| Summary: | <p>The temperature dependence of most chemical reactions is accurately represented by the Arrhenius equation,</p> <p>k = Ae<sup>-E/RT</sup></p> <p>where k is the rate constant, A is a temperature independent entropy factor, often called the frequency factor, and K is the activation energy. This equation provides a useful method for the expression of results in a kinetic investigation.</p> <p>It is found that when changes in rate are brought about through moderate variations of react ant structure, or of solvent medium, the values of A and E change, but not in a random manner. Three types of behaviour are found.</p> <p>In some eases A remains constant while E varies as reactants are modified, this type of behaviour often being found, when a substituent is separated from the reaction centre by a benzene ring. In other cases A and B vary together, and it is often found that large values of E are associated with large values of A, and vice versa, so that the resultant change in rate is less than that indicated by the variation of either parameter alone, and a sort of 'compensation' is found. The third type of behaviour, in which K remains constant while variations of A control the rate, is occasionally found when solvents are varied through a related series or when the structure of the solvent is varied by the application of pressure. The various types of behaviour are reviewed in section I of the thesis, together with some of the theories advanced in explanation. It is suggested that the type of behaviour sometimes shown on solvent variation in which E remains constant while variations in A modify the rate, could arise from differences in the order of the pure solvent, and evidence is presented to support this view.</p> <p>The experimental work reported in the thesis consists of a study of the formation and decomposition of quaternary ammonium salts based on dimethyl aniline, both in nitrobenzene solution and in solvents consisting of mixtures of chlorobenzene with nitrobenzene. Special attention has been given to the variations in rate brought about through changes in A and E by modifications in the structure of the reactants or in the solvent medium. The experimental methods used in following the reactions at different temperatures include measurement of the conductivity of the solutions, and the aqueous extraction of the quaternary salt from the reaction mixture, followed by the titration of ionisable halide by Volhard's method. These methods are discussed in section II of the thesis.</p> <p>Section III contains an account of the problems connected with the kinetic behaviour of the reactions and the methods adopted to overcome difficulties arising from the unusual nature of this in some cases. It was found impossible satisfactorily, to represent the results for some of the reactions by a simple rate law, and this proved to be a major difficulty. The cause was traced to the decomposition of the salt first formed, together with the subsequent reaction of the decomposition products. These complications made it impossible in certain cases to use measured values of the reaction equilibrium in the calculation of the results. Two methods were used to overcome the difficulty, either the drawing of an accurate initial tangent or, when this was net possible, graphical extrapolation of the simple second order constants back to zero reaction.</p> <p>Some of the reactions, especially those in solvents of low polarity, were found to have equilibrium constants corresponding to very little reaction, a result sometimes obscured by reaction arising from the decomposition products.</p> <p>Quaternary ammonium salts were found to be insoluble in some of the solvents used, and the product occasionally separated cut from the solution, and the reactions became heterogeneous. This caused trouble in two ways. In some cases the experimentally measured rate corresponded to the rate of crystallisation, and net the rate of quaternization, whilst in other cases the surface of the solid product noted as a catalyst.</p> <p>Section IV of the thesis contains the record and discussion of results for the reactions in nitrobenzene solution. Although structural changes in the reactants provide examples conforming to all three of the types of behaviour discussed in section I, the major effect is to change the rate through a variation of E. The value of A also varies, so that there is some degree of 'compensation'. This compensation effect is found when the alkyl groups attached to nitrogen are varied, the halogen is varied, and in reactions with ethyl iodide, when ring substituents in the base are varied. When however, ring substituents in the base are varied in reactions of methyl iodide, the value of A does net change, and a comparison of the series of reactions for methyl and ethyl iodide throws doubt upon some established ideas about the stereochemistry of the transition complex.</p> <p>Not all of the reactions were attended by troublesome subsequent or side reactions, and in some cases equilibrium constants could be measured, and the entropies and enthalpies of reaction calculated. Consideration of these lends support to the views of the transition complex obtained from the kinetic results.</p> <p>Three of the quaternization reactions were studied in solvent mixtures containing nitrobenzene and chlorobenzene. The rate of the forward reaction proves to be mainly dependent on the activation energy, which rises as the solvent becomes progressively less polar, so that the rate in chlorobenzene is a little less than one seventh of the rate in nitrobenzene. The rate of the reverse reaction is, however, greater in the solvents of low dielectric constant, as a result of the low stability of the salt. This leads to the unusual circumstance in which slower reactions have a shorter "half life" than faster ones.</p> <p>The conclusions reached in the earlier sections are reviewed and collected in section VI, where some suggestions for additional work in this field are presented.</p>
|
|---|