Water Sanitation Health

Managing water in the home

Accelerated health gains from improved water supply

Chemical methods of water treatment

1 Chemical coagulation, flocculation and precipitation

A number of chemical methods are used for water treatment at point-of-use or entry and for community water systems. These methods can be grouped into several main categories with respect to their purpose and the nature of the technology. The main categories to consider here are: (1) chemical pre-treatments by coagulation-flocculation or precipitation prior to sedimentation or filtration, (2) adsorption process, (3) ion exchange processes and (4) chemical disinfection processes. All of these processes can contribute to microbial reductions from water, but the chemical disinfection processes are specifically intended to inactivate pathogens and other microbes in water. Therefore, chemical disinfection processes appropriate for household water treatment in the developing world will be the focus of attention in this section of the report. Other chemical methods for water treatment will be examined for their efficacy in microbial reductions and their applicability to household water treatment.

1.1 Introduction
Chemical precipitation or coagulation and flocculation with various salts of aluminum (e.g., alum), iron, lime and other inorganic or organic chemicals are widely used processes to treat water for the removal of colloidal particles (turbidity) and microbes. Treatment of water by the addition of chemical coagulants and precipitants has been practiced since ancient times, even though the principles and physico-chemical mechanisms may not have been understood. Sanskrit writings refer to the use of vegetable substances, such as the seed contents of Strychnos potatorum and Moringa oleifera, which are still in use today for household water treatment (Gupta and Chaudhuri, 1992). Judeo-Christian, Greek and Roman records document adding "salt", lime, "aluminous earth", pulverized barley, polenta as precipitants to purify water. Although alum and iron salts are the most widely used chemical coagulants for community drinking water treatment, other coagulants have been and are being used to coagulate household water at point of use, including alum potash, crushed almonds or beans and the contents of Moringa and Strychnos seeds. Table 12 lists some the coagulants that have been and are being used for water treatment at the community and household level, their advantages and disadvantages and their costs.

Table 12. Chemical coagulants for water treatment and their advantages, isadv

Coagulant Community/Household Use Advantages Disadvantages Cost* Comments
Alum (aluminum sulfate, etc.), alum potash Yes/rare-moderate Community use common; simple technology Difficult to optimize without training and equipment Moderate? Proper use requires skill
Iron salts (ferric chloride or sulfate) Yes/rare Same as Alum Same as Alum Moderate? Proper use requires skill
Lime (Ca(OH2)), lime+soda ash (Na2CO3), caustic soda (NaOH) Yes/rare-moderate Same as Alum Same as Alum; pH control and neutralization a problem; hazardous chemicals Moderate to high? Softeners; not applicable to many waters
Soluble synthetic organic polymers Yes/no-rare Improve coagulation with alum and iron salts Same as Alum; hard to dose; need training & equipment; hazardous chemicals High Use with other coagulants; limited availability
Natural polymers (carbohydrates) from seeds, nuts, beans, etc. Rare/Yes
(in some developing countries) Effective, available and culturally accepted in some places Source plant required; training and skill required; cultural acceptability; may be toxic Low Traditional use based on historical practices

*Estimated Annual Cost: low is 0.01 per liter (corresponds to about $100, respectively, assuming household use of about 25 liter per day)

Chemical coagulation-flocculation enhances the removal of colloidal particles by destabilizing them, chemically precipitating them and accumulating the precipitated material into larger "floc" particles that can be removed by gravity settling or filtering. Flocculation causes aggregation into even larger floc particles that enhances removal by gravity settling or filtration. Coagulation with aluminum or iron salts results in the formation of insoluble, positively charged aluminum or iron hydroxide (or polymeric aluminum- or iron-hydroxo complexes) that efficiently attracts negatively charged colloidal particles, including microbes. Coagulation-flocculation or precipitation using lime, lime soda ash and caustic soda is used to "soften" water, usually ground water, by removing (precipitating) calcium, magnesium, iron, manganese and other polyvalent, metallic cations that contribute to hardness. However, reductions in microbial contaminants as well as turbidity, and dissolved and colloidal organic matter are also achieved in this process.

1.2 Microbial reductions by coagulation-flocculation
Optimum coagulation to achieve maximum reductions of turbidity and microbes requires careful control of coagulant dose, pH and consideration of the quality of the water being treated, as well as appropriate mixing conditions for optimum flocculation. Lack of attention to these details can result in poor coagulation-flocculation and inefficient removal of particles and microbes. Under optimum conditions, coagulation-flocculation and sedimentation with alum and iron can achieve microbial reductions of >90 to >99% for all classes of waterborne pathogens (Sproul, 1974, Leong, 1982, Payment and Armon, 1989). However, poor microbial reductions occur (<90%) when coagulation-flocculation or precipitation conditions are sub-optimal (Ongerth, 1990). Even greater microbial reductions (>99.99%) can be achieved with lime coagulation-flocculation or precipitation if high pH levels are achieved in the process (pH >11) to cause microbial inactivation as well as physical removal.

