Comparative CFD Study of Bubble Trap Inlet Geometry for Ex-Vivo Heart Perfusion Systems

Abstract

This study focuses on the computational fluid dynamics (CFD) analysis and optimization of inlet geometry of a bubble trap designed for ex-vivo heart perfusion systems. The main objective was to examine how variations in inlet height and angulation affect bubble trapping efficiency and flow characteristics within the chamber. For that reason, different device configurations were designed in SolidWorks and analyzed using CFD simulations in ANSYS Fluent, employing an Eulerian–Lagrangian framework, where the continuous perfusate phase was modeled with the Navier–Stokes equations and the dispersed air bubbles were tracked using the Discrete Phase Model (DPM) under two-way coupling to account for phase interaction. A total injection of 1,500 spherical air bubbles consisting of six size groups of 250 bubbles within a size range of 50–500μm, was introduced at the inlet of the bubble trap to represent entrained air entering the system and six geometrical configurations were tested by varying the inlet height and angulation, while maintaining constant chamber volume and flow conditions. Contours of pressure, temperature, and turbulent kinetic energy, together with inlet and outlet data on pressure, velocity, and temperature, as well as velocity streamlines, were examined to characterize the hydrodynamic behavior within the chamber. The results demonstrated that inlet configuration had an influence on flow circulation and bubble trapping efficiency, as it was observed that the higher inlet position enhanced recirculation and delayed bubble escape, while inlet angulation redirected the flow toward the chamber walls, further improving bubble entrapment and reducing the likelihood of direct bubble transport to the outlet. Among all tested geometries, the 45° angled high inlet configuration achieved the highest bubble trapping efficiency, while the low inlet configuration without angulation exhibited the lowest efficiency. Furthermore, in all simulations it was showed that bubble size significantly affected escape behavior, as smaller bubbles (50–100 μm) were more likely to escape across all geometries, while larger bubbles (≥400 μm) were effectively trapped. Additionally, pressure and temperature analyses showed small variations across all cases, with a pressure drop ranging between 1.25 and 1.28mmHg and temperature loss of approximately 0.007–0.009 °C, across the different geometries.