No abstract

No abstract

No abstract

Cerebrospinal fluid (CSF) is a Newtonian fluid and can, therefore, be modelled using computational fluid dynamics (CFD). Previous modelling of the CSF has been limited to simplified geometric models. This work describes a geometrically accurate three dimensional (3D) computational model of the human ventricular system (HVS) constructed from magnetic resonance images (MRI) of the human brain. It is an accurate and full representation of the HVS and includes appropriately positioned CSF production and drainage locations. It was used to investigate the pulsatile motion of CSF within the human brain. During this investigation CSF flow rate was set at a constant 500 ml/day, to mimic real life secretion of CSF into the system, and a pulsing velocity profile was added to the inlets to incorporate the effect of cardiac pulsations on the choroid plexus and their subsequent influence on CSF motion in the HVS. Boundary conditions for the CSF exits from the ventricles (foramina of Magendie and Lushka) were found using a "nesting" approach, in which a simplified model of the entire central nervous system (CNS) was used to examine the effects of the CSF surrounding the ventricular system (VS). This model provided time varying pressure data for the exits from the VS nested within it. The fastest flow was found in the cerebral aqueduct, where a maximum velocity of 11.38 mm/s was observed over five cycles. The maximum Reynolds number recorded during the simulation was 15 with an average Reynolds number of the order of 0.39, indicating that CSF motion is creeping flow in most of the computational domain and consequently will follow the geometry of the model. CSF pressure also varies with geometry with a maximum pressure drop of 1.14 Pa occurring through the cerebral aqueduct. CSF flow velocity is substantially slower in the areas that are furthest away from the inlets; in some areas flow is nearly stagnant.

In this paper we study the effects of dynamic wetting on the immiscible displacement of a high viscosity fluid subject to the radial injection of a less viscous fluid in a Hele-Shaw cell. The displaced fluid can leave behind a trailing film that coats the cell walls, dynamically affecting the pressure drop at the fluid interface. By considering the nonlinear pressure drop in a boundary element formulation, we construct a Picard scheme to iteratively predict the interfacial velocity and subsequent displacement in finite-mobility-ratio flow regimes. Dynamic wetting delays the onset of finger bifurcation in the late stages of interfacial growth and at high local capillary numbers can alter the fundamental mode of bifurcation, producing vastly different finger morphologies. In low mobility ratio regimes, we see that finger interaction is reduced and characteristic finger breaking mechanisms are delayed but never fully inhibited. In high mobility ratio regimes, finger shielding is reduced when dynamic wetting is present. Finger bifurcation is delayed, which allows the primary fingers to advance further into the domain before secondary fingers are generated, reducing the level of competition.