1.3 Alum and iron coagulation
Because coagulation-flocculation treatment with alum, iron and other coagulants requires knowledge, skills to optimize treatment conditions, it is generally considered to be beyond the reach of most consumers. Most authorities consider such treatment to be best performed in specialized central facilities by trained personnel. This type of treatment is less likely to be performed reliably at point-of-use for household water treatment. Furthermore, the limited availability and relatively high costs of alum and ferric salts in some places present additional obstacles to widespread implementation of this technology at the household level.

Despite the caveats and limitations, alum coagulation and precipitation to remove turbidity and other visible contaminants from water at the household level has been traditionally practiced for centuries in many parts of the world (Jahn and Dirar, 1979; Gupta and Chaudhuri, 1992). When potash alum was evaluated for household water treatment in a suburban community in Myanmar by adding it to water in traditional storage vessels (160L capacity) at 500 mg/L, fecal coliform contamination was reduced by 90-98% and consumer acceptance of the treated water was high (Oo et al., 1993). The ability of the intervention to reduce diarrheal disease was not reported. In another study, alum potash was added to household water stored in pitchers of families with an index case of cholera and intervention and control (no alum potash) households were visited to 10 successive days to track cases of enteric illness (Khan et al., 1984). Illness among family members was significantly lower (p < 0.05) in intervention households (9.6%) than in control households (17.7%). The authors concluded that household water treatment by adding a pinch of alum potash was effective in reducing cholera transmission during outbreaks and was an appropriate and low cost (1 cent per 20 liters) intervention.

1.4 Seed extract coagulation-flocculation
Coagulation-flocculation with extracts from natural and renewable vegetation has been widely practiced since recorded time, and appears to be an effective and accepted physical-chemical treatment for household water in some parts of the world. In particular, extracts from the seeds of Moringa species, the trees of which are widely present in Africa, the Middle East and the Indian subcontinent, have the potential to be an effective, simple and low-cost coagulant-flocculent of turbid surface water than can be implemented for household water treatment (Jahn and Dirar, 1979; Jahn, 1981; Jahn, 1988; Olsen, 1987). The effectiveness of another traditional seed or nut extract, from the nirmali plant or Strychnos potatorum (also called the clearing nut) to coagulate-flocculate or precipitate microbes and turbidity in water also has been determined (Tripathi et al., 1976; Able et al, 1984). Microbial reductions of about 50% and 95% have been reported for plate count bacteria and turbidity, respectively. Despite the potential usefulness of Moringa oleifera, Strychnos potatorum and other seed extracts for treatment of turbid water, there has been little effort to characterize the active agents in these seed extracts or evaluate the efficacy as coagulants in reducing microbes from waters having different turbidities. The chemical composition of the coagulant in Strychnos potatorum has been identified as a polysaccharide consisting of a 1:7 mixture of galactomannan and galactan. These findings suggest that such seed extracts may function as a particulate, colloidal and soluble polymeric coagulant as well as a coagulant aid. The presence of other constituents in these seed extracts are uncertain, and there is concern that they may contain toxicants, because the portions of the plant also are used for medicinal purposes. Also, little has been done define, optimize and standardize conditions for their use. Furthermore, there appears to be little current effort to encourage or disseminate such treatment for household water or determine its acceptability, sustainability, costs and effectiveness in reducing waterborne infectious disease.

1.5 Summary
The results of several studies suggest that alum and other coagulation-flocculation or chemical precipitation methods can be applied at the household level to improve the microbiological quality of water and reduce waterborne transmission of diarrheal disease in developing countries. However, further studies are needed to determine if this type of treatment can be effectively, safely, and affordably applied for household water at point of use by the diverse populations living in a variety of settings. Furthermore, it is uncertain if household use of coagulation-flocculation can be optimized to provide efficient and consistent microbial reductions on a sustainable basis. Therefore, household water treatment by coagulation-flocculation and precipitation is not be widely recommended at this time. More information is needed on the effectiveness, reliability, availability, sustainability and affordability of these processes when applied at the household level. However, newer approaches to treatment of collected and stored household water have combined chemical coagulation-flocculation with chemical disinfection to achieve both efficient physical removal as well as inactivation of waterborne microbes. These systems offer great promise as effective, simple and affordable household water treatment technologies. These systems and their performance are described in a later section of this report.

2 Adsorption processes

2.1 Introduction
Adsorption processes and adsorbents such as charcoal, clay, glass and various types of organic matter have been used for water treatment since ancient times. Some of these adsorption processes tend to overlap with either filtration processes, because the media are often used in the form of a filter through which water is passed, or coagulation processes, because they may be combined with chemical coagulants. Therefore, adsorption processes can be carried out concurrently with filtration or coagulation. The candidate media potentially used for adsorption treatment of household water are shown in Table 13.

