By Rick Andrew

Arsenic is a naturally occurring metal that’s present in many residential water sources. Many consumers have no worries about arsenic as surveys of U.S. drinking water indicate that about 80 percent of water supplies have less than 2 micrograms per liter (µg/L) of arsenic. There are some consumers, however, who have significant concerns, as 2 percent of water supplies exceed 20 µg/L of arsenic. These concerns arise from documented potential adverse effects of arsenic on human health. In fact, inorganic arsenic has been utilized as an effective poison since ancient times. Remember the old Cary Grant movie, “Arsenic and Old Lace?” High oral doses of over 60,000 µg/L can cause death due to acute health effects. Lower doses of inorganic arsenic (about 300 to 30,000 µg/L) can be a problem, too, as they can cause stomach and intestinal irritation including stomachache, nausea, vomiting and diarrhea. Other effects of oral consumption of inorganic arsenic may include decreased production of red and white blood cells. This condition can lead to abnormal heart rhythm, blood-vessel damage, fatigue and impaired nerve function.

Aside from these effects of arsenic consumption, probably the most prevalent effect of long-term oral exposure to inorganic arsenic is skin damage. The skin may develop dark spots and discoloration, and corn or wart-like areas may appear on hands, feet and other areas. These corn or wart-like areas have the potential to develop into skin cancer. As if that isn’t bad enough, oral ingestion of arsenic may also increase the risk of lung, liver, bladder, prostate or kidney cancer. Because of these links to cancer, the U.S. Department of Health and Human Services (HHS) has determined that inorganic arsenic is a known carcinogen. The International Agency for Research on Cancer (IARC) has determined that inorganic arsenic is a human carcinogen, and the U.S. Environmental Protection Agency (USEPA) and the National Toxicology Program (NTP) have classified inorganic arsenic as a known human carcinogen.

Because of these numerous potential health effects of oral consumption of inorganic arsenic, on Jan. 22, 2001, the USEPA adopted a new standard of 10 µg/L as a maximum contaminant level (MCL) for arsenic to begin in October 2001. The previous MCL was 50 µg/L. Public water systems must comply with the 10 µg/L standard by Jan. 23, 2006.

The result of the new MCL for arsenic is that many communities must improve their water treatment capabilities for arsenic reduction to be in compliance with this federal requirement. Many individual well owners are also suddenly faced with the fact that their well water doesn’t meet the new MCL for arsenic. Not only are these consumers confronting a health concern over the levels of arsenic in their well water, but they also face difficulties in selling their homes if the water isn’t safe to drink.

Small communities and individual well owners alike are searching for proven point-of-use (POU) technologies and products to help them treat unacceptable levels of arsenic. This need for proven arsenic reduction performance in POU water treatment devices is supported by a number of ANSI/NSF Drinking Water Treatment Unit (DWTU) standards that address arsenic reduction methods, claims and requirements.

Different standards for different technologies
Several different residential water treatment technologies have the potential to significantly reduce the concentration of arsenic in potable water. Distillation, reverse osmosis (RO), and media-based filtration technologies can all be used. These technologies provide distinctly different options for treating arsenic. Distillation uses the fact that water boils at a significantly lower temperature than arsenic and has a lower vapor pressure. RO uses pressure and ionic charge and size to separate water from dissolved ions such as inorganic arsenic. Media filtration uses various chemical interaction forces to cause target contaminants to adsorb within the media instead of remaining dissolved in the water.

Because of these differences in treatment technologies, each has a different standard for arsenic reduction (see Table 1). These different standards involve different test methods for determining the arsenic reduction capability of the test systems, such that the method is appropriate for the technology.

Due to its atomic structure and position in the periodic table, arsenic occurs naturally in two oxidation states: trivalent arsenic—arsenic III, As (III) or arsenite—and pentavalent arsenic—arsenic V, As (V) or arsenate. In natural groundwater, arsenic may exist as trivalent arsenic, pentavalent arsenic or a combination. Trivalent arsenic is considered more harmful than pentavalent arsenic and is generally more difficult to treat.

