Sand filters remove suspended particles, turbidity, bacteria, protozoa, many viruses, dissolved iron and manganese, a portion of natural organic matter, and even microplastics from water. They have been a cornerstone of municipal and household water treatment for over a century, helping end cholera and typhoid epidemics in the early 1900s. But the list of what sand filters catch, and how effectively, depends heavily on the type of filter, the grain size of the sand, and how long the filter has been in operation.
Turbidity and Suspended Solids
The most visible job a sand filter performs is removing turbidity, the cloudiness caused by tiny particles suspended in water. Sand grains create a tortuous path that traps sediment, clay, silt, algae, and other particulates as water moves through the bed. Experimental filters using fine sand have achieved around 89% turbidity removal on average, though performance varies with design and operating conditions.1PubMed. Bacterial, viral and turbidity removal by intermittent slow sand filtration for household use in developing countries: experimental investigation and modeling Rapid sand filters used in municipal plants typically aim for filtered water turbidity well below 1 NTU (nephelometric turbidity unit), and optimized backwash strategies can keep turbidity spikes during the filter’s restart phase to a fraction of that target.2Journal of Water Process Engineering. Investigation of backwash strategy on headloss development and particle release in drinking water biofiltration
The mechanism is more than simple straining. Research on particles in the 0.75 to 10 micrometer range shows that physical and chemical interactions between particles and sand grains, like electrostatic attraction and van der Waals forces, play a significant role alongside the sieving effect. An interesting finding is that a filter’s particle removal efficiency actually increases as more particles accumulate in the bed, because the captured particles themselves become additional collectors.3Journal AWWA. Physical—chemical mechanisms in slow sand filters This “ripening” effect means a freshly cleaned filter is often its least effective state.
Bacteria and Protozoa
Sand filters are remarkably effective against many waterborne pathogens, though the degree of removal depends on what kind of organism you are dealing with. When operated correctly, slow sand filters can remove more than 99.9% of Giardia and Cryptosporidium cysts from water.4Water Research. Slow sand and diatomaceous earth filtration of cysts and other particulates These protozoan parasites are relatively large, on the order of several micrometers, which makes them susceptible to physical straining and attachment to sand grains.
Bacteria removal is also strong. Household slow sand filters in developing countries achieved an average bacterial removal of about 96% under typical conditions, climbing to about 98.5% with optimal design (fine sand and low flow rates).1PubMed. Bacterial, viral and turbidity removal by intermittent slow sand filtration for household use in developing countries: experimental investigation and modeling That sounds impressive, and it is a massive improvement over untreated water. But it also means a few percent of bacteria pass through, which is why sand filtration alone is generally not considered sufficient for producing drinking water without additional disinfection.
Real-world performance can slip under adverse conditions. A field study of a slow sand filtration plant found an average 93% removal of Giardia cysts, but Cryptosporidium removal averaged only 48%, with the parasite detected in nearly half of filtered samples. Cold water temperatures and deviations from accepted design standards were flagged as likely causes.5Journal AWWA. Removing Giardia and Cryptosporidium by Slow Sand Filtration Cryptosporidium oocysts are notoriously tough, resistant even to chlorine, so their partial passage through sand filters under suboptimal conditions is a real concern for water utilities.
Viruses
Viruses are far smaller than bacteria or protozoa, and sand filters generally have a harder time with them. Experimental intermittent slow sand filters removed MS2 bacteriophage (a common virus surrogate in research) by an average of about 0.54 log units under standard conditions, which translates to roughly 71% removal. Under the best design configuration, removal improved to about 0.94 log units, or roughly 89%.1PubMed. Bacterial, viral and turbidity removal by intermittent slow sand filtration for household use in developing countries: experimental investigation and modeling That is nowhere near the multi-log reductions required by drinking water standards, which is why virus removal through sand filtration depends on biological activity within the filter rather than straining alone.
In an established slow sand filter, both the surface biological layer (known as the schmutzdecke) and deeper layers of the sand bed contribute to viral removal. Research has confirmed that retention rates for viruses are highest in the top layer of the filter, driven by a combination of physical-chemical factors like smaller pore sizes and the presence of iron and manganese oxides, as well as biological factors including dense biofilm.6Water Research. The contribution of deeper layers in slow sand filters to pathogens removal Without that mature biological community, virus removal drops sharply. This is one reason why slow sand filters need time to “mature” before they perform at their best, and why disrupting the schmutzdecke can temporarily degrade water quality.
