Efficacy of Copper/Silver Ion Generation with Reduced Chlorine Concentrations on Disinfection and Operation of a Municipal Swimming Pool.

 ABSTRACT

The disinfection of swimming pool water in the Town of Brookline, Massachusetts' Municipal Swimming Pool Recreational Facility using chlorine concentrations according to the provisions of Chapter V of the Massachusetts Sanitary code (1.0 ppm free available chlorine) is compared with the use of Copper/Silver ion generation with the use of low levels of free available chlorine (0.4 ppm). Comparisons are made usin5 standard methods for the detection of coliform bacteria and heterotrophic bacteria using the standard plate count method. In addition, comparisons are made relative to trihalomethane production (THM) under conditions of chlorine disinfection alone and copper/silver ion.

I. Bathing Water Microorganisms and Traditional Disinfection Techniques

Microorganisms, including bacteria, protozoa and viruses occur naturally in recreational waters. Some of these microorganisms can be pathogenic, i.e., capable of causing human disease, and are, therefore, of legitimate public health concern. In swimming pools, these microorganisms may be introduced into the water by "carriers" and transmitted to other bathers via mechanisms of ingestion, inhalation, or broken skin. The literature is rife with instances of disease transmission involving the use of recreational waters (1).

Proper treatment of swimming pool water is essential to protection of the public health from diseases spread by microorganisms as discussed above. Swimming pool water treatment methodologies have traditionally incorporated the agglomeration of microorganisms and other water impurities utilizing chemicals such as aluminum sulfate and subsequent filtration through a medium such as sand or diatomaceous earth. However, due to a variety of reasons, including but not limited to the wide variation in size of microorganisms, the efficiency of the agglomeration technique and break-through in the filter medium, this treatment technology cannot be solely relied upon for water purification purposes. Effective disinfection of properly treated swimming pool waters has traditionally been accomplished by the addition of chlorine and chlorine compounds (2)(3)(4). In larger pools, chlorine is normally dispensed through a calibrated flow meter as gaseous chlorine or in aqueous solution as sodium hypochlorite, while in smaller residential pools, chlorine is dispensed as solid calcium hypochlorite. In all cases the active disinfecting agent is with low chlorine levels. Finally, the results of bather satisfaction using a bather survey technique are discussed under conditions of each test protocol. hypochlorous acid (HOCL). To insure effective elimination of all pathogenic microorganisms and to provide excellent general sanitary quality of swimming pool waters, State Health and/or Environmental regulatory agencies require that the bathing water contain an excess amount of free available chorine (FAC) above the amount which enters into reactions designed to inhibit the activity of microorganisms. In Massachusetts, swimming pool operators are currently required to maintain FAC levels of 1.0 ppm at all times when a pool is in use.

II. Disinfection Using Copper/Silver Ion Generation

The use of metallic ions in water disinfection techniques is not new. The early Greeks used copper and silver goblets and vessels for drinking and storage purposes (5). The low solubility of these metals served as a natural, controlled release mechanism which added trace amounts of these ions to the water. Such amounts were high enough to purify the liquid without causing objectionable taste.

More recent use of copper and silver ions to inactivate microorganisms is well documented (6)(7)(8)(9). In addition to bacteria, they also are effective in controlling viruses, algae and fungi in the part per billion ( ppb) range. Copper and silver ion disinfection of swimming pool water has several advantages over chlorine, viz., the ions are chemically stable and do not undergo the destructive reactions of aqueous chlorine; they do not form objectionable by-products such as chloramine or Trihalomethanes (THM); they do not escape from the water by volatilization as chlorine does.

Maintaining ppm range concentrations of copper and silver ions in swimming pool water in a convenient and reproducible manner is accomplished by electrolytic generation of the ions. electrolytic ion generators consist of a positively charged anode consisting of the metals to be ionized and a negatively charged cathode. The electrodes are housed in a chamber through which the water to be purified flows. The anode and the cathode are connected to a power source and a weak electrical charge flows between them, releasing silver and copper ions from the anode. The metals ion concentration is precisely controlled by varying both the flow rate of water through the chamber and the current to the electrodes.

Experimentation and publication by Gerba and others (10)(11)(12)(13)(14) indicates that 300-400 ppb of copper and 40 ppb silver combined with 0.1 ppm -0.4 ppm of chlorine is more effective in controlling a host of microorganisms, including coliform, than the use of higher levels of chlorine. The research points toward a synergistic effect when water containing microorganisms is subjected to copper/silver ion treatment with low levels of chlorine.

III. Test Protocol

The Town of Brookline Municipal Swimming Pool was chosen as the site for the current study because it is well maintained and operated by the Recreation Department staff is well regulated by the Brookline Health Department staff and it is utilized year round by the citizens of the Town and the surrounding community and organized high school swimming programs.

Chapter V of the Massachusetts Sanitary Code requires swimming pool waters to be disinfected using chlorine at a rate resulting in a FAC of 1.0 ppm. The Code also allows for alternative treatment technologies, but, at present, since they are viewed only as supplementary disinfectants, they must still conform with the above chlorine standards. If an operator wishes to operate below this minimum FAC requirement, then that is handled as a matter of variance issued by the local Board of Health after a public hearing and subsequent approval of that variance by the Massachusetts Department of Public Health (MDPH).

The Brookline Recreation Department applied to the Brookline Commissioner of Public Health on February 20,1996, for a variance to conduct a ten (10) week study using copper/silver ionization with reduced chlorine levels. On March 5,1996, a public hearing was held and on March 21,1996 the Brookline Health Department issued a variance for test purposes. On March 27,1996 the MDPH approved the variance. .

Under terms of the issued and approved variance, the test was to be carried out under a protocol which is attached to this report (Appendix n. The protocol in summary required Crystal Water Systems to conduct a two (2) week `Baseline Period" study whereby physical, chemical and bacteriological data and samples would be gathered for analysis by a certified laboratory (G&L Labs of Quincy, Ma.). During this period, no changes would be made relative to pool operation, i.e. chlorine dosage would remain such that a 1.0 ppm FAC was maintained.

At the end of the `Baseline Period", the installed copper/silver generator would be activated. When the copper levels in the pool water reached 300 ppb, chlorine levels would be reduced to 0.4 ppm. The facility would then be operated in this manner for a period of eight (8) weeks, during which data gathering, sampling and analyses would be intense.

An integral part of the test protocol was the design and implementation of a consumer reaction questionnaire which is attached (Appendix II).

