The use of a jet-stirred Continuously Stirred Tank Reactor (CSTR) to study the homogeneous gas phase partial oxidation of methane to methanol

Direct partial oxidation of methane to methanol is a process that has the potential to reduce the energy use and the cost of methanol prodcution by eliminating synthesis gas generation as an intermediate step. This process has been the subject of considerable study, as manifested by the scope of work reported in a number of recent reviews [1-3]. Most of the studies reported have been carried out in tubular flow reactors that were empty or contained inert packing. To compliment the experimental studies, theoretical work has been carried out, largely with a view to obtaining a suitably tractable model for computer experiments to be performed, at far less cost and over a greater range of parameter space [1]. It has been pointed out that non-isothermal modelling of tubular reactor systems is not possible without assumptions, as spatial gradients exist and the required transport properties such as diffusion coefficients and thermal conductivities of the numerous transient species involved in the most comprehensive methane oxidation mechanism are not known. However, modelling the reaction in a CSTR is feasible, as spacial effects are absent and the steady state, oscillation and bifurcation behaviour can be clearly seen without physical complications [4]. While the use of a continuously stirred tank reactor (CSTR) for this reaction at high pressure has been suggested by other workers [5], to the best of our knowledge, only one experimental study using a CSTR for this reaction, at pressure, has been reported [6]. However, the equipment used in this study (a commercial, mechanically stirred, cylindical autoclave with a volume of 240ml) proved less than ideal, due to the substantial thermal resistance of the metal vessel, the glass liner and pocket of air between the two, the uneven heating outside the vessel, and the dissipation of heat along the stirrer shaft. These factors made it impossible to measure the heat transfer coefficient of this vessel, essential for modelling the reaction system. In our desire to validate the nonisothermal CSTR model experimentally, these factors and the desire to use shorter residence times required by the model, led to the use of a cylindical jet stirred CSTR, based on the design of David, Houzelot and Villermaux [7]. In this paper, we report the design and validation of the CSTR, and the results of experimental studies carried out using the reactor. Experiments were carried out using residence times of 10 to 30s, pressures ranging from 1.5 to 5.0 MPa and reactor oven (ambient) temperatures up to 450°C. Results are presented and compared with those predicted by the model, as well as those reported in previous studies.