Syngas Utilization through High Pressure Solvent Assisted Fischer-Tropsch Synthesis

Currently, the high oil price combined with a low gas price presents a good opportunity for the development of new gas-related markets. This opportunity is also driving a significant boost for the research and development of gas conversion technology. At Texas A&M University at Qatar we are exploring to improve the Fischer-Tropsch process through the use of supercritical solvents. In this process, we employ hydrocarbons with a suitable critical point  to assist the Fischer-Tropsch catalyst in converting synthesis gas (a mixture of hydrogen (H₂) and carbon monoxide (CO)) to liquid and gaseous hydrocarbon products¹.The main techno-economic challenge facing this process is the high pressure requirement with the solvent flow and separation. The pressure of the reactor bed which can go all the way up to 80 bar, could significantly increase pumping, materials and maintenance and other related operating costs for the GTL plant. The concomitant benefit however is that the high pressure may assist in flash separation of the solvents and the other middle distillate products of the synthesis. In addition, it was shown that the solvent assists in a more efficient heat removal through its high heat conductivity, as well as presenting a significant heat buffer for the generated heat from the reaction exotherm through the increased heat capacity of the medium. This results in a much more even temperature distribution profile throughout  the bed², which contributes to longer catalyst lifetimes and improved control of the reaction parameters, allowing higher overall reactor temperatures. The solvent also assists in improved removal of waxes³ surrounding the catalyst particles and may even aide in removal of waxes from the catalyst pores. This should lead to higher gas diffusion rates through the catalyst particle and thus higher catalyst particle effectiveness. Furthermore, solvent properties are highly tunable around the critical point, providing another tool for control during operation. Of further interest, one additional benefit has been reported where the supercritical solvent assists in generating a more active phase of Cobalt (hcp) for the Fischer-Tropsch synthesis.⁴

Modeling

We are currently exploring the use supercritical solvents in the Fischer-Tropsch reaction using a two sided approach. One the one hand we are investigating and developing models to describe in detail the local conditions in the reactor bed.⁵ This includes the use of intraparticle transport equations, as well as overall 1-D reactor bed mass, heat and energy balances.

The result of using these equations is highly dependent on the choice of the correct material properties (eg. thermal conductivity and effective diffusivity), about which not much is known for these systems. In addition, the high pressures and the non-ideality of the reactor medium requires use of EoS calculations in order to find information about the local concentrations and about the phase behavior. Used methodology include the Peng-Robinson and Soave-Redlich-Kwong equations of state. Although they are not perfect, they can provide significant insight, especially in the inlet conditions of the reactor bed, where no water is present. The intraparticle diffusion calculations lead to insight to the overall effectiveness of the particle, through inspection of the concentration gradients within the particle. In our method we interlink all calculations to arrive at overall bed behavior by taking into account the effectiveness of particles at local conditions. Preliminary modeling results using kinetic expressions from the literature⁶ have shown that even though the use of supercritical fluids does not lead to higher conversions per se, it does lead to a much lower exotherm

Despite that the average temperature in the gas-phase reactor bed is much higher, the overall conversion profile remains comparable. Comparison of this data indicates that this is due to the improved catalyst particle effectiveness in the supercritical phase. In our model we assume two cases, in the gas phase reaction the catalyst particle is essentially filled with wax, whereas with the solvent assisted reaction the catalyst particle is filled with mainly supercritical hexane. Current investigations are towards employing specialized EoS’ that account for fluid behavior in confined spaces (such as pores) as well as detailed MRI studies that can measure the liquid hold-up inside the pores.

Experimental efforts

The other aspect of our investigations is the use of our custom design automated reactor unit to perform experimental testing campaigns on the activity and product selectivity of our catalysts. Guided by the modeling efforts we can validate the reactor bed behavior and analyse the outcomes in order to improve our modeling. We are currently quantitatively assessing the difference between supercritical, solvent assisted and regular gas phase Fischer-Tropsch reactions at 80 bars (20 bar for the gas phase reaction without solvent) to establish the effect of solvent and total pressure on the activity whilst keeping residence time and partial pressures of syngas constant.

The next step is to further develop an accurate expression for the kinetics by performing a set of kinetic experiments under very low conversion conditions where we have accurately calculated the non-ideal inlet conditions through the use of the EoS calculations. Our online GC set-up is capable of complete analysis of permanent gases and hydrocarbons in a single injection, thus enabling analysis without discrepancies due to flashing processes in conventional bench scale Fischer-Tropsch units. With this we aim to move on to detailed mechanistic kinetics schemes for a more complete process description.

Using these results we hope to build a detailed kinetic expression that helps to visualize the overall reactor bed behavior for systems containing more catalyst and increased reactor bed lengths. Our goal in the end, is to be able to accurately assess the potential for scale up towards plant or pilot plant size. Here a total energy optimization including product separation, and solvent recycling will be of vital importance to the quality of the assessment.

References

1.            Yokota K, Fujimoto K. Supercritical phase Fischer-Tropsch synthesis reaction. Fuel. 1989;68:255-256.

2.            Fan L, Fujimoto K. Fischer-Tropsch synthesis in supercritical fluid: characteristics and application. Appl. Catal., A. 1999;186:343-354.

3.            Jacobs G, Chaudhari K, Sparks D, et al. Fischer-Tropsch synthesis: supercritical conversion by using a Co/Al2O3 catalyst in a fixed bed reactor. Fuel. 2003;82:1251-1260.

4.            Elbashir NO, Dutta P, Manivannan A, Seehra MS, Roberts CB. Impact of cobalt-based catalyst characteristics on the performance of conventional gas-phase and supercritical-phase Fischer–Tropsch synthesis. Applied Catalysis A: General. 2005;285(1-2):169-180.

5.            Froment GF, Bischoff KB, De Wilde J. Chemical Reactor Analysis and Design. 3rd edition ed. New York: John Wiley & Sons:; 2011.

6.            Ma WP, Jacobs G, Keogh R, Yen CH, Klettlinger JLS, Davis BH. Fischer-Tropsch Synthesis: Effect of Pt Promoter on Activity, Selectivities to Hydrocarbons and Oxygenates, and Kinetic Parameters over 15%Co/Al2O3. Acs Sym Ser. 2011;1084:127-153.