IV. Purpose of the Test

The test, under the protocol described, was carried out to determine whether copper/silver ion generation used in conjunction with substantial reduction in chlorine usage, provides the same or better public health protection as the use of high levels of chlorine alone. The test was also designed to determine if the use of lower levels of chlorine have any positive effects on bather satisfaction.

V. The Swimming Pool

The Brookline Municipal Swimming Pool was constructed in 1958. Three distinct areas comprise the total pool complex viz., the diving area the wading area, and the lap pool. The three pools have a total volume of 245, 880 gallons.

Water supply for the pools is obtained from the Town of Brookline municipal water supply system which obtains its water from the Massachusetts Water Resources Authority (MWRA), a regional water supplier for the entire metropolitan Boston area. MWRA water is obtained totally from surface supplies, is not presently filtered, and is treated only with chlorine for disinfection, soda ash for corrosion control and sodium fluoride for control of dental caries. The Brookline municipal pool water is filtered through enclosed low pressure sand filters having a total capacity of 520 gallons per minute (gpm). The entire content of the pools are, therefore, filtered every eight hours.

Sodium hypochlorite in 10% aqueous solutions used for the disinfection purposes. It is fed through an electrically operated pump calibrated to maintain a FAC level in the pool water of not less than 1.0 ppm. Average chlorine usage is 9-10 gallons per day, providing an average chlorine dosage of 1.8 ppm. when chlorine alone is used for disinfection purposes. During the Pilot Phase of the study after introduction of copper/silver ions at the desired level, chlorine usage was reduced to less than 3 gallons per day 2.67g.p.d.)

VI. The Cooper/Silver Generator

The copper/silver ion generator is a Water System Model CWS 3001. The unit is NSF approved per Standard 50-1992 and UL listed. The system consists of two components: a controller and two flow cells which contain six copper/silver electrodes each. The controller measures 19 ½" x 15 ¾" x 8 3/8" and weighs 25 lbs. Input power to the controller is 110/220 volts 50/60 hertz; while Output is 16 volts Max 10 amps. It is a Class 2 Transformer. The Flow cells are constructed of high pressure Schedule 80 PVC and measure 18" long. The ratio of copper/silver is 99:1, respectively. This ratio allows the copper/silver ions imparted into the pool to fall within EPA Drinking Water Regulations which is a NSF requirement. The System was installed on July 18,1996. The electrodes were inserted as an offset to a separate water loop fed off the main pool line that feeds the Dectron dehumidifier on the roof.

VII. Baseline Period Testing

On June 25,1996 baseline data relative to chlorine residual (FAC), coliform bacteria total heterotrophic plate count and trihalomethane (THM) began to be gathered and continued until July 3l, 1996 - a period of 5.5 weeks. This represents an almost three-fold increase in data relative to the test design protocol, but was thought to be prudent considering that a FAC concentrations ranged from a minimum of 0.3 ppm to 3.10 ppm.

Review of the data indicates that as expected coliform bacteria appeared to be well controlled with only a single sample showing one (1) coliform colony at a time when average FAC was 1.9 ppm.

Relative to heterotrophic bacteria, an average of 90.8 colonies/ml were determined during this period.

An average concentration of 121.2 micrograms per liter (ug/L) of trihalomethanes was detected during the Baseline Period.

Bather load was on average of 288 persons/day during the Baseline Period.

VIII. Ionizer Ramp UP Period

From August l, 1996 through September 16,1996, the copper/silver ion generator was introduced into the water treatment system, while chlorine dosage levels were reduced. The ion generator responded well to ramp-up, reaching a level of 0.3 ppm by August 17,1996, the pool was fully ionized. However since the pool was to be closed (August 25,1996 - September 10, 1996) for its annual maintenance program, the Pilot Phase part of the test was delayed until the reopening. Following the reopening of the pool and restoring the copper levels (see (1) below), The chlorine levels were reduced to the 0.4 ppm level as stipulated in the protocol and the Pilot Phase commenced.

IX. Pilot Test Periods

On September 17,1996, data began to be gathered relative to the actual performance of copper/silver ionization with reduced levels of chlorine and continued for an uninterrupted period of four weeks. During the period from October 19,1996 through October 23,1996, the Boston area received almost nine (9) inches of rainfall causing severe generalized flooding conditions throughout the area. The Brookline Municipal Pool Building was a victim of this flooding and the swimming pool water treatment room was inundated causing failure of all pumps. The pool, with ionizer fully operational, and copper levels restored to 0.3 ppm was not in operation until November 1l, 1996.

Upon restoration of all pool equipment, the Pilot Period was once again commenced on November 1l , 1996, and ran uninterrupted until December 21,1996, a period of six weeks. Pilot test period in the aggregate provide ten weeks of data.

During this time period FAC levels averaged 0.52 ppm with an occasional excursion to 1.0 ppmbut for the most part remained in the 0.4 - 0.5 ppm range.

Copper/silver ion levels remained consistent at the 0.3 ppm level with only occasional readings of 0.2- 0.25 ppm.

No coliform bacteria colonies were developed throughout the Pilot Period Test. Relative to heterotrophs, an average of 20.2 colonies were counted during this period. 48.5 ug/L of THM was detected as an average during the test period. Bather load during this period averaged 202 persons/day.

X. Discussion

A) Free Residual Chlorine (FAC)

It is interesting to note the extremes in chlorine concentrations during the Baseline Period data collection. Chlorine concentrations ranged from a maximum level of 3.10 ppm to a minimum of 0.3 ppm. The average maximum level was 2.7 ppm and the average minimum level was 0.6 ppm which is a spread of 4.5 times. This spread indicates two things: i) the chlorine residual is unstable and is quickly subject to the influence of chlorine demand (bather load); and ü) the chlorine feed system either reacts slowly to changes in chlorine residual or that high/low set point signals are set too widely apart.

Data collected during the Pilot Period demonstrated chlorine concentrations ranging form a maximum level of 1.3 ppm to a minimum of 0.2 ppm. The average maximum level was 1.0 ppm and the average minimum level was 0.4 ppm, which is a spread of only 2.5 times. This suggests that the chlorine residual was more stable when copper/silver ions were being used.

B) Coliform Bacteria

Coliform bacteria levels appeared to be well controlled throughout the duration of the study whether using conventional high levels of chlorine alone or when using copper/silver ions in conjunction with reduced levels of chlorine. The only coliform event detected during the study occurred during the Baseline Period data collection when chlorine levels were at an average of 1.9 ppm FAC.