Table 13. Adsorbents for Water Treatment and their Advantages, Disadvantages and Costs for Household Use

Adsorbent Community/Household Use Advantages Disadvantages Cost* Comments
Clays Rare/rare-moderate Some efficiently adsorb microbes; adaptable to many treatment formats Some adsorb microbes poorly; availability limited Low to moderate Use as an adsorbent or coagulant
Charcoal (C), Activated Carbon (AC) Moderate/Moderate; (AC more in developed world; C more in developing world) Adaptable to many treatment formats; charcoal often readily available Poor microbe adsorption; can degrade microbial quality Moderate (C ) to high (AC) Used as adsorbents or coagulants; use varies regionally; C use based on traditional practice
Crushed organic matter: seeds, rice, etc. No-very rare/Rare-moderate in some countries Ditto charcoal and carbon Poor microbe adsorption; can degrade microbial quality Low Used as adsorbent or coagulant

*See footnote to Table 11 for explanation of cost basis.

2.2 Clay adsorption
Clay continues to be used as an adsorption medium for household water treatment in some regions and countries, with applications as clay particles in suspension, as filters (usually fired ceramic) or in conjunction with a chemical coagulant. Porous, fired ceramic clay filters (and adsorbers), typically as candles or other vessels have been described in a previous section of this report. The use of clay in conjunction with chemical coagulants also has been described elsewhere (Lund and Nissen, 1986; Olsen, 1987). When used alone, clays can decrease turbidity and microbes in water by about 90-95%. However, some microbes may not efficiently or consistently adsorb to certain, which reduces the overall efficiency of clay adsorption as a household water treatment process. Furthermore, the use of clay particles as suspensions in water is limited by the availability of the material and by the need to control the process so that the particles will settle, either alone or in the presence of a coagulant or coagulant aid. The use of such technology for clay adsorption requires training and is best supported by specialized equipment to carry out and monitor treatment effectiveness. Therefore, clay adsorption is not well suited for household water treatment.

2.3 Charcoal and activated carbon adsorption
Charcoal and activated carbon have been used extensively as adsorbents for water treatment in the developed and developing world. The main application is the reduction of toxic organic compounds as well as objectionable taste and odor compounds in the water. In developed countries granular or powdered activated carbon are used in community water treatment and granular or pressed carbon block is typically used for point-of-use or household water treatment (AWWA, 1999; LeChevallier and McFeters, 1990. Although fresh or virgin charcoal or activated carbon will adsorb microbes, including pathogens, from water, dissolved organic matter in the water rapidly takes up adsorption sites and the carbon rapidly develops a biofilm. Therefore, carbon is not likely to appreciably reduce pathogenic enteric microbes in water over an extended period of time. If anything, carbon particles are prone to shedding heterotrophic plate count bacteria and other colonizing microbes into the product water, thereby reducing the microbial quality. In many point-of-use devices the carbon is impregnated or commingled with silver that serves as a bacteriostatic agent to reduce microbial colonization and control microbial proliferation in the product water. Fecal indicator bacteria, such as total and fecal coliforms, and opportunistic bacterial pathogens, such as Aeromonas species are capable of colonizing carbon particles and appearing in product water. For these reasons, activated carbon is not recommended as a treatment method to reduce pathogenic microbes in drinking water. Additional treatment, such as chemical disinfection, often is needed to reduce microbe levels in carbon-treated water. Mixed media containing carbon along with chemical agents effective in microbial retention have been developed and evaluated. For example, carbon filters containing aluminum or iron precipitates have been described, and these filters have achieved appreciable microbial reduction in laboratory scale tests (Farrah et al., 2000). Therefore, it is possible that granular activated carbon filter media prepared with chemical agents more effective in retaining microbes may eventually become more widely available for point-of-use treatment of household water. However, the conventional charcoal and activated carbon media currently available for water treatment are not recommended for use at the household level to reduce microbial contaminants. Only charcoal or activated carbon media that been combined with other materials to improve microbial reductions should be considered for household treatment of collected and stores water and then only if there are performance data or certifications to verify effective microbial reductions.

2.4 Vegetative matter adsorbents
Historically, other vegetative matter has been used as an adsorbent for water treatment, as has been previously noted (Baker, 1948). Of these other plant media, burnt rice hulls seems to have been the most widely used in recent times (Argawal and Kimondo, 1981; Barnes and Mampitiyarachichi, 1983). The application of this material has been in the form of a granular medium filter, either alone or in conjunction with another filter medium, such as sand or activated (burnt) coconut shell. Use of this water treatment medium is still limited, primarily to those parts of the world where rice agriculture is widely practiced and where other filter media are not readily available at low cost. However, these adsorbent materials and their technologies require further development, evaluation and dissemination before they can be recommended for household water treatment in other parts of the world.

3 Ion exchange processes

3.1 Introduction
Ion exchange processes in water treatment have been used primarily for softening (hardness removal) in both community and point-of-use treatment and for disinfection in point-of-use treatment. Some ion exchange resins are used to deionize, disinfect or scavenge macromolecules from water. The main classes of ion exchangers used in water treatment and their advantages, disadvantages and costs are summarized in Table 14.

