl substituents on the imine carbon atoms. Polycyclic tetraza compounds result from the condensation of the cyclic tetraamines with glyoxal. A group of 20,21 and 22-membered sexadentate ligands having N402 donor sets, and 20-membered ligands with N4S2 and N6 donor sets have also been studied; e.g.: 12,13,25,26,27,28-Hexahydrotetrabenzo (e,i,o,s) (11,14,1,4,8,17) dithiatetraazacycloeicosine. Reduction of the azomethine linkages was achieved for all these ligands and cationic copper(ll) complexes have been prepared and characterised. An X-ray structure determination showed that one of the N402 tetraamine ligands uses only three of its nitrogen atoms in forming a complex with copper(II).

The investigation has consisted of two elements, one concentrating on the stabilities of metal complexes in solution, and the other on the co-ordination chemistry of some new N-donor macrocyclic ligands.

Unsymmetrical compounds of the general type: ArCH2(S)nCH2CH2OH (n = 1-4, Ar - o- or p-ClC6H4, o- or p-MeC6H4, Ph, p-MeOC6H4) have been prepared as analogues of the trisulphide PhCH2SSSCH2CH2OH which occurs naturally in the stem and root of the plant Petivcris alliacea.

A new five-membered cyclic tertiary diphosphine has been synthesised, PhPCH(=CH2)PPhCH2CH2 (X). Cis and trans isomers were observed; and the trans isomer (Xb) has been isolated and fully characterised by P-31, H-1 and C-13 n.m.r. spectroscopy.

The photoreaction of pentacarbonyliron with 1,3-diene/monoolefin mixtures results in C-C coupling of the two organic substrates, forming a six-carbon chain which is coordinated to the Fe (CO) 3 moiety in a 1-a,4-6-n3 fashion. Using various 1,3- dienes and monoolefins carrying methyl and carboalkoxy substituents, the wide applicability of this process is demonstrated.

This thesis describes the determination of the crystal structures of one organo-metallic and three organic compounds. Originally the research programme had envisaged working on substances which were obtained by the corrosive action of hydrocarbon oils on metallic lead, but after some considerable time had been spent on attempting to crystallise these rather intractable materials, the research was switched to deal with other compounds of industrial importance which were more readily available in the Polytechnic. Of the four crystalline compounds investigated, two were connected with the inhibition of the corrosion of steel, one was an additive used in the polymer industry, and one was a molecular complex which had exhibited some unusual physico-chemical properties. X-ray intensity data were collected in several different ways. One crystal structure was solved from visually estimated photographic data, two sets of data were collected on our. Stoe Stadi-2 diffractometer which uses Weissenberg geometry, and one set which comprised over 7000 intensities, the largest data-set ever worked on in this Department, was recorded on the CAD-4 diffractometer at Queen Mary College, University of London, by arrangement with Dr M Hursthouse. A variety of techniques were used in working out the crystal structures. One was solved by the MULTAN direct methods computer program, two by the use of the NRC direct methods program using 12 relationships, and one by "trial-and-error" because none of the other methods produced any meaningful results. Refinement of all four structures was carried out using the NRC suite of programs implemented on our Polytechnic computers (DEC-IO, DEC-20, and VAX 11/750). Anisotropic thermal vibrations were allowed for, and hydrogen atom positions were calculated resulting in the conventional R-factors of 6.7, 6.1, 6.6 and 6.1% for the four structures. No abnormalities were encountered in the results for three of the structures, but the picric acid - naphthalene complex turned out to be disordered with some of.the picric acid molecules occupying an alternative orientation in the unit cell which resulted in displacement of the naphthalene molecules also, and the possibility of proton transfer within the picric acid molecule which was confirmed by differences in bond lengths and anisotropic temperature parameters.

M(cup)n complexes have been prepared (M = Cu, Ni, Fe, Co, Mn, or Zn ; n=2, M-=Fe ; n=3 cupH = cupferron) and structures have been proposed on the basis of spectroscopic, thermal and magnetic techniques. Their behaviour towards the Lewis bases pyridine, triphenyl- phosphine and hydroxylamine has been investigated. All of the M(cup)2 complexes, except when M = Zn or Fe, affords adducts on reaction with pyridine, whereas the complex Fe(cup)3 underwent an internal redox reaction. Internal redox behaviour was also observed when the complexes Cu(cup)2 and Fe(cup)3 were treated with triphenylphosphine. On the basis of the nature of the products arising from these reactions and from a kinetic study of the reaction of Cu(cup) with the phosphine, mechanistic proposals have been made.

