Arsenic Removal Arsenal
An overview of treatment options for successfully cleaning up contaminated drinking water supplies in compliance with the new stricter arsenic standard
The U.S. Environmental Protection Agency (EPA) has historically regulated arsenic at 50 parts per billion (ppb), but the agency will lower the maximum contaminant level (MCL) for arsenic to 10 ppb by 2006. Some states are even setting their own limits well below this level. EPA estimates one in 20 (or roughly 4,000) of the 74,000 U.S. systems that must comply with the new standard will need to install additional treatment processes or adopt different measures to meet the requirements. Nearly 97 percent of those are small systems serving communities of fewer than 10,000 people.
According to EPA, the new arsenic standard will increase the average American's water bill by $32 annually. Those living in smaller communities may see a substantially greater hike, ranging from $58 to $327 per year. EPA predicts the new regulation will cost local communities approximately $200 million annually.
The new arsenic standard will not protect private well owners, and EPA is not sure how many of the 2.5 million private well users consume water with arsenic that exceeds the new MCL. Installing a well treatment system that serves 500 people is approximated to cost $160,000 and $27,000 in capital and annual operating costs, respectively.
Some public works are adopting treatment avoidance measures as the most economical way to comply with the upcoming arsenic regulation. Such practices include abandoning existing water supplies with high arsenic concentrations in favor of new water supplies that have lower arsenic concentrations. Another solution includes blocking the flow of water from zone(s) within an aquifer that have higher arsenic concentrations, reducing the overall arsenic concentration produced from the well to, in some cases, below acceptable levels. Where these treatment avoidance strategies work, the capital cost and operating costs are lower than the cost of removal treatment. However, implementing treatment avoidance strategies is not a viable option for all public works.
Luckily, several treatment options for arsenic removal exist. Pairing the appropriate removal method with an existing system depends on several factors including:
- The nature of the arsenic,
- Arsenic concentration,
- Presence and concentration of additional contaminants,
- Existing plant equipment and processes,
- Site conditions,
- Water source,
- Flow rate, and
- Residuals disposal means.
Arsenic Chemistry and Removal Mechanisms
Arsenic exists in two valence states: arsenite (As III) and arsenate (As V). Arsenite has a neutral charge, so it is not readily removed by most of the treatment processes. Arsenate, on the other hand, has a negative charge and is more thoroughly removed by the treatment processes described in this article.
Waters with a high arsenite concentration usually need to be pre-oxidized. This converts the hard-to-treat arsenite into the more easily removable arsenate. Chlorine, a preferred pre-oxidant, also provides disinfection capacity. As the majority of arsenic-bearing water supplies are groundwaters with low influent dissolved organic carbon levels, chlorinated organic disinfection byproducts are not of significant concern. Potassium permanganate is another oxidant that may be used to oxidize other present contaminants such as manganese.
For water supplies with naturally occurring iron, the pre-oxidation process using chlorine or potassium permanganate will oxidize soluble iron as well as arsenite. The resulting iron floc provides a suitable particle for co-precipitation of arsenic. If the concentration of naturally occurring iron is insufficient for adequate arsenic removal, additional coagulant can be added after the oxidation process.
Coagulants such as ferric or aluminum salts are fed into the water to co-precipitate arsenic so that it can be removed through settling and/or filtration. Ferric coagulants tend to be more effective at arsenic removal and, more importantly, are less pH-sensitive than alum. Alum and aluminum salt coagulants can be effective in binding arsenic, but the process is highly pH-sensitive, with optimum pH often found in the 6.0 to 6.5 range. If aluminum salt coagulant is used, acid dosing may be needed to achieve this optimum pH range. Ferric coagulants are effective over a much wider pH range and are therefore most commonly used for arsenic co-precipitation.
Several studies investigating the stability of sludges derived from ferric coagulants have determined that the sludges are stable, based on the toxic characteristics leaching procedure (TCLP) protocol. The results, which may need to be duplicated on a site-by-site basis, indicate that ferric sludges from arsenic removal processes would be classified as nonhazardous waste under the Resource Conservation Recovery Act and therefore can be disposed of in sanitary landfills. Public works are encouraged to contact their local regulatory approval agency for guidelines on disposing of arsenic residuals.
Iron-based Adsorption Media
Stringent U.S. and European arsenic regulations have prompted manufacturers to develop and promote more iron-based adsorption media. These specialized media operate similarly to granular activated carbon contactors. The media is placed in a pressurized treatment vessel in a fixed bed adsorber. Raw water passes through the media that adsorbs the arsenic, which means that the arsenic adheres to the surface of the media. Backwash is performed infrequently to prevent compaction and to remove any particulate that may be present in the supply. This process requires minimal operator attention, compared to other arsenic removal processes.
