Study On Reaction Kinetics Of Direct Coal Liquefaction By Using The Autoclave

Direct coal liquefaction (DCL) is a strategic technology to relieve the shortage of oil and to ensure the energy security in China. To improve the DCL technology in China, for the need of the industrialization of DCL key technology in the future, the state supported"973"program:"The fundamental research on direct coal liquefaction for large-scales"has been approved, and"The hydro-liquefaction kinetics of direct coal liquefaction"is one item of the program. This paper is a part of this item.Here the hydro-liquefaction kinetics of Shenhua coal is studied by using the autoclave.First the investigation of the condition of the 500 ml horizontal vibrated autoclave available in the laboratory was carried out to study the kinetics. Study shows that the reaction pressure is linear with the initial H2 pressure; when the initial pressure is 8.4 MPa, the reaction pressure is 19.0 MPa, which is of the same pressure with the continuous system. The conversion is increased with the elevating of the initial pressure.The customary batch operation of autoclave is not suitable to study the reaction kinetics, so the flash heating can effectively decrease the heating time to ensure the accuracy of heating, and the heating time is reduced to 48 min from 80 min. Two sets of data are obtained in batch operation at 430℃and 450℃. Two kinetic modes are established to simulate the data. The modelⅠsuggests that the coal can be consisted of three components: the high reactive component that can directly be converted to oil, asphaltene, preasphaltene and gas; the low reactive component that can only be converted to asphaltene and preasphaltene; and unreacted one can not reacted in the conditions. The result shows that the reaction rate constant of the high reactive one is 12.5 times of the lower one, which matches with the assumption. Because of the low rate of heating, in the conditions of different temperature, the beginning of the reactants are different, so that the related parameters are incomparable. Therefore the modelⅠis not appropriate to study the kinetics in this vibrated autoclave. ModelⅡsuggested that the coal is divided into the reacted and the unreacted according to the reactivity, and the former is composed with some components which react in first order. This reasonable assumption is accord with the configuration of the coal and more scientific. In comparison, the latter is much more accurate although both regressions are quite approximate. To reduce the heating time and to simulate the conditions of the demonstration project much better, a bench-scale continuous autoclave direct coal liquefaction unit is built up. The 2 L reactor is an autoclave which feeds from bottom, and discharge from top. The level of high temperature separator is controlled by the weight of the feed in the separator. The gas holdup of the autoclave in high temperature is measured, which is linear with the gas flow rate. In limited time, three steady runs, which are at 410℃,420℃,430℃with a fixed feed and gas flow rate, are carried out and the yields of different products are obtained.At 430℃, the stirred reactor is modeled as a continuous stirred-tank reactor (CSTR); the conversion is simulated by the discrete model. The average conversion of CSTR is the product of the residence time distribution and the conversion of the batch reactor. The simulated conversion at 430℃is 79.19%, is some different from that of the continuous system, 77.08%; however, the simulation results of the microfluid model of asphaltene and oil yields are out of accord with the real ones. This disaccord is ascribed to the slow heating rate of the batch autoclave experiment.