Bubbling Fluidized Bed Reactor

Presentation By Nancy Verma. Bubbling Fluidized Bed Reactor.

Types of reactor. Bubbling Fluidized Bed Circulating Fluidized Bed Flash Reactor Annular Fluidized Bed.

Bubbling Fluidized Bed Reactor.

Principle. Fluid Passes through Bottom with low velocity first to settle down the solid material on the porous plate called Distributor. Fluid passes through the voids of the solid material. This is known as Packed Bed Reactor. After some time when the solid material gets balanced with the Fluid force this stage known as Incipient Fluidization. And this velocity known as minimum fluidization velocity. After this minimum velocity surpassed the reactor become Fluidized bed reactor..

A fluidized bed appears much like a vigorously boiling liquid; bubbles of gas rise rapidly and burst on the surface. Bubbles contains small amount solids. From top collected out..

Pass gas upward through a bed of fine particles. For superficial (or inlet) gas velocities u, much in excess of this minimum the bed takes on the appearance of a boiling liquid with large bubbles rising rapidly through the bed. In this state we have the bubbling fluidized bed, BFB. Industrial reactors particularly for solid catalyzed gas-phase reactions often operate as bubbling beds with gas velocities u, = 5 - 30 u,,. Calculations show that the conversion in bubbling beds may vary from plug flow to well below mixed flow, see Fig. and for many years the perplexing and embarrassing thing about this was that often we could not reliably estimate or guess what it would be for any new situation. Because of this, scale-up was cautious and uncertain, and preferably left to other s..

Conversion of reactant in BFB is usually poorer than for both plug flow and mixed flow.

It was soon recognized that this difficulty stemmed from lack of knowledge of the contacting and flow pattern in the bed: in effect, the bypassing of much of the solids by the rising bubble gas. This led to the realization that adequate prediction of bed behavior had to await a reasonable flow model for the bed. Since the bubbling bed represents such severe deviations from ideal contacting, not just minor ones as with other single-fluid reactors (packed beds, tubes, etc.), it would be instructive to see how this problem of flow characterization has been attacked. A wide variety of approaches have been tried ..

Dispersion and Tanks in series Model. The first attempts at modeling naturally tried the simple one-parameter models; however, observed conversion well below mixed flow cannot be accounted for by these models so this approach has been dropped by most workers ..

RTD Models. The next class of models relied on the RTD to calculate conversions . But since the rate of catalytic reaction of an element of gas depends on the amount of solid in its vicinity, the effective rate constant is low for bubble gas, high for emulsion gas. Thus any model that simply tries to calculate conversion from the RTD and the fixed rate constant in effect assumes that all elements of gas, both slow and fast moving, spend the same fraction of time in each of the phases. As we will show when we treat the details of gas contacting in fluidized beds this assumption is a shaky one, hence the direct use of the RTD to predict conversions ..

Contact Time Distribution Models. To overcome this difficulty and still use the information given by the RTD, models were proposed which assumed that faster gas stayed mainly in the bubble phase, the slower in the emulsion. Gilliland and Knudsen (1971) used this approach and proposed that the effective rate constant depends on the length of stay of the element of gas in the bed, thus short stay means small k and long stay means large k. This approach has also been discarded..

Two-Region Models. Two-Region Models. Recognizing that the bubbling bed consists of two rather distinct zones, the bubble phase and the emulsion phase, experimenters spent much effort in developing models based on this fact. Since such models contain six parameters, see Fig ., many simplifications and special cases have been explored (eight by 1962,15 by 1972, and over two dozen to date), and even the complete six-parameter model of Fig. has been used. The users of this model,.

Two-phase model to represent the bubbling fluidized bed, with its six adjustable parameters, v,, V,, ( DIuL ),, ( DIuL ),, m,, K..

those dealing with FCC reactors, claim that this model fits their data beautifully. However, they had to choose different sets of parameter values for each crude oil feed, in each of their FCC reactors. Also some of the values for their parameters made no physical sense, for example, a negative value for V, or v,. With this as the situation we should also discard this type of model which gives a perfect fit but predicts nothing, and brings no understanding with it. The reason is that we have no idea how to assign values to the parameters for new conditions. Thus this is just a curve-fitting model, and we should be able to do better..

Hydrodynamic Flow Models. Hydrodynamic Flow Models. The discouraging result with the previous approaches lead us reluctantly to the conclusion that we must know more about what goes on in the bed if we hope to develop a reasonable predictive flow model. In particular we must learn more about the behavior of rising gas bubbles, since they probably cause much of the difficulty. Two developments are of particular importance in this regard. The first is Davidson's remarkable theoretical development and experimental verification [see Davidson and Harrison (1963) for details] of the flow in the vicinity of a single rising bubble in a fluidized bed which is otherwise at minimum fluidizing conditions ..

What he found was that the rise velocity of the bubble u,, depends only on the bubble size, and that the gas behavior in the vicinity of the bubble depends only on the relative velocity of rising bubble and of gas rising in the emulsion u,. In the extremes he found completely different behavior , as shown in Fig. For catalytic reactions we are only interested in fine particle beds, so let us ignore the large particle extreme from now on Now , for the fine particle bed gas circulates within the bubble plus a thin cloud surrounding the bubble. Thus the bubble gas forms a vortex ring and stays segregated from the rest of the gas in the bed..

As an example if the bubble rises 25 times as fast as the emulsion gas (not all that uncommon because this ratio is 100 or more in some industrial operations), then the cloud thickness is just 2% of the bubble diameter. This is the regime which interests us..

-z Bubble In fine particle beds Fast bubble Chuded bubble Emulsion In large Micle Slow bubble Cloud bubble -'Emulsion* Extremes of gas flow in the vicinity of rising gas bubbles in BFBs..

Size of particle. Some very fine particle may not be fluidized Depends upon size particle and density Geldart’s classification according to size particle and density ..

THANK YOU.