Intraparticle coking

This condition is a consequence of the high vapor pressure of the light hydrocarbon fraction. Because of the rapid transport out of the particle, this fraction has no possibility to coke. The bitumen, on the other hand, is viewed as a high boiling liquid which can undergo intraparticle coking. The... 

The results of the mechanism described above are (1) that V and Ni concentrate on the surfaces of catalyst particles (as shown by EDS), (2) a high degree of metal sintering occurs on the catalysts particles (as observed by SEM) and (3) active metals are very sensitive to oxidation-reduction treatments. We interpret the lower MAT catalytic activity (w.r.t. the zeolite surface area) as due to a high coke-making tendency of the catalyst leading to pore mouth plugging and increased intraparticle diffiisional constraints. 

A comprehensive model for the steam reformer should be developed. This model should take into consideration the characteristics of the combustion chamber as well as the details of the processes taking place in the catalyst tubes steam reforming reaction, coke formation, reoxidation of Ni and rereduction of NiO, mass and heat transfer between the catalyst pellets and the bulk gas (both external and intraparticle), heat transfer between the catalyst tubes and the combustion chamber. .. etc. 

A series of experiments varying temperature, micro-sphere size and time on stream have been performed in a fixed fluidised bed microactivity reactor to study the role of intraparticle diffusion in commercial fluid catalytic cracking (FCC) catalysts, particularly on gasoline yield and catalyst deactivation by coke deposition, for the cracking of a vacuum gas oil. Additionally, a mechanistic model that describes interface and intrapartide mass transfer interactions with the cracking reactions, has been used to study the combined influence of pore size and intraparticle mass diffusion on the deactivation of FCC catalysts and the gasoline yield. 

First studies on the influence of intraparticle diffusional mass transfer on catalytic reactions, and about deactivation of cracking catalysts by coke deposition, started with Thiele [1] and Voorhies [2], respectively. To-date, Thiele s analysis remains valid, however the approach followed by Voorhies in which coke formation is expressed as a function of time on stream (fos), although still used in many studies [3], is not adequate. It has been stated that deactivation, due to the coverage of active sites by coke deposition and to pore blockage by coke growth [3, 4], should be directly related to coke itself emd not to tos [5]. In this way, coke formation is linked to the operating conditions, the nature of the feedstock and the type of catalyst. 

In practice, some constraints are usually imposed on the lumped model. For example, certain chemical species (e.g., coke precursors) need to be kept unlumped. In this and similar situations, the lumping matrix needs to reflect the constraint a priori. Li and Rabitz also extended their analysis to include nonisothrmal effect and intraparticle diffusion. ... 

The present effort makes no attempt to match in scope the previous review we shall confine ourselves to work concerning chemical poisoning and coking as the primary mechanism of deactivation but retain the classification according to scale — individual kinetics and mechanism, intraparticle problems, and chemical reactor problems. Sintering has been admirably covered in a recent review (2), and the subject of automotive exhaust catalysis (which is almost wholly an exercise in catalyst mortality) will be treated in one forthcoming (3). 

There were no effects of pore dimension on the deactivation rates (due to coke formation) of fresh catalyst over the range investigated. Further studies of deactivation rates after one and two hours of utilization at 205 C also revealed no influence on pore structure. This would rule out intraparticle mass transport as controlling deactivation rates as well as the occurrence of any pore blockage resulting from coking in this reaction. 

Experiments on larger size particles have also involved H-mordenite, but with cumene cracking as the reaction (13). Relations between coke content, activity, and intraparticle diffusivity were investigated on 1/16 in. Norton Zeolon extrudates for 230-250 C and space velocities from 0.2 to 0.65 wt/wt-hr. Effective diffu-sivities were determined (with SF via chromatography) as a function of reaction time and coke content with the results shown in Figure 1. Diffusivity decreased twofold for reaction times of 2 hours or longer, but remained essentially constant after that. 

Intraparticle diffusion is considered in the apparent rate coefficients. Catalyst deactivation by coke occurs during the first 100 h of time-onstream and then reaches equilibrium. 

---