16 Apr Going Low NTU!
Why Granular Media Filtration Works Best for Water Treatment
April 17, 2019 – For over a century, due to numerous mechanisms, granular media filtration has provided reliable removal of pathogens, contaminants, and nuisance particles in water and wastewater and has added resiliency to public health and the environment. The filtered wastewater turbidity limit of <3.0 NTU is acceptable for non-potable reuse in numerous US states; however, the drinking water regulatory limit is an order of magnitude lower (0.3 NTU), requiring significantly-improved treatment. Unlike other types of filters (e.g., “disk” or compressible media), the multiple removal mechanisms of granular media filtration ensure effluent requirements can be repeatably met.
Particle removal in a granular media (e.g., sand, anthracite) water or wastewater filter is performed by several mechanisms (see Figure 1), to include:
Flocculated particles with density greater than water will settle on the surface of the media or horizontal filter surfaces
As fluid streamlines between media particles constrict, particles come in contact with the media and are removed
Particles below 1-2 µm are captured similar to ion exchange. Particle charge neutralization during coagulation is necessary for effective media attachment
Particles larger than the media pore space are removed mechanically
Straining is the main filtration removal mechanism for relatively-large, flocculated, wastewater secondary suspended solids; thus, nominal 10 µm filters (e.g., disk) with little media depth are effective. Since these type filters are predominantly used on flows less than 15 MGD, reject volumes from routine backwashing are relatively unimportant. However, using these types of filters for higher-capacity wastewater treatment facilities create significant retreatment volumes as well as added and more complex maintenance.
Reliance on straining as the removal mechanism in granular media filters is inadvisable. Should solids be captured only at the top of a granular media bed, rapid head loss would necessitate overly-frequent backwashing. Pretreatment (i.e., coagulation, flocculation, sedimentation) processes must be optimized to remove relatively-large particles and allow filtration vertically through the media bed depth. Deeper designs of larger-sized granular media can compensate for straining potential and produce acceptable effluent. Or, use of two or three stratified layers of medias with increasing density and decreasing size (from top to bottom) removes larger solids near the bed top and smaller solids further down.
In granular media filtration, interception and diffusion provide the mechanisms for removal of small, sub-10 µm particles (e.g., colloids, poliovirus, and Cryptosporidium). Standard practice is to design filtration sand with an effective size of 0.5 mm (500 µm). In Figure 2, one can see with 500 µm diameter (spherical) media that 75 µm diameter particles will pass through the pore space! Without interception and diffusion in granular media beds, drinkable water would be scarce
The force of flowing water through a granular media bed shears away some material before it becomes firmly attached. These particles penetrate deeper into the media bed, increasing removal efficiency and filter run time before backwash. Minimum particle removal efficiency occurs for particle sizes of ±1 µm, the transition point between interception and diffusion (see Figure 3). Reduced particle removal efficiency occurs at higher filtration rates, particularly for smaller particles. Further, as the media bed becomes clogged, surface shear increases to a point at which particles “break through” the bottom of the filter media, causing effluent turbidity to increase—the filter must then be backwashed.
Particle removal via interception and diffusion is highly dependent on the surface area of the granular media. Typical filtration sand with porosity of 0.40 will have 9.17×109 particles per cubic meter with gross surface area of 7.20×103 m2/m3 (assuming spherical shape). Because media particles shield each other, even an assumption of only 1% of the surface (72 m2) being effective yields far more area than without any depth of media (1 m2). This provides explanation for the ineffectiveness of many wastewater filtration technologies in potable water applications.
Figure 1: Adapted from Figure 5-2, Operational Control of Coagulation and Filtration Processes (M37). American Water Works Association (AWWA), 2011.
Figure 2: Adapted from Figure, p.10 of notes handout, Optimizing Filter Performance: Surveillance, Data Analysis and Maintenance. AWWA, ACE17 Conference, SUN01 Workshop, 2017.
Figure 3: Adapted from Figure 5-3, Operational Control of Coagulation and Filtration Processes (M37). AWWA, 2011.
Other References:
Tchobanoglous, George, and Franklin Burton. Wastewater Engineering: Treatment, Disposal, and Reuse. McGraw-Hill, 1991.
Droste, R. L. Theory and Practice of Water and Wastewater Treatment. Wiley, 1997.
Stevenson, David G. Water Treatment Unit Processes. Imperial College Press, 1997.
Make Sure To See All Our Video & Blog Posts!
Resilient – Reliable – Sustainable – Permanent
Get In Touch With Orthos