In the early 2000s, the contamination of drinking-water sources with arsenic in the United States was publicly well known and covered frequently in the media because the U.S. Environmental Protection Agency (EPA) lowered the maximum contaminant level (MCL) from 50 micrograms per liter (µg/L) to 10 µg/L. Since then, concerns about arsenic in drinking water have been replaced in the public consciousness with concerns over lead and polyfluoroalkyl substances (PFAS). Unfortunately, this does not mean the public health danger from arsenic in drinking water has been resolved. In fact, the threat may be growing due to more frequent and widespread drought conditions, not only in the western United States but globally as we experience climate change.
Arsenic, a naturally occurring element, may be colorless, odorless, and tasteless, making it an invisible danger that could infiltrate our water systems. Long-term exposure to arsenic at even low levels in drinking water has been linked to serious health issues, such as cardiovascular diseases, high blood pressure, diabetes, and various cancers.
The World Health Organization estimates that 140 million people in at least 70 countries drink water with arsenic levels above 10 µg/L.1 This situation may worsen due to recent drought conditions, leading to increased groundwater pumping from aquifers for irrigation and human consumption. Public health officials and researchers throughout the western U.S.2, Spain3, and Australia4 have documented increasing arsenic concentrations in groundwater, which they believe is caused by over-pumping aquifers. A U.S. Geological Survey modeling study predicted that drought conditions have increased the number of people in the U.S. who have been exposed to arsenic levels above the MCL to 4.1 million, up from 2.7 million. 5
Fortunately, arsenic can be readily removed from drinking water by point-of-use (POU) or point-of-entry (POE) water treatment solutions, including reverse osmosis (RO) and active media filtration. The performance of these water-treatment systems can be verified through independent, third-party certification to NSF/ANSI 53: Drinking Water Treatment Units – Health Effects or NSF/ANSI 58: Reverse Osmosis Drinking Water Treatment Systems. Both standards outline arsenic-reduction requirements, detail testing methods that can be employed for certification, and require certified products to contain educational documents for end-users.
Forms of Arsenic
Arsenic occurs in water in two forms known as oxidation states. These are pentavalent arsenic, also called As (V), As (+5), or arsenate; and trivalent arsenic, also known as As (III), As (+3), or arsenite. These forms have different chemical properties and therefore respond differently to treatment technologies.
As (III) is typically more challenging than As (V) to remove from drinking water with most filtration methods. The good news is that chemically changing As (III) to As (V) is easy using oxidizing disinfectant chemicals such as chlorine, or exposure to oxygen in the air. Once the chemical change has taken place, the water can be treated more easily.
Verifying Active Media Filtration
with NSF/ANSI 53
NSF/ANSI 53 allows two options for an arsenic reduction performance claim: a general arsenic-reduction claim or pentavalent arsenic reduction. As very few water supplies are contaminated only with As (III), NSF/ANSI 53 does not allow for a performance claim of As (III) reduction alone. A claim of pentavalent arsenic reduction only is included in the standard because it is more common to find water supplies with arsenic present only in the pentavalent form, and because As (III) can be easily converted to As (V) prior to filtration treatment.
The arsenic-reduction claim requires successful testing for reduction of both As (III) and As (V), while the pentavalent arsenic reduction claim requires only As (V) reduction testing. Additionally, since filtration media can respond differently to arsenic at different pH levels, the standard requires arsenic-reduction testing to be conducted at both pH 6.5 and pH 8.5.
The arsenic-reduction claims are unique in that manufacturers have two options for the influent challenge level: 300 µg/L or 50 µg/L. The 50 µg/L influent option is provided so consumers of water supplies that met the previous MCL can choose a treatment device that provides adequate arsenic reduction for their needs.
The water characteristics outlined in Figures 1 and 2 specify the concentrations of various ions present in the test water in addition to arsenic. These substances are commonly present in drinking water and may impact the filtration media’s ability to remove arsenic. Therefore, requiring a specific test-water recipe enhances the repeatability of the test and represents real-world conditions.