How the Schmutzdecke Makes Filtration Biological
One of the most distinctive features of slow sand filtration is the schmutzdecke, a thin biological mat that develops on the surface of the sand bed over days to weeks. This layer is alive with bacteria, algae, protozoa, and other microorganisms that actively consume or break down contaminants in the water passing through. It transforms what might otherwise be a purely physical process into something closer to a living treatment system.
The contribution of this biological activity extends well below the surface. Studies have shown that E. coli removal in an established slow sand filter is driven by active biofilm-led processes including predation, grazing, and enzymatic inactivation at multiple depths within the sand bed. Physical mechanisms like straining and attachment to biofilm-coated sand also contribute, but the biological processes are central to how effectively the filter handles bacteria.6Water Research. The contribution of deeper layers in slow sand filters to pathogens removal
Biofilm growth is not always welcome, though. In intermittent sand filters, the same biological activity that removes contaminants eventually leads to clogging, as the biomass (dominated by heterotrophic bacteria) builds up in the top layers of the filter. This clogging zone does offer strong removal of biodegradable organic matter and ammonia, but if left unchecked, it reduces flow to a trickle and the filter needs maintenance.7PubMed Central. Managing biofilm growth and clogging to promote sustainability in an intermittent sand filter (ISF) Operators of slow sand filters periodically scrape off the top few centimeters of sand to restore flow, effectively resetting the schmutzdecke and beginning the ripening process again.
Iron and Manganese
Dissolved iron and manganese are among the most common aesthetic complaints in groundwater. They are not typically dangerous at the concentrations found in most wells, but they stain laundry, discolor water, and produce unpleasant tastes. Sand filtration is one of the standard methods for removing both.
The process works by oxidizing dissolved iron and manganese (which are invisible in water) into solid forms that the filter can then trap. This oxidation can happen naturally at the sand surface, especially when the sand grains become coated with iron and manganese oxides over time. These coatings act as catalysts, speeding up the conversion of dissolved metals to particles. Research has shown that manganese dioxide, particularly the delta form, plays a crucial role in this auto-catalytic process.8PubMed. Oxidants-assisted sand filter to enhance the simultaneous removals of manganese, iron and ammonia from groundwater: Formation of active MnOx and involved mechanisms
Some systems add an aeration step before the sand filter to pre-oxidize the metals, which makes the filter’s job easier. One study of an aeration-manganese sand filter system found that it removed all dissolved iron and about 90% of manganese from the water, while also reducing fouling on downstream membrane filters.9Journal of Water Process Engineering. Aeration-manganese sand filter-ultrafiltration to remove iron and manganese from water: Oxidation effect and fouling behavior of manganese sand coated film If you have orange or brown staining in your water, a properly configured sand filter with an aeration pretreatment step can solve the problem.
Arsenic
Arsenic contamination in groundwater affects tens of millions of people, particularly in South and Southeast Asia. Standard sand filters were not originally designed with arsenic in mind, but they remove a meaningful fraction of it as a side effect of iron removal. When groundwater contains both arsenic and dissolved iron, the arsenic binds to iron particles as they oxidize and precipitate, and the sand filter then traps those particles. A field study of 43 household sand filters in Vietnam’s Red River Delta found average arsenic removal of about 80% from water containing 10 to 382 micrograms per liter of arsenic.10PubMed. Arsenic removal from groundwater by household sand filters: comparative field study, model calculations, and health benefits
Eighty percent removal is significant for highly contaminated sources, but it may still leave arsenic above the WHO guideline of 10 micrograms per liter if the source water is heavily contaminated. For more consistent and thorough arsenic removal, sand coated with iron oxides has been studied as a specialized medium. Laboratory tests of iron oxide-coated sand achieved 99% arsenic removal with a relatively simple column design that could be adopted in small communities.11PubMed. Assessment of arsenic removal efficiency by an iron oxide-coated sand filter process The iron coating provides a reactive surface that adsorbs both arsenate and arsenite, the two common forms of arsenic in water, more effectively than plain sand.
