Coke reforming, catalyst behavior

Although the deactivation of Industrial catalysts is often due to two or more different causes, the modeling of simultaneous deactivation phenomena has not been widely studied (refs. 1, 2). The occurrence of two different deactivation processes not only adds another level of complexity to the determination of the intrinsic kinetic behavior but also complicates the interpretation of the experimental results. In our previous studies regarding the thloresistance of naphtha reforming catalysts (refs. 3, 4) we have shown that the activity decay caused by the presence of sulfur compounds in the feed is often accompanied by coking. In this situation, the thioresistance cannot be obtained in a simple way from the deactivation curves. The characteristics of the sulfur poisoning have to be deduced from the overall deactivation rate. 

The Pt-Re system has been studied extensively since the 1970s because adding Re to AhOs-supported platinum catalysts increases the resistance to deactivation of the catalysts used in naphtha reforming by preventing coke deposition. By using carbonyl precursors, well-defined bimetalhc species have been prepared. A proper characterization of these species allowed a relationship to be established between their structure and their catalytic behavior. Table 8.3 shows several Pt-Re bimetaUic catalytic systems prepared using different carbonyl species in which Pt-Re interactions determine the catalytic behavior. 

The purpose of our paper is to compare the behaviors of platinum,rhenium, iridium and rhodium under the same coking conditions (Cyclopentane, 400 C)h He and Ir were chosen as they are usually added to Pt in reforming bimetallic catalysts. Rh was studied since it is the most active metal in hydrocarbon steain reforming (Refs. 5-7). 

The model was applied in order to investigate the influence of various parameters on the performance of the FBMR with oxygen addition. Although the results showed that autothermal operation can be achieved by using a O2/CH4 feed ratio of approximately 0.3, the interaction between the different parameters is quite complex. For instance, in methane reformers an important parameter is the steam/carbon ratio. However, when feeding also oxygen, the steam becomes also a product of the oxidation reaction and this makes the prediction of the reactor behavior a bit more complicated. Furthermore, an important conclusion of the work is that oxygen addition reduces the coke formation and consequently the catalyst deactivation. 

---