Standard 62: Drinking Water Distillation Systems
Based on a study conducted by NSF International in 1991, total dissolved solids (TDS) may be used as a surrogate for reduction of a number of contaminants, including arsenic, when testing distillation systems. The TDS consists of 1,000 milligrams per liter (mg/L) of sodium chloride (NaCl) in chlorine-free deionized water. Two test systems are challenged repeatedly, typically over a period of a week, with multiple samples of influent water and product water collected and analyzed. Each sample point must result in a reduction in TDS of at least 99 percent for test systems to meet the requirement and make the TDS reduction and arsenic reduction claim.

Alternatively, two test systems may be tested directly with a challenge of 0.30 mg/L trivalent arsenic (from sodium arsenite) in water from a chlorinated public water supply containing 200-500 mg/L TDS. Because trivalent arsenic in water is typically more difficult to treat, effective reduction of both trivalent and pentavalent arsenic can be verified by challenging with trivalent arsenic. The same repeated challenge schedule as for the TDS reduction test is used. The systems must reduce the arsenic such that the arithmetic mean of all product sample results and 90 percent of individual product water sample results are less than or equal to the MCL of 0.010 mg/L.

Standard 58: RO Drinking Water Treatment Systems
RO water treatment systems aren’t very effective at reducing levels of trivalent arsenic from water; however, RO systems can be highly effective at reducing pentavalent arsenic. This doesn’t mean RO cannot be used to treat trivalent arsenic. In fact, a free chlorine residual will oxidize trivalent arsenic to pentavalent arsenic with minimal contact time. Ozone and potassium permanganate can also be effective oxidizing agents for trivalent arsenic. The oxidized pentavalent arsenic can then be effectively treated by the RO system. For these reasons, Standard 58 doesn’t have a test method or a claim for trivalent arsenic reduction. The standard instead requires literature be provided to consumers advising them of the differences between trivalent and pentavalent arsenic, and the need for effective oxidation of any trivalent arsenic that may be present in their water source prior to RO treatment.

For testing, sodium chloride is added to chlorine-free deionized water to achieve a concentration of 750 mg/L. Pentavalent arsenic, in the form of sodium arsenate, heptahydrate, is added to make up challenge water of 300 µg/L. Two test systems are operated for seven days, and multiple samples of influent and effluent are collected throughout to evaluate system performance. The test systems must reduce the arsenic such that the arithmetic mean of all product water sample results and 90 percent of the individual product water samples are less than or equal to the MCL for arsenic of 10 µg/L.

Alternatively, a 50 µg/L challenge of pentavalent arsenic may be used instead of the 300 µg/L. The philosophy for the 50 µg/L challenge level is that consumers of water from supplies previously in compliance with the 50 µg/L arsenic MCL would be able to choose systems demonstrated to treat water at that level or below to the new requirement of 10 µg/L.

Standard 53: Drinking Water Treatment Units-Health Effects
Media filters are evaluated under this standard. Because specific media may be designed for specific applications, it’s recognized some media filters may be effective at reducing both trivalent and pentavalent arsenic, and some may be effective at reducing only one or the other. Separate test methods are necessary to determine effectiveness of the media system for reducing trivalent arsenic and pentavalent arsenic in potable water.

Last year, a test method for evaluating media filters for pentavalent arsenic reduction was incorporated into Standard 53. This method was the result of a great deal of research and testing by a dedicated task group of water treatment manufacturers, professionals from the regulatory community, and users of treated water. It was a complicated process because media adsorption of arsenic may be affected by a number of water chemistry parameters. The group had to devise a test method to take into account many complex water chemistry effects and incorporate specific confounding ions into the test water to ensure any systems tested for conformance using the method would perform well in the vast majority of real-world applications.

