OZONE,  CHLORINE DIOXIDE,
          AND  CHLORAMINES AS
      ALTERNATIVES TO CHLORINE
          FOR  DISINFECTION OF
            DRINKING WATER
          STATE-OF-THE-ART
        WATER SUPPLY RESEARCH
 OFFICE OF RESEARCH AND DEVELOPMENT
U,S,  ENVIRONMENTAL PROTECTION AGENCY
       CINCINNATI, OHIO 45268

            NOVEMBER 1977

-------
         OZONE, CHLORINE DIOXIDE,

           AND CHLORAMINES AS

        ALTERNATIVES TO CHLORINE

           FOR DISINFECTION OF

             DRINKING WATER
           State-of-the-Art
         Water Supply Research
   Office of Research and Development
U.S. Environmental Protection Agency
        Cincinnati, Ohio 45268
            November 1977

-------
                                 TABLE OF CONTENTS
                                                                        Page
INTRODUCTION                                                             1
GENERATION AND USAGE OF ALTERNATE DISINFECTANTS                          3
     Ozone                                                               3
          Generation                                                     3
          Usage                                                          4
     Chlorine Dioxide                                                    6
          Generation in Aqueous Solution                                 6
          Usage                                                          7
               Chlorine Dioxide Use in the Unted States                  8
               Chlorine Dioxide Use in Canada                            8  .
               Chlorine Dioxide Use in Europe                            8
     Chloramines, Combined Chlorine, or Chloramination                   9
          Generation                                                     9
          Usage                                                          9
     Summary                                                            10
MICROBIOLOGICAL ASPECTS OF DISINFECTION                                 11
     Comparative Disinfection Efficiency                                12
     Chemical Basis for Disinfectant Activity                           14
     Comparative Resistance of Microorganisms                           15
     Protective Effects of Particulate Matter                           19
     Influence of Water Quality                                         20
     Stability During Water Distribution                                20
     Summary                                                            20
MEASUREMENT OF DISINFECTANT RESIDUALS                                   24
     Free Chlorine      '                                                25
     Ozone                                                              25
     Chlorine Dioxide                                                   25
     Chloramines                                                        26
          Free Chlorine with Chloramines                                26
          Chlorine with Chlorine Dioxide                                27
          Ozone with Chlorine                                           27
          Complex Combinations                                          27
     Summary                                                            29

-------
                               TABLE OF CONTENTS
                                  (Continued)

BY-PRODUCTS AND END PRODUCTS OF DISINFECTION                          30
   Chemistry                                                          30
      Free Chlorine                                                   30
      Ozone                                                           33
      Chlorine Dioxide                                                34
      Chloramines                                                     43
   Health Effects                                                     44
      Chlorine                                                        44
         Toxicology                                                   44
         Epidemiology                                                 47
      Ozone                                                           48
      Chlorine Dioxide                                                50
         Toxicity                                                     50
         Hemolytic Anemia                                             51
      Chloramines                                                     55
      Comparative Studies                                             56
   Summary                                                            57
COST OF DISINFECTION                                                  58
   General Considerations                                             58
   Basis of Cost Estimates                                            59
   Chlorine                                                           59
   Ozone                                                              61
   Chlorine Dioxide                                                   63
   Chloramines                                                        64
   Summary                                                            66
DISCUSSION                                                            68
SUMMARY                                                               73
ACKNOWLEDGMENTS                                                       73
REFERENCES                                                            74

-------
              OZONE, CHLORINE DIOXIDE AND CHLORAMINES AS ALTERNATIVES


                 TO CHLORINE FOR DISINFECTION OF DRINKING WATER


                                State-of-the-Art
Compiled by James M. Symons, Chief, Physical and Chemical Contaminants
Removal Branch (P&CCRB), Water Supply Research Division (WSRD),  Municipal
Environmental Research Laboratory (MERL)

Written by:

     J. Keith Carswell, Research Engineer, P&CCRB, WSRD, MERL
     Robert M. Clark, Systems Analyst, P&CCRB, WSRD, MERL
    . Paul Dorsey, Research Assistant, P&CCRB, WSRD, MERL
     Edwin E. Geldreich, Chief, Microbiological Treatment Branch (MTB),  WSRD, MERL
     W. Paul Heffernan, Research Pharmacologist, Laboratory Studies Division,
       Health Effects Research Laboratory (HERL)
     John C. Hoff, Research Virologist, MTB, WSRD, MERL
     0. Thomas Love, Jr., Research Engineer, P&CCRB, WSRD, MERL
     Leland J. McCabe, Research Epidemiologist, Field Studies Division,  HERL
     Alan A. Stevens, Research Chemist, P&CCRB, WSRD, MERL
All in the Office of Research and Development, U.S. Environmental Protection
Agency, Cincinnati, Ohio

                                INTRODUCTION
     Although free chlorine is a good disinfectant, the problems of the formation

of halogenated organics, particularly chloroform and related trihalomethanes,

during its use have caused the entire subject of drinking water disinfection

to be reviewed.  Three other disinfectants are used by various water utilities

throughout the world; ozone, chlorine dioxide, and chloramines.  If any

disinfectant or combination of disinfectants is to replace free chlorine as

the most commonly used disinfectant, several criteria must be met,  as follows:

     It must be easily generated and be in widespread use;

     It must be a good biocide;

     It must provide an easily measured residual;

     It must produce fewer undesirable by-products than does free chlorine; and

     It must be cost effective.

-------
                                       - 2 -
     The water supply research program, in support of a possible Agency




regulation of trihalomethane concentrations, is currently evaluating the




disinfectants ozone, chlorine dioxide, and chloramines as free chlorine




alternatives with respect  to these characteristics.  This document will




be a progress report on  this evaluation.  The report will be organized




by sections, covering the  individual criterion noted above.  Each section




will contain a summary review of the literature and a summary of research




findings to date (Fall 1977) based on in-house and extramural projects.




Because research is continuing in most areas, many of the studies discussed




are unfinished and will  be presented as progress reports.  Detailed




papers will be published in the technical literature on these studies as




they are completed.  In  total, however, this report will summarize the




Agency's knowledge on the  subject of disinfectant alternatives at this




writing.




     Although this paper focuses on the possibility of changing disinfectants




as a solution to the disinfection by-product problem, by-product formation




is as much a result of the organic content of the water being disinfected as




it is the result of chlorine being used.  Therefore, another, and possibly




preferable, option, which  is beyond the scope of this paper, would be to




reduce concentrations of disinfectant by-products by chlorinating water that




had been treated, possibly with granular adsorbants, to remove the natural




organic materials with which chlorine reacts.  This would hsrve the additional




benefit of reducing the  concentration of synthetic organics in finished




waters as well as limiting the biological nutrients being carried into




distribution systems.

-------
                                      - 3 -
               GENERATION AND USAGE OF ALTERNATE DISINFECTANTS



     Based on a review of the literature and a research grant sponsored

by the Water Supply Research Division (WSRD) the following information

is presented to demonstrate that the three alternate disinfectants

considered in this paper meet the first criterion of suitability as a

replacement for chlorination, that generation technology exists and that

their use is sufficiently widespread that they should be considered

"common practice."

OZONE

Generation

     Lawrence and Cappelli in their recent review article  list three

methods used for ozone generation:

     a.  silent electric discharge in air or oxygen

     b.  electrolysis of water

     c.  ultraviolet radiation of air or oxygen

Of these three, only (a) is used commercially because of its higher

generation efficiency (maximum of 5-6 weight percent ozone).

     In the usual generation process, low pressure, clean dry air (or

oxygen) is passed between large area electrodes separated by an air gap

and a dielectric (usually glass) across which an alternating potential

of 15,000 volts (or greater) is maintained.  In this type of ozone

generator, the production efficiency depends on the rate of gas flow,

the applied voltage and the cooling water (or air) temperature.
           2
     Mignot has reviewed in detail the numerous methods used for contacting

ozone with the liquid to be treated.  These methods include:

     a.  Liquid dispersion in an atmosphere of ozonized air.

     b.  Fine orifice injection of ozonized air at the bottom of a co-
          or countercurrent contact chamber, with or without contact
          media and/or performated septa.

-------
                                      - 4 -
     c.  An injector operating on the total flow of water to be treated
          to aspirate low pressure ozonized air into a contact chamber
          and create mixing turbulence.

     d.  Similar to (c), with the injector operating on a portion of
          the total water flow

     e.  Over-ozonation of a portion of the water to be treated with
          mixing with the remaining non-ozonated portion

     f.  Porous media injection of ozonated air into a single or multiple
          (series) contact chamber with co- or counter-current flow

     g.  Injection of ozonized air by porous diffusers in a pressurized
          contact chamber with recovery of the residual ozonized air for
          re-injection at the chamber entrance (the Torricelli system).

     h.  Turbine diffusion of ozonized air without recirculation of the water
          to be treated.

     i.  Same as (h), with recirculation of the water to be treated

     Selection of an appropriate method of contacting is a very important

aspect in the design of an ozone treatment system.

Usage
     Lawrence and Cappelli  have listed a number of common uses of ozone

in drinking water treatment.  They include:

     a.  Disinfection

     b.  Precipitation of iron and manganese

     c.  Destruction of sulfites

     d.  Destruction of many surfactants

     e.  Removal of turbidity

     f.  Oxidation of taste, odor and color producing compounds

     g.  Oxidation of other organic compounds

-------
                                      - 5 -
     Hill , in reviewing the development of water and wastewater ozonation,

lists several facts on the early use of ozone for drinking water treatment:

     a.  First permanent ozone installation — 1893 at Oudshoon, Holland
          (Rhine River)

     b.  Other early ozone installations — 1901 at Wiesbaden, Germany;
          1902 at Paderborn, Germany; 1906 at Nice, France

     c.  By 1916 there were 49 water treatment plants (26 in France)
          using ozone

     d.  By 1940, there were 119 plants, with 90 in France and 14 in Italy.


     The International Ozone Institute has summarized what was then (1976)
                                                                            4
known of the use of ozone for drinking water treatment in the United States.

This information includes:

     a.  Two water treatment plants are using ozone in operation.
          Whiting, Indiana began using ozone in 1941 for taste and odor
          control.  Strasburg, Pennsylvania has been using ozone since
          1973 for disinfection.

     b.  Monroe and Bay City, Michigan have ozonation facilities under
          construction.

     c.  New York City, Boston, Washington D.C., Louisville, and the State
          of Vermont have conducted research and/or pilot plant studies
          on ozonation in drinking water treatment.

     d.  Haverhill, Massachusetts and Laconia, New Hampshire have conducted
          pilot plant studies.

     e.  Vineland, New Jersey has received State approval to conduct a
          pilot-scale demonstration project

     f.  Boca Raton, Florida is considering an ozonation study.

     In 1976, the WSRD funded a Research Grant titled:  "Status of

Ozonation and Chlorine Dioxide Technologies for Treatment of Municipal

Water Supplies."   This project is a fact finding, state-of-the art

survey of municipal water treatment practices involving the use of ozone

and chlorine dioxide in the United States and selected foreign countries.

-------
                                       - 6 -







     Thus far this study has developed some preliminary information on




the use of ozone in other countries.  Nineteen water treatment plants




(18 in Quebec, 1 in the Northwest Territory) have been identified in




Canada.




     In Europe, Asia and Africa, a total of 1034 water treatment plants




using ozone have been identified through a search of the literature and




through contact with the manufacturers of ozone generation equipment.




The vast majority of these plants are in France (577) , with West Germany




(247) and Switzerland (135) each having substantial numbers.  Nineteen




other countries have so far been identified as having at least one ozone




facility.




CHLORINE DIOXIDE




Generation in Aqueous Solution




     Chlorine dioxide, like ozone, is usually generated and used on




site, although products referred to as "stabilized" chlorine dioxide are




available on the Market.  Chlorine dioxide  is usually generated in




aqueous solution for water treatment, as gaseous chlorine dioxide generation




is not well suited to this application.  White  notes that chlorine




dioxide would not be available for use in water or wastewater treatment,




if it were not for the availability of solid sodium chlorite.  This




chemical was first made availble in commercial quantities by Matheson




(now Olin-Matheson) in 1940.




     Two methods are commonly used to generate chlorine dioxide for




water treatment:




     a.  From chlorine and sodium chlorite




     C12 +  H20 —> HOC1 + HC1                                     Eq.(l)




     HOC1 + HC1 + 2NaCK>2 —> 2C102 + NaCl + H20                   Eq. (2)





     This method is commonly used where a gas chlorinator is available.

-------
                                         - 7 -



     b.  From sodium hypochlorite and sodium chlorite


     NaOCl + HC1 — > NaCl + HOC1                                 Eq.(3)
     HC1 + HOC1 + 2NaC102 — > 2C102 + 2NaCl + H20                Eq.(4)


     In either method, the addition of a mineral acid can be used to


control the reaction stoichiometry, and thus eliminate an excess of


either free chlorine (prevent trihalomethane formation) or chlorite


(toxicity problems) .


     In the case where a gas chlorinator is used, the chlorinator effluent


must be at pH 3.5 or lower, and contain at least 500 mg/£ chlorine, if

                                                           Q
complete reaction of the sodium chlorite is to be effected.   To achieve


this condition, the usual practice is to set the chlorine-technical


sodium chlorite ratio at 1:1 to provide excess chlorine.  If excess


chlorine is not desired, the chlorine- technical sodium chlorite ratio is


set at the theoretical 1:3 ratio and acid is added to maintain the pH at


3.5 or lower.


     Although not as common in practice, chlorine dioxide can also be


generated from sodium chlorite and acid.  The following equation illustrates

                                      9
this, using sulfuric acid as the acid.


     10 NaC102 + 5H2S04 — > 8 C102 + 5 Na2S04 + 2HC1 + 4 H20     Eq.(5)


This reaction has the advantage of producing "chlorine-free" chlorine dioxide


easily.


Usage


     White  reports the first known use of chlorine dioxide as a water


treatment process in the United States occurred at Niagara Falls, New


York in 1944. Chlorine dioxide was used to control phenolic tastes and


odors from industrial wastes as well as tastes and odors from algae and


decayed vegetation. The 1963 Inventory of Municipal Water Supplies


showed that only 8 of 11,590 water treatment plants in the Inventory


used chlorine dioxide.

-------
                                      - 8 -




     Information is still being gathered on the WSRD extra-mural project




and a final report is not due until early in 1978.  Some information on




the use of chlorine dioxide is available, however.  It should be considered




preliminary in nature at this writing (Fall 1977).




Chlorine Dioxide Use in the United States




          103 facilities have been identified as using or have used




          chlorine dioxide




               27 - Georgia




               24 - Ohio




               18 - Pennsylvania




               8 - Michigan




     Many of the treatment facilities in Georgia and Pennsylvania are




thought to be small ground water systems with well-head chlorine dioxide




disinfection prior to distribution.  Most of the United States facilities




are using chlorine dioxide for taste and odor control (phenols), for




iron and manganese oxidation or for final disinfection.




Chlorine Dioxide Use in Canada




     Ten facilities have been identified as using chlorine dioxide.  All




are in the Province of Ontario and all use chlorine dioxide for taste




and odor control.  Three of these plants are said to have "severe" taste




and odor problems.




Chlorine Dioxide Use in Europe




     In Europe, chlorine dioxide is used in several thousand water




treatment facilities, most of them of very small size.  Countries where




extensive use is known include:  West Germany, Switzerland, France, and




to a lesser extent, Belgium.  Italy and The Netherlands also are known




to have a small number of chlorine dioxide facilities.  At most locations,




chlorine dioxide is used for final disinfection prior to distribution.




The principal reason for its extensive use ir that it does noi: Impart a chlorinous

-------
                                     - 9 -
taste to the water.  When completed, the final report from the WSRD



research grant  will provide more extensive information on the use of



both ozone and chlorine dioxide in drinking water treatment.



CHLORAMINES, COMBINED CHLORINE, OR CHLORAMINATION



Generation



     Combining ammonia with chlorine to form chloramines has been variously



called the chloramine process, chloramination and combined residual



chlorination.  The usual goal in this water treatment process is to form



monochloramine according to the reaction:



     HOC1 + NH  —> NH Cl + H70 (monochloramine)              Eq.(6)
              J       £,      £,


     Reaction conditions to optimize production of monochloramine include:



     a.  A pH range of 7 to 8.



     b.  A chlorine-to-ammonia ratio of 5:1 (by weight), or less.  The



          natural free ammonia content of the water, if any, should be



          included in determining this ratio.  The preferred ratio,



          according  to White , is 3:1 (by weight), although widely



          varying ratios are found in practice.



     At higher chlorine-to-ammonia ratios, or at lower pH values, other



cahloramine species will form according to the reactions:



     NH2C1 + HOC1 —> NHC12 + H20    (dichloramine)          Eq. (7)



     NH2C1 + HOC1 —> NCI  + H20 (trichloramine, or nitrogen trichloride)   Eq.(8)



     The production of these latter species is related to the familiar



chlorination "breakpoint" phenomenon.  Their presence may contribute to



taste and odor problems in the finished water.

