Numerical simulation of buoyancy-driven flow in a human stomach geometry: Comparison of SPH and FVM models

The primary functions of the stomach, including mixing, digestion, and emptying, are affected by the spatial distribution of content properties, and how this distribution changes in response to gravity and wall contraction forces. One aspect that is not well studied is the buoyancy effects that result from content with heterogenous density. A buoyancy-driven flow in a non-deforming stomach is simulated using the Smoothed Particle Hydrodynamics (SPH) and Finite Volume Method (FVM). An aqueous liquid layer is initially placed above a fatty layer of the same volume. The buoyancy effect driven by gravity and strongly influenced by the stomach shape, causes the fat to rise to the top layer and the aqueous liquid to sink to the bottom. Rayleigh-Taylor Insta-bility (RTI) is well-captured in the SPH model in the initial interface flow development. Some detailed flow behaviours are slightly different between the two models once the bulk turnover flow is established, but both models show that the separation process happens rapidly (within 6 s) in the curved geometry of the stomach. The final phase separation is more complete in the FVM compared with the SPH. Sensitivity analysis is conducted in both models to examine the solution accuracy. This work suggests that buoyancy driven flow effects, which have not been investigated previously, occur on a much shorter timescale (6 s) than peristaltic (20 - 60 s) and digestion timescales (hours). This result may help basic understanding of digestive processes and be used to guide food and drug design.

Automated summary

This study investigates the buoyancy effects of a mixture of fat and aqueous liquid in the stomach using Smoothed Particle Hydrodynamics (SPH) and Finite Volume Method (FVM). The results show that fat quickly rises to the top layer and liquid sinks to the bottom layer in the curved geometry of the stomach. These buoyancy-driven flow effects occur on a much shorter timescale (6 s) than peristaltic (20 - 60 s) and digestion timescales (hours), and may have implications for understanding digestive processes and guiding food and drug design.