Table 14. Ion Exchangers and their Advantages, Disadvantages and their Advantages and Disadvantages for Household Use

Exchange Resin Community/Household Use Advantages- Disadvantages Cost* Comments
Softening resins Yes/Yes Easy to use Do not inactivate microbes; spent resin replacement and disposal required; unavailable in some parts of the world High Lack of microbial reduction makes them unsuitable for microbial reductions in household water treatment
Deionizing Resins Yes/Yes Inactivate microbes; easy to use Not recommended for drinking water; spent resin replacement and disposal required; unavailable in some parts of the world High The effects on long-term consumption of deionised water on health are not fully understood.
Iodine Disinfection (tri-iodide and penta-iodide) No/Yes Inactivates microbes; easy to use Risk of soluble iodine leaching into water; spent resin replacement and disposal required; unavailable in some parts of the world High Difficult to determine useable life without added technology; impractical and limited availability in developing world
Adsorbent and scavenging resins No/Yes Easy to use Not likely to inactivate microbes; microbial colonization and release a concern; not available in some parts of the world High Difficult to determine useable life without added technology; impractical and limited availability in many parts of the world

*See Table 11 footnote for explanation of cost basis

3.2 Softening, deionizing and scavenging resins
Ion exchange typically employs synthetic polymeric resins that must be centrally manufactured in specialized production facilities. The costs of these synthetic resins are relatively high and their availability in developing countries is limited. Ion exchange using natural zeolites has been applied to softening, chemical adsorption and other purposes in water treatment. However, natural zeolites have only limited availability worldwide, they require mining and processing systems that may be beyond the capacity of developing countries, and they have not been widely evaluated or used for microbial reductions in drinking water. The effects on long-term consumption of deionised water on health are not fully understood.

Softening resins are intended to remove hardness and they do not remove or inactivate waterborne microbes on a sustained basis. Furthermore, softening resins often become colonized with bacteria, resulting in excessive bacterial levels in product water, and they also increase the levels of sodium in the product water. In developed countries, point-of-use water treatment systems employing softening or scavenging resins often include addition treatment methods to reduce microbial loads in product water. Softening resins are relatively expensive, require regular monitoring and frequent replacement or recharging (regeneration of exchange capacity of spent resin); therefore, they are not practical for widespread household use to treated collected stored water. Because of their inability to reduce microbes, their complexities and other limitations, as described in Table 14, softening and scavenging resins are not recommended for household water treatment.

3.3 Ion exchange disinfection
Ion exchange disinfection is primarily with iodine in the form of tri-iodide or penta-iodide exchange resins. Portable and point-of-use iodine exchange resins have been developed and extensively evaluated for inactivation of waterborne pathogens, primarily in developed countries. Most of these are in the form of pour through cups, pitchers, columns or other configurations through which water is passed so that microbes come in contact with the iodine on the resin. Point-of-use iodine resins have been found to extensively inactivate viruses, bacteria and protozoan parasites (Marchin et al., 1983; 1985; Naranjo et al., 1997; Upton et al., 1988). While such iodine exchange disinfection resins are both effective and convenient, they are too expensive to be used by the world's poorest people and the production and availability of these resins is limited primarily to some developed countries. As described in the next section of this report, other chemical disinfection methods besides ion exchange halogen resins are available and preferred for household treatment to inactivate microbes in collected and stored drinking water.

4 Chemical disinfection processes

4.1 Chemical disinfectants for drinking water
Chemical disinfection is considered the essential and most direct treatment to inactivate or destroy pathogenic and other microbes in drinking water. The abilities of chemical disinfectants to inactivate waterborne microbes and reduce waterborne infectious disease transmission have been well known since the germ theory was validated in the mid-19th century. However, it was not until the late 19th and early 20th centuries that chlorine became widely recognized as an effective, practical and affordable disinfectant of drinking water. Subsequently, ozone and chlorine dioxide were developed as drinking water disinfectants and their ability inactivate waterborne pathogens was determined.

Today, chemical disinfection of drinking water is widely recognized as safe and effective and is promoted and practiced at the community level as well as at point-of-use. The preferred and most widely used chemical disinfectants of drinking water are all relatively strong oxidants, namely free chlorine, ozone, chlorine dioxide, chloramines (primarily monochloramine), and oxidants generated by electrolysis of sodium chloride solution (primarily or exclusively free chlorine). Additional chemical disinfectants sometimes used for drinking water are acids and bases; these agents inactivate microbes by creating either low or high pH levels in the water, respectively. The combined use of multiple treatment processes or "barriers" is a widely embraced principle in drinking water science and technology that is widely applied in community drinking water supplies, especially for surface waters. This approach also has been adapted to water treatment at the household level by the use of combined chemical treatments that are designed to chemically coagulate, flocculate, filter and disinfect the water. The advantages, disadvantages, costs and practicalities of these disinfectants for household treatment at the household levels are summarized in Table 15.

