Diatomaceous earth filters have been used for many decades in a variety of industries such as the drinks and pharmaceutical sectors. The process depends upon a filter cake of diatoms, Kieselgur, expanded perlite or a cellulose  based products being laid evenly on series of internal candles or plates. The filter cake is normally only a few milli meters thick. so it is important that the cake is applied properly otherwise there will be a short circuiting of the media. 

When the differential pressure reaches a certain point, the filters need to be back-washed.  Only a small amount of water is required for the application  to discharge all of the filter cake along with the solids removed by the cake. This procedure will be required once or twice every week in a public pool, in reality it is often required to back-wash at more frequent intervals.  In order  to try and prolong the time interval between back-washes the filter media can be injected into the filter to build up a cake  and reduce blinding.   It is importabt to note that many water companies in the UK will not accept a discharge to sewer of diatomaceous earth or perilte.

Diatomaceous earth is no longer recommended because the small particle size and high free silica component is extremely dangerous. Perlite is a much safer alternative to diatomaceous earth,  however the very fine particles still represent a potentially serious health hazard, indeed any small particle cloud ( especially silica) in the atmosphere represents a respiratory health issue or even an explosion hazard.

Performance issue

The system can give a very good performance in terms of solids removal, certainly filtration down to a less than 5 microns, so the performance at removing the oocysts of Cryptosporidium can be very good. Indeed  the diatomaceous earth filter was first developed by the US Army as a light weight filter for the mechanical filtration of drinking water.  The system was first adopted  for the treatment of municipal drinking water in 1947,  however since then there have only been around 200 systems installed worldwide for drinking water, compared to 100 of thousands of sand filters.  There is obviously a reason for this disparity in numbers of systems installed which is explained as follows;

While the filters have a low capital cost and high flow rate per unit area,  the process is not generally sustainable for drinking water or swimming pool water treatment .  There are health and safety issues with the dust and cost implications in chemicals and changing of the filter cake.  If the filter cake is discharge to sewer it will be classified as an industrial discharge and there will be high costs associated with the discharged solids loading, either to sewer or to specialist disposal sites.

Chlorine or oxidising chemicals are required in order to give a  residual disinfection,  however the chlorine or oxidation demand of the water must first be satisfied in order to leave a residual that will prevent horizontal transmission of disease among the bathers.  The organic and nitrogen waste loading in swimming pools is approximately  20% solids and 80% in solution.  The components in solution include  proteins , fats amino acids etc etc,  exert the principal chlorine demand in swimming pools.  The dissolved components and bacterial organic nutrients as well as  phosphates can be removed by injection of coagulation and flocculation chemicals such as NoPhos or APF.  The coagulants / flocculants bring the dissolved components out of solution and form a small particle that can then be removed by a media type filter.  It is essential that coagulation and flocculation chemicals are used before media bed filtration with sand or AFM .  In addition to removing the dissolved components,   flocculation makes the micron and sub -micron particles much larger which in turn makes it much easier for sand or AFM media to remove the particles from the water.  Indeed it is quite easy to achieve a nominal filtration rate of of less than 0.1 microns using flocculation followed by AFM filtration at the correct flow rate.

Filters based on diatomaceous earth or similar technologies can not be used in combination with coagulation and flocculation because the filter cake blocks very quickly making it impractical to use the technology.   This means chlorine consumption is a great deal higher and more toxic chlorine reaction products are produced as a consequence of using a DE or perilte system.  The  chlorine will not manage  to instantly oxidize the organics, so  the pool will also tend to run with high dissolved organic concentration as well phosphates which promotes  the growth of bacterial  biofilm on the inside off  the pipes and on the tiles and drains.  There is therefore a much greater risk from MRSA, Legionella , Mycobacterial aerosols and respiratory disease as  well as the toxicological issues associated  with trichloramines and trihalomethanes in the atmosphere.

The capital cost of  diatomaceous based systems  and plant room requirements  are generally lower, however the running costs are  great deal higher in terms of chemicals, and associated products.  It is estimated  that there  would be a 3 to 5  year return in capital cost made out of revenue savings if a properly engineered AFM system is used as opposed to a diatomaceous type system.  However the main issue is the much better water and atmospheric air quality from using AFM and  the improved safety and well being of not only the public but the staff working in the facility. Indeed  new legislation from the European Commission and tighter controls on water and atmospheric quality may make it difficult to operate diatomaceous type technology in the near future because they will simply not be able  to comply with water quality and atmospheric quality criteria.