Modelling Helps Improve Safety in the Production of Teflon
We know it as the non-stick coating Teflon, but chemists know it as PTFE (polytetrafluoroethylene). This substance has revolutionized the plastics industry and created a new branch with annual sales in the billions of dollars. Because PTFE is non-toxic, biologically inert and has excellent resistance to chemicals, organic solvents, acids, and alkalis, it is used for piping and valves for the processing of aggressive chemicals and substances. Although PTFE is one of the most useful plastics ever discovered, the polymerization process whereby tetrafluoroethylene (TFE) gas is converted into the solid PTFE can be dangerous. In addition to being highly flammable, TFE belongs to the small group of decomposable gases that are capable of exothermic reactions without the need of an oxidant. Under specific conditions that can even occur during the production process usually when local temperatures reach the range of 500 K an exothermic dimerization of TFE gas can start, leading to the self-heating of the gas phase. In some cases, this can initiate an explosive decomposition reaction.
In an effort to help its member companies better understand how to improve safety in PTFE production facilities and prevent future accidents, the industry organization PlasticsEurope has been subsidizing experimental research for several years. As part of this initiative, Dr. Fabio Ferrero and Dr. Martin -Kluge at the BAM Federal Institute for Materials Research and Testing in Berlin, Germany have been working on the development of a mathematical model of the self heating of TFE. Created using COMSOL Multiphysics, it is the only CFD code used to study this phenomenon.
Prior to using multiphysics simulation, the team at BAM had addressed this issue by conducting tests on small autoclaves where they would determine the Maximum Ignition Temperature of Decomposition (MITD), which depends on the initial pressure and on the vessel geometry. They performed tests for a number of initial pressure vessel volume conditions and interpolated between them. However, it was not possible to conduct such tests on industrial-sized autoclaves because of the very high effort required to deal with the amount of gas and due to the extensive setup and necessary manpower. These restrictions led to the desire for a mathematical model that could predict the behavior of the studied phenomenon in large autoclaves. The model was designed to simulate the self-heating process of TFE to determine the MITD for TFE at elevated pressures, which would help reveal the critical conditions reSsponsible for the self-ignition of TFE.
A particular challenge in developing the model was to identify a suitable chemical reaction mechanism. Until now, researchers have concerned themselves only with the dimerization reac-tion where two TFE molecules “bump” into each other to create a new bigger molecule; this is the main reaction that releases energy to the gas and can result in a runaway condition if the heat builds up to critical levels. However, using this reaction alone resulted in a model with poor correlation to some of the experimental results. After extensive study and research, the team identified a set of reactions that took place in the heated gas phase, but didn’t know which ones were important for the self-heating that leads to explosion.
To test which reactions contributed to runaway, the team used the Chemical Reaction Engineering Module in COMSOL. They were able to include all of the reactions identified as possibly being important in a single model, and were able to identify the six reactions that were needed for an accurate model. The other reactions were ignored because it was determined that they take place at temperatures above the self-ignition temperature, meaning they start only if the system is already experiencing a runaway under the given conditions of the elevated initial pressure. To create more detailed simulations, they utilized the fluid flow, heat transfer, and reaction kinetics interfaces in COMSOL to help isolate hot spots that could initiate the undesired reactions.
Validation of the model was achieved by comparing the experimentally determined MITD with the simulated MITD given by the model, and good agreement was found. The research team has meanwhile conducted a series of validation tests using industrial-sized re-actors which has shown a good correlation to the model. The benefits of having a validated model are plentiful. Companies could determine if, for a given reactor, a specific pressure/temperature setting is not safe and consequently adapt the process. Plant engineers could use the model to determine if and how they should change their process conditions. Furthermore, with a working model, engineers can study additional aspects such as determining the geometric dependence of the self-ignition temperature. Future research for the team includes studying the effects of forced convection, adding piping of various diameters, different flow regimes, and vessels orientations as well the effect of internal features/obstacles.
For more information
Email: [email protected]