Table 15. Chemical disinfectants for drinking water supplies: advantages, disadvantages and costs for household use

Disinfectant Community/Household Use Advantages Disadvantages Cost* Comments
Free chlorine (NaOCl, Ca(OCl)2 Yes/Yes (worldwide, but not in some regions) Easy to use; effective against most pathogens; stable residual Not available worldwide; some users object to taste and odor Low The most widely used drinking water disinfectant; proven technology
Electro-chemically generated oxidant from NaCl Yes/Yes (limited distribution) Easy to use; effective against most pathogens; stable residual Not available worldwide; some users object to taste and odor (mostly chlorine) Low Practical for worldwide use; can generate on site by electrolysis of NaCl; proven technology
Chloramines (monochloramine) Yes/Rare (less widely used than free chlorine; must react free chlorine with ammonia Stable residual Less effective microbiocide than free chlorine; requires skill and equipment to generate on-site; household use impractical Moderate More difficult to use than free chlorine; potentially available where free chlorine is used but requires ammonia source
Ozone Yes/Rare(less widely than free chlorine; mostly in Europe) Highly micro-biocidal; No residual; Generate onsite; hard to use; need special facilities and trained personnel; hazardous High Not practical for household use in many regions and countries
Chlorine Dioxide Yes/Rare (much less use than free chlorine; for individual use by acidifying chlorite or chlorate) Highly micro-biocidal; Poor residual; generate on-site; some technologies require special facilities, trained personnel and are hazardous; toxicologic concerns High Can be generated on-site by reacting chlorate or chlorite salts with acids; reactants may not be available and some are hazardous
Acids (especially lime juice and mineral acids) and hydroxide (caustic) Limited/Limited (in community systems mineral acid and base for pH control; lime (CaO) and soda ash for chemical softening; in household Treatment lime juice for inactivation of V. cholerae Acids inactivate V. cholerae & some other bacteria; limes and chemicals widely available Limited microbiocidal activity; CaO use requires special facilities and trained personnel and is hazardous; CaO process difficult to control High for CaO; low-moderate for lime juice Lime juice has been reported to be effective for cholera control at the household level; Chemical acids and lime precipitation not practical for household use
Combined chlorina-tion, coag-ulation-flocculation-filtration systems Yes/Yes As sequential processes in community systems and as combined processes in household systems Highly effective for microbe reductions Availability now limited; requires some training and skill; efficacy varies with water quality; High Limited availability and higher cost (compared to chlorine) are barriers to household use in some countries and regions

*See Table 11 footnote for explanation of cost basis.

Iodine, silver, copper, quaternary ammonium compounds and some other chemical agents have been proposed and are sometimes used to inactivate waterborne pathogens. However, none of them are considered suitable for long-term use to disinfect drinking water for various important and valid reasons. Iodine is difficult to deliver to water and can cause adverse health effects, silver and copper are difficult to deliver to water and primarily only bacteriostatic, and quaternary ammonium compounds are limited in availability, costly and not effective against viruses and parasites. However, iodine, either dissolved in water or in the form of an iodinated exchange resin, has been used for short-term water treatment by outdoor recreationists (campers, hikers, etc), field military personnel, and persons displaced by natural disasters and human conflicts (wars and other societal disruptions). Silver is used as a bacteriostatic agent for point-of-use or household water treatment by storing water in vessels composed of silver or passing water through porous or granular filter media impregnated with silver. However, the extent to which silver alone inactivates microbes in water is limited, bacteria may develop silver resistance and many microbes, such as viruses, protozoan cysts and oocysts and bacterial spores, are not inactivated at silver concentrations employed for point-of-use drinking water treatment. Therefore, these agents are not recommended for routine disinfection of household water.

4.2 Factors influencing disinfection efficacy
The ability to inactivate waterborne microbes differs among the commonly used disinfectants as follows (from most to least potent): ozone > chlorine dioxide $ electrochemically generated oxidant $ free chlorine > chloramines. It is also noteworthy that these disinfectants differ in their stability and ability to persist in water to maintain a disinfectant residual. Ozone is a gas and the least stable in water; it is unable to provide a stable disinfectant residual. Chlorine dioxide is a dissolved gas in water and capable of persisting typically for periods of hours. Free chlorine is stable in water and can persist for days if there is no appreciable chlorine demanding material in the water. Because electrochemically generated oxidant from NaCl is primarily free chlorine (about 80 to nearly 100%, depending on the electrolysis conditions), it is relatively stable in water. Chloramines (primarily monochloramine) are the most stable of the listed disinfectants in water and can persist for many days.

Waterborne microbes also differ in their resistance to the chemical disinfectants used for drinking water as follows (from greatest to least resistant): parasites (protozoan cysts and oocysts and helminth ova) $ bacterial spores $ acid-fast bacteria (notably the Mycobacteria) > enteric viruses > vegetative bacteria. Within each of these major microbe groups there are differences in the resistance of different sub-groups and specific species or strains of microbes. In addition, the resistance of waterborne microbes to inactivation by chemical disinfectants is influenced by their physical and physiological states. Microbes in the form of aggregates (clumps) or embedded within other matrices (membrane, biofilm, another cell, or fecal matter, for example) are protected from being reached by chemical disinfectants and by the oxidant demand of the material in which they are present. This makes the microbes more resistant to inactivation.