C) Heterotrophic Plate Count

The heterotrophic plate count data provide the most significant information regarding the capabilities of the two disinfection strategies. This data provides a measure of the total heterotrophic bacteria in the pool water. It looks at a much larger population of bacteria than the important but more limited group of organisms detected in the coliform procedure. Therefore, the heterotrophic plate count provides a better measure of the overall sanitary condition of the pool water.

The Pilot Period study showed far lower numbers of heterotrophic bacteria detected than during the Baseline Period data collection. An average of 90.8 colonies/ml were found in samples collected during the Baseline Period, while only 20.2 colonies/ml were detected while using copper/silver ionization. This represents a 78% reduction in bacterial population.

D) Trihalomethanes (THM)

Trihalomethanes (THM) are a group of halogenated hydrocarbons which have been found to be potentially cancer causing. THM are produced when chlorine is introduced into water containing organic constituents and can either be ingested or absorbed through the skin. Therefore, THM concentrations in drinking water are regulated by the United States Environmental Protection Agency (LTSEPA) at a level of 100 ug/ml. There is no regulated maximum concentration for THMs in swimming pools.

An average concentration of I21.2 ug/L was found in samples collected during the Baseline Period, while an average concentration of only 48.5 ug/L was found during the Pilot Period of the study. This represents a 150% reduction in THM concentrations.

XI. User Satisfaction

In order to gain insight into whether the introduction of copper/silver ions with reduced levels of chlorine has any effect on the bathing experience of users of the swimming pool, a survey was designed, implemented and analyzed by Opinion Dynamics of Cambridge, Ma. The full report is attached (Appendix II].

It is well understood that bathers who use swimming pools using chlorine alone as a disinfectant have experienced a variety of unpleasant side effects, including but not limited to, bleaching of skin, hair and bathing suit material; eye, nose and throat mucous membrane irritation; unpleasant odors; and skin irritation and rashes. Such pool users seem to understand that these unpleasant effects are the price to be paid for assurance that the pool water is free from parthenogenic organisms. In fact, some persons are so sensitive to chlorine as to cause them to avoid using swimming pools.

The survey found a marked increase in bather satisfaction swimming in water disinfected by copper/silver ions and low levels of chlorine, by a margin of 76% - 2%. Users experienced a very positive reaction to copper/silver ionization in that the incidence of eye irritation was cut by 16%; objectionable odors by l0o/a; bleaching of hair by 6%; and skin irritation by 4%.

XII Conclusions and Recommendations

Review of the data generated during the course of the study clearly supports the premise that copper/silver ionization technology is an effective and superior alternative to conventional swimming pool water disinfection by use of high levels of chlorine alone. The technology provides a high level of bacteria control, while lower chlorine levels result in substantial reduction of the production of trihalomethanes and increase substantially the enjoyment and satisfaction of the swimming experience for the pool user.

It is strongly recommended that the Massachusetts Department of Public Health after review of this evidence appropriately revise Chapter V of the State Sanitary Code to allow municipal Boards of Health to permit the use of copper/silver ionization technology with reduced levels of chlorine for indoor swimming pool disinfection.

BIBLIOGRAPHY

1. V.J.Cabelli. Water Sci. Tech. 21 13-21 (1989).

2. J.G. Jacangelo, V.P. Oliveri, and K Kawat.a. Mechanism of Inactivation of Microorganisms by Combined Chlorine. American Water Works Research Foundation. Denver, CO.

3. J. CarTell Morris in "Water Chlorination Environmental Impact and Health Effects" Volume 1. Robert L. Jolley (ed.) Ann Arbor Science Publishers Inc., Ann Arbor, Ifch. (1979).

4. Robert J. Jolley and James H. Carpenter in "Water Chlorination Environmental Impact and Health Effects" Volume 4. Robert L. Jolley (ed.) Ann Arbor Science Publishers Inc., Ann Arbor, Mich. (1983).

5. Showcase USA, 3rd Quarter, 52-4 (1990).

6. C.W. Chambers, C.M. Proctor, and P.W. Kabler. J. Am. Water Works Assoc. 54, 208- 216 ( 1962).

7. D.O. Cliver, W.K. Foell, and J.M. Goepfert. Final Technical Report. Contract NAS 9-9300. Food Research Institute, University of Wisconsin, Madison.

8. K. ZVuhrmann and F. Zobrist. Schwiez. Z. Hydrol. 20 218-254 (1958).

9. RA MacLeod, S.C. Kuo, and R Geliuas. J. Bacteriol. 93 961-9 (1967).

10. S. M Kutz et al, Proceedings Fourth Conference on Progress in Chemical Disinfection. GiIbeTt Janauer (Chairman) Dept. of Chemis-try, SUNY Binhampton.

11. M.T. Yahya et al, J. Environ. Health SI 282-285 (1989).

12. M.T. Yahya et al, Can J. Microbiol. 36 109-116 (1990).

13. L.K Landeen, M.T. Yahy and C.P. Gorba. Appl. Environ. Microbiol. 55 3045-3050 (1989).

14. M.T. Yahya et al, Int. J. Environ. Health Res. 1 76-86 (1991).

Recreational Water Disinfection by Copper/Silver Ions (Synergistically with Chlorine)

CHARLES W. BEER, Ph.D.

 ABSTRACT

 The disinfection of water used for recreational purposes with electronically generated copper and silver ions is reviewed. Ionizer technology, the chemistry of aqueous chlorine, disinfection mechanisms for metal ions and chlorine, and advantages and disadvantages of metal ion and halogen-based disinfection systems arc discussed. A number of studies arc cited which demonstrate the synergistic effect of low levels of chlorine (0.4 ppm) with electronically generated copper and silver ions. These disinfection systems show the rapid inactivation rates associated with higher (>1 ppm) chlorine levels coupled with stable, long-lived residuals of the metal ions. The metal ion residuals required for effective synergy with chlorine are within drinking water standards.

WATERBORNE PATHOGENS

Microorganisms occur naturally in potable and recreational waters ( 1 ) as well as in hot water systems, cooling towers, and decorative fountains (?-13). These microorganisms may be protozoa, bacteria or viruses and may be pathogenic.

Waterborne pathogens can pose a significant health risk in pools or spas (14-18). They may be shed by bathers (19-?0) and transmitted to nearby swimmers via ingestion, inhalation, or contact with the eyes, nasal passages, oral cavity, or broken skin (21). Other pathogens, most notably Legionella, may be transmitted via aerosols from decorative fountains (2) or heated spas (22). A recent outbreak of Legionellosis occurred aboard a cruise ship and was caused by an improperly disinfected Whirlpool Spa (23-25).