Free available chlorine must be present in As (V) reduction test waters to ensure the arsenic remains in the pentavalent form. Conversely, the As (III) reduction test water requires that very little dissolved oxygen be present to ensure the arsenic remains in the trivalent form.
The filter systems are tested to make sure treated water samples do not contain more than the MCL of 10 µg/L of arsenic over the claimed life of the filter. Two systems are operated in the testing laboratory at the highest achievable flow rate with 60 pounds per square inch inlet pressure. The systems are cycled on and off so that water is flowing 50 percent of the time. The test units are operated in this manner 16 hours per day, with an eight-hour rest period overnight. To ensure a margin of safety, the systems are tested to 200 percent of the manufacturer’s claimed capacity, or 120 percent if they include a filter-change indicator. This product testing method ensures they will perform effectively under heavy usage beyond the marketed product life.
Leveling Up POU RO Systems
with NSF/ANSI 58
RO membranes have limited effectiveness in treating As (III) but are much more effective in treating As (V). Therefore, NSF/ANSI 58 offers a performance claim for the reduction of As (V) but not As (III). Consistent with a typical approach to treating groundwater contaminated with As (III), the As (V) reduction claim applies only to system usage on water supplies with a free chlorine residual present, or water supplies demonstrated to contain only As (V). A chlorination device just upstream of the RO system is usually used to ensure that all arsenic in the water is oxidized to the pentavalent form before passing through the RO system.
Unlike under NSF/ANSI 53, As (V) reduction testing under NSF/ANSI 58 occurs only at pH 7.5 because the removal of arsenic by RO is not significantly affected by the pH of the water supply. The test water is pure water that has been treated by RO and deionization, to which sodium chloride is added at a concentration of 750 milligrams per liter. As with NSF/ANSI 53 testing, the manufacturer may choose to test with an As (V) influent concentration of either 50 µg/L or 300 µg/L. In either case, the system must reduce the arsenic to 10 µg/L or less in all treated water samples collected throughout the test to earn certification.
Figure 1: Pentavalent Arsenic Reduction Test Water
Reagents are added to reverse osmosis deionized (RO/DI) water to produce test water with the following characteristics:
| Parameter | Target Value |
| Mg2+ | 12 mg/L |
| NO3–N | 2 mg/L |
| F- | 1 mg/L |
| SiO2 | 20 mg/L |
| PO43–P | 0.04 mg/L |
| Ca2+ | 40 mg/L |
| As (V) | 0.05 mg/L or 0.30 mg/L |
| Temperature | 20°C |
| Turbidity | < 1 NTU |
| Free available chlorine | 0.25-0.75 mg/L |
| pH | 6.5 and 8.5 |
Figure 2: Trivalent Arsenic Reduction Test Water
Reagents are added to RO/DI water to produce test water with the following characteristics:
| Parameter | Target Value |
| Mg2+ | 12 mg/L |
| NO3–N | 2 mg/L |
| F- | 1 mg/L |
| SiO2 | 20 mg/L |
| PO43–P | 0.04 mg/L |
| Ca2+ | 40 mg/L |
| As (III) | 0.05 mg/L or 0.30 mg/L |
| Temperature | 20°C |
| Turbidity | < 1 NTU |
| pH | 6.5 and 8.5 |
| Dissolved oxygen | < 0.5 mg/L |
RO systems are tested over the course of a week. Operation and sampling cycles are varied during the test to allow for performance evaluation under differing usage conditions. An example of this can be seen in the required steps for an arsenic-reduction test on a typical POU RO system with an automatic shut-off valve and storage tank:
- Completely empty a full storage tank and collect a sample, then allow the tank to refill.
- Empty a full storage tank to the point at which the automatic shut-off valve is activated, collect a sample, then allow the tank to refill.
