Session: 11-18-01: Fundamental Issues in Fluid Mechanics/Rheology of Nonlinear Materials and Complex Fluids/Plasma Flow

Paper Number: 173643

Benchmark Simulations and Matched Experiments of Non-Newtonian Flows Through Complex Geometries 

Non-Newtonian fluids are common across a wide range of biomedical and industrial applications. Historically, many studies have simplified analysis by assuming Newtonian behavior and compensating for deviations with conservative safety factors. While such approximations can offer rough estimates of parameters like drag forces, they fail to capture the complex dynamics inherent to non-Newtonian behavior—particularly as systems scale down or when high fidelity in flow characterization is required. Accurate modeling of non-Newtonian flows is further complicated by significant numerical challenges, especially in realistic scenarios involving interactions with impermeable walls and complex geometries. These conditions, which are ubiquitous in practical systems, demand a more robust and precise simulation framework. Developing such a methodology is essential for advancing our understanding of non-Newtonian fluid dynamics, particularly in wall-bounded, vorticity-dominated flows. A reliable simulation approach will enable more accurate, physics-based design across a broad spectrum of engineering applications.

This study presents a comprehensive set of CFD simulations of the well-established 4:1 planar contraction geometry, a canonical benchmark for studying complex flow behavior. The simulations encompass a range of rheological conditions and are broadly divided into two fluid models: power-law and viscoelastic (Oldroyd-B model). These models represent commonly encountered non-Newtonian behaviors in both industrial and biomedical applications, making them particularly relevant for evaluating flow instabilities and deformation under contraction. The evolution of flow through the contraction is analyzed by systematically varying the power-law index (n) for shear-thinning and shear-thickening behaviors, and the Weissenberg number (Wi) for elastic effects in viscoelastic fluids. To isolate rheological effects and avoid numerical instabilities or solver-induced anomalies, both the Reynolds number and Weissenberg number are constrained to a maximum of 0.05. Quantitative analysis focuses on key flow characteristics, including the dimensionless sizes of lip and corner vortices—critical indicators of flow separation and recirculation—as well as the non-dimensional, steady-state velocity profile along the mid-plane of the contracted section. These parameters enable consistent comparison across varying rheological regimes. All simulations are performed using the open-source CFD platform OpenFOAM, employing the RheoTool library specifically designed for complex rheological models. The choice of solver is informed by its proven reliability and prior successful applications in modeling viscoelastic and multiphase flows (e.g., Habla et al., 2012; Omowunmi et al., 2013; Favero et al., 2010; Habla et al., 2011). This setup provides a robust framework for investigating non-Newtonian flow dynamics in confined, wall-bounded geometries.

The numerical simulations will be validated through Particle Image Velocimetry (PIV) experiments conducted on a custom-fabricated glass model, designed to match the geometric parameters of the computational setup. Experimental measurements will be performed for four representative rheological cases—two corresponding to limiting conditions for power-law fluids and two for Oldroyd-B fluids. The same quantitative metrics used in the simulations, including vortex size and velocity profiles, will be used to calibrate the numerical model and assess its accuracy. This validation step will ensure the reliability of the solver before extending the study to the full range of rheological parameters.

Presenting Author: Aarthi Sekaran SUNY Polytechnic Institute

Presenting Author Biography: Aarthi Sekaran graduated with a Ph.D. in Mechanical Engineering and a concentration in thermal and fluid sciences from Texas A&M University in 2012. She has since held positions as a research associate and a research fellow at JNCASR (India) and the University of Texas at Austin. Her areas of interest include physics and modeling of fluid flows with an emphasis on flow instabilities and flows through complex geometry. She primarily uses CFD in conjunction with simple experiments to study fundamental flow phenomena that occur in the context of natural, biological and industrial flows. She also has a keen interest in engineering education, particularly flow visualization applications and using video games for fluid dynamics education.

Authors:

Aarthi Sekaran SUNY Polytechnic Institute
Ahmed Abdelaal SUNY Polytechnic Institute
Shitiz Seghal National Oilwell Varco