The quality of the water to be disinfected also influences microbial inactivation by chemical disinfectants. Particulate, colloidal and dissolved constituents in water can protect microbes from inactivation by reacting with and consuming the chemical disinfectant. The microbiocidal activity of some chemical disinfectants is influenced by the pH of the water, with generally better inactivation at low pH than at high pH for free chlorine, for example. Further details of the water quality factors influencing microbial inactivation are presented in detail elsewhere (Sobsey, 1989).

4.3 Free chlorine treatment
Of the drinking water disinfectants, free chlorine is the most widely used, the most easily used and the most affordable. It is also highly effective against nearly all waterborne pathogens, with notable exceptions being Cryptosporidium parvum oocysts and Mycobacteria species (Sobsey, 1989). At doses of a few mg/l and contact times of about 30 minutes, free chlorine generally inactivates >4 log10 (>99.99%) of enteric bacteria and viruses. For point-of-use or household water treatment, the most practical forms of free chlorine are liquid sodium hypochlorite, solid calcium hypochlorite and bleaching powder (chloride of lime; a mixture of calcium hydroxide, calcium chloride and calcium hypochlorite). Bleaching powder is less desirable as a drinking water disinfectant because it may contain other additives that are undesirable in drinking water (detergents, fragrances, abrasives, etc). and because it is somewhat unstable, especially if exposed to the atmosphere or to water.

In addition to the well-documented evidence that free chlorine effectively inactivates waterborne microbes and greatly reduces the risks of waterborne disease in community water supplies, there is considerable evidence of the same beneficial effects in point-of-use and household water supplies. Table 16 summarizes the results of carefully designed intervention studies documenting the ability of free chlorine to reduce microbes and to reduce household diarrheal disease when used to disinfect household drinking water in developing countries. The results of these studies show conclusively that chlorination and storage of water in either a "safe" (specially designed) vessel or even a traditional vessel reduces diarrheal disease by about 20-48% and significantly improves the microbial quality of water (by reducing thermotolerant (or fecal) coliforms, E. coli, V. cholerae and other microbial contaminants). Most of the recent studies on chlorination of household water have been done by the US Centers for Disease Control and Prevention (CDC) and its many partners and collaborators around the world (CDC, 2001). The CDC intervention includes a hygiene education component and the use of a plastic, narrow-mouth water storage vessel with a spigot designed to minimize post-treatment contamination. Therefore, the beneficial effects of this chlorination system in the form of improved microbial quality and reduced diarrheal and other infectious diseases may also include the positive effects of the improved storage vessel as well has improved hygienic practices. However, some water chlorination studies listed in Table 16 tested only the effect of chlorination or chlorination plus the use of an improved water storage vessel for better protection of the chlorinated water during storage. These studies of the water intervention only also demonstrate improved microbiological quality of chlorinated water and reduced diarrheal and other infectious diseases.

Table 16. Efficacy of Chlorination and Storage in a Specialized Container (Safewater System) to Disinfect Household Water: Disease Reduction and Improvement in Microbial Quality

(a) Water storage and source:
(b) Significant difference in disease burden in intervention household members than in control household members. In some cases only certain age groups were studied or scored for an effect. cW = Water intervention only. SH = Sanitation and health intervention.
(c) Water and Service Level: Household stored water or other water; water service levels: on-plot (On) or off-plot (Off), communal (C), yard (Y), surface water (S), ground water (G), other water (O), mixed sources (M)
(d) CDC vessel: Plastic (high-density polyethylene), about 20-L capacity, valved spigot to dispense water, 6-9 cm opening to fill and clean, handle to carry and re-position.
(e) 12-L jerry can: plastic, 12-L capacity, medium opening for filling, cleaning and dispensing water.
(f) ORS = Oral rehydration solution.
(g) Narrow-necked vessel with a spout

Only a few studies have not demonstrated the ability of household water chlorination to reduce diarrheal illness. One such study compared microbial water quality and diarrheal illness in households chlorinating drinking water stored in clay pots at a dose of 6.3 mg/l versus matched control households not chlorinating in rural village in Northeastern Brazil (Kirchhoff et al., 1984). Bacterial contamination of water in households chlorinating was significantly lower than in non-intervention households (70 versus 16,000 colonies per 100 ml, p < 0.001). However, diarrhea rates were not significantly different between the intervention and control households. Several factors may account for the lack of differences in diarrheal illness rates between the two sets of households. First, it is unclear whether or not the treated water was actually used for drinking purposes and at what consumption level. Second, the drinking water for all families was heavily contaminated pond water, so it may have continued to harbor appreciable levels of pathogens despite chlorination. Third, chlorine dose was not rigorously controlled and chlorine residual in stored household water was not measured. Incorrect chlorine dosing may have resulted in either too little chlorine to reduce pathogens or too much chlorine that caused families to stop using the water or drop out of the study. Eight of 9 families who dropped out of the study reported objectionable taste as the reason for dropping out. Additionally, families were not involved in the intervention process; they were passive recipients of it and received no hygiene education. Finally, because overall sanitary conditions were poor and socio-economic status was very low, a single intervention only on drinking water may not have had sufficient impact on overall pathogen exposure to observe a significant decrease in diarrheal illness, especially in a short-term study of only 18 weeks.

