Liquid holdup operating

High recovery of a volatile component by a batch operation is required. Liquid holdup is much lower in a packed column. 

In bubble-flow operation, the gaseous phase moves upwards as discrete bubbles. The liquid phase may be in either co- or countercurrent flow. The liquid holdup is relatively high. 

The experimental and theoretical work reported in the literature will be reviewed for each of the five major types of ga s-liquid-particle operation under the headings Mass transfer across gas-liquid interface mass transfer across liquid-solid interface holdup and axial dispersion of gas phase holdup and axial dispersion of liquid phase heat transfer reaction kinetics. 

In this section, four other papers may be finally referred to, which, in addition to liquid holdup, deal with various other aspects of liquid flow in trickle-bed operation. 

The central difficulty in applying Equations (11.42) and (11.43) is the usual one of estimating parameters. Order-of-magnitude values for the liquid holdup and kiA are given for packed beds in Table 11.3. Empirical correlations are unusually difficult for trickle beds. Vaporization of the liquid phase is common. From a formal viewpoint, this effect can be accounted for through the mass transfer term in Equation (11.42) and (11.43). In practice, results are specific to a particular chemical system and operating mode. Most models are proprietary. 

To operate the larger column at a reduced rate probably will not be too difficult if a bubble cap is specified. These columns have wide stable operating ranges. The only difficulties are that the tray efficiency may be low and the liquid holdup relatively large. This latter is especially undesirable if, as has been noted, the liquid is flammable, or if some undesired reaction takes place at the elevated temperatures within the column. 

Liquid holdup is critical in the downflow operation of fixed beds, in contrast to the upflow operation where the liquid occupies practically the whole external free void volume of the bed. Total liquid holdup ht consists of two parts static h, and dynamic holdup liA. Static holdup is related to the volume of liquid that is adherent to the particles surface, whereas dynamic holdup is related to the flowing pari of the liquid. 

In a reactor completely filled with liquid, the wetting efficiency is 100% or, in other words, the external wetting of the catalyst is complete (Burghardt et al., 1995). While it is true that when a fixed bed is completely filled with liquid wetting is complete (wetting efficiency is unity), the opposite is not true in a trickle bed, a portion of the bed voids will be always occupied by the gas phase. Thus, while in a well-operated trickle bed the wetting efficiency could be unity, its total liquid holdup based on the void volume is always lower than the bed voidage, i.e. the bed is never completely filled with liquid. 

The effectiveness of a fixed-bed operation depends mainly on its hydraulic performance. Even if the physicochemical phenomena are well understood and their application in practice is simple, the operation will probably fail if the hydraulic behavior of the reactor is not adequate. One must be able to recognize the competitive effects of kinetics and fluid dynamics mixing, dead spaces, and bypasses that can completely alter the performance of the reactor when compared to the ideal presentation (Donati and Paludetto, 1997). The main factor of failure in liquid-phase operations is liquid maldistribution, which could be related to low liquid holdup in downflow operation, or other design problems. These effects could be critical not only in full-scale but also in pilot- or even in laboratory-scale reactors. 

For the evaluation of the particle Peclet number and the liquid holdup, the correlations proposed by Inglezakis et al. are used, i.e. eqs. (3.313) and (3.332), respectively. The Biot number, liquid holdup, and bed Peclet number for downflow operation versus relative volumetric flow rate are presented in Figure 4.35. 

It is obvious that for the whole flow-rate range, the rate-controlling mechanism is expected to be the solid diffusion control (Hi > 41). Furthermore, the flow can be characterized as ideal plug flow for flow rates above 2.15 BV/h, where PeL is higher than about 100. However, the liquid holdup is very low (56.83%) and this could be proved a serious problem, even with the use of a liquid distributor at the top of the bed. In order to have a satisfying liquid holdup, i.e about 80%, the relative flow rate should be about 5.62 BV/h. Then, by means of a liquid distributor at the top of the bed, it is possible to achieve a holdup near 100%. Thus, for downflow operation the limits of the relative flow rate are (BV/h)... 

On the other hand, in upflow operation, Pe, is higher than 140 for the whole relative flow-rate region up to the value of 6.66 BV/h. At the same time, the liquid holdup is always... 

Figure 4.35 Biot number, liquid holdup, and bed Peclet number for downflow operation versus relative volumetric flow rate.

From the hydraulics perspective, if scale-up is based on the same superficial velocity, Peh will be higher in a large bed in downflow operation due to the higher bed height, whereas the liquid holdup will be low due to the low velocity, which is frequently used in laboratory beds. This leads to problems and special efforts are required to improve the liquid holdup, for example, a special distributor design. These problems are absent in upflow operation. 

Figure 5.2-25. Influence of the total reactor pressure and of the liquid flow-rate on the dynamic liquid holdup for the single water operation (after Wammes et al. [34]).

Since liquid does not completely wet the packing and since film thickness varies with radial position, classical film-flow theory does not explain liquid flow behavior, nor does it predict liquid holdup (30). Electrical resistance measurements have been used for liquid holdup, assuming liquid flows as rivulets in the radial direction with little or no axial and transverse movement. These data can then be empirically fit to film-flow, pore-flow, or droplet-flow models (14,19). The real flow behavior is likely a complex combination of these different flow models, that is, a function of the packing used, the operating parameters, and fluid properties. Incorporating calculations for wetted surface area with the film-flow model allows prediction of liquid holdup within 20% of experimental values (18). 

During vaporization or condensation, thermal and hydraulic performances depend essentially on the two-phase flow structure. Furthermore, as very often in industrial processes, the heat exchanger operates in thermosyphon or under natural circulation knowledge of the pressure drop and liquid holdup is of major importance. But there is very little information on two-phase flow characteristics in compact geometries. 

Depending on the gas and liquid residence times required, the reactor could be operated horizontally or vertically with either downflow or upflow. Weikard (in Ullmann, Enzyklopaedie, 4th ed., vol. 3, Verlag Chemie, 1973, p. 381) discusses possible reasons for operating an upflow concurrent flow tubular reactor for the production of adipic acid nitrile (from adipic acid and ammonia). The reactor has a liquid holdup of 20 to 30 percent and a residence time of 1.0 s for gas and 3 to 5 min for liquid. 

Countercurrent Flow The gas flows up countercurrent with the downflow liquid. This mode of operation is not as widely used for catalytic reactions since operation is limited by flooding at high gas velocity at flooding conditions increasing the liquid flow does not result in increase of the liquid holdup. 

For single separation duty, Farhat et al. (1990) considered the operation of an existing column for a fixed batch time and aimed at maximising (or minimising) the amount of main-cuts (or off-cuts) while using predefined reflux policies such as constant, linear (with positive slope) and exponential reflux ratio profile. They also considered a simple model with negligible liquid holdup, constant molar overflow and simple thermodynamics, but included detailed plate to plate calculations (similar to Type III model). 

In the first one,the apparent reaction rate is empirically related to the operating conditions on the basis of experimental observations. The models suggested by Henry et al. (3) and by Mears (4) belong to this category. The apparent reaction rate is assumed to be proportional to the liquid holdup in the first model and to the catalyst irrigation rate in the second one. These hydrodynamic quantities are estimated using empirical correlations based on experiments. 

---