(334b) Ionic Liquids as Sustainable Extractants in Petrchemicals Processing

Introduction Only a few ionic liquids can be successfully applied for the separation of aromatic hydrocarbons from mixtures of aromatic and aliphatic hydrocarbons. Ionic liquids have been identified as promising solvents to replace conventional solvents in liquid-liquid extraction of aromatic hydrocarbons requiring less process steps and less energy consumption, provided that the mass based aromatic distribution coefficient and/or the aromatic/aliphatic selectivity are higher than those of the current state-of-the-art solvents such as sulfolane. Only a small number of ionic liquids is able to combine higher mass based distribution coefficients with selectivities comparable or higher to that of the benchmark sulfolane. The most suitable ionic liquids from our evaluation are [3-mebupy]B(CN)₄, [3-mebupy]C(CN)₃, [3-mebupy]N(CN)₂ and [bmim]C(CN)₃. The aromatic distribution coefficients of these ILs are a factor of 1.2 to 2.5 higher and the aromatic/aliphatic selectivities are up to a factor of 1.9 higher than with sulfolane. Pilot Plant Experiments The results from laboratory equilibrium experiments have been validated with two ionic liquids ([4-mebupy]BF₄ and [3-mebupy]N(CN)₂) and sulfolane in our 6 m high pilot plant extraction column. Successful introduction of RTILs into extraction operations also requires knowledge on their hydrodynamic and mass transfer characteristics, because the viscosity and the density of ILs are usually higher than those of conventional solvents. Common extraction contactors may not be suitable for separation processes with ILs as extractants. Therefore, mechanical energy has to be used to enhance mass transfer into ILs. Hydrodynamics (drop size, hold-up and operational window) and mass transfer efficiency determine the column performance in an extraction process. The solvent (ionic liquid or sulfolane) was the dispersed phase and was fed to the top of the column and the extract phase was collected from the bottom settler. The heptane/toluene phase was fed from the bottom and the raffinate phase was collected from the top settler. Regenerated ionic liquid was used in the extraction experiments. From the concentration profiles of toluene over the column length can be concluded that with sulfolane as solvent more toluene is extracted than with [4-mebupy]BF₄, although the distribution ratios of toluene in sulfolane and [4-mebupy]BF₄ are comparable on weight basis (0.22 kg/kg for [4-mebupy]BF₄ vs. 0.26 kg/kg for sulfolane). The IL [3-mebupy]N(CN)₂ shows a slightly higher mass transfer of toluene than sulfolane. The raffinate phase with this IL contains less toluene than with sulfolane and the extract phase with the IL has a higher toluene content than with sulfolane. The performance regarding the mass transfer of toluene of the regenerated IL after several regenerations of the ionic liquid is increased, the reason yet unknown. The IL [4-mebupy]BF₄ is a better solvent than sulfolane from a hydrodynamic point of view, because the operational region is larger. However, based on mass transfer characteristics, it can be concluded that sulfolane outperforms [4-mebupy]BF₄. This is due to the large difference in viscosity of these two solvents. The IL [3-mebupy]N(CN)₂ shows a slightly better mass transfer of toluene than sulfolane. Next to the toluene/heptane separation, also experiments with model FCC, LCCS and diesel feeds were carried out. In addition, refinery streams from the BP refinery in Rotterdam were used. The performance with the actual refinery stream as feed is comparable with that of the model feeds. The removal of benzene and toluene from the actual refinery FCC stream was 81 and 71% respectively, and from the model feed 89 and 75%, respectively. Ionic liquids can be suitable extractants for several compounds, as is proven in batch equilibrium experiments. Presently, it is also shown that a continuous extraction using ionic liquids in a pilot plant produces good results. The ionic liquid [3-mebupy]N(CN)₂ has been found to be the best ionic liquid in terms of mass based capacity and can outperform sulfolane for the separation of toluene/n-heptane. Economic evaluation An economic evaluation was made for the separation of aromatic compounds from the feed of a naphtha cracker with the IL [4-mebupy]BF₄, extended to other ILs ([emim]C₂H₅SO₄, [4-mebupy]CH₃SO₄, [bmim]AlCl₄, [emim]AlCl₄ and [3-mebupy]N(CN)₂) and compared to that with sulfolane. The separation of toluene from a mixed