As summarized in Table 16, many studies have shown that the microbiological quality of stored household water can be significantly improved and diarrheal disease can be significantly reduced by adding chlorine to water stored in a household vessel. Recent studies have attempted to overcome the limitations and uncertainties of previous efforts by employing uniform and fully articulated systems of chlorine production, distribution and dosing, an improved, standardized household water storage container, and the inclusion of participatory education, motivation and behavior modification components (Mintz et al., 2001; USA CDC, 2000). A simple and low cost system of adding chlorine to collected household water stored in a dedicated, narrow-mouth plastic container (preferably with a valved spigot) has typically reduced waterborne microbes by >99% and reduced community diarrheal disease, including cholera, by as much as 17-90% (Handzel, 1998; Mintz et al., 1995; 2001; Quick et al., 1999; Semenza et al., 1998). To make the system sustainable efforts are made to have users purchase concentrated sodium hypochlorite that is produced in the community by electrolysis of NaCl solution. The concentrated sodium hypochlorite solution is added to household water stored in a specially designed, rectangular, 20-liter plastic vessel with a moderate diameter, screw cap opening for filling and cleaning and a separate valved spigot to dispense the treated, stored water. The treatment and storage technology is accompanied and supported by an education, motivation (through social marketing) and behavior modification system to achieve community and household participation and improve hygiene behaviors related to household water use. This approach to providing microbially safe household drinking water, called the "CDC Safewater" system, has been successfully implemented in numerous communities, countries and regions in different parts of the world, including Latin America (Bolivia, Ecuador, Nicaragua and Peru), Central Asia (Uzbekistan), the Indian subcontinent (Pakistan), and Africa (Zambia and Madagascar) (Luby et al., 2001; May and Quick, 1998; Mong et al., 2001; Quick, et al. 1996; 1999; Semenza et al., 1998). In addition, similar systems have been used to disinfect water for oral rehydration therapy (ORT) solutions and for water and beverages marketed by street food vendors (Sobel et al., 1998; Daniels et al., 1999).

Consumer education, participation and social marketing are considered essential and integral to achieving acceptance and sustainability for this and other household drinking water treatment systems (Thevos et al., 2000). In addition, pilot and feasibility studies are also encouraged, as is economic, social and political support from donor agencies, NGOs, government agencies, the private sector and other sources. Such activities are recognized as essential for designing, mobilizing for, implementing and assessing this and other water quality management systems at the household level. Further considerations of the role of these factors in household water treatment systems are discussed in a later section of this report.

4.4 Chloramine treatment
Disinfection of water with chloramines or ammonia-chlorine is widely practiced in community water supplies in order to provide a long-lasting disinfectant residual and to reduce tastes and odors associated with the use of free chlorine in some drinking water supplies. Chloramination also reduces the formation of free chlorine by-products that are considered toxic, such as trihalomethanes. However, compared to free chlorine, ozone and chlorine dioxide, chloramines are relatively weak oxidants and germicides. Based on the product of disinfectant concentration, C, and contact time, T, (CxT) it takes about 10 to 100 times more chloramine than free chlorine to inactivate an equivalent amount of most waterborne microbes. Chloramination is also more difficult to apply to water than free chlorine because it requires the combined addition of controlled amounts of both free chlorine and ammonia. Treatment of water with chloramines is not commonly practiced at point-of-use and is not recommended for household water treatment. This is because it requires the availability of both free chlorine and ammonia, it is complicated to properly apply both free chlorine and ammonia to water at the required doses, the resultant disinfectant is weak and slow acting, and the cost of using both chlorine and ammonia is higher than the cost of using chlorine alone.

4.5 Ozone
Ozone has been used as a drinking water disinfectant since the early 20th century. It has gained popularity for community water supplies in developed countries because it is a strong oxidant capable of rapidly and extensively inactivating a variety of waterborne pathogens, including chlorine-resistant Cryptosporidium parvum oocysts. Ozone is a highly reactive gas that must be generated on site using electricity. It requires specialized equipment to deliver to water at required doses, and care must be taken to prevent safety hazards from the release of ozone gas by the treated water. Because ozone is rapidly consumed by dissolved and particulate constituents in water, achieving appropriate ozone doses in actual practice requires careful attention to water quality as well as the ability to monitor both ozone dose and ozone residual in the treated water. Therefore, this method of drinking water disinfection is suitable primarily for community or other centralized water systems where the specialized equipment and delivery systems required for its use can be properly applied by trained personnel. Although small, point-of-use ozone treatment systems are available for consumer use they are relatively expensive and difficult to maintain, require electricity and therefore are not recommended for household water treatment.

