Simulation and experimental validation: waste cooking oil transesterification using rushton and elephant ear impellers
The performance of two impellers (Rushton and Elephant ear) for waste cooking oil (WCO) transesterification were experimentally and numerically evaluated inside a unbaffled and baffled stirred tank reactor (STR). Three parameters (impeller bottom distance (C), speed, (N) and temperature) were i...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/7087/ http://irep.iium.edu.my/7087/ http://irep.iium.edu.my/7087/1/755-064.pdf |
| Summary: | The performance of two impellers (Rushton and Elephant
ear) for waste cooking oil (WCO) transesterification were
experimentally and numerically evaluated inside a
unbaffled and baffled stirred tank reactor (STR). Three
parameters (impeller bottom distance (C), speed, (N) and
temperature) were investigated for fatty acid methyl esters
(FAME) yield using a Taguchi orthogonal array (OA)
experimental design. The mixing study is based on the �-�
and large eddy simulation (LES) turbulent model using a
multiple reference frame (MRF) approach and the result
was validated by particle image velocimetry (PIV)
measurement. Overall flow structure in the baffled in STR
was isotropic at C = 20, 25 and 30 mm for Rushton and to
a lesser extent at the STR centre in the unbaffled STR.
Mean radial velocities of the Elephant ear impeller was
about 25-30% of the Rushton impeller for 600 rpm and
85-90% in baffled STR at C = 30 mm in the unbaffled
STR. CFD results of �-� and LES results were
qualitatively similar at mid-impeller plane than for the
other parts of the STR. However, none of the model could
capture the vortex trailing encountered with Rushton
impeller. Experimental mixing time, a function of FAME
yield and liquid velocities were found to be dependent on
stirrer speeds, impeller bottom distance and bulk flow
pattern. The study demonstrated the effect of different
impeller types used in the laboratory on FAME yield and
attributed to flow characteristics. This model is being used
to evaluate other conditions of the reactions. |
|---|