Natural Organic Matter
Natural organic matter, the dissolved plant and soil material that gives some water sources a tea-like color, is more than an aesthetic nuisance. When organic matter reacts with chlorine during disinfection, it can form disinfection byproducts that pose health risks. So removing organic matter before chlorination is a real priority in drinking water treatment.
Sand filtration removes a modest portion of this material. A comparison study found that sand filtration removed about 22% of UV-absorbing organic compounds, 20% of humic-like substances, and 26% of protein-like substances from source water.12PubMed. A comparison study of sand filtration and ultrafiltration in drinking water treatment: Removal of organic foulants and disinfection by-product formation Those are not dramatic numbers, but sand filtration also reduced the formation of certain disinfection byproducts downstream, performing slightly better than ultrafiltration membranes at lowering chlorite concentrations. Sand filters appear to be particularly effective at capturing medium-sized organic molecules, while membranes do better with larger ones.
For substantially higher organic matter removal, dual-media filters that combine a layer of anthracite coal on top of sand roughly double the efficiency. Pilot-scale trials showed dual-media filters removed about 65% of total organic carbon, compared to about 34% for sand-only filters at the same flow rate.13Desalination and Water Treatment. Removal of organic matter from drinking water by single and dual media filtration: a comparative pilot study The anthracite layer, being lighter and coarser, captures larger flocs in the upper portion of the filter while allowing finer particles to penetrate deeper into the sand layer below, making better use of the full depth of the filter bed.
Microplastics
Sand filtration is increasingly being studied for its ability to trap microplastics, the tiny plastic fragments and fibers that have been found in virtually every water source tested. For larger microplastics (those between 500 and 5,000 micrometers), sand filters are quite effective, removing about 81% of particles in that size range. But as microplastic size decreases, so does the filter’s ability to catch them. Removal dropped to only about 17% for particles in the 100 to 200 micrometer range, and sand grain size was a major factor: filters with the finest sand (0.5 to 1.0 mm) removed about 90% of microplastics overall, while coarser sand (1.5 to 2.0 mm) only managed about 56%.14PubMed. Sustainable sand filtration strategies for microplastic removal in irrigation water Fibrous microplastics were much easier to trap than rounded particles, at about 85% versus 30%, likely because fibers snag on sand grains while smoother particles slip through.
At the very smallest end of the microplastic spectrum, rapid sand filtration in tap water treatment shows a different picture. Research on particles smaller than 10 micrometers found that rapid sand filters achieved about 98% removal of these tiny fragments.15PubMed. Rapid sand filtration for <10 μm-sized microplastic removal in tap water treatment: Efficiency and adsorption mechanisms This seems counterintuitive given the struggle with 100-200 micrometer particles in the other study, but the explanation lies in the type of filter: rapid sand filters in municipal plants operate after chemical coagulation and flocculation, which agglomerates tiny particles into larger clumps that the filter can then catch. The irrigation-focused study used sand filtration without those pretreatment steps. The lesson is that sand filtration’s performance with microplastics depends enormously on what treatment happens upstream.
What Sand Filters Do Not Remove Well
For all their versatility, sand filters have genuine blind spots. Dissolved chemicals that do not readily attach to sand grains or form particles pass through largely unaffected. This includes most dissolved salts, nitrates, fluoride, and most synthetic organic chemicals like pesticides, pharmaceuticals, and industrial solvents. A sand filter will not reduce the hardness of your water, remove chlorine taste, or address contamination from fuel spills.
Volatile organic compounds, persistent organic pollutants, and endocrine-disrupting chemicals are essentially invisible to conventional sand filtration. These molecules are small, often nonpolar, and do not interact strongly with sand surfaces. Removing them typically requires activated carbon adsorption, advanced oxidation processes, or membrane technologies that operate on fundamentally different principles than granular filtration.
Even for contaminants that sand filters do remove, the level of removal may not be sufficient on its own. As discussed earlier, virus removal by sand alone is rarely adequate to meet drinking water safety standards, and Cryptosporidium removal can falter in cold conditions. Sand filtration works best as one step in a multi-barrier treatment train, usually preceded by coagulation and followed by disinfection. Expecting a sand filter to single-handedly produce safe drinking water from a heavily contaminated source is asking too much of the technology.
Rapid Versus Slow Sand Filters
These two types of sand filters work at fundamentally different speeds and through somewhat different mechanisms, which matters for what they remove. Slow sand filters process water at rates typically measured in fractions of a meter per hour. That slow pace gives biological processes time to work, which is why the schmutzdecke becomes so important. The biological component gives slow sand filters an edge in pathogen removal without chemical pretreatment, and they are widely used in smaller communities and developing-country settings precisely because they need less chemical input and less operator expertise.
Rapid sand filters, the type found in most large municipal water plants, push water through at rates roughly 10 to 50 times faster. At those speeds, biological activity in the bed has far less time to act on contaminants, so rapid filters depend more heavily on chemical pretreatment (coagulation and flocculation) to form larger particles that can be physically trapped. They need regular backwashing, where water is pumped backward through the bed to flush out accumulated material. Different backwash strategies produce different results: adding air scour and extended terminal subfluidization wash phases can cut head loss by 11 to 18% and dramatically reduce the turbidity spike that occurs when the filter comes back online.2Journal of Water Process Engineering. Investigation of backwash strategy on headloss development and particle release in drinking water biofiltration
Flow direction also affects performance. Downflow filters (water entering from the top) and upflow filters behave differently when it comes to how quickly turbidity breaks through and how evenly particles are distributed through the bed. Upflow filtration tends to distribute captured particles more evenly across the filter depth, which in principle makes better use of the entire sand bed. However, once breakthrough begins, upflow filters see turbidity rise faster than downflow designs, and the two modes respond differently to backwashing, requiring different maintenance approaches.16Separation and Purification Technology. Downflow and upflow sand filtration for removing turbid particles with diverse morphological and surface-charge characteristics: Comparisons of filtered water quality and head loss distribution
Sand Filtration in Historical Context
Sand filtration is among the oldest engineered water treatment methods still in widespread use. Its adoption in the 19th century marked one of the most consequential public health interventions in history. The connection between filtered water and reduced disease was observed before germ theory was fully established. By the early 1900s, sand filtration had helped end cholera and typhoid epidemics in cities that adopted it.17Journal AWWA. From Slow Sand to High‐Rate Processes, Filtration Remains the Primary Protection
The technology has evolved considerably since then. Rapid sand filtration emerged to keep pace with growing urban water demand, dual-media and multi-media filters improved organic matter capture, and engineered coatings like iron oxide on sand grains extended the filter’s reach to contaminants like arsenic. But the fundamental principle, passing water through a bed of granular material and relying on a combination of straining, adhesion, and biological activity, remains essentially unchanged. In an era of advanced membrane filtration and ultraviolet disinfection, sand filters still anchor most of the world’s drinking water plants, not because they do everything, but because they reliably do a lot at modest cost and complexity.
Household and Small-Community Sand Filters
If you are considering a sand filter for your home, well, or small community, it helps to know where the technology’s sweet spot lies. A biosand filter, the household version of a slow sand filter, is a concrete or plastic box filled with layered sand and gravel. Once the biological layer matures (usually after a few weeks of daily use), it provides substantial reduction in bacteria, protozoa, and turbidity. For people drawing water from rivers, ponds, or shallow wells in areas without centralized treatment, a biosand filter can be transformative.
The practical limits are real, though. Household filters using fine sand and low flow rates have achieved roughly 98.5% bacterial removal and about 89% virus removal under optimal conditions, but those conditions include consistent daily use, warm enough temperatures, and proper construction.1PubMed. Bacterial, viral and turbidity removal by intermittent slow sand filtration for household use in developing countries: experimental investigation and modeling Sand grain size is one of the most critical design factors: coarser sand lets water through faster but catches less. If you build or buy a biosand filter, the quality and size of the sand matters more than almost any other variable.
For well water in developed countries, the usual application is not pathogen removal (chlorination or UV handles that) but rather iron and manganese reduction, sediment removal, or turbidity control. These are cases where sand filtration does well, especially when paired with an aeration step for iron-heavy water. If your concern is chemical contamination, nitrates, or total dissolved solids, a sand filter is the wrong tool and you will want to look at activated carbon, reverse osmosis, or ion exchange systems instead.