Murray's law revisited: Qu\'emada's fluid model and fractal trees

TL;DR

This paper revisits Murray's law for blood vessel design, extending it to shear-dependent Quémada fluids, and explores how viscosity effects influence optimal vessel and fractal tree geometries.

Contribution

It generalizes Murray's law to include shear-dependent viscosity models like Quémada's, and derives extended laws accounting for phase separation effects in microcirculation.

Findings

Murray's law is universal for Quémada's fluids without phase separation.

Phase separation effects lead to extended versions of Murray's law.

The study impacts the understanding of optimal arterial network geometries.

Abstract

In 1926, Murray proposed the first law for the optimal design of blood vessels. He minimized the power dissipation arising from the trade-off between fluid circulation and blood maintenance. The law, based on a constant fluid viscosity, states that in the optimal configuration the fluid flow rate inside the vessel is proportional to the cube of the vessel radius, implying that wall shear stress is not dependent on the vessel radius. Murray's law has been found to be true in blood macrocirculation, but not in microcirculation. In 2005, Alarc\'on et al took into account the non monotonous dependence of viscosity on vessel radius - F{\aa}hr{\ae}us-Lindqvist effect - due to phase separation effect of blood. They were able to predict correctly the behavior of wall shear stresses in microcirculation. One last crucial step remains however: to account for the dependence of blood viscosity on shear rates. In this work, we investigate how viscosity dependence on shear rate affects Murray's law. We extended Murray's optimal design to the whole range of Qu\'emada's fluids, that models pseudo-plastic fluids such as blood. Our study shows that Murray's original law is not restricted to Newtonian fluids, it is actually universal for all Qu\'emada's fluid as long as there is no phase separation effect. When phase separation effect occurs, then we derive an extended version of Murray's law. Our analyses are very general and apply to most of fluids with shear dependent rheology. Finally, we study how these extended laws affect the optimal geometries of fractal trees to mimic an idealized arterial network.