-------
                                    - 10 -

     Ammonia may be added  to the water before or after the addition of
chlorine.  In the former case, pre-ammoniation can prevent the formation
of tastes and odors resulting from the reaction of chlorine with phenols
or other substances.  In the latter, post-ammoniation permits a free
chlorine residual to exist within the treatment plant, followed by the
establishment of a lasting chloramine residual in the distribution
system.  The ammonia-chlorine process finds its widest use in this application,
according to White .
Usage
     One of the first small scale and subsequently full scale practical
applications of combined chlorine as a disinfectant in water treatment
was in Ottawa, Ontario  (Canada).  Joseph Race, chemist and bacteriologist
for the City of Ottawa  experimented with the material as a disinfectant
during the winter of 1915-1916 because "the price of bleach advanced to
extraordinary heights."   About the same time the City of Denver, Colorado
(USA) was successfully  applying a combination of chlorine and ammonia to
                                                        12
control nuisance organisms in their distribution system.    By 1918,
chloramines were being  used in the Catskill system of the water supply
for New York City because  in previous use chlorine had disasterous
                 13        7
effects on trout.    White further reports that chloramines were most
popular between 1929 to 1939 at which time their popularity decreased because
of the discovery of breakpoint chlorination along with the inability to
purchase ammonia during World War II.
     The 1963 Inventory of Municipal Water Supplies   shows only 308 of
the 11,590 water treatment plants in the Inventory using an ammonia-
chlorine process.   The current unpublished inventory lists about 40,000
water supplies, however no differentiation on the type of disinfectant
used is recorded.

-------
                                       - 11 -






SUMMARY



     These three disinfectants are judged to be sufficiently common in




practice to be seriously considered as alternates to chlorination.  Other




disinfectants such as silver,  iodine,  pH, heat, ionizing radiation, permanganate,




ultra-violet radiation,  bromine chloride, ferrate,  and so forth are recognized to




exist but will not be discussed in this paper as their use for treating




drinking water is minimal at the present time (Fall 1977).

-------
                                       - 12 -

                    MICROBIOLOGICAL ASPECTS OF DISINFECTION


     The primary reason  for  the use of disinfectants in potable water
treatment is to kill or  inactivate pathogenic microorganisms that may be
present.  The material presented in this section describes the relative
merits of chlorine, ozone, chlorine dioxide and chloramines as disinfectants
from the microbiological standpoint, the second basic criterion of an
acceptable disinfectant.
COMPARATIVE DISINFECTION EFFICIENCY
     In considering the  relative merits of the various chemicals used
for disinfection of potable  water that are reviewed in this paper, the
major factor used  to compare the different agents is their disinfection
efficiency.  This  may be expressed in terms of relative disinfection
concentrations needed to attain the same disinfection rate or relative
disinfection rates produced  by the same concentration of disinfecting
agent.  Because of the lack  of standardization of methods by which
disinfection research has been conducted, the variation in resistance
shown by different isolates  of the same microorganism species and variation
in resistance between species, only generalized statements about the
comparative efficiency of disinfectants can be made.  In this context
the disinfecting agents  under consideration may be ranked as follows in
order of decreasing efficiency .
            C102  (>) HOC1  (>)  OCr(>) NHC12  (» NH2C1       Eq. (9)
     ozone  (>)  chlorine  dioxide  (>) hypochlorous acid  (>) hypochlorite ion (>)
        dichloramine  (>) monochloramine
This order  is derived  from the results of numerous disinfection investigations
and provides a  general overview  of  the comparative efficiency of various
disinfectants for inactivation of microorganisms.

-------
                                   - 13 -



     A further complicating factor, particularly with attempts to rank



the free chlorine and combined chlorine disinfectants, results from



changes in chemical species that occurs with changing pH, temperature,



and concentration. For instance, at pH 9.5 - 10 the usual pH for investigation



of OC1  disinfection, HOC1 is present in low concentrations and because



of its high relative germicidal efficiency compared to OC1, some of the



inactivation measured may be caused by the HOC1 rather than OC1 .



Similarly, in the case of chloramines, an equilibrium state exists as



follows:


               <—

     NH  + HOC1—> NH Cl + H2 NHC12 + H20                      Eq.  (10)





     Thus, a low concentration of HOC1 may be present and may contribute



to chloramine disinfection activity observed.   This equilibrium is



further influenced by temperature, pH, the presence of other chemical species,



and disinfectant concentrations present. Because of the effect of increasing



temperature on the rate at which chemical reactions occur,  the overall



effect of increasing temperature is one of increasing disinfection



rates, although the effect may not be the same for different disinfectant



species.  For instance, the relative effective ratio of OC1  to HOC1 for



destruction of Endamoeba histolytica cysts was found to be 1 to 150 at



3°C and 1 to 300 at 23°C.  That is, HOC1 was 300 times as effective as



OC1~ at 23°C , but only 150 times as effective as OC1~ at 3°C.


                                                              14
     Although such effects usually are slight, Scarpino et al.   in



discussing some unusual disinfection results, speculated that a borate



buffer system used in one of their studies may have had a major effect



on the HOC1—>OC1 equilibrium, may have suppressed ionization, or may have



resulted in the formation of previously undescribed viricidal forms.



Using this buffer system, their resultc indicated that OCT was seven



times as effective as HOC1 against poliovirus t>pe

-------
                                     - 14 -
In the same buffer system, HOC1 was 50 times as effective as OC1  in




inactivating Escherichia  coli.  Because no further investigations of




this phenomenon have been published, the reason for these findings is




unknown.




     Rather than the linear ranking of Equation (9) shown above, division




of these disinfectants  into two groups based on their disinfection




efficiency is probably  a  better approach.  One group, consisting of




those agents that are considered very powerful disinfectants, includes




ozone, chlorine dioxide,  and HOC1.  The other group, consisting of




weaker disinfectants includes OC1  , monochloramine and dichloramine.




CHEMICAL BASIS FOR DISINFECTANT ACTIVITY




     Chemical disinfectant activity is dependent on contact between the




disinfecting agent and  the microorganism surface followed by reaction




with the surface and/or penetration of and reaction with vital internal




constituents resulting  in death or inactivation of the microorganism.




In the case of the disinfectants under consideration the reactions that




take place are oxidative  in nature.  Thus, oxidation potential is an




important factor with regard to disinfecting capabilities.  Not all




chemicals having a high oxidation potential are good disinfectants,




however.  For example,  Morris   pointed out that gaseous oxygen, 0_ has




a higher oxidation potential than many effective disinfectants.  The




statement of Peleg   "...substances that do not possess a high oxidation




potential will not be germicidally active. Conversely, chemical species




that have high oxidation  potential may [or may not] possess germicidal




potential", reiterates  this point.  Therefore, chemicals with a high




oxidation potential cannot be assumed a_ priori to be good disinfectants.

-------
                                      - 15 -





     Another factor involved in disinfectant activity relates to the




chemical dissociation of the disinfecting agent and the presence of a




charge on the germicidally active portion of the molecule.  An ion such




as OC1  is a relatively poor disinfectant because the presence of the




negative electrical charge prevents its approach to and diffusion through




cell membranes of microorganisms.  All three of the very effective




germicides, ozone, chlorine dioxide, and HOC1 have high oxidation potentials




and also are undissociated.  The germicidally reactive ozone chemical




species is not known, but it may be the OH (hydroxy radical) or the HO-




(hydroperoxyl radical).  Other species such as 0_  (ozonide radical),




0- (superoxide radical), and 0  (oxide radical) have been demonstrated,




but their role in germicidal activity is not known  .   Most experimental




results indicate that the disinfection efficiency of ozone varies with




pH and that it is somewhat more efficient at a lower pH (6.0) than at a




higher pH (10.0). This is probably because of the greater stability of




ozone in water at lower pH's.




COMPARATIVE RESISTANCE OF MICROORGANISMS




     Based on the results of many previous studies, the inherent resistance




of different kinds of microorganisms also may be ranked.    As in the




case of comparative ranking of disinfectants, the figures are not precise,




but do provide a general impression of relative resistance.  The rankings




in order of decreasing resistance are as follows:




     protozoan cysts > enteroviruses > enteric bacteria      Eq. (11)




     In a currently funded WSRD research grant at the University of




Illinois involving an intensive examination of the chlorine resistance




of 6 different enteroviruses, their resistance to HOC1 (pH 6.0) was




found to vary by a factor of five in the time required for a four log




reduction.

-------
                                       - 16 -
The time required for a four log reduction of these same viruses using OC1



 (pH 10.0) ranged from 10 to 150 times that required for equivalent


                  18
reduction by HOC1.    These results indicate that changing the pH had



different effects on the chlorine resistance of different viruses.



     Further indication of this phenomenon is shown in chlorine dioxide



disinfection studies.  In contrast to chlorine, chlorine dioxide does



not react with water to hydrolyze, although it does disproportionate to



CIO   (chlorite), and CIO ^(chlorate).   The chlorine dioxide molecule


                                                              19
remains in the same form over a pH range of at least 4 to 8.4.    The



rate at which E. coli is inactivated by chlorine dioxide increases,


                                               19
however, as the pH increases within this range.    A similar effect with


                                       20
polio virus was shown by Cronier et al.    They showed that chlorine



dioxide inactivated poliovirus 1 over  three times as last at pH 9.0 as



at pH 7.0.  Thus, both E. coli and poliovirus 1 are much more sensitive


                                                             21
to chlorine dioxide at a slightly alkaline pH.   Moffa et al.   in



studying the disinfection of storm water, found a diphasic inactivation



curve when using chlorine dioxide.  Although chlorite and chlorate



concentrations were not measured, they attributed the second slower



phase of inactivation to the disappearance of chlorine dioxide and the



appearance of chlorite.  They concluded therefore that chlorite had "—



weak bactericidal and viricidal properties—."



     Differences in disinfection resistance between different isolates


                                                          22
of the same virus type have also been shown.  Kelly et al.   showed that



the time required for a three log reduction of several different poliovirus



1 isolates varied from 3 to 6 minutes during exposure to 0.2 to 0.3 mg/Jl


                                                 23
of free chlorine.  Recent studies by Bates et al.   indicate that progeny



of poliovirus 1 repeatedly cycled through chlorine disinfection are

-------
                                       - 17 -






somewhat more resistant to chlorine than the original parent stock



virus.  The resistant progeny required about 2.5 times the length of



exposure required by the parent stock for inactivation to the same



level.



     The existence of extremely resistant polio 1 viruses has also been

                 r\ I

reported recently  .   At free chlorine levels of about 0.5 mg/£ and  pH 7.1,



these isolates showed a decline of about one log in 2 minutes with little,



if any, further decrease in titer after 30 minutes exposure.  Thus,



these viruses apparently show a complete resistance to chlorine for



upwards of 30 minutes, rather than merely a difference in time required



for complete destruction.  Isolates of these resistant viruses have been



obtained and further experiments are in progress in the laboratories of



Dr. Engelbrecht of the University of Illinois and Dr. Sharp of the



University of North Carolina, both funded by WSRD research grants.



     Aggregation or clumping of viruses has long been considered to be a



factor involved in increased viral resistance to disinfectants.  Until



recently, however, no evidence of this has been available.  Through a



series of WSRD research grants at the University of North Carolina,



evidence of this phenomenon and information on the factors involved in


                                                       25-29
virus aggregation and deaggregation have been obtained.       Although



the effects are complex and appear to be different with different viruses,



some general information has been developed from these studies as follows:



     1.  Although differences in halogen resistance of up to 300 fold



          between aggregated and dispersed viruses have been shown, the



          differences thus far shown are merely differences in rates of



          inactivation.  No evidence of complete resistance to inactivation



          because of aggregation has been found.

-------
                                        - 18 -
     2.   pH and metallic cations are important determinants with regard



          to aggregation.  Aggregation does not appear to be related to



          virus isoelectric point characteristics.



     3.   Induced aggregates of some viruses are quite stable although



          those of other viruses can be dispersed easily.  Natural virus



          aggregates appear to be more stable than induced aggregates.



     Of the most resistant forms of microorganisms of concern, protozoan



cysts, only Endamoeba histolytica has been extensively studied with



regard to disinfection.  Most of this research has been conducted using


                                            30
chlorine in various forms.  Newton and Jones   showed that ozone was a



more effective cysticidal agent than chlorine.  Although chlorine dioxide



would be assumed to be an effective cysticidal agent, research findings



on this point could not be found. The resistance of E. histolytica cysts



to chlorine is such that White  states, "The best protection against



this disease (amebiasis) is removal of the organisms from water supplies



by filtraiton.  They are large enough to be trapped by sand filters, and



require high doses of free residual chlorine (10 ppm) and long contact



time (one hour or more) to effect disinfection."   According to Palin ,



the range of free residual chlorine recommended by the National Research



Council for insuring cyst destruction within a 30 minute contact time



ranged from 2 mg/£ to 70 mg/£ depending on the pH and temperature of the



water being treated.  Despite this and the lack of universal filtration



of water supplies originating from source waters that may contain E.



histolytica cysts, waterborne outbreaks of amebiasis are rare in the



United States.



     Outbreaks of giardiasis, a disease caused by Giardia lamblia, a



protozoan that also forms cysts, have, however, been occurring with



increasing frequency in recent years.  In contrast to E. histolytica, a



method for determining viability of G. lamblia cysts has not yet been

-------
                                       - 19 -





developed.  Although they are assumed to be highly resistant to disinfection,



little is known about the resistance of these cysts to various disinfectants.



A WSRD funded research grant at the University of Oregon Medical School



in Portland on development of methods for determining viability is in



progress and current results indicate that a method for determining



cyst viability in vitro will soon be available for use in disinfection


                              32
studies on this microorganism.



PROTECTIVE EFFECTS OF PARTICULATE MATTER



     As indicated previously, for disinfection to be effective, contact



must occur between the disinfectant and the microorganisms to be inactivated



or killed.  Because source waters may contain a variety of inorganic and



organic particulates, some of which may originate as fecal wastes, and



because of the tendency of microorganisms, particularly viruses, to



adsorb to various kinds of particles, removal of particulate matter,



usually measured by light scattering, has been considered very important



during water tratment.  Although the major reason for removal of particulate



matter (turbidity) has been the concern about its interference with



disinfection, little direct evidence of such interference was available



until recently.



     Previously laboratory studies had established that enteric bacteria



and enteroviruses ingested by aquatic nematodes found in some water


                                                           33
supplies were protected against very high doses of chlorine   .   In



another instance, persistence of coliforms in a water supply system



containing a substantial level of residual chlorine was attributed to



ingestion and survival of the coliforms in small crustaceans present in


         34
the water  .   More recent studies indicate that virus adsorbed onto



surfaces of particles such as clay remain exposed to the disinfectant

                                                             •5 |- o f

and disinfection rates are affected only slightly, if at all.  '    Further,



viruses precipitated with aluminum phosphate, or flocculated with calcium

-------
                                   - 20 -
carbonate or alum or either unprotected or are protected to only a minor


       Q X	O Q

degree.        Viruses associated with cell debris, however, are well



protected from inactivation, showing survival for up to 50 minutes at


                                     37
HOC1 concentrations of 1.5 - 2.0 mg/Jl  .  Protection of coliforms associated



with primary sewage effluent solids .for at least 60 minutes at HOC1


                                              38
concentrations of 0.5 mg/S, also has been shown



      All of these studies have been conducted using free chlorine as



the disinfectant.  Similar studies sponsored by WSRD research grants are



in progress at the Unversity of Cincinnati using chlorine dioxide and



at the University of Maine using ozone and studies using chloramines are



planned. Based on the comparative disinfection efficiency of various



disinfectants, one might  speculate that the same order of efficiency



would be maintained in inactivating microorganisms associated with



particulate matter.  The  efficiency of ozone, in particular, is difficult



to predict because of its extremely reactive nature and consequent rapid



disappearance after addition to water.



INFLUENCE OF WATER QUALITY



     The organic content  of water influences the disinfectant demand of



water.  Waters with lower organic content require the application of



lower doses of disinfectant to achieve good disinfection.  Figures 1 and



2 show that the lower organic content of the effluent from granular



activated carbon beds resulted in a reduced requirement for disinfectant,



either ozone or chlorine  dioxide, when compared to a typical filter



effluent. This has important implications when considering the problem



of by-products discussed  later in this paper.



STABILITY DURING WATER DISTRIBUTION



     In addition to their disinfection efficiency as defined above,



another important consideration is the stability of these agents over a



period of time and their  ability to remain as residual disinfectants

-------
             - 21 -
cc

O
X

00


H
<

_l

E
O
O
UJ

<

o.

a
oc

o
z

H
co
          DUAL MEDIA EFFLUENT, pH=7.3
           FILTRATION/ADSORPTION

           GRANULAR ACTIVATED CARBON  EFFLUENT

              pH=7.9

             GAC AGE: 8 WEEKS IN SERVICE

              OZONE CONTACT TIME=6 MIN.
    0.1
                     0.2     0.3     0.4

                      APPLIED OZONE, mg/L
0.6
Fig. 1. POST DISINFECTION WITH OZONE

-------
                    - 22 -
IWUUU

1






1000(
CO
tr
i <
CIO 2 CONTACT TIME = 30 min.
pH = 7.0 - 8.1
, TEMPERATURE = 22 - 26°C
\
\
\
\
\
\
^ \
\ \
* JL.
\ \
\ \
A \ *
« \\ \ \
"* \V
< \ vW^DUAL MEDIA EFFLUENT
•^ I * »
^ I \ \\ _^ss^^ FILTRATION/ADSORPTION
^_ & \ \ i ^-^-^=j"-— ^
Z T .L-— =====¥*==::=^^^ GRANULAR ACTIVATED
§ 100 tlr^ \v CARBON EFFLUENT
0
LU
_i
O.
Q
CC
^t
0
z

CO
10










1
\ \ \\ GAC AGE: 24 WEEKS IN
\ \ \\ SERVICE
\ \ ^
' , \\
\ \ \
\ 1 \\
\\ s
\ \ *
- \\
\ \
\
\ \ \^ NONE DETECTED AT
\ \ ^^^\ 0.2 mg/l CIO 2 DOSE
i '. ^ 	 * \
\ \ — ' ^
\ ' \
; \ N
\ v
\ \ \
\ ^ i *, i i
    0.1     0.2     0.3     0.4     0.5
         APPLIED CHLORINE DIOXIDE, mg/l

Fig. 2. POST DISINFECTION WITH CHLORINE DIOXIDE

-------
                                    - 23 -




to prevent growth of microorganisms in storage and distribution systems



after initial disinfection during water treatment.  Although the major



concern regarding the regrowth of microorganisms is related more to



harmful effects on the distribution system and aesthetic effects such as



taste, odor and appearance rather than to potential health hazards,



regrowth is not a trivial problem.  Further, the maintenance of a biocidal



residual to the consumers tap not only keeps the system clean and protects



against some cross-connection contamination, but also its sudden disappearance



is a rapid indication of distribution system problems.  On the other



hand, the application of chlorine can depress the pH and aggravate corrosion



problems in some instances.



