Implementation of Reaction Kinetics into Reactor Model Followed by Validation of the Reactor Model

Abstract

This master thesis project was performed in collaboration with Johnson Matthey. Johnson Matthey is a worldwide company working with sustainable technologies. The site in Perstorp, where this work was performed, focus on formaldehyde production. In order to continuously improve chemical processes, simulation and modeling are advantageous tools. They can give a better understanding of the processes and the optimum operation conditions. By using reactor modeling, design improvements can be investigated, and the efficiency of the plant can be increased.

The aim with this project was to implement a kinetic model into an existing reactor model and to further on validate the reactor model against data from the pilot reactor. The model was developed in Aspen Custom Modeler.

The kinetic model was described by the power law, but at an early stage, numerical errors were obtained with this kinetic model. The power law equations were replaced with another kinetics model from literature. The kinetic model from literature was more complex and more equations were added, including the surface fraction of the components. The development of the kinetic model was an ongoing project and two kinetic models were developed from the literature kinetic model. The rebased model including a reference temperature and the second one with refitted parameters to the experimental data provided from Johnson Matthey.

The rebased model was validated against the pilot reactor and the outlet composition and temperature were compared. The results showed deviation when compared with the pilot plant data, with relative deviation for the temperature profile of +15% and relative deviation for the outlet composition of +2.6% for formaldehyde. A possible explanation for the deviation could be higher reaction rates for reaction 1 and 2. However, the reactor model showed similar behaviour of the pilot plant data regarding the hot spots and similar range of concentration for the main components in the outlet, which shows great potential describing the pilot plant in the future after final validation.

The different layers of catalyst along the tube were implemented in the reactor model and the inert dilution was taking into consideration. When analysing the result in the radial direction, the reactor model showed expected results for the composition and temperature, i.e. constant values for the composition in the radial direction and warmer temperature in the center of the tube in comparison to the wall. This result indicates that the kinetic model was successfully implemented in 2D.

To improve the reactor model, the refitted kinetic model can be evaluated further. This kinetic model with refitted parameters showed problems with convergence and due to the time limit of this project, the convergence problem could not be solved. The mass transfer limitation within the catalyst particle should preferably be implemented as well in order to describe the correct behaviour in the reactor tube.

Popular Abstract

In order to continuously improve chemical processes, it can be advantageous to have a digital model of the process. The model aims to imitate the real process and to describe it as well as possible. With a digital model, the efficiency of a process can be improved and production cost decreased. It can be useful and time saving to run a digital version instead of a real reactor to evaluate how different factors, such as inlet concentration and temperature, affect the performance of the reactor. In this project a model is created to describe a physical reactor and mainly how the reactions occur in the reactor.

Inside a reactor is where the magic of a process occurs - reactants are connected and formed into the desired products. To describe what is happening inside a reactor, experimental measurements has to be performed and from this, a kinetic model can be developed. A kinetic model consists of a set of equations which aim to describe the reactions taking place inside a reactor. In this master thesis, the equations were programmed and a digital kinetic model was developed.

To see how well the digital model performed, a validation was performed. A validation is a comparison against measured data from a physical reactor in order to see how well the digital version describes the real unit. In this case a reactor from the laboratory was used for comparison, called a pilot plant reactor. The correct parameters, such as temperature, pressure, etc., were implemented into the reactor model in order to match the pilot plant data.

The results when running the digital model were compared with measured data from a pilot plant reactor in order to determine how well the model described the pilot reactor. The results evaluated the outlet composition and the temperature profile along the reactor tube. Similar behaviour was obtained for both the composition and the temperature, when compared with the pilot plant data. The absolute value when comparing the temperature deviated slightly and can be explained by different effects. Higher reaction rates can cause a higher temperature in the reactor - when the reactions takes place, heat are released. The reactor tube are cooled from the outside, by a cooling jacket surrounding the tube. If the cooling system is not modeled correctly, a higher temperature can be obtained in the reactor.

The equations were successfully programmed and implemented, and a digital kinetic model was developed. With some further improvement, this reactor model has a very good potential to give matching results compared to the pilot reactor.

• author: Svensson, Rebecka
• supervisor: Helena Svensson
• organization: Chemical Engineering (M.Sc.Eng.)
• course: KETM05 20211
• year: 2021
• type: M1 - University Diploma
• subject: Technology and Engineering
• keywords: Modeling, Reaction kinetics, Chemical Engineering
• language: English
• id: 9055998
• date added to LUP: 2021-06-28 11:45:18
• date last changed: 2021-06-28 11:45:18