Catalyst wetting reactor performance

At low liquid flowrates incomplete wetting of the catalyst by the liquid may occur, as illustrated in Fig. 4.19, leading to channelling and a deterioration in reactor performance 34 . 

An appropriate model for trickle-bed reactor performance for the case of a gas-phase, rate limiting reactant is developed. The use of the model for predictive calculations requires the knowledge of liquid-solid contacting efficiency, gas-liquid-solid mass transfer coefficients, rate constants and effectiveness factors of completely wetted catalysts, all of which are obtained by independent experiments. 

In ideal trickle flow reactors, all particles in the catalyst bed take part in the overall reactor performance, since each is surrounded (wetted) by the liquid phase that flows around it. Situations in which the liquid flows preferentially through a certain part of the bed, while the gas phase flows predominantly through another part, should be avoided [23]. In this case, part of the bed is not contacted by the liquid reactant at all and docs not contribute to the overall conversion. To avoid this maldistribution, Gierman [20] proposed the following criterion for the wetting number Wtx for co-current downflow operation ... 

The above issues associated with prediction of trickle-bed reactor performance has motivated a number of researchers over the past two decades to perform laboratory-scale studies using a particular model-reaction system. These are listed in Table I. Although a more detailed summary is given elsewhere (29), a general conclusion seems to be that both incomplete catalyst wetting and mass transfer limitations are significant factors which affect trickle-bed reactor performance. 

A few reactor models have recently been proposed (30-31) for prediction of integral trickle-bed reactor performance when the gaseous reactant is limiting. Common features or assumptions include i) gas-to-liquid and liquid-to-solid external mass transfer resistances are present, ii) internal particle diffusion resistance is present, iii) catalyst particles are completely externally and internally wetted, iv) gas solubility can be described by Henry s law, v) isothermal operation, vi) the axial-dispersion model can be used to describe deviations from plug-flow, and vii) the intrinsic reaction kinetics exhibit first-order behavior. A few others have used similar assumptions except were developed for nonlinear kinetics (27—28). Only in a couple of instances (7,13, 29) was incomplete external catalyst wetting accounted for. 

S catalyst wetting may occur. These adversely affect the reactor performance. 

Mears24 suggested that the fact that (4-6) correlated the data was fortuitous. He questioned the validity of Eq. (4-5) for the packed-bed trickle-bed reactor, since this equation was derived from the data taken for the flow over a string of spheres. He argued that the dependence of reactor performance on velocity in pilot-scale reactors is due to incomplete catalyst wetting at low flow rates. For a first-order reaction, he modified Eq. (4-4) as... 

It is difficult to ascertain whether the poor performance observed in pilot-scale trickle-bed reactors is due either to ineffective catalyst wetting or to the axial dispersion effects, because both these effects are physically realistic and both occur under similar operating conditions (i.e., low liquid flow, large catalyst size, and shorter beds). It should be noted, however, that the criterion for removing the axial dispersion effect is available. A similar criterion for removing ineffective catalyst wetting is, however, presently not available. 

Another approach to evaluate the performance of a trickle-bed reactor (particularly a pilot-scale reactor) is to incorporate the RTD with intrinsic kinetics. Since the liquid holdup, catalyst wetting, or the degree of axial dispersion can all be obtained from the RTD, this approach is not exclusive of the ones described above. For a first-order reaction, if the residence-time distribution E(t) and the degree of conversion are known, they can both be related by an expression... 

Mears,53 Paraskos et al.,66 Montagna and Shah,38 and Montagna et al.59 have recently shown that ineffective catalyst wetting can cause the reactor performance to be dependent on the liquid velocity. The y used a correlation of Puranik and Vogelpohl69 for the effectively wetted surface area of the packing to explain the effects ofliquid hourly space velocity and the length of the catalyst bed on the performance of bench-scale HDS reactors. 

Several other reports have also shown the importance of effective catalyst wetting on the performance of a bench-scale trickle-bed reactor. Hartman and Coughlin37 concluded that for sulfur dioxide oxidation in qojjntercurrejQt trickle-bed reactor packed with carbon particles, the catalyst was not completely wet at low liquid flow rates (of the order of 5 x 10 4 cm s-1). Sedricks and Kenney86 found that, during catalytic hydrogenation of crotonaldehyde in a cocurrent trickle-bed reactor, liquid seeped. into dry palladium-on-alumina... 

The effect of catalyst wetting on the performance of a bench-scale trickle-bed reactor was also theoretically evaluated by Sylvester and Pitayagulsarn.94,9S Using the method of moments of Suzuki and Smith,93 they developed a procedure to show the combined effects of axial dispersion, external diffusion, intraparticle diffusion, and surface reaction on the conversion for a first-order irreversible reaction in an isothermal trickle-bed reactor and evaluated the effect of catalyst wetting on these combined effects. 

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. 

The performance of trickle-bed reactors may be affected by many factors, such as interphase mass transfer, intraparticle diffusion, axial dispersion and incomplete catalyst wetting. Therefore, knowledge about these influenced factors is important for their mathematical description by an unsteady-state reactor model. Until now, the literature analysis shows the experimental and theoretical understanding of trickle-bed reactors under unsteady-state-operation conditions has improved, but not considerably. The following studies are focused on the trickling regime under unsteady-state-operation conditions. 

The developed dynamic reactor model for the simulation studies of the unsteady-state-operated trickle-flow reactor is based on an extended axial dispersion model to predict the overall reactor performance incorporating partial wetting. This heterogeneous model consists of unsteady-state mass and enthalpy balances of the reaction components within the gas, liquid and catalyst phase. The individual mass-transfer steps at a partially wetted catalyst particle are shown in Fig. 4.5. 

Sphere-packed monolith reactors for catalytic gas/liquid reactions were studied by Bauer et al. [31]. The authors used this concept for a structured trickle bed reactor and observed a significant increase in the reactor performance, which can be explained by an increased specific surface area and a uniform wetting of the particles. This in turn diminishes the risk of local catalyst overheating, which may harm the catalyst stability and the product selechvity. 

At low liquid flow rates, flow maldistributions such as channeling, bypassing and incomplete catalyst wetting may occur. These adversely affect the reactor performance. 

For trickle beds or monoliths, the catalyst is fixed in space and hence, a uniform concentration is available along the reactor axis. The problem of nonuniform catalyst wetting, however, needs to be addressed to achieve the desired performance. 

Figure 13.33 shows a comparison of semi-industrial and bench-scale performance at LHSV = 0.2 h and 380° C. It is evident that the semi-industrial scale outperforms the bench scale as a result of the effect of liquid flow rate on catalyst utilization. The use of wetting efficiency as the scale-up parameter allows accounting for such an effect on reactor performance. From the previous analysis, it was determined that a wetting efficiency of 0.7 correlates sufficiently well bench-scale and semi-industrial reactor performances. 

Montagna, A., Y. T. Shah. The Role of Liquid Holdup, Effective Catalyst Wetting, and Backmixing on the Performance of a Trickle-Bed Reactor for Residue Hydrodesulfurization. 

Alkanes and Alkenes. For this study, C150-1-01 and C150-1-03 were tested under primary wet gas conditions with ethylene, ethane, propylene, and propane being added to the feed gas. This study was made in order to determine whether these hydrocarbons would deposit carbon on the catalyst, would reform, or would pass through without reaction. The test was conducted using the dual-reactor heat sink unit with a water pump and vaporizer as the source of steam. All gas analyses were performed by gas chromatography. The test was stopped with the poisons still in the feed gas in order to preserve any carbon buildup which may have occurred on the catalysts. 

---