toluene/n-heptane stream was used to model the aromatic/aliphatic separation. Most ethylene cracker feeds contain 10 to 25% of aromatic components, depending on the source of the feed (naphtha or gas condensate). The aromatic compounds are not converted to olefins and even small amounts are formed during the cracking process in the cracker furnaces. Therefore, they occupy a part of the capacity of the furnaces and they put an extra load on the separation section of the stream containing C5 ? C10 aliphatic compounds. If a major part of the aromatic compounds present in the feed to the crackers could be separated upstream of the furnaces, it would offer several advantages: higher capacity, higher thermal efficiency, and less fouling. The improved margin will be around ? 20/ton of feed or ? 48 million per year for a naphtha cracker with a feed capacity of 300 ton/h, due to lower operational costs. For a naphtha feed of 300 t/h containing about 10% aromatic hydrocarbons, the total investment costs in the sulfolane extraction were estimated by UOP, supplier of the process, to be about M? 86 and with [4-mebupy]BF4 about M? 56, including an IL inventory of M? 20. In the calculations, an ionic liquid price of ? 20/kg was used and BASF, a major producer of imidazole, one of the primary products for ionic liquids, has indicated that it is indeed possible to reach a level of ? 10 ? 25/kg with production on a large scale. From a confidential process scheme for an extraction of aromatic hydrocarbons from a reformate feed, containing about 82 wt% aromatics, with sulfolane, the sizes of the major equipment, the heat load of the heaters, coolers, heat exchangers, reboilers, condensers, etc., the investment costs and the scale-up/scale down capacity exponents were obtained. From these data, it is apparent that the most expensive equipment in the sulfolane process is not the columns, vessels etc., but the heat exchangers, reboilers and coolers, which make up almost 65 % of the total investment. Sulfolane has to be regenerated with high pressure steam, while the regeneration of ILs requires low pressure steam. The investment and annual costs for the separation of 10% aromatics from a cracker feed with sulfolane and several ionic liquids are shown in the figure. Figure. Investment and variable costs for extraction with ionic liquids. 1 [emim]C₂H₅SO₄, 2 [mebupy]BF₄, 3 [mebupy]CH₃SO₄, 4 [3-mebupy]N(CN)₂, 5 [bmim]AlCl₄, 6 [emim]AlCl₄. Data about the sulfolane process from UOP. A higher aromatic/aliphatic selectivity means a purer aromatic product, less extraction of aliphatics and a lower number of extraction stages. A large aromatic distribution coefficient means a lower S/F ratio with, consequently, smaller extraction and regeneration units with lower investment costs, including a smaller IL inventory, and lower energy costs. Higher mass based aromatic distribution coefficients than 0.7 hardly lead to lower costs, as can be seen in Figure 4. At a toluene distribution coefficient of about 0.6 to 0.7 kg/kg, the investment costs can be reduced to about M? 25 ? 30 and the annual costs will be in the range of M? 16 ? 18/year. Therefore, suitable ILs for the extraction of aromatics from mixed aromatic/aliphatic streams must have a high aromatic distribution coefficient, higher than 0.3 kg/kg, and an aromatic/aliphatic selectivity of at least that of sulfolane. Conclusions Ionic liquids can replace conventional solvents in liquid-liquid extraction of aromatic hydrocarbons, provided the mass based aromatic distribution coefficient and/or the aromatic/aliphatic selectivity are higher than those with sulfolane. The main conclusion of the process evaluation is that ILs that show a high aromatic distribution coefficient with a reasonable aromatic/aliphatic selectivity could reduce the investment costs of the aromatic/aliphatic separation by a factor of two. The best ILs for the separation of aromatic and aliphatic hydrocarbons are [3-mebupy]B(CN)₄, [3-mebupy]C(CN)₃, [3-mebupy]N(CN)₂ and [bmim]C(CN)₃, which all show a factor of 1.2 to 2.5 higher aromatic distribution coefficients and aromatic/aliphatic selectivities up to a factor of 1.9 higher than with sulfolane. Based on the performed analysis, it can be concluded that industrial application of ILs for aromatics extraction has not materialized yet, because most of the reported ILs do not provide higher mass based aromatic distribution coefficients and/or higher aromatic/aliphatic selectivities than those achieved by conventional solvents such as sulfolane.