Wetted catalyst pellets

FA Hessari, SK Bhatia. Reaction rate hysterisis in a single partially internally wetted catalyst pellet Experiment and modelling. Chem Eng Sci 51 1241-1256,1996. 

Most common in practice is a trickling flow that occurs at relatively low gas and liquid flow rates. Under these conditions in trickle-bed reactors (TBRs) complete liquid films around the particles would break up into partial films, rivulets and droplets and a so-called partial wetting trickling regime exist. It is easier for hydrogen to enter the no wetted or very thin wetted part of partially wetted catalyst pellets because the mass-transfer resistance between a gas and the particle sur-... 

The purpose of this paper is to summarize previous interpretations of the effect of incomplete catalyst wetting on trickle-bed performance and to develop a model for the effectiveness factor for partially wetted catalyst pellets. In the case of a reaction... 

Recently (21) another approximate model for the catalysts effectiveness factor in trickle bed reactor has been proposed. In this model the effectiveness factor for a partially wetted catalyst pellet in a trickle-bed reactor for a reaction occurring only in the liquid filled pore region of the pellet is defined by ... 

DUDUKOvic AND MILLS Catalyst E ffectiveness in Trickle-Bed Reactors 391 Mathematical Model For Reaction In Partly Wetted Catalyst Pellets... 

It is instructive to consider first a siiqtler problem, namely that of a reaction on a partly externally but coiqtletely internally wetted catalyst pellet. This case is of Interest in particular for hydrocarbon feeds which presumably readily wet the internal pore structure. Clearly if the kinetic rate is very slow the reduction in the "supply area" i.e. external area wetted through which reactants arrive to the pellet will hardly affect pellet utilization but for higher kinetic rates a reduction in utilization due to Incomplete external wetting should become apparent. 

Catalysts are prepared by impregnation by spraying a solution of a metal salt onto pellets of a porous support until incipient wetness. The pellets are then dried and calcined to transform the metal into insoluble form. 

Cylindrical catalyst pellets, whether hollow or not, are often given shape by extrusion of a paste of wet catalyst (Chapter 3). During this process the diameter of pores in the radial direction will become smaller, whereas for pores in the longitudinal direction the length will decrease. This can result in a radial effective diffusivity being smaller than the longitudinal one. Thus cylindrical catalyst pellets can be anistropic. 

Industrially, hydrodesulfurization of oil fractions, like aU hydroprocessing, is carried out catalytically in a fixed bed trickle flow unit. The catalyst is stacked in a packed bed and gas (hydrogen) and liquid (oil) are fed downstream concurrently. The reactor operates in the trickle-flow regime, in which the catalyst pellets are fully wetted with the liquid and both gas and liquid flow along the external surface. 

Stream formation in large-diameter reactors and wall channeling in small-diameter reactors can lower reactor performance. Often the catalyst is not fully exploited owing to incomplete wetting by the liquid and low mass-transfer rates together with low residence times within the catalyst pellets. 

It is convenient to compare the features of mass transfer in a trickle-bed (completely wetted pellet) and in catalyst pellets under condition of capillary condensation. In trickle-bed reactors, the interfacial gas-liquid surface (5j) is slightly less (due to porosity) than the external surface of pellet 5r, Si < S sq, and is proportional to (1 — s) sq/R. 

The recommended procedure for scale-up of trickle-bed reactors is, first, to establish in the laboratory the rate of reaction for single catalyst pellets which will include the effect of wetting efficiency [12]. If there is a soluble gaseous reactant, the rate should account for mass transfer from both the gas- and liquid-covered surfaces of the pellet. This basic rate data then can be used with intrareactor mass and, if necessary, energy conservation expressions to design the large-scale reactor. This second step should include the liquid distribution. The required mass and energy transport rates will limit application of this approach because the majority of the literature is concerned exclusively with nearly atmospheric conditions. 

These are used for screening of fines from lumps of raw material (limestone) or finished product (dried salt from wet lumps) and removal of fines from catalyst pellets (screening of catalyst during aimual shutdown of a sulphuric acid plant). 

In this formulation, it is assumed that the overall reaction rate is the sum of the reactions that are occurring on the wetted and non-wetted portions of the catalyst pellet surface. Over the wetted surface, the reactants reach the surface of the catalyst through a liquid film which may be under mass transfer control. 

Biot number of the dry and wet catalyst, kg (or kb) V /D Sg, dimensionless Inlet and outlet concentrations of key components, kg mole m 3 Gas concentration in the liquid, kg mole Pellet dimension (volume/surface area), m Diameter of catalyst pellet, m Molecular diffusivity, m s ... 

Liquid holdup in bed porosity, dimensionless = Bed void fraction, dimensionless = Catalyst pellet void fraction, dimensionless = Effectiveness factor of totally wetted catalyst, dimensionless = Catalyst wetting efficiency, dimensionless = Overall effectiveness factor, dimensionless = Overall efficiency of a trickle bed, dimensionless... 

Comparison of Eqs. (96) and (97) shows that the temperature difference across the film surrounding the catalyst pellet must be very low for a fully wetted particle, but could be important for a non-wetted particle. The design engineer must ensure that scale-up of reactor diameter for highly exothermic reactions does not diminish heat transfer from the reactor, or increase evaporation of liquid and generation of hot spots. To test for these effects, a pilot plant should be operated so that evaporation can occur leading to the development of dry zones. When this condition is found detailed axial temperature measurements should be taken. 

As previously explained, problems arise from an uneven distribution of the liquid around the catalyst pellets in TBRs In fact only a part of the external particle surface is effectively wetted by the flowing liquid the remaining zones are partly in contact with semistagnant liquid pockets and partly covered by an almost motionless thin film The chosen reaction is not significantly exothermic and the catalyst pores can therefore be assumed completely filled by liquid No particular problem exists if the controlling reactant is fed with the liquid because the most active zones of the catalyst are very likely those characteri- ed by a flowing liquid and the conversion rate increases as the liquid flow... 

The second case implies that the pores in the catalyst pellets are not interconnected and that the fraction of internal wetting corresponds directly to external wetting. This in general is not the case when dealing with real catalysis and hydrocarbon feeds which readily wet internal pore structures (22). 

---