Olefin Recovery from Chemical Industry Waste Streams

Abstract

For a period of four years a f e r acceptance of all items tu be delivered under this contract, the Goverrirrier~t agrees to use these data for Government purposes only. and they shall not be disclosed outside tlie Governtnerit (including disclosure for procurernent purposes) during srtrli period witiiout permission of the Contractor, e-rcept that, subject to the foregoing use and disclosure proltibifions, srtcli data may be disclosed for use by sirpport contractors. Afer the aforesaid 4-year period, tlw Governrnent has a royalty-free license to use, and to autliori:e others to use on its beha& these data for Governrnent purposes, but is relieved of all disclosure proliibitioris and assurnes no liability for unarttliorized use of these data by third parties. This Notice shall be affied to ariy reproductions of tliese data, in whole or in part. EXECUTIVE SUMMARYThe objective of this project was to develop a membrane process to separate olefins from paraffins in waste gas streams as an alternative to flaring or distillation. Flaring these streams wastes their chemical feedstock value; distillation is energy and capital cost intensive, particularly for small waste streams.We have discovered novel, robust, plasticization-resistant perfluoropolymer membranes that selectively permeate the olefin component of olefidparaffin gas mixtures. These membranes are made from perfluoropolymers such as Hyflon@ AD (Solvay Solexis, Thorofare, NJ), the copolymer of tetrafluoroethylene and 2,2,4-trifluoro-5-trifluoromethoxy-1,3-dioxole, and CytopTM (Asahi Glass, Japan), poly(perfluorobuteny1 vinyl ether). The configuration of these membranes was optimized and the membranes were successfully scaled up in our commercial casting machine. Defect-free threeand four-inch spiral-wound modules were prepared. Membrane performance was evaluated in bench-scale membrane stamp and membrane module tests. These parametric tests showed that the optimized membranes had consistently good laboratory performance at commercially relevant operating conditions. In contrast, membranes made from other materials did not perform well under these conditions. Perfluoropolymer membrane modules were evaluated at three plants. The first field test was performed at a Chevron Phillips slurry reactor polypropylene plant in Pasadena, TX. The membrane system recovered propylene from a reactor purge stream that containedpropane and other inerts. The second and third field tests, performed at Equistar ethylene crackers in Channelview, TX and Chocolate Bayou, TX, separated propylene from propane upstream of a C,-splitter. The results obtained in the first two field tests were used to optimize the configuration of the membrane. An optimized Cytop-based membrane was used in the third field test. This membrane maintained a propylene/propane selectivity of 3 during 12 weeks of operation, but selectivity dropped to 2 in the last two weeks of the field test.Our economic analysis showed that membranes with a propylene/propane selectivity of at least 3 and propylene pressure-norm...

Modeling of the Separation of Propene/Propane Mixtures by Permeation through Membranes in a Polymerization System

“…This is the main reason the separation of propene/propane mixtures is usually carried out by cryogenic distillation. However, this separation method is highly energy demanding; very low temperatures and high pressures are needed to accomplish a reasonable separation, which results in high operational costs. − Because the boiling points of propene and propane are so close, large distillation columns, containing up to 180 plates, are needed, and high reflux rates are demanded . Therefore, there exists great interest in the development of more economical propene/propane separation processes .…”

“…Only recently has the separation of mixtures containing olefins and paraffins using polymeric membranes gained importance, and the available information reported on this subject is steadily increasing. Most of the studies, however, deal with the determination of diffusivity and solubility data of the gases in various polymer matrixes, − ,− with the exception of the work of the work of Burns and Koros, where upper bounds for selectivity/permeability are calculated from a mathematical model and compared to the available literature data.…”