(733d) Case Study on the Application of Aspen HYSYS to Simulate and Optimise the Process of Plastics for the Production of Fuels

The global production of plastics was 311 million tonnes in 2014, and is expected to double over the next 20 years. Between 22-43% of the plastics is discarded in landfills resulting in wasting valuable resources, taking up valuable space and blighting communities. Recovering plastic wastes for recycling or for energy generation has the promise to decrease these problems. The pyrolysis of plastic wastes can provide a solution for these problems, where the plastics undergo a cracking process to produce lower molecular weight products. The products have similar compositions to gasoline, kerosene and diesel compounds, depending on the feedstock and process conditions.

However, the main problem facing plastics pyrolysis research is the complexity of the process with enormous numbers of reactions and hence, there are no models available to describe the complete process. Even that several models (mostly kinetic models) have been developed to describe this process, they have been unsuccessful. Overall, the lack of comprehensive understanding of this process dissuades its transformation into a profitable industry. The main aim of the project is to contribute to pollution prevention and treatment of plastic wastes by providing a valuable mechanism that will improve the research on plastics pyrolysis.

Low-density polyethylene (LDPE) is a type of polyolefin plastic, which is a common domestic plastic waste. Polyolefins account for 57% of the total amount of plastics present in household waste, of which polyethylene is the most abundant type of this group. The objectives of this project are to create hypo components that represent real LDPE, to design an effective and a complete computer-based modelled process that can simulate an actual LDPE pyrolysis process, to establish a method for predicting the pyrolysis products of LDPE with great accuracy, to determine the optimum operating conditions of LDPE pyrolysis for the production of liquid hydrocarbons for fuel generation and to predict the energy requirements for LDPE pyrolysis process.

In this study, two simulation frameworks were proposed in order to design and evaluate the pyrolysis process of low-density polyethylene to produce liquid hydrocarbons that are suitable for fuel production. The first simulation framework (Simple Model) was based on a simplified pyrolysis process where the products were combined into three groups (gas, oils and char), whilst the second one (Complex Model) was based on a simplified pyrolysis process where the detailed product distribution of the liquid fraction was obtained.

Aspen HYSYS was chosen as the modelling software to design and simulate the two frameworks of LDPE pyrolysis process. Aspen HYSYS is a market-leading process modelling tool for conceptual design and optimisation for oil and gas production, gas processing and petroleum refining industries. Since the products of LPDE pyrolysis process are mainly composed of hydrocarbons, HYSYS is suitable tool to model, predict and estimate the products obtained from this process.

LDPE was firstly developed as a number of hypo components in HYSYS. The software estimation methods were accurate for describing a complex material such as plastic. These components were classified into a number groups due to the variability of the composition of LDPE from plastic to plastic. By using HYSYS, the properties of the simulated plastic could be modified, to meet the experimental properties of the actual plastic either by changing the number groups or by changing their compositions. The created hypo components had properties very close to the real ones and exhibited an accurate behaviour when they undergone the pyrolysis process.

The results obtained from the simulation of the Simple Model agreed with the experimental data. This accuracy allowed for the simulation of the process at temperatures that were not achieved experimentally. The behaviours and properties of the three products of the Simple Model at different temperatures were accurately described. The classification of a large number of hydrocarbons into three different groups and the representation of each group with as a single hydrocarbon did not adversely affect the simulation accuracy of LDPE pyrolysis process. Although the material balances of these components were different from real reactions, the pyrolysis process was simulated with greater accuracy in comparison with published results.

The results of the Complex Model simulation showed a clear improvement compared to those of the Simple Model, whilst maintaining a great agreement with the experimental data. Moreover, the description of the pyrolysis process was more detailed, improving the quality of predicting the compositions of the products. Increasing the number of the products also resulted in a more detailed process flowsheet, providing more complete operational specifications, including the energy duties needed for the LDPE pyrolysis process. Pyrolysis products have different applications and the determination of the optimum application for each product is necessary to increase the profitability of the process. Thus, each component needs to be taken into account and managed accordingly.

It was observed, from the simulation results, that it is possible to describe the LDPE pyrolysis in HYSYS with more accuracy than other methods that have been studied previously. Both simple and complex models were proven to be accurate for simulating LDPE pyrolysis process. Overall, both models had shown their appropriateness to overcome the research challenges of plastics pyrolysis.