An Overview of Arsenic Treatment Concerns and Applications for Small and Very Small Systems
By Larry Henke
Arsenic continues to be one of the most extensive environmental poisonous chemical elements throughout the world. It is found at high concentrations in Taiwan, China, Argentina, Chili, Mexico, parts of Europe, in the United States and most notably in West Bengal, India and neighboring Bangladesh.
Since the mid 1980s, millions of people have been exposed to dangerous levels in their drinking water, thus spurring a high level of activity in all public health efforts for the treatment and prevention of arsenic-related illness. In Bangladesh alone, in excess of 120 million people have been exposed to levels greater than 0.010 ppm, the WHO (and US EPA) MCL, posing a potential health emergency of unprecedented scale.
Arsenic falls into the periodic table of the elements in Group 15 (V) along with nitrogen and phosphorous. It is considered a metalloid and as such has both metallic and nonmetallic properties. It is a brittle, grey metal in its pure form, but in nature is most often found with other metals, such as iron, copper, silver and nickel and in combination with oxygen and sulfur.
Its atomic number is 33 and its atomic weight is 74.9, placing it as heavier than iron, nickel and manganese but lighter than silver, lead or gold. It has four common redox states, -3, 0, +3 and +5, but the most inorganic forms are the +3 and +5 state when in combination with oxygen as an arsenite or an arsenate.
Arsenic is used in the production of pesticides, as protective coating for wood and in industries such as electronics, glass and paper manufacture.
Studies on the geological mechanisms that mobilize arsenic from the earth and into the water continue to be investigated. Although there can be inorganic pathways, such as in the instance of highly acidic mine tailing water, bacteria have been found that can reduce inorganic arsenic (As5) to the more soluble form (As3). Thus oxidized arsenic is mobilized from stable minerals into the groundwater.
Arsenic is the 20th most common element in the Earth’s crust, 14th in the sea and 12th in the human body. Arsenic is found naturally in over 150 minerals, one of which is arsenopyrite, which includes both iron and sulfur; another is realgar (AsS).
Arsenic as poison
The effects of arsenic on the human body have been known for over four millennia; its legacy as a toxic substance, both intentionally and accidentally, has been documented in fiction and in history. Arsenic has a deserved reputation for mischief, but it also has a history as a medicine.
Hippocrates used arsenic salts to ‘cure’ many afflictions, including cancer, ulcers, rheumatism, TB, asthma and VD, lending a curious irony to his oath: “First, do no harm.” Arsenic salts and tonics such as Fowler’s, De Valagin’s and Donovan’s solutions were invented in the 1700s to cure a spectrum of maladies, the use of which continued well into the 20th century. Even today, arsenic has a use in treating leukemia and colorectal cancer.
As an acute poison, arsenic has an LD50 (the median lethal dose, or lethal 50 percent of the time), which equals 763 mg/kg by ingestion of pure substance. The other forms of arsenic have greater or lesser toxicities; e.g., an LD50 of 15 mg/kg for arsenic trioxide, 1.5 mg/kg for arsine gas and higher LD50 doses for various organic forms.
Arsenic readily combines with organic molecules to form arsenosugars or methylated forms of arsenic (mono- or di-methyl-ated arsenic: MMA or DMA). When ingested, the human body produces a biochemical reaction with inorganic arsenic to form dimethylarsenic, which can then be excreted from the body.
Higher concentrations of arsenic overwhelm the body’s rejection mechanisms and arsenic can then interfere with cellular respiration by replacing phosphorous in adenosine triphosphate (ATP). ATP is an essential organic chemical that controls the energy storage and use in every living cell, plant or animal.
When arsenic replaces the phosphorous, the mechanism of cellular energy control is disrupted and the cell dies, resulting in organ failure and possibly death to the organism. This is only one example of the effect of arsenic in living things and researchers are studying many other arsenic pathways that can occur in living organisms.
Acute exposures result in quickly identifiable symptoms: gastric and pulmonary distress, vomiting and bloody diarrhea; chronic effects are not as rapidly pronounced, but are much more insidious. The early effects of long-time exposure may not be noticed. The person may be asymptomatic for up to five years.
Chronic level effects are determined by age, gender and weight among other factors. After five or so years, the individual may have darkening of skin or dark spots on the hands and feet. After 10 years of continued exposure at low levels, arsenic might produce internal disease such as diabetes or attack the cellular structure of the neurological system; the liver, kidneys or bladder and other organs may be destroyed or become cancerous.
Routes of exposure
Inadvertent routes of exposure include environmental exposures to pesticides and herbicides. Until the 1940s these major pesticides and herbicides contained arsenic and many golf courses still use arsenic-based chemicals.
Another common path is the ash, smoke or sawdust of CCA-treated wood. While CCA (copper chromated arsenic) ‘pressure-treated’ wood was voluntarily limited to non-residential uses after 2003, many old structures have timbers treated with the chemical.
