Gas-solid flow regimes

Spray Flow

Gas-liquid flow regimes

FIGURE 1.9 Some flow regimes of multiphase systems (from Krishna, 1994).

transfer from gas phase to liquid phase is determined by the mass transfer coefficient, interfacial area and concentration driving force for mass transfer. Each of these aspects is intimately related to the underlying fluid dynamics. Overall interfacial area and its distribution within the reactor is controlled by local gas volume fraction and local bubble size distribution. Local as well as global mean and turbulent flow fields control local bubble size. Local mass transfer coefficient is also dependent on local turbulence. Gas volume fraction and its distribution are also intimately related to overall fluid dynamics and interphase coupling of gas and liquid phases. One can imagine that the method of gas introduction (design of gas distributor, location, size and so on) can significantly affect the gas distribution and therefore the overall fluid dynamics of gas-liquid reactors. The flow field will also be very sensitive to reactor internals. This is not typical for gas-liquid reactors only but, in fact, is true for all multiphase reactors. For all multiphase reactors, hardware details and operating conditions have very significant impact on the resulting flow and therefore on the reactor performance. Even small-scale hardware details like the design of feed nozzles, gas distributors and baffles may have a dramatic influence on flow structure. The issues of scale-up and scale-down become much more complicated for multiphase reactors, because not all the relevant properties can be scaled proportionately. For example, in the case of gas-liquid-solid reactors, the relative dimensions of solid particles, gas bubbles and the reactor are bound to change with different reactor scales. Therefore, for reliable scale-up, interphase mass and heat transfer, which ultimately depends on microscopic fluid dynamics near the interface, overall flow patterns and mixing need to be analyzed at all the considered scales separately. An iterative methodology needs to be adopted to design most of these multiphase reactors. Without referring to any specific reactor type, the overall reactor engineering methodology for designing and validating an industrial reactor is shown in Fig. 1.10.

It is often necessary to carry out a small number of iterations over the sequence of steps shown in Fig. 1.10. Reaction engineering with idealized models is used to

Reactor Selection: Flow regimes, Desired mixing, RTD, Heat transfer and so on ■

Preliminary Reactor Configuration: Reactor sizing, vessel configuration (diameter, height to diameter ratio, baffling, jacket/coils, distributors, other internals), Evaluation of design targets (mixing time, transport rates, operating flows & flow regimes), Evaluation of role of mean flow and shear, Evolve hardware details of full scale reactor

Reactor Selection: Flow regimes, Desired mixing, RTD, Heat transfer and so on ■

Preliminary Reactor Configuration: Reactor sizing, vessel configuration (diameter, height to diameter ratio, baffling, jacket/coils, distributors, other internals), Evaluation of design targets (mixing time, transport rates, operating flows & flow regimes), Evaluation of role of mean flow and shear, Evolve hardware details of full scale reactor

Scale Up/ Validation / Final Design: Establishing scale-up rules, Extrapolation using reactor simulation tools, Evolution of configuration of a commercial reactor, Validation and refinements of this configuration based on more detailed flow modeling

Reaction Engineering with Idealized Models: Liquid / slurry phase- complete mixing Gas phase- complete mixing or plug flow No heat transfer limitations Reactor volume for different degrees of mixing and for different values of mass transfer coefficient Heat transfer area for different values of overall heat transfer coefficients

Scale Down/ Reactor Simulations: Pilot scale reactor(s) design (may not be geometrically^ similar), Evaluation of influence of key hardware and operating parameters on RTD, mixing and rates of transport processes, Interpretation/ fitting of pilot scale experimental data- estimation of model parameters, Simulations of pilot and large scale reactors

Commercial Reactor

FIGURE 1.10 Methodology of reactor engineering.

understand the upper and lower bounds on performance and to identify important factors which control the performance. Studies using idealized models are also helpful in determining the desired performance targets for transport processes such as mixing, mass and heat transfer. Engineering creativity, experience and accumulated empirical information is generally used to evolve preliminary reactor configurations. Reactor simulation models are then developed to evaluate these different reactor configurations. Using a conventional methodology, a reactor engineer has to rely on experimental and semi-empirical tools to obtain a knowledge of fluid dynamics, which is essential when addressing many crucial design issues. Several reviews and books have been published which analyze published empirical correlations to estimate parameters necessary for the design and simulation of reactors (for example, see Shah, 1991: stirred reactors; Deckwer, 1991: bubble column reactors; Kunii and Levenspiel, 1991: fluidized bed reactors and so on). Wherever such available information is not adequate, experiments on pilot scale reactors are designed and carried out. The usefulness of pilot scale studies depends on how well these pilot reactors mimic the fluid dynamics and mixing in proposed large-scale reactors. For obvious reasons, such conventional design methods using empirical correlations have rather limited reliability. Uncertainty associated with extrapolation may be unacceptably high. Such methods are not able to relate details of reactor hardware with reactor performance. This non-capability narrows down the choice of reactor configurations. New ideas and new reactor configurations are often sidelined in favor of proven and conventional reactor types when there is an excessive reliance on empirical information and experiments.

To achieve and retain a competitive edge, it is becoming more and more important to address the third question 'what is the best way to carry out the desired transformation?' and to design the reactor hardware and operating protocols accordingly. To answer this question, it is necessary for a reactor engineer to establish accurately the relationship between reactor hardware and reactor performance. Computational flow modeling tools can make substantial contributions in establishing such a relationship.

CFM tools can accelerate the reactor engineering tasks shown in Fig. 1.10 with minimum experimentation on pilot scales and with enhanced confidence levels. Use of CFM for reactor engineering is briefly discussed in Section 1.3 and Parts III and IV of this book. Traditional methods of flow modeling, which rely either on experimental investigations or on making drastic simplifications of the flow problem to allow analytical solutions, are proving to be increasingly inadequate for this purpose. These methods have served us well to establish the reactor engineering profession, which has shaped the present chemical industry. However, to make further progress, it is essential to make creative use of the best possible tools available to reactor engineers. This is especially true for multiphase reactors. The ultimate wish of any reactor engineer is to know the complete history of all the fluid elements flowing through the reactor. This was considered hitherto not possible for most practical reactors. Recent advances in understanding the physics of flows, numerical techniques and digital computers can make tremendous contributions to realizing this ultimate wish by enabling simulations of complex flows in industrial equipment. Of course, there are still many unanswered questions and problems which need to be overcome to realize this dream. It is, however, important that reactor engineers are aware of the potential of these recent advances and are equipped to apply computational flow modeling tools creatively in their endeavors to develop innovative and better reactor hardware and operating protocols. The major features of these advances which are relevant to reactor engineering are discussed in the next section.

0 0

Post a comment