Feed Effluent Heat Exchanger design

Feed/effluent heat exchangers are used in many industrial processes to warm up the fluid before the reactor and to cool it down after treatment at high temperature. The conventional design of such heat exchangers is based on shell-and-tube units. But to increase the thermal effectiveness of the heat exchangers, the required heat length becomes very important, and high pressure drop will occur. 

The feed-effluent heat exchanger is assumed to be single-pass, countercurrent shell-and-tube design. Three partial differential equations are used for the temperatures of gas on tube side, gas on shell side, and the tube metal Eqs. (6.11), (6.12), and (6.13), respectively. The overall heat transfer coefficients on both tube and shell sides, U, and U are constant and are equal to 0.284 kJ s-1 m-2 K-1. Equal heat transfer area per volume is assumed for the shell and tube sides (A,/V, = As/Vs) and is 157 m2/m3, based... 

The impact of several design parameters has been explored. Control performance worsens when the steady-state economic optimum design, consisting of a large feed-effluent heat exchanger and a small furnace, is used. The most robust control is obtained when a small FEHE and a large furnace are employed. 

Fig. 10 shows a typical SO2 multistage reactor system manufactured by Zieren-Chemiebau. This particular design has feed-effluent heat exchangers and cold-shot interstage cooling. 

As we have seen in Example 1.2, operation at Pi is inherently unstable and we need to design a stabilizing controller, which should be very robust to maintain stable operation. The design of such controller may not always be possible because either the reaction is extremely sensitive or its mechanism and parameters are poorly known. In such a case we need to redesign the reactor system and make it inherently stable (Figure 25.2c) or eliminate the feed-effluent heat exchange. 

The last example is a gas-phase process with a tubular reactor, gas recycle compressor, feed-effluent heat exchanger, condenser and separator. The steady-state design of this process leads to an uncontrollable system if the reactions are highly temperature sensitive. We demonstrate that changing the design produces a much more easily controlled process. We consider a complete plant, not just the reactor in isolation. 

If a design is used where the feed/effluent heat exchanger is located in a separate pressure shell, then the converter outlet temperature becomes equal to the catalyst bed outlet temperature, and the converter outlet system must be designed for this higher temperature. In such systems the waste heat may be recovered at higher temperature so that it can be used, for example, for production of high pressure steam. 

Under autothermal operation the hot effluent of a fixed-bed reactor is used to heat up the cold feed to the ignition temperature of the catalytic reaction. Since no other addition or removal of heat takes place, autothermal operation is restricted to reaction systems which all together are exothermic. The conventional reactor design consists of an adiabatic packed-bed reactor coupled with a countercurrent heat exchanger (Fig. 23A). 

---