     The level of cross connection protection attained by maintenance of



a disinfectant residual has been the subject of much conjecture.  A WSRD



funded research grant in progress at the Johns Hopkins University is


                                   39
providing some information on this.     The results of batch studies in



which E. coli, polio 1 and f2 bacteriophage mixed with sewage were added



in proportions of up to 5 percent by volume to water containing various



levels and types of chlorine residual indicated that both free and



combined chlorine residual at equivalent concentrations eventually gave



comparable degrees-of inactivation of E. coli.  A free residual at pH 6,



however, provided inactivation in 2 minutes whereas an equivalent concentration



of combined residual required two hours to inactivate E. coli.  Neither



free residual nor combined residual at pH 8 consistently inactivated



polio virus or f2 bacteriophage to the lower sensitivity limit of the test.



     The distribution system problems associated with the use of combined



clorine residual or no residual have been documented in several instances.  '   '



In these cases the use of combined chlorine is characterized by an



initial satisfactory phase in which chloramine residuals are easily

-------
                                     - 24 -





maintained throughout the system and bacterial counts are very low.  Over



a period of years, however, problems may develop, including increased bacterial



counts, dropoff of chloramine residuals, increased taste and odor complaints, and



reduced main carrying capacity.  Conversion of the system to free chlorine residual



results initially in an increase of taste and odor problems resulting from sloughing



off of accumulated organic material and problems in maintaining a free chlorine



residual.  Following this phase, a free chlorine residual is established throughout


                                                                             40
the system, bacterial counts are lower and taste and odor complaints decline.



Of the disinfectants under consideration, a residual level can be maintained with



all except ozone, which, because of its high reactivity, disappears very rapidly.



SUMMARY



     In considering the overall biocidal or germicidal advantages and disadvantages



of these disinfecting agents as alternatives to free chlorine, which is currently



advocated, from a microbiological point of view chlorine dioxide would be best.



It is highly efficient, is most effective in the pH range in which water



treatment is practiced, and can be maintained as a residual disinfectant in



distribution systems.  Chloramines have disadvantages with respect to their



much lower disinfection efficiency, possible failure to completely inactivate



pathogens present, their ineffectiveness in providing a barrier to cross



connection contamination, and their failure to prevent regrowth in distribution



systems in some cases.  Ozone, although extremely efficient as a disinfectant,



cannot be maintained as a residual in distribution systems.  In the future, the



effectiveness  of these three alternate disinfectants and the influence of



particulates and mixing intensity on disinfection will continue to be studied



in both in-house and extramural bench-, and pilot- and field scale projects



sponsored by WSRD.

-------
                                          - 25 -



                  MEASUREMENT OF DISINFECTANT RESIDUALS



     In order to properly control the use of a disinfectant, the residual



concentration of the disinfectant must be easily measurable, and where



distribution system protection is needed, this residual must be maintained



and monitored, the third criterion of an acceptable alternate to chlorine.



FREE CHLORINE



     For measurement of free chlorine residuals (HOC1+OC1~+C12) the 14th


                           42
edition of Standard Methods   lists and describes in detail eight methods



of residual analysis: Two iodometric methods, Phenylarsine oxide —



Amperometric Titration, Stabilized Neutral Orthotolidine (SNORT), DPD



Ferrous Titrimetric, DPD Colorimetric, Leuco Crystal Violet, and Syringaldazine



(FACTS).  The relative usefulness of each method plus known possible



interferences in the analyses are described in the reference.  Each



method is based on the oxidizing capacity of aqueous chlorine and measures



in some quantitative way the oxidation products of the reaction of the



chlorine residual with the method reagent materials.



OZONE



     Because all of the alternate disinfectants under consideration are



also oxidizing agents (ozone, chlorine dioxide, chloramines) some version



of most of the above methods is applicable to the analysis of each of the


                                             42
alternates as well.  Indeed, Standard Methods   lists an iodide—iodine



titrimetric method as the quantitative approach to ozone residual measurement



where concentrations are over one mg/£.  An amperometrically determined


                                                   43
end point extends the lower range of this analysis.

-------
                                        - 26 -

CHLORINE DIOXIDE



     lodimetry, a modified  orthotolidine, and amperometric titration are


                                             42                     44
recommended for chlorine  dioxide measurements   .   Recently, Miltner


                          45  46               47                    48
reviewed the work of Palin   '   , Adams, et al.   and Hong and Rapson



and ultimately applied  the  DPD  titrimetric procedure to the measurement



of aqueous chlorine dioxide during his work describing the prevention of


                                                                   49
formation of trihalomethanes  by use of this alternate disinfectant.



CHLORAMINES


                                                                  42
     Amperometric titration is  considered a standard of comparison   for



free or chloramine residuals, although some of  the other chlorine methods



such as the DPD methods are also applicable to  the determination of



chloramines.



COMBINATION OF DISINFECTANTS



     Thus, the measurement  of chlorine and alternate disinfectant residuals



per se is not a particularly  difficult analytical task where considerations



are given to the interferences  and limitations  described in each of the



methods referenced.  The  measurement of individual disinfectants is only



slightly more complicated when  disinfectants are used in combination



with one another. Commonly  encountered examples of this are chloramines



in the presence of chlorine (chlorine used with ammonia or where ammonia



or organic amines are already present when chlorine is applied), chlorine



in the presence of chlorine dioxide (chlorine dioxide is often produced



by the action of excess chlorine on chlorite),  and chlorine in the



presence of ozone.  Even  these  three examples have relatively straightforward



analytical solutions:


                              42
Free Chlorine with Chloramines



     Free chlorine (HOC1, Cl- or OC1~) is measured by either of the DPD



methods or by amperometric  titration followed by sequential measurement



of iodine liberated by  the  reaction of the respective mono- and dichloramine



with added iodide under specified pH conditions.

-------
                                      - 27 -



Chlorine with Chlorine Dioxide


                                                                 44
     After a thorough investigation of published methods, Miltner   selected



the DPD approach as most reliable and convenient for measuring chlorine



dioxide in his work.  When DPD is used (either titrimetric or colorimetric)



two separate measurements are necessary to determine chlorine and chlorine


                                                                44
dioxide concentrations, and two approaches to this are possible.     Malonic



acid can be added to an aliquot of the sample to react with residual


                                                45 46
chlorine prior to a chlorine dioxide measurement  '   and then two DPD



measurements, one before and one after malonic acid addition, will yield



total chlorine dioxide-plus-chlorine and chlorine dioxide respectively.



Chlorine is obtained by difference.  The other approach is to purge the



chlorine dioxide from solution by gas stripping leaving the chlorine in


        48
solution  .   In this approach the chlorine is measured directly and the



chlorine dioxide determined by difference.  In either case the analysis



for these two species in simple combination is relatively uncomplicated.



Ozone with Chlorine


                                   43
     According to Dailey and Morrow  , phenyl arsine oxide—amperometric



titration of iodine liberated upon the reaction of both chlorine and



ozone with added iodide followed by a similar titration of a second



aliquot previously treated with ammonia will yield the concentrations of



total ozone-plus-chlorine and chlorine alone respecitvely.  The ozone



concentration is obtained by difference.  This approach depends upon the



reaction of ozone and chlorine with ammonia to form nitrate (unreactive)



and monochloramine (reactive with iodide) respectively.



Complex Combinations



     From the above, the conclusion might erroneously be reached that



the problems of disinfectant residual measurement are fairly well solved.

-------
                                       - 28 -






This is not necessarily the  case, however.  Where multiple disinfectants




are used, where more than  two or three active species need to be quantified,




or possibly where certain  organic materials are present (see next section),




the number of analytical steps and techniques as well as the calculations




necessary become much greater and the procedure more complex.  For



                44
example, Miltner   attempted to measure free chlorine, chlorine dioxide,




and all expected inorganic reaction products resulting from the action




of these species with the  organic materials in the source water.  The




goal was to generate a material balance of inorganic chlorine and oxychlorine




species to aid in assessment of toxicological hazard (to be discussed




later) from products (for  example, chlorite and chlorate) possibly




resulting from the widespread application of chlorine dioxide as a




disinfectant alternative to  chlorine.




     Miltner was able to use a combination of five procedures, two




required in triplicate, to construct a material balance of free chlorine,




chlorine dioxide, chlorite,  chlorate and chloride in a few simple cases




in rather dilute solutions.  Follow up work conducted over a wide concentration




range   indicated that many  problems with this total analysis still




exist, especially with the measurements of chlorate and the relatively




small contributions of chloride above background by the applied disinfectant.




     Additionally, because of the differences in disinfecting power of




each of the free chlorine  species, HOC1 and OC1  (see previous section)




distinguishing between them  under any given set of treatment conditions




rather than simply measuring the "free residual" might be desirable.




This represents a different  problem from that posed by the need to




measure oxychlorine species  of different oxidation states.  Johnson   has




developed a membrane electrode that claims promise, but it has not yet




been generally accepted as a practical tool in the drinking water industry.

-------
SUMMARY




     If such differentiation of oxidizing species as in the two examples




cited above are finally deemed to be unnecessary, disinfectant residual




measurements are likely to remain relatively simple when proper consideration




is given to potential interferences.  Future work by WSRD will focus on




the possible interferences of organic compounds when measuring chlorine




dioxide residuals.

-------
                  BY-PRODUCTS AND END PRODUCTS OF DISINFECTION



     The occurrence of  trihalomethanes in finished drinking water has



been demonstrated to be widespread and a direct result of chlorination


                                       52-54
practices at the water  treatment plant.       Because of the health



significance of these trihalomethanes (discussed later in this section)



an alternate disinfectant  is being sought.  The fourth criterion any



alternate must meet is  the creation of less undesirable by-products and



end-products than chlorine.



CHEMISTRY



Free Chlorine



     Chloroform results from the generalized reaction:



     Free Chlorine + "Precursor" —> Chloroform52"56          Eq.(12)



     This occurs to some extent in any water treatment plant where



chlorination for disinfection is practiced and natural organic (humic)



material is present to  act as precursor.  The reaction is not instantaneous



and occurs over a period of a few days until either chlorine or precursor



is exhausted.  In the presence of natural bromide, the reaction products



include some mixed-halogen trihalomethane species (bromodichloromethane,


                                    52
dibromochloromethane) and  bromoform.    This occurs in most chlorinated


              54
drinking water  , even  where bromide concentrations in the source water



are small. Iodine-containing species have also been observed  , presumably



because of the presence of natural iodide.  Because the chemical reactions



for formation of these  bromine- (and iodine-) containing trihalomethanes

                                                                         CO

are probably mechanistically similar to that for formation of chloroform,



the trihalomethanes, including chloroform, can be discussed as a group.

-------
                                     - 31 -






     Because trihalomethane formation is not an instantaneous reaction,



as long as a free chlorine residual is present, trihalomethane concentrations



will increase in the water as it flows through a water treatment plant



(unless removed during treatment) usually reaching some value higher



than that which would be observed if an analysis for trihalomethane



species was performed immediately after sampling at the first point of



chlorination.  (A more complete discussion of these concepts has been


                    59
presented elsewhere.  .) Further, the consumer is likely to receive



water with trihalomethane concentrations higher than those leaving the



plant because the reaction proceeds in the distribution system.  Additionally,



not only are the concentrations of trihalomethanes time dependent, but



the rate of reaction is dependent on pH, precursor concentration, nature



of precursor(s), temperature, and to tome degree, free chlorine concentration



early in the chlorination process.  '  '    Finally, the ratio of chloroform



to other trihalomethanes is highly dependent on the bromide content of


                 /• o f o

the source water.   '     Typical concentrations of trihalomethanes


                                                 54
formed range from < 1 yg/£ to a few hundred yg/£.



     In addition to the trihalomethanes, numerous other compounds have



been found in chlorinated waters and their presence attributed to the


                                                              64
chlorine applied under those circumstances.  For example, Rook   has



characterized more  than twenty by-products of chlorination of Rotterdam



storage reservoir water. Jolly   and Glaze   have reported the presence



of a large number of carbon-substituted chlorinated compounds in chlorinated



waste waters. Nitrogenous compounds react with chlorine during both



drinking water and wastewater disinfection to yield the nitrogen-substituted



chloramines. Currently, Morris of Harvard University has a WSRD Research



Grant to study the influence of pH and time on the reactions of chlorine with

-------
                                       - 32 -
nitrogenous compounds likely to be found in natural waters.  In addition



to the chlorine demand reactions, the production of chloroform in these



reactions is also being determined.  Phenols have long caused a problem



with odors in chlorinated waters by forming the highly odorous chlorophenols.



 Generally speaking, however, the concentrations of these compounds, if



formed upon chlorination of drinking water, are much lower in finished



drinking water than are the concentrations of the trihalomethanes at



locations where a free chlorine residual is maintained. Individual chlorinated



compounds other than the trihalomethanes are not generally found at



concentrations in finished drinking water higher than found inx the



corresponding source water.  Additional confirmation on this is being



sought through a thorough study of the occurrence of halogenated hydrocarbons



in the Passaic River watershed and the rate of change of these (increase



or decrease) concentrations through the water treatment process at



Little Falls, New Jersey.  The work is being carried out by J.V. Hunter



at Rutgers, The State University under a WSRD research grant.

                                                                     /TQ
     The recent findings concerning the carcinogenicity of chloroform   has



stimulated EPA efforts to aid in reducing human exposure from the drinking


                                              69
water source to these chlorination by-products  .  Any efforts to reduce



trihalomethane concentrations should be expected to result in a presumably



beneficial decrease in formation of the other chlorinated by-products.



For example, Sontheimer et al.   reported an increase in total organic



chlorine concentration, as concentrated by adsorption on ground granular



activated carbon and measured by a pyrolysis technique, of 186 yg/£ after



breakpoint chlorination of Ruhr River water, although the chloroform



concentration only rose 6 yg/£ and the total trihalomethane concentration



only increased 14 pg/£.  Insights as to the reduction in organo-chlorine



load obtained as a consequence of control cf f~~"" 1 on"-''li"~r ^ormation may

-------
                                     - 33 -






be gained as a result of understanding the mechanism of chlorination




reactions of humic substances leading to trihalomethane production.




This is currently under detailed review through a WSRD research grant to




R.F. Christman, et al., at the University of North Carolina.




Ozone




     In-house research by EPA's Water Supply Research Division   has




shown that disinfection level (<5 mg/£) ozone doses, under the conditions




tested, will not:




     a.  form trihalomethanes (i.e., oxidize chloride to chlorine with




          subsequent reaction with trihalomethane precursors)




     b.  reduce the concentrations of trihalomethanes already formed prior




          to ozonation




     c.  reduce the concentration of trihalomethane precursors as measured


                                                   59
          by the trihalomethane formation potential




     At the present time plans have not been made to attempt to' identify




other by-products from ozonation in-house, although Chian of the University




of Illinois has found that upon ozonation, 2-propanol was converted to




acetone that in turn was oxidized to acetic and oxalic acids.  Trace




amounts of formaldehyde and formic acid were also detected in the ozonated




acetone solution.  Further ozonation of acetic acid resulted in the




formation of glyoxylic acid that was oxidized readily to oxalic acid.




Extramural research is now being sponsored by the WSRD on ozone oxidation




by-products and end-products.  A research grant to Glaze at North Texas



State University is investigating the use of ozone with and without




ultraviolet radiation in drinking water treatment.  The removal of




selected compounds (trihalomethane precursors, trihalomethanes, polycyclic

-------
                                          - 34 -





aromatic hydrocarbons  (PAH's), polychlorinated biphenyls (PCB's) and



others) and the oxidation/photolysis products formed are being studied



over a 3-year period.  At  this time  (Fall 1977) preliminary results are



not available.



Chlorine Dioxide


                                     49
     Some results of work  by Miltner   in the WSKD laboratory to monitor



trihalomethane production  with time  after chlorine dioxide, chlorine,



and chlorine dioxide with  chlorine were separately applied to pilot



plant settled water are  shown in Figure 3.  The upper curve represents



the action of chlorine alone; the curve coincident with the abcissa



represents the application of chlorine dioxide alone; and the curve



between these two represents the action of chlorine to form trihalomethanes



in the presence of chlorine dioxide. These curves demonstrate two important



points:  (a) chlorine dioxide does not cause the formation of trihalomethanes



and (b) chlorine dioxide plus excess chlorine (as is often the case in



water treatment) results in the formation of lower concentrations of



trihalomethanes than does  the same amount of chlorine alone.  Therefore,



with consideration only  to minimizing trihalomethane formation, chlorine



dioxide is a possible alternative to chlorine as a disinfectant in water



treatment.  Although this  is the case, consideration must still be given



to other by-products that  may be formed by chlorine dioxide treatment.



