Previous and current ODA supported work involving the applicant (Prof. Nigel Graham) has demonstrated the value of fabric-protected slow sand filters. The potential use of natural, indigenous fabric materials will enhance the feasibility and sustainability of water treatment technology. The project will evaluate the feasibility of natural fabrics for this purpose and maake recommendations at the end of the study.

At the start of the research period (July 1995) a review of the range and types of natural tropical fibres took place, and studies were carried out into their respective properties. Using textile theory, and the results of previous research studies at Imperial College, it was possible to determine whether fibres of a certain type could be manufactured into fabrics with suitable filtration properties.

Once a limited number of potential fibre types had been identified, it was necessary to obtain samples of raw fibres, and fabrics manufactured to a specification theoretically ideal for filter protection to carry out preliminary laboratory experimentation. It was envisaged that the major problem in the use of natural fabrics as protection for slow sand filtration would be the micro-biological degradation of the fabrics. Hence, degradation experiments were carried out to quantify the problem.

In the laboratory, using the samples of manufactured fabrics from jute, abaca, sisal, coir, and flax, filtration and permeability experiments were carried out to predict the likely performance of these fabrics at full scale.

The findings from the laboratory work were used in the selection of fabrics for application to pilot plant trials, carried out at the Thames Water Shalford water treatment works, Guildford, Surrey. Fibre types at this stage were limited to jute and abaca, together with the incorporation of coir. Long-term pilot plant trials would enable the performance and behaviour of filters protected with these selected fabrics to be monitored in an environment similar to that of the intended final application and also compared to those of filters with no fabric protection. Parameters of interest at this stage were rate of head loss development, and the filtered water quality.

It was possible to carry out two complete sets of experimentation at pilot plant scale. This led to the conclusion that the objectives of the research were able to be met, i.e., an extension of slow sand filter run time and the protection of the sand bed was possible, by placing multiple layers of intermediate-to-high density jute fabrics and a single layer of low density coir fabric on top of a slow sand filter.

The project has demonstrated that slow sand filter run times can be extended with the careful selection of natural filtration fabrics. We have also confirmed that through the use of natural fabric protection, the penetration of influent solid material into the sand can be eliminated. Hence, the objectives of the research have been successfully met.

From preliminary calculations the implications of these findings are a possible 45% saving in annual operational costs, as a result of the reduced frequency of cleaning requirements, and a potential for the reduction in the thickness of the sand bed.

However, the full implications of using natural fabric protected slow sand filtration are not yet known. It is widely known that micro-biological degradation of organic materials is much more rapid in hotter climates. It may be that the degradation of natural fabrics in a hot climate will be rapid enough to prevent a filter run time extension to be achieved. The work described here was carried out in a cold/temperate climate, and it did not appear that there were any counter-beneficial factors restricting the use of natural fabrics in a water treatment scenario, i.e. the fabrics did degrade slightly but not to a state where filtration performance was affected. This means that we can only at this stage recommend the use of jute fabrics in such climatic conditions as we have worked in. Ultimately the benefits of natural fabric protected slow sand filtration are intended to be made available in developing countries. This cannot be yet recommended until a thorough investigation in a typical developing country climate is carried out.

1. Luxton,P.J. and Graham, N.J.D. (1998). 'Preliminary Evaluation of Natural Fabrics in Slow Sand Filtration', J Water SRT - Aqua, 47, (3), 95-106.

2. Luxton, P. and Graham, N.J.D. (1996). "Natural Fabrics in Slow Sand Filtration", 22nd WEDC Conference, 9-13 September, New Delhi, India.

Tel: 0171 594 6121
Fax: 0171 594 6124
Email: n.graham@ic.ac.uk