Researcher(s)
- Tiffany Jung, Chemical Engineering, University of Delaware
Faculty Mentor(s)
- William Hartt, Chemical Engineering, University of Delaware
- Alexandra Bayles, Chemical Engineering, University of Delaware
Abstract
Static mixers (ex. SMX and Kenics) are used for mixing polymers and other materials, measuring rheological properties, etc. and are characterized by unique geometries. Pressure drop is important since it drives fluid flow and is related to properties such as density, viscosity, etc. These properties can be combined into the Reynolds number, often plotted against the friction factor, giving rise to quantities such as KL for laminar static mixer flows. Computational fluid dynamics (CFD) digitally models fluid flow using conservation of mass, energy, and momentum, making it suitable for complex geometries. Since static mixers require significant investment of time and resources, using CFD is less costly and faster. In this poster, CFD was used to simulate laminar flow through an SMX mixer and calculate pressure drop, later used to understand fluid-pipe interactions (characterized by the Darcy friction factor). The friction factor was then compared to data obtained experimentally, therefore leading to the primary question of this research: how well does CFD data compare to experimental?
Simulations were run in Ansys Fluent at different Reynolds numbers for glycerol-water solutions to collect pressure drop data. This data was then used to calculate the Darcy friction factor which was plotted vs. the inverse Reynolds number. The CFD data was compared against experimental data obtained by another member of Bayles lab and literature data.
The trend of the CFD data seemed to match the pattern of the experimental data relatively well. The simulation data was also compared against literature values and found to be different by a factor of 2.We hypothesize that the experimental data is in transitional flow rather than laminar due to deviation from the linear CFD and literature models. Future work will be conducted to model other static mixers (ex. Kenics) as well as non-Newtonian fluids (fluids that change viscosity).