Comparison with Semibatch Operation

Comparison between the heat exchanged per unit of volume during oxidation experiment in the Shimtec reactor and the maximal heat exchanged in a classical batch reactor (with a double jacket) highlights the effectiveness of the former. Indeed, in oxidation reaction experiments, a mean value of the heat exchanged per unit of volume in the HEX reactor is estimated with utility stream temperature of 47 °C  

the quantity of heat that could be removed in batch reactors whose volume varies from 11 to 1 m is calculated. In order to compare with experimental results, the temperature gradient is fixed at 45 °C (beyond which water in the utility stream would freeze and another cooling fluid should be used). The maximum global heat-transfer coefficient is estimated at an optimistic value of 500 W m K h The calculated value of the global heat transfer area of each batch reactor. A, is in the same range as the one given by the Schweich relation [35]  

Since the final volume of the reactant mixture (equal to the volume of the reactor) and that of the mass fraction of sodium thiosulfate are known, the total mass of sodium thiosulfate that has to be added is also known. 

Clearly, the oxidation reaction could not have been implemented in a pure batch operating reactor. Indeed, heat removal capacity would not have been sufficient (100—1200 kW m removed versus 20 x 10 kW m generated). As a consequence, a semibatch mode is necessarily required. Besides, Table 12.10 shows that the feeding times are much higher than the residence time of the Shimtec reactor (around 15 s). 

As expected, heat exchanged per unit of volume in the Shimtec reactor is better than the one in batch reactors (15-200 times higher) and operation periods are much smaller than in a semibatch reactor. These characteristics allow the implementation of exo- or endothermic reactions at extreme operating temperatures or concentrations while reducing needs in purifying and separating processes and thus in raw materials. Indeed, since supply or removal of heat is enhanced, semibatch mode or dilutions become useless and therefore, there is an increase in selectivity and yield. 

---

In comparison with a batch reactor, the gradual or intermittent addition of a reactant (say, B in A + B -> products, Figure 12.3(a)) in semibatch operation can result in improved control of T, particularly for a reaction with a large... 

Example 5-5 Hexamethylenetetramine (HMT) is to be produced in a semibatch reactor by adding an aqueous ammonia solution (25 wt % NH3) at the rate of 2 gpm to an initial charge of 238 gal (at 25°C) of formalin solution containing 42% by weight formaldehyde. The original temperature of the formalin solution is raised to 50°C in order to start the reaction. The temperature of the NH4.OH solution is 25°C. The heat of reaction in the liquid phase may be, assumed independent of temperature and concentration and taken as —960 Btu/lbbf HMT. If the reactor can be operated at a temperature of 100°C, the rate of reaction is very fast in comparison with the rate of heat transfer with the surroundings. Temperatures higher than 100°C are not desirable because of vaporization and increase in pressure. 

In comparison with the continuous mode of operation, the mean size was found to depend to a greater degree on the mixing conditions on all scales in the semibatch mode. 

Deterministic optimization has been the common approach for batch distillation operation in previous studies. Since uncertainties exist, the results obtained by deterministic approaches may cause a high risk of constraint violations. In this work, we propose to use a stochastic optimization approach under chance constraints to address this problem. A new scheme for computing the probabilities and their gradients applicable to large scale nonlinear dynamic processes has been developed and applied to a semibatch reactive distillation process. The kinetic parameters and the tray efficiency are considered to be uncertain. The product purity specifications are to be ensured with chance constraints. The comparison of the stochastic results with the deterministic results is presented to indicate the robustness of the stochastic optimization. 

---