Slurry reactors interfacial area

The other major type of catalytic reactor is a situation where the fluid and the catalyst are stirred instead of having the catalyst fixed in a bed. If the fluid is a liquid, we call this a slurry reactor, in which catalyst pellets or powder is held in a tank through which catalyst flows. The stirring must obviously be fast enough to mix the fluid and particles. To keep the particles from settling out, catalyst particle sizes in a slurry reactor must be sufficiently small. If the catalyst phase is another Hquid that is stirred to maintain high interfacial area for reaction at the interface, we call the reactor an emulsion reactor. These are shown in Figure 74. 

Liquid-solid interfacial area in slurry reactors... 

In mass-transfer correlations, the volumetric mass-transfer coefficient is expressed using the gas-liquid interfacial area per unit volume of slurry (or expanded column or reactor, VR) (Koide, 1996 Kantarci el al., 2005 NTIS, 1983) ... 

An accurate evaluation of kxa is complicated by the heterogeneous nature and poor definition of contaminant/soil systems. Some success has been achieved in modeling mass transfer from a separate contaminant phase. During degradation these nonaqueous phase liquids (NAPLs) often dissolve under conditions where phase equilibrium is not achieved and dissolution is proportional to k a. Experimental determinations and correlations for k-p depend on interfacial area of the NAPL and liquid velocity at the interface (Geller Hunt, 1993). For adsorbed contaminants, kxa varies with soil composition and structure, concentration and age of contamination, and therefore with time. For example, slurry reactor tests indicate that the rate of naphthalene mass transfer decreases with time, with media size, and with aging of the tar prior to testing (Luthy et al., 1994). 

With a finely divided solid catalyst as typically used In the Flscher-Tropsch synthesis In slurry reactors It Is generally agreed that the major mass-transfer resistance, If It occurs, does so at the gas-liquid Interface. There are considerable disagreements about the magnitude of this resistance that stem from uncertainties about certain physical parameters, notably interfacial area, but also the solubility and mass transfer coefficients for H2 and CO that apply to this system. However when this resistance Is significant, the concentrations of Hg and CO in the liquid in contact with the solid catalyst become less than they would be otherwise, which not only reduces the observed rate of reaction but can also affect the product selectivity and the rate of formation of free carbon. 

Particularly in co-current down-flow operation, very high gas-slurry interfacial areas per cubic meter of reactor volume can be realized [6]. To improve the solids suspension and/or to improve mass or heat transfer, in many cases a stirrer is added to the system. Particularly where pure gases have to be absorbed in the slurry and no gases are produced, the gas may be sparged into the liquid via a hollow shaft stirrer, sucking the gas from the free board above the slurry. 

Power or energy dissipated in the aerated suspension has to be large enough (a) to suspend all solid particles and (b) to disperse the gas phase into small enough bubbles. It is essential to determine the power consumption of the stirrer in agitated slurry reactors, as this quantity is required in the prediction of parameters such as gas holdup, gas-liquid interfacial area, and mass- and heat-transfer coefficients. In the absence of gas bubbling, the power number Po, is defined as... 

Gas holdup is an important hydrodynamic parameter in stirred reactors, because it determines the gas-liquid interfacial area and hence the mass transfer rate. Several studies on gas holdup in agitated gas-liquid systems have been reported, and a number of correlations have been proposed. These are summarized in Table VIII. For a slurry system, only a few studies have been reported (Kurten and Zehner, 1979 Wiedmann et al, 1980). In general, the gas holdup depends on superficial gas velocity, power consumption, surface tension and viscosity of liquids, and the solid concentration. The dependence of gas holdup on gas velocity, power consumption, and surface tension of the liquid can be described as... 

As indicated previously, to obtain meaningful estimates of gas absorption rates, the gas-liquid interfacial area must be known locally. The present work is concerned with the effect of gas-liquid interfacial area on the performance of a slurry bubble column reactor. 

In general, the gas holdups and kLa for suspensions in bubbling gas-liquid reactors decrease substantially with increasing concentrations of solid particles, possibly because the coalescence of bubbles is promoted by presence of particles, which in turn results in a larger bubble size and hence a smaller gas-liquid interfacial area. Various empirical correlations have been proposed for the kLa and gas holdup in slurry bubble columns. Equation 7.46 [24], which is dimensionless and based on data for suspensions with four bubble columns, 10-30 cm in diameter, over a range of particle concentrations from 0 to 200 kg m 3 and particle diameter of 50-200 pm, can be used to predict the ratio r of the ordinary kLo values in bubble columns. This can, in turn, be predicted for example by Equation 7.41, to the kLa values with suspensions. 

In slurry systems, similar to fluidized beds, the overall rate of chemical transformation is governed by a series of reaction and mass-transfer steps that proceed simultaneously. Thus, we have mass transfer from the bubble phase to the gas-liquid interface, transport of the reactant into the bulk liquid and then to the catalyst, possible diffusion within the catalyst pore structure, adsorption and finally reaction. Then all of this goes the other way for product. Similar steps are to be considered for heat transfer, but because of small particle sizes and the heat capacity of the liquid phase, significant temperature gradients are not often encountered in slurry reactors. The most important factors in analysis and design are fluid holdups, interfacial area, bubble and catalyst particle sizes and size distribution, and the state of mixing of the liquid phase. ... 

