| Summary: | <p>This work is concerned with the investigation by electron spin resonance (e.s.r.) of organic free radicals in solution. All the radicals investigated contained oxygen functional groups and were all produced by one-electron oxidation or reduction. The thesis is presented in five Parts, <em>Part 1</em> being an introductory Part in which the basic principles of e.s.r. are discussed and the apparatus described. A flow apparatus, which is able to detect transient free radicals with lifetimes of <em>ca</em>. 10<sup>-2</sup> sec., is described; this apparatus has been used to detect many of the radicals investigated in this work.</p> <p><em>Part 2</em> describes the oxidation of monohydric phenols, ArOH, by a one-electron oxidising agent, ceric sulphate, to produce phenoxy radicals, ArO<sup>&bullet;</sup>. The e.s.r. spectrum of the unsubstituted phenoxy radical itself, PhO<sup>&bullet;</sup> , has been recorded, together with the spectra of some 50 other substituted phenoxy radicals. In a large majority of cases the spectra have been analysed successfully and coupling constants, arising from interaction of the free electron with both the ring protons and the α—protons of alkyl-substituents, have been obtained. It was found that the total spin density associated with the ring carbon atoms C<sub>2</sub>-C<sub>6</sub> is <em>ca</em> 0.80, thus indicating that <em>ca</em>. 20% of the spin density is associated with the phenoxy oxygen atom and the carbon atom to which it is attached.</p> <p>Qualitative trends in the coupling constants of the ring protons were observed and discussed; for example, it was found that the algebraic sum of the <em>ortho</em> and <em>meta</em> coupling constants was approximately constant with different substituents at the <em>para</em> position. This would suggest that the only effect of the <em>para</em> substituents is to cause a redistribution of the spin density between the <em>ortho</em> and <em>meta</em> positions.</p> <p><em>Part 3</em> describes the oxidation, both with potassium ferricyanide in alkaline solution and with ceric sulphate in acid solution, of resorcinol and substituted resoroinols, using the flow apparatus in each case. Different spectra were obtained from the oxidation of the same resorcinol under the different conditions. The spectra obtained by alkaline oxidation have been characterised as the <em>meta</em>-semiquinones, <em>i.e.</em> <em>m</em>-<sup>−</sup>0&bullet;C<sub>6</sub>H<sub>4</sub>•0<sup>&bullet;</sup> ; this being the first time such radicals have been described. Coupling constants have been unambiguously assigned by obtaining the spectra of <em>m</em>-semiquinones substituted with a carboxyl group at each ring position in turn.</p> <p>It was found that only <em>ca</em>. 18% of the total spin density can be associated with both oxygen atoms. This situation is quite different from that obtaining in <em>ortho</em>- or <em>para</em>-semiquinones, where some 60% of the spin density is associated with the oxygen atoms. Suggestions have also been made as to the radical species present in acid solution; a singly or doubly protonated radical, <em>i.e.</em> [•0&bullet;C<sub>6</sub>H<sub>4</sub>&bulet;OH] or [HO&bullet;C<sub>6</sub>H<sub>4</sub>&bulet;OH]<sup>+</sup> respectively, being most likely.</p> <p><em>Part 4</em> describes examples of the production of secondary radicals from catechol and some 2-(<em>o</em>-hydroxyphenyl)-l,4-benzo- quinones in alkaline dimethylformamide/water solutions. The e.s.r. spectrum of the <em>o</em>-semiquinone from catechol was found to quickly change to one which contained a superimposed spectrum due to a secondary radical. This radical was shown to be dibenzo-[l,4]-dioxin-2,3-semiquinone (i).</p> <p align="center"><img alt="Images for dibenzo-[l,4]-dioxin-2,3-semiquinone (i) and dibensofuran-l,4-semiquinones (ii)"/></p> <p>Secondary radicals were also observed on reduction of some 2-(,em>o</p>-hydroxyph0nyl)-l,4-benzoquinones, or oxidation of the corresponding quinol, under the same conditions as used for the oxidation of catechol. In this case the secondary radicals were shown to be dibensofuran-l,4-semiquinones, i.e. (ii). Possible mechanisms for the production of these radicals have been suggested. <p><em>Part 5</em> deals with the e.s.r. investigation of the radicals produced by the oxidation of simple substituted catechols. Two methods, <em>i.e.</em> aerial oxidation in alkaline dimethylformamide/water and oxidation with alkaline ferricyanide in water, were found, which in many cases permitted the spectra of the simple <em>ortho</em>- semiquinones to be recorded. In several cases secondary radicals were formed in aqueous solution, and the characterisation of these has been described.</p> <p>It was found that the secondary radicals were probably formed by nucleophilic hydroxyl anion attack on the <em>o</em>-quinone to produce a trihydroxybenzene system. 4-Hethylcatechol was shown to give the radical from l,2,4-trihydroxy-5-methylbenzene, whilst 3,4- dihydroxybenzaldehyde, for example, gave the pyrogallol semiquinone from 2,3,4-trihydroxybenzaldehyde. Possible reaction mechanisms have been discussed and an explanation offered for the difference in reactivity between the two substituted catechols.</p>
|
|---|