4.6 Chlorine Dioxide
Chlorine dioxide is used primarily as a bleaching agent and has gained some use for disinfection of both community and point-of-use drinking water supplies in developed countries. Chlorine dioxide is a relatively strong germicide capable of inactivating most waterborne pathogens, including Cryptosporidium parvum oocysts, at practical doses and contact times. For community water treatment, chlorine dioxide is generated on-site from the reaction of sodium chlorite with chlorine gas or from the reaction of sodium chlorite with acid. Producing chlorine dioxide from free chlorine and sodium chlorite is technically demanding and requires specialized equipment. For point-of-use treatment of water, chlorine dioxide is produced on site from the reaction of sodium chlorite with acid. Such use is primarily for disinfecting temporary or informal water sources used by outdoors enthusiasts and others requiring short-term applications in developed countries. The toxicity of chlorine dioxide and its by-products, such as chlorite, limits the use of this disinfectant, because the amount of toxic by-products is difficult to control or measure. In addition, the generation of chlorine dioxide from sodium chlorite and acid is relatively expensive, compared to free chlorine. For these reasons, chlorine dioxide is not widely used and is not recommended for long-term disinfection of household drinking water.

4.7 Combined point-of-use treatment systems
The combined application of chemical coagulation-flocculation, filtration and chlorine disinfection is widely practiced for community water treatment in developed countries, especially for surface sources of drinking water. In combination, these processes have been shown to dramatically reduce microbial contaminants in drinking water, produce product water that meets international guidelines and national standards for microbial quality and embody the principles of a multiple barrier approach to drinking water quality (LeChevallier and Au, 2000). Because of the relative complexity of these processes, they are more difficult to implement at point-of-use for household drinking water supplies in developed countries. However, purification of water at point-of-use using tablets or powders that combine a coagulant-flocculent and a chemical disinfectant have been described (Kfir et al., 1989; Rodda et al., 1993; Powers, 1993; Procter & Gamble Company, 2001). In South Africa commercial tablets containing chlorine in the form of Halazone p-triazine-trione or dichloro-S-triazine-trione and either aluminum sulfate or proprietary flocculating agents have been developed, evaluated and promoted for community and household water treatment, as well as emergency water treatment. For household use on non-piped, household water supplies it is recommended that the tablets be added to water in a 20-liter bucket. The mixture is stirred to dissolve the tablet and flocculate, then allowed to stand unmixed to settle the floc and then supernatant water is poured through a cloth filter into another bucket. When these tablets were tested for efficacy in reducing bacteria, viruses and parasites, they were found to achieve extensive reductions and meet US EPA requirements for a microbiological water purifier. According to the manufacturer, the cost of treatment with these tablets is low. Epidemiological studies of the effectiveness of these systems to reduce waterborne diarrheal and other diseases have not been reported.

The Procter and Gamble Company in cooperation with the USA CDC and other collaborators reports the development and evaluation of a combined flocculent-disinfectant powder supplied in a small packet that is added to a 10-liter volume of household water by consumers. The powder contains both coagulants and a timed-release form of chlorine. After stirring briefly, contaminants settle to the bottom of the container and the supernatant water is poured through a cloth filter into another container for safe storage and use. Initial studies document dramatic reductions of microbial as well as some chemical contaminants in water, and field epidemiological studies to determine reductions in household diarrheal disease are under way in Guatemala villages. The cost of the treatment is estimated at US$0.01 per liter.

Overall, combined coagulation-flocculation and chlorine disinfection systems have shown considerable promise as microbiological purifiers of household water. Currently, they have not come into widespread use and their worldwide availability is limited at the present time. However, further studies that document efficacy in reducing diarrheal disease and improving microbial quality are apparently forthcoming for some of these systems. Such data documenting performance and the commercial availability of the materials through widespread marketing and distribution create the potential for this technology to be not only scientifically supportable but also widely available in many parts of the world. The relatively high costs of these combined systems may limit their use by some of the world’s poorest people, but market studies also are under way to determine consumers' willingness to pay. Therefore, these combined systems may prove to be appropriate technologies for household water treatment in many settings for the large segment of the world's population now collecting and storing water for household use.

4.8 Lime juice disinfection of V. cholerae
Drinking water disinfection by lowering water pH with lime juice is effective in inactivating V. cholerae and in reducing cholera risks (Dalsgaard et al., 1997; Mata et al., 1994; Rodrigues et al., 1997; 2000). Adding lime juice to water (1-5% final concentration) to lower pH levels below 4.5 reduced V. cholerae by >99.999% in 120 minutes (Dalsgaard et al., 1997). Lime juice also killed >99.9% of V. cholerae on cabbage and lettuce and was recommended for prevention of cholera by addition to non-acidic foods, beverages and water (Mata, 1994). Epidemiological studies during cholera outbreaks in Guinea-Bissau showed that lime juice in rice foods was strongly protective against cholera and laboratory studies showed that the presence of lime juice inhibited V. cholera growth in rice foods. These studies indicate that adding lime juice to water, beverages and other foods (gruels, porridges, etc.) has the ability to inactivate V. cholera and reduce disease risks. Therefore, the use of lime juice in water and foods is a potentially promising household treatment to control cholera transmission. Further studies to better characterize the efficacy of this treatment and its ability to reduce cholera transmission are recommended.