     Considerable evidence exists that chlorine dioxide reacts with



organic material during  water treatment and therefore is likely to



produce organic by-products:

-------
       F.A. Cl = FREE AVAILABLE CHLORINE
       pH = 7.4
       T = 24°C
               / 1.5 mg/l F.A.CI
               f    y
                   1.3 mg/l CIO2 + 1.5 mg/l F.A.CI
                               CIO 2 ALONE
         10     20    30    40     50     60

                    CONTACT TIME, HOURS
70
80
FIGURE 3.  TRIHALOMETHANE FORMATION BY CIO 2  AND
          EXCESS FREE AVAILABLE CHLORINE,
          ERC PILOT PLANT SETTLED WATER
          (FROM MILTNER, R. J., 1976)O
                                                                          I
                                                                          OJ

-------
                                        - 36 -

     1.  Because chlorine dioxide is a good disinfectant (see earlier
section), some reaction  is  taking place between the cell components of
the organism and the chlorine dioxide.
     2.  Even though chlorine dioxide does not react with ammonia most
waters exhibit a chlorine dioxide demand similar to that of chlorine.
     3.  At applied chlorine dioxide concentrations higher than those
encountered in drinking  water treatment, identifiable by-products have
              72
been isolated.
     4.  Chlorine dioxide destroys phenolic compounds when the oxidant
is used  for taste and odor  control in water supplies.
     5.  Most importantly,  as was shown earlier, the presence of chlorine
dioxide  reduces the formation of trihalomethanes by chlorine.  This and
                                  49
other evidence obtained  by  Miltner   indicated that chlorine dioxide
reacts with natural humic acid.  This is not surprising because chlorine
                                                                  73
doxide is effective for  reducing color in drinking water supplies.
     Before an informed  decision can be made as to the relative desirability
of free  chlorine and chlorine dioxide with regard to the formation of
organic  by-products a detailed  investigation and identification of the
products formed during disinfection with both chlorine and chlorine
dioxide  was considered necessary and was initiated in the WSRD laboratory.
The possible organic by-products arising from the use of chlorine dioxide
as a disinfectant in drinking water was first considered on the basis of
the existing literature,followed by an on-going laboratory study to
determine whether extrapolations from the literature describing work
where concentrations of  oxidant and organic material were generally high
were valid.

-------
                                       - 37 -




     A b.rief review of the pertinent literature was recently presented


          74
by Stevens   although a much more extensive and complete review is


                    72
available elsewhere.     Briefly, the literature describes chlorinated



and non-chlorinated derivatives including acids, epoxides, quinones,



aldehydes, disulfides, and sulfonic acids, that are products of reactions



carried out under conditions vastly different from those experienced at



water treatment plants.


            74
     Stevens   reviewed only a few reaction types that were suspected to



be important in water treatment practice.  Considered as reacting species



were saturated aliphatic hydrocarbons, olefins, amines, and aromatic



compounds.  Based on the literature, chlorine dioxide would not be



expected to react with saturated aliphatic hydrocarbons in aqueous



media.  Chlorine dioxide was reported to react with olefins to produce a



variety of chlorinated and non-chlorinated products:  aldehydes, epoxides,



chlorohydrins, dichloro-derivatives, and chloro- and unsaturated-ketones.



Amines react with chlorine dioxide in a manner very different from that



of chlorine, chlorine dioxide does not produce chloramines.  In general,



the respective aldehydes are formed from amines in the order of reactivity:


                              72
tertiary > secondary > primary



     The best known reactions of aqueous chlorine with aromatic compounds



in the water treatment field are those that occur with phenols.    Chlorine



reacts rapidly with phenol to form mono-, di-, and tri-chloro derivatives.



These compounds are highly odorous and are slowly decomposed by excess



chlorine.  Other phenolics and substituted aromatics can also be chlorinated.



The formation of chlorophenols by chlorine treatment is one of the chief



reasons that chlorine dioxide has been used in drinking water disinfection



applications.

-------
                                        - 38 -





     According  to  the  literature, chlorine dioxide usually does not form



odorous compounds  with phenol, but through a complex mechanism, forms



the quinones and chloroquinones, and when in excess, oxalic and maleic


      72
acids.     Chlorine dioxide  treatment of phenols can cause chlorine



substitution or ring cleavage or both depending on the phenol reacted



and the conditions of  the reaction.  The chlorinated products will



generally be of different structure and are either less odorous or are



formed in much lower yield than those formed by chlorination.  For



example, vanillin  reacts at  pH 4 with chlorine dioxide to give the non-



chlorinated g-formymuconic acid monomethyl ester.   Veratryl alcohol,


                                       7 ? 76
however, produces  4,5-dichloroveratrole  '   when reacted with chlorine



dioxide, which is  one  of the products expected from chlorination of



veratryl alcohol.



     Some of the more  detailed investigations of the reactions of chlorine



dioxide with phenols (ring cleavage or ring retention with or without



chlorine substitution) have  been accomplished by Glabisz and Paluch at


                                             78
the Polytechnic University,  Szczecin, Poland.    Glabisz  states



that, at least at  concentrations of one mg/& and above, the character of



the reaction products  of phenols falls into two groups.  The first is



the group in which the ring  structure is retained and the end products



are quinones.  This group is made up of para-dihydric and monohydric



phenols that are not para substituted.  The second group is characterized



by those phenols which undergo ring cleavage to give carboxylic acids as



end products.  Examples include para alkyl phenols and ortho or meta



dihydric phenols.

-------
                                     - 39 -
     In general, monohydric phenols reacting with chlorine dioxide undergo



chlorination along with oxidation,  and those of the first group form the


                                                  78
chloroquinones, as well as chlorophenols.   Glabisz   considers this to be



somewhat similar to the reaction of chlorine with these phenols.   Although



chlorine dioxide tends to favor oxidation over chlorination,  the relative



amount of chlorinated versus oxidized products depends on the relative



amounts of both chlorine dioxide used and phenols present.  Excess chlorine



dioxide favors oxidation.



     Because of the potential for undesirable by-product formation resulting



from chlorine dioxide disinfection of drinking water, an in-house investigation



was begun at the WSRD Laboratory to determine if by-products of the type



predicted by the literature, where reactions described were carried out at



generally higher concentrations, would prevail in the drinking water



disinfecting conditions.  This work is being carried out in two phases:



     1.  The search of purgeable compound gas chromatographic data (for



example from trihalomethane analyses) for differences between chlorine dioxide



treated- and untreated waters.



     2.  Development and use of a more elaborate analytical scheme to detect



products of more diverse nature, specifically those expected from reactions



of phenolic compounds.



     The semi-quantitative results of the first phase have already been


                 74
briefly described   where C0 through C0 aldehydes were noted to increase in
                           /          o


concentration after treatment of a natural water with chlorine dioxide.



In that work, no other dramatic differences were observed between treated



and untreated samples with regard to compounds amenable to the purge-and-trap


      *                 u-     i  •       54,57,79
type of gas chromatographic analysis used.

-------
                                        - 40 -





     Phenol was selected as  the model compound for the beginning of the



second phase primarily because of the polyphenolic nature of humic


         80
materials   (see Figure 4)  (trihalomethane precursors) that make up a



large fraction of the organic material present in natural waters where



trihalomethane formation is  a problem.  This work is not yet complete,



but chlorophenols have been  produced at high (5/4) molar ratios of



phenol to chlorine dioxide.  Lower ratios (5/14, 1/14) do not produce



chlorophenols, but hydroquinone was isolated from all reaction mixtures.



Other organic by-products such as oxalic and maleic acids, and 2,6- and



2,5-dichloro-p-benzoquinone  were not immediately identifiable, although



total organic carbon concentration data indicate that the phenol is not



completely converted to carbon dioxide.  To date (Fall 1977) no gas



chromatographable compounds  in this category have been identified in



chlorine dioxide-treated natural waters, and humic- and fulvic acid



solutions that were not present in the untreated sample (detection



limits estimated to be in the range of 5 - 10 ug/& as phenol) .



     Some circumstantial evidence now seems to indicate formation of



poly quinone-like material from phenol as well as from at least one


                                                                       42
source of aqueous humic material, however.  The DPD colorimetric method



was used to determine the chlorine dioxide demand of a highly colored



natural water.  An interference equivalent to approximately 0.3 mg/5, of



chlorine dioxide was observed below the point on the demand curve where



residual concentration began to rise as dose increased. A similar effect



was observed after chlorine  dioxide treatment of phenol solutions (see



above). The compound p-benzoquinone, a model for possible organic byproducts



of chlorine dioxide reactions and the oxidized form of hydroquinone (see



above) was also found to produce a color with DPD without chlorine dioxide treatment,



so an alternative test was sought for the determination of chlorine



dioxide that might not be subject to quinone interference.  The FACTS

-------
 COOH
OH
                                        OCH
OH  O
                       OH  HO^OH
                                               N
FIGURE 4. A PROPOSED HUMIC STRUCTURE (FROM
         CHRISTMAN & GHASSEMI,)^

-------
                                       - 42 -

                                                                        42
test, using syringaldazine with spectrophotometric measurement at 530 nm

was found to be a suitable replacement for DPD.  This reagent did not give a

color with p-benzoquinone.  In a chlorine dioxide demand determination on

a second colored water, however, a water similar to but from a different source

than that used previously, the syringaldazine method gave a higher blank (reagents +

water, no chlorine dioxide) reading than the DPD method.  Although the curves

compared well when background was subtracted, neither curve showed the 0.3 mg/£

"false positive" seen with the previous sample.

     A chlorine dioxide demand test on a second sample of the first highly

colored raw water gave results similar to the first test reported above

and again showed "false positive" concentrations below the "breakpoint"

of the demand curve without the DPD test.  Chlorite was found to be major

inorganic product of chlorine dioxide treatment of this water (see Table I) .

                                   TABLE I

            Chlorine Dioxide Treatment of a Colored Natural Water
CIO  Dose     CIO  Residual       CIO   Cone.     Contact Time  23 Hour.
mg/K             mg/£              mg/x,            Hours        Chlorine
                                                                Demand
                                                                 mg/£
13.13            1.19              6.33             23           11.94
The presence of chlorite did not cause this apparent chlorine dioxide residual

using the DPD method.  The same effect has not been observed on other raw

water sources or treated humic- or fulvic acid solutions.  This general

problem requires considerably more work and will be the subject of a future

more detailed report.

-------
                                       - 43 -




     Inorganic by-products may also be of concern.   As noted above



evidence obtained at the WSRD laboratory indicates  that a large portion



of the applied chlorine dioxide remains as chlorite rather than chloride



although accurately determining what percentage of  this results from



simple disproportionation in contrast to being a reduction product from



organic reactions (see previous section on measurement of disinfectant



residuals) has not been  possible.



Chloramines


                                                                   54
     During the 1975 National Organics Reconnaissance Survey (NORS)   10 of



the 80 utilities sampled disinfected with chloramines.  The sum of the



concentrations of the trihalomethanes found, expressed as ETHM, in the



finished water of these supplies ranged from 1 to 81 yg/£ with an average



of 19 yg/£.  For comparison, those utilities that were using breakpoint



chlorination had  THM concentrations ranging from 1 to 472 yg/£ in the



finished water with an average concnetration of 72  yg/£.  A similar


                                  Q I

study conducted in Ontario, Canada   on 48 water utilities included 10



that were disinfecting with combined chlorine, but  the report does not



contain sufficient details on treatment practices to assess the impact



of combined chlorine on trihalomethanes.  Stevens et al.   in a laboratory



experiment, applied both free and combined chlorine for varying contact



times to samples of untreated Ohio River water.  Figure 5 shows the



ETHM formation in the chlorinated sample reaching 160 yg/£ after 72



hours, yet the companion sample having combined chlorine formed only 16



yg/&  ETHM during the experiment.


                                                             54
     The reason some of the water utilities in the  NORS study   using



chloramines had relatively high  ETHM's is probably because of free



chlorine being used prior to the addition of ammonia.  For example, in



the Kansas City Water Works the  ETHM concentration at the time of the


                 54
NORS was 34 yg/£    This is not surprising because  free chlorine is

-------
                                    - 44a -





dosed about 5 hours before the ammonia is added.  Another example of


                               82
this effect is shown by Tuepker   for the St. Louis County Water Company.



There, like Kansas City, the water is softened and chlorinated with



approximately an eight hour contact time before ammonia is added.  In



that situation the chloroform concentration averaged 49 yg/£ over a 5



month sampling period.


                                83
     According to Lambert et al.   trihalomethane production begins only



when monochloramine, and presumably dichloramine, is destroyed by free



chlorine. Further, once water is ammoniated and the disinfectant exists



as combined chlorine the trihalomethane reaction essentially stops.



This is shown by the plateau on the combined chlorine curve in Figure 5.



Finally, the trihalomethane levels reaching the consumer should be



similar to the trihalomethane concentrations .leaving the treatment plant

                                                           Q O

when combined chlorine is used as the disinfectant. Tuepker   has shown



this to be the case where he reported no measurable differences between



the trihalomethane concentrations leaving the treatment plant and those



measured in the distribution system.  As noted previously, free chlorine,



on the other hand,  continues  to react with precursors in the distribution



system and consumers in outlying reaches can receive water with  trihalomethanes



considerably higher than the treatment plant effluent concentrations.



The use of combined chlorine during water treatment is currently being



investigated in a WSRD Research Grant to Professor Leland Harms, South



Dakota Schoo,l of Mines and Technology at the Huron, South Dakota Water


               84
Treatment Plant   and at the Louisville Water Treatment plant as part of


                                                                85
the EPA project with the Ohio River Valley Sanitation Commission  .



HEALTH EFFECTS



Chlorine



Toxicology



     Concern over the occurrence of chloroform and other halogenated

-------
                       - 44b -
180 -
               COMBINED CHLORINE
                       --

            24        48        72
              CONTACT TIME, HOURS

 FIGURE 5: COMPARISON OF FREE AND COMBINED
          CHLORINE ON THE  FORMATION OF
          TRIHALOMETHANES IN OHIO RIVER
           (AFTER STEVENS, ET

-------
                                     - 45 -

hydrocarbons in drinking water, primarily as a result of chlorine
disinfection, has precipitated investigations to evaluate the consequence
of exposure to these compounds on human health.
            i
     Chloroform has been used medicinally for anestheia, as a cough
suppressant, a carminative, a flavoring agent, and as a local irritant
in liniments.  Chronic ingestion of doses of 23-37 mg/K have been reported
to produce some reversible hepatotoxic effects while ingestion of 0.3 to
                                                                 o/:
0.96 mg/K for 1 to 5 years resulted in no observed hepatotoxicity
     Chloroform is rapidly absorbed from the gastrointestinal tract and
other trihalomethanes may be expected to be absorbed and metabolized in
a similar manner.  Human exposure to chloroform'may result from its
presence in ambient air, from industrial exposure and in drinking water.
A Threshold Limit Value (TLV) of 10 ppm in air had been established to
               87      '
protect workers   and long term clinical studies using a cough suppressant
containing chloroform noted some reversible hepatotoxicity when about 2
                                                           88
grams of chloroform had been consumed per day for 10 years.    A short
term, few animal study with mice gave an indication that chloroform
                                      89
might be carcinogenic in that species.
     In mid-1977 the National Institute of Occupational Safety and Health
issued a "Revised Recommended Chloroform Standard" (unnumbered and undated)
reducing the TLV to 2 ppm and its use in drugs, cosmetics and food
                                                             90
packaging has been banned by the Food and Drug Administration   because
the potential risk associated with its use are greater than its benefits.
      89
water.
     Laboratory studies have produced LDc-f. toxicity data following acute
oral exposure to chloroform at levels of 120 to 1350 mg/K in rats, male
                              91 92
mice and male and female dogs.  '    No teratogenicity was determined in an oral
                                     93
exposure study with rats and rabbits.    Evidence of the animal

-------
                                     - 46 -
carcinogenicity of chloroform has been confirmed by several studies that
have been extensively reviewed  in the Report of the National Academy of
                                                  94
Sciences (NAS) titled "Drinking Water and Health."     They have found
chloroform capable of producing malignant and metastatic neoplasms in at
least one feeding study  in mice and has produced tumors in both rats and
mice in other studies.   Utilizing the standard protocol for carcinogenic
testing, a dose response relationship was found for epithelia tumors of
the kidneys and renal pelvis in the rat and hepatocellular carcinomas in
                                                f Q
mice in studies by the National Cancer Institute  , and the latency
period for the carcinogenic effect decreased as the dose increased. Another
     95
study   showed hepatocellular carcinoma in femal rats and one strain of
male mice.   These studies established that chloroform is carcinogenic
to the animals under those experimental conditions and therefore, presents
a potential carcinogenic risk to humans.
                                                         94
     The National Academy's Safe Drinking Water Committee   calculated
the upper 95% confidence estimate of risk at 3 x 10   at an average two
liters consumption of water with 20 yg/£ of chloroform for a lifetime.
(one excess case of cancer for  every 33,333 persons exposed for a lifetime).
The NAS suggested that strict criteria be applied when limits for chloroform
in drinking water are established.                The use of maximum
tolerable dose in the toxicological studies and the mathematical extrapolation
to human population effects have an air of unrealness, but are based on
a philosophy of safety using what data are available.
     Before an alternative disinfectant for drinking water can be considered
for use, the potential toxicity of not only the disinfectant itself, but
also the potential toxicity of  the by-products, both inorganic and
organic, of the reactions between the disinfectant and water constituents,
must be evaluated.  For  example, sodium hypochlorite was reported to be
a weak base substitution mutagen in Salmonella typhimurium.