- Draw 5 percent of the daily production rate of the unit from a full storage tank, collect a sample, then allow the tank to refill.
- Allow to sit with no operation and no water withdrawn over a 48-hour stagnation, then completely empty the tank and collect a sample, allowing the tank to refill afterward.
Solutions for Private Well Owners Impacted by Arsenic
Due to the treatment differences between As (III) and As (V) and the potential need for chemical oxidation as pretreatment, arsenic treatment of groundwater can be confusing to end-users, especially homeowners. Recognizing this, NSF/ANSI 53 and NSF/ANSI 58 require specific information, including an arsenic fact sheet and a performance data sheet, to be included in product literature for certified systems with an arsenic-reduction claim. These documents must include the following information:
The forms of arsenic present in water and an explanation of their health effects.
The procedure for determining whether the user’s source water contains pentavalent arsenic and whether the system is effectively removing arsenic following installation.
The specific arsenic removal claim, including influent concentration and treatment capacity for an active media system.
The water-quality conditions under which arsenic-
removal performance may be limited.
The identification of the arsenic-removal component of the system, frequency of replacement, and source of replacement component(s).
This information provides a useful and readily available guide for end-users to understand how their treatment system functions and what steps must be taken to maintain it.
While many rural homeowners face the challenge of well water contaminated with arsenic, POU/POE devices can provide a solution to help protect their health. Products certified to NSF/ANSI 53 or NSF/ANSI 58 for arsenic reduction allow end-users to feel confident that a system can reduce arsenic to below the MCL over its lifetime. Certified products contain information on arsenic contamination and how the treatment system works, and instructions on properly maintaining the system so it will continue to provide effectively treated drinking water.
Arsenic contamination in drinking water is a persistent and global concern that we cannot afford to overlook. Despite the shift in public focus to other water-related issues, arsenic remains a potent threat, exacerbated by droughts induced by climate change. With millions of people worldwide at risk from elevated arsenic levels, the water industry must remain committed to addressing this critical issue. Industry solutions such as POU and POE water-treatment systems, certified under NSF/ANSI 53 and NSF/ANSI 58 standards, offer hope for mitigating this threat and ensuring safe drinking water.
References
- World Health Organization. “Arsenic.” Fact sheets. December 7, 2022. www.who.int/news-room/fact-sheets/detail/arsenic
- Melissa Bailey, “Once ‘Paradise,’ Parched Colorado Valley Grapples with Arsenic in Water,” NPR, May 22, 2023, www.npr.org/sections/health-shots/2023/05/22/1177346122/risky-arsenic-levels-scarce-water-colorado-san-luis-valley
- Juan Carlos García-Prieto et al. “Impact of Drought on the Ecological and Chemical Status of Surface Water and on the Content of Arsenic and Fluoride Pollutants of Groundwater in the Province of Salamanca (Western Spain),” Chemistry and Ecology 28, no. 6 (December 2012): 545-60.
- J. Appleyard et al. “Arsenic-Rich Groundwater in an Urban Area Experiencing Drought and Increasing Population Density, Perth, Australia,” Applied Geochemistry 21, no. 1 (January 2006): 83-97.
- Melissa A Lombard et al. “Assessing the Impact of Drought on Arsenic Exposure from Private Domestic Wells in the Conterminous United States,” Environmental Science & Technology 55, no. 3 (February 2021): 1,822-31.
About the author Mike Blumenstein is the senior technical manager in NSF’s Residential Water Certification Program. He holds a bachelor’s degree in microbiology and a master’s degree in environmental health, both from the University of Michigan. He can be reached at (800) NSF-MARK or by email at [email protected].
About the company NSF is an independent, global services organization dedicated to improving human and planet health by facilitating standards development and providing world-class testing, inspection, certification, advisory services, and digital solutions to the food, water, health sciences, and consumer goods industries. NSF operates in 180 countries and is a World Health Organization Collaborating Centre on Food Safety, Water Quality, and Medical Device Safety.