CHLORINE DISINFECTION

Proper disinfection of water used for recreational purposes is important, since these waters are continually inoculated with microorganisms which are naturally occurring as well as those which are transmitted from bather to bather. Disinfection in swimming pools has additionally been performed by chlorine and chlorine compounds (26): bromine compounds are gaining acceptance for disinfection of spas and hot tubs.

Chlorine is normally dispensed in larger pools as chlorine gas, in intermediate size pools as liquid sodium hypochlorite solution, and in small residential pools as solid calcium hypochlorite. In all cases, the active disinfecting agent is hypochlorous acid, HOCl, which is formed from gaseous chlorine by the hydrolysis reaction (27,28)

Cl₂ + H₂O à HOCl + Cl^- + H⁺ ( 1 )

and from hypochlorites via the reaction

OCl^- + H₂O à / ß HOCI + OH^- (2).

Reaction (1) generates hydrogen ion which decreases the pH while reaction (2) generates hydroxyl ion which increases the pH. Reaction (1) requires caustic or soda ash to be added to balance the pH while reaction (?) requires acid or sodium bisulfate to balance the pH.

Hypochlorous acid is a weak acid which dissociates according to the reaction

HOCl à / ß H⁺ + OCl^- (3).

The dissociation is pH dependent; at pH values between 7.3 - 7.7, the range recommended for swimmer comfort and proper pool maintenance, the amounts of HOC1 and OCl' are roughly equal (28). The concentration of hypochlorous acid, HOCI, plus the concentration of hypochlorite ion, OC1-, is referred to as the free available chlorine or chlorine residual. The recommended chlorine residual is 1-? ppm for pools used mainly by adults. Higher residuals arc recommended for wading pools used by small children (29).

Organic chlorine is also used to disinfect pools. These arc organic compounds which dissociate in the presence of water to form hypochlorous acid. Most are based on chlorides of isocyanuric acid. A number of newer halamine water disinfectants have been reviewed by Worley and Williams (30).

CHEMISTRY OF AQUEOUS CHLORINE

In addition to reactions (1) - (3) which describe the formation and dissociation of hypochlorous acid, a number of other important chemical reactions occur for this compound as well as the hypochlorite ion. These include decomposition reactions via disproportionation or decomposition to oxygen, formation of chloramines and their subsequent reactions, and oxidation of organic species present in water (28-29).

DECOMPOSITION REACTIONS

These reactions consume hypochlorous acid or hypochlorite ion to form the chlorate ion or molecular oxygen. Because these compounds are so reactive, they are unstable and react with themselves to form products which are inefficient disinfectants. The disproportionation reaction is given by:

3H0C1 - 3H` + ?Cl + ClOz ' (4a)

or

30C1' - 2C1' + C10; (4b)

The rate of these reactions depends upon the square of the HOCI or Ocl concentration and is pH dependent, having a maximum at pH 6.7.

The decomposition reaction is given by:

2H0C1 - ?H` + 2C1' + O (Sa)

or

20C1' - ?Cl' + O (5b)

The rate of reaction (Sa) is greater than that of (3b) and is 8reatly accelerated in sunlight.

FORMATION OF CHLORAMINES

Ammonia and Ammonium compounds occur in perspiration and urine and are thus added to the pool environment by bathers. Ammonium compounds arc also present in fertilizers and -are added to the pool environment by runoff. Chloramines are formed from ammonia by reaction with HOCl:

NH3 + HOC1 - hTH:Cl T HO (6a) ,

NH:CI + HOC1 - NHCl: + HO (6b) .

and NHCl. + HOC1 - NCl3 + HO (6c) .

At normal pool pH, monochloramine, NH Cl, constitutes > 90 7o of the chloramine formed, the balance being dichloramine, NHCI2.

Chloramines themselves are disinfecting agents, but are much slower acting than free chlorine. Chlorine in the form of chloramine is referred to as combined chlorine. Combined chlorine is not an effective disinfectant and is often associated with odor and eye and skin irritation.

Chloramines are removed from the pool by adding excess HOCl ("superchlorination") to form dichloramine via reaction (6b), which then decomposes to nitrogen via the following reaction:

2NHCl + HO -- N + HOC1 + 3H` + 3C1' (7).

Organic chloramines are formed by the reaction of organic amines, RNH:, with HOCl or NH2Cl. Little is known about the occurrence of nitrogen-containing organics in water or their reactions with chlorine. It is believed that organic chloramines are responsible for the combined chlorine residual after superchlorination to remove chloramines has been carried out.

REACTION WITH ORGANIC COMPOUNDS

Aqueous chlorine, in either the form HOC1 or OC1-, can participate in three basic chemical reactions with the organics normally found in swimming pool environments. These reactions are: oxidation, addition, and substitution. These reactions are responsible for the disinfection of microorganisms by chlorine as well as the generation of new organic compounds from organics occurring in swimming pools.

Oxidation occurs with a wide variety of organic compounds. Some examples are the oxidation of urea to nitrate, sulfhydryl groups to sulfuric acids, or carbohydrate moieties to organic acids.

Addition reactions involve the formation of halohydrins across carbon double bonds. These reactions are probably important for tanning lotion residues which contain oils and sun blocks containing C=C double bonds.

HOC1 is a good elcctrophile which can add to activated phenyl rings and hetcrocyclics such as uracil. OCl' reacts with the hydrogen's alpha to a carbonyl group to substitute a chlorine. After hydrolysis, this produces a molecule of chloroform. Bromine and iodine-containing disinfectants similarly produce bromoform or iodoform; mixtures of chlorine, bromine or iodine-containing disinfectants produce trihalomethanes (THVI's) containing a mixture of Cl, Br, and I atoms.

MECHANISM OF CHLORINE DISINFECTION

A number of authors have proposed mechanisms for the inactivation of bacteria and viruses by chlorine, however, the mechanism for chlorine disinfection remains unclear. Both hypochlorous acid and hypochlorite ion are strong oxidizing agents which can attack cell walls and impair cellular membrane functionality. Venkobachar et al (31) found that chlorine changes the membrane permeability of E. coli, allowing proteins and RNA to leak out. At higher chlorine levels, they observed leakage of DNA.

