Sodium Deactivation of Fluid Catalytic Cracking Catalyst

Abstract

The mechanism of FCC catalyst deactivation by sodium is addressed in this paper. In commercial units, sodium is found to deactivate the matrix surface area significantly but no significant trend was observed for the effect of sodium on the zeolite surface area of equilibrium catalysts. On the other hand, the effect of sodium is more pronounced on zeolite and much less severe on matrix surface area in the typical laboratory deactivation protocol. The differences are explained by the mobility of sodium on catalysts.Significant interparticle migration of sodium is observed both commercially and in laboratory deactivated catalysts. The loading of sodium is associated with the available sites of the particular fraction of that catalyst.In commercial units, sodium preferentially migrates to the freshly added catalyst due to its greater availability of exchange sites. The effect of catalyst sodium and feed sodium are simulated in the laboratory and their effect on catalyst activity and cracking yields are discussed.Sodium on fluid cracking catalyst, FCC, comes from the raw materials used in the catalyst manufacturing process as well as salt contamination in the feedstock. Sodium can deactivate cracking catalysts by poisoning the acid sites on the matrix and zeolite and by promoting sintering of silica-alumina (1). Sodium can act synergistically with vanadium to accelerate the destruction of zeolite (2).The relationship between cracking activity and sodium level of zeolites is complicated and seems to depend on the type of feedstock and catalyst

Continuous Age Distribution Method for Catalytic Cracking 2. Understanding Nonidealities

“…In an earlier paper, we took the hypothesis that refinery zeolite collapse kinetics were separable into tetrahedral framework aluminum (AlO 4 – , Al T ) and silicon (SiO 4 , Si T ) contributions. The separability hypothesis is common in the study of differential equations and made reasonable in the present case because, as noted above, E-cat dealumination is much faster than Si T collapse. ,, After subtracting Al T , the earlier-derived expression for the fractional retention of zeolite framework Si T (eq 6) explained the full range of refinery and pilot plant data available in the literature. − Refineries, however, sometimes employ mixtures of catalysts, which can include two different zeolites. Some of our earlier laboratory results (Figures 3, 5, and 7 in ref ) also appear to be consistent with a zeolite mixture hypothesis.…”

“…We find that 10–20% of the original Si T is quickly lost in the newest portion of the age distribution, but most or all of the second type of zeolite is predicted to be retained in the refinery. This latter result does not represent well density separation data from refineries. − However, if we adopt the presteamed kinetic parameters for cat. E as the second zeolite and optimize the remaining three parameters to best fit the nonpresteamed data for cat.…”

“…N (open symbols), both of which were presteamed and otherwise unrelated to the published data. 9,11 While the agreements at τ R /τ Si = 0.409 are reasonably good, the deviations from the targeted properties becomes systematically worse for the more severe deactivations defined by the larger values of τ R /τ Si . These deactivations are run at systematically higher temperatures (eq 9), 7 suggesting a problem with the CADM calibration may exist at high temperatures.…”

“…That profile had been derived for the first-order decay of the tetrahedral silica (Si T ) portion of the zeolite, the zeolite being the most active and therefore the most important ingredient in the catalyst. Although fundamental kinetic and reactor models had been used and first-order Si T decay was confirmed by Pine, as well as by fitting published equilibrium catalyst density separation data, − we nevertheless found substantial deviations from ideal first-order Si T decay . The deviations were evident both by the classical Arrhenius method and by implementing the new continuous age distribution method (CADM) experimentally .…”

“…The results can then be compared to theory for first-order decay (eq 6) 7 and published experimental data. 7,9,11 To obtain the results in Figure 5, we first fitted the earlier eq 6 7 to the cat A density separation data of Palmer and Cornelius, 9 and of Zhao and Cheng 11 (full symbols), obtaining τ R /τ Si values of 3.57 and 0.409, respectively. Since their data were incomplete, we had assumed values of MSA and zeolite retention as reported before, 7 the choice of which values effectively erased any potential unreported effects of metastable zeolite.…”