Surface spinels

When supported on kaolinite, Ni observability is independent of the thermal pretreatment used to age the catalyst however, its speciation or interaction with this clay change after steaming. In fact, whereas 2% Ni on kaolin is approximately 60% reducible after calcination, the steamed sample is not, probably because of the formation of a stable Ni-silicates or even a surface spinel phase like Ni-aluminate. 

The results show that catalytic activity is higher when the catalyst is reduced in a N -Hj mixture between 500 and 573 K and that it decreases when steam is present in the reduction mixture. For samples reduced in H -HjO flow, treated at high temperature or taken from an industrial reactor, lower activity is evident for the LTWGS reaction. Such deactivation in the samples reduced in Hj-HjO stream may be caused by the formation of a Cu-Zn alloy and by changes in the Cu oxidation state, as determined by XRD characterization. However, the deactivation observed here in the reoxidized-aged samples could not have been caused by the observed decrease in Cu crystal size. In this case, the decrease in activity may be related to a spinel surface species observed in the XRD pattern (Fig. 2d). Laine et al. [14] have proposed that an increase in dispersion accompanied by deactivation can be explained by the formation of a highly dispersed but less active phase, possibly a CUAI2O4 surface spinel . 

In the aged-reactivated catalyst reduced at 500 K Cu is redispers but the catalyst has a lower activity and stability than the fresh samples. This observation can be related to surface spinel species observed by XRD analysis... 

Cation vacancies and interstitials (0001) twins and stacking faults (1120), (10T2) and (1 i53) twins <1150> 0001 screw dislocations, grain boundaries, surface steps, surface spinel precipitates. 

Cu2+ ions stabilized at the surface of SiOz and AI2O3 give rise to reduction peaks at 300-380 C. The reduction of surface spinel in CU/AI2O3 is indicated at 462°C. The reduction of copper silicate is not finished even at 700 C. 

At steaming conditions at high temperature formation of the blue nickel aluminium spinel (Reaction R46 in Table 4.2) may start at temperatures above about 700 C [389], but a less well-defined interaction between nickel oxide and t - or y-alumina is apparent already at lower temperatures. It is possible to form a surface spinel below 600°C, which may hardly be identified by X-ray methods alone. 

Transmission electron micrographs show hectorite and nontronite as elongated, lath-shaped units, whereas the other smectite clays appear more nearly equidimensional. A broken surface of smectite clays typically shows a "com flakes" or "oak leaf surface texture (54). High temperature minerals formed upon heating smectites vary considerably with the compositions of the clays. Spinels commonly appear at 800—1000°C, and dissolve at higher temperatures. Quartz, especially cristobalite, appears and mullite forms if the content of aluminum is adequate (38). 

An example is a LEIS study on a specific spinel, namely ZnAl204, for which cations (Zn) in tetrahedral sites are expected [3.145] to be less stable and therefore move to sites below the surface where they are better shielded, yielding a lower LEIS signal. This has been confirmed by Brongersma et al. [3.146] (Fig. 3.59). This figure shows that LEIS is very sensitive to Zn, as shown by LEIS from ZnO, but for the spinel no Zn is visible in the surface. 

Further investigations of spinel formation reactions are to be found in the literature [1], but the above representative selection illustrates a number of typical features of these rate processes. Following migration of cations from one constituent onto the surfaces of the other, the process is limited by the rate of diffusion across a barrier layer. While obedience to a particular kinetic expression is sometimes reported, the data available are not always sufficiently precise to enable the fit found to be positively... 

In principle, reaction schemes similar to that given in the preceding paragraph may be developed for other comparable rate processes, for example spinel formation. However, Stone [27] has pointed out that, where the barrier phase is not an efficient ionic conductor, the overall reaction may be controlled by the movement of a single cation and anion. In addition, there is the probability that lattice imperfections (internal surfaces, cracks, leakage paths [1172], etc.) may provide the most efficient route to product formation.]... 

---