Abstract
Biological filtration systems offer a sustainable alternative to existing engineered solutions. In this computational work, we seek to optimize the surface coverage by an array of hairs to capture particles in channels. A variety of aquatic organisms rely on arrays of hairs to interact with their fluidic environments. The hair functionality can vary from sensing to smelling, filtration to flow control depending on the species considered. Among those organisms are filter-feeders that rely on suspension-feeding, one of the most widespread feeding mechanisms and one of the oldest. Baleen whales are filter-feeders that catch their prey by using the baleen, a complex structure composed of plates and bristles in their mouth. The hairs are hollow cylindrical structures with a diameter of a few hundred micrometers that can extend over tens of centimeters. The baleen filters out the prey while letting the seawater through. The baleen is composed of flexible and elongated structures whose properties fit the feeding habits of the whale.
The porosity of the structure depends on the flow feature. Effectively, the flow can tune the filter properties, which sets biological filters apart from their engineered counterpart. Previous mechanical studies have shown that an array of hairs can either act as a sieve, allowing all the fluid to flow through it, or as a rake, forcing the fluid to flow around it instead. As the speed increases, the behavior shifts from rake to sieving for a given hair spacing. From a filtration perspective, the rake regime is not favorable as particles do not enter the array. For a fixed fluid velocity, the flow transitions from rake to sieve as the spacing between the hairs in the array increases. Our recent work has also demonstrated that the confinement of the channel influences the sieve to rake transition. The filtration mechanisms that filter-feeder organisms use to capture food particles exhibit complex fluid-structure interactions that have yet to be leveraged in engineered systems. To guide the development of hair-covered surfaces capable of trapping particles in channel flows, we investigate how different geometric factors affect the fluid transport and capture of particles by the array. In previous work, a small number of hairs, typically 25, were considered. Here, we vary the array geometry, the Reynolds number of the flow, and the surface coverage to study the transport through this confined porous structure. We compare arrays based on their optimal efficiency and the (sub-optimal) operating conditions which make the filter versatile.
