Simulating the Connection Between Fluid Droplets and a Solid Surface
The behaviour of fluids in different circumstances are in immensely difficult thing to simulate. Fluid dynamics is an entire field of study, typically requiring petaflop computers, and anything but real-time simulation.
Real-time simulation of fluids is of course possible, but is largely limited to still bodies of water, or wave formations. As such any research that increases our knowledge of how to simulate fluids under different conditions is immensely valuable to us.
Dr. Adham Hashibon at the Fraunhofer Institute for Mechanics of Materials in Freiburg, Germany , has been working on one such problem. This work is intended for use in designing window panes, corrosion coatings and microfluidic systems in physical medical labs, but is useful for simulation work for the exact same reason.
The research was directed at understanding what characterizes surfaces that do the best job of cleaning themselves and why. To do this, the behaviour of fluid droplets had to be simulated in an entirely new way. Rather than just considering the water itself, the properties of the surfaces the water interacts with had to be taken into consideration.
Our simulation shows how various liquids behave on different surfaces, no matter if these are flat, curved or structured, explains Dr. Hashibon How liquid behaves on a surface is influenced by a great deal of parameters, including the surface characteristics of the material as well as its structure, but also by substances dissolved in the liquid. We have taken all this into account to different degrees of detail within the simulation so that we are able to clearly reproduce our experimental findings.
The software analyses what goes on within a given droplet how the individual water molecules interact with each other, how a droplet is attracted by the surface and how it resists the air. This is still way beyond the capability of real-time simulation of course, and if anything pushes real-time complex interactions with water and similar fluid systems further into the future. But, it produces vastly more accurate simulations than even a particle swarm is capable of.
The simulation system developed using these principles, simulates the exact form every liquid droplet takes when it hits a given surface, and determines whether it will distribute itself over the surface, or contract to form droplets in order to minimize contact with the surface based on this interaction. When droplets are added in larger numbers and begin to merge, the same simulation process is able to calculate the flow behaviour in terms of how liquids move across different surfaces, whereby the determinant factors at different scales of measurement are integrated, from atomic interactions to the impact of microscopic surface structure.
The uses in non-real-time simulations are legion, and the benefit is immense. Currently particle swarms are used to simulate biological soft tissues. Poke one area, and all the surrounding particles deform in response, according to their specific properties. It is a very good way of simulating how soft tissue will react, but it is not perfect and cannot accurately simulate the flow of the body's fluids through cuts in the simulated flesh. This method can do that.
Which means it is useful both for showing how the fluid will react when a vein is breached, or the lymphic system is ruptured. Better, it opens up the possibility of being able to diagnose from the pattern of blood in a body undergoing surgery, where the leak actually is. Potentially saving surgical time and patient lives.
These uses are in the future, as larger scale simulations and swifter calculation speeds are necessary for such applications. However, the basic principle is sound. It also means, we can simulate exactly how water will react to your clothes and skin depending on the nature of the surface they encounter, even if we cannot do such with any great speed just yet.