McFeters and Camper (32) found that reaction of cell surface enzymes and chlorine reduced metabolic efficiency. They showed that chlorine- injured cells had impaired transport systems and were incapable of moving glucose and amino acids across the cell membrane. Roller et al (33) showed that inactivation of bacteria by chlorine was due to oxidation of sulfhydryl groups, impairing enzymatic activity.

LeChevallier et al (34) studied the invasiveness of chlorine-injured Yersinia enterocolitica. They found that two requirements appeared necessary for invasiveness: (1) the organism must be viable and metabolically active, and (?) the organism must have certain surface components to induce engulfment. Chlorine was found to alter certain characteristics of the bacterial outer membrane so that it had a reduced capacity to bind to the host cells. Oxidation of sulfhydryl groups or substitution of chlorine into the amino acid sequences of the bacterial cell wall proteins by chlorine may have been responsible for the reduction in binding capacity. A similar study carried out by Terzieva and McFeters (35) on Campylobacter jejuni showed a 6-fold decrease in cell invasion after chlorine treatment.

Sletten (36) reported that HOC1 was more effective than OC1' in inactivating bacteria. This may be due to the fact that HOCl is electrically neutral and capable of diffusing through bacterial cell walls. Once inside, it can react with proteins that are inaccessible to the charged OC1 ion.

The inactivation of viruses by chlorine compounds has been reviewed by Thurman and Gerba (37). They stated that there are currently two schools of thought on the inactivation mechanism: one school assumes the proteins of the viral capsid arc attacked while the other assumes attack of the viral nucleic acids. Attack on the protein capsid would block or destroy host cell receptors on the viral outer surface. Attack of the viral nucleic acid would prevent the virus from replication properly.

In general, viruses are more difficult to inactivate than bacteria, due to the simpler nature of the virus. There are no metabolic processes occurring within a virus that can be impaired by reaction with chlorine.

DISADVANTAGES OF CHLORINE DISINFECTION

Chlorine has been the water disinfectant of choice since the early 1900's. It has a rapid inactivation rate for most microorganisms, and is sufficiently long-lived to provide a protective residual for continued disinfection. Finally, it is relatively cheap and readily available.

However, a number of drawbacks of using aqueous chlorine as a disinfectant have come to light recently. Of greatest concern is the production of trihalomethanes (THM's) by the reaction of chlorine (or bromine or iodine) with organic materials (see "Reaction with Organic Materials" above). These compounds have been linked to cancer in laboratory animals. By-products of water chlorination have also been associated with rectal and bladder cancer in humans through epidemiological studies (38,see reprint by Morris et al). The EPA has currently set a limit on THM's in drinking water of 100 parts per billion (ppb). Water in swimming pools can have THM concentrations several times the EPA limit (39-40).

The danger in swimming in pools contaminated with THM's is (41) not from ingestion of THM's, but from absorption of THM's through the skin. Brown et a! calculated that skin absorption of aromatic hydrocarbons during the course of bathing or swimming would exceed the dose obtained from drinking water containing; the same concentration. Since THM's are volatile, inhalation is also another route to exposure. Several studies have measured the increase in THMI levels in blood plasma following swimming in THM-contaminated water (42-45). One study carried out in Denmark found that swimmers training for two hours in water contaminated with 150 ppb of chloroform absorbed 5 mg of this compound (45).

Since THM's are volatile, they leave the pool environment shortly after formation and go into the atmosphere, eventually reaching the ozone layer. There they participate in photochemical reactions which liberate halogen atoms which contribute to ozone depletion.

Another disadvantage of chlorine disinfection in swimming pools is the short lifetime of chlorine residual, especially on hot, sunny days. Reactions (4)-(7) consume chlorine and require constant monitoring to maintain the free chlorine residual in the 1-2 ppm range.

Attempts to stabilize the chlorine residual during hot weather using cyanuric acid brings along its own set of problems. Cyanuric acid stabilizes the chlorine by combining with it to form mono, di, and trichloroisocyanurates (46). The relative amounts of each compound is determined by the pH and free chlorine concentration. Thc chloroisocyanuratcs act as a reservoir for free chlorine according to the following reaction:

H3Cy + xHOCl ClxH .XCy + xH:O (8)

Cyanuric acid itself is toxic and has a 100 ppm concentration limit. It's use has been banned in New York State. Fitzgerald and Der Vartanian (47) measured the kill time for streptococcus faecalis in a solution containing 0.3 ppm chlorine and varying amounts of cyanuric acid. They found that cyanuric acid increased the kill time significantly and the increase was proportional to the concentration of cyanuric acid.

Failure to maintain the proper chlorine residual, either due to negligence, equipment malfunction or bather overload can have serious health consequences. An account is given of a gastroenteritis outbreak in North Dakota which resulted from the free chlorine residuals of a pool and spa dropping to 0 ppm over a several day period, presuma5ly due to bather overload 48,sce reprint by Holmes et al).

A number of microorganisms have developed resistance or survival mechanisms to chlorine disinfection. Legionella pneumophilia is quite resistant to chlorine as is poliovirus. The attachment of bacteria to surfaces to form biofilms provides a high degree of resistance to free chlorine disinfection (50).

Another drawback to usin5 aqueous chlorine for swimming pool disinfection is the irritation swimmers' eyes, nasal passages, and skin as well as the odor in the pool area associated with high levels of chlorination. An odor associated with chlorine usually occurs at concentrations greater than 0.6 ppm (51). This may be due to chloramines rather than hypochlorous acid itself, however, the odor is usually associated with pool chlorination. The irritation to eyes and mucus membranes associated with chlorination may not be due to chlorine itself but rather chloramines, THM's, or improper pH or hypotonicity of the pool water. Chiswell and Wildsoet (52) have reviewed the subject of eye irritation in swimming pools and have begun a series of experiments to identify the sources of irritation.

Finally, the issue of safety in handling gaseous chlorine must be addressed. Chlorine gas is a highly toxic and reactive material which is under pressure. It is a Class III Oxidizer and requires special storage. Switzerland has passed legislation to prevent incidents from occurring in chlorine disinfection facilities in public swimming pools (53).

DISINFECTION USING COPPER AND SILVER IONS

Copper and silver ions have been used for centuries for disinfecting water. The early Greeks used copper and silver vessels to store water and Greek royalty stored their water in silver flagons and drank from silver goblets (54). The low solubility of these metals in water served as a natural controlled-release mechanism which added trace amounts of these ions to the contents of the vessel. The amounts were high enough to purify the liquid without causing objectionable taste or health problems to the users.

