By Natalya Eagan-Rosenberg

The U.S. Environmental Protection Agency (USEPA) has proposed lowering the maximum contaminant level (MCL) for arsenic in Drinking Water Standards from 50 parts per billion (ppb) to either 20, 10 or 5 ppb, based on the arsenic’s potential as a carcinogen. As a result of the impending MCL change, NSF and the drinking water treatment unit (DWTU) industry—also referred to as the point-of-use/point-of-entry (POU/POE) industry—see a need to produce safe, reliable drinking water treatment units that reduce arsenic to levels at or below the new standard.

Manufacturers are anticipating greater demand for arsenic reduction devices as consumers realize arsenic levels in their drinking water, thought to be safe, may now be above the MCL. Demand will also increase if the USEPA allows municipalities that meet certain conditions to incorporate home DWTU systems into their programs for providing safe drinking water.

Task group
Although an arsenic reduction protocol was reinstated in ANSI/NSF Standard 58 in September 1999, reduction claims that can be made under this standard are limited to reverse osmosis (RO) technologies used on chlorinated systems. An NSF task group has been formed of volunteers from the DWTU industry—as well as state and federal regulatory agencies—to work on reinstating arsenic claims in Standard 53, DWTU-Health Effects, and expand the current claims available under Standard 58. As a result of this effort, the option to be certified for arsenic reduction will soon be extended to any technology that can reduce arsenic to a level at or below the new MCL when tested under the conditions outlined in the standards.

Some of the elements of the draft protocols under discussion by the task group will probably be similar to both ANSI/NSF 53 and 58 and are outlined below.

Types of arsenic
Arsenic exists in many forms, but the two most commonly found in drinking water are pentavalent arsenic (As-V) and trivalent arsenic (As-III). In general, arsenic in water that has been exposed to air, or some other oxidizing agent, is found as As-V. As-V is usually easier to remove. This is because As-V is present as an ionic species at pH 7.5 and many technologies remove ionic (charged) species easier than neutral species. Technologies that remove As-V, such as RO, are well understood and tend to be less expensive since they involve less research and development. Water that’s distributed by a municipality should also contain arsenic only in As-V since the disinfection required by the municipality will oxidize As-III to As-V. For these reasons the task group is suggesting technologies have the option to be certified for removal of As-V only.

As-III is commonly found in groundwater that hasn’t been exposed to air or other oxidizing agents and is more difficult to remove. Most technologies that remove As-III will also remove As-V. Technologies designed to remove As-III must either first convert the arsenic to As-V and remove it by established technologies or incorporate a way to reduce uncharged As-III species. This means if the type of arsenic in the water is unknown, or the water is known to contain As-III, a treatment system capable of removing total arsenic (III and V) is needed.

New influent challenge
ANSI/NSF 58 currently requires an influent arsenic challenge of 300 ppb; however, the task group is considering giving manufacturers the option of challenging the test unit with an arsenic concentration of either 500 ppb or 50 ppb. Analysis of USGS occurrence data by the task group showed 95 percent of source waters that contain arsenic—and are used for domestic purposes—should have arsenic concentrations of less than 50 ppb. Also, there are many functioning arsenic reduction systems that produce an effluent stream at or below 50 ppb, and these systems will need to be modified to reduce the arsenic further to the new MCL. It’s anticipated homeowners with levels in the 50-ppb range will need to reduce arsenic but won’t require a system designed to remove higher levels.

Meanwhile, a number of areas in the country experience levels well above 300 ppb and it’s imperative systems claiming to reduce arsenic in these situations have their performance verified to protect public health. These two “tiers” would be the basic claims that can be made under the DWTU standards. As with any of the contaminants covered in the NSF standards, a manufacturer who has a product that can reduce arsenic from an even greater concentration will have the option of testing at a higher level.

The effluent limit will be set at the new MCL level wherever it falls. Any unit that is certified under one of the DWTU standards must reduce arsenic to a concentration at or below the level set by the USEPA.

