Methanation of biosyngas in a fluidized bed reactor

Doctoral thesis, 2006

Rising prices for fossil fuels and concerns over the climate change increase the interest in renewable energies. Synthetic natural gas (SNG) can be an option, filling the request for alternatives in the transportation sector, where the dependence on the oil is very strong. The advantages of synthetic natural gas (SNG) as fuel are the high energy density, the clean combustion and the already existing natural gas distribution grid. The perspective for a realization in the near future of an industrial size SNG production from biomass is promising as already a sound scientific basis for the methane synthesis exists. In the 70`s and 80`s of the last century methanation of synthesis gas was investigated intensely. At that time coal gasification and synthesis of SNG was considered as a possibility to increase the security of energy supply. With the participation of PSI, a project team was set up that has the vision to realize a first 20 MW plant converting “Wood to SNG” until the end of this decade. A systems analysis identified key issues, for the optimization of the combination of gasification technology and fuel synthesis. Calculations using ASPENPLUS®, furthermore, showed the expected chemical efficiency from wood to SNG to be in the range of 60 to 65% and determined promising operation conditions for the experimental campaigns. State of the art for the methane synthesis from producer gas is a multistage fixed bed process. To control the reaction conditions and to avoid catalyst deactivation by carbon deposition, tar reforming, water gas shift and methanation reaction are performed in different operation units. Goal of the technology development is an efficient synthesis process for SNG from biogenic producer gases, which is economically attractive. From literature it is known, that fluidization of the catalyst has a regenerating effect, limiting the carbon deposition. Based on fluidized bed technology, a once through SNG synthesis concept was developed incorporating tar reforming, water gas shift and methanation in a single fluidized bed. To proof the reactor concept, a lab scale fluidized bed methanation system was constructed. After several preliminary tests, an experimental campaign was performed with a slipstream of the FICFB-gasifier (Fast Internally Circulating Fluidized Bed) in Güssing (Austria) in 2003. The producer gas of this industrial 8 MWth gasification power plant with a H2/CO ratio of 1.4 and a content of light tars (Benzene, Toluene, Naphthalene, etc.) in the order of several g/m3 is used to run a gas engine which has an output of 2 MWel. As the producer gas is nearly nitrogen-free but methane-rich it is also considered as very suitable for methanation. During this campaign, the methanation catalyst was tested using three different gas cleaning strategies. The campaign started with an additional cleanup, consisting of an ammonia scrubber, an active carbon filter and a ZnO bed for desulphurization. Stepwise, the active carbon filter and ammonia scrubber were removed form the cleaning system. At the end of the campaign the gas cleaning for the methanation catalyst consisted only of the existing product gas cleaning of the power plant and a ZnO bed. During this campaign the catalyst was in operation for 120 h. Under the final conditions without additional cleaning, more than 98% CO conversion and 99% tar conversion was achieved resulting in an overall chemical efficiency of 63%. The experiences gained in this campaign and the promising results set the basis for the construction of a 10 kW mini pilot. The new test rig was designed for unmanned operation, needed to perform on-site long-duration experiments. The setup was commissioned in summer 2004 and tested with synthetic product gas for promising operation conditions. Finally, a 100 h test was performed as a preparation for the on-site experiments in Guessing. In September 2004, the setup was transferred to Guessing and linked to the FICFB gasifier. Until the end of 2004, two 200 h long duration experiments in Guessing were performed. Results from the campaign 2003 could successfully be reproduced. A second proof of concept could be done. The results are encouraging even though the catalyst deactivated after 200 hours, probably due to a combination of carbon deposition and sulphur poisoning by organic sulphur species. In the last part of the thesis, we concentrated on the carbon exchange processes in the fluidized bed, facilitating the methanation with integrated WGS and reforming, and limiting the rate carbon deposition. By means of in-situ measurements of the axial gas phase concentration profiles with a moveable sampling probe, strong carbon exchange processes between the catalyst and the gas phase were shown. These exchange processes structure the bed into three zones: Carbon deposition, predominantly by CO dissociation, at the inlet; predominant gasification of solid carbon species from the catalyst in the following zone, and predominant carbon deposition by methane dissociation in the upper part of the bed. By analyzing the carbon balance, locally up to 20 % excess carbon was found in the gas phase, mainly in form of methane. Due to an intensive catalyst mixing, the build-up of unreactive carbon is prevented by regeneration in the middle zone of the reactor. The result of the two campaigns (2003 and 2004) was the technical basis for an EU project within the 6th Framework program of DG-TREN. Within this project a 1 MW SNG process development unit (PDU) will be erected and operated. The project started May 2006.