Thermogravimetry (TG) and differential scanning calorimetry (DSC) were used to study the effect of sand, silica and kaolinite on crude oil combustion. Three distinct regions, namely distillation and two combustion/cracking regions were observed on all TG curves. Thermogravimetric curves were analyzed using an Arrhenius-type kinetic model and a ratio method to obtain kinetic parameters. Activation energy and reaction order were obtained from this analysis. The reaction order seemed to be insensitive to the presence of granular materials. However, a significant reduction of activation energy was caused by addition of kaolinite to the crude oil, indicating that the kaolinite had a catalytic and surface area effect on crude oil combustion/cracking reactions.

A fully automated in-situ combustion apparatus supported by a minicomputer was designed, constructed, and tested.Results obtained from four adiabatic dry combustion runs to investigate the effect on clay on crude oil combustion are reported. Sand mixtures of varying clay (kaolinite) content were saturated with crude oil and water. The fourth run was performed with amorphous silica powder in the sand mixture for comparison with clay results.We concluded that the large surface area of the clays was a major contributor to the fuel deposition process. However, different oxygen utilization efficiencies obtained from both types of sand mixtures indicated that mechanisms controlling the combustion reaction also depended on the composition of the porous matrix.A thermogravimetric analyzer (TGA) and a differential scanning calorimeter (DSC) were used to obtain kinetic data on the effects of kaolinite type clay on crude oil combustion. The addition of kaolinite clay or silica powder changed the shape of the crude oil TGA/DSC thermograms significantly, but sand had no effect. The major effect on DSC thermograms was a shifting of the large amount of heat produced from a higher to lower temperature range. Reduction of activation energy caused by the addition of kaolinite clay to the crude oil indicates both catalytic and surface area effects on combustion/cracking reactions. Introduction In-situ combustion is a thermal recovery process in which a portion of the crude oil is coked and burned in situ to recover the remaining oil. Design of the process involves experimental evaluation of process variables in laboratory experiments. Variables sought experimentally for the design of the process are usually fuel availability, air requirement, oxygen utilization efficiency, combustion peak temperature, combustion front velocity, effect of porous matrix, and kinetic parameters. Four methods have been used to obtain design data for in-situ combustion projects. These include (1) adiabatic in-situ combustion tube runs, (2) isothermal reactors, (3) flood pot tests, and (4) thermal analysis techniques.This paper describes an investigation of the effect of clay on in-situ combustion involving results from adiabatic combustion tube runs and thermal analysis methods. Part 1 describes the minicomputer-based insitu combustion system developed as part of the research program. Part 2 demonstrates application of the system to study the effect of clays on the in-situ combustion process. Combustion tube runs described in Part 2 are supplemented with thermal analysis methods to evaluate the effect of clay on in-situ combustion of a Kansas crude oil. Part 1-Development of an Automated In-Situ Combustion Tube Adiabatic tube runs have been the most commonly used approach for studying in-situ combustion. Since heat loss is small to nil in thick reservoirs, in-situ combustion is assumed to occur under adiabatic conditions. Adiabatic conditions in tube runs can be achieved either by insulating the tube or by reducing the temperature gradient between the sandpack and the environment surrounding the tube, or both. To attain adiabatic conditions in a partially or noninsulated tube, the temperature of the surroundings must be raised to that of the sandpack as the combustion front moves along the tube. Heater bands with proportional heat loads controlled by individual controllers are used. This requires a large number of controllers to control the temperature of the outside SPEJ P. 493^