The numerical modelling of the loading behaviour of fibrous filters.

Filters are used widely to remove particles from the air. This could be in order to protect people working in industries where harmful particles are produced. In this case people may use filters in the form of a face mask to protect themselves or there may be a large industrial filter that removes the particles in the area in which the people work. Other uses for filters include keeping the air clean in schools, airplanes, hospitals etc. They therefore have a very important role and hence we must understand how they work. There are different types of filters but one of the most important is filters made up of fibres, known as fibrous filters. These filters are made up of a large number of layers; each layer is made up of very thin fibres with a lot of space between them. Air containing particles passes through the filter and some of the particles in the air will hit the fibres and stay on them, hence removing the particles from the air. As time goes on the number of particles collected by the fibres increases and this affects how the filter works. Eventually the filter may get to the point where it is clogged and no air can pass through it. In this case it will either need to be replaced or cleaned which is an expensive business.
The amount of harmful particles removed by the filter depends on many things, in particular the number and size of fibres in the filter, the size of the particles and the speed at which the air moves through the filter. If we understood how the performance of the filter depends on these factors then we could design and use them in such a way that they achieved the best performance for the situation. Gaining this understanding has been the subject of much work by people in the past. Generally this work has involved considering the filters at the beginning of their lives when they are clean and no particles have been collected. However as particles collect they affect the filter performance and so we need to understand just how the build-up of particle deposit in the filter affects its performance.
To date the PI has been involved in developing mathematical models that show how small particles deposit on the fibres and how this affects the air flow through the filter and also further particle deposits. Now we would like to increase that knowledge so that particles of all sizes can be considered. The aim of the current proposal is that the PI should collaborate with Professor Zaripov's group at Kazan Federal University who are also developing a model for aerosol filtration. The Russian study has adopted techniques that enable the study of the build up of large particles. By combining the expertise of the two studies the most efficient model for the particular filtration situation will be identified. A range of particle, and filter, properties relevant to aerosol filtration will be considered.

The proposed collaboration will result in a robust numerical model that can predict the performance of a fibrous filter as particulate is deposited. The results of which will benefit filter manufacturers and users. Manufacturers will have the potential to theoretically examine fibrous filter media performance, efficiency, loading characteristics, etc. and relate these parameters to filter structure. Users will be able to determine optimum operating conditions.