Variants of iron-based media include granular ferric hydroxide, granular ferric oxide, iron hydroxide-coated sand, metallic iron (referred to as zero valent iron), sulfur/iron mixtures (referred to as sulfur-modified iron) and many others. Several of the ferric-based materials can sufficiently adsorb arsenite, making pre-oxidation unnecessary. Many of these materials are at the field-trial stage, but others are already being used in full-scale applications throughout Europe and the United States.
Sorption processes with single-use media provide a simple treatment solution for small installations and wellhead applications that have low or moderate arsenic concentrations with no treatment process in place. Water supplies that have iron or manganese should consider using alternative treatment processes that specifically remove these contaminants as well as arsenic.
The primary limitations -- effective media lifetime and consequent media replacement cost -- affect applications at higher flow rates and at higher concentrations. Most of the residuals in this process are solid waste, which alleviates concerns about disposing of liquid waste. And, as the spent ferric sorbents have passed TCLP testing and therefore are classified as nonhazardous waste, spent media may be sent to sanitary landfills.
Empty bed contact time, loading rate and media bed lifetime capacity are key design parameters for iron-based sorbents. Bed lifetime depends on arsenic concentration, competing ion concentrations, arsenic species and pH, and is specific to a particular type of media. Operating conditions may also influence the observed lifetime since mixing the bed during backwash can cause it to behave as a stirred tank batch adsorber rather than as a fixed bed adsorber. Most iron-based media have capacity to remove other contaminants such as selenium, vanadium, antimony and others.
A broad series of processes are included in the filtration category for arsenic removal. These processes generally oxidize arsenite (if required), co-precipitate the arsenic with iron or aluminum salts and filter the resulting solids. A pressure filter is the most common filtration device for groundwater supplies, as it eliminates the need to re-pump the water after treatment. Gravity filtration may be more economical for larger public works or for those with a groundwater storage tank. The filtration process simply separates the formed solids from the water. A wide range of products is available to best meet specific project requirements.
For project planning purposes, the filtration processes described below can readily be bench-and pilot-scale tested to determine removal efficiency and overall process design.
The filtration processes require regular backwashing that generates a solids-laden liquid waste. In cases where disposal of backwash waste to sewer is impractical, the management of the backwash waste stream proves a major disadvantage. This process is best applied in areas where it is easy to dispose of liquid waste.
Commonly used to remove iron and manganese, oxidation/filtration also effectively removes arsenic. The process minimizes chemical costs since naturally occurring iron is used for co-precipitation. At moderate levels of arsenic contamination, even a relatively low iron concentration 0.3 milligrams (mg) or higher adsorbs enough arsenic to adequately remove the contaminant. However, to be effective, the iron content of the water must be stable or at least remain above the required threshold for arsenic removal.
The backwash waste produced contains highly stable iron-arsenic sludge that can typically be disposed of in a landfill.
This process is similar to the oxidation/filtration removal process except the coagulant is added after oxidation to increase arsenic removal efficiency. Laboratory, pilot-plant tests and full-scale operating plants have shown coagulation and filtration to be an effective treatment process for arsenate, iron and manganese removal. Pre-oxidation is necessary for arsenite-laden water supplies.
Naturally occurring iron helps remove arsenic and, as a result, impacts the amount of coagulant used. Similarly, water conditions may affect the process reaction time, and additional detention prior to filtration may be required. Ferric salt such as ferric chloride or ferric sulfate is the most common coagulant selected.
As with oxidation/filtration, the backwash waste produced contains highly stable iron-arsenic sludge that can typically be disposed of as nonhazardous waste in a sanitary landfill.
This coagulation/filtration treatment process uses a microfiltration membrane system instead of a granular media filter. Ferric coagulant is typically used. Pre-oxidation may be required, depending on the arsenic species. The process also removes waterborne pathogens such as cryptosporidium and giardia.
Oxidant-tolerant microfiltration membranes are necessary because of the pre-oxidation requirement. The key economic driver in using a microfiltration filter is design flux, which is the flow rate per unit membrane area. The flux dictates the size of the equipment, and is sensitive to the ferric solids load. Depending on the water chemistry, it may be worthwhile to reduce the ferric dose by adjusting the pH, thus optimizing the membrane area requirement.
Coagulant-assisted microfiltration systems require periodic backwash and produce sludge as waste similar to granular media filtration.
With high influent arsenic concentration, the coagulant dose necessary for adequate arsenic removal results in impracticably short filter run times. Elevated levels of other contaminants such as iron and manganese also contribute to increasing solids load. Adding sedimentation before filtration reduces solids load, which extends filter operation and minimizes backwash events.
