October 2001: Volume 43, Number 10
Bottled Water Filters: A Class of Their Own
by Shannon Murphy
ANSI/NSF Standard 53 covers water treatment units and the evaluation of those products’ capability to remove health-related contaminants from drinking water. Under the standard, there are two main types of certification—the complete system and the components. A complete system is defined as a complete water treatment device, including all components needed to connect it to a potable water supply. These products are capable of obtaining a number of contaminant reduction claims as detailed in the standard.
Components, on the other hand, can be certified to the standard and can only be listed for material extraction and—if a pressure-bearing component exists—structural integrity. The standard, as currently written, doesn’t allow for components to claim any reduction capabilities for a number of reasons.
Fitting the bill
A filter is designed to reduce, or remove in the case of Cryptosporidium, contaminants from the water based upon a number of scientific principals. Controls such as flow rates, contact time and the general ability to seal properly play a critical part in the ability of a filter to function properly. These parameters as well as others are also critical for a filter to successfully meet the capacity claims of the system.
For example, a filter capable of obtaining a 1,000-gallon capacity for volatile organic compound (VOC) reduction at 0.5 gallons per minute (gpm) may not be able to meet the same capacity of a system that has a flow rate of 1.0 gpm. This is why it’s critical to use only replacement elements as specified by the certified system.
Likewise, a filter certified as a component cannot be listed for any reduction claims since it hasn’t been evaluated within a defined housing. Just because a filter looks like it will fit into a “standard” housing doesn’t mean that it will function or seal properly. A housing and cartridge must be designed to fit each other in order to obtain the reduction performance as stated on the data plate.
Exceptions to the rule
As with many things in life, there is an exception to this rule. Bottled water filters meeting the 222- and 226-millimeter (mm) designs are capable of becoming certified for Cryptosporidium removal. Why is it then that this replacement element is certified for a reduction claim when it’s obviously not a complete system? The answer is based upon the bottled water industry and the standards they have set up within their industry.
In 1997, the bottled water industry came to NSF with a concern and a request. They had been getting a lot of pressure from their consumers to prove that their bottled water didn’t contain Cryptosporidium. Consumers wanted testing through a third-party laboratory. Upon this increase in consumer awareness, the bottled water industry went to the filter manufacturers and asked if they had third-party testing to prove their filters removed Cryptosporidium. The bottled water filter manufacturers and the bottled water industry then came to NSF to develop a certification for these types of filters. The Drinking Water Treatment Unit (DWTU) Joint Committee agreed to work with the industry in order to provide answers for its consumers.
A filter meeting the dimensional requirements of a “222” or “226” cartridge is specifically designed to fit into a filter cup and housing designed to hold those filters. As developed, only filters that have a redundant filtration element sealing mechanism such as the 222 and 226 double o-ring seals are capable of being tested. These filters can have different locking mechanisms; however, the double o-ring and the 222-mm and 226-mm top dimension remains standard. Since this industry standard already exists, replacement elements for these standard cups and housings can be certified.
Road to certification
The protocol was developed in 1998, under the supervision of the DWTU Joint Committee task group, which consists of bottled water industry and filter manufacturer members. The task group developed the protocol and then came back to NSF and the Joint Committee to have it validated. At the conclusion of this validation process, the Joint Committee decided the standard could permit the certification of specific bottled water filters.
The task group identified that the testing protocol would have to be different than the protocol used for Cryptosporidium removal for other point-of-use/point-of entry (POU/POE) systems. Unlike a household unit where the system would see a constant inlet pressure—and as the filter is plugged up with particles, the flow rate would be reduced—bottling water plants have a specific production rate that they must maintain. For bottled water plants, as the filter plugs up, the pressure delivered to the filter is increased in order to maintain the desired gpm for production goals.
With this in mind, the task group worked out a system where the filter would initially see the rated service flow of the filter as specified by the manufacturer. The manufacturer’s rated service flow is then maintained throughout the test. Samples are collected at the start up—25, 50, 75, 100 and 150 percent of the manufacturer’s recommended maximum pressure drop at the rated service flow. Currently, the standard has a maximum pressure drop capacity of 90 pounds per square inch (psi). Therefore, the maximum pressure differential allowed under the standard is 60 psi. With this in mind, a bottled water filter with a rated maximum pressure drop of 60 psi contains samples that will be collected at start up, 15, 30, 45, 60 and 90 psi. Additionally, immediately prior to collection of the effluent samples, a pressure pulse is administered to the system. This is meant to replicate the usage patterns of a filter such as in a bottling plant when operating a fast-acting fill valve. This also provides additional stress on the filter media, which may cause a release of microsphere particles.
It’s interesting to witness how the increased awareness of consumers can become a driver for third-party certification. As we’ve seen, through consumer requests, the bottled water industry recognized that there was a need, which the DWTU Joint Committee was able to provide solutions through development and validation testing of a protocol. There’s now the existence of certification of a component with performance claims for bottled water filtration. As we’ve seen, with the right mixture of industry standards, teamwork, protocol development and validation, certification organizations and the industry can be quick to respond to consumer and market requirements.
About the author
Shannon Murphy is operations manager of the Drinking Water Treatment Unit program at NSF International in Ann Arbor, Mich. His bachelor’s degree from Concordia University in Montreal, Canada, is in biology. His master’s degree from Wayne State University in Detroit is in biology with an emphasis on limnology. Murphy can be contacted at (800) 673-6275, (734) 827-7144 (fax) or email: firstname.lastname@example.org