Consumer help
The task group is aware that allowing claims for different influent concentrations and different types of arsenic are liable to cause considerable confusion for a consumer. For this reason, the group is working on developing improved consumer information on arsenic in the instructions, performance data sheet and data plate that are required with each product certified under ANSI/NSF standards. The current concept is that an “arsenic fact sheet” will be required. Details of the fact sheet have not been discussed but NSF anticipates it will, at the least, simply explain the exact claim being made and the need for each customer to verify that his/her system is actually reducing arsenic. It may include an explanation of the difference between As-V and As-III, and how the effectiveness of the device depends on source water characteristics.

Increased monitoring
There is concern that no amount of literature will ensure that the consumer uses/installs the correct technology for his/her arsenic problem. As an additional safety measure, the task group is considering requiring that each unit come with the equipment and instructions needed to test the product water and verify the treated arsenic level is below the MCL. One avenue of accomplishing this is to supply the consumer with a sample bottle to fill and send to an accredited lab for arsenic analysis. Another option includes a kit so that the consumer can test their own product water for arsenic after the reduction unit is installed. It is anticipated that if technology is developed to monitor arsenic as it exits the treatment system, it could also fulfill the requirement. In addition to verification that the system works upon installation, the standard would probably require re-testing at discrete intervals since arsenic in groundwater can change species and concentration over time.

Simulating conditions
Much of the research conducted by the task group has focused on developing challenge water that would mimic field conditions. This is important because many ions commonly found in drinking water interfere with the arsenic removal technologies covered in ANSI/NSF 53. At this point, the background ions (Mg+2, SO4-2, NO3-N, F-, SiO2, PO4 and Ca+2) have been included at average occurrence levels in the proposed challenge water, because it’s unlikely that a natural water would contain “worse case” levels of all possible competing ions.

Also influencing the effectiveness of many arsenic reduction technologies is pH and for that reason testing is proposed at pH 6.5 and 8.5 to cover the range of commonly found source waters.

As noted earlier, one way of reducing As-III is to convert it to As-V before utilizing technology that removes As-V only. With this in mind, the first draft of the Standard 53 protocol includes the option to certify an “oxidizer unit” that converts As-III to As-V. The unit could be marketed as part of a total arsenic removal unit or as a stand-alone unit to be used with an As-V removal unit. The protocol for performance testing of the oxidizer unit has not been discussed, but it would determine the conversion capacity of the unit. Systems claiming total arsenic reduction that incorporate an oxidation unit would have the oxidation technology tested with a protocol that determines conversion capacity of the oxidizer unit.

RO options
Arsenic removal by RO is less dependent on pH than some technologies. Still, since ROs have most difficulty removing non-charged species, the conservative pH for RO testing is thought to be 6.5. This is because As-V is in a more neutral form at pH 6.5.

Similar to ANSI/NSF 53, RO manufacturers who wish to market their system under a total arsenic reduction claim would have the option of attaching a “preoxidizer.”

RO would still undergo the 58 protocol testing to remove As-V. The preoxidizer would be tested under the new 53 proposal and the capacity rating of the preoxidizer would be stated in the literature.

The draft protocols, which incorporate these proposed changes, were discussed at the Sept. 26-27 meeting of the Arsenic Task Group. Although there may be significant changes to the protocol outlined here, NSF anticipates completing this project next year so that consumers can easily obtain the right DWTU unit for their arsenic problems.

About the author
Natalya Eagan-Rosenberg is a project manager in the Standards Department at NSF International in Ann Arbor, Mich. Her bachelor’s degree from the University of California, Berkeley, is in environmental science. Her master’s degrees from the University of Michigan are in resource ecology & management and environmental health, with an emphasis on water quality. Eagan-Rosenberg can be contacted at (800) 673-6275, (734) 827-6831 (fax) or email: [email protected]

If you have questions concerning this column, or if there’s a topic you would like addressed, please let us know. Contact “Water Matters” at: WC&P Magazine, 7522 N. La Cholla Blvd., Tucson, AZ 85741; (520) 323-6144, (520) 323-7412 (fax) or email: [email protected]


Comments are closed.