The aim of this work was to develop innovative membranes able to separate carbon dioxide and eventually other undesirable compounds from raw biogas. Such membranes should be stable in an aggressive environment and resistant to humidity present in biogas. Therefore, three completely different types of membranes were investigated in this work, namely a supported ionic liquid membrane, a water condensing membrane [1] and a water-swollen thin film composite membrane [2]. This work deals mainly with a model mixture, but raw biogas taken from a sewage plant was used to complete the results of the work. Biogas is produced by anaerobic digestion of organic waste, and consists mainly of methane, carbon dioxide, and a small amount of corrosive gases (water vapor, hydrogen sulfide, ammonia, and mercaptanes). The methane present gives biogas the potential to become an alternative source to classical fuels. Unfortunately, the composition of biogas, typically 50-70 vol% methane and 30-50 vol% carbon dioxide, depends on its origin and on the season. Consequently, it is most commonly used in ancillary combined heat and power plants connected to biogas sources, such as farms or sewage plants, where a change in the composition of biogas is not a problem. For use as a fuel, the best source of biogas is that produced in sewage plants, because it has generally the highest methane content and it is easily accessible. Many different methods have been studied to purify biogas to engine-fuel quality. Water scrubbing, polyethylene glycol scrubbing, or molecular sieves are used to remove carbon dioxide. Pressure-swing absorption is also very common. Hydrogen sulfide, which is problematic because of its corrosive effect, is captured on impregnated active coal or by absorption. Membrane separation represents the latest approach to biogas purification. Polymeric membranes made of silicone rubber [3] and cellulose acetate have already been described [4]. Polyimide membranes [5,6] are very popular and polyether block amide membranes have also been tested [7]. Most of these membranes are effective for CH4/CO2 separation, but the majority cannot be used for biogas purification because they are damaged by aggressive gases. Nevertheless, they have already been applied for inert gases [8]. A promising novel class of gas separation membranes is represented by ionic liquid membranes. Their main advantages are high fluxes through the membranes and a very good selectivity [9]. Many different ionic liquids have been used to separate methane from carbon dioxide [10-11] and their effectiveness has been proved. However, ionic liquids appear to be too expensive for biogas treatment on an industrial scale. Recently we have proposed a new method of membrane separation, using a so-called "condensing-liquid membrane" [1]. This type of membrane has a significant advantage over the usual liquid membrane. Unwanted and toxic gases are removed from its continuously refreshed surface with condensed water to avoid contamination of the perm-selective membrane; furthermore, condensed water passing through the membrane ensures selectivity of the whole separation. The method is in fact based on a liquid (water in this case), condensing on a hydrophilic membrane as a result of the temperature difference of the membrane and the water-saturated biogas feed. The feed gas mixture is saturated by water vapor. The membrane has to be cool enough to make the liquid condense on the surface. Various operational conditions were followed and their effect on the separation of methane from unwanted gases was monitored. Another type of membrane based on a similar principle is the swollen hydrophilic thin film composite membrane. The condensing water on the membrane or on the swollen hydrophilic thin film composite membrane creates a separation barrier, which separates polar gasses (CO2, H2S) and CH4 from biogas on the basis of their higher solubility in water. In order to achieve spontaneous condensation of water, the membrane temperature must be below the dew point of the biogas feed. The contact of the membrane surface with water causes swelling of the polyamide thin film in the case of composite membranes. During the impregnation, the porous support is also saturated with water, but this water tends to evaporate during the experiment. Thus, the membrane has to be cooled down enough to make the liquid condense on its surface. A binary mixture was used to see the performance of the membrane treated by the new method. The upstream pressure was kept at 500kPa to achieve the highest possible permeation flux (within apparatus limits).. Preliminary tests revealed that the permeation fluxes of methane and carbon dioxide in raw biogas increase with increasing temperature. At a pressure difference of 400kPa on the reverse osmosis membrane (a swollen skin layer on a dry support) is possible to obtain from raw biogas even more than 95 vol. % of CH4 in retentate stream (Fig. 1.). Fig. 1. Dependence of CH4 and CO2 concentration in retentate during raw biogas separation with water condensing on reverse osmosis membrane. The two plateaus correspond to partial and complete wetting of the membrane surface, where the highest selectivity is reached in the completely wetted membrane. In conclusion we propose a new effective method for upgrading of raw biogas to the same quality as that of fuel standard natural gas, based on membrane separation processes. In our single stage method with a temperature below the dew point of the raw biogas feed, condensing water on the swollen hydrophilic thin film composite reverse osmosis membrane promotes the formation of a very thin selective water layer. The significant difference in solubility and permeability of methane and of raw biogas impurities (carbon dioxide, hydrogen sulfide) in and through the water layer results in an effective CO2/CH4 separation. The presented work represents an innovative approach to enable relatively inexpensive production of biomethane from sewage biogas. All our data were compared with the supported ionic liquid membrane and with literature data in Robeson plot. References 1.M. Poloncarzová, J. Vejrazka, V. Veselý, P. Izák, Angewandte Chemie Int. Ed., 50 (2011) 669-671. 2.M. Kárászová, J. Vejrazka, V. Veselý, K. Friess, A. Randová, V. Hejtmánek, L. Brabec, P. Izák, Sep. Pur., in press. 3.F. Wu, L. Li, X. Zhihong, T. Shujuan, Z. Zhibing, Chem. Eng. J. 2006, 117, 51-59. 4.H. M. Ettouney, G. Al-Enezi, S. E. M. Hamam, R. Hughest, Gas Sep. Purif. 1994, 8, 31-38. 5.J. Zhang, J. Lu,W. Liu, Q. Xue, Thin Solid Films 1999, 340, 106-109. 6.J. D. Wind, D. R. Paul, W. J. Koros, J. Membr. Sci. 2004, 228, 227-236. 7.S. Sridhar, R. Suryamurali, B. Smitha, T. Aminabhavi, Colloids Surf. A 2007, 297, 267-274. 8.F. F. Krull, C. Fritzmann, J. Membr. Sci. 2008, 325, 509-519. 9.A. Corti, D. Fiaschi, L. Lombardi, Energy 2004, 29, 2025-2043. 10.X. Hua, T. Jianbin, A. Blasig, S. Youqing, M. Radosz, J. Membr. Sci. 2006, 281, 130-138. 11.J. E. Bara, C. J. Gabriel, E. S. Hatakeyamaa, T. K. Carlisle, J. Membr. Sci. 2008, 321, 3-7. 12.P. Scovazzo, D. Havard, M. McShea, S. Mixon, D. Morgan, J. Membr. Sci. 2009, 327, 41-48.