Modeling catalyst distribution

Let s note three important aspects followed from the model [4] application for description of reesterification reaction. At first as reesterification reactions with TBT and in it absence proceed in identical conditions, then from the comparison of figure 1 kinetic curves follows, that the reaction fractal-like behaviour in TBT presence is due to local fluctuations of catalyst distribution in reactive medium. Secondly the division of reaction duration into short and long... 

There is, nevertheless, some evidence (35, 36), based in NiNaY and NiMo/ alumina/Y model catalyst systems, that the amount of coke formed is reduced with increasing intimacy of mixing of the two functions at the submicron level. This concept is further supported by the reported relatively high performance of NiW/ASA (amorphous silica-alumina) cogel HC catalysts which, it is claimed, exhibit an excellent distribution of the NiW hydrogenation function throughout the catalyst particles (37). 

Supported metal catalysts generally show an increase in catalytic activity compared to the pure oxide or metal. Yet these systems are not well characterised, owing to the fact that such catalysts typically consist of a range of different supported metal sites, from small clusters to monolayer islands, all with non-uniform distributions in size and shape. One way to begin to understand such complex systems is to attempt to capture some essential part of the full system by developing model catalysts experimentally or using computer modelling techniques. This chapter concentrates on the latter but in the context of the relevant experimental data. 

This discrepancy is partially attributed to the very short contact time through the membrane layer for any significant additional conversion due to the catalytic membrane layer. It may also be the result of the actual catalyst distribution in the membrane and support layers. It has been pointed out that the reaction conversion of cyclohexane to benzene in a porous catalytic membrane reactor is extiemely sensitive to the distribution of the Pd catalyst through the membrane [Cannon and Hacskaylo, 1992]. The last point is supported by the observation that, when the membrane is not catalytic and the remaining conditions arc the same, experimental data and model prediction agree [Becker etal., 1993]. 

The above modeling study showed therefore the importance of having a proper knowledge of the pore texture and catalyst distribution of the catalytic filter, since they can seriously affect its performance. This suggests, in line with Ref. 40, the need of a proper characterization of the porous structure of the catalytic filters, concerning pore connectivity, pore size distribution, presence of deadend pores, etc., since each of these features might play a primary role in reactor performance. On the basis of such characterization work, valuable information could be drawn in order to choose or optimize the preparation routes. 

Recently, Yates and Rowe (YIO) have observed, on the basis of their model for catalyst distribution in the freeboard region, that this region can usually exert a considerable influence on the course of the reaction. Their observation is essentially parallel with the concept of the successive contact mechanism. However, they use the bubbling bed model in calculating the reaction in the dense phase, so that the effect of directly contacting catalyst seems to be corrected two times, first partially in the dense phase and then in the freeboard region (see Section VII,A,3). 

Several other recent modelling membrane reactor studies are also worth discussing, Varma and coworkers [127] have analyzed the effect that nonuniform catalyst distribution on the membrane itself (for CMR and PBCMR applications) and in the catalyst bed (for PBMR applications) has on membrane reactor performance. Conventional membrane reactor models were utilized by a number of groups to model their experimental data. Shu and co-workers [33]... 

The model catalysts were prepared by HV deposition of an amorphous SiO film (by evaporation of SiO in 10 Pa of oxygen), onto which Pt was deposited by high vacuum evaporation, as described earlier (7). In order to get catalysts of different dispersion, the mean thickness of deposited Pt was varied between 0.1 and 1 nm. By TEM inspection of catalyst specimens the particle density and the particle size distribution were obtained, from which data the platinum surface area and the dispersion were calculated. Additionally a conventional 6,3 % Pt/Sio catalyst (EUROPT-1, d = 1.7 nm) was used in the experiments. 

It is difficult to deduce what gold particle morphologies arise from heterogeneous chemical reduction of HAuCU. Understanding of the model catalysts is much easier. In brief, a) nucleation of gold clusters occurs at surface defects that act as traps b) on AI2O3, there are two kinds of traps at <0.8 and >1.6eV c) the defect density is ca. 3 x 10 sites per cm (10 monolayer) and d) when the clusters grow to >600 atoms, they leave the traps. This can explain the bimodal size distribution of the clusters. Atomistic definition of these traps is needed. 

---