Macrokinetic processes for slurry systems are sketched on Table 7. The main points are the characteristics of the three phase dispersion (fluid holdups, interfacial areas, bubbles and catalyst particles size distributions), the state of macromixing of fluids which can be defined through the concept of residence time distribution, the state of micromixing of fluids which for the gas phase shall determine the degree of coalescence of bubbles, the heat transfer between the reactor and the environment. 

The most important fluiddynamic parameters of aerated stirred slurry reactors are energy dissipation and pumping efficiency of agitator and gas throughput, gas-holdup and mean bubble diameter produced (i.e. interfacial area resulting) and flooding characteristics. ... 

In general, it is also possible in a slurry reactor that a significant mass transfer resistance 1/kgag can be observed at liquid-particle interface. However, due to the small catalyst particles which are usually applied in the FT slurry process the interfacial area a is large, ag is given by... 

Effect of gas-liquid mass transfer Different opinions on the importance of mass transfer limitations have been uttered by Satterfield and Huff (81,82,94) and Zaldl et al. (54) and Deckwer et al. (87). Satterfield and Huff (81) assxuned a bubble diameter of eUaout 2 mm and concluded that the FTS in slurry phase may be significantly limited by gas-liquid mass transfer. New experimental results do, however, confirm that in the molten paraffin system bubble diameters are less than 1 mm (53-55). With this low value and the high gas holdup observed in FT liquid f ase, i.e., eq. (8), high interfacial areas are obtained. Therefore slgniflccuit mass transfer can be excluded and for the catalyst systems studied until now the FT process in slurry phase is mainly reaction controlled (54,87). In addition, it follows from the reactor model used by Satterfield and Huff (81) that the relative mass transfer resistance is given by (95)... 

In this chapter it was shown that the major engineering parameters which might affect the performance of a FT slurry reactor can be estimated from rather reliable correlations. There are, however, some controversial results in the literature which concern gas holdup and interfacial area (bubble diameter). Additional studies would be valuable for further clarification of this point. However, one can state, at least, that gas holdup and interfacial area are surprisingly large in the... 

Table 6.4 shows the macroscopic attributes that may be achieved in these three-phase slurry reactors from a mechanical and configurational standpoint. For instance, in stirred slurry reactors, the action of the stirrer causes chopping of the bubbles emanating from the distributor, and hence the steady-state bubble size achieved in the reactor is largely determined by the breakage caused by the stirrer action. In slurry bubble columns and three-phase fluidized beds, however, fine bubbles emanate from the distributor, and as they rise, coalescence dominates and the bubbles increase in size (causing reduction in interfacial area... 

While the lower order models described in Section 6.3 are useful for the quick prediction of the overall performance of a reactor, these models often rely on simplified flow approximations and often fail to account for change in the local fluid dynamics or transport processes during the presence of internal hardware or changes in flow regimes. Moreover, these models are also based on empirical knowledge (as discussed in Section 6.4) of several parameters such as interfacial area, dispersion coefficients, and mass transfer coefficients. Some of these limitations may be avoided by using CFD models for simulations of gas-liquid-solid flows in three-phase slurry and fluidized bed. 

The two common types of slurry reactors are the bubble column (section 4.6.1.2) and the stirred gas liquid contactor (section 4.6.1.3) (Sometimes the bubble column slurry reactor is called a gasiliquid solid fluidized bed see Fan, 1989). Even when a very fine dispersion of bubbles is made, the bubble size usually exceeds the particle size of the solids by a factor of 10 to 100 or more. That means that even for relatively low volume fractions of solids the solid/liquid interfacial area is much greater than the gasAiquid interfacial area. The consequence is that in most situations the mass transfer resistance is concentrated in the film around the bubbles. An important aspect of slurry reactors is that the mass transfer around the bubbles can be influenced by the presence of the solids. The influence can be positive or negative, depending on the prevailing mechanism. Four mechanisms have been identified ... 

Three-phase packed bed reactors generally have a lower specific capacity than slurry reactors, for two reasons Much larger catalyst particles are used, so that for rapid reactions, with diffusion or mass transfer limitations, much larger catalyst volumes are required. Also, the maximum specific gas/liquid interfacial area is generally smaller. On the other hand, the volumetric mass transfer coefficients at the gas/liquid and at the liquid/solid interfaces are of comparable magnitude, so they are better adapted to one another. Heat transfer rates to the walls are quite limited. 

In the classical slurry reactor processes analysed by Satterfield and Huff, mass transfer limitations were already manifest at relatively low reactor productivities obtained at low space velocities (typically about 200 N 1 gas per litre of slurry per hour) and low catalyst concentrations of typically 50 g catalyst per litre of slurry). In view of the gas-liquid mass transfer limitations, the occurrence of bubble coalescence is an important phenomenon in large bubble columns since the formation of big bubbles reduces the gas-liquid interfacial area and hence the effectiveness of mass transfer from gas to liquid. 

---