Recovery of bitumen from tar sands by a thermally coupled fluidized-bed process

Tar sands, biomass, heavy oil, oil shale, and coal are considered to be the primary contributors to the emerging synfuel industry. On a global basis, Meyer and Fulton (1982) estimate the total heavy oil and bitumen resource to be more than 4 trillion barrels, with in-place reserves estimated at 671 billion barrels of oil. These resources are enormous when compared to the known conventional light oil reserves. Bitumen deposits represent about 70% of the total world resource of bitumen and heavy oil. Approximately 71% of the total heavy oil resource is believed to be located around the Orinoco Heavy Oil Belt in eastern Venezuela, whereas the estimate by Meyer and Fulton (1982) places nearly 68% of the world bitumen resource in the western provinces of Canada. The bitumen resource exists usually in the form of oil- impregnated rock deposits, also referred to as tar sand, oil sand, or bituminous sands. Although Canada and the Soviet Union possess nearly 90% of the global tar sand resource, the 54 billion-barrel bitumen resource in the United States represents a significant fossil-fuel source. Even though tar sand deposits are known to exist in 22 of the 50 states, major deposits of 1 million barrels of in- place oil or more exist in only nine states. Of the total United States tar sand resource, over 37% is estimated to be located in the state of Utah (Ritzma, 1982). Most hydrocarbon-recovery processes for tar sands can be categorized as either in situ processes or as processes involving mined tar sands. Canada has been the site of several commercial operations for tar sand processing, consisting mostly of surface-recovery operations. However, activity has recently heightened towards the development of processes that will be both economically and environmentally feasible for the recovery of hydrocarbons from foreign and domestic tar sands. As of 1982, there were at least 60 tar-sand-related pilot plant studies being conducted in the U.S. alone (Meyer and Fulton, 1982). The different surface-extraction processes for the recovery of oil from tar sand deposits can be categorized into four general areas: solvent processes, hot water and solvent-assisted water processes, thermal processes, and other, nonrelated processes. Selection of the best surface recovery process depends on such variables as specific characteristics of the tar sand deposit, type of solids handling required, degree of pretreatment needed, physical characteristics of the tar sand, capital and operating costs, and other technical and economic factors. Several significant features of a thermal process distinguish it from the other surface-extraction processes. First, the thermal process supplies the opportunity for power cogeneration by the use of waste heat Second, the thermal process involves a partial upgrading, which, to a degree, replaces the delayed coking steps of other processes. Third, water requirements for most thermal processes are modest compared to those for water or solvent-assisted water processes. Weeks (1977) presented the first work relating to the thermal process discussed here. His equipment consisted of a two-staged fluidized-bed reactor. Coked sand from the upper bed was transferred by gravity to the lower combustion bed, while the oil products of the coking bed passed through a product recovery system. Sensible heat of the hot gasses, leaving the combustion bed and fluidizing the upper bed, provided the major portion of the energy required for bitumen pyrolysis in the upper bed. Propane, or other combustible gases, were used to supply any extra energy required for pyrolysis. Runs were made using tar sand from the Tar Sand Triangle deposit of Utah. The process was then modified by Jayakar (1979) to utilize liquid-metal heat pipes to permit more efficient thermal coupling of the two fluidized beds. A single liquid-potassium heat pipe was installed in the first laboratory unit to extract from the combustion bed the heat needed to maintain the necessary temperature for pyrolysis in the upper bed. The latest modifications of the process, completed in the study reported here, entail the design, construction, and operation of a larger unit using three liquid-metal heat pipes. Moreover, the product recovery system was modified to provide more efficient synthetic-crude collection. Addition of an air preheater and an additional fluidized-bed heat-recovery section was also completed. A data acquisition and control system was adapted to the scaled-up reactor. Runs were made with four different Utah tar sands to demonstrate the controlability and efficiency of the scaled-up process.