Separation of alkane / alkene gaseous mixtures by adsorption unto microporous carbons

Gases are strongly attracted to surfaces although not all in the same way. The difference in chemical composition of both the gases and the surfaces upon which they adsorb will determine the actual affinity for the adsorption of a gas unto a surface. Carbons, both natural, and man-made, may be appropriately modified (by activation) to have a relatively large available surface in the form of microscopic pores. The shape and size distribution of these pores along with the final nature of the surfaces after activation will ultimately determine the adsorption characteristics of a given gas in a pore. With mixtures, the problem becomes more complicated, since now other factors come into play such as the interactions amongst the gases and their preferential adsorption on the surface. This work porposes to study a particular type of mixture made up of small hydrocarbons (alkanes) and their unsaturated counterparts (alkenes). The alkenes are particularly useful as starting points for polymerization reactions which ultimately lead to the production of plastics and petrochemicals. However, in may scenarios, it is necessary to obtain them in pure form before such reactions take place. The conventional separation technique is distillation. While feasable, distillation involves an large energetic and technological effort for these type of mixtures. Other technologies, such as the use of membranes achieve low selectivies (poor separation) and/or present other technical problems This work focuses on the proposal that adsorption on carbons may be a feasable, economical and green alternative. The choice of the appropriate adsorbent is, however, the key to a sucessful separation. Only limited information can be obtained from experiments, since activated carbons are dificult to characterize and the experiments can not focus on a single variable at a time (e.g. the results are hard to interpret). Theoretical developments are still incipient and do not allow much more than the description of simpler systems. We propose to study this problem by modelling, from a molecular point of view and using the appropriate statistical and computational tools, the adsorption of ethylene/ethane and propylene/propane mixtures unto well-defined carbon slit pores. Since we are performing molecular modelling on a computer, we may vary at will the pore size, pressure, composition and temperatures to obtain the conditions at which the separation is maximized. This information can be used as a starting point for the synthesis and design of new and high-performance adsorbents.