-------
                                       - 47 -
Epidemiology


                                                            54
     When the first nationwide data were available from NORS   on chloroform



concentrations, they were compared with cancer mortality (all sites) in



the cities served by the supplies sampled and a significant correlation



noted. These data have continued to be analyzed in more sophisticated



ways.  An association with bladder cancer and the trihalomethanes other


                                          97
than chloroform is the most solid finding.



     Data on water quality for carrying out epidemiological studies have



been lacking and other classifications of exposures have had to be used



in order to proceed with an evaluation of the problem.  The studies of


                              98
the Environmental Defense Fund   comparing Parishes (counties) were the


                                                 99
first to create a public interest. A recent paper   used this technique



of comparing mortality of counties served mostly by surface or ground



water in Ohio.  A more elaborate study of this type    was conducted by



the University of North Carolina for all the counties in the Ohio River



Valley.  These studies found that some aspects of cancer death rates



were higher in counties where most of the people used surface water as



compared to counties where most of the people used ground water.  The



assumption in these studies was that surface waters are more heavily



contaminated than ground waters or that more chloroform is produced when



surface waters are chlorinated.



     Another general classification can be made of drinking water —



chlorinated or not chlorinated.  Three studies have used this dichotomy



as a basis for comparison.  The Johns Hopkins study    covered a small



population for which many variables were controlled and only obtained


                                                                      102
suggestive results.  The University of California at Los Angeles study



demonstrated the extreme mobility of people in that area and was not



definitive.  The Columbia University study    had a population of adequate

-------
                                      -  48  -
size without excess mobility and  uncovered  some  interesting results.  Males



living in  the  chlorinated  drinking water  areas of Erie, Rensselaer, and



Schenectady Counties  and females  living in  the chlorinated drinking water



areas of Erie  and  Schenectady Counties are  at a  greater risk of gastrointestinal



and urinary tract  cancer mortality than are individuals living in non-



chlorinated drinking  water areas.



     The epidemiological studies  are  suggestive  that the toxicological studies



are reasonable, but may have in this  case underestimated the cancer risk.  Studies



with more  definitive  water quality data and a more rigorous design are now



being conducted.



Ozone



     Ozone is  a powerful oxidant  and  as such, can be expected to oxidize



many compounds present in  surface waters.   Whether these oxidation products



are more harmful or less harmful  than the reaction products of chlorine is



not known  at this  time (Fall,  1977).  Numerous studies, summarized below,



have been  performed to evaluate the potential toxicity of ozonated water and



wastewater, the majority of such  studies  concerning themselves with the



formation  of carcinogenic,  mutagenic, and teratogenic substances from the



ozonolysis of  organic contaminants.



     Comparative toxicity  studies  of  ozonated and chlorinated waters with



a defined  chemical composition of  organic materials such as pesticides or



detergents indicated  chlorine treated water exhibited a higher toxicity



than ozone treated water,  but the  greater toxicity to the cell cultur system



was speculated to  be  because of the residual chlorine rather than oxidative


    ,     104
products.

-------
                                    - 49 -





     Cytochrome P-450 reduction, as an indication of liver necrosis, was



evidenced in one strain of mouse subjected to intraperitoneal injections



of the products of ozonation, but was not apparent after injections with



the products of chlorination, however, the results were reversed with a


                          104
different strain of mouse.     In studies conducted to determine if



ozone would decrease the cytotoxicity of a synthetic military hospital



wastewater containing methanol, ethanol, isopropanol, acetone and acetic



acid, ozone produced products were more toxic to test cells than non-



ozonated waste.     Reversion of histidine dependent mutant strains of



Salmonella typhimurium by the products of ozonation of organic compounds



bearing various functional groups showed that ethanol and a mixture of



ethanol, acetaldehyde, and hydrogen peroxide were mutagenic and the



ozonation of benzidine resulted in a form that no longer required metabolic



activation for mutagenicity.  Positive results were not obtained in



similar bioassays for samples of unconcentrated secondary treated municipal



effluent that were a) untreated, b) chlorinated, c) chlorinated/dechlorinated,



d) chlorobrominated, or e) ozonated.



     Another study currently (Fall 1977) being conducted by Dr. Gruener



in progress at the Hebrew University in Jerusalem showed these preliminary



results on reused water. A reverse osmosis concentrate of reused water



was ozonated and then tested in the Salmonella mutagenicity test.  The



same water sample was checked before ozonation and found to be negative.



Water ozonated for 15 minutes was mutagenic in the two strains used:



TA100 (base pair substitution) and TA98 (frameshift).  No activation by



microsomal fraction was needed to cause mutations in TA100.   In the



strain TA98, although an effect existed in the absence of the liver



fraction (about three times the control value), a four-fold

-------
                                      - 50 -





increase in the number of colonies occurred compared to the control without the



liver fraction in the presence of phenobarbital induced liver fractions;  the



number of colonies was 9-10  times higher in the presence of methylcholanthrene



induced- and Aroclor 1254 induced liver fractions.  Clearly, the toxicity



problems are so complex  that it cannot be stated at this time if ozone



is better or worse than  chlorine from the point of view of by-product



formation.



Chlorine Dioxide



     Although chlorine dioxide possesses many desirable characteristics



as a disinfectant it is, none-the-less, a powerful oxidant capable of



oxidizing and chlorinating many compounds with which it comes in contact.



The properties of the reaction by-products as well as chlorite and



chlorate ion (end products), - the reduction products of chlorine dioxide



oxidation, are being thoroughly evaluated by the Health Effects Research



Laboratory (HERL) both in-house and  through extramural research .



Toxicity



     Considerable disagreement exists among literature sources concerning



the toxicity of chlorine dioxide     and the majority of reports are



concerned with exposure  in  air, principally from the paper pulp industry,



describing symptoms of local contact irritation and two cases of illness


                                                                    108
including one fatality from exposure to 19 ppm inside a bleach tank.



     The effects, on rats,  of chronic administration of chlorine dioxide


                 109
in drinking water    showed  no discernable toxic effects at 10 mg/Ji or



less over a two year period  although rats consuming 100 mg/£ showed a



higher mortality rate at the end of  the two year period.  Enger



reported that laboratory workers, consuming one liter of water containing



0.7 mg chlorine dioxide, experienced a feeling of weakness and gastrointestinal

-------
                                      - 51 -





troubles. Musil    supposed that the symptoms were caused by:  a)  the



chlorine dioxide itself, b) the final products from the reaction between



chlorine dioxide and substances in the water, or c) unreacted chlorite



from the generation of the chlorine dioxide.



     The toxicity of chlorite ion arising from the reaction of chlorine



dioxide and organic matter or the disproportionation of chlorine dioxide



in water, similarly, is not wall understood.  One of the reported advantages



of chlorine dioxide disinfection is the ability'to maintain a residual


                           112                                 113
in the distribution system.     Conversely, Myhrstad and Samdal    found



that the chlorine dioxide dosed was completely converted to chlorite



within 35 minutes in water from a system using microstraining and ozonation



for treatment.  The potential for chlorite ion to be present in water



exists and its toxicity must be evaluated.



     Studies have determined the LD _ of chlorite ion, for the rat, to



be 140 mg/K  1 and for sodium chlorite, 182 mg/K   ; 208 mg/£ and 50.6mg/£



for bluegill and rainbow trout, respectively    and for mallard ducks,



100 mg/K.     Rats exposed to 8 mg/£ chlorite or less in drinking water


                                                        109
had no discernible toxic effects over a two year period.



     Hopf    proposed that the toxicity of chlorite might be approximately



the same as chlorate and reported symptoms of a "weak furry feeling" on



the tongues of subjects who drank sodium chlorite concentrations of more



than 0.3 mg/£.



Hemolytic Anemia



     The propensity of certain chemicals and drugs to induce hemolytic



anemia has been a subject of much interest for almost a century.  The



majority of these substances are aromatic compounds containing amino-,



nitro-, or hydroxy groups, although a number of inorganic compounds such

-------
                                      - 52 -




as hydroxylamines, nitrites, chloramines and chlorates are also active.



Such compounds usually  cause red cell injury without evidence of specific



toxicity to other cells or  tissues.  The hemolytic process is characterized



by two distinctive features in  the affected red cells; a) the appearance



of brownish or greenish derivatives of hemoglobin, including methemoglobin



and, b) the formation of water-insoluble stainable granules called Heinz


                            118
Bodies within the red cell.



     Methemoglobin is an oxidation product of the normal blood pigment,



hemoglobin, in which the ferrous iron is oxidized to the ferric state



and is unable to combine reversibly with oxygen and carbon dioxide,



causing a shift in the  oxygen dissociation curve thereby hindering the



transfer of oxygen from the blood to the tissues.  Abnormal amounts of



methemoglobin will result in a  functional anemia with cyanosis.     Musil



claimed that chlorite ion,  as an oxidizing agent, carries the oxidation



of hemoglobin to methemoglobin  in vivo.   He recommended a concentration



of 0.0 mg chlorite per  liter because of the hazard of methemoglobinemia



in nursing babies. He further recommended that prior to the use of



chlorine dioxide, other methods of treatment, such as activated carbon,



should be considered.   Based on these observations, the Norwegian Health



Authority has recommended the absence of chlorite in drinking water.



     The repeated administration of methemoglobin producing chemicals



may cause hemolytic anemia, the exact mechanism of which is unknown as



the chemicals themselves may not be directly hemolytic.  Data dervied



from experiments conducted  at HERL indicated that in vitro, chlorite and



chlorine dioxide will oxidize hemoglobin to methemoglobin and, in the



intact human red cell,  oxidizes hemoglobin with corresponding increases in



the incidence of spicule cells  of the acanthrocyte and stromatocyte



type, morphological abnormalities that are associated with hemolytic



disorders, after certain "threshold" levels are exceeded.

-------
                                      - 53 -





     Continuous monitoring of the methemoglobin formed in cats from a



single oral dose of sodium chlorite resulted in a dose response that



indicated that chlorite was absorbed into the blood within 10 minutes,



reached a methemoglobinemic maximum in 1-1/2 to 2 hours and was reduced



at a rate of 16 percent per hour.



     Although methemoglobin formation from a single dose of a compound



does not necessarily indicate toxicity, an increasing degree of in vivo



lysis of red cells was noted at doses above 20 mg/K and cats given 500



mg/£ chlorite in water exhibited a sharp decrease in hemoglobin content



and packed cell volume after 2 weeks exposure, although methemoglobin



was not detected.  Similarly, rats given chlorite in water for 30 days



resulted in significant (p<0.05) increases in red cell count at doses of



0.9 to 1.6 mg/K/day; decreases in hemoglobin and packed cell volume at



doses of 11-16 mg/K/day; and decreased hemoglobin, packed cell volumes



and red cell count at doses of 28-50 mg/K/day.  Although methemoglobin



was not detected in any of these animals, probably because the "threshold"



level was not exceeded, a single dose of 50 mg/K resulted in excess of



50 percent of the animal's hemoglobin being oxidized to methemoglobin



within 30 minutes.



     Exposure of erythrocytes to mild oxidative stress converts hemoglobin



to methemoglobin, a reaction that is reversible because metabolic pathways



for the reduction of methemoglobin is present in normal erythrocytes.



Further oxidative stress, probably involving the oxidation of -SH groups



of the globin, causes irreversible damage with the formation of mixed



disulfides between glutathione and the hemoglobin.  Additional damage to



the erythrocyte membrane from further exposure to oxidative stress has


                                                              120
been postulated to lead to the destruction of the erythorocyte



In the HERL laboratory, the erythrocyte survival rate of cats drinking



100 mg/& chlorite in water (1.6 mg/K/day) was found to be reduced by

-------
                                 - 54 -
23.6 percent and the survival rate in cats drinking 500 mg/£, (5 mg/K/day)



was reduced 27.2 percent,  as compared to control drinking chlorinated



tap water, (p<0.05).


                                                       121
     Conversely, in other  experiments conducted by HERL    in which an



assessment of the effects  of chlorine dioxide in drinking water on the



hematopoietic system of African Green Monkey was made by exposing male



and female monkeys to  20 and 200 mg/£ chlorine dioxide (avg. 1.72 and



8.96 mg/K/day, respectively) for 13 weeks, no statistical deviation from



normal hemoglobin, hematocrit, red cell count, reticulocyte, methemoglobin



or osmotic fragility values occurred.  A series of other hematological



parameters, for example, Heinz Bodies, differential count, red cell and



leukocyte morphology were  within normal limits and serum bilirubin and



serum enzymes monitored throughout the study showed no abnormalities or



statistically significant  trends, indicating a stable homeostatic response.



     Animal response data  is difficult, at best, to extrapolate to the



human.  Species differences exist in the response to methemoglobinizing



agents.  The cat, for  example, is generally recognized as the species



most susceptible to methemoglobinemia, and is followed by man, dog,



rat , rabbit and monkey in decreasing order of susceptibility to aromatic


                              122
amine methemoglobin formation.     Segments of the human population also



possess variations in  methemoglobin susceptibility and sensitivity to



oxidative stress to the hematopoietic system.  Hemolytic anemia is



frequently a serious problem among patients undergoing long-term hemodialysis.


                                                                     123
Newborn infants are all susceptible to drug induced hemolytic anemias    caused



by an abnormality that persists for only a week or two.  Genetically



altered hemoglobins may confer abnormal sensitivity to nitrites and



other oxidizing agents that cause methemoglobinemia.  The abnormal



pigment, hemoglobin H, is  unusually sensitive to oxidation.  The life span

-------
                                      - 55 -


of these affected erythrocytes is about 40 days instead of the usual 120



days.



     The enzyme deficienty that results in familial idiopathic methemoglobinemia



has been known for a long time.  Individuals with this condition are



deficient in the ability to reduce methemoglobin.  Studies have shown



that erythrocytes with decreased glucose-6-phosphate dehyrogenase activity,



an enzyme deficiency present in 13 percent of U.S. Black males, are more


                                       124
vulnerable to oxidant induced hemolysis    which may begin with precipitation


                              125
of hemoglobin within the cell.      To continue the research in this



area,  studies are being initiated to determine, first, if animals deficient



in glucose-6-phosphate dehydrogenase activity are more susceptible to



oxidative stress caused by exposure to chlorite and chlorine dioxide in



water.



     In addition to these controlled laboratory studies, HERL is presently



investigating the feasibility of conducting epidemiological studies on



populations exposed to chlorine dioxide in their drinking water.  Although



numerous treatment plants use chlorine dioxide for periodic taste and



odor control, few use it consistently or in concentrations exceeding



1-2 mg/£.  Appropriate sites are being sought at this time.  Also,



a community water supply in Massachusetts utilized chlorine dioxide



exclusively during World War II.  A retrospective mortality study in



which the rates and causes of death and the incidence of neonatal jaundice



for this community could be compared with similar areas utilizing conventional



chlorine disinfection is being considered.



Chloramines



     Only two reports of combined chlorine toxicity were found in the



literature. Chloramines have been found responsible for two epidemics of



acute hemolytic anemia characterized by Heinz Bodies in dialyzed uremic

         -i o/:

patients.     The effected patients had undergone dialysis in two hospitals



that had recently adopted a reverse osmosis technique for purifying dialysis

-------
                                      - 56 -



water.   Patients in a third hospital,  which utilized activated  carbon



filtered water,  were not affected.   Also monochloramine  has  been shown



to  be a weak mutagen when reversion of trpC to trp  in B.  subtilis was


                  127
used  as an assay.



.Comparative Studies



      A clinical  trial is being formulated that will assess the  human



tolerance to each of the proposed alternate disinfectants, including



chlorine, and some of the end products such as chlorite  and  chlorate.



These trials will initially be conducted on normal healthy male subjects,



followed by controlled exposure to  levels of chemicals comparable to



those used in water treatment practices to individuals possessing some



of  the deficiencies that may render them more susceptible  to oxidative



stress than the  normal human population. The results of  these studies



will  determine the absorption, distribution, and metabolism  of  these



agents and provide a thorough assessment of human  tolerance  and effects,



if  any.



      The comparative toxicity of by-products from  the three  proposed



alternative disinfectants and chlorine is being evaluated  by HERL in a



project in which organic concentrates  in aqueous solution  obtained by a



100-fold reverse osmosis concentration technique,  from undisinfected



coagulated, settled and dual-media  filtered Ohio River water and different



samples of similar dual-media filtered water disinfected with chlorine,



chlorine dioxide,  ozone, or chloramine will be tested for  cytotoxicity



in  both mamallian cells in culture  and bacterial cells,  bacterial mutagenesis,



rat mortality, carcinogenic initiation/promotion,  effects  on DNA binding,



and on Xenobiotic metabolic systems.   In the first test, the concentrates



from  the four disinfectants were negative for bacterial  mutagenesis and



rat mortality and only the concentrate from the system disinfected with



chloramines showed any significant  bacterial cytotoxicity.   At  this time



(Fall,  1977)  the carcinogenic initiation/promotion studies are  just starting.

-------
                                    - 57 -
SUMMARY




     At the present time (Fall 1977) chlorine is known to produce trihalomethanes




and possibly other chlorinated by-products.   Ozone, chlorine dioxide, or




chloramines on the other hand, either do not cause trihalomethane formation,




or cause reduced quantities of trihalomethane to be produced.  Evidence




is growing, however, that as strong oxidants, both ozone and chlorine




dioxide do produce non-trihalomethane by-products, and inorganic end




products in the case of chlorine dioxide.  Chloramines have not yet been




studied in this regard. As yet, however, none of the organic by-products




of ozone or chlorine dioxide have been demonstrated to be formed at




concentration levels comparable to the trihalomethanes.  Both in-house




and extramural research is continuing in the area.




     Chloroform has been demonstrated to be a carcinogen in some animal




tests and epidemiological studies support the concept of some adverse




effects from chlorination.  Similar data are not currently (Fall, 1977)




available for ozone, chlorine dioxide and chloramines.   Preliminary




studies both by EPA and others have not shown any definitive dramatic




adverse effects from the three proposed alternates, but the completion




of on-going research is needed to add confidence for the continued use




of these alternatives.  One possible result of these studies could be




the recommendation of a maximum acceptable dose for any or all of these




disinfectants.

-------
                                          - 58 -






                               COST OF DISINFECTION




     If all other factors are equal, drinking water treatment unit processes




should be selected on a minimum cost basis.  In the case where the least




expensive unit process has an unacceptable risk, however, a higher cost




process may actually have a more favorable benefit/cost ratio than a minimum




cost process and might be selected.  The fifth criterion of a disinfectant




alternate to chlorine should be cost-effectiveness.  The costs of the alternate




disinfectants are discussed below.




GENERAL CONSIDERATIONS




     The costs in this section are intended for the development of




planning estimates only and not for the preparation of bid documents or




detailed cost estimates. Exact capital and operating costs are highly




variable from location to location within the United States, even for




plants of the same size and design.  Variables — such as local costs of




land, materials and labor; state or regional differences in building




codes; and existing facilities suitable for modification — may accentuate




the differences in treatment costs for similar plants to reduce chloroform




and other trihalomethanes in drinking water.




     These cost data are presented in such a way as to enable the planner




to make adjustments to the reported costs when local information is




available. For example, operation and maintenance costs can be reduced




if the delivered cost of chemicals is less than the costs upon which the




estimates were based.  A cost index is used to provide a baseline for




projecting costs and for estimating escalation because of inflation.

-------
                                      - 59 -


BASIS OF COST ESTIMATES

     Much of the basic information utilized in this report was obtained from

data developed by the Systems and Economic Evaluation Section (SEES) of EPA's
                             -i o o -I on
Wastewater Research Division.   '     For example, computations for chlorine

costs were performed using a computer program developed by SEES, but operational

modifications were assumed in the analysis to reflect conditions unique

to water supply.

     Costs reported as Capital Costs include:

          a.  construction for site preparation,

          b.  plant construction,

          c.  legal, fiscal and administrative service,

          d.  interest during construction, and

          e.  start-up costs.

     Costs reported as Operation and Maintenance Costs include:

          a.  chemical costs,

          b.  labor costs, and

          c.  operation and maintenance costs, such as utilities, annual

          replacement of expendable items, etc.

CHLORINE

     In this analysis, a number of variables have been evaluated as to

their effect on the cost of chlorination.  Baseline or standardized

values for a set of design parameters were assumed and the cost

of chlorination was calculated.   The parameters and their associated

"standardized" levels are shown in Table II.

-------
                                       - 60 -


                                     TABLE II

                 Design Parametes  for  the Cost of Chlorination
     Design Pa ramet e r  (Variable)

     Chlorine Dose
     Chlorine Contact  Time
     Construction Cost Index
     Cost of Chlorine
     Wholesale Price Index
     Direct Hourly Wage Rate
     Amortization Interest Rate
     Amortization Period
     Capacity Factor
                                Level

                             2 mg/1
                             20 min.
                             273.8
                             300 $/ton
                             194.6
                             $5.58/hr
                               7%
                             20 yr.
                             70 percent
     Assuming design flow rates of 1, 5, 10, 100 and 150 mgd, at an

operating to design ratio of 0.7 and the design parameter values in

Table II, the unit costs for chlorination are shown in Table III.

                                    TABLE III

                      Unit Chlorination Costs (j5/1000 gal)*
                                 Design Capacity
Item
1 mgd
5 mgd
10 mgd
100 mgd    150 mgd
Capital Cost 2.19
Operating Cost 1.06
0.88
0.56
0.62
0.46
0.26 0.24
0.32 0.31
Total Cost
3.25
1.44
1.08
0.58
0.55
*0perating costs are calculated on 70% on design capacity.

-------
                                     - 61 -
OZONE




     The standardized design parameters listed in Table IV were used as




the basis for analysis.   Because of increased biocidal activity the




disinfectant dose of ozone was assumed to be one-half the chlorine dose




used in Table II.




                                  TABLE IV




                    Design Parameters for Ozonation Costs









              Design Parameters (Variables)           Level




              Ozone dosage                           1 mg/£




              Ozone contact time                     20 min.




              Electrical Power Cost                  $0.03/kw-hr




              Construction cost index                273.8




              Cost of Oxygen                         $0.046/lb.




              Wholesale  Price Index                  194.6




              Direct Hourly Wage Rate                $5.58/hr




              Amortization Interest Rate                7%




              Amortization Period                    20 yr




              Capacity Factor                        70 percent




     Costs based on these parameters have been calculated for  ozone generated




from both air and oxygen.  These costs are shown in Table V and VI.

-------
                                          -  62  -





                                        TABLE V




        Unit Costs  for Disinfection with Ozone - Air  (j^/lOOO  gal.)







                                               Design  Capacity
Item
Capital Cost
Operating Cost
Total Cost
1
2
2
5
mgd
.90
.75
.56
5
1
1
2
mgd
.36
.05
.41
10
1.
0.
1.
mgd
11
77
88
100 mgd
0
0
1
.76
.40
.16
150
0
0
1
.73
.38
.11
mgd



Operating costs are  based on 70%  of  capacity.





                                     TABLE VI





      Unit Costs  for Disinfecting with Ozone - Oxygen (<£/1000  gal)







                                               Design Capacity
Item
1 mgd 5 mgd 10 mgd 100 mgd
150 mgd
Capital Cost                4.46       1.50       1.08         0.61         0.58




Operating Cost              2.87       1.17       0.88         0.52         0.49









Total Cost                  7.33       2.67       1.96         1.13         1.07

-------
                                        - 63 -
CHLORINE DIOXIDE

     In this cost analysis, the dose of chlorine dioxide was assumed to

be half the dosage of chlorine to achieve equivalent disinfection results.

Therefore, 1 mg/£ of chlorine dioxide would achieve disinfection results

equivalent to those achieved by 2 mg/£ of chlorine, see Table II.  If

0.5 mg/£ of chlorine is combined with 1.6 mg/£ of technical grade sodium

chlorite, a 1 mg/£ dosage of chlorine dioxide will result.  The equation

below shows this relationship (assuming 80 percent pure NaClO^ yields

1.3 mg/£ of reactive material):
2 NaC10
                    — > C102 + 2 NaCl
Mol.Wt. (181)   (71)    (135)     (117)

       1.3 mg/£  0.5 mg/£ 0.9 mg/£ 0.8 mg/£

Therefore, the chlorine feeding system and contact basins were estimated

for a 0.5 mg/£ dosage of chlorine with the rest of the standardized

values fixed at the levels shown in Table VII.

                                      TABLE VII

               Design Parameters for the Cost of Chlorine Dioxide
     Design Parameters (Variable)                  Level

     Chlorine Dose                                 0.5 mg/£

     Contact Time                                  20 min.

     Cost of Sodium Chlorite                       $0.70/lb

     Chlorine Cost                                 $300/ton

     Construction Cost Index                       273.8

     Ratio of Sodium Chlorite Concentration to
       Chlorine Concentration                       3.2

     Wholesale Price Index                         194.6

     Direct Hourly Wage Rate                       $5.56/hr

     Amortization Interest Rate                    7%

     Amortization Period                           20 yrs.

     Capacity Factor                               70 percent

-------
                                      - 64 -





                                    TABLE VIII




         Unit Cost of Disinfecting with Chlorine Dioxide (^/lOOO gal.)*










                                            Design Capacity
Item
Capital Cost
Operating Cost
1 mgd
1.90
1.55
5 mgd
0.76
1.18
10 mgd
0.51
1.12
100 mgd
0.22
1.03
150 mgd
0.20
1.02
Total Cost             3.45      1.90       1.63        1.25       1.22




*0perating Costs are based on 70% of design capacity.









CHLORAMINES




     The standardized levels for cost estimation are shown in Table IX.




Assuming design flows of  1, 5, 10, 100 and 150 mgd and an operating to design




ratio of 0.7 and the design parameter values shown in Table IX, the unit




costs for chloramines are shown in Table X.

-------
                                     - 65 -






                                   TABLE IX




                    Design Parameters for Chloramine Costs




          Design Parameters (Variable)             Level




          Chlorine Dose                            0.5 mg/£




          Contact Time                             20 min.




          Construction Cost Index                  273.8




          Cost of Chlorine                         $300/ton




          Cost of Ammonia                          $200/ton




          Ratio of Chlorine to Ammonia Cone.       5.0




          Wholesale Price Index                    104.6




          Direct Hourly Wage Rate                  $5.58/hour




          Amortization Rate                        7%




          Amortization Period                      20 yrs.




          Capacity Factor                          70 percent




                                    TABLE X




                    Unit Chloramine Costs (^/lOOO gal.)*




                                             Design Capacity
Item
Capital Cost
Operating Cost
1 mgd
1.70
0.63
5 mgd
0.62
0.25
10 mgd
0.42
0.19
100 mgd 150 mgd
0.17 0.15
0.10 0.10
Total Cost             2.33        0.87         0.61       0.27       0.25




*0perating costs are calculated on 70% of design capacity.

-------
                                        - 66 -
SUMMARY




     Unit costs for five disinfection processes have been developed and




are summarized below in Table XI.  Costs for chlorination, chlorine dioxide and




ozonation are developed on  the assumption of equivalent disinfecting capability.




The costs for chloramines are calculated on the assumption that the values




shown in Table IX yield 3 mg/2, of monochloromine, which is not equivalent in




disinfecting power to  the other disinfectants.  Therefore to obtain equivalent




biocidal activity chloramines must be used with some other disinfectant,




such as chlorination or in  high quality waters where high biocidal activity




is not needed.




                                  TABLE XI




            Summary of Unit Costs for Disinfectants (^/lOOO gal.)




                                             Design Capacity
Item 1 mgd
Chlorination
2 mg/£ 3.25
Cost of Ozone - Air
1 mg/£ 5.65
Cost of Ozone - Oxygen
1 mg/£ 7.33
Chlorine Dioxide
1 mg/fi, 3.45
Cost of Chloramine
3 mg/£* 2.33
5 mgd 10 mgd 100 mgd 150 mgd
1.44 1.08 0.58 0.55

2.41 1.88 1.16 1.11

2.67 1.96 1.13 1.07

1.90 1.63 1.25 1.22

0.87 0.61 0.27 0.25
*Not equivalent  to  the  other disinfectants in biocidal activity.

-------
                                         - 67 -
     To change from one disinfectant to another represents a minimal investment.




This is illustrated by Table XII.  The data in Table XII is derived from




the assumption that a typical family of 4 consumers would use 117 thousand




gallons of water in the home per year (80 gal/capita/day - a generous




figure).   Using data from Table XI an annual cost for disinfection per family




is presented in Table XII, based on system size.




                                  TABLE XII




Cost to a Family of Four in Dollars/Year for the Disinfection Unit Process




                                           Design Capacity




Item                  1 mgd      5 mgd       10 mgd      100 mgd     150 mgd
Chlorination
Ozonation -
Ozonation -

Air
Oxygen
Chlorine Dioxide
Chloramines*

$3.
$6.
$8.
$4.
$2.
80/yr
61/yr
58/yr
04/yr
73/yr
$1.
$2.
$3.
$2.
$1.
68/yr
82/yr
12/yr
22/yr
02/yr
$1.
$2.
$2.
$1.
$0.
26/yr
20/yr
29 /yr
91/yr
71/yr
$0.
$1.
$1.
$1.
$0.
68/yr
36/yr
32/yr
46/yr
32/yr
$0.
$1.
$1.
$1.
$0.
28/yr
29/yr
25/yr
43/yr
29/yr
*Not equivalent in biocidal activity to the other disinfectants




     Therefore the annual cost per family for any disinfectant is small, and




the cost increment for changing disinfectants is minimal.

-------
                                     - 68 -




                                  DISCUSSION




     Choosing the optimum or ideal  disinfectant is a difficult task as



each of the candidates discussed  in this paper has advantages and disadvantages:



See Table XIII.  Other disinfectants are currently (Fall 1977) in minor



use and were not discussed in this  paper.  In this country, historically, the



disinfectant of choice was chlorine.  Chorine was inexpensive and very



effective in controlling  waterborne disease and although it sometimes



imparted taste and odor to water, it did not produce any obvious effect



on people's health.  Subsequent studies showed humans had a high tolerance


                   130
to chlorine itself.     Later chloramines were substituted in some cases



to prevent taste and odor,  although some biocidal activity was sacrified.



Outside of the United States the  disinfectants ozone and chlorine dioxide



were widely used.  Although slightly more expensive, they were effective,



did not impart a taste and odor to  treated water, and did not cause any



obvious effects on the health of  consumers.  In practice, the disinfectant



dosage was often less than with chlorine, which reduced the additional cost.



     Currently (Fall 1977)  the situation is changing.  Improved analytic



techniques for organic chemicals  have shown that chlorine produces at least



one carcinogenic by-product,  and  epidemiological studies indicate some



adverse effect on health  may be caused by drinking chlorinated water.



Therefore, the U.S. Environmental Protection Agency is considering establishment



of limits for the concentrations  of disinfection by-products, particularly



trihalomethanes.  Should  such regulations be promugulated, they will cause



changes in drinking water treatment practices in this country.

-------
                                       - 69 -
     The concentrations of trihalomethanes can be controlled either by




changing disinfectants such that less trihalomethanes are produced or




by treating the water, for example with granular absorbants, such that chlorine




could still be used, but trihalomethane concentrations would be reduced.  Under




these latter circumstances, the benefits of chlorination could be maintained




without the corresponding undesirable organic by-products being produced




and no alternate disinfectant would be required.  This approach would have




the additional benefit of removing synthetic organics found in raw water




as well as controlling concentrations of biological nutrients entering




the distribution system.  Another treatment approach would be removal of




the trihalomethanes themselves by treatment.




     The details of the final regulation, if promulgated, will influence




which of the above choices are available as alternate treatment procedures.




If fairly high concentrations of trihalomethanes are permitted, free chlorine




can continue to be used in many circumstances, possibly followed by




ammonia to form chloramines.  If the permissible concentrations of trihalomethanes




are low, either now or in the future, the use of chlorine will be limited




only to waters with low concentrations of precursors, either naturally or




after treatment, and alternate disinfectants or treatment for removal of




trihalomethanes will have more widespread use.  Further, if in an attempt to




limit the formation of all disinfection by-products, the regulation were to




contain a total organic carbon concentration or disinfectant demand limit,




chlorine could have a continuing role in disinfection in some instances.




     Considering only the disinfectants themselves, their major characteristics




are presented in Table XIII.

-------
                                             - 70 -

                                           TABLE XIII

                             SUMMARY OF DISINFECTANT CHARACTERISTICS
                      Chlorine
     Use &     Widespread  in U.S.
     Genera-   Off-site
     tion      generation
                                           Ozone
                                     Widespread outside
                                     U.S.   On-site
                                     generation
                                                      Chlorine  Dioxide
                                                    Widespread outside
                                                    U.S.   On-site
                                                    generation
                                                               Chloramines
                                                              Limited use
                                                              On-site
                                                              generaticn
C
cti
  CO
W 4->
4-1 O
O 3
3 T3
T3 O
O S-i
l-l p<
(X I
I  -X)
>^ C
pq W
     Biocidal
     Activity
          Affec.ted by  pH
          HOC1 - third in
          activity
          OC1  - weak
     Measur.
     &
     Persist.
     of
     Residual
Health
   o10 mgd
a) co § 100 mgd
o 0^150 mgd
                    Little pH influence
                    Most active
          Eight methods
          HOC1 & OC1~
          not easily
          distinguished
          Persists
                             42
                    Titration method
                    Does not persist
                                                42
     Chemistry
           Trihalomethanes     Some,  but not yet
           other               well  characterized
           chlorinated         No trihalomethanes
           compounds, some
           of which are
           odorous
CHC1_ a carcinogen None positive yet,
epidemiology       Still under study
supports some
adverse effect
                 2  mg/JL
                                   mg/£
                               Air
                                           Oxygen
                 3.
                 1.
              25
             .08
           0.58
           0.55
,65
,88
,16
                                    1.11
7.
1,
1.
1.
33
96
13
07
                                                          Little  pH influence
                                                          Second  in activity
                                         Affected by pH
                                         NHC12 - Fourth in
                                         activity
                                                                               NH2C1 - Fifth
                                                                               activity
                                                                            in
                                                          DPD & FACTS
                                                          lodimetry, Mod.  OT
                                                          Amperiometric
                                                          titration
                                                          CIO  & CIO    can
                                                          be distinguished
                                                          Certain by-products
                                                          may interfere with DPD
                                                          Persists - slowly
                                                          changes to C"10~2 & C10~3
                                         Amperometric
                                         titration
                                         DPD
                                         Persists
                                         Some, but not yet
                                         well characterized
                                         No_trihalomethanes
                                         C102 & C10~3 are
                                         end products
None positive yet,
Both by-products
and end-products
under study

1 mg/&

3.45
                                                                               Unknown
                                                                               No trihalomethanes
                                                                               if no chlorine
                                                                               present
                                                                               None  positive yet,
                                                                               Studies  beginning
                                                              3 mg/A*
 ,63
 .25
 .22
                                                              2.33
                                                              0.61
                                                              0.27
                                                              0.25
    oj *Not  equivalent in biocidal activity to the other disinfectants.

-------
                                       - 71 -
     Because many on-going studies are not complete, the choice of a

disinfectant alternative to chlorine must be made on the basis of incomplete

information, if circumstances dictate that an alternate disinfectant is the

proper approach.  Although each of the alternatives have advantages and

disadvantages, at the present time (Fall 1977), four alternatives are worthy

of consideration:  1) chlorine dioxide; 2) ozone plus chlorine dioxide or

chloramines to maintain a residual; 3) chlorine followed shortly by ammonia

to stop the trihalomethane formation reaction; or 4) chlorine dosed at

a concentration less than the chlorine demand followed by chlorine dioxide or

chloramines to maintain a residual.  Each of these options would reduce

trihalomethane concentrations, would provide adequate initial disinfection

and would provide a biocide*  in the distribution system.  Although theoretically

attractive, chloramines are such weak biocides, that any option based on

their use should be evaluated prior to recommendation, something that

has not yet been accomplished for all situations.

     Whether required by a regulation or not, another prudent step would

be to limit the dose of any disinfectant, thereby reducing the potential

for the formation of any undesirable by-products and end-products regardless

of the disinfectant used.  In a given situation, if the disinfectant

demand is much greater than 1 mg/£, treatment of the water prior to

disinfection to reduce the disinfectant demand, see Figures 1 and 2,

would be generally beneficial.  As an extension of this concept, as

noted above, chlorine could continue to be used as the disinfectant if

the water being chlorinated were properly treated with a granular adsorbant

to remove organic matter.

     *Currently  (Fall 1977) some State regulatory agencies specify a "chlorine"

residual in the distribution system, something that must be taken into consideration

where applicable.  Another possible constraint to use of an alternate disinfectant

is Section 141.21(h) of the National Interim Primary Drinking Water Regulations that

discusses the substition of "chlorine" residual measurements for some required

microbiological contaminant samples.

-------
                                       - 72 -
     Although controling disinfectant demand is a prudent concept,  the problem




of what to use as a raw water disinfectant, where one is needed or  mandated,




remains.  The control of growths, such as fresh water clams,  in raw water




transmission lines, the prevention of slime growths in flocculating and




settling basins or the destruction of ammonia without producing undesirable




by-products and end-products is a practical problem that has  yet to be




completely evaluated.  Periodic shock chlorination or the use of small doses




of chlorine, ozone or chlorine dioxide may be effective for controlling




microorganism development.

-------
                                      - 73 -




                                        SUMMARY




     Providing a wholesome drinking water to the consumer is the primary




goal of all water purveyors.  Effective killing or inactivating of pathogenic




organisms through disinfection is the most important process in drinking




water treatment and distribution.  Because chlorination has been so effective




in controlling waterborne disease over the past 70 years in this country, any




departure from current practice must be made cautiously to avoid any




deterioration in the microbiological quality of the Nation's drinking water.




All other criteria by which to judge any new treatment proposal is secondary




to the maintenance of a water free of all pathogens.




     Several options exist for disinfecting water without by-product




contamination.  Treatment of water to remove by-product precursors or




treatment to remove formed by-products are possibilities.  This paper has




focused on a third possibility, replacement of chlorine as the disinfectant.




     Although all of the on-going research cited in this paper is not complete,




enough is known at this time (Fall 1977) to suggest that "chlorine-free"




or "low-chlorine-concentration" chlorine dioxide either as the primary




disinfectant or in combination with ozone as the primary disinfectant,




preferably used in waters with a disinfectant demand of about 1 mg/& or




less, would be two possible acceptable alternates to chlorine.  Although




not the only possible suggestions, these two meet all of the five criteria




established at the beginning of this paper as necessary for any alternative




to chlorine as a disinfectant.




                              ACKNOWLEDGEMENTS




     The authors and the compiler of this paper wish to thank Mr. Gordon G.




Robeck for his helpful suggestions during review and Mrs. Maura M. Lilly for




typing the several drafts and the final paper.

-------
                                     - 74 -

                                  REFERENCES
1.-   Lawrence, J. and Cappelli, F.P., "Ozone in Drinking Water Treat-
     ment: A Review.", The Science of the Total Environment, _7_, 2,
     99-108, (Mar 1977).

2.   Mignot, J., "Practices for Contacting Ozone with Liquids to be
     Treated.", In: Proceedings of the Second International Symposium
     on Ozone Technology, Montreal, Canada,(May 11-14, 1975), 15-46.

3.   Hill, A.G., "The Development of Ozonation: Application to Water
     and Wastewater Treatment.", Department of Chemical Engineering,
     Louisiana Tech. University, Ruston, Louisiana, (Dec 1976), mimeo,
     66 pp.

4.   OZONews—, _3, 1, Jan 1976, International Ozone Institute, Syracuse,
     N.Y.

5.   Miller, G.W., "Status of Ozonation and Chlorine Dioxide Tech
     nologies for Treatment of Municipal Water Supplies.", EPA Research
     Grant No. R804385-01, Washington, D.C., (In Progress).

6.   Ward, W.J., "Chlorine Dioxide— A New Selective Oxidant/Disinfectant
     for Wastewaters.", In: Proceedings of the Forum on Ozone
     Disinfection, Chicago, Illinois, (June 2-4, 1976), 382-393.

7.   White, G.C., "Handbook of Chlorination.", Van Nostrand Reinhold,
     New York, N.Y., (1972), 744 pp.

8.   Anon., "Treatment of Water Supplies with Chlorine Dioxide.",
     Olin Corporation, Stamford, Conn., (undated), mimeo, 10 pp.

9.   Gall, R.J., "Chlorine Dioxide-An Overview of its Preparation,
     Properties and Uses.", Hooker Chemicals and Plastics Corporation,
     Niagara Falls, N.Y., (1976), 36 pp.

10.  Anon., "Statistical Summary of Municipal Water Facilities in the
     U.S.—Jan. 1, 1963.", PHS Publ. No. 1039, U.S. Dept. of
     Health, Education and Welfare, Public Health Service, Division
     of Water Supply and Pollution Control, Washington, D.C. 20201.
     (1965).

11.  Race, J., "Chlorination and Chloromines.", JAWWA, _5,3,79,
     (Mar 1918).

12.  Anon., "Chloramine at Denver Solves Aftergrowth Problem.", Eng.
     News-Record, 210,  (Aug. 2, 1917).

-------
                                     - 75 -

13.  Hale, F.E., "The Chloramine Process as Applied to the Catskill
     (ESOPUS)  Water.", JAWWA.  ^, 6,  804, (June 1919).

14.  Scarpino, P.V., Berg,  G., Chang, L.L., Kahling, D.,  and Lucas, M.
     "A Comparative Study of the Inactivation of Viruses  in Water by
     Chlorine.", Water Research, 6^  No.  8, pp. 959-965,  (Aug.  1972).

15.  Morris, J.C. "Disinfectant Chemistry and Biocidal Activities.",
     In: Proc. National Specialty Congress on Disinfection., Amherst,
     Mass., (1970), pp 609-633.

16.  Peleg, M., "The Chemistry of Ozone in the Treatment  of Water.",
     Water Research, 10, No. 5, pp.  361-365, (May, 1976).

17.  Farooq, S., Chian, E.S.K., and Engelbrecht, R.S., "Basic Concepts
     In Disinfection with Ozone.", J. Water Pollution Control Federation,
     4£, No. 8, pp. 1818-1831, (Aug. 1977).

18.  Engelbrecht, R., "Virus Sensitivity to Chlorine Disinfection of
     Water Supplies.", EPA Res. Grant No. R803346, Univ.  of Illinois,
     (In Progress) .

19.  Benarde, M.A., Israel, B.M., Olivieri, V.O., and Granstrom, M.L.,
     "Efficiency of Chlorine Dioxide as a Bactericide.",  Applied
     Microbiology, 13, No.  5,  pp. 776-780, (Sep. 1965).

20.  Cronier, S., Scarpino, P.V., Zink, M.L., and Hoff,  J.C.,
     "Destruction by Chlorine Dioxide of Viruses and Bacteria in Water.",
     Abstr. Ann. Meeting American Society for Microbiology, (1977),
     p. 238.

21.  Moffa, P.E., and Smith, J.E., "Bench-scale high-rate disinfection
     of Combined Sewer Overflows with Chlorine and Chlorine Dioxide.",
     EPA Report No. S-80240, Adv. Waste Treat. Res. Lab,  NERC Cincinnati,
     Ohio,  (Nov. 1974), 181 pp.

22.  Kelly, S. and Sanderson,  W.A., "The Effect of Chlorine in Water
     on Enteric Viruses.",  American Journal of Public Health, 48,
     No. 10, pp. 1323-1334, (Oct. 1958).

23.  Bates, R.C., Shaffer,  P.T.B., and Sutherland, S.M.,  "Development
     of Polio-virus Having Increased Resistance to Chlorine
     Inactivation.", Applied and Environmental Microbiology, In Press,
     34:6  (Dec. 1977).

24.  Shaffer, P.T.B., Metcalf, T.G. and Sproul, O.J., "Chlorine
     Resistant Viruses from Drinking Water.", Presented at the National
     Environmental Engineering Conference of the American Society of
     Civil  Engineers, Vanderbilt University, Nashville, TN, July 11, 1977

-------
                                     - 76 -
25.  Sharp, D.G., Floyd, R., and Johnson, J.D., "Initial Fast Reaction
     of Bromine on Reovirus in Turbulent Flowing Water.", Applied
     and Environmental Microbiology, 31, No. 2, pp. 173-181, (Feb. 1976)

26.  Floyd, R. , Johnson, J.D., and Sharp, D.G., "Inactivation by
     Bromine of Single Poliovirus Particles in Water.", Applied and
     Environmental Microbiology, 31, No. 2, (Feb. 1976).

27.  Sharp, D.G., "Virus Particle Aggregation and Halogen Disinfection
     of Water  Supplies.", EPA Report No. 600/2-76-287, Municipal
     Environ.  Res. Lab, Cincinnati, OH, (Dec. 1976) pp. 59.

28.  Floyd, R. , and Sharp, D.G., "Aggregation of Poliovirus and
     Reovirus  by Dilution in Water.", Applied and Environmental
     Microbiology. 33, No. 1, pp. 159-167, (Jan. 1977).

29.  Young, D.C., and Sharp, D.G., "Poliovirus Aggregates and Their
     Survival  in Water.", Applied and Environmental Microbiology,
     .33, No. 1, pp. 168-177, (Jan. 1977).

30.  Newton, W.L. and Jones, M.G., "The Effect of Ozone in Water
     on Cysts  of Endamoeba histolytica.", American Journal of
     Tropical Medicine, 29_, No. 5, pp. 669-681, May 1949).

31.  Palin, A.T., "Water Disinfection - Chemical Aspects and Analytical
     Control.", In: Disinfection Water and Wastewater.  J.D. Johnson,
     Ed., Ann Arbor Sci., Ann Arbor, Mich., 1970, 425 p.

32.  Meyer, E.A. "Determination of Giardia Cyst Viability.", EPA
     Res. Grant No. R804898, Univ. of Oregon Health Sciences Center,
     (In Progress).

33.  Chang, S.L., Berg, G., Clarke, N.A., and Kabler, P.W., "Survival
     and Protection Against Chlorination, of Human Enteric Pathogens
     in Free-living Nematodes Isolated from Water Supplies.",
     American Journal of Tropical Medicine, 9_,  No.  1, pp. 136-142,
     (Jan. 1960).

34.  Tracy, H.W., Camarena, V.M., and Wing, F., "Coliform persistence
     in Highly Chlorinated Waters.", Journal of the American Water
     Works Association, 58, No. 9, pp. 1151-1159, (Sep. 1966).

35.  Boardman, G.D., "Protection of Waterborne Viruses by Virtue of
     Their Affiliation with Particulate Matter.", Ph.D. Thesis.,
     Univ. of Maine, 196 p., (Dec. 1976).

-------
                                    - 77 -
36.  Stagg, C.H., Wallis, C., and Ward, C.H., "Inactivation of Clay-
     Associated Bacteriophage MS-2 by Chlorine.", Applied and
     Environmental Microbiology, 33, No. 2, pp. 385-391, (Feb. 1977).

37.  Symons, J.M., and Hoff, J.C., "Rationale for Turbidity Maximum
     Contaminant Level.", In:  Proc. AWWA Technol. Conf., Atlanta,
     Georgia, (Dec. 1975), pp. 2A-4a, 1-15.

38.  Hoff, J.C., "Relationship of Turbidity to Disinfection of Potable
     Water.", USEPA Office of Water Supply, Symposium on Eval. of
     Microbiol. Stand, for Water, (1977), In press.

39.  Kruse, C.W. and Olivieri, V.P., "Biological Evaluation of the
     Benefits of Maintaining a Chlorine Residual in Public Water
     Systems," EPA Res. Grant No. R804307, Johns Hopkins Univ. (In
     Progress).

40.  Brodeur, T.P., Singley, J.E., and Thurrott, J.C.-, "Effects of a
     Change to Free Chlorine Residual at Daytona Beach, Florida.",
     In: AWWA Technol. Conf. Proceedings, San Diego, Calif., (Dec.
     1976), pp. 3A-5, 1-12.

41a. Vendryes, J.H., "Experiences with the Use of Free Residual
     Chlorination in the Water Supply of the City of Kingston, Jamaica.",
     In: Proc. AIDIS Cong, of Washington, D.C., (1962).

41b. Buelow, R.W. and Walton, G., "Bacteriological Quality vs. Residual
     Chlorine," JAWWA, 63, No. 1, 28-35 (Jan. 1971).

42.  Anon., "Standard Methods for the Examination of Water and Waste-
     Water."  14th ed., Amer. Pub. Health Assn., New York, N.Y., (1975).

43.  Dailey, L., and Morrow, J., "On Stream Analysis of Ozone Residual,"
     In:  Proceedings, First International Symposium on Ozone for Water
     and Wastewater Treatment, Vol.  I, Rice, R.G., and Browning, M.E.,
     Eds., International Ozone Institute, (1975).

44.  Miltner, R.J., "Measurement of Chlorine Dioxide and Related
     Products"  In:  Proceedings AWWA Water Quality Technology Conf.,
     Dec. 6-7, 1976, San Diego, Calif.  Paper No. 2A-5; AWWA, Denver,
     Colorado, 1977.

45.  Palin, A.T., "Methods for the Determination, in Water, of Free
     and Combined Available Chlorine, Chlorine Dioxide and Chlorite,
     Bromine, Iodine and Ozone, Using Diethyl-p-phenylene Diamine
     (DPD)."  Jour. Inst. Water Engr., 21. 537, (1967).

46.  Palin, A. T.,  "Analytical Control of Water Disinfection with
     Special Reference to Differential DPD Methods for Chlorine, Chlorine
     Dioxide, Bromine, Iodine and Ozone."  Jour. Inst. Water Engr.,
     _28, 139-154, (1974).

-------
                                       - 78 -

47.  Adams, D., Carter, J., Jackson, D. and Ogelby, J.,  "Determination
     of Trace Quantities of Chlorine, Chlorine Dioxide,  Chlorite and
     Chloramines in Water."  Proc. Soc. Water Treatment  and Exam.,
     JL5, 117, (1966).

48.  Hong, C. and Rapson, W., "Analyses for Chlorine Dioxide, Chlorous
     Acid, Chlorite, Chlorate, and Chloride in Composite Mixtures."
     Can. Jour. Chem. 46, 2061, (1968).

49.  Miltner, R.J., "The Effect of Chlorine Dioxide on Trihalomethanes
     in Drinking Water."  M.S. Thesis, University of Cincinnati,
     August 1976.

50.  Seeger, D.R., USEPA Cincinnati, personal communication, 1977.

51.  Johnson, J.D., "Measurement and Persistence of Chlorine Residuals
     in Natural Water"  In:  Proceedings of the  Conference on the
     Environmental Impact of Water Chlorination,  CONF-751096,
     R.L. Jolley, Ed., Oak Ridge National Laboratory, Oak Ridge,
     Tennessee (1976).

52.  Rook, J.J., "Formation of Haloforms During Chlorination of
     Natural Waters,"  Water Treatment and Examination,  23, Part
     2, 234-243, (1974).

53.  Bellar, T.A., Lichtenberg, J.J. and Kroner, R.C., "The Occur-
     rence of Organohalides in Chlorinated Drinking Water,"  Jour.
     AWWA, 66. No. 12, 703, (Dec. 1974).

54.  Symons, J.M., Bellar, T.A., Carswell, J.K., DeMarco, J., Kropp,
     K.L., Robeck, G.G., Seeger, D.R., Slocum, C.J., Smith, B.L., and
     Stevens, A.A., "National Organics Reconnaissance Survey for
     Halogenated Organics in Drinking Water," Water Supply Research
     Laboratory and Methods Development and Quality Assurance Labora-
     tory, National Environmental Research Center, U.S.  EPA, Cincinnati,
     Ohio, Jour. AWWA, 67, 634, (Nov. 1975).

55.  Stevens, A.A., Slocum, C.J., Seeger, D.R., Robeck,  G.G.,
     Chlorination of Organics in Drinking Water.  In: Proceedings,
     Conference on the Environmental Impact of Water Chlorination,
     CONF-751096, R.L. Jolley, Ed., Oak Ridge National Laboratory, Oak
     Ridge, Tennessee.  See also:  Jour. AWWA, 68. No. 11, 615,  (Nov. 1976)

56.  Love, O.T., Jr., Carswell, J.K., Miltner, R.J. and Symons,  J.M.,
     "Treatment for the Prevention or Removal of Trihalomethanes  in
     Drinking Water,"  In:  Symons, J.M., "Interim Treatment Guide
     for  the Control of Chloroform and Other Trihalomethanes,"
     Water Supply Research Division, Municipal Environmental Research
     Laboratory, U.S. EPA, Cincinnati, Ohio (June 1976) App. 3

-------
                                      - 79 -
57.  Coleman, W.E.,  Lingg,  R.D.,  Melton,  R.G.,  Kopfler,  F.C.,  "The
     Occurrence of Volatile Organics in Five Drinking Water Supplies
     Using Gas Chrbmatography/Mass Spectrometer (GC/MS),"  In:
     "Identification and Analysis of Organic Pollutants  in Water,"
     1st ed, Keith,  L.  H.,  Ed.,  Ann Arbor Science Publishers,  Inc.,
     Ann Arbor, Michigan, (1975)  Chapter 21.

58.  Gould, E.S., "Mechanism and Structure in Organic Chemistry,"
     Holt, Reinhart & Winston, New York,  (1964).

59.  Stevens, A.A.,  Symons, J.M., "Measurement of Trihalomethane and
     Precursor Concentration Changes" Jour. AWWA, 69, No. 10,  546-554,
     (Oct. 1977).

60.  Rook, J.J., "Haloforms in Drinking Water," Jour. AWWA, 68, No. 3,
     168-172 (Mar. 1976).

61.  Seeger, D.R., 1976, U.S.  EPA, Cincinnati,  personal  communica-
     tion

62.  Bunn, W. W., Haas, B.B.,  Deane, E.R., and Keopfler, R.D.,
     "Formation of Trihalomethanes by Chlorination of Surface  Water,"
     Environmental Letters, 10,  No. 3, 205-214, (1975).

63.  Moore, L., 1976, U.S.  EPA,  Cincinnati, personal communication.

64.  Rook, J.J., "Chlorination Reactions of Fulvic Acids in Natural
     Waters,"  Environmental Science & Technology, 11, No. 5,  478,
     (1977).

65.  Jolley, R.L., "Chlorine-Containing Organic Constituents,  in
     Chlorinated Effluents,"  Jour. WPCF, 47, No. 3, 601-618,  (1975).

66.  Glaze, W.H., and Henderson,  J.E., IV, Formation of  Organo-
     chlorine Compounds from the Chlorination of a Municipal Secondary
     Effluent, Jour. WPCF.  47, No. 10, 2511-2515, (Oct.  1975).

67.  Burttschell, R.H., Rosen, A.A., Middleton, F.M., and Ettinger,
     M.B., "Chlorine Derivatives of Phenol Causing Taste and Odor,"
     Jour. AWWA, ,51, 205-214,  (Feb. 1959).

68.  Anon., "Report on the  Carcinogenesis Bioassay of Chloroform,"
     Carcinogen Bioassay and Program Resources  Branch, Carcino-
     genesis Program, Division of Cancer Cause and Prevention,  National
     Cancer Institute.

69.  Train, R.E., U.S.  EPA  News Release,  March 29, 1976.

-------
                                     - 80 -
70.  Sontheimerj H. , Heilker, E., Jekel, M. , Nolte, H. and Vollmer,
     F.H., "The 'Mulheim Process1 - Experience with a New Process
     Scheme for Treating Polluted Surface Water, " Jour. AWWA,  In
     Press.

71.  "Interim Treatment Guide for the Control of Chloroform and
     Other Trihalomethanes," Water Supply Research Division, Municipal
     Environmental Research Laboratory, Office of Research and
     Development, U.S. Environmental Protection Agency, Cincinnati,
     Ohio 45268, June 1976.

72.  Gordon, G., Kieffer, R.G., and Rosenblatt, D.H., "The Chemistry
     of Chlorine Dioxide:  In:  Progress in Inorganic Chemistry,
     Vol. 15, Lippard, S.J., Ed., Wiley - Interscience, New York,
     (1972), p. 201.

73.  Black, A.P. , and Christman, R.F. "Chemical Characteristics of
     Fulvic Acids," Jour. AWWA, 55, 897 (1963).

74.  Stevens, A.A., Seeger, D.R., Slocum, C.J., "Products of Chlorine
     Dioxide Treatment of Organic Materials in Water."  To be pub-
     lished in:  "Proceedings of Workshop on Ozone/Chlorine Dioxide
     Oxidation Products of Organic Materials, Nov. 17-19, 1976,
     Cincinnati, Ohio.

75.  Morris, J.C., "Formation of Halogenated Organics by Chlorination
     of Water Supplies,"  EPA-600/1-75-002, U.S. Environmental
     Protection Agency, Washington, D.C., (1975).

76.  Dence, C.W., Gupta, M.K., and Sarkanen, R.V., "Studies on
     Oxidative Delignification Mechanisms, Part II.  Reactions  of
     Vanillyl Alcohol with Chlorine Dioxide and Sodium Chlorite.",
     Tappi, 45, 29, (1962).

77.  Dence, C.W., and Sarkanen, K.V., "A Proposed Mechanism for the
     Acidic Chlorination of Softwood Lignin," Tappi, 43, 87, (1960).

78.  Glabisz, U. , "The Reactions of Chlorine Dioxide with Components
     of Phenolic Wastewaters - Summary.  Monograph 44, Polytechnic
     University, Szczecin, Poland (1968).

79.  Bellar, T. A., and Lichtenberg, J. J., "Determining Volatile
     Organics at Microgram-Per-Liter Levels by Gas Chromatography,"
     Jour. AWWA, 66, No. 12, 739-744, (Dec. 1974).

80.  Christman, R.F., and Ghassemi, M., "Chemical Nature of Organic
     Color in Water,"  Jour. AWWA, 58, 723, (1966).

-------
                                   - 81 -
81.  Sraillie, R.D., Nicholson, A.A., Meresz, 0., Suholke, W.K.,
     Rees, G.A.V., Roberts, K., and Fund, C., "Organics in Ontario
     Drinking Waters, Part 11, A Survey of Selected Water Treatment
     Plants," OTC 7703, AAN-7504, Ontario Ministry of the Environ-
     ment, 1977.

82.  Tuepker, J.L., "Sampling and Analysis of Chloro-Organics in the
     Distribution System,"  Proceedings 4th Annual AWWA Water Tech-
     nology Conference, Dec. 1976, San Diego, California, AWWA, Denver,
     Colorado, 1977.

83.  Lambert, M., Vilagines, R., Montiel, A., and Derreumaux, A.,
     "A Comparative Study of Halomethanes Formation During Drinking
     Water Treatment by Chlorine or its Derivatives"  Presented at
     the 96th Annual AWWA Conference, Anaheim, CA., May 8, 1977.

84.  Harms, L.  "Preventing Haloform Formation in Drinking Water,"
     South Dakota School of Mines and Technology.  EPA Research Grant
     No. R805149-01 (In progress).

85.  Razor, W., "Organic Substances in the Ohio River and Associated
     Water Supplies,"  Ohio River Valley Water Sanitation Commission
     (ORSANCO) EPA Grant No. 804615-02 (In progress).

86.  DeSalva, S., Volpe, A., Leigh, G., and Regan, T., "Long Term
     Safety Studies of a Chloroform-Containing Dentrifice and Mouth
     Rinse in Man," Food Cosmet. Toxicol., 13, 529-532 (1975).

87.  Anon. "Criteria for a Recommended Standard — Occupational
     Exposure to Chloroform," National Institutes of Occupational
     Safety and Health, HEW Publication No. (NIOSH) 75-114, U.S.
     Government Printing Office, Washington, D.C. (1974).

88.  Wallace, C.J., "Hepititis and Nephrosis Due to Cough Syrup
     Containing Chloroform," Calif. Med., 73, 442 (1959).

89.  Eschenbrenner, A., "Induction of Hepatomasin Mice by Repeated
     Oral Administration of Chloroform with Observations on Sex
     Differences," J. Nat. Cancer Inst.,  5, 251-255 (1945).

90.  Federal Register, 41, No. 126 (June, 1976). p. 26842

91.  Kimura, E.T., Ebert, D.M. and Dodge, P.W. "Acute Toxicity and
     Limits of Solvent Residue for Sixteen Organic Solvents," Toxicol.
     Appl. Pharmacol.. lj), 699-704 (1971).

92.  Hill, R.N., Clemens, T.L., Liu, D.K., Vesell, E.S., and
     Johnson, W.D., "Genetic Control of Chloroform Toxicity in Mice,"
     Science, 190, 159-161 (1975).

93.  Thompson, D.J., Warner, S.D., and Robinson, B.V., "Teratology
     Studies on Orally Administered Chloroform in the Rat and Rabbit,"
     Toxicol. Appl. Pharmacol. 29, 348-357 (1974).

-------
                                     - 82 -
94.  Anon. Drinking Water and Health, A Report of the Safe Drinking
     Water Committee, Advisory Center on Toxicology Assembly of Life
     Sciences, National Research Council, National Academy of Sciences,
     Washington, D.C. (1977).

95.  Roe, F.J.C., "Preliminary Report of Long-Term Tests of Chloroform
     in Rats, Mice and Dogs," Unpublished Report (1976).

96.  Wlodkowski, T.V., Rosencranz, H.S. "Mutagenicity of Sodium
     Hypochlorite for Salmonella typhimurium" Mutation Res., 31, No. 1,
     39-42 (1975).

97.  Cantor, K.P. "The Epidemiological Approach to the Evaluation of
     the Health Effects of Water Chlorination"  Presented at Conference
     on Water Chlorination:  Environmental Impact and Health Effects,
     Oct. 31 - Nov. 4, 1977, Gatlinburg, Tennessee.
98.  Page, T., Harris, R.H., and Epstein, S.S., "Drinking Water and
     Cancer Mortality in Louisiana", Science, 193, 55-57, (July 2,
     1976).

99.  Kuzma, R.J., Kuzma, D.M. and Buncher, C.R., "Ohio Drinking Water
     Source and Cancer Rates"  Am. J. of Public Health, 67, No. 8,
     725-729 (1977).

100. Salg, J., "Cancer Mortality Rates and Drinking Water in 346
     Counties of the Ohio River Valley Basin,"  Report on purchase
     order No. 5-03-4528J U.S. EPA, Cincinnati, Ohio, mimeo.  (1977).

101. Kruse, C., "Preliminary Report:  Chlorination of Public Water
     Supplies and Cancer, Washington County, Maryland." Report on
     purchase order No. CA-6-99-3284 U.S. EPA, Cincinnati, Ohio, mimeo.
     (1977).

102. Spivey, G.H., Sloss, E., and Mah, R.A., "Cancer and Chlorination
     of Drinking Water (Los Angeles County)"  Part II, Report on
     purchase order No. CA-6-99-3349J  U.S. EPA, Cincinnati, Ohio,
     mimeo.  (1977).

103. Alavanja, M. , Goldstein, I., and Susser, M., "Case Control Study
     of Gastrointestinal Cancer Mortality in Seven Selected New York
     Counties in Relation to Drinking Water Chlorination."  Report on
     purchase order No. CA-7-2805-J.  U.S. EPA, Cincinnati, Ohio,
     mimeo.  (1977).  Also presented at Conference on Water Chlorination:
     Environmental Impact and Health Effects, Oct. 31 - Nov. 4, 1977,
     Gatlinburg, Tennessee.

-------
                                     - 83 -
104. Hartemann, J.,  "Biochemical Aspects of the Toxicity Involved by
     the Ozone Organic Oxidation Products in Water," presented at
     101 Workshop,  Cincinnati, Ohio, (Nov. 1976).

105. Kinman, R.,  et al., "Effects of Ozone on Hospital Waste Water
     Cytotoxicity." presented at 101 Workshop, Cincinnati,  Ohio,
     (Nov. 1976).

106. Simmon, V.F.,  Eckford, S.L., "Methods for Evaluating the Muta-
     genic Activity of Ozonated Chemicals," presented at 101 Workshop,
     Cincinnati,  Ohio, (Nov. 1976).

107. Haller, J.F. and Northgraves, W.W., "Chlorine Dioxide and Safety"
     TAPPI, ^38, No.  4, 199-202, (1955).

108. Elkins, H.B.,  "The Chemistry of Industrial Toxicology," John
     Wiley and Sons, Inc., New York, N.Y. (1950) p. 87.

109. Haag, H.B. "The Effect on Rats of Chronic Administration of
     Sodium Chlorite and Chlorine Dioxide in the Drinking Water"
     Report to The Mathieson Alkali Works, Niagara Falls, N.Y.
     (Feb. 7, 1949).

110. Enger, M., "Treatment of Water With Chlorine Dioxide to Improve
     Taste," Gas  und Wasser Fach, 3.4, 340-343 (Apr. 1, 1960).

111. Musil, J., Kontek, Z., Chalupa, J., and Schmit, P., "Toxicological
     Aspects of Chlorine Dioxide Application for the Treatment of
     Water Containing Phenol," Sf vys. Sk. Chem-Technol. Praze, 8,
     327-345 (June 1963).

112. Augenstein,  H.W., "Use of C10? to Disinfect Water Supplies"
     JAWWA, 66_, 716-717 (Dec. 1974J.

113. Myhrstad, J.A.  and Samdal, J.E., "Behavior and Determination of
     Chlorine Dioxide," JAWWA, 61, 205-208 (April 1969).
114. Sperling, F., "LD5n Determination of Sodium Chlorite"  Report
     No. OM-B5 to Olin Mathieson Chemical Corp., Baltimore, Md.
     (Oct. 8, 1959).
115. Macek, K.J., "Acute Toxicity of Sodium Chlorite to Bluegill
     (Lepomis macrochirus) and Rainbow Trout (Almo gairdner:)"
     Report to Olin Water Services, Stamford, Conn., Dec. 1971.

116. Calandra, J.C. "Acute Oral Toxicity Study with Sodium Chlorite
     in Mallard Ducks."  Report No. 1BT-J2118 to Olin Corporation,
     New Haven, Conn.  (Jan. 9, 1973).

117. Hopf, W., "Experiments with Chlorine Dioxide Treatment of
     Drinking Water", Gas und Wasser Fach, 108, No. 30, 852-854, (1967)

118. Jandl, J.H., Engle, L.K. and Allen, D.W., "Oxidative Hemolysis
     and Precipitation of Hemoglobin, I.  Heinz Body Anemias as an
     Acceleration of Red Cell Aging,"  J. Clin. Invest. 12, No. 39,
     1818-1836, (Dec. 1960).

-------
                                      - 84 -
119. Davidson, I. and Henry, J.B., "Clinical Diagnosis by Laboratory
     Methods," 14th ed. Saunders, Philadelphia, PA  (1962) p. 135.

120. Jaffe, E.R., "Introduction  to Discussion of Red Cell Metabolism:
     Exp. Eye Res., 11, 306-309,  (1971).

121. Tardiff, R.G., Bercz, J.  "The Assessment of Chlorine Dioxide
     in Drinking Water on the Hematopoietic System  of African Green
     Monkey."  Project No. TAB-006  U.S. EPA, Health Effects
     Research Lab., Cincinnati,  OH (Aug. 1977).

122. Spicer, S.S., "Species Differences in Susceptibility to Methemo-
     globin Formation," J. Pharm. Exp. Ther., 99, 185-194 (1950).

123. Goldstein, A., Aronow, L.,  Kalman, S., "Principles of Drug Action:
     The Basis of Pharmacology,"  2nd Ed., John Wiley & Son, N.Y., N.Y.,
     (1974).  pp. 461.

124. Calabrese, E.J., Kojola, W.H. and Carnow, B.W., "Ozone:  A Possible
     Cause of Hemolytic Anemia in Glucose-6-Phosphate Dehydrogenase
     Deficient Individuals," J.  Tox. Environ. Health, 2, 709-712, (1977).

125. Kiese, M., "Methemoglobinemia:  A Comprehensive Treatise," CRC Press
     Inc., Cleveland, Ohio (1974), p. 34.

126. Eaton, J.W., Kolpin, C.F. and Swofford, H.S.,  "Chlorinated Urban
     Water:  A Cause of Dialysis  Induced Hemolytic  Anemia,"  Science,
     181, No. 4098, 463-464 (Aug. 3, 1973).

127. Shih, K.L., Lederberg, J.,  "Chloramine Mutagenesis in Bacillus
     subtilis,"  Science, 192, No. 4244, 1141-1143  (June 11, 1976).

128. Patterson, W.L., and Banker, R.F., "Estimating Costs and Manpower
     Requirements for Conventional Wastewater Treatment Facilities
     for the Environmental Protection Agency," Black and Veatch,
     Consulting Engineers, Kansas City, Missouri (1971).

129. Eilers, Richard G. and Smith, Robert, "Executive Digital Computer
     Program for Preliminary Design of Wastewater Treatment Systems."
     NTIS-PB222765 (report NTIS-PB222764 (card deck), (November 1970).

130. Muegge, O.J., "Physiological Effects of Heavily Chlorinated Drinking
     Water,"  Jour. Am. Water Works Assn.. 48, No.  12, 1507-1509 (Dec. 1956)

                                                  6 U.S. GOVERNMENT PRINTING OFFICt 1977- 7 57 - 140 /6601

-------