The more recent use of copper and silver ions to inactivate microorganisms is well documented (55-58). In addition to bacteria, they also control viruses, algae, and fungi. They are as effective at the part per billion (ppb) level. Current EPA maximum levels for these metals arc 1 ppm for copper and 50 ppb for silver. At these levels, their inactivation rates are low compared to chlorine (59).

Copper and silver ion disinfection of swimming pool water has several advantages over chlorine but several disadvantages as well. The ions are chemically stable and do not undergo the destructive reactions that aqueous chlorine does, thus it is easier to maintain an effective residual dose. They do not form objectionable by-products such as chloramines or THM's as chlorine does. However, maintaining their concentrations in the ppb range using standard metering techniques would prove difficult to implement reproducibly and conveniently. Also, their relatively slow rates of inactivation would make it difficult to keep up with the high ioburden levels which occur during; times of peak bather load.

The problem of maintaining ppb concentrations of the ions in a convenient and reproducible manner was solved by using electrolytic generation of the ions. Dorroh first used the technique in the 1930's to generate silver ions for swimming pool disinfection (60). More recently, compact electrolytic silver ion generators were developed as pan of the water systems aboard Apollo spacecraft (61).

Electrolytic generators consist of a positively charged anode consisting of the metal(s) to be ionized and a negatively charged cathode. The electrodes are housed in a chamber through which the water to be purified flows. The anode and cathode are connected to a DC power source of a few volts and a weak electrical current flows between them. This causes the following electrode reactions to occur:

Ag à Ag + e Anode (9a)

H2O + e- à 1/2 H2 + OH Cathode (9b)

The metal ion concentration can be precisely controlled by varying the flow rate of the water through the chamber and varying the electrical current to the electrodes. The rate at which metal ions are produced by the electrode is directly proportional to the applied current to the electrodes. The rate at which the ions are removed from the cell is directly proportional to the water flow rate through the cell. The concentration of metal ions in the effluent from the cell is given by:

C = k i/F , ( 10)

where C = Concentration (in ppb),

i = current

F = Water flow rate,

and k = constant.

The Apollo silver ion generator (61)was tested extensively on distilled water and sweat condensate using E. coli and Staph aureus test organisms. E. coli in distilled water showed complete kill within 90 minutes and Staph aureus within 24 hours at a silver ion level between 50-100 ppb. Staph aureus had nearly a 1009c kill rate in 4 hours in sweat condensate at pH 7.6-7.75 and silver concentrations of 100-146 ppb.

A number of ionizers for swimming pool disinfection are now commercially available (62).They no longer employ pure silver electrodes as the ionizer developed for the Apollo program, rather they now use 80/?0 or 90/10 copper silver electrodes. Ionizers are gaining commercial acceptance and there are currently standards which govern their manufacture (63).

Gerba and co-workers were probably the first to perform a microbiological evaluation of copper-silver ionizer (59,see reprint by Kutz et al). They measured the reduction in cultures of E. coli, Streptococcus faecalis, Pseudomonas aeruginosa, Legionella pneumophilia, Salmonella typhi, Klebsiella terrigena, and Staphylococcus aureus in well water containing 400 ppb copper and 40 ppb silver. Low levels of chlorine (0.1-0.4 ppm) were also added to the copper/silver solutions. They found that the rates of inactivation for the copper/silver solutions were much slower than those for free chlorine, however when low levels of chlorine were added to the copper/silver solutions, the inactivation rates were faster than for free chlorine.

Subsequent publications by Gerba and coworkers on bacterial systems corroborated their initial findings that copper/silver ions combined with low levels of chlorine (0.2 ppm) inactivated bacteria faster than did free chlorine. In one study, they found that the combined metal ion/chlorine system had a 0.7 log or greater reduction in bacterial number than did the system containing free chlorine (64, see reprint by Yahya et al 1989). The metal ion concentration changed very little during the course of the experiment, however the chlorine concentration dropped by a factor of two.

In a definitive paper, Gerba and co-workers actually simulated indoor and outdoor pool environments and made additions of urine and bath water to simulate bather load (65.see reprint by Yahya et al 1990). They compared 400 ppb copper/40 ppb silver containing 0.3 ppm chlorine as a disinfectant to 1.0 ppm chlorine and found no significant differences between the two systems. The test was carried out for four weeks and they found the number of total colifroms, staphylococci, and pseudomonas were all less than drinking water standards. The implication of this work is that significantly lower chlorine residuals can be used with systems containing copper/silver ions.

Landeen et al (66, see reprint) used various levels of electrolytically generated copper and silver ions combined with low levels of chlorine (0.1-0.4 ppm) to inactivate Legionella pneumophilia. Legionella ordinarily is difficult to control with chlorine. At room temperature, they found that the inactivation rate for systems containing both metal ions and chlorine was significantly greater than for systems containing chlorine alone.

Gerba and coworkers also investigated inactivation of viruses with the copper/silver ion system with and without added chlorine. In one study (49,sec reprint Yahya 1992), they used electrolytically generated ions (400 ppb copper/40 ppb silver) to inactivate coliphage MS-2 and poliovirus. They found the race of inactivation was slow when only the metal ions were used, however, when 0.3 ppm chlorine was added, the inactivation rate increased 100-fold. In another study (67), they investigated inactivation of the same two viruses in copper, galvanized, and plastic pipes. The galvanized pipes had the highest kill rate for MS-2 virus while copper had the highest kill for poliovirus. Once again, an increase in rate of inactivation was observed when 0.? ppm chlorine was added to the copper solution.

McFeters and coworkers (68, see reprint Pyle et al) investigated the efficacy of copper and silver ions with iodine in the inactivation Pseudomonas cepacia. They used low concentrations of electrolytically generated copper and silver ions ( 100 ppb copper/11 ppb silver ) with 200 ppb iodine. Rapid inactivation was attained with the copper/silver/iodine system. Regrowth of the microorganism occurred after iodination but not when the copper/silver system was used.

The studies cited above all provide strong evidence of a synergy between low levels of copper and silver with chlorine or iodine, resulting in a disinfectant system that has a rapid rate of inactivation. This rate of inactivation is comparable to chlorine only-systems with halogen levels several times higher. Although no studies have yet been performed on metal ion/bromine systems, the implication is that a similar synergy should exist due to bromine's position between chlorine and iodine on the periodic table.