The chemistry of 1,2-quinone monooximes hi« been reviewed. The synthesis of 5-amlno, 5-alkylamino and 5-acylamino substituted 1,2-benzoquinone monooximes was systematically studied. Thus, It has been found that 5-amino-1,2-benzoquinone-2-oxime (5-AqoH), 5-ethylamino-4-methyl 1-1,2- benzoquinone-2-ox I me (5-Et-4-MeqoH), 5-hexylamino-1,2-benzoquinone-2-oxime (5-HxqoH) and 5-heptylamino-1,2-benzoquinone-2-oxime (5-HptqoH) were best prepared by the reaction of the corresponding phenol with sodium nitrite In the presence of concentrated hydrochloric acid. Under these conditions, N-nitrosation was inhibited and the 1,2-benzoquinone- 2-oxlme hydrochlorides were formed. The 5-acylaminc1,2-benzoqulnone-2- oximes, 5-propionylamino-1,2-benzoquinone-2-oxime (5-PrqoH), 5-butyrylamino-l,2-benzoquinone-2-oxime (5-BuqoH), 5-pentanoylamino-1,2-benzoquinone-2-oxime (5-PeqoH), and 5—heptanoylamino-1,2-benzoquinone-2-oxime (5-HpqoH) were obtained from their metal complexes, since the direct reaction of the phenol with sodium nitrite failed to give the compounds giving the corresponding l,4-benzoquinone-4-oximes,3- proplonylamino-1,4-benzoquinone-4-oxime (3-PrqoH), 3-butyrylamino-1,4- benzoquinone-4-oxlme (3-BuqoH), 3-pentanoylamino-1,4-benzoquinone-4- oxime (3-PeqoH) and 3-heptanoylamino-1,4-benzoquinone-4-oxime (3-HpqoH) instead. Spectroscopic analysis has shown the compounds to exist in the quinone oximic rather than In the nitrosophenollc form. X-ray crystallographic studies of 5-HxqoH, 5-Et-4-MeqoH and 5- Et-4-MeqoH.HCI have shown ail three compounds to have some 1,4-oxime- imino character. The synthesis of nlckel(ll), copper(ll), palladium(ll) and platinum(ll) complexes of the acylamino and alkylamino substituted 1.2- benzoqulnone-2- oximes both by the direct and nitrosation methods was examined. The nitrosation method gave rise to complexes NI(qo)2.nH20, Cu(qo)2.nH20 and Pd(qo)2 but failed to give the corresponding platinum complexes. The latter were only obtained by the direct method. The nickel(ll) and oopper(ll) complexes reacted with pyridine and 2,2-dlpyrldyl to give adducts of the type NI(qo)2(py)2 Cu(qo) (py) and M(qo) (dipy) respectively. The palladium and platinum complexes however failed to react with either of the Lewis bases. Magnetic moment studies showed the hydrated complexes and Lewis base adducts to be magnetically dilute and thus monomeric. Such studies, as well as IR and LSIMS mass spectral analysis of the anhydrous complexes Ni(qo) and Cu(qo) , showed them to be associated both in the solid State and in solution. The reaction of selected 1,2-quinone monooximes (qoH) and metal 1.2- quinone monooximates (M(qo)2; qoH = 1,2-nap ht hoqu I none-1-oxime (1-nqoH), 1,2-naphthoquinone-2-oxime (2-nqoH), 3-butyrylamino-1,2-benzo-quinone-2-oxime, 4-chlora-1,2-benzoquinone-2-oxlme (4-ClqoH), 4-bromo-1,2-benzoquinone-2-oxime (4-BrqoH), 5-acetylamino-1,2-benzoquinone-2-oxime (5-AcqoH), 5-hydroxy-1,2-benzoquinone-2-oxlme (5-HqoH); M = Nl and Cu) with dimethyl acetylenedicarboxylate (DMAD) was systematically examined. The reaction of 1-nqoH and 5-HqoH gave rise to nucleophile addition products cis- and trans-(0-1,2-dicarbomethoxyethenyl)-1,2-naphthoquinone-1 -oxime, trans-(0-dicarbomethoxyethenyI)-5-hydroxy-1,2-benzoquinone-2-oxime. The yields were enhanced by the presence of small amounts of alkali and alkaline earth metal chlorides. With M(1-nqo)2 and M(Buqo)2 (M = Ni, Cu) 1,4-oxazines were the only products formed. However, M(4-Clqo)2 and M(4-Brqo)2 1,4-benzoxazinones and 1,4-benzoxazines were isolated. The reaction of the structurally related 1,2-dloxlmes emd metal 1,2-dioxlmates M(dmgH)2 and M(dagH)2» (dmgH2 = dimethylglyoxime; dagH_ = 1,2-dlaminoethanedlone dIoxIme; M = NI, Cu) with DMAD was also examined. The dimethylglyoxime and its metal complexes failed to afford any adducts while the 1,2-diaminoethanedione dioxIme and its metal complexes gave nucleophilic addition products cis- and trans-bls(0-1,2-dicarboxyethenyl)-1,2-diaminoethanedione dioxime. Mechanisms for the reaction of DMAD with both the quinone monooximes and the 1,2-dioximes have been proposed.