Bimetallic catalysts find increasing applications in the industry, e.g. in the petroleum industry for catalytic reforming of petroleum fractions and in the environmental sector for the removal of harmful exhaust gases from automobiles. Recent theoretical calculations and surface science experiments have contributed significantly to the understanding of bimetallic alloy systems [1]. The improved fundamental understanding was used to design a steam-reforming catalyst consisting of supported Ni particles with Au alloyed into the surface of the nickel. The chemical properties of the surface alloy could be tuned by varying the Au/Ni atomic ratio in the surface of the catalyst metal alloy particle. In this way it was possible, for the first time, to prepare a bimetallic catalyst based on a fundamental knowledge of the structure, segregation, and reaction processes at a metal surface [2].

Au-Ni nano-particle catalysts were prepared and characterised by a combination of in situ transmission EXAFS in combination with on-line mass spectrometry (MS), transmission electron microscopy (TEM), X-ray powder diffraction and thermogravimetric analysis (TGA). The EXAFS measurements were performed at beamline BM29.

Figure 118 shows the Fourier back-transformed XAFS spectrum obtained from a reduced MgAl2O4 supported Ni catalyst (16.5 weight % Ni) modified with 0.3 weight % Au. Only if we allow that Au has Ni as nearest neighbours at Ni interatomic distances the data can be fitted properly. Because Au and Ni are immiscible in the bulk, this demonstrates that Au is alloyed into the Ni surface layer.

After reduction in pure hydrogen the catalysts were exposed to a diluted n-butane gas at 550°C and an on-line MS measured the steam-reforming activity of a Ni and a Ni/Au catalyst. In Figure 119 it can be seen that the pure Ni catalyst deactivates rapidly, whereas the conversion for the Ni/Au sample is almost constant. The deactivation of the Ni catalyst is typical for this type of catalyst under extreme steam-reforming conditions. It is associated with the formation of graphite, which can be observed by TEM and by the weight increase in a TGA setup. No weight increase was observed for the Au/Ni catalyst during steam reforming.

In conclusion, the combination of a fundamental theoretical understanding of structure and reactivity and several experimental in situ techniques has resulted in the design of a new catalyst for the steam-reforming reaction.

[1] L.P. Nielsen, F. Besenbacher, I. Stensgaard, E. Lægsgaard, C. Engdahl, P. Stoltze, K.W. Jacobsen, J.K. Nørskov, Phys. Rev. Lett., 71, 754 (1993).
[2] F. Besenbacher, I. Chorkendorff, B.S. Clausen, B. Hammer, A.M. Molenbroek, J.K. Nørskov, I. Stensgaard, Science, 279, 1913 (1998).

A.M. Molenbroek (a,b), J.K. Nørskov (a), B.S. Clausen (c).

(a) Center for Atomic-scale Material Physics, Technical University of Denmark, Lyngby (Denmark)
(b) Present address: Haldor Topsøe Research Laboratories, Lyngby (Denmark)
(c) Haldor Topsøe Research Laboratories, Lyngby (Denmark)