MECHANISM OF COPPER/SILVER DISINFECTION

McFeters and coworkers have performed a number of studies which have shed light on the mechanism of copper injury to bacterial cells. In their initial work on copper-induced damage, Domek et al (69, see reprint 1984 ) observed that 649c of the coliform population in l4 drinking water samples analyzed were injured. Using a statistical model, they traced the damage to the copper ion concentrations (mean value.= 160 ppb) in the distribution system. They conducted laboratory experiments and confirmed that levels of copper as low as 25 and 50 ppb caused 909o injury in 6 and 2 days respectively. ·They subjected copper-injured E. coli cells to physiology studies in which they monitored oxygen uptake. They found that the injured cells had significantly lower oxygen uptake than healthy cells and associated the damage to the respiratory chain. They hypothesized that the damage was caused by copper binding to the sulfhydryl groups of respiratory enzymes in thc cell membrane.

In a follow-up study, Domek & McFcters (70, sec reprint 1987 ) performed a series of elegant experiments to study the metabolism of copper-injured E. coli. They used C-13 NMR and Gas Chromatography to study differences in metabolism between injured and healthy bacteria. They also performed oxygen uptake studies which corroborated their previous work.

Thc NMR and GC studies of metabolism were carried out on E. coli cells grown under aerobic and anaerobic conditions, using glucose and succinate as nutrients. The studies showed inhibition of glycolysis and tricarboxylic acid activity in the copper-treated cells in comparison to healthy cells. Again, damage to cell surface enzymes by copper was cited as the cause of reduced metabolic activity.

Singh and McFeters (71, see reprint) performed in vitro studies on mice to determine the survival and virulence of chlorine and copper-injured Yersinia enterocolitica microorganisms. They found that fewer of the copper-damaged cells in orally-inoculated mice survived attack by gastric pH than did the chlorine-damaged cells. The virulence of the copper-damaged cells was therefore significantly lower than the chlorine-dama8ed cells, which were similar in virulence to undamaged controls.

The study conducted by Pyle, Broadaway, and McFeters (68) on the copper/silver/iodine system indicated that there were two populations of bacteria in the system exposed to iodine alone; the majority of the bacteria which were inactivated by the treatment and a small population that was resistant to it. This latter population was responsible for regrowth after the iodination was stopped. The bacteria exposed to the metal ions did not exhibit regrowth. The authors described this as possibly being due to different modes of action of the iodine and the metal ion systems.

Thurman and Gerba have reviewed the molecular mechanisms of copper and silver ion disinfection of bacteria and viruses (7?). They first considered how the transfer of metal ions across cell walls might occur as well as tactics employed by the cell to reduce the toxic effects of metal ions.

Three possible mechanisms have been proposed for inhibition of microorganisms by silver: interference with electron transport, binding to DNA, and interaction with the cell membrane. The formation of complexes with sulfhydryl groups can inactivate cell surface enzymes and interfere with respiration at the cell membrane.

Copper is known to attack respiratory enzymes in bacteria. presumably by binding to groups containing; sulfhydryl, amine, and carboxyl moeities. Copper is also thought to facilitate hydrolysis or nucleophilic displacement reactions in peptide chains or nucleic acids. Finally, copper is able to chelate with phosphate groups and this may result in the opening of double helices.

Although a number of studies exist which document changes in metabolism and inactivity of microorganisms which have been damaged by metal ions or halogens, there has not yet been a definitive work which links the changes caused by the metal ions or halogens on a molecular level. There is an even greater paucity of data on changes in metabolism and infectivity of microoroanisma damaged by synergistic combinations of metal ions and halogens.

Little is known or has been proposed about the changes occurring on the molecular level caused by metal ion/halogen combinations as well. The most probable molecular mechanism is a two-step mechanism proposed by Taylor (73) to explain how two different chemicals may result in more efficient disinfection when used together or sequentially. Kutz et al (59) proposed that attack by halogen on the cell walls may make them more permeable to charged metal ions which are normally ;excluded by healthy cells.

ADVANTAGES OF COMBINED COPPER/SILVER/HALOGEN DISINFECTANTS

There are a number of distinct advantages to using electrolytically generated copper and silver ions in combination with low levels of chlorine in place of the higher chlorine levels now used to disinfect swimming pools and spas. One advantage is that they give a fast inactivation rate, comparable to chlorine at the higher dose levels. This is due to the unique synergy between the two modes of disinfection.

The fact that a lower chlorine residual can be employed to achieve the same degree of disinfection in itself has many advantages. One advantage is that a lower chlorine residual should be easier to maintain. The rates of the decomposition reactions (4a,b) and (5a,b) · depend upon the square of the HOCl (or OCl') concentration. When a lower chlorine residual s able to be used, the rate of the decomposition reactions should be lower, thus making the lower residual easier to maintain. This should also minimize the use of chlorine stabilizers such as Cyanuric acid. The use of chlorine is also reduced and with it the acids and bases used to maintain the pool pH. Finally, labor associated with chlorine and pH monitoring and adjustment should be reduced.

Maintaining the chlorine residual at 0.4 ppm also results in lower eye, mucus membrane, and skin irritation and less "chlorine" odor in the pool area as well as the area surrounding indoor pools and spas.

Given the stability of the metal ion residual and the case of maintaining a lower chlorine residual, it should be easier to keep up with periods of bather overload than with a conventional pool chlorinating system. It should also be safer because microorganisms exposed to a combination of metal ion and chlorine disinfection tended to be less invasive than microorganisms exposed to chlorine only. Also. it should be more difficult for microorganisms to develop resistance to two different disinfection mechanisms instead of one. Finally, silver ion is easily adsorbed to surfaces and prevents the formation of biofilms (59).

Since chlorine still must be used in conjunction with ionizers, the formation of THM's and chloramines is unavoidable. However, since lower amounts of chlorine arc used and consumed in pools with ionizers, presumably there should be lower amounts of these compounds being formed. As more ionizers are being used in pool disinfection, no doubt studies will be conducted in the future to determine if this is the case.

BIBLIOGRAPHY

1. V.J. Cabclli. Vater Sci. Tcch. 21,13-21 (1989).

2. Statc of California Dcpt. of Hcalth Serviccs-Future California Morbidity Notc on Lc ionclIosis from a Decorative Fountain. Gcorge W. Ruthcrford, M.D., Chicf, Infcctious Discase Branch.

