In conventional water treatment plants traditional slow sand filters are typically third stage treatment after settling and rapid sand filters. This is because in poor quality water water high in turbidity, algae and iron in particular headloss would develop necessitating frequent cleaning. As a result of this limitation on the quality of water slow sand filters can effectively treat and the perception of them as being an outdated technology, their use declined with the advent of the rapid sand filter in North America.

However, the technology is experiencing a resurgence as of late particularly due to their proficiency at removing Giardi, Crypto and dissolved organics. This is especially true in the United States due to the Surface Water rule and the more stringent regulations. The resurgence is a result as well as driver of the evolution of SSF systems that has transformed them into a simple, reliable and cost-effective treatment for small utilities.

COMPONENTS OF A TYPICAL SLOW SAND FILTER:
Typical conventional slow sand filters are open vessels made of concrete anywhere from 2.5 to 4 m deep.
Raw water enters from above and sits in a reservoir on top of the sand bed. The reservoir maintains a constant head above the media which provides the pressure to drive the water through the sand.

The sand bed ranges from 0.6 to 1.2 m deep. Various types of silica or quartz sand are used provided they are in the range of about 0.1 to 0.3mm effective size and is reasonable uniform.

The underdrain is typically a system of porous pipe through which the filtered water exits the filter. It is covered by graded course gravel which supports the sand and prevents fines from clogging the underdrain.

And of course the filter has to have a system of control valves to adjust the flow, allow for backfilling which is allowing water up through the underdrain system upon startup or backwashing hydraulic cleaning of the bed is used in some of the more modern SSF systems.

HOW DOES IT WORK?
Water is purified by a slow sand filter through two main processes physical-chemical and biological. Microorganisms are too small to be removed by physical processes alone. As raw water enters the filter it sits in the reservoir for sometimes several hours where settling of larger particles occurs. The water then percolates through the sand bed at a slow rate where a variety of other physical processes such as screening and attachment occur that work to purify the water.

The slow flow rate is the key to the effectiveness of the SSF. This slow flow of the water through the filter allows microorganisms to form biofilms and colonize the sand bed. When this occurs the filter is said to have ripened meaning that micro-organisms are present and performing biological processes that remove or transform contaminants in the influent water. Upon startup a slow sand filter is run to drain until it has ripened.

A layer of slimy microbial biomass will grow on top of the sand bed and within the top couple of inches. This slime layer is referred to as the schumtzdecke german for dirty skin. The schmutzdecke is where the bulk of the biological activity that purifies incoming water occurs. This purification is believed to occur in two ways predation and grazing of higher organisms on lower order (or smaller) organisms. Predators present in the biomass graze on bacteria that is attached to the sand grains. Alternatively predators that are in the water will graze on suspended particles and bacteria as they flow through the filter.

Although the schmutzdecke is a primary mechanism for purification, it accumulates during filtration resulting in headloss development. When a certain amount of headloss is experienced the filter must be scrapped which involves draining down the filter and removing the top couple inches of sand. After a number of scrapings have taken the bed down to its minimum depth; it must be resanded this can be with new sand or used sand that is washed and replaced.

DISTINGUISHING CHARACTERISTICS OF SLOW SAND FILTERS:

-a slow flow rate approximately 0.04 0.42m/h

-no chemical pre-treatment as chemicals would likely harm the biomass

-when a prescribed amount of headloss occurs the sand bed is scrapped and resanded

-uniformity of sand grain size

-the formation of the biomass layer on top of the sand bed

-no backwashing required

-relatively long filter runs as they are typically employed as a third stage of treatment

SIMPLE, LOW COST, EFFECTIVE TREATMENT, BUT...
Traditional slow sand filters offer simple, low cost and effective treatment but they also have some drawbacks.

Certain conditions can impair filter performance.

-intermittent operation stopping and starting of filters can result in stagnant water in the reservoir on top of sand and depletion of dissolved oxygen and nutrients necessary to sustain the biomass. The result can be a decline in performance.

-cooler temperatures can hamper biological activity particularly when the bed is ripening or in systems where the biomass is not abundant and firmly established

- micro-organisms require key nutrients, trace minerals and vitamins to thrive and when they are not present in the influent water the biomass will struggle and the performance, therefore will not be optimal

- some surface waters may pose a particular challenge to slow sand filters as they can be prone to sudden fluctuations in quality which can upset the biological activity and effect filter efficiency

Another drawback of conventional slow sand filters is the maintenance procedure. Filter scraping and resanding, even if it is a mechanized process, is generally quite laborious, requires many manhours and results in extended filter downtime by the time the filter is re-ripened and ready to go back online again.

Finally, unless the water being treated is of reasonably good quality, the filter may have to be cleaned quite frequently.

While these disadvantages to conventional slow sand filters are well noted, their recognized potential as an effective treatment system prompted considerable research and development leading to evolution.

