Periodic Reporting for period 1 - IMPRECSIM (Improving wet plastic recycling through innovative lagrangian particle-fluid simulations)
Période du rapport: 2015-06-01 au 2016-11-30
Mechanical plastic recycling is currently one of the weakest points in the recycling system, because only a low percentage of plastic is reused compared to the amount of recovered metal, glass and waste paper. Recycling of wastes generally requires the separation of different materials. As the raw mixture of plastic waste usually includes various kinds of plastics (e.g. Acrylonitrile-butadiene-styrene (ABS), Polyethylene terephthalate (PET), Polystyrene (PS), Polyethylene (PE), Polypropylene (PP), Polyvinyl chloride (PVC)), the separation process should classify waste into a number of reclaimable plastic fractions, so as to meet the requirements for the purity and cleanliness of a polymer type that are needed in a high-quality plastic recycling process. The separation of different polymers by type is necessary because contamination in recycling of one type of plastic by another type can severely lower the quality of plastic products and cause serious processing problems.
Wet particle separation is used widely in mineral processing as well as plastic recycling to separate mixtures of particulate materials into further usable fractions due to density differences. Despite its wide usage wet particle separation processes are often attributed to operational problems especially if density differences of the feed material are low. Objective of this project was to develop a numerical method with which the wet plastic particle separation process can be simulated and to apply the developed framework for modelling of a technical wet separation process.
A comparative study on mesh-based and mesh-less (SPH) coupled CFD-DEM (Computational Fluid Dynamics – Discrete Element Method) methods to model particle-laden flow was performed. In general, results obtained using both approaches agreed well with analytic reference results. Numerical differences were found mostly due to difference in computed fluid fractions that result in different drag forces. The proposed new model to account for boundary conditions in the SPH approach was demonstrated to produce accurate results in the presented verification tests. They proved to be convenient and stable in the context of DEM-SPH simulations.
In the second part of the project, the developed framework was applied to the modelling of a wet separation process involving a sink-float drum separator for plastic recycling. The numerical analysis of polyethylene terephthalate (PET) particle separation from polypropylene (PP) particles in the rotating drum was performed. The influence of different operational and design parameters, such as the rotational velocity, the number of lifters, the feed rate etc., was analysed. Numerical results show, that the use of the separating vertical walls, lower rotational velocity, higher number of the lifters and the higher feed velocity by water increases the purity of the separated particles.
Dissemination of the results:
1) Published paper in a journal:
D. Markauskas, H. Kruggel-Emden, R. Sivanesapillai, H. Steeb. Comparative study on mesh-based and mesh-less coupled CFD-DEM methods to model particle-laden flow. Powder Technology. 305 (2017) 78-88.
A version of the final peer-reviewed manuscript accepted for publication is available on arXiv.org: arxiv:1603.06808
2) Presentation of the results in a conference:
Talk: Numerical analysis of wet separation of particles by density differences. Presenter: D. Markauskas, Co-author: H. Kruggel-Emden. 14th International Conference of Numerical Analysis and Applied Mathematics, ICNAAM 2016, 19-25 September 2016, Rhodes, Greece.
3) Accepted manuscript for publication in a conference proceeding:
D. Markauskas, H. Kruggel-Emden. Numerical analysis of wet separation of particles by density differences. Proceedings of 14th International Conference of Numerical Analysis and Applied Mathematics, ICNAAM 2016, 19-25 September 2016, Rhodes, Greece, 4 p.
A version of the final peer-reviewed manuscript accepted for publication is available on arXiv.org: arxiv:1609.08421
4) Submitted manuscript for publication in a journal:
D. Markauskas, H. Kruggel-Emden, V. Scherer. Numerical study of wet plastic particle separation using coupled DEM-SPH method. Manuscript is submitted for the review in the journal Powder Technology.
The numerical modelling has been applied for the wet plastic particle separation for the first time. The lack of earlier research in this area can be explained by the lack of reliable, predictive computational methods being available that would be applicable for such analysis. A coupled DEM-SPH framework, developed during this research, is a powerful tool enabling a detailed analysis of this process with regards to operational parameters and design specifications. By numerical modelling time consuming extensive experimental investigations could be avoided and especially separation of plastic mixtures of similar density at elevated throughput could be optimized.