The production of CCA wood continues for industrial applications. Burning of this wood is unlawful in all states, but people continue to be exposed through handling or sawing. Although children can be exposed by playing on or around play structures containing CCA treated wood, testing shows that there is limited concern; most exposure can be limited by hand washing.
Exposure through food is often a worry; arsenic has been found in a large variety of foodstuffs, from shellfish and other seafood to rice and other grains, as well as lettuce, radishes, mushrooms, wine and even licorice. In many foods, however, the level of arsenic is very low and the intake only occasional, so the overall exposure is relatively small.
It is most often present in food as organic arsenic and humans are better able to process and eliminate organic forms. Still, foods such as rice cooked in arsenic-laden water can add to a diet already high in arsenic through drinking water; it is drinking water that is the path of greatest concern.
Rapid medical attention is essential if acute arsenic poisoning is suspected. Although there are several available treatments, stabilization of body functions is required.
Often the patient must have gastric lavage, flushing the toxic substance from the digestive tract. In addition, kidney and neurological functions need to be monitored and supported. Hemodialysis, chelation therapy and the use of drugs such as dimercoprol have been used successfully in some instances.
Chronic intoxication can be more difficult to treat depending on the length of exposure. The first and probably most important step in helping those who have been chronically exposed is to seek alternate sources of water. Failing that, it is possible to eliminate the arsenic from the drinking and cooking water.
Once the poisoning has progressed to the stage where the effects are noticeable, the long-term health outcome is grim. When an entire population is affected, the long-term public health consequences are catastrophic.
Detection and measurement
Public health officials recognized early the need for accurate and rapid water analysis, especially in remote areas. Analytical manufacturers responded by developing improved field-testing methods and equipment, resulting in several methods that, while not equivalent to laboratory methods, are quick and relatively reliable. There remains a need to test millions of tube wells (although it is uncertain how many wells there are, it is estimated there are more than 10 million in Bangladesh alone).
A commonly used field test strip can be employed to screen wells for arsenic, with more accurate follow-up conducted when indicated. Several tests on the market rely on converting any arsenic present in a sample of water to arsine gas, which in turn changes the color on a small indicator patch of mercuric bromide. This patch can be read visually and compared to a set of standards or can be read by an electronic instrument for greater accuracy.
The test is simple, requires little training and takes slightly more than one-half hour for the chemistry to occur. Rapid mobile screening can be conducted by field staff; however, care must be exercised to avoid either false positives or negatives.
Studies conducted in Asia have seen error rates from two to five percent, which is not acceptable for public health. False positives can alarm the community, especially when no other source of water is readily available, or conversely when wells with arsenic are missed. This emphasizes the need for field training and careful supervision of field staff whether professional or volunteer.
The ‘gold standard’ for arsenic detection and quantification remains the AAS or MS-ICP where detection limits and the margin of error can be less than 1.0 µg/L. In the case of US EPA Method 200.8, ICP-MS detection of 0.4 µg/L, with Method 1632, AAS limits of two nanograms per liter are possible.
These tests are conducted on laboratory instruments that can be expensive to obtain and operate, require trained operators and take time for testing and interpretation. There remains a need for more accurate field-testing methods.
Water treatment is a primary focus of public health. Recognized treatment technologies for arsenic include coagulation/precipitation, iron-arsenic co-precipitation and filtration, containment by adsorptive media, ion exchange and membranes. The efforts to find a simple, foolproof, inexpensive, effective and reliable method have been explored worldwide.
A $1,000,000 prize was offered by the Grainger Foundation and won by a professor at George Mason University who devised a batch process system that used materials readily available to rural India and Bangladesh. Abul Hassam, originally from Bangladesh, arranged three pitchers or ‘Kolshi’ containing local river sand, milled iron turnings (zero-valent iron) and brick chips.
This system is now being distributed throughout the region. Figures 1 and 2 illustrate various other treatment processes commonly used.
Coagulation/precipitation through the addition of aluminum or iron salts or by lime softening, are methods generally restricted to large scale, municipal treatment plants. They each require a high level of operator training, continued chemical addition that may vary as the water composition varies and have waste product implications.
Of the coagulant chemicals, those that are iron based seem to be more effective in reducing arsenic to levels below the US EPA MCL of 10 µg/L. These technologies are not as appropriate for small systems because of the level of operator involvement.
In developing nations or in small communities, attention has focused on adsorptive media. Some of the media are unusual materials restricted to a specific locality such as dried hyacinth root, jute, laterite clay, sawdust, orange waste and newspaper pulp.
Fly ash from coal plants has been studied even though fly ash can itself be high in arsenic. Among the media proposed and tried are a variety of activated carbons formed from various sources, ranging from coal to wood to bone and to forms of biomass such as corncobs, fruit pits and seaweed. While activated carbon can reduce levels of particulate arsenic, it is costly and introduces waste disposal problems.
Iron is a central component of many successful adsorptive media. Iron coatings on sands, glauconites, charcoals and other substrates have been employed. Amorphous iron hydroxide, stepwise charcoal and zero-valent iron (such as in the 3-Kolshi system above) and carbon followed by steel wool have all been proposed and tested with varying success.