EVOLUTION:
The evolution of slow sand filters has been aimed at:
-extending the filter runs
-improving treatment within a wider range of water quality and operating conditions
-and simplifying maintenance

Filter cleaning procedures have been developed that reduce the amount of labour and filter downtime required. Several mechanized systems such as conveyor belts, traveling bridges and modified tractors with skimming blades have been used to simplify the scrapping and resanding procedure. One method that appears to be gaining popularity is wet-harrowing which is simply raking the top of the sand bed and draining off the debris rather than scrapping.

A method of resanding called trenching has also been developed that minimizes the disturbance to the bed and reduces the re-ripening time. This involves moving the used sand to one side of the filter, adding new sand, and then replacing the old sand on top. This way the micro-organisms stay in the filter and can quickly resume efficient particle removal.

Hydraulic cleaning of slow sand filters is perhaps the most advantageous innovation with respect to time and labour savings. Backwashing involves the low velocity application of wash water up through the underdrains and light air scour to avoid excessive disturbance of the biologically active sand bed. The debris is then drained off the top of the filter and it can be returned to service in perhaps as little as an hour or two.

To help extend filter runs some facilities are using filter mats. These are sheets of non-woven synthetic material that is placed on the surface of the sand bed. When headloss does occur the material can be removed and cleaned and no scraping is required.

Filters are also being constructed out of materials other than concrete. Molded polyethylene tanks are increasingly being used for this application and offer the advantages of being more cost effective, they allow for modular design and therefore can be expanded if capacity increases and they simplify the engineering and design requirement.

SSF treatment trains have also evolved to include other unit processes such as ozonation, roughing filtration and biological activated carbon polishing.

THE MODERN SSF TREATMENT TRAIN:
A modern SSF treatment train may be designed something like this:

Ozone pre-treatment for initial disinfection and oxidation of metals and organics. Ozone is becoming a common treatment prior to slow sand filtration as, if applied correctly, there is no residual that may harm the micro-organisms. Ozone provides effective inactivation of oocysts such as Crypto and Giardia as well as viruses. It also breaks up larger organics into smaller particles that can then be biodegraded by the micro-organisms in the filter.

Roughing filters composed of course gravel media are also becoming a common component in SSF trains. They can reduce the solids loading on the sand filters, increase the filter run length, and allow the sand filters to operate at higher hydraulic loadings. The use of a roughing filter also increases performance of the slow sand system in cold water conditions by providing additional biological activity and increased retention time that can compensate for any loss of efficiency.

The slow sand filters provide reduction of turbidity, heavy metals and organics.

And finally an activated carbon filters can be added to some SSF systems to provide a polishing process that can remove further contaminants particularly organics, taste and odour compounds, pesticides, herbicides, trihalomethane precursors, and ozonation by-products. GAC filters in slow sand systems also develop biological activity and can be referred to as BAC (biologically activated carbon). The advantage of BAC over GAC is an extended bed life. As contaminants are adsorbed onto the carbon granules the micro-organisms consume them continually freeing up adsorption sites.

The above picture demonstrates an example of a modern SSF system. It is modular in design, therefore expansion can occur if required. The filters are cleaned hydraulically, by backwashing with treated water and light air scour. Therefore operation is simple, there is very little process adjustment and no chemical use other than chlorine for final disinfection residual. On a daily basis, the operator needs 15 minutes to check chlorine and turbidity levels and a filter cleaning takes about 1 to 2 hours. Frequency of filter cleaning is variable, however a roughing filter may be cleaned once every week to two weeks, a sand filter once a month and a BAC filter every three months.

THE ADVANTAGES:
With these advancements of SSF in mind the acknowledged advantages include:
-that it is a very effective treatment for a wide range of contaminants
-its chemical-free
-they are simple to operate and produce good quality drinking water with minimal process adjustment
-the addition of other components such as ozone, roughing and BAC filtration have made them effective under a wider range of conditions including cooler weather
-minimal time and expertise is required to operate these systems
-the operating costs are low
-there are no residuals meaning that there are no issues having to dispose of spent or contaminated media
-there is minimal water wasted relative to other forms of treatment eliminating concerns about water supply or expansion of the lagoon
-and finally they are modular which eases design requirement and expansion

THE DISADVANTAGES:
Even still, no treatment system is perfect and there can be some drawbacks associated with slow sand filtration as well.
-they occupy a relatively large footprint, therefore a larger building may be required
-the impact of cold water on performance can be minimized by the addition of the other components, however, ripening of the filters can take longer when influent water is cool
-SSFs that do not use alternative media cannot treat some contaminants including sulfates, nitrates, hardness and total dissolved solids
-and they can have difficulty removing silty colloidal matter

TYPICAL PERFORMANCE:
This chart shows some of the reported removal efficiencies of various contaminants:

DOLLARS & SENSE:
SSF systems are touted as the simple, reliable and cost effective treatment system for small communities. Several factors contribute to the low cost to operate these systems including:
-no chemicals required during treatment
-reduced chlorine required to maintain residual
-they are gravity-based systems so there is low energy input
-minimal labour required particularly with the systems that are hydraulically cleaned
-and there is no replacement of media

The capital costs of these SSF treatment trains more then competitive to other systems including the cost associated with a larger building requirement. In fact, much of the time the building required is of such a simplistic design even these potential additional costs are kept to a minimum.