Kinetics of Crude-Oil Coke Combustion

Abstract

Summary Coke is a solid or semiliquid material that deposits on the sand-grain surface area and is eventually burned as a fuel during an in-situ combustion process. Its combustion is the main source of energy to sustain the fire front. This study investigates the effects of different variables-such as specific surface area, oxygen partial pressure, and oil content-on the coke combustion by thermogravimetric analysis (TGA). Each variable was varied while the others were kept constant. The TGA and the derivative thermogravimetric (DTG) curves were subjected to kinetic analysis. The rate equation produced indicated that the rate of coke combustion was proportional to the coke content yet to be burned, oxygen partial pressure, and sand-grain specific surface area. The rate constant followed an Arrhenius-type equation for the temperature dependency. The rate equation was tested for the range of specific surface areas of 0. 126 to 24.3 m /g [615 to 119,000 ft/lbm], oil content of 10 to 58 wt%, and oxygen partial pressure of 5 to 50 kPa [0.05 to 0.50 atm]. The model's validity was tested for various crude oils from different geographical locations in the U.S., Canada. and Latin America. Introduction Coke combustion is the main source of energy in the in-situ combustion process. In-situ combustion, commonly known as fireflooding, is a thermal oil recovery technique in which the oil is ignited underground. It consists of injecting compressed air, or air enriched with oxygen with or without recycled gases- In the case of wet combustion, water is also injected to scavenge heat from the burned zone and/or to quench the front partially. The flow of air and produced gases causes an oil bank to move through the reservoir toward the producing wells. In the field, ignition is achieved either by electrical means or by a gas burner; in some cases, ignition starts just by continuous injection of the air, a process called "auto-ignition." The fuel for the in-situ combustion process is a coke-like residue that deposits on the sand-grain surface area during the process. Whether the combustion can be sustained depends on the rate at which the fuel coke is formed from the original oil and the rate at which this fuel is burned. The fire front is not easy to re-establish after it is extinguished. Hence, the mechanisms by which combustion can be sustained should be fully understood. Fuel deposition depends to a great extent on the low-temperature- oxidation (LTO) reactions. Reaction between oxygen and petroleum hydrocarbons carried out at temperatures below about 300 deg. C [572 deg. F] are LTO reactions. In conventional combustion reactions of hydrocarbons, the main reaction products are water and CO, while in LTO reactions, most of the products are in the form of alcohol, aldehydes, ketones, acids, and peracids. Application of thermal analysis techniques in studying crude oil combustion is not new. Tadema seems to be the first investigator who applied differential thermal analysis (DTA) to crude oil combustion and recognized two distinct exothermic reactions. Burger and Sahuquet used DTA to study the effect of metal oxides on crude oil combustion. The effect of reservoir rock and its clay content on crude oil combustion was also studied by other investigators using the DTA technique. Bae applied both DTA and TGA techniques to a variety of crude oils and concluded that DTA and TGA techniques were useful tools in studying the effect of certain variables in fireflooding. Kinetics of Crude Oil Combustion The kinetics of the process plays a dominant role in determining whether a self-sustained burning can be established. The in-situ combustion process has been widely studied in the laboratory and in field operations. Most laboratory studies have been carried out to determine the effect of overall kinetic parameters on the combustion-tube tests as a preparation for field tests. Different mathematical models have also been developed to predict the related characteristics. Recent studies are more concerned with process kinetics. A meaningful kinetic model is obviously required to produce meaningful prediction from mathematical modeling. Evidence in the literature indicates that a meaningful kinetic model should reflect the effect of sand-grain surface area and oxygen partial pressure in addition to the matrix oil content. Coke combustion is a heterogeneous reaction. The rate constant of a heterogeneous reaction is expected to be a function of the surface area of the matrix. In addition to the sand-grain specific surface area, the oxygen partial pressure of the flowing gas can also be quite important. This effect is felt not only in the kinetics of crude oil combustion but also indirectly in the economic evaluation of the in-situ combustion project through the cost of the air compression. The oxygen-enriched fireflood process is a manifestation of this effect. This study investigates the effect of different variables-such as specific surface area, oxygen partial pressure, and oil content-on the crude-oil coke combustion by TGA. Method of Approach Small samples of crude oil mixed with sand were subjected to TGA. Each variable-such as specific surface area of the sand grains, oxygen partial pressure of the purging gas, and the oil content of the sand mixture-was varied independently while the others were kept constant. Consistent with earlier findings, the DTG curves produced in the presence of sand grains revealed three major transitional stages: distillation, low-temperature combustion, and coke combustion. Each region of the DTG curve was subjected to kinetic analysis. This paper is concerned only with the kinetics of the coke combustion. Crude oil distillation has been discussed elsewhere 16 and details of the work can be found in Ref. 17. Equipment and Materials Equipment used in this study consisted of a DuPont 951 thermogravimetric analyzer with R90 programmer and a Hewlett Packard 7046A X-Y recorder. The study was focused mainly on crude oil from the Iota field, Allen County. KS, with a gravity of 0.94 g/cm [19.3 deg. API] and a viscosity of 530 mPa s [530 cp] at 25 deg. C [77 deg. F]. However, to test the validity of the model, TGA/DTG curves were also generated for eight additional crude oil samples supplied by Air Products and Chemicals Inc., Arco Oil and Gas Co. Inc., and Esso Resources Canada Ltd. These crude oils were from different locations of the U.S., Canada, and Latin America. Their exact locations were withheld by the suppliers. Table 1 summarizes the granular materials used in this study and their specific surface areas. The sample weight for all the TGA runs ranged from 10 to 50 mg. Sample size depended on crude-oil/sand-grain mixture. Larger sand grains required larger sample sizes to ensure the sample representation of the mixture. The kinetic data generated proved to be independent of the sample size in the range studied. Crude oil content was varied between 10 and 58 wt %. Oil content below 10 wt% did not produce a reasonable TGA/DTG curve simply because of the small total weight loss. Oil contents greater than 58 wt% produced DTG curves similar to those of crude oil in the absence of solid grain materials; i.e., surface-area effect was not felt because of the large oil content. Commercial-grade oxygen and nitrogen were used to reach the required oxygen concentration. The two gases were blended with a gas proportioner. The oxygen concentration was determined with an oxygen analyzer. The oxygen partial pressure of the purging gas was varied between 5 and 50 kPa [0.05 and 0.50 atm]. The lower limit was established by the lack of significant exothermic reactions. SPERE P. 201^