As with other metals tested under this standard, the method requires arsenic be tested at two different water pH levels, 6.5 and 8.5. This requires two separate tests, one at each pH level. Metals have different solubility characteristics under acidic and basic conditions, and testing at both pH levels ensures the media will be an effective adsorbent across a range of typical pH values. Because of the sensitivity of the pentavalent arsenic media adsorption process, the arsenic challenge water must be built up from RO-treated, deionized water. Specific reagents are added to achieve very specific water chemistry (see Table 2).

Media have limited capacity to adsorb contaminants. Accordingly, the contaminant reduction testing under the standard is based on the manufacturer’s rated capacity for the system. If the system has a performance indicator that informs the user when system capacity has been reached and it’s time to change the replacement element, testing must be performed to 120 percent of rated capacity. Otherwise, testing is conducted to 200 percent of capacity. Multiple samples of influent and effluent are collected throughout the test. Two test systems must reduce the 300 µg/L pentavalent arsenic challenge (added as sodium arsenate, heptahydrate) to below the 10 µg/L MCL at every sample point for both systems at both pH levels of testing.

As with Standard 58 for RO systems, the 50 µg/L pentavalent arsenic challenge and claim are an option under Standard 53. Earlier this year, the first media filtration systems conforming to these requirements for pentavalent arsenic reduction were certified by NSF. It’s anticipated more manufacturers of media filtration products will develop special media blends designed for arsenic reduction and more of these products will be certified.

Coming soon: As (III) reduction
The same task group that developed and validated the pentavalent arsenic reduction method now included in Standard 53 is continuing its work, now focusing on trivalent arsenic reduction. This task is even more daunting than the pentavalent arsenic reduction method. In addition to all of the complex water chemistry issues surrounding the interaction of arsenic with media, trivalent arsenic in solution must be kept away from dissolved oxygen or it will oxidize to pentavalent arsenic. For this reason, the proposed method for testing media filters for trivalent arsenic reduction involves starting with RO-treated, deionized water, building up the required challenge water chemistry through the addition of precise amounts of specific reagents (see Table 3). The challenge water is then constantly purged with nitrogen to bring the dissolved oxygen level below 0.5 mg/L throughout the course of the test.

As with Standard 58 for RO systems and pentavalent arsenic under Standard 53, the 50-µg/L trivalent arsenic challenge and claim are proposed to be an option under Standard 53. A round-robin study involving testing identically manufactured media filters according to this proposed protocol was recently conducted by a number of laboratories. The task group is currently evaluating the results of this study. The group must determine if results support the method being a robust, repeatable and consistent one that’s appropriate for adoption into the standard, or if modifications and improvements are required.

It’s anticipated the method will be fully developed and balloted through the DWTU Joint Committee into the standard by the end of this year. Once this has occurred, it will be possible to establish conformance to Standard 53 for trivalent arsenic reduction. Once the standard contains methods for conformance of systems to the claims of pentavalent arsenic reduction and trivalent arsenic reduction, a total arsenic reduction claim will be possible under this standard.

Small communities and individual well owners need guidance in their quest for proven POU technologies and products to help them treat arsenic. This guidance is provided by Standards 62, 58 and 53. By searching out products that conform to one of these standards for arsenic reduction, consumers can select from a variety of technologies and know each product has been tested appropriately to assure proven performance. Whether their situation and preferences steer them toward a distiller, an RO system or a media filtration system, consumers can know they have choices in these technologies and products that will all help them to accomplish their goal—safe drinking water untainted by unacceptable levels of arsenic.

About the author
Rick Andrew is the technical manager of the Drinking Water Treatment Units Program at NSF International. He has been with NSF for over four years, working with certification of residential drinking water products. His previous experience was in the area of analytical and environmental chemistry consulting. Andrew has a bachelor’s degree in chemistry and a master’s degree in business administration from the University of Michigan. He can be reached at (800) 673-6275 or email:


Comments are closed.