Effect of Preparation Variables On the Performance of Supported Iron Fischer-Tropsch Catalysts

The typical iron FT catalyst used by Sasol is unsupported iron promoted with copper, potassium and silica (Fe/Cu/K/SiO₂). Bukur et al. [1] reported a weight-time yield of 450 mmol (CO+H₂)/g_Fe/h/MPa and a C₂₊ hydrocarbon productivity of 0.86 g_HC/g_Fe/h for 100Fe/3Cu/4K/16SiO₂ at 260 °C and 2.2 MPa.  Despite their high activity and favorable selectivity, unsupported iron catalysts are generally too mechanically weak to be used in slurry bubble column reactors (SBCRs): the most thermally efficient and economical reactors [2].

This study chooses four important variables in the preparation of alumina-supported Fe catalysts and holds all other variables constant to systematically investigate the effect of each variable.  Catalysts were prepared to examine the effects of (1) iron loading level, (2) potassium loading level, (3) impregnation method (aqueous Incipient wetness or non-aqueous slurry), and (4) timing of potassium addition.

Table 1 below catalogs the values of each of the 4 preparation variables for the six catalysts of this study.

Table 1.  Preparation variable values for each of the six catalysts studied in this paper

	Variable 1		Variable 2		Variable 3				Variable 4
Catalyst		Fe Loading (weight%)		K Loading	Aqueous/       Non-aqueous			Deposition Method		K Impregnation Timing
K1		20		4K/100Fe	NA			SI1		Sequential impregnation
K2		40		4K/100Fe	NA			SI		Sequential impregnation
K3		20		4K/100Fe	A			IWI2		Sequential impregnation
K4		20		8K/100Fe	A			IWI		Sequential impregnation
K5		20		4K/100Fe	A			IWI		Co-impregnation
K6		20		4K/100Fe3	A			IWI		Directly on support plus Sequential impregnation
	1SI = Slurry Impregnation (50% acetone, 50% isopropanol)
	2IWI = Incipient Wetness (aqueous)
	3Plus 0.2 weight% added directly onto the support

Adding K to the support which increased it's basicity largely increased the activity of the catalyst. The co-impregnation preparation (K5) significantly enhanced the activity since iron particles may have better contact with potassium. Doubling the potassium level (K4 vs. K3) slightly decreased the activity and methane selectivity indicating a threshold for potassium level of around 4K/100Fe. Although K5 and K6 were the most active catalysts, stability analysis conducted at 250C revealed that neither of these two catalysts was very stable, i.e., K6 particularly deactivates to nearly half its former rate in only 175 hours on stream.  The non-aqueous catalysts were the most stable.

These alumina-supported catalysts had activities ranging from 211-267 mmol(CO+H₂)/g_cat/h/MPa), and are nearly three-fold more active than the silica-supported catalyst reported by Bukur (101 mmol(CO+H₂)/g_cat/h/MPa), and 5-fold more active than Bukur's alumina-supported (50 mmol(CO+H₂)/g_cat/h/MPa) catalyst.

References:

1- Bukur, D.B. and X. Lang, Highly Active and Stable Iron Fischer-Tropsch Catalyst for Synthesis Gas Conversion to Liquid Fuels. Ind. Eng. Chem. Res., 1999. 38(9): p. 3270–3275.

2- Xu, J., C.H. Bartholomew, J. Sudweeks and D.L. Eggett, Design, synthesis, and catalytic properties of silica-supported, Pt-promoted iron Fischer–Tropsch catalysts. Topics in Catalysis, 2003. 26(1-4): p. 55.