Application of calorimetric testing and dynamic simulation to predict and control violently reactive chemical reactions

The purpose of this thesis is to demonstrate the use of calorimetric testing and dynamic simulation to predict and prevent the runaway reactions in a chemical process. A simple distillation process involving a complex runaway reaction is used to demonstrate this concept. An Accelerated Rate Calorimeter (ARC) is used to obtain reaction data (i.e. heats of reaction, reaction rates, reaction by products, etc.). The reaction data is then used to develop a dynamic simulation of the distillation process for the purpose of evaluating failure scenarios that may trigger a runaway reaction. Finally, the simulator is used to assess the performance of different emergency safety systems (such as emergency shutdown systems, quench systems, dump systems, etc.) to prevent a potential runaway reaction.

Three failures scenarios (loss of cooling, loss of vacuum, and excess heat) are simulated in the refining process. The simulation results indicate that a typical emergency shutdown strategy (ESD) will prevent vessel over-pressurization in two of the three cases. For loss of vacuum, however, the emergency shutdown system, by itself, is ineffective for preventing vessel over-pressure. The simulation indicates that a reduction of the ESD initiation temperature and the addition of an emergency dump system can significantly reduce the potential for vessel over-pressurization.