Reaching Higher Productivity In a Fixed-Bed Fischer-Tropsch Reactor at Laboratory Scale and Pilot Scale

It is commonly accepted that the productivity of Fischer-Tropsh reaction is one of the keys to economic efficiency of the whole XTL process. The expression for calculation of productivity is shown below in equation (1) as mass of liquid product per volume of catalytic bed in period of time (kg/(m³´hr).

                                                                                                                           (1)

where  P_FT is productivity of the bed; M_liq is weight of liquid Fischer-Tropsch product released from the reactor bed (usually accounted as sum of hydrocarbon from pentane and heavier or C₅₊); V_bed is the volume of the catalytic  bed; and t stands for the period of time necessary for the release of M_liq .

Although there are many other types of productivity definition in literature, the one shown in equation (1) is the best for understanding relation between catalyst properties and reactor performance parameters. The productivity is closely related with the catalyst activity but not equal to it. Due to severe diffusion limitations and heat transfer complications, which are typical for Fisher-Tropsch reaction, catalysts with very high activity may show quite low productivity if placed into a catalytic bed. The above-mentioned limitations and complications can be relatively easily removed if one uses a catalyst in the form of small micrometric grains diluted with thermally conductive neutral agent such as quartz sand or silicon carbide powder. The active centers of the catalyst would release a large amount of product under such conditions. However, the volume of such diluted bed V_bed  would be too large and hence P_FT would not reach any significant level as is obvious from equation (1). So this work does not consider experiments with diluted catalytic beds.

Higher productivity means smaller reactor size and, hence, lower capital investment in XTL plant. Also higher productivity means that smaller amount of the catalyst needs to be exchanged for fresh one during regular reloading.

This work reports the results of the study devoted to the development of a fixed bed with significantly increased productivity 240 to 400 kg/(m³´hr). The research includes:

a)      The mathematical modeling of the bed;

b)      Formulation of pre-conditions for higher productivity, including requirements to the catalyst, requirements to the reactor and requirements to the temperature and pressure of the process;

c)      The results of developing and laboratory testing pelletized catalyst capable of manifesting productivity higher than 240 kg/(m³´hr)  in a fixed bed;

d)     The results of designing, building and starting up a scaled-up pilot GTL unit ( ¼ bbld capacity) with a highly productive Fischer-Tropsch reactor.

The mathematical model describes stationary process occurring in a fixed pelletized bed of a Fischer-Tropsch reactor. The model integrates the following contributing parts:

·         Chemical kinetics of paraffine and olefine synthesis from CO and H2 at the surface of a Co-based catalyst;

·         Diffusion model of heat and mass transfer inside quasi-spherical catalytic particles including capillary condensation of water inside pores of the particles;

·         Heat and mass transfer model of inter-particle medium of the bed;

·         The pressure drop correlation of vapor-liquid flow inside porous medium;

·         model for Mass transfer of vapor phase components through the liquid film, which is formed at the surface of the catalytic particles and at the points of interconnection between particles during the process;

·         Modern equation of state (PC-SAFT) for estimation of thermodynamic properties of vapor and liquid phases in three-phase systems (gas -water-rich phase – hydrocarbon-rich phase).

 The model allows estimation of thermal stability of spherical catalytic particle with internal heat generation and diffusion resistance of gaseous component.

This model was integrated into a dialogue computer software, which allows calculation of temperature profile along and across the reactor as well as pressure gradient, velocity vector field and the map of concentration distribution of components in the reactor.

The analysis of the calculations showed that a highly productive catalytic bed should include a catalyst with high thermal conductivity and advanced pore system. It was also shown that a chosen highly productive catalyst can manifest high productivity if aspect ratio of a fixed bed is high enough. The value of aspect ratio determines an optimal value of the diameter of a fixed bed.   Another important result of the calculations was a conclusion that high productivity can be reached at gas hour space velocity (GHSV) exceeding certain limit, which most usually varies between 2000 hr^-1 and 4000 hr^-1.

We developed a catalyst in compliance with recommendations of the mathematical model. This is a cobalt-based catalyst, which is manufactured by impregnation of a pelletized composite thermally conductive support. The catalyst composition and manufacturing procedure are described in the patent applications PCT/RU2010/000323 and PCTRU2010/000429.

The catalyst was tested at a laboratory test rig in a fixed pelletized bed (pellet size 2.5 mm) at the pressure of 21 bar. The testing procedure included activation by hydrogen at 400°Ñ and consequent measurement of catalytic properties in a stream of synthesis gas at the temperatures 150 to 250°Ñ and GHSV from 1000 hr^-1 to 5000 hr^-1. An extract from experimental results is shown in Table 1.

Table 1. Extract from experimental records of testing the highly productive palletized catalyst.

(Conditions: H₂/CO ratio in feedstock 1.98, P = 21 bar)

No	GHSV (hr-1)	" (°Ñ)	Conversion, %	Productivity (g/( m3´hr))
1	1000	230	86	112
2	2000	233	85	244
3	3000	233	82	335
4	4000	235	72	361
5	5000	237	50	276

It can be seen from Table 1 that the catalyst is indeed capable of producing more than 240 kg/(m³´hr)  of liquid Fischer-Tropsch product in a fixed bed. The data of Table 1 suggest that optimal conditions for operation of Fischer-Tropsch process with this catalyst are as follows:

• GHSV 3000 hr^-1
• Temperature of flow 233°Ñ

The results of testing the developed catalyst allowed us to design, build and start up a scaled-up pilot unit for modeling a complete XTL process. The pilot unit with capacity of ¼ bbld models a full cycle of XTL process with natural gas as a feedstock. The unit includes a reactor of desulfurization of feedstock; a reactor for steam-CO₂ reforming, heating-cooling station, compressor station and two parallel highly productive Fischer-Tropsch reactors. The unit in the form of a truck-transportable block was complete and commissioned in October 2010.