(494i) Relationship between Flow Pattern in Cyclone Classifier and Classification Process

Size classification is an important unit operation in process industries, such as mineral processes, food processing, resources recycling and chemical engineering. Various types of turbo air classifiers have been developed. Air classification is a dry-state classification method which separates the powder by air flow. To date, both dynamic turbo- and static air classifiers are widely used for dry classification. The turbo air classifier usually works with turbine, which mainly generates a forced vortex for classifying the materials. It is well known that both capital cost and operational energy consumption of turbo air classifier are much higher than that of static air classifier. In additional, the static air classifier is more applicable for harsh environment such as high temperature. The improvement for the classification efficiency of the static air classifier is worth studying. With the rapid development of the computational fluid dynamics (CFD), extensive numerical simulations were focused on air classifier and achieved much success.

The objective of this research is to investigate the relationship between the gas flow pattern and classification performance of cyclone classifier. Two different cyclone classifiers are designed and manufactured. The simulated results show that the tangential velocity profile resembles a Rankine vortex in the cyclone classifier. The traditional reverse-flow cyclone classifier has a higher maximum tangential velocity compared to the double-vortex cyclone classifier. The maximum tangential velocity is 2.5~2.72 times the inlet velocity in traditional reverse-flow cyclone classifier while that is 1.6~1.83 times the inlet velocity in double-vortex cyclone classifier. The result suggests that the centrifugal force acting on particles is larger in reverse-flow cyclone classifier. The axial velocity distribution differs greatly in the two type classifiers. Air goes downwards near the wall and upwards near the centre in the reverse-flow cyclone classifier. However, for the double-vortex cyclone classifier, the axial velocity value is positive near the wall above the air inlet and gas forms upstream. The upward airstream helps to clean the coarse particles and decreases the fines content in coarse particles. The experimental results with fluidized catalytic cracking catalysts demonstrate that the collected coarse powder increases averagely by 4% for the reverse-flow cyclone classifier, compared to the double-vortex cyclone classifier. This result can be well explained by the simulated velocity profile, which indicates that the tangential velocity has larger value in the reverse-flow cyclone classifier. However, the classification accuracy is higher for the double-vortex cyclone classifier. The results suggest that a suitable classification flow field may not be very strong in intensity, but recirculation of the fine particles is also important. The findings are useful to better understanding and designing the cyclone air classifier.