Die oksidasiekinetika van 'n Fischer-Tropsch-was / Adriaan Stephanus Luyt

• Main Author: Luyt, Adriaan Stephanus
• Language: other
• Published: 2015
• Online Access: http://hdl.handle.net/10394/13338

The oxidation of hydrocarbons in the liquid phase plays an important part in the manufacturing of several products and it was therefore investigated intensively. The emphasis was on the mechanism of the process and, although mechanistic uncertainties still exist, several reaction schemes with mutual similarities were suggested. A limited amount of investigations were however carried out on the oxidation of waxes. The reason for this is that waxes consist of a series of longchain paraffins which is a much more complicated system. The present investigation was aimed at the kinetic analysis of a suitable wax in the interest of industry as well as science. It was carried out on Sasolwaks M, a medium-hard Fischer-Tropsch-wax with a relatively small carbon number distribution, in the absence of initiators and catalysts. The induction period, which is normally eliminated by these substances, could therefore be distinguished as a macroscopic stage and be studied as such. No description of the induction period could be found in the available literature. The oxidation of the wax was carried out in a special bubble column reactor using normal as well as oxygen enriched air. The kinetic analysis was based on oxygen consumption, product formation and paraffin consumption. The oxygen consumption was measured using an in-line-technique which was developed specifically for this investigation and which is based on the combined working of a rotameter, a flow meter and an oxygen detection meter. The product formation and paraffin consumption were both monitored using sample withdrawal techniques. The individual product types were analysed titrimetrically, while the unoxidised paraffins were determined by a pyrolysis-thinlayer chromatography technique. The process parameters (induction times, observed rate constants, activation enthalpies and entropies) for the consumption of oxygen and paraffins as well as the formation of products (peroxides, alcohols, carbonyls, acids, esters) could be obtained by the processing of the measured results (linear and polynomial regression, graphic extrapolation) according to the method of initial rates (differential method) or the integral method. The linearity requirement for the pseudo-first order graphs was fixed on a minimum of one half-life in accordance with the deviations which may be seen as a result of secondary reaction steps. The distinction between the macroscopic stages was made possible by the superposition of the concentration-time curves for the different starting materials and product types at the different reaction temperatures, and with due allowance for the obtained process parameters a qualitative description could be given for each. The induction period is the stage before the oxygen consumption reaches a maximum rate and during which the primary oxidation products (peroxides, alcohols, carbonyls) are formed at the maximum rate. The after-induction period, which starts when the oxygen consumption reaches its maximum rate, is characterized by the formation of secondary oxidation products (acids, esters) at the maximum rate as a result of the decomposition of the formed peroxides. The available kinetic data were used to give a theoretical description of the wax oxidation process by means of mechanism proposal and computer modellation. The course of the process is according to the proposed reaction scheme a free radical mechanism which consists of a number of chain reaction steps (initiation, propagation, termination, branching) . The establishment of a maximum concentration peroxide as a result of simultaneous formation and decomposition plays an important role in the scheme. The deduced rate equations and the calculated activation parameters could be correlated satisfactorily with the experimental results and it was shown thereby that the oxidation mechanism for waxes does not differ essentially from the one for hydrocarbons. The set up of an empirical computer model was less successful, but a number of important criteria for future investigations of this kind were obtained. The process analysis was facilitated by obtaining additional, semi-kinetic information. The small influence of the frit pore size on the oxidation rate, together with the low energy requirement for oxygen consumption and the slower oxidation of waxes with a higher viscosity, confirms that the taking up of air by the wax is diffusion controlled. The results with a few process accellerators (initiators, catalysts, oxidised wax, acids) show that not only is the induction period shortened or eliminated, the oxidation mechanism can also be changed. The necessity of investigating the role of process accellerators in the future is thereby emphasized. This is one of a number of aspects which are missing for a total image of the wax oxidation process, but for which perspectives were opened in the present investigation. === Proefskrif (DSc)--PU vir CHO, 1985