Capable of handling a higher influent content of contaminants and a wider range of water conditions, the coagulation/sedimentation/filtration process is appropriate for existing surface water treatment plants requiring arsenic removal or for large groundwater systems opting for centralized treatment.
Robinson and Leible have reported excellent arsenic removal in such systems, specifically citing a Microfloc® Tricon installation that treats 9 million gallons per day (mgd) of groundwater by pre-oxidation, coagulation, adsorption clarification and filtration (1999). They report reducing influent arsenic of 150 ppb to 6 ppb in the treated water.
Despite increasing the process complexity, the sedimentation step improves the overall performance. Additionally, this system increases detention time during which reactions occur.
Both the sedimentation and filtration sections generate sludge waste streams sufficiently stable for landfill disposal.
Under appropriate conditions, activated aluminais an effective arsenic sorbent that can be operated in either a single-use or regenerated mode. Sorption capacity is pH-dependent, with slightly acidic conditions generally favored. Many anionic species such as selenium, fluoride, chloride and sulfate compete with arsenic for adsorption sites and, if present at high levels, will reduce the economic effectiveness of the process. Pre-oxidation and pH adjustment are normally required to maximize removal and sorption capacity. While activated alumina can be regenerated, significant loss of capacity (of 10 percent or more) can occur on each cycle, leading to future replacement of the media.
This process may be more appropriate for use in larger water treatment applications, due to the incomplete nature of regeneration and the problems of handling and disposing of chemical streams. Smaller sites may prefer to treat activated alumina as a once-through, disposable sorbent. Waste produced by the activated alumina process will consist of concentrated arsenic in solution.
Pre-oxidation of arsenite to arsenate followed by the exchange of anions (ions with negative charges) is normally required to maximize arsenic removal. Anion exchange can effectively remove arsenic, generally using strong base anion exchange resins in the chloride form. Sulfate and other anions compete with arsenic and can greatly reduce the operating cycle if present in elevated concentrations.
Pilot-testing indicates that the brine regeneration solution can be reused as many as 20 times with no impact on arsenic removal, provided that some salt is added to the solution to provide adequate chloride levels for regeneration. Regenerate reuse reduces the amount of waste for disposal, but increases the ultimate arsenic concentration of the spent brine.
Spent brine disposal options are critical in assessing the viability of ion exchange for specific treatment situations. The brine can be treated with ferric coagulant to remove the arsenic from the liquid waste. Smaller sites may not prefer this treatment process, due to regeneration and the handling of chemical waste streams.
Reverse Osmosis and Nanofiltration
These high-pressure membrane systems provide effective removal of arsenic and many other dissolved species such as calcium, magnesium, nitrates and sulfates. These technologies are ideal for point-of-use and point-of-entry applications at low flow rates, particularly when arsenic is just one of several water quality parameters requiring treatment.
For sites with several water quality issues to address, particularly if these are related to dissolved solids, reverse osmosis or nanofiltration provide complete treatment in a single process step. If a high concentration of undissolved solids is present in the supply, pretreatment for removal may be required.
Larger flows lose 15 percent to 30 percent of feed flow as reject and require 50 pounds per square inch (psi) to 150 psi operating pressure even for modern high-efficiency, low-pressure reverse osmosis operating on low total dissolved solids feedwater. While the reject flow is high, the concentration of arsenic in the liquid waste is lower than in many of the other processes. This is an important consideration when disposing of the liquid waste to a sanitary sewer, as some wastewater treatment plants prefer an increased flow with lower concentration versus small batch flows with higher concentrations.
Waste produced by the reverse osmosis and nanofiltration systems will consist of concentrated arsenic in solution.
Lime is added to water to raise the alkalinity and, when combined with media filtration, effectively removes arsenic through adsorption-co-precipitation. Able to handle a wide range of contaminants and water conditions, lime softening applies to groundwater and surface water supplies with high hardness, iron, manganese and arsenic. Existing plants that use lime softening should be able to simply increase pH in the reaction basin or make other process modifications to meet arsenic removal standards.
The lime softening process is unlikely to be chosen solely for arsenic removal, due to operational complexities and the cost of equipment and operation. However, the softening stage decreases solids load, which increases the filter run time between backwashes and provides the added benefit of reducing hardness. The softening basins and filter backwash waste produce sludge waste.
Arsenic is a growing problem in communities great and small, urban and rural. Newly discovered contaminated sites appear in the news every day.
The good news is there are many arsenic treatment removal options available. Industry professionals are becoming more educated about the issue and are sharing that knowledge with community leaders and public works.
Contaminated sites should be evaluated individually to determine which solution best meets their needs.
This article originally appeared in the 02/01/2004 issue of Environmental Protection.