Researchers from Group T of KU Leuven investigated and optimized a sink-float separator utilized for sorting plastics prior to their recycling. A computational fluid dynamics (CFD) model was developed to model the flow field of a lab-scale sink float separator, which was also used to validate the model by Particle Image Velocimetry (PIV).
The model has proven to be able to correctly predict the flow field inside the separator with a maximum error of 5% on the average velocity magnitude. Additionally, different densities were introduced in the model to predict the yield and the purity for separating different plastic types, which was again validated by experiments with plastic granules, and depicted a maximum error of 1% in the prediction of the yield.
Eventually, the model was utilised to conduct a parametric analysis on the density range which the separator can efficiently separate and the optimal injection location at which the plastics need to be injected. The key takeaway of this study is that AdRem sink-float separators can achieve high purities of above 98% for plastics with a minimum relative density difference of 0.04 with respect to the fluid’s density.
Second takeaway is that the air-bubbles attachment is the main factor limiting separation efficiency up to 7% due to the hydrophobic properties of various plastic types, thus changing their apparent density.
Based on this study, the separator design has been optimised for improved wetting prior to injection and optimal location and method of injection in the separator, allowing for further enhancement of the plastic sorting efficiency.
This research was done within the project Plastics Recycling 4.0 (HBC.2020.2750) funded by Flanders Innovation & Entrepreneurship agency (VLAIO) in cooperation with the recycling company Galloo and sorting equipment constructor AdRem.
Read the full article here: Investigation of a Sink-Float plastic separator through Computational Fluid Dynamics and Particle Image Velocimetry