3. T.J. Dondcro ct al. N. Eng. J. Med. 302, 365-370 (1980).

4. C.B. F7iermans, G.E. Bettinger, and A.W. Frank. Water Res. I6, 903-909 (1982).

5. C.B. Flicrmans et al. Appl. Environ. Microbiol. 41, 9-16 (1981).

6. W. Habicht and H.E. Muller. Zentralblatt Bakteriol. Parasitenkd. InfektionlQ. Hyg. 186, 79-88 ( 1988).

7. C.M. Helms et al. Ann. Intern. Med. 99, 172-178 (1983).

8. J.B. Kurtz et al. J. Hyg. 88, 369-381 (198?).

9. G.K. vforris et al. Ann. Inter. vled. 90, 664-666 (1979).

10. P.W. Muraca, V.L. Wu, and J.E. Stout. J. Am. Water Works Assoc. 80, 78-86

cl9ss.

11. J.E. Stout, V.L. Yu, and M.G. Best. Appl. Environ. Microbiol. 49, 221-228 (1985).:?.

12. J.O. Tobin, R.A. Swann and C.L.R. Bartlett. Br. Med. J. 282, 515-517 (1981). .

13. R.lt. Wadowsky et al. Appl. Environ. Microbiol. 43, 1104-1110 (1982).

14. P.L. Seyfried. Am. J. Public Health. 75, 1071-1075 ( 1985).

15. D.J. D'Allessio et al. Am. J. Epidemiol. 113, 533-541 (1981).

16. H.B. Hawley et al. J. Am. Vtcd. Assoc. ??6, 33-36 (1973).

17. J.S. Koopman et al. Am. J. Epidemiol. 11, 163-17? (198?).

18. W.J. Martone et al. A. J. Epidemiol. 1 I l, ??9-237 (1980).

19. W.L. vlallmann. Am. J. Public fiealth, 5?, ?001-2008 (1962).

20. E.E. Gcldrcich. Am. Soc. Microbiol. Ncws. 47, 23-?7 (1981).

21. S. Pitlik, S.A. Bcrer, and D. Humincr. Rev. Infcct. Diseasc. 9, 54-62 (1987).

22. L.E. Withcrcll ct al. J. Environ. Hcalth. 46, 77-79 (1983).

23. Boston Globc, July 20,1994. p.4.

24. New York Timcs, July 20,1994. p. A l. .

25. Morbidity & Mortality Weckly Report, Auoust 12, 1994. p.574.

26. J.G. Jacanbelo, V.P. Olivcri, and K. Kawata. Mechanism of Inactivation of Microorganisms by Combined Chlorine. American Water Works Research Foundation. Denver, CO.

27. J. Carrell Morris in "Water Chlorination. Environmental Impact and Health Effects" Volume l. Roben L. Jolley (ed.) Ann Arbor Science Publishers Inc., Ann Arbor, Mich. (1979).

28. Roben J. Jolley and James H. Carpenter in "l.Vater Chlorination. Environmcntal Impact and Health Effects" Volume 4. Robert L. Jolley (ed.) Ann Arbor Scicnce PubIishers Inc., Ann Arbor, Mich. (1983).

29. J. Dawes. Desig n and Planning of Swimming Pools. CBI Publishing Co, Inc. Boston, VIA. (1979).

30. S.D. Worley and D.E. Williams. Critical Reviews of Environmental Control. 18, 133 (1988). .

31. C. Venkobachar, L. Iyenar, and A.V.S. Prabhakara Rao. Water Res. 1 l, 727- 729( 1977).

32. A.K. Camper and G.A. McFeters. Appl. Environ. Iicrobiol. 37, 633-641 (1979).

33. S.D. Roller, V.P. Olivieri and K. Kawata, Water Res. 14, 635-641 (1980).

34. M.V. LeChevallier, D.J. Schiemann, and G.A. IIcFeters. Appl. Environ. Microbiol 53. 138-1364 (1987).

35. S.I. Terzieva and G.A. McFeters. Int. J. Environ. Health Res. 2, 24-? (1992).

36. O. Sletten. J. Am. Water Works Assoc. December. 690-697 (1974).

37. R.B. Thurman and C.P. Gerba. Adv. in Appl. Microbiol. 33, 75-105 (1988).

38. R.B. Morris M.D. ct al. Am. J. of Public Health. 8?, 95-963 (199?).

39. J.A. Bccch ct al. Am. J. Public Health 70, 79-8? (1980).

40. J.A. Beech in "Chemistry of Water Rcuse" Vol. 1. W.J. Cooper (cd.) Ann Arbor Scicnce Publishers Inc., Ann Arbor, Mich. (1981).

41. H.S. Brown, D.R. Bishop and C.A. Rowan. Am. J. Public Health 74, 479-484 1984).

42. G. Agazzotti ct al. Io. Mod. 88, 39-54 (1987).

43. G. Aggazzotti et al. Arch. Environ. Health 45, 175-9 (1990).

44. J. Copaken in "Water Chlorination. Environmental Impact and Health Effects" vol. 6. Ann Arbor Science Publishers, Inc. Ann Arbor Mich. (1990).

45. P. Koss and K. Radiengaard, Z. Gesamte Hyg. Ihre Grenzgcb. 34, 231-2 (1988).

46. C.M. Feldstein, J. Rickabauh, and R.J. Miltner in "Water Chlorination. Environmental Impact and Health Effects" vol.5. Ann Arbor Science Publishers, Inc., Ann Arbor, Mich. (1985).

47. G.P. Fitzgerald and vI.E. DerVartanian, Appl. Microbiolocy 15, 504-9 (1967).

48. S.E. Holmes et al, J. Environ. Health, 51286-8 (1989).

49. M.T. Yahya, T.M. Straub, and C.P. Gerba, Can J. Microbiol. 38, 430-5 (1992).

50.. M.W. LcChevallier, C.D. Cawthon, and Ramon G. Lee, Appl. Environ. Microbiol. 54. 649-654 (1988).

51. i I.A. Clarke and D. Berman in "Disinfection, Sterilization, and Preservation". S.S. Block (ed.). Lea and Fabiger, Phila., Pa. (1983).

52. B. Chiswell and C.F. Wildsoet, bVater Sci. Technol. 21, 241-4 (1989).i. A. Luescher, Umwclttechnik 27, 5-7 (1993).

53. Showcase USA, 3rd Quarter, ?-4 (1990).

54. C.W. Chambers, C.M. Proctor, and P.W. Kabler. J. Am. Water Works Assoc. 54 208-216 (1962).

Copper and E. coli

The disease-causing bacterium E. coli 0157 is killed within hours of its contact with copper surfaces, according to preliminary studies by the Center for Applied Microbiology and Research in Wiltshire, England. Scientists also are investigating whether copper plumbing could reduce E. coli found in drinking water.