New Kinetic Models for Thermal Cracking of Crude Oils in In-Situ Combustion Processes

Abstract

Summary. This paper presents two kinetic models for representing the thermal cracking of crude oils, which incorporate the cracking rate parameters and stoichiometric coefficients to correlate experimental data. parameters and stoichiometric coefficients to correlate experimental data. The models presented show that the first-order kinetics generally accepted for pure components are unsatisfactory for multicomponent systems characterized by pseudocomponents. We conclude that three corrections to the existing first-order model are needed for modeling thermal cracking of mixtures. First, the apparent reaction order is always greater than one. Second, the reaction order is a decreasing function of temperature. Third, coke may also be formed from intermediate products. These corrections are incorporated into the models. In the first model, crude oil is split into two pseudocomponents, while in the second model, crude oil is represented by three pseudocomponents. The models can be easily extended to any number of pseudocomponents. The models can be easily extended to any number of pseudocomponents. pseudocomponents. The models successfully correlated experimental data of four systems available in the literature. Furthermore, it was confirmed that coke is not always the same source of the fuel burned in an in-situ combustion process. process. Introduction It is generally believed that in in-situ combustion processes, the combustion zone is preceded by a cracking or processes, the combustion zone is preceded by a cracking or superheated steam zone where coke is formed from the thermal cracking (pyrolysis) of crude oil. The kinetics of the cracking reaction may be a crucial process mechanism affecting the performance of combustion processes because it not only produces solid-like coke for combustion but also upgrades the remaining oil, which affects the vaporization behavior. As a result, the cracking reaction will strongly influence the total amount of fuel available in the combustion zone. The effects of cracking reactions on the fuel deposition mechanism and the fuel composition have been discussed elsewhere. The reaction mechanisms of hydrocarbon cracking are very complex. Even for a pure component, it is almost impossible to describe the mechanism precisely. Nevertheless, it is possible to use simple global rate expressions to represent the reaction rate of the reactant. For pure hydrocarbons, it is well established that the cracking reaction can be properly modeled by a first-order rate expression, although self-inhibition (decreasing first-order rate constant with increasing conversion) was generally observed. This is caused by the formation of olefins, which are known to be good inhibitors of free-radical reactions. As a general rule, the reaction rate constant for normal paraffins increases with increasing carbon number. while the activation energy decreases with increasing carbon number. Global rate expressions were also applied to the pyrolysis of gas oil and crude oils by use of first-order pyrolysis of gas oil and crude oils by use of first-order kinetics. Most of these studies lumped the multicomponent mixture into one oil component, while McNab et al. assumed that only the heavy-oil fraction contributes to the cracking reaction and arbitrarily chose 80% of the residue of the distillation of the original crude as the heavy-oil component. This two-pseudocomponent approach was adopted by Henderson and Weber. In all the above studies, only the reaction rates of the crude oil components were considered, and no attempt was made to correlate the stoichiometry of the reaction. In an attempt to match the product distribution of Athabasca bitumen pyrolysis, Hayashitani et al. constructed a number of complex kinetic models. The oil was divided into three to five components, and six to eight first-order reactions were included in each model. They found that these complex models cannot satisfactorily correlate all aspects of experimental behavior of cracking reactions. They concluded that during the course of cracking reactions, each pseudocomponent might have changed its characteristics, A comprehensive reaction scheme for catalytic cracking of gas oils was recently developed by Jacob et al., who simulated the gas-oil cracking using four pseudocomponents (heavy oil, light oil, gasoline, and gas plus coke). pseudocomponents (heavy oil, light oil, gasoline, and gas plus coke). For each of the light- and the heavy-oil components, the pseudocomponent was further split into paraffinic, pseudocomponent was further split into paraffinic, naphthenic, aromatic, and aromatic substitute groups. As a result, the reaction scheme involves 10 species (components) and 20 first-order reactions. The Arrhenius constants and the activation energies for these reactions were assumed to be universal constants (i.e., independent of gas-oil composition), This model was claimed to work well for a wide spectrum of gas oils. However, the applicability of this model to heavy oils with a wider range of composition remains to be proved. P. 54

Recovery of light oil by air injection at medium temperature: Experiments

“…There is a large body of literature describing the use of HPAI to recover oil [1,2,10,11,34,37,38,39], which is appropriate at high pressures (>100 bars) in reservoirs at large depths. However, at shallower depths (300 ∼ 1000 m), an alternative is to inject at medium pressures (10 ∼ 90 bars) for light and medium oil in heterogenous low permeability reservoirs.…”

“…However, it is suggested [47] that only about 20% of the light oils are good candidates for undergoing full oxidation at temperatures greater than 400 • C, and that the other 80% only incorporates the oxygen in the oil by low temperature oxidation (LTO) at lower temperatures. The mechanism for combustion of light or medium oils is fundamentally different from that for the combustion of heavy oils [1,2,9,10,11,34,37,38,46,30], in which high temperatures are achieved. In the high temperature oxidation (HTO), the heat conducted out of the reaction zone converts the oil to coke before it is combusted.…”

“…At initiation of the experiment the nitrogen is replaced by air and the combustion process starts [1,2,9,10,11,34,37,38,46]. For ramp temperature oxidation tests, such as the one we performed, the reactor is heated over the whole length of the tube.…”

Diffusive Effects on Recovery of Light Oil by Medium Temperature Oxidation

“…It is the purpose of this work to determine the effects of capillary pressure diffusion, longitudinal heat conduction, and mass diffusion on the combustion recovery process. Combustion for low and medium viscosity oil (Abou-Kassem et al 1986;Akin et al 2000;Bruining et al 2009;Brigham 1997, 2003;Kok and Karacan 2000;Lin et al 1987Lin et al , 1984Xu et al 2004) is described by different mechanisms. For light oil combustion, coke formation is usually disregarded, although it can occur (Khoshnevis Gargar et al 2010).…”

Recovery of Light Oil by Air Injection

“…There is a large body of literature describing the use of HPAI (high pressure > 100 bars air injection) to recover oil [1,3,11,13,14,36,39,40,56,59]. HPAI was first introduced in 1979 in the Buffalo field [17].…”