【正文】 (26)whereQM = hiAi(T ? TM) (27)andQ0 = h0A0(TM ? TJ) (28)The energy balance for jacket fluid of batch and semibatch reactor: (29)The behaviour of the jacket fluid has been considered as a perfect mixing as it is indicated by Luyben [14].3. Plant descriptionThe experimental measures of pH, conductivity, voltage and temperature have been obtained in a pilot scale reactor (Fig. 1). It consists of a 5l glassjacketed reactor provided with a data acquisition system based on GPIB bus and PC software. The flow rate of the circulating fluid in the jacket may be prefixed or controlled by differential on–off valves, providing an alternative heating–cooling fluid (hot water or cold water). Table 1 indicates the state of the valves for the different modes of operation. All physically available analogue inputs and outputs as well as all virtual channels are automatically monitored andTable1State of the valvesValvesActionIaIIbIIIcV1Reactive A inletX X(?)V2Reactive B inletXV3Jacket fluid inletXXV4Jacketfluid outletXXV5Products outletXa Charge the reactor.b Reaction development (?) only in semibatch mode.c Empty the reactor (when the experiment is finished).process values are stored. It is also possible to obtain online configured curves on the display screen.4. Development of the experimentsPreliminary experiments to obtain the kinetic expression were carried out at different conditions for each reaction and in batch mode of operation. Then, the two reactions indicated were carried out in the two modes of operation: batch and semibatch. For all the experiments, the jacket was filled with water circulating at a flow of 10?5 m3 s?1. The selected setpoint temperature in the reactor was 308 K, and the volume of reactant A used was l to mix with l of reactant B, to obtain 4 l of reacting mass. Reactant A is NaOH in the saponification reaction and H2O2 in the oxidation reaction. Experiments 1 and 2 correspond to the saponification reaction in batch and semibatch mode of operation respectively, and Experiments 3 and 4 correspond to the oxidation reaction in batch and semibatch mode of operation respectively. Table 2 indicates the operation conditions. Experiments for the oxidation reaction were carried out in similar conditions for the reactant volumes to those used in the saponification experiments. In these conditions initial concentrations in the reactor are rather diluted. For the reaction of thiosulfate with hydrogen peroxide, several works (Szeifert et al. [15]) use more concentrated solutions with a ratio cB0/cA0 = 2/3. Experiments 5 and 6 were examples of these conditions (Table 3). The initial concentrations of the reactants were half of those indicated in the table, because the volume of the two solutions was the same. These experiments were used to validate the mathematical model which was implemented in a software Table 2Operation conditions for Experiments 1–4 Saponification reactionOxidation reactionExperiment1Experiment2Experiment3Experiment4T （ K）TJO（K）cA（M）11cB（M）22FO（m2s1）107107Table 3Operation conditions for Experiments 5–8T（K）Tset（K）TJO（K）cA（M）cB（M）Fo（m2s1 ）Experiment5106Experiment6293106Experiment7313106Experiment8313106module written in FORTRAN 77 language. Experiments 7 and 8 were carried out with similar conditions to those of Experiment 5, but introducing the setpoint temperature at 313 K.5. Results and discussionThe Arrhenius equation was obtained for the saponification and oxidation reactions. These results show a good agreement with those found by other investigators (Ortiz et al. [16]。 Cohen and Spencer [7]). The two reactions show a very different reaction rate and thermal behaviour. Indeed, the oxidation reaction is faster than the saponification reaction because: ,and (Ea/R)acid–base = , in front of 1011 m3 kmol?1 s?1, and 9156K for the oxidation reaction. To pare the behaviour of the batch and semibatch mode of operation, the temperature profiles are shown in Fig. 2 for both reactions. For the saponification reaction, the initial slope and the overshoot, in the batch mode, are much more pronounced than in semibatch mode. For the oxidation reaction, this difference is even more important. In the batch mode it is impossible to control the temperature evolution. For this reason this reaction will not be studied in this operation mode in this work. It is important to remark that for the saponification reaction the jacket fluid was initially hotwater (Table 2), because the reaction heat is very low and if jacket fluid would have been cold water, it would have not been possible to see the temperature rise. To follow the concentration evolution of reactants, for the saponifaction reaction the pH was measured. Using Eqs. (14)–(19) it was possible to convert pH profiles into concentration profiles. The corresponding concentration evolution of reaction mixtures for batch and semibatch reactors is shown in Figs. 3 and 4, respectively. The feed of A is stopped after 1800 s when the stoichiometric input value is reached. It is possible to see the accumulation of NaOH added (CA) in the reaction mixture. Once the maximum is reached, reaction evolution is similar to that of a batch reactor. With the same initial conditions, simulation was carried out based on mass balance equations (Eqs. (6)–(13)) as it is indicated in a previous work (Grau and Puigjaner [17]).Concentration profiles for the oxidation reaction using experimental data were obtained only for the adiabatic mode of operation. Using Eq. (23) the conversion was obtaine