Many of the more exotic methods of adsorption employing rare or expensive materials or processes may not be practical for arsenic removal from drinking water per se, but may be instrumental in the deeper understanding of arsenic and thus may lead to more novel methods in the future. In some cases, adsorptive media can be used in single-family residences.
Strong base ion exchange resins have also been used successfully in small systems. The usual concerns with competitive ions such as sulfate and nitrates along with fouling substances such as iron and manganese apply. High levels of sulfate or high TDS can limit ion exchange effectiveness.
Oxidation and co-precipitation with iron
When the ground water has natural iron, oxidation and co-precipitation followed by filtration can be a cost-effective means of removing both contaminants. The rule of thumb ratio of iron to arsenic is 20:1. Oxidants include chlorine, ozone and potassium permanganate; oxygen alone while often sufficient for iron alone, will not oxidize arsenic.
Fenton’s reagent (iron and hydrogen peroxide) has been demonstrated as a possible method for small systems. In all cases the use of appropriate filtering media is necessary to follow the oxidation step. The co-precipitation oxidizes iron and arsenic from Fe2 to Fe3 and As3 to As5; they attach to each other and are then filtered from the water stream.
Filter media range from simple filter sands, manganese greensand, iron-coated sands and glauconites and ceramic media. The filters are configured in traditional designs with intermittent backwash based on time, volume or pressure loss.
Membranes are a proven method of arsenic removal and both reverse osmosis and nanofiltration are proven methods. While UF and MF can be used in some cases, membrane performance can be improved by the proper pretreatment of the raw water with coagulation chemistry, prefiltration and softening.
Predicting arsenic remediation
Small systems are generally very cost sensitive. Small communities or small public water supplies often do not have the taxing and bonding authority of large municipal or industrial water users, so design considerations become more challenging.
It is essential that arsenic treatment be designed carefully with built-in monitoring and assurance steps at once mindful of installation and start-up costs, continuing costs for chemicals, operator training and waste disposal. In smaller communities the operator has other, often more immediately pressing duties.
A complete water analysis often consists of determining both arsenic totals and the species (As3 and As5), along with contributing factors such as iron, manganese, hardness, TDS and confounding factors such as sulfate, nitrates and phosphorous. The pH, reduction-oxidation potential (ORP), temperature of the water and alkalinity can affect the selection of appropriate treatment. Figure 3 indicates the pH and Eh ranges for the various species of arsenic and can aid in determining if pH or oxidation levels are sufficient.
Testing can be conducted in two phases: the first to aid in determining an appropriate treatment choice and the second to narrow the choice and ensure arsenic removal. The following are those parameters considered essential, although not all are required for some methods of treatment: alkalinity; aluminum (Al); arsenic (As); arsenite (As[III]); arsenate (As[V]); calcium (Ca); chloride (Cl); fluoride (F); iron (Fe); iron (Fe2); iron (Fe3); magnesium (Mg); manganese (Mn); nitrate (NO3); nitrite (NO2); orthophosphate (PO4); oxidation reduction potential (ORP/Eh); pH; silica (Si); sulfate (SO4); total dissolved solids (TDS) ; total hardness; total organic carbon (TOC) and turbidity.
For example, if ion exchange is selected, then competitive ions of sulfate, nitrate and chloride should be noted. In the case of activated alumina, then aluminum, manganese and iron should be considered.
For membrane processes, any fouling agents such as turbidity, iron, TOC and hardness should be measured. Bench studies to elevate the ORP with chlorine or permanganate or bench work using small columns can be informative.
Laboratory bench studies can be helpful in determining if a selected process is likely to work, but the more certain way is through a pilot study.
Pilot testing should be conducted on site using the same process as that selected. Oxidation method and pretreatment chemistry and media, a small membrane system, or a small ion exchange column can offer insights into the performance.
Jar testing along with other more sophisticated laboratory studies conducted on site and in real time may determine if arsenic levels are reduced to below the MCL. Field arsenic tests should be augmented by certified laboratory testing to ensure compliance.
Pilot studies, bench testing and laboratory testing of water add to the overall cost of an installed treatment system, but the consequences are too grave to skip steps in arsenic treatment design. Unlike iron, manganese and hard water problems, arsenic represents a danger to public and individual health, the consequences of which may not be evident until long after the initial and/or prolonged exposure. The liabilities for non-performance are too severe, both for the designer and more for those obliged to drink the water.
About the author
Larry Henke is an independent water treatment consultant in the Minneapolis, MN area, serving industrial filter and disinfection applications and small non-community and community public drinking water systems. A graduate of the University of Minnesota, he has more than 20 years of experience in the water treatment industry. Henke is a member of the American Water Works Association and the National Ground Water Association, as well as WC&P’s Technical Review Committee. He can be reached at (612) 599-7477 or at firstname.lastname@example.org.