Heat Removal from Exothermic Reactors

Given an exothermic reaction operation, an important first step is to compute the adiabatic reaction temperature, that is, the maximum temperature attainable, in the absence of heat transfer. Note that this can be accomplished readily with any of the process simulators. Furthermore, algorithms have been presented for these iterative calculations by Henley and Rosen (1969) and Myers and Seider (1976), among many sources. 

Consider the reaction of carbon monoxide and hydrogen to form methanol  

With the reactants fed in stoichiometric amounts at 25°C and 1 atm, calculate the standard heat of reaction and the adiabatic reaction temperature. 

In ASPEN PLUS, the RSTOIC subroutine is used with a feed stream containing I Ibmol/hr CO and 2 Ibmol/hr H2 and the PSRK method (Soave-Redlich-Kwong equation of state with Holderbaum-Gmehling mixing rules). To obtain the heat of reaction, the fractional conversion of CO is set at unity, with the product stream temperamre at 25°C and the vapor fraction at 1.0. The latter keeps the methanol product in the vapor phase at 2.44 psia, and hence both the reactants and product species are vapor. The heat duty computed by RSTOIC is —38,881 Btu/hr, and hence the heat of reaction is AH, = —38,881 Btu/lbmol CO. 

To obtain the adiabatic reaction temperature for complete conversion, the heat duty is set at zero and the pressure of the methanol product stream is returned to 1 atm. This produces an effluent temperature of 1,158°C (2,116°F), which is far too high for the Cu-based catalyst and the materials of construction in most reactor vessels. Hence, a key question in the synthesis of the methanol process, and similar processes involving highly exothermic reactions, is how to lower the product temperature. In most cases, the designer is given or sets the maximum temperature in the reactor and evaluates one of the heat-removal strategies described in this section. 

---

The highly exothermic nature of the butane-to-maleic anhydride reaction and the principal by-product reactions require substantial heat removal from the reactor. Thus the reaction is carried out in what is effectively a large multitubular heat exchanger which circulates a mixture of 53% potassium nitrate [7757-79-1/, KNO 40% sodium nitrite [7632-00-0], NaN02 and 7% sodium nitrate [7631-99-4], NaNO. Reaction tube diameters are kept at a minimum 25—30 mm in outside diameter to faciUtate heat removal. Reactor tube lengths are between 3 and 6 meters. The exothermic heat of reaction is removed from the salt mixture by the production of steam in an external salt cooler. Reactor temperatures are in the range of 390 to 430°C. Despite the rapid circulation of salt on the shell side of the reactor, catalyst temperatures can be 40 to 60°C higher than the salt temperature. The butane to maleic anhydride reaction typically reaches its maximum efficiency (maximum yield) at about 85% butane conversion. Reported molar yields are typically 50 to 60%. 

Since most of the reactions are exothermic, liberating 120 to 200 kJ/mol of reactant, the well-mixed liquid medium allows rapid convective heat removal from the reactor. 

An equation relating temperature T and conversion Xa is required to design the non-isothermal reactors. This relationship between temperature T and conversion is obtained by setting up a heat balance equation around the reactor (Section 3.1.5.3). In certain cases, reactor temperature T is deliberately varied with conversion by regulating the heat supply to the reactor or heat removal from the reactor. One such case is the non-isothermal reactor in which a reversible exothermic reaction is carried out. In the case of a reversible exothermic reaction, there is an optimum temperature T for every value of conversion x at which the rate is maximum. A specified conversion Xaj will be achieved in a CSTR or a PFR with the smallest volume or in a batch reactor in the shortest reaction time if the temperature in the reaction vessel is maintained at the optimum level. This optimal temperature policy in which temperature is varied as a function of conversion x,i is known as the optimal progression of temperature presented in the following section. 

Until recently, the temperature control of highly exothermic reactions using the microreaction systems was mainly based on the removal of heat in order to prevent hot spot formation and thermal runaway [29]. More recently, however, research has focused on techniques that enable microreactors to be heated because they can efficiently dissipate the heat. If a microheat exchanger is integrated into a microreactor, both effects can be combined, that is, either enabling fast heat supply in the reactor or heat removal from the reactor [30]. In practice, strongly exothermic reactions such as nitration, oxidation, chlorination, and even fluorination with elementary fluorine (in microreactors made of nickel) can be carried out in microreactor systems under nearly isothermal conditions [31]. 

Because the reaction is exothermic, heat must be removed from the reactor to keep the temperature from increasing out of control. The heat transfer rate of removal is ... 

Influence of reactor size/shape on process performance via heat transfer. Heat transfer between a reaction mixture and its surroundings strongly depends on the size of the reactor. The amount of heat evolved during an exothermic reaction is proportional to the volume of the reaction mixture (Qgcncrated V), ie. the effective volume of the reactor, whereas the amount of heat removed from the mixture is proportional to the heat-transfer surface area (firemovod A,). Morc precisely, the amount of heat transferred is given by ... 

In the auto-refrigerated reactor shown below, an exothermic reaction A —> B is carried out using a low boiling solvent C. The heat of reaction is removed from the reactor by vapourising the solvent, condensing the vapour in the reflux condenser and returning the condensate as saturated liquid to the reactor. The total holdup of liquid in the reactor is maintained constant, but the temperature of the reactor is controlled by regulating the mass flow of vapour to the condenser. The example is taken from the paper of Luyben (1960). 

Example 1.3. Our third example illustrates a typical control scheme for an entire simple chemical plant. Figure 1.5 gives a simple schematic sketch of the process configuration and its control system. Two liquid feeds are pumped into a reactor in which they react to form products. The reaction is exothermic, and therefore heat must be removed from the reactor. This is accomplished by adding cooling water to a jacket surrounding the reactor. Reactor elHuent is pumped through a preheater into a distillation column that splits it into two product streams. 

The formation of phthalic anhydride is highly exothermic, and even with the most careful design the heat removal from packed bed reactors can become uncheckable, leading to temperature runaways, meltdowns, and even explosions. If the chief engineer of those reactors had been required to sit on the reactor during start up, there would be fewer chief engineers about. ... 

This mode is used industrially for exothermic reactions such as NH3 oxidation and in CH3OH synthesis, where exothermic and reversible reactions need to operate at temperatures where the rate is high but not so high that the equilibrium conversion is low. Interstage cooling is frequently accomplished along with separation of reactants from products in units such as water quenchers or distillation columns, where the cooled reactant can be recycled back into the reactor. In these operations the heat of water vaporization and the heat removed from the top of the distillation column provides the energy to cool the reactant back to the proper feed temperature. 

Almost all polymerization processes are highly exothermic, and this heat must be removed from the reactor. The temperature must be very tightly controlled in a polymerization process because each step has a highly temperature-dependent rate coefiflcient so the properties of the resulting polymer will depend sensitively on temperature and its variations in the reactor. 

The active Cu containing catalyst is also very air sensitive (like the Ni reforming catalyst) and will spontaneously oxidize generating uncontrolled reaction heats. Thus it must be passivated before discharged and exposed to air. A small amount of air is added to the reactor and the temperature monitored. This process is continued until the exotherm is small enough that the catalyst can be safely removed from the reactor. 

We conclude that most reaction systems in the chemical industries are exothermic. This has some immediate consequences in terms of unit operation control. For instance, the control system must ensure that the reaction heat is removed from the reactor to maintain a steady state. Failure to remove the heat of reaction would lead to an.accumulation of heat within the system and raise the temperature. Forreversible reactions this would cause a lack of conversion of the reactants into products and would be uneconomical. For irreversible reactions the consequences are more drastic. Due to the rapid escalation in reaction rate with temperature we will have reaction runaway leading to excessive by-product formation, catalyst deactivation, or in the worst case a complete failure of the reactor possibly leading to an environmental release, fire, or explosion. 

Step 3 of our plantwide control design procedure involves two activities. The first is to design the control loops for the removal of heat from exothermic chemical reactors. We dealt with this problem in Chap. 4, where we showed various methods to remove heat from exothermic reactors and how to control the temperature in such reactors. At that point we assumed that the heat was removed directly and permanently from the process (e.g., by cooling water). How-ever. it is wasteful to discard the reactor heat to plant utilities when we need to add heat in other unit operations within the process. Instead, a more efficient alternative is to heat-integrate various parts of the plant by the use of process-to-process heat exchangers. 

To ensure exothermic heat removal from the process, we are constrained by the process design to assign two control loops. We must control reactor inlet temperature with the furnace and control reactor exit temperature with the quench flow. Only by adjusting fuel to the furnace do we allow the reactor heat to be dissipated to the cooler. And only when the quench loop works do we guarantee that the furnace is operational. Because of this design we do not need a bypass line around the process-to-process heat exchanger. 

The exothermic reactions occur in a reactor containing tubes packed with a precious metal catalyst on a silica support. Heat is removed from the reactor by generating steam on the shell side of the tubes. Water flows to the reactor from a steam drum, to which makeup w-ater (boiler feeder water BFW) is supplied. The steam leaves the drum as saturated vapor. The reactions are irreversible and the reaction rates have an Arrhenius-type dependence on temperature. 

When two reactions on the two opposite sides of a membrane are coupled or conjugated through the membrane, not only species balances including reaction and permeation need to be maintained but also heat balance needs to be established. From an engineering standpoint, it is desirable to have one of the coupled reactions endothermic while the other exothermic and the heat transfers through the shared membrane. If the supply and demand of heat is not properly balanced as it often occurs, net heat input or removal from the reactor will be required. This poses a technical challenge. 

---