By Rick Andrew
Carbon block filters have been around for many years and are present in a significant percentage of point of use (POU) products sold. This is no accident. Carbon block technology can include a broad range of water treatment performance capabilities, depending on the ingredients used in the mix and the specific processing techniques and conditions employed to manufacture the carbon block.
One of the basic features of the carbon block is that it uses adsorptive media such as powdered activated carbon and possibly metals sorbents, and through highly sophisticated manufacturing processes, produces an end product that has a rigid structure to it. This combination of adsorptive media and rigid structure allow the carbon block to have both chemical reduction and mechanical filtration properties.
The Standards for Carbon Block Filters
The operative Standards for carbon block filters are NSF/ANSI 42 and 53. These standards have identical requirements for material safety and structural integrity, but cover different types of contaminant reduction performance.
NSF/ANSI 42 addresses reduction of aesthetic contaminants, including those that affect drinking water, such as taste, odor, color, and appearance, which may in turn affect acceptance of public or private drinking water. NSF/ANSI 53 addresses reduction of contaminants that have known health effects.
Many carbon block filtration systems are certified to both NSF/ANSI 42 and 53. This can be accomplished through testing for material safety and structural integrity, requirements that both Standards have in common, and then for various claims under both NSF/ANSI 42 and 53. The claims made on carbon block filter systems under these Standards are completely up to the manufacturer, and are of course also dependent on the filter system to meet the requirements for making the claim.
Chemical Reduction
NSF/ANSI 53 defines chemical reduction as, “The reduction in the quantity of one or more specified organic or inorganic contaminants in drinking water.” The realm of organic and inorganic contaminants is a very broad one, including such familiar contaminants as chlorine, chloramine, volatile organic contaminants, various pesticides and herbicides, and metals including lead.
Chemical reduction involves adsorption and/or absorption by the media in the carbon block. This leads to questions about the capacity of the carbon block, which is impacted by the chemical properties of the contaminant, the concentration of the contaminant, the maximum allowable effluent concentration, the flow rate, and the operational cycle. All of these factors are taken into account in the test methods for chemical reduction in NSF/ANSI 42 and 53.
Testing involves operating the system at the manufacturer’s specified flow rate under NSF/ANSI 42, or at the highest achievable flow rate with an initial clean system inlet pressure of 60 psi under NSF/ANSI 53. The flow through the system is cycled on and off using valves, typically using a 10 minute on and 10 minute off cycle known as “50/50” cycling. At the manufacturer’s discretion, a “10/90” cycle can be used, which is typically 2 minutes of flow followed by 18 minutes without flow. See Figure 1 for a photo of a typical carbon block filter test stand.
Figure 1. Typical Chemical Reduction Test Stand
The concentration of the contaminant in the influent challenge water under NSF/ANSI 42 is set at a typical value based on occurrence.
The concentration of the contaminant in the influent challenge water under NSF/ANSI 53 is typically at the 95th percentile of contaminant occurrence, based on U.S. Geological Survey data or other sources of occurrence data. This means that 95% of those water sources identified to be contaminated with the chemical in question have a concentration equal to or lower than the challenge concentration used in the Standard. When this method suggests a low concentration, or when occurrence data is not available, a concentration of three times the maximum allowable effluent concentration is used.
The maximum allowable effluent concentration for aesthetic contaminants under NSF/ANSI 42 is set based on USEPA secondary maximum contaminant levels or other aesthetic thresholds. The maximum allowable effluent concentration for health contaminants under NSF/ANSI 53 is set at regulated levels based on USEPA or Health Canada limits, or at other health effects concentrations when contaminants are not regulated by these agencies.
Samples of the influent and effluent water are collected at specified intervals throughout the test, typically at six different sample points. Performance for aesthetic contaminants under NSF/ANSI 42 must be acceptable to 100% of the manufacturer’s rated capacity based on volume. Performance for health contaminants under NSF/ANSI 53 must be acceptable to 200% of the manufacturer’s rated capacity based on volume. For those systems incorporating a performance indication device, which is a volume-based filter change indicator, performance must be acceptable to 120% of the manufacturer’s rated capacity based on volume.
Mechanical Filtration
NSF/ANSI 53 defines a mechanical filtration system as, “A system that mechanically separates particulate matter from water.” In contrast to chemical reduction claims, of which there are a vast number included in NSF/ANSI 42 and 53, there are only four mechanical filtration claims, per Figure 2.
Mechanical filtration is dependent on the pore size of the media, the sealing integrity of the filter into the housing, the size and physical nature of the particles being filtered, and the pressure drop of the system. All of these factors are addressed in the test methods for mechanical filtration in NSF/ANSI 42 and 53.
Testing involves collecting samples of the influent and effluent water as the system becomes progressively more and more clogged, until the specified end point. Under NSF/ANSI 42, the end point is a 50% reduction in flow rate, whereas under NSF/ANSI 53, the flow rate must be reduced by 75% before the end point is reached.
Very specific test dust is used to clog the filters. There are actually four different test dust specifications used, depending on the test, per Figure 2. Note that for particulate reduction and turbidity reduction, the dust is the challenge particle. This means that the amount of test dust in the influent and effluent are compared to determine the outcome of the test. For particulate reduction, a laser particle counter is used to count particles in the applicable size range. For turbidity reduction, a nephelometric turbidimeter is used to measure turbidity caused by the test dust.
For asbestos reduction, actual asbestos fibers are used to determine the outcome of the test through analysis by electron microscope. And for cyst reduction, either live Cryptosporidium or polystyrene microspheres are used to determine test results through analysis by epifluorescent microscope. At the sample points, the challenge water containing the test particles is substituted for the test dust clogging water for several cycles, and the samples are collected and analyzed.
The amount of the test particles in the challenge water is based on the required percent reduction and the analytical sensitivity. For example, the claim of cyst reduction requires 99.95% reduction of the challenge particles. With a minimum 50,000 challenge particles per liter in the test water, this means that a count of a minimum of 25 challenge particles in the effluent will be sufficient to determine failure.
50/50 operating cycles are used for mechanical filtration testing, with the effluent samples being collected at the beginning of the on cycle. This is important, because the sudden initiation in flow at the beginning of the on cycle can strain the seals of the system and create a sensitive point in the operational cycle that must be closely evaluated.
Sophisticated Technology, Sophisticated Test Methods
Carbon block filters combine the properties of adsorptive media and physical filtration, and as such represent a very sophisticated water treatment technology. Likewise, the test methods included in NSF/ANSI 42 and 53 are designed specifically around the treatment process and limitations of the technology, to provide conservative and robust measures of performance in the laboratory that will translate into a significant gauge of capability in a variety of real world POU applications.
About the author
Rick Andrew is the Operations Manager of the NSF Drinking Water Treatment Units Program. Prior to joining NSF, his previous experience was in the area of analytical nd environmental chemistry consulting. Andrew has a bachelor’s degree in chemistry and an MBA from the University of Michigan. He can be reached at 1-800-NSF-MARK or email: [email protected] .