Selection of media grain size

Why filter media size is important

It is very important to select media with correct grain sizes. The reasons for specifying boundaries for grain size have to do with:

• Mechanical filtration mechanisms
  A slight increase in treatment efficiency has been observed with decreasing sand size which, indicates the importance of straining and adsorption. Higher removals tend to be due to smaller interstices between smaller sand, as well as the larger surface area available of the smaller sand size, which allows more adsorption to take place (Nam, Timmons, Montemagno and Tsukuda, 2000 [ref.01]Ref.01: Nam, T. K.; Timmons, M. B.; Montemagno, C. D.; Tsukuda, S. M. (2000) Biofilm characteristics as affected by sand size and location in fluidized bed vessels. Aquacultural Engineering 22, pp. 346-9). In the same way, having a sand that is too fine will lead to rapid clogging.
• Cleaning
  In coarser sand with a proportionally higher flow rate solids in the raw water are able to penetrate deeper into the sand making cleaning more difficult.
• Biological filtration mechanisms
  A smaller size of sand will have a larger total surface area available for biofilms to grow on, and therefore more biofilm can come into contact with the raw water. This therefore improves treatment effectiveness (Buzunis, 1995 [ref.02]Ref.02: Buzunis, B.J. (1995) Intermittently Operated Slow Sand Filtration: A New Water Treatment Process. MSc Thesis, University of Calgary, Canada.). Indeed, a greater combined surface area speeds up chemical reactions (surface catalysis) (Huisman and Wood, 1974 [ref.03]Ref.03: Huisman, L; Wood, W.E. (1974). Slow Sand Filtration. WHO, Geneva, Switzerland. pp. 31-4. Available from WHO). A study by Logan et al (2001 [ref.04]Ref.04: Logan, A.J.; Stevik, T.K.; Siegrist, R.L.; Rønn, R.N. (2001). Transport and rate of Cryptosporidium parvum oocysts in intermittent sand filters. Wat. Res. Vol. 35, No. 18, pp.4359–4369.) into Cryptosporidium oocyst removal suggested that the larger surface area associated with fine sand as opposed to coarse sand, together with the accompanying increased residence time were more important factors than pore size per se. The higher flow rates observed in coarser sand lead to poorer bacteriological filtration. This poorer filtration occurs because there is less contact time for biological predation on potential pathogens by the biological layer before the water passes through. In addition, flow rates may cause thinner and sparser biofilms attached to sand grains. Lessons can be learned from a study on biofilms from reactor vessels using 0.60 mm sand size (Uniformity Coefficient of 1.8) that were found to be thinner and smoother. This which was attributed to higher shear environments, in contrast to the thick, rough, porous films measured on 0.23 mm (UC of 1.4) sand samples. In these samples, the biofilm volume per sand unit area was found to be three times that of the 0.60 mm sand size. The same study found that the finer sand beds with 0.23 mm sand also had close to three times the biofilm surface area per unit sand area compared to the 0.60 mm sand beds (Nam, Timmons, Montemagno and Tsukuda, 2000 [ref.01]Ref.01: Nam, T. K.; Timmons, M. B.; Montemagno, C. D.; Tsukuda, S. M. (2000) Biofilm characteristics as affected by sand size and location in fluidized bed vessels. Aquacultural Engineering 22, pp. 346-9). While the flow rates in the study were much higher than slow sand filtration rates, and therefore flow characteristics will be different, the principles are possibly transferable: flow rate can affect biofilm development, which in turn can affect filtrate quality.

Effective size and uniformity coefficient explained

In order to select the correct grain size it is necessary to measure the Effective Size (or D10) and Uniformity Coefficient (or K). Both are used in defining filter media, in this case to know whether a type of media is or is not suitable for slow sand filtration.

The effective size of a given sample of sand is the particle size (in millimetres) where 10% of the particles in that sample (by weight) are smaller, while 90% are larger. Usually this is denoted as the D10.

The size distribution is represented by the Uniformity Coefficient, which enables you to see how well graded your sand sample is (that is, whether there is a whole different range of sizes, or whether most of the sample is only one size). This is done by taking the D60 and dividing by the D10. For slow sand filtration, some degree of uniformity is desirable in order to ensure that the pore sizes between the grains are reasonably regular and that there is sufficient porosity (Huisman and Wood, 1974 [ref.03]Ref.03: Huisman, L; Wood, W.E. (1974). Slow Sand Filtration. WHO, Geneva, Switzerland. pp. 31-4. Available from WHO).

Both effective size and uniformity coefficient are both used to define any particular sample. However, theory is not much use without practical ways to find this out in the field. Here we explain how to find out the effective size and uniformity coefficient yourself using 2 methods:

1. Sieve analysis. To be able carry out a sieve analysis, the following materials are needed:

• 3-cycle logarithm paper.
• Set of sieves for sand analysis. A plastic set is available from www.geosupplies.co.uk;
• Electronic scales with the ability to weigh 200 grams accurately to within 0.1 gram. Click here to download a worked example of how to carry out a sieve analysis.

2. Sand grain analysis using settlement jar.

This is a simple method using a transparent jar. A sand sample is mixed with water and then allowed to settle. If you can estimate the grain sizes using grain size tables such as those in Brassington (1988 [ref.05]Ref.05: Brassington, R. (1988) Field Hydrogeology. John Wiley and Sons Ltd.), and calculate the proportion of each sediment to the total depth of sand in the jar, a grain size distribution can be plotted similar to that of a standard sieve analysis. From this graph, the effective size and uniformity coefficient can then be calculated. It must be noted that this assessment is therefore done on volume of sand size, rather than weight.

In case it is not possible to carry out a grain size analysis, a field method to determine whether the sand size is more or less correct is to measure flow rates after the filter is installed. If flow rates are within acceptable limits, it is likely that the sand size is within the correct range.

References:

Ref 01: Nam, T. K.; Timmons, M. B.; Montemagno, C. D.; Tsukuda, S. M. (2000) Biofilm characteristics as affected by sand size and location in fluidized bed vessels. Aquacultural Engineering 22, pp. 346-9

Ref 02: Buzunis, B.J. (1995) Intermittently Operated Slow Sand Filtration: A New Water Treatment Process. MSc Thesis, University of Calgary, Canada.

Ref 03: Huisman, L; Wood, W.E. (1974). Slow Sand Filtration. WHO, Geneva, Switzerland. pp. 31-4. Available from WHO

Ref 04: Logan, A.J.; Stevik, T.K.; Siegrist, R.L.; Rønn, R.N. (2001). Transport and rate of Cryptosporidium parvum oocysts in intermittent sand filters. Wat. Res. Vol. 35, No. 18, pp.4359–4369.

Ref 05: Brassington, R. (1988) Field Hydrogeology. John Wiley and Sons Ltd.