Slow sand filtration - sand supply and control of the filtration rate

[Hayley_99]

PhD researcher here (background in biological sciences, but taking PhD in civil engineering).

TLDR: Does anyone know where to source suitable sand for slow sand filtration (e.g., imports from Portugal)? Also, how is filtration rate mostly controlled in slow sand filters—via the outlet valve or resistance to flow from small porosity?

Our project is on slow sand filtration and the removal of pharmaceuticals but we are running into problems manufacturing a large-scale testing column.

Initial research confirmed that an effective size (D10) smaller than 0.25 mm is ideal (slowing filtration rate = increased filter efficiency), but smaller again is ideal. It is proving difficult to find such a sand as most stone suppliers local to us (Ireland) cannot provide a sand with a small enough particle size. The sand we are recieving from them typically ends up around 0.30 mm (despite being labelled as smaller than this, and for sand filtration). Chatting to the director of sales from one of these quarries, he said that their company was the supplier to most water treatment companies in the north and south of Ireland so we thought we were on the money. However, lab testing shows the permeability and flow rate through these sands is much too fast (nowhere near the 0.2 - 0.4 m/h recommended). My suspicion is that these sands are being purchased for pressure and rapid gravity filters, not *slow* sand filters. My first question is this - if you are a slow sand filter designer/operator/etc., where do you source your sand from? We had rumours of sand being imported from Portugal.

My second question is this, most studies on slow sand filtration do not report physically how they controlled the filtration rate, so how is this done? Is this by a control valve at the outlet? I feel like this is too simple and totally negates the principle of treatment by the water percolating through the sand under gravity. Yes, the smaller the pore size means pores between grains are smaller and therefore there is more resistance to the flow of water thus a slower filtration rate. But one test of sand with a D10 of 0.15 mm was still flowing much too quickly (0.7 m/h).

 

[bimr]

Filter sand generally is sourced from river bed alluvial deposits.

Contact:

Northern Filter Sand - Northern Filter Media

Northern Filter Media, Inc. is the leading US manufacturer of filter sand for the potable water, waste-water and pool industries since 1914. Northern’s Filter Sand was formed from a natural glacial deposit formed millions of years ago along the Mississippi River Valley. Northern Filter Sand...

www.northernfiltermedia.com

Intermittent sand filters are operated by applying effluent on a periodic or intermittent basis until a predetermined head loss limited to the available freeboard is reached. At that point, the bed is drained and cleaned. Removal and replacement of the top layer of clogged sand has been shown to provide the most effective means of restoring the filtering ability of the sand.

 

[Compositepro]

I have never heard of "slow sand filtration". Sand filters are generally classified as depth filters, which work by capturing particles by impingement. Therefore, higher flow velocity works better. It sounds as if you are relying on sedimentation rather than filtering. Perhaps you should look into clarifiers that use crossflow through a sedimentation/settling tank.
Also, particle size measurement and definition are more complex subjects than most scientists imagine. Only a single sphere can be characterized by a single number. Distributions can be an average or a mean based on number, surface area, or mass of particles. Without defining every one of these, the numbers do not mean a thing.

 

[pointsman]

The primary filtering element of a slow sand filter is the biological layer that rests above the sand, which must be ripened (grown) before the filter becomes effective. If you are missing the biological layer then your filtration rate will be much higher than literature values obtained from mature filters.

An experimental challenge will be establishing a reproducible biological layer. Over time, the layer builds up thickness and must be periodically cleaned. You should also ensure that composition of the microbial community that forms on your experimental filter is representative of the communities of real filters.

It might be interesting to perform functional analysis of the community to determine which metabolic pathways contribute to breakdown of pharmaceuticals.