(324a) Deactivation and Promotion of Co Catalysts During Fischer-Tropsch Synthesis: a Computational and Experimental Study

Authors: 
Xu, J., National University of Singapore
Tan, K. F., National University of Singapore
Chang, J., Institute of Chemical and Engineering Sciences (ICES), A-STAR

Introduction

Fischer-Tropsch synthesis (FTS) converts synthesis gas, a mixture of CO and H2, into various hydrocarbons. Supported Co catalysts are preferred for FTS because of their good hydrocarbon selectivity, high CO conversion, and low water gas shift activity [1]. However, supported Co catalysts are known to deactivate under realistic FTS conditions [2], and it is hence desirable to improve the stability of Co catalysts. A number of deactivation mechanisms have been proposed [1], including the formation of carbidic and graphitic coke [2,4,5].

In this work, the relative stability of various forms of deposited carbon and the effect of boron on the carbon deposition mechanism [6] were evaluated using periodic Density Functional Theory (DFT-PBE). To confirm the first principles based predictions, a series of Co catalyst promoted with various concentrations of boron were synthesized, characterized and tested.

Results and Discussion

Terrace sites were modeled using a Co(111) slab, while step sites were created by removing rows of Co atoms from the Co(111) surface [7]. Carbon binding energies, as well as the Gibbs free energy for the CO + H2 -> C + H2O reaction under FTS conditions are shown in Table 1. Carbide formation, in particular a clock reconstruction initiating from the step sites, and graphene formation nucleating from the step sites, were found to be the most stable forms of deposited carbon on the Co surface. Surface and subsurface carbon on the Co terraces are less favorable under typical FTS conditions. Boron binds strongly to the step sites and to subsurface sites. Boron at the step sites is found to significantly reduce the carbon binding energy both at and near the step edges, hence reducing the driving force for graphene nucleation and initiation of a clock reconstruction.

To confirm the promotional effect of boron on the catalyst deactivation, boron promoted Co catalysts were prepared and tested under typical FTS conditions (240 °C, 20 bar, H2/CO=2). Co catalysts were prepared using wet aqueous impregnation to produce 20 wt% Co supported on γ-Al2O3. Boron was introduced via boric acid following a similar procedure. X-ray Photoelectron Spectroscopy characterization of the reduced catalyst indicates the formation of a thin Co boride film, consistent with the simulation data. Catalysts promoted with 0.5 wt% boron were found to show promising activity and stability, and were subjected to extended tests. Fig. 1a compares the effect of 0.5 wt% boron on the CO conversion for Co catalysts during FTS. The deactivation rate for the unpromoted catalyst compares well with values reported for a slurry bubble column [1]. Boron promotion significantly reduces the deactivation rate from 1.7x10-3 hr-1 to 2.7x10-4 hr-1. Transmission Electron Microscope (TEM) images of the unpromoted catalyst after 200 hours on stream show the formation of graphitic carbon (Fig. 1b). No graphitic carbon could be detected on the promoted catalyst.

Conclusions

Based on Density Functional Theory calculations, boron was proposed to enhance the stability of Co catalysts during FTS by reducing graphene nucleation and initiation of a clock reconstruction from the step sites. Promotion of a 20 wt% supported Co catalyst by 0.5 wt% boron was found to reduce the rate of deactivation by a factor of 6.

Table 1. Effect of boron promotion on the carbon binding energy on a Co surface.

Site

Carbon binding energy / DGrº(500 K, 20 bar) (kJ/mol)

Without boron

With 50% boron at steps

Step

-673 / -9

-471 / +192

Step-edges (clock reconstruction)

-697 / -33

-579 / +84

Terrace hollow sites

-650 / +14

N.A.

Figure 1. Effect of boron promotion on the CO conversion for Co/alumina catalysts as a function of time on stream (A). TEM image for the unpromoted catalyst after 200 hours on stream (B).

References

[[1]] A.M. Saib, A. Borgna, J. van de Loosdrecht, P.J. van Berge, J.W. Niemantsverdriet, Appl. Catal. A, 312 (2006) 12.

[2] D.J. Moodley, J. van de Loosdrecht, A.M. Saib, M.J. Overett, A.K. Datye, J.W. Niemantsverdriet, Appl. Catal. A, 354 (2009) 102.

[3]   Iglesia, E., Soled, S.L., Fiatto, R.A., Via, G.H., J. Cat., 143 (1993) 345.

[4]   G.A. Beitel, A. Laskov, H. Oosterbeek, E.W. Kuipers, J. Phys. Chem., 100 (1996) 12494.

[5]   G.A. Beitel, C.P.M. de Groot, H. Oosterbeek, J.H. Wilson, J. Phys. Chem., 101 (1997) 4035.

[6]    J. Xu, M. Saeys, J. Cat., 242 (2006) 217.

[7]    J. Xu, M. Saeys